Information Technology for Patient Empowerment in Healthcare 9781614514343, 9781614515920

Table of contents :
Contents
Preface
About the editors
Introduction
List of contributing authors
Acknowledgements
Part I: Information Technology for Patient Empowerment: 360° Perspectives
1 Patient-centered healthcare, patient engagement and health information technology : the perfect storm
2 Placing patients at the center of patient-centered care : a healthcare provider system perspective of a powerful new technology-enabled “language”
3 Using health IT to engage patients in choosing their doctors, health plans and treatments
4 Old media to new in health: from information to interactivity
5 Policy context and considerations for patient engagement with health information technology
Part II: Current and Future Information Technology Solutions for Patient Empowerment
6 Patient portals can enable provider-patient collaboration and person-centered care
7 Data model for integrated patient portals
8 Telehealth: connecting patients with providers in the 21st century
9 Patient-controlled sharing of medical imaging data
10 Patient empowerment via technologies for patient-friendly personalized language
11 Finding and understanding medical information online
12 Electronic media for engaging patients in the research consent decision process
13 Patient engagement at the point of care: technology as an enabler
14 Supporting active patient self-care
15 Using patient-reported outcomes to drive patientcentered care
Index

Citation preview

Grando, Rozenblum, Bates Information Technology for Patient Empowerment in Healthcare

Information Technology for Patient Empowerment in Healthcare Edited by Maria Adela Grando, Ronen Rozenblum, David W. Bates

Editors Maria Adela Grando Arizona State University, Department of Biomedical Informatics 13212 East Shea Boulevard Scottsdale, AZ 85259 USA

David W. Bates Brigham and Women’s Hospital, DGIM 1620 Tremont Street Boston, MA 02120 USA

Ronen Rozenblum Brigham and Women’s Hospital, DGIM 1620 Tremont Street Boston, MA 02120 USA

ISBN 978-1-61451-592-0 e-ISBN 978-1-61451-434-3 e-ISBN (EPUB) 978-1-61451-955-3 Set-ISBN 978-1-61451-435-0 Library of Congress Cataloging-in-Publication data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.dnb.de. The publisher, together with the editors and authors, has made efforts to ensure that all information presented in this work (programs, applications, amounts, dosages, etc.) reflects the standard of knowledge at the time of publication. Despite careful manuscript preparation and proof correction, errors can nevertheless occur. Authors, editors and publisher disclaim all responsibility for any errors or omissions or liability for the results obtained from use of the information, or parts thereof, contained in this work. The citation of registered names, trade names, trademarks, etc. in this work does not imply, even in the absence of a specific statement, that such names are exempt from laws and regulations protecting trademarks etc. and therefore free for general use. © 2015 Walter de Gruyter Inc., Berlin/Boston/Munich Typesetting: PTP-Berlin Protago-TEX-Production GmbH, Berlin Printing and binding: CPI books GmbH, Leck Cover image: Ioana Davies/Stock Photo/123rf ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Contents Preface | VII About the editors | IX Introduction | XI List of contributing authors | XIII Acknowledgements | XVII

Part I: Information Technology for Patient Empowerment: 360° Perspectives Ronen Rozenblum, Paula Miller, Disty Pearson and Ariane Marelli 1 Patient-centered healthcare, patient engagement and health information technology: the perfect storm | 3 C. Martin Harris and Gene Lazuta 2 Placing patients at the center of patient-centered care: a healthcare provider system perspective of a powerful new technology-enabled “language” | 23 David Lansky and Stephanie Glier 3 Using health IT to engage patients in choosing their doctors, health plans and treatments | 39 Michael L. Millenson and Jane Sarasohn-Kahn 4 Old media to new in health: from information to interactivity | 59 Asaf Bitton, Michael Poku and David W. Bates 5 Policy context and considerations for patient engagement with health information technology | 75

Part II: Current and Future Information Technology Solutions for Patient Empowerment Mary Jo Deering and Cynthia Baur 6 Patient portals can enable provider-patient collaboration and person-centered care | 93 Mary McNamara 7 Data model for integrated patient portals | 113

VI 

 Contents

Roy Schoenberg 8 Telehealth: connecting patients with providers in the 21st century | 125 Yaorong Ge and J. Jeffrey Carr 9 Patient-controlled sharing of medical imaging data | 141 Mehnaz Adnan, Jim Warren and Hanna Suominen 10 Patient empowerment via technologies for patient-friendly personalized language  | 153 Jeff Harwell, Christopher Pentoney and Gondy Leroy 11 Finding and understanding medical information online | 165 Yukari Schneider, Maria Adela Grando, Jihad S. Obeid and Wajeeh Bajwa 12 Electronic media for engaging patients in the research consent decision process | 179 Catalina Danis, Martha Jean (Marty) Minniti, Sasha Ballen, Marion Ball, Scott Cashon, Margaret Piscitelli, Marjorie Miller and Robert Farrell 13 Patient engagement at the point of care: technology as an enabler | 199 Maarten van der Heijden, Marina Velikova and Peter J.F. Lucas 14 Supporting active patient self-care | 227 Eyal Zimlichman 15 Using patient-reported outcomes to drive patient-centered care | 241 Index | 257

Preface Not so long ago, health care was organized around what physicians did; it is now undergoing a fundamental re-organization around meeting the needs of patients. This change is not based upon the sudden evolution of health care providers to some more altruistic state, or the seizure of power by patient advocates. This change reflects progress in our understanding of the real challenges we face in health care delivery, and enhancement of our ability to meet them. For practical purposes, the earlier physician-centric era can be defined as the period before the 1999 publication of To Err is Human by the Institute of Medicine (IOM). Before that report, the widespread assumption in health care was that quality was basically just fine, and not really measurable in any case. A further assumption was that most actions by health care providers were beneficial to patients – and that the more providers did, the more good they would be doing. Accordingly, health care providers, their leadership, and their governing boards focused on doing as much as possible, and on maximizing the revenues they received for their activities – so that they could invest in the resources they needed to do even more. The IOM report revealed, however, that all was not well in health care, that errors occurred with unsettling regularity, and that many had devastating consequences. Quality could in fact be measured, and the data that emerged revealed variability and unreliability. These painful insights meant that health care could no longer be organized around maximizing clinician activities that were reimbursed under fee-for-service payments. Instead, the paradigm that has steadily won acceptance is that health care should be organized around reliably meeting the needs of patients, and doing so as efficiently as possible. This change in focus does not devalue the hard work of clinicians; it simply means that this work needs to be conducted with a more appropriate goal than maximizing fee-for-service revenue. That goal is meeting patients’ needs – it is producing health itself, not just producing health care. The pursuit of that goal has become increasingly realistic – indeed, it is defining the life work of many of the “best and brightest” of young health care providers today, as well as their predecessors. How to measure what matters to patients is more deeply understood, as is the way in which clinicians need to work with each other and with patients themselves in order to enhance outcomes. Indeed, I believe that a vibrant health care marketplace is emerging in which providers truly compete around creating value for patients – and that providers who may have been skeptical or oblivious to market needs in the past are realizing that such competition will reward providers who are most innovative and effective. Their patients will win as well, with better and more affordable care. In the exciting period that lies just ahead, more will be needed than simply connecting patients to clinicians, and clinicians to each other. The health care systems

VIII 

 Preface

that will be most effective in meeting patients’ needs will be those that can actually design their “human-ware” around that purpose. This book provides deep insights into how information technology can and will support that redesign. In all likelihood, the health care providers that are most effective and efficient in making these changes will be rewarded with market share, business success, and pride in the quality of care they are delivering. Thomas H. Lee, MD, MSc. Chief Medical Officer, Press Ganey Associates, South Bend, IN Professor of Medicine, Harvard Medical School and Professor of Health Policy and Management, Harvard School of Public Health, Boston, MA

About the editors Maria Adela Grando Maria Adela Grando, PhD is an Assistant Professor at the Department of Biomedical Informatics at the Arizona State University and an Adjunct Assistant Professor of Medicine at the Mayo Clinic, College of Medicine. She has over 7 years of experience on applying methodologies from artificial intelligence for the development of information technologies to support clinical decision processes. Dr. Grando holds a Master in Computer Science from Argentina, and a PhD in Formal Languages and Applications from Rovira i Virgili University (Spain). She completed two post-doctoral training research fellowships in Biomedical Informatics at the UK Cancer Research (Oxford University, Edinburgh University and the University College London) and at the Division of Biomedical Informatics at the University of California, San Diego. Dr. Grando’s research at UK Cancer Research was concentrated on building technologies to support the decision process of clinicians. However, in the last 4 years, she has shifted her research towards the deployment of tools to empower patients to make healthcare decisions. Dr. Grando has led multiple projects in the development of tablet-based electronic informed consent systems to engage and educate patients to make better informed decisions when deciding to participate in research studies. In particular, she has focused on building technologies to support patients’ desire to have granular control over sharing healthcare data for research. Currently, she is spearheading the development of several evidence-based smartphone applications to support behavioral change, including apps for engaging users on self-management activities to improve disease control, and apps for building family management skills to improve parent-children interactions. Dr. Grando is an active member of the American Medical Informatics Association. She has authored more than 40 publications in the field of medical informatics, including articles in the leading journals in the field.

X 

 About the editors

Ronen Rozenblum Dr. Ronen Rozenblum is a researcher, lecturer and entrepreneur. He is the Founding Director of the Unit for Innovative Healthcare Practice & Technology and Director of Business Development of the Center for Patient Safety Research and Practice at Brigham and Women’s Hospital in Boston. Dr. Rozenblum has a faculty appointment at Harvard Medical School where he is a health services researcher. He is an expert in patient-centered care, patient experience and engagement, and health information technology. His research, lectures, and contributions have been recognized nationally and internationally. Dr. Rozenblum holds a Master’s degree in Public Health (MPH) and a PhD in Business Administration, Health Management and Economics. In 2011, he completed his post-doctoral research fellowship in patient-centered care, patient experience, quality of care, and medical informatics at Harvard Medical School. Dr. Rozenblum has a wide breadth of expertise and practical knowledge in developing innovative methods and information technologies to enhance patient-centered care, patient experience and engagement. He has trained many clinicians to implement these tools in clinical settings and has presented his work at numerous national and international conferences. He lectures on patient-centered care in a HarvardX course entitled “Improving Global Health: Focusing on Quality and Safety” at the Harvard School of Public Health. He is a member of the Patient Experience Strategy Committee at Brigham and Women’s Hospital, the Medical Advisory Board of the Adult Congenital Heart Association in the U.S., the Abstract Committee of the International Society for Quality in Healthcare (ISQua) Conference and the Israeli National Patient Experience Survey Committee. In addition to his academic pursuits, Dr. Rozenblum has been involved in establishing and managing startup companies in the healthcare industry and has considerable experience in business development and management of such entities.

About the editors 

 XI

David W. Bates Dr. Bates is an internationally renowned expert in using information technology to improve clinical decision-making, patient safety, quality-of-care, cost-effectiveness, and outcomes assessment in medical practice. A practicing general internist, Dr. Bates is Chief Quality Officer at Brigham and Women’s Hospital in Boston where he is also Chief of the Division of General Internal Medicine. He is a Professor of Medicine at Harvard Medical School, and a Professor of Health Policy and Management at the Harvard School of Public Health, where he co-directs the Program in Clinical Effectiveness. He also serves as Medical Director of Clinical and Quality Analysis for Partners HealthCare. Dr. Bates is a graduate of Stanford University, and the Johns Hopkins School of Medicine. He began his fellowship in general internal medicine at Brigham and Women’s Hospital in 1988, and he received an M.Sc. in Health Policy and Management from the Harvard School of Public Health in 1990. He has been elected to the Institute of Medicine, the American Society for Clinical Investigation, the Association of American Physicians and the American College of Medical Informatics, and is past chairman of the Board of the American Medical Informatics Association. He serves as external program lead for research in the World Health Organization’s Global Alliance for Patient Safety. He is the president of the International Society for Quality in Healthcare (ISQua). He is the editor of the Journal of Patient Safety. Dr. Bates’ special research interests include clinical decision-making and affecting physician-decision-making, particularly using computerized interventions; quality of care and cost-effectiveness and medical practice; and outcome assessment. He has published over 700 peer-reviewed papers.

Introduction Every day millions of people enter hospitals, ambulatory care centers or clinics and assume the role of “patient”. This can be a scary transition – entering the healthcare system can create enormous fear and uncertainty, and often requires relinquishing control. Many patients may not understand their diagnoses, or the risks of the care they are about to receive. Yet it is increasingly clear that patients have better experiences and outcomes and often use less healthcare resources when they understand more about their diseases and care, especially when they are empowered to and responsible for managing it. This book, Information Technology for Patient Empowerment in Healthcare, is about engaging patients and their families in new and innovative ways to support the management of their own health. This takes place in the context of the recent revolution in health information technology (HIT) – we have gone from living without HIT to using it routinely. However great our progress, the systems of today do not yet come close to leveraging the full potential of HIT to engage patients. In this book, leading figures in this area present their perspectives on how this can be accomplished. This will involve enabling patients and family members to participate actively in their care, to self-manage their medical problems and to improve communication with their healthcare providers by using patient-facing HIT tools. These tools, which range from personal health records to mobile applications, among many others, enable healthcare providers to partner with their patients, and together, build care regimens that optimize quality of care and health outcomes. They have the potential to help each of us – as we all eventually make this transition to being a patient – to have more control in an inherently vulnerable position, and to make better decisions about our health and well-being; about our destiny. We attempt to keep the patient and family experience and voice at the center of the conversation about HIT’s role in patient engagement and empowerment by weaving together the perspectives of multiple key stakeholders on the various tools and aspects of this emerging domain. Linked through the thread of facilitating patient empowerment, this book consists of contributions from patients, family members and patient advocates who discuss their experiences, expectations, satisfactions and frustrations with health care delivery; renowned clinicians, healthcare organization leaders and top industry managers who advocate for a new generation of patient-centered technologies; policymakers who play a central role in shaping healthcare reform; researchers from major universities around the world who propose and test cutting-edge technologies for patient care; and information technology experts who have crafted practical solutions to real patient needs. Together, these perspectives paint a picture of care that is being transformed by HIT, leading to greater patient engagement, empowerment and, ultimately, to improved quality of care and better clinical outcomes.

XIV 

 Introduction

This book tells its story in two parts. The first part of the book aims to provide a 360 degree perspective on information technology for patient empowerment by incorporating the diverse perspectives of patients and their family members, clinicians, researchers, healthcare organization leaders, HIT industry mangers and policymakers. The first five chapters of this book provide in-depth presentations of the transformation of patient and clinician behavior and interaction, evolution of patient-centered care and patient engagement, healthcare services activities, industry strategies, the role of the media and policy regulations in the information technology age. Thus, this part takes a critical view of the existing needs, challenges and opportunities for improving patient empowerment and engagement through HIT, mapping out what has been accomplished and what work remains to truly transform the care we deliver and engage patients in their care. This part provides the backdrop for the second half of the book. The remainder of the book covers the primary current and future information technology solutions for patient empowerment through specific HIT case studies. This section discusses information techniques to further personalize the access to healthcare data through patient portals; secure and scalable systems to support patients’ needs for ownership, control and sharing of health information; information retrieval and natural language processing methodologies to improve health literacy and education; electronic informed consent approaches for enhancing patients’ decision making processes; mobiles, kiosks and web-based technologies to support self-care management; and patient-reporting healthcare outcomes systems, among other functions. Each of these 10 chapters explores the existing gaps, strengths, weaknesses and potential implications of these practical solutions to real patient problems. This collection of timely case studies introduces the reader to a wide perspective on the direction patient empowerment technologies are headed. Patients have never previously been so empowered to shape their own health care. Patient engagement and empowerment via HIT appear destined to become of major value to both the public and healthcare organizations. Thus, in 10 years, the question may not be if and how to empower patients via HIT, but how we ever lived without these tools. We think this book is unique in its focus on the intersection between HIT and patient engagement and empowerment, and believe that it represents a timely and significant contribution to the literature in this field. We hope that it will be of value to policymakers, healthcare providers and administrators, consultants and industry managers, researchers and students and, not least, to patients and their family members.

List of contributing authors Mehnaz Adnan, PhD Research Assistant, National Institute for Health Innovation, University of Auckland, Auckland, New Zealand Wajeeh Bajwa, PhD Program Director, Regulatory Knowledge and Research Support, Clinical and Translational Science Institute, University of Florida, Gainesville, FL Marion Ball, EdD, FAHEMA, FHIMSS, FCHIME, FAAN, FACMI, FMLA Senior Advisor, Healthcare and Life Sciences Institute, IBM, Yorktown Heights, NY Professor Emerita, Division of Health Sciences and Informatics, School of Medicine, Johns Hopkins University, Baltimore, MD Sasha Ballen IT Analyst, Innovative Health Advisors LLC, Doylestown, PA David W. Bates, MD, MSc Chief Quality Officer, Senior Vice President and Chief Division of General Internal Medicine and Primary Care, Brigham and Women’s Hospital, Boston, MA Professor of Medicine, Harvard Medical School and Professor of Health Policy and Management, Harvard School of Public Health, Harvard University, Boston, MA Cynthia Baur, PhD Senior Advisor for Health Literacy, Office of the Associate Director for Communication, Centers for Disease Control and Prevention, Atlanta, GA Asaf Bitton, MD, MPH, FACP Assistant Professor of Medicine, Brigham and Women’s Hospital, Boston, MA Assistant Professor of Health Care Policy, Harvard Medical School, Boston, MA

J. Jeffrey Carr, MD, MSc Cornelius Vanderbilt Endowed Chair in Radiology and Radiological Sciences and Professor of Radiology for Biomedical Informatics and Cardiovascular Medicine, Vanderbilt University Medical Center, Vanderbilt University, Nashville, TN Scott Cashon Vice-President, Health Information Technology, CarePartners Plus LLC, Horsham, PA Catalina Danis, PhD Research Staff Member, Insights-driven Wellness Services Group, Health Informatics Department, IBM, Yorktown Heights, NY Mary Jo Deering, PhD President, Deering Health Associates, Bethesda, MD Robert Farrell, MS, MPhil. Research Staff Member, Social Computing Research, IBM, Yorktown Heights, NY Yaorong Ge, PhD Associate Professor, College of Computing and Informatics, University of North Carolina at Charlotte, Charlotte, NC Stephanie Glier, MPH Senior Policy Analyst, Pacific Business Group on Health, San Francisco, CA Maria Adela Grando, PhD Assistant Professor, Department of Biomedical Informatics, Arizona State University, Scottsdale, AZ Adjunct Assistant Professor of Medicine, College of Medicine, Mayo Clinic, Rochester, MN C. Martin Harris, MD Chief Information Officer, Chair, Enterprise Data Governance, Cleveland Clinic, Cleveland, OH Executive Director, eCleveland Clinic, Cleveland, OH

XVI 

 List of contributing authors

Jeff Harwell PhD Candidate, Center for Information Systems and Technology, Claremont Graduate University, Claremont, CA David Lansky, PhD President and Chief Executive Officer, Pacific Business Group on Health, San Francisco, CA Gene Lazuta Director, Strategic Communication, Secure Online Services, Cleveland Clinic, Cleveland, OH Gondy Leroy, PhD Associate Professor, President AIS SIG-Health, Management Information Systems Eller College of Management, University of Arizona, Tucson, AZ Peter J.F. Lucas, MD, PhD Professor, Institute for Computing and Information Sciences, Radboud University Nijmegen, Nijmegen, The Netherlands Ariane Marelli, MD, MPH, FRCPC, FACC, FAHA Professor of Medicine, McGill University, Montreal, Canada Founder and Director, McGill Adult Unit for Congenital Heart Disease Excellence and Associate Director, Academic Affairs and Research Cardiology, McGill University Health Centre, Montreal, Canada Mary McNamara, MLIS PhD Candidate, Department of Bioengineering, University of California Los Angeles, Los Angeles, CA Michael L. Millenson, BA President, Health Quality Advisors LLC, Highland Park, IL Adjunct Associate Professor of Medicine, Feinberg School of Medicine, Northwestern University, Chicago, IL Marjorie Miller, BS Ed President and Senior Consultant, HealthPower Associates Inc, Philadelphia, PA

Paula Miller, MSN, RN Member Outreach Manager, Adult Congenital Heart Association, Philadelphia, PA Martha Jean (Marty) Minniti, RN, BS Sr VP and Co-Founder, CarePartners Plus, Horsham, PA Jihad S. Obeid, MD Associate Professor and Associate Director Biomedical Informatics Program, South Carolina Clinical and Translational Research Institute, Medical University of South Carolina, Charleston, SC Disty Pearson, PA-C Physician Assistant, Boston Adult Congenital Heart and Pulmonary Hypertension Service Boston Children’s Hospital, Brigham and Women’s Hospital, Boston, MA Christopher Pentoney, MA Doctoral Student, Applied Cognitive Psychology, Claremont Graduate University, Claremont, CA Margaret Piscitelli, BSN, RN, CCRN Staff Registered Nurse, Department of Nursing Informatics, Jefferson School of Nursing, Thomas Jefferson University, Philadelphia, PA Michael Poku, MBA Medical Student, Vanderbilt School of Medicine, Vanderbilt University, Nashville, TN Ronen Rozenblum, PhD, MPH Founder and Director, Unit for Innovative Healthcare Practice and Technology and Director of Business Development, Center for Patient Safety Research and Practice, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA Jane Sarasohn-Kahn, MA, MHSA President, THINK-Health LLC, and Founder, Health Populi blog, Phoenixville, PA

List of contributing authors 

Yukari Schneider, PhD, MPH Research Affiliate, College of Journalism and Communications, University of Florida, Gainesville, FL Roy Schoenberg, MD, MPH CEO, American Well, Boston, MA Board of Directors, American Telemedicine Association, Washington, DC Hanna Suominen, Adj/Prof, PhD Senior Researcher, Machine Learning Research Group, NICTA, Canberra, Australia Adjunct Research Fellow, College of Engineering and Computer Science, The Australian National University, Canberra, Australia Professional Associate, Faculty of Health, University of Canberra, Canberra, Australia Maarten van der Heijden, PhD Postdoctoral Researcher, Institute for Computing and Information Sciences, Radboud University Nijmegen, Nijmegen, The Netherlands

 XVII

Marina Velikova, PhD Research Fellow, Embedded Systems Innovation by TNO, Eindhoven, The Netherlands Jim Warren, PhD Professor and Chair of Health Informatics, Department of Computer Science, University of Auckland, Auckland, New Zealand Eyal Zimlichman, MD, MSc Deputy Director and Chief Quality Officer, Sheba Medical Center, Ramat Gan, Israel Research Associate, Center for Patient Safety Research and Practice, Brigham and Women’s Hospital, Boston, MA

Acknowledgements We are grateful to the many people who contributed to this book. First and foremost, we thank each of the 41 authors who contributed their valuable expertise and experience. We also thank our colleagues, Anita Murcko, David Kaufman, Bret Anderson, Conny Morrison and Christine Powers, for helping us review the chapters and for their ongoing support with the book's preparation. This book was inspired by the work of various organizations and people alongside whom we have worked to promote patient-centered care and patient empowerment. We are grateful to the Mayo Clinic for their support of our research on the development of technologies for patients. Mayo Clinic founders’ mission and value statement, “the needs of the patient come first”, has inspired and will keep inspiring our efforts to build technologies to empower patients. We also thank Brigham and Women’s Hospital and Partners Healthcare, which have long been international leaders in HIT, and pioneers in patient computing. We thank the Gordon and Betty Moore Foundation for their support and funding, which has contributed significantly towards developing innovative health information technologies that are promoting patient engagement now at Brigham and Women’s Hospital, as well as the other hospitals in the Libretto Consortium. Additionally, we would like to thank the Adult Congenital Heart Association for being a strong voice for patients and for their work towards increasing patient-centered care over the last few years. Their work has inspired us. We also thank the Patient-Centered Outcomes Research Institute (PCORI), the Agency for Healthcare Research and Quality (AHRQ), the Commonwealth Fund, the International Society of Quality in Health Care (ISQua) and the American Medical Informatics Association (AMIA) for sponsoring research, facilitating its discussion, and for their thought leadership in this area. Finally, we would like to dedicate this book to the patients whose experience we work so tirelessly to improve. It is a role that we all will one day assume and it is with that acknowledgement that we also thank all the individual people, patient advocates, professionals and organizations which promote patient-centered care, patient empowerment and patient engagement. Their commitment to enhancing these key domains of quality of care will make for a brighter future for us all. Maria Adela Grando Ronen Rozenblum David W. Bates

Part I: Information Technology for Patient Empowerment: 360° Perspectives

Ronen Rozenblum, Paula Miller, Disty Pearson and Ariane Marelli

1 Patient-centered healthcare, patient engagement and health information technology: the perfect storm An integrated perspective from patients, families, clinicians and researchers Abstract: This chapter will provide the backdrop for a volume that brings together a range of viewpoints on health information technology and patient engagement. Though few in the field would argue that engaging patients in their care is not important, few have written about patient engagement with patients, instead of about them. In this chapter, we – patients, family members, clinicians and researchers – have intentionally worked together to integrate our perspectives. By way of background, we define information technology as the term was originally coined; we listen to a narrative from the perspective of a family member and we look at the evolution of the traditional clinical encounter into the modern era. In the second part of this chapter, we characterize patient-centered care and patient engagement in terms of a definition of concepts and describe its current status. Part three focuses on patient-facing health information technology tools that enable patient engagement and empowerment, and their potential effect on healthcare outcomes. We end our chapter with remarks on the opportune confluence of factors bringing us to this point in time. Finally, we go back to the patient’s voice to illustrate the power of social media and imagine how social networks may shape the future of our field. We begin and end with the patient and family voice through personal narratives drawn from the authors in this chapter. This underscores the personal nature of the content of this book for us as people as we interact with the healthcare system and each of us becomes a part of this conversation at some point in our lives. We hope our readership will be as excited as we are to be at the crossroads where we find ourselves: patient-centered healthcare, patient engagement and health information technology – the perfect storm.

1.1 Background In 1958, in the Harvard Business Review, the term “information technology” was coined in a prophetic piece entitled “Management in the 1980s”. Referring to a new technology beginning to take hold in American business, Leavitt and Thomas wrote:

4   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

The new technology does not yet have a single established name. We shall call it information technology (IT). It is composed of several related parts. One includes techniques for processing large amounts of information rapidly and it is epitomized by the high-speed computer. A second part centres around the application of statistical and mathematical methods to decision-making problems; it is represented by techniques like mathematical programming and by methodologies like operations research.[1]

The origin of the word information, dating back to the late 14th century, relates to the “act of informing” from Old French informacion, enformacion "information, advice, instruction", and from Latin informationem (nominative informatio) with the meaning of "knowledge communicated". Almost fifty years after ‘information technology’ came into being, the term ‘health information technology’ (HIT) was defined. HIT refers to “the application of information processing, involving both computer hardware and software that deals with the storage, retrieval, sharing, and use of healthcare information, data, and knowledge for communication and decision making”.[2] Since then, HIT has made its mark on efforts to improve quality of care. As David Bates, a pioneer in the field of HIT wrote in 2002, “excellent information technology and high-quality health care are closely linked. All of the healthcare organizations […] are recognized as quality leaders due to excellent clinical outcomes, which have been achieved in part because of their information systems”.[3] The ability to become informed or to inform is one of the most significant tools empowering human evolution. That empowerment has been magnified in light of the information technology revolution that has occurred over the last 50  years. In the healthcare space, its recent impact has been magnanimous. A personal narrative helps to illustrate this profound progression: A family member’s personal narrative: It was 1984. My world changed forever when our first son was born, but not at all in the way I had expected. He was eight hours old when he turned very blue and was moved to the intensive care unit. The wonder of his birth turned into terror and fear for his survival. He had a severe heart defect. They started him on a new medicine that would keep him alive temporarily, but he stopped breathing and was put on a respirator. When I heard his nurse report to the receptionist that I could not see him yet, as he had suffered a “respiratory arrest”, I held onto the receptionist’s desk as the world turned black; I almost went down. They let us hold him before he went to surgery. My husband and I held hands and held onto him in a mixture of love for our beautiful baby we had waited the last nine months for and the horror that we might never hold him again. We sought to understand. We searched. We went to the medical library where we poured through volumes of printed material. We could find only one article about my son’s “defect”. The article began: “These poor unfortunate infants… all died”. I stopped reading. Caregivers were wonderful and provided us with a robust support system, but there were no other resources to turn to.

1 Patient, family, clinician and researcher perspectives   

   5

We were powerless, with nowhere to turn for more information and no way of understanding what my son’s heart defect would mean for his future and our life. We felt alone and terrified. It was 2011. Twenty-seven years later, my husband was diagnosed with cancer. Once again our world stopped. But it was a different world. We searched, but this time we found. We found information. Using the web, we had access to medical papers about his cancer, the natural history of his disease, the risks and benefits of the therapeutic options, expert opinions and consensus statements. We were able to research providers. We found the internationally recognized experts and centers for his cancer. Though the terror of the diagnosis was present, we were empowered by online resources that were offered to us by the providers, enabling us to view personal health information and other educational data. My husband presented to his surgical consult with academic papers in hand and debated the pros and cons of the varied approaches with the surgeon. He was able to understand the options and discuss them with his care providers with a common understanding, an equal footing and a sense of truly shared decision-making and direction of his own care. Between these two time points of experience as a patient’s family member, three decades expired and we witnessed the birth of the health information technology revolution. Access to online information provided us with a level of unprecedented control. Health information technology completely changed our experience and gave us a level of comfort that was not possible when our son was born.

This narrative relates a family member’s journey between two clinical encounters twenty-seven years apart; the first with the birth of a child with severe heart disease and the second with a spouse who had cancer. The difference between experiences is the result of immense progress in HIT. In parallel, the narrative reveals a fascinating, natural progression of language from ‘they’ (healthcare providers) to ‘we’ (patient, family and provider). During the time that spanned the two poles of the experience described, the author articulates the transition towards becoming part of the healthcare system and, in some way, becoming part of the solution. Access to information through the use of information technology was the catalyst to that empowerment. This ultimate transition from ‘they’ to ‘we’ is the clearest illustration of patient engagement and empowerment. We now travel back in time to understand the anatomy of the complex relationship between the clinician (then, most commonly a physician) and the patient. We will demonstrate why the ‘they’ vs. ‘we’ culture has been so embedded in the medical profession’s framework.

1.1.1 Evolution of the traditional clinical encounter into the modern era Traditionally, the clinical encounter begins when a person seeks council for an ailment. Fundamental to this encounter is the patient narrative, during which the clinician seeks to take a ‘history’. In this sense, the degree to which a clinical encounter has been ‘patient-centered’ is fairly consistent over the last 250 years.

6   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

The scientific method took root during the 17th century’s “Age of Enlightenment” when reason began to dominate Western civilization. Until that time, ‘sickness’ was the subject of a person. In the early 1800s, however, the ‘sick man’ disappeared, replaced with a concept of ‘the patient’, at least in part due to the emergence of hospital medicine.[4] The subject or the person thus became the object of his/her ailment. [5] In the latter half of the 19th century, commentary on the role of the patient history in the process of clinical assessment was always centered on the clinician. As illustrated in a quote written by Samuel Fenwick, a prominent physician in the 1800s,[5] a shift occurred from what the patient thought or felt to what the doctor found: It is generally difficult for the young student to guide the patient’s account in such a way as to derive the necessary information from the details. Most persons ramble in describing their symptoms and many insist on giving their own or other persons’ opinions as to the nature of their disease, instead of confining themselves to the narration of facts.[6]

Thus the patient’s narrative provided a source of facts to be uncovered by the skilled physician who took control of the history.[5] While the patient sought comfort, the doctor sought information. Clinicians of the time believed that the “patient needs the physician in a way that the physician does not need the patient”.[7] Thus, the clear difference in what the doctor and patient expected from the clinical encounter accounted for, and to some extent continues to account for, the tension that prevails in this complex relationship. This difference in expectation and the growing gap in expertise set the stage for the power differential and paternalism that grew over the 19th century. Medical advances were changing the practice of medicine. Early medical technology became more widespread during the first half of the 20th century, with the invention of tools such as the stethoscope and the X-ray machine. This technology allowed the clinician to gain physical proximity to the inside of a patient’s body, widening the knowledge gap, and the power differential, between the patient and clinician. Despite its strong roots in traditional medical culture, the clinical encounter was not immune to the rapidly evolving social sciences of the 20th century. The once simple, one-dimensional clinical encounter was bound to rapidly evolve. Social scientists were learning how to question the public about its opinion during the l950s.[8] By the 1970s, behavioral sciences began highlighting the importance of interpersonal skills, sensitizing clinicians to the importance of their behavior relative to the needs of patients.[8] Communication sciences evolved in the 1980s against the backdrop of the information technology making medical advice freely available for the lay person.[8] The ‘patient’ who had once been a simple ‘sick man’ was making way for the appearance of the ‘consumer’ or ‘client’, reflecting the growing value of consumerism in a society increasingly interested in service delivery.[9] Thus, the social, behavioral and communication sciences converged onto the clinical encounter, empowering the patient with new tools by which to evaluate the clinical encounter. The notion that the

1 Patient, family, clinician and researcher perspectives   

   7

patient could and should expect to be satisfied has more recently become common, thereby legitimizing patient satisfaction as an increasingly important performance measurement in the delivery of healthcare and a dimension of quality. The evolution of the modern clinical encounter, as illustrated in the narrative above, can only be fully understood in the context of parallel progress in HIT. Yet the complexity of the clinician-patient encounter in the growing e-health landscape persists. For most clinicians, medicine remains a humanistic science that relies heavily on the clinician’s ability to establish good communication with the patient. This is corroborated by common clinical experience and studies that support the notion that effective communication between clinician and patient results in an increased adherence to therapy and enhanced overall outcomes.[10] As such, clinicians commonly believe that the interposition of HIT tools between themselves and the patient will impair this rapport. Indeed, focus groups have revealed that physicians who were asked to introduce HIT into their practices objected to using tools that minimized the physician-patient relationship.[11] However, studies also reveal that physicians have a poor understanding of patients’ health-related values and beliefs. For example, studies have shown a significant difference between what physicians perceive to be important for the patient and what patients value in terms of treatment options, highlighting the need for partnership and shared development of a care plan.[12, 13] At the same time, as exemplified in the earlier personal narrative, the contemporary patient often comes to the physician bringing information obtained online prior to the consultation. Although clinicians may be favorable to discussing the patient’s use of health related internet information,[14] studies indicate that the patient’s use of this type of information during the clinical encounter often requires clarification and interpretation by physicians to avoid misleading conclusions.[15] Although the modern clinical encounter has undergone a major paradigm shift from a paternalistic clinician-centered interaction to a more symmetrical patient-centric partnership, further enhancement of the clinician’s acceptance of the important role of the patient will be critical to facilitating patient empowerment. While it may seem paradoxical, it is quite possible that medicine, even as a humanistic science, will be helped by HIT and patient empowerment. The power to record and transmit information in real time should help the clinician get closer to the patient’s core by acquiring more detailed, personal and relevant information. In addition, the patient and family narrative that remains the cornerstone of this rapport can now be expressed over a variety of complementary technology platforms. These tools can help to bridge the gaps in communication and understanding between clinicians and patients, allowing clinicians to provide patient-centered care and enhance patient engagement.

8   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

1.2 Patient-centered care and patient engagement Patient-centered care (PCC) and patient engagement have become central components of the modern clinical encounter. Below we define PCC and patient engagement, review the current trends and discuss why these have become key dimensions of healthcare quality. We describe policy-based initiatives and examine why gaps in performance still prevail. Finally, we discuss determinants of successful implementation of PCC and patient engagement, as well as challenges that healthcare organizations are still facing today.

1.2.1 Definition of terms and concepts Although there are various proposed definitions of PCC that encompass the core concepts, none are globally accepted. Nevertheless, many current definitions highlight the central role that patients should play in healthcare delivery. The Institute of Medicine defines PCC as “care that is respectful and responsive to individual patient preferences, needs and values, and ensuring that patient values guide all clinical decisions”.[16] A number of frameworks describing the key dimensions of patientcentered healthcare have emerged. In 1993, the Picker Institute identified eight dimensions of PCC: respect for patients' preferences and values; emotional support; physical comfort; information, communication and education; continuity and transition; care coordination; involvement of family and friends; and access to care.[17] In addition, based on a thorough review of nine models of PCC, Carol Cronin identified six core concepts that appeared in multiple definitions of PCC: education and shared knowledge; involvement of family and friends; collaboration and team management; sensitivity to nonmedical and spiritual dimensions; respect for patient needs and preferences; and the free flow and accessibility of information.[18] Although most models of PCC place an emphasis on the importance of the patient's and family's engagement in their care, there is a lack of consistency in terminology related to the concept of patient and family engagement.[19] Some definitions of patient engagement focus on individuals’ behavior relative to their health care and some on the relationship between patients and healthcare providers.[20] The Center for Advancing Health defines engagement as “actions individuals must take to obtain the greatest benefit from the healthcare services available to them”.[21] Angela Coulter defines patient engagement as a set of reciprocal tasks by both individuals and healthcare workers who “work together to promote and support active patient and public involvement in health and health care and to strengthen their influence on healthcare decisions, at both the individual and collective levels”.[22] Most recently, a group of researchers from four institutions in the U.S. collaboratively defined patient and family engagement as “an active partnership between health professionals and patients and families working at every level of the healthcare system to improve

1 Patient, family, clinician and researcher perspectives   

   9

health and the quality, safety, and delivery of healthcare. Arenas for such engagement include but are not limited to participation in direct care, communication of patient values and goals, and transformation of care processes to promote and protect individual respect and dignity”.[23] Furthermore, according to them, “Patient and family engagement comprises five core concepts: collaboration; respect and dignity; activation and participation; information sharing; and decision making”.[23] While this definition was designed for the intensive care unit environment, the specific domains identified may also be applicable to other clinical settings. The term “patient engagement” is sometimes used interchangeably with “patient empowerment”, “patient partnership” and “patient activation”. While the terms are connected, they are not synonymous and may occur separately. Patient empowerment is usually related to capacity. Patients may feel empowered or take action to promote health, which may lead to engagement, or they may become empowered as a result of engagement with providers. Patient activation is defined as “understanding one's own role in the care process and having the knowledge, skills and confidence to take on that role”.[24] In this chapter, we use the term “patient engagement” to denote a broader concept that includes patient empowerment, patient partnership and patient activation. In attempting to capture both the continuum and levels of engagement from the patient and family perspective, Carman and colleagues propose a multidimensional framework which starts from the clinical encounter and incorporates healthcare structure and policy along a continuum from consultation to partnership and shared leadership.[20] In sum, despite the variation in definitions of PCC, common to all is the focus on the importance of incorporating patients’ needs and perspectives into care delivery. Based on the literature, six core concepts of PCC emerge consistently including: education and shared knowledge; involvement of family and friends; collaboration and team management; sensitivity to nonmedical and spiritual dimensions; respect for patient needs and preferences; and free flow and accessibility of information. Finally, although patients’ engagement with their care is now considered a key dimension of PCC, further work is needed to standardize terminology related to the concept of patient and family engagement.

1.2.2 Why now and why does it matter? Patient-centered care and patient engagement have drawn increasing interest in recent years, highlighting the shifting roles of patients and families in healthcare as they become more active and empowered. One of the most important ways in which patient-centered healthcare gained explicit recognition was in a 2002 Institute of Medicine report, "Crossing the Quality Chasm: A New Health System for the 21st Century", in which patient centeredness was considered a dimension of quality.[16] Triggered in part by this report, PCC and patient engagement have become an increas-

10   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

ing priority on the national agenda in the United States and play a prominent role in the Patient Protection and Affordable Care Act.[16, 25, 26] As a result, meaningful collaboration with patients and families and their active participation in the care process and decision-making are now considered key elements of healthcare quality and delivery performance. Motivated further by public reports on patient experience, many healthcare organizations have strived to become more patient-oriented and use patient surveys to assess their progress.[25–27] Organizations have implemented various performance measures related to PCC. The United States and the United Kingdom, as well as other countries, are using patient experience survey programs to systematically collect patient experience feedback at a national level. The “Hospital Consumer Assessment of Healthcare Providers and Systems” (HCAHPS) is now nationally required by all providers treating Medicaid and Medicare patients in the United States.[28] The intent of the HCAHPS initiative was to provide a standardized survey instrument and data collection methodology for measuring patients’ perspectives on hospital care. The results are available to the public, which allows valid comparisons to be made across all hospitals. The current requirement to publicly report scores on HCAHPS ties the amount of reimbursement directly to level of service performance. Similarly, the National Health Service (NHS) in the United Kingdom publishes information on the specialist options available and performance indicators from national survey data for each hospital. This allows patients to rate and comment on their healthcare providers online, and based on this information, to choose which provider they wish to access. The above policies and reporting requirements, as well as the emerging trend of consumerism in healthcare, have spurred a groundswell of activity around creating partnerships with patients and families and engaging them in the care process in the United States and around the world. Part of the impetus for supporting PCC and patient engagement initiatives is the growing body of evidence that these can lead to greater patient satisfaction, improved clinical outcomes (both as reported by patients and measured more traditionally), health service efficiency and improved health-related business metrics.[20, 29–36] Specifically, studies have indicated that patient engagement and shared decisionmaking leads to improvement in self-management and treatment adherence.[24, 34, 37–44] Other benefits associated with patient engagement have been demonstrated in more efficient health services utilization rates of diagnostic tests, referrals, emergency department visits and hospital attendance.[43, 44] Consistent with this notion, higher levels of patient activation have been shown to be correlated with lower predicted per capita cost ratios.[34] In sum, it appears that PCC and increased patient engagement are associated with positive health outcomes, as well as reduced health services utilization rates and costs. Together, these domains have become a priority in many countries around the world, where they play an increasingly prominent role in the recent healthcare

1 Patient, family, clinician and researcher perspectives   

   11

reforms. In turn, this has led to the development of interventions and surveys to improve and assess PCC in healthcare organizations.

1.2.3 Where are we now? In recent years, numerous regulators and organizations from various countries have incorporated these domains into national documents and frameworks. Over the last decade, healthcare organizations have initiated strategies to enhance PCC and engage patients in the care process. These include staff development, proactive leadership, collecting and reporting patient feedback, redesigning and co-designing service delivery, creating Patient and Family Advisory Councils (PFAC) and implementing patients’ rights charters. National and international organizations have also provided frameworks and tools to support adoption of these strategies. These include the Institute of Medicine (IOM), the Center for Medicare & Medicaid Services (CMS), the Joint Commission on Accreditation of Healthcare Organizations ( JCAHO), the National Committee for Quality Assurance (NCQA), the Agency for Healthcare Quality and Research (AHRQ), the World Health Organization (WHO), the Institute for Healthcare Improvement (IHI), the Institute for Patient- and Family-Centered Care, Planetree, the Picker Institute and the National Health Services (NHS) in the United Kingdom. Yet, despite these expanding initiatives, many healthcare organizations have faced barriers when attempting to transform their organizational culture from provider-focused to patient-centric and still fall short of achieving high scores on patient experience.[25, 45, 46] According to the Center for Studying Health System Change, levels of patient engagement vary considerably in the United States, with less than half of the adult patient population achieving high levels of activation.[42] Numerous factors may modify the advancement of PCC and patient engagement initiatives. Studies have shown that healthcare organizations successfully incorporated PCC as a strategic investment priority mainly through committed leadership, active measurement and feedback on patient experience, and engagement of patients and staff.[47] A contributing factor to the current gaps in performance of health organizations is insufficient institutional support for clinicians to manage and improve these dimensions of healthcare quality. Patients’ perceptions of the care they received has been shown to be mainly influenced by their interactions with healthcare providers, particularly nurses and physicians.[48, 49] Thus, achieving successful PCC and high levels of patient engagement requires frontline clinicians to be engaged in this cultural transformation. But how do health organizations enhance provider engagement? In a multi-center study during which more than 1,000 physicians and nurses at four academic hospitals in Denmark, Israel, the United Kingdom and the United States were surveyed, insufficient frontline clinician engagement and leadership support for PCC improvement initiatives was observed.[45] Only 9% of clinicians stated that their department had

12   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

a structured plan to promote improvement of patient experience. In addition, only one-third recalled having received feedback from their hospital management regarding patient experience status. It was also discovered that, while nearly all clinicians believed that improving patient experience during hospitalization was achievable, only 38% remembered targeted actions that were conducted to improve it. Thus, thirteen years after the Institute of Medicine's Quality Chasm report called for the fundamental improvement and redesign in PCC, these findings raise concerns as to whether today’s clinicians have an active role or are being engaged by hospital management in enhancing PCC. Together with consistent results from a survey conducted in congenital heart disease outpatient clinics, these data suggest that healthcare organizations should take a more active role in developing approaches to engage clinicians in the PCC improvement process and ensure routine feedback about patient experience is systematically provided.[50] Increasingly, patient experience and patient engagement should become a strategic investment priority in order to shift healthcare organizations towards a culture of PCC. In sum, numerous healthcare organizations have recently initiated strategies to enhance PCC and engage patients in the care process. Nonetheless, barriers have been identified in attempting to transform culture at the organizational level from provider-focused to patient-centric. As a result, initial survey scores on patient experience and other PCC metrics have been suboptimal. Importantly, increased institutional support for clinicians to manage and improve these dimensions of healthcare quality will help contribute to the improvement of performance of, and consequent reimbursement to, health organizations.

1.3 Health information technology to engage patients Communication access points in healthcare have transitioned away from one-onone interpersonal interaction to a wide array of technological platforms that enable a more complex exchange of information. This change promises to be transformative. The growing demand for improved and more efficient communication between healthcare providers and patients has created an impetus to harness HIT and consumer e-health tools to promote patient engagement and empowerment. These tools have the potential to transform care into an active collaboration between providers and patients, with the expectation that this change will lead to improved standards of care. Examples of promising patient-facing technology include personal health records, patient portals, mobile health technologies, personal monitoring systems, patient-provider secure e-mail messaging, internet-based resources for health information, education and consultation, and social media networking websites. We now examine the evidence base available to analyze the impact of patientfacing HIT tools on outcomes. We then discuss the trends aimed at promoting the use

1 Patient, family, clinician and researcher perspectives   

   13

of HIT for patient engagement and the challenges of wide-scale implementation and uptake of HIT.

1.3.1 The impact of patient-facing HIT tools on health outcomes The importance of patient-facing HIT tools stems from one significant challenge: to improve quality by enabling patients to take a more active role in their care. Some tools give patients the opportunity to be more responsible for their care by providing them with the ability to access health information, choose healthcare providers and manage their health care. Other tools allow patients to communicate directly with their care team, coordinate care across caregivers and interact with other patients with similar health conditions, creating a broader and more connected healthcare network. Engaging patients in new ways has been shown to directly impact patient behavior that promotes positive health outcomes, patient satisfaction, care delivery efficiency and improved quality of care.[51–61] For example, a recent literature review found that access to health records appeared to enhance patients’ perceptions of care plan control without increasing, and possibly even reducing, patient anxiety.[62] In other studies, access to patient health records have been shown to increase use of preventive services, enhance patient-provider communication and improve self-management of clinical outcomes, including blood pressure and control of blood glucose levels.[51, 53, 55, 57, 59, 60, 63–65] HIT tools also have the potential to improve patient safety through increased surveillance of medication and improved management of chronic disease. Finally, HIT systems can also help to reduce costs, for example by avoiding test duplication.[66] These examples illustrate the promise of patient-facing HIT tools to shift the balance of power and responsibility from healthcare providers towards a partnership between providers and patients. Despite the benefits of patient-facing technologies, quantifying and understanding the impact of these technologies on health outcomes require further study. Several studies to date have shown mixed results on the effect of information technology on quality of care.[67, 68] A systematic literature review that explored the impact of HIT tools on patient satisfaction found that while there is some evidence that HIT improves patient satisfaction, interventional studies, in particular randomized control trials, were not consistent in these findings.[69] Another systematic literature review found that the effect of patients' access to their medical records on measures of the IOM quality domains showed mixed results for effectiveness, patient-centeredness and efficiency, and were understudied for patient safety, timeliness and equity.[62] This marked inconsistency in the literature indicates that further study is needed to achieve a thorough understanding of the impact of HIT and the factors that mediate it. Of the patient-facing HIT tools mentioned above, social media has a significant potential to truly engage patients in their care, as this technology is driven by ‘user-

14   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

generated’ content.[70] Patients use social media to share their opinions or rate a provider, treatment or healthcare policy.[71] Social media can also serve as a new means of communication between provider and patient, a place for patients with similar conditions to congregate and a dissemination point for general health information by healthcare organizations. On a macro level, social media gives researchers and policy leaders access to large amounts of data generated directly from the consumer. For example, researchers have used data from Twitter to accurately track influenza outbreaks, as compared to traditionally reported disease levels. While there have been a handful of relevant studies, further research will need to be conducted in order to evaluate the impact of these tools on levels of patient engagement and healthcare quality, as well as the accuracy and usage of user-generated ratings and reports. In sum, patient-facing HIT tools have already begun to make their mark on quality improvement initiatives of healthcare providers. Despite some mixed results, data are increasingly emerging to support the role of HIT in improving PCC outcomes delivery, health services efficiency and health outcomes. While positive associations between HIT and improved health outcomes are promising, it remains to be determined specifically which technologies will make the most significant contributions to quality improvement.

1.3.2 Current trends and challenges In step with the early success of these technologies, a consensus is emerging: engagement of patients with their own health care through HIT is essential to healthcare reform in the United States. The National Coordinator for Health Information Technology has identified the adoption and meaningful use of HIT by healthcare providers and patients as a key factor in improving the nation's health system.[52] Consistent with this notion, incentive programs have been created to promote consumer access to health data. Perhaps the most important ones, Medicare and Medicaid Electronic Health Record (EHR) Incentive Programs, give financial inducements to health professionals and hospitals to encourage “meaningful use” of EHRs in order to improve patient care. Policy makers, with the support of patient advocates, have successfully argued that adoption of patient portals should be part of meaningful use-related policies.[72] Thus, as part of the staged mandatory implementation of HIT in healthcare delivery, providers complying with meaningful use guidelines need to offer patients the ability to view, download and transmit their health information online. Motivated by such policy initiatives to accelerate the adoption and meaningful use of HIT, healthcare organizations have begun to implement, use and promote e-health tools. [66, 73] Typically, the strategies they use to encourage uptake are multi-faceted: both directed at raising consumer awareness of availability and gaining provider acceptance via easing the change processes necessary to embed e-health tools in usual practice.[74] However, while federal level initiatives and healthcare organizations

1 Patient, family, clinician and researcher perspectives   

   15

activities are considered necessary to increase the availability of HIT, there is little data to support the notion that they increase patients’ uptake of HIT and e-health tools. The limited data available suggest that ongoing patient usage rates of HIT are low.[66, 75] Moreover, it remains unclear whether patients provided with access to more information are able to make optimal healthcare choices. Patients may lack the necessary comprehension that is required to inform and drive good health choices. The uptake of HIT will depend on the complex balance between the ability of patients to adapt to new models of healthcare delivery, and the healthcare system’s ability to adapt to changing patient needs – both of which are, and will be, increasingly driven by market forces and policy. Challenges related to HIT adoption can be categorized into factors related to patients, clinicians and healthcare systems. From the patient perspective, numerous factors impact the ability to use HIT tools. Specifically, lack of HIT awareness, limited health literacy, lower socio-economic status, older age, inadequate computer skills and unmet technical support needs have been identified as negatively affecting patients’ use of HIT.[75] From the clinicians’ perspective, although there is an increasing use of HIT to engage patients, barriers to adoption persist. These result from a wide variation in attitudes towards HIT, which can include a perceived decrease in efficiency and concerns over clinicians' own ability to adapt to HIT infrastructure.[76] On a macro level, the healthcare setting itself is evolving into a multi-faceted ecosystem in which traditional hospitals or clinics are becoming an increasingly small part. The growing number of electronic platforms providing parallel, complimentary or alternative ways to access health care will lead to a fluctuation of boundaries between traditional and less traditional clinical infrastructures. This may force the accelerated use of HIT tools, such as telehealth, to enable a choice of communication platforms between providers and patients. In sum, trends in policy, influenced by policy makers and advocacy groups, have taken the lead in establishing measures to promote the use of HIT. Specifically, policies related to meaningful use of HIT were implemented in order to accelerate adoption. Challenges to HIT adoption are numerous, ranging from factors related to patients, clinicians and healthcare organizations. Limited access to HIT remains a key determinant of ongoing low adoption rates. Nonetheless, it is evident that engaging patients with HIT will play an essential role in the reform of our healthcare system. HIT tools that enable patient engagement will continue to grow in importance as their potential is further understood and harnessed.

1.4 The perfect storm – transitioning to the future In this last section, we demonstrate that we are at a crossroads of healthcare trends, where the increasing interests in PCC, patient engagement and HIT create “the perfect storm”. We then go back to the patient and listen to the voice of a second narrative

16   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

to steer us in the direction of the future. In the future, we imagine how evolutionary dynamics may shape the growth of healthcare networks resulting from PCC, patient engagement and HIT, influenced by social media in the evolving landscape.

1.4.1 The “perfect storm” Over time, patients have become central to the process of healthcare delivery. This shift is beginning to have a major impact across a range of dimensions of healthcare, including healthcare quality. One of the most important criteria of quality of care today is related to the ability to engage patients and families in the care process and create a culture of patient-centered healthcare: a culture where patients and healthcare providers collaborate as a team, share knowledge and work towards common goals; a culture where clinicians perceive the patients as equal partners; a culture where healthcare providers incorporate patient and family voices into care delivery. At the same time, patient-facing HIT tools, such as personal health records and mobile health technologies (applications for mobile phones and tablets), have created new opportunities for patients and families to participate actively in their care, to selfmanage their medical problems and to improve communication with their healthcare providers. The domains of patient-centered healthcare, patient engagement and HIT have come together in a singular and timely way, the consequences of which will be powerful on quality of care. These circumstances have created a major shift in how patients and healthcare providers connect – in effect, the “perfect storm”.

1.4.2 Transitioning from the 21st century to the future A patient’s personal narrative: Born with a complex heart defect in 1951, I grew up fraught with anxiety, fear, loneliness and sadness. Particularly difficult were the unknown factors. What would happen to me? Would I need more surgery? Answers to questions from different healthcare providers I encountered were the same – “I don’t know” or “there’s no one out there who knows how to take care of you”. Pediatric heart specialists didn’t want to see me because I was “too old” and general cardiologists didn’t have the answers but often pretended that they did. Before the advent of the Internet, there were few, if any, ways to research the disease I had grown up with and no opportunities to find a medical provider who knew my disease. Public libraries had no relevant books and medical libraries were not accessible, even if they were to have the information. Imagine my surprise upon first browsing the internet to find that there were, in fact, doctors who were beginning to recognize that we adults, growing up with a pediatric onset chronic disease, needed special care. I was amazed to find that one of these doctors had been practicing since 1973. No one I encountered was aware of this.

1 Patient, family, clinician and researcher perspectives   

   17

In the late 1990s, approaching my late forties with failing cardiac health, I was certain that life as I knew it was coming to an early end. I believed that I would never see my children grow up and that I would certainly become disabled and nonproductive. These were not options I was willing to accept. Not only was I a mother, but I was also a well-educated professional with an active career. But how was I to gain the knowledge that I needed to change the trajectory of my health? In 1999, at the age of 48, I first found the courage to talk about my heart defect and then to reach out to others who might be able to help. For me, the world wide web and an ‘AOL’ chat room were not only game changers but life changers. There I virtually met others, like myself, who had lived with heart defects, wanted to know more and came together because they did not believe they were the only people like themselves in the world. It was this one simple encounter in cyberspace that allowed me to begin taking the steps necessary to change my health and my life. I hold steadfast to my belief that the internet, through a chat room, empowered me to advocate for myself and to become a partner in the delivery of my health care.

This personal narrative exemplifies the profound and personal impact that HIT can have on engagement and empowerment to manage one’s disease state and harness the perspective of many similar patients in the pursuit of health and well-being. It is only from the patient’s perspective that the unique needs are best understood and strategies to improve outcomes are best achieved. But historically, as we have shown, the information balance about disease states and challenges faced by patients leaned heavily towards providers, making patient-centered care an afterthought. In recent decades, however, the internet has taken the first steps towards redressing this disconnect. Initially, the principle use of the internet by patients may have been simply to seek health information. However, this personal narrative makes clear that the reference information function of the internet was just the beginning. It is increasingly evident that social media represents a compelling force. In addition to the rapidly increasing number of social media tools, the sharp rise in the population using them have aggregated to magnify their potential impact. It seems that social media will have the power to dramatically transform our healthcare culture and systems by introducing a new level of patient engagement.[70] In imagining the future of healthcare, a pertinent question to ask is whether we can predict the growth and impact of social media networks in the delivery of care. Martin Nowak, an evolutionary biologist from Harvard University, in a book entitled Evolutionary Dynamics  – Exploring the Equations of Life, states that, “Where ever information reproduces, there is evolution”.[77] In this book, mathematical biology used to study Darwinian evolution and traditionally applied to a variety of biological networks is applied to networks that harbor information. Hence the concept emerges that ideas reproduce themselves within networks of information. Of particular interest is Erez Lieberman’s[78] application of evolutionary dynamics to social media networks.[79] The power of social media to impact change relates to three dimensions of networks which can be modeled and quantified. The first dimension lies in the fact that networks harbor and incubate ‘replicator’ ideas. Applied to the healthcare

18   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

space, modeled ideas related to emerging HIT tools will replicate themselves and multiply. The second is in a phenomenon called ‘network reciprocity’ where the actual network modifies the behavior of its members. This means that as the internet evolves and social media becomes embedded in our healthcare system, patients and clinicians will themselves undergo transformative changes that will shape the evolution of healthcare in real time. The aforementioned personal narrative makes clear that making relevant information available to patients can have a significant impact on engagement and empowerment of their health status, making adherence to care plans and better outcomes more promising. Lastly, social networks amplify the capacity to influence and cooperate with one another. Much like the patient activation illustrated in the narrative, social media networks can enable clinical encounters not only between clinicians and patients but with other providers, family members and organizations to optimize patient support systems. If we apply evolutionary theory to the dynamics of social media networks and their aggregated potential impact on patients and patient empowerment, the results may be astounding. Ultimately, evolutionary dynamics reflect the emergence and refinement of ideas that impact major populations. This process has begun to occur in healthcare and will continue to transform it. The possibilities to positively impact patient engagement and empowerment via HIT are endless. Now is the best time to leverage healthcare information technology networks and engage patients in their care in an effort to promote a culture of PCC and ultimately to improve the value of care delivered. It takes a perfect storm to optimize the confluence of change, opportunity and growth. We believe that the perfect storm is here. This is the beginning of the future. “The lights went dim, and the glass cubicle was no longer empty. A figure occupied it in a wheelchair. It said, “I am Hari Seldon”. The voice was old and soft. “It is fifty years now since the Foundation was established – fifty years in which you have been ignorant of what you were working toward. It was necessary that you be ignorant, but now the necessity is gone.” Isaac Asimov, Foundation Trilogy, 1951

References [1] [2]

Leavitt HJ, Whisler TL. Management in the 1980's. Harvard Business Review. 1958;36(6):41–8. Brailer D, Thompson T. Health IT strategic framework. Washington, DC: Department of Health and Human Services; 2004. [3] Bates DW. The quality case for information technology in healthcare. BMC Med Inform Decis Mak 2002;2:7. [4] Jewson ND. The Disappearance of the Sick-Man from Medical Cosmology, 1770–1870. Sociology 1976;10:225–244. [5] Gillis J. The history of the patient history since 1850. Bull Hist Med. 2006;80(3):490–512. [6] Lawrence C. The meaning of histories. Essay review. Bull Hist Med. 1992;66(4):638–645.

1 Patient, family, clinician and researcher perspectives   

[7]

[8] [9] [10] [11] [12]

[13]

[14]

[15]

[16] [17] [18] [19] [20]

[21]

[22] [23] [24]

[25] [26] [27]

   19

Schleifer R, Vannatta J. The Patient-Physician Relationship: The Scene of Narration The Chief Concern of Medicine. Fourth Edition ed. Ann Arbour: University of Michigan Press; 2013:137–168. Stoeckle JD, Billings JA. A history of history-taking: the medical interview. J Gen Intern Med. 1987;2(2):119–127. Sitzia J, Wood N. Patient satisfaction: a review of issues and concepts. Soc Sci Med. 1997;45(12):1829–1843. Stewart MA. Effective physician-patient communication and health outcomes: a review. CMAJ. 1995;152(9):1423–1433. Shortliffe EH. Doctors, patients, and computers: will information technology dehumanize health-care delivery? Proc Am Philos Soc. 1993;137(3):390–398. Street Jr RL, Haidet P. How well do doctors know their patients? Factors affecting physician understanding of patients' health beliefs. Journal of General Internal Medicine. 2011;26(1):21–27. Pieterse AH, Henselmans I, de Haes HC, Koning CC, Geijsen ED, Smets EM. Shared decision making: prostate cancer patients’ appraisal of treatment alternatives and oncologists’ eliciting and responding behavior, an explorative study. Patient Educ Couns. 2011;85(3):e251–259. Giveon S, Yaphe J, Hekselman I, Mahamid S, Hermoni D. The e-patient: A survey of israeli primary care physicians’ responses to patients’ use of online information during the consultation. Israel Medical Association Journal. 2009;11(9):537–541. Sommerhalder K, Abraham A, Zufferey MC, Barth J, Abel T. Internet information and medical consultations: experiences from patients’ and physicians’ perspectives. Patient Educ Couns. 2009;77(2):266–271. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press; 2001. Gerteis M, Edgman-Levitan S, Daley J, Delbanco TL, eds. Through the Patient’s Eyes: Understanding and Promoting Patient-centered Care. San Francisco: Jossey-Bass; 1993. Cronin C. Patient-Centred Care — An Overview of Definitions and Concepts. Washington DC: National Health Council; 2004. Gallivan J, Kovacs Burns KA, Bellows M, Eigenseher C. The many faces of patient engagement. J Participat Med. 2012;4:e32. Carman KL, Dardess P, Maurer M, Sofaer S, Adams K, Bechtel C, et al. Patient and family engagement: a framework for understanding the elements and developing interventions and policies. Health Aff (Millwood). 2013;32(2):223–31. Center for Advancing Health. A new definition of patient engagement: what is engagement and why is it important? [Internet]. Washington (DC): CFAH; 2010 [cited 2014 Aug 3]. Available from: http://www.cfah.org/file/CFAH_Engagement_Behavior_Framework_current.pdf Coulter A. Engaging patients in healthcare. New York (NY):McGraw-Hill Education; 2011. p. 10. Brown SM, Rozenblum R, Aboumatar H, Fagan MB, Milic M, Lee BS, et al., Defining patient and family engagement in the intensive care unit. Am J Respir Crit Care Med. 2015;191(3):358–60. Hibbard JH, Stockard J, Mahoney ER, Tusler M. Development of the Patient Activation Measure (PAM): conceptualizing and measuring activation in patients and consumers. Health Serv Res. 2004;39(4 Pt 1):1005–26. Jha AK, Orav EJ, Zheng J, Epstein A. Patients’ perceptions of hospital care in the United States. N Eng J Med 2008, 359:1921–1931. US Government. The Patient Protection and Affordable Care Act. 2009. UK Department of Health. Helping the NHS Put Patients at the Heart of Care: The Patient and Public Engagement Support Programme 2009–2010. London: Department of Health; 2009.

20   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

[28] HCAHPS: Patient’s Perspectives of Care Survey. [Internet] US Center for Medicaid and Medicare Services. 2012. [cited 2014 Aug 3]. Available from: http://www.cms.gov/Medicare/QualityInitiatives-Patient-Assessment-Instruments/HospitalQualityInits/HospitalHCAHPS.html [29] Manary MP, Boulding W, Staelin R, Glickman SW. The patient experience and health outcomes. N Engl J Med 2013;368(3):201–203. [30] Glickman SW, Boulding W, Manary M, Staelin R, Roe MT, Wolosin RJ, et al. Patient satisfaction and its relationship with clinical quality and inpatient mortality in acute myocardial infarction. Circ Cardiovasc Qual Outcomes 2010;3:188–195. [31] Isaac T, Zaslavsky AM, Cleary PD, Landon BE. The relationship between patients’ perception of care and measures of hospital quality and safety. Health Serv Res 2010;45:1024–1040. [32] Charmel P, Frampton S. Building the business case for patient centered care. Health Financ Manage 2008;62:80–85. [33] Meterko M, Wright S, Lin H, Lowy E, Cleary PD. Mortality among patients with acute myocardial infarction: the influences of patient centered care and evidence-based medicine. Health Serv Res 2010;45:1188–1204. [34] Hibbard JH, Greene J, Overton V. Patients with lower activation associated with higher costs; delivery systems should know their patients’ scores. Health Aff 2013;32(2):216–222. [35] Coulter A, Ellins J. Effectiveness of strategies for informing, educating, and involving patients. BMJ 2007;335(7609):24–7. [36] Holman H, Lorig K. Patients as partners in managing chronic disease. Partnership is a prerequisite for effective and efficient health care. BMJ 2000;320(7234):526–7. [37] Hibbard JH, Mahoney ER, Stock R,Tusler M. Do increases in patient activation result in improved selfmanagement behaviors? Health Serv Res. 2007;42(4):1443–63. [38] Mosen DM, Schmittdiel J, Hibbard J, Sobel D, Remmers C, Bellows J. Is patient activation associated with outcomes of care for adults with chronic conditions? J Ambul Care Manage. 2007;30(1):21–9. [39] Fowles JB, Terry P, Xi M, Hibbard J, Bloom CT, Harvey L. Measuring self management of patients’ and employees’ health: further validation of the Patient Activation Measure (PAM) based on its relation to employee characteristics. Patient Educ Couns. 2009;77(1):116–22. [40] Hibbard JH, Greene J, Tusler M. Plan design and active involvement of consumers in their own health and healthcare. Am J Manag Care. 2008;14(11):729–36. [41] Becker ER, Roblin DW. Translating primary care practice climate into patient activation: the role of patient trust in physician. Med Care. 2008;46(8):795–805. [42] Hibbard JH, Cunningham PJ. How engaged are consumers in their health and health care, and why does it matter? Res Briefs. 2008;(8):1–9. [43] Greene J, Hibbard JH. Why does patient activation matter? An examination of the relationships between patient activation and health-related outcomes. J Gen Intern Med. 2012;27(5):520–6. [44] Remmers C, Hibbard J, Mosen DM, Wagenfield M, Hoye RE, Jones C. Is patient activation associated with future health outcomes and healthcare utilization among patients with diabetes? J Ambul Care Manage.2009;32(4):320–7. [45] Rozenblum R, Lisby M, Hockey P, Levtizion-Korach O, Salzberg C, Efrati N, et al. The patient satisfaction chasm: the gap between hospital managements and frontline clinicians. BMJ Quality Safety 2013;3:242–250. [46] Rozenblum R, Lisby M, Hockey P, Levtizion-Korach O, Salzberg C, Lipsitz S, et al. Uncovering the blind spot of patient satisfaction: an international survey. BMJ Quality Safety 2011;20(11):959–965. [47] Luxford K, Safran DG, Delbanco T. Promoting patient-centered care: a qualitative study of facilitators and barriers in healthcare organizations with a reputation for improving the patient experience. Int J Qual Health Care 2011;23:510–15.

1 Patient, family, clinician and researcher perspectives   

   21

[48] Frampton S, Guastello S, Brady C, Hale M, Horowitz S, Bennett Smith S, et al. The Patientcentered Care Improvement Guide. Derby, Connecticut, USA: The Planetree Association and The Picker Institute; 2008. [49] Hospital Pulse Report. South Bend, Indiana, USA: Press Ganey Associates; 2010. [50] Rozenblum R, Gianola A, Ionescu-Ittu R, Verstappen A, Landzberg M, Gurvitz M, et al. Clinicians’ Perspectives on Patient Satisfaction in Adult Congenital Heart Disease Clinics-A Dimension of Health Care Quality Whose Time Has Come. Congenit Heart Dis. 2014. [51] Tang PC, Ash JS, Bates DW, Overhage JM, Sands DZ. Personal health records: definitions, benefits, and strategies for overcoming barriers to adoption. J Am Med Inform Assoc 2006;13:121–6. [52] Ricciardi L, Mostashari F, Murphy J, Daniel JG, Siminerio EP. A national action plan to support consumer engagement via e-health. Health Aff (Millwood) 2013;32(2):376–84. [53] Ross SE, Lin CT. The effects of promoting patient access to medical records: a review. J Am Med Inform Assoc 2003;10:129–38. [54] Zhou YY, Kanter MH, Wang JJ, Garrido T. Improved quality at Kaiser Permanente through e-mail between physicians and patients. Health Aff (Millwood) 2010;29(7):1370–5. [55] Tang PC, Lansky D. The missing link: bridging the patient-provider health information gap. Health Aff (Millwood) 2005;24:1290–5. [56] Finkelstein J, Knight A, Marinopoulos S, Gibbons MC, Berger Z, Aboumatar H, et al. Enabling patient-centered care through health information technology [Internet]. Rockville (MD): Agency for Healthcare Research and Quality; 2012 Jun [cited 2014 Aug 3]. (Evidence Report No. 206, Contract No. 290-2007-10061-I). Available from: http://www.ncbi.nlm.nih.gov/books/ NBK99854 [57] Delbanco T, Walker J, Bell SK, Darer JD, Elmore JG, Farag N, et al. Inviting patients to read their doctors’ notes: a quasi-experimental study and a look ahead. Ann Intern Med. 2012;157(7):461–70. [58] Goldzweig CL, Towfigh AA, Paige NM, Orshansky G, Haggstrom DA, Beroes JM, et al. Systematic review:secure messaging between providers and patients, and patients’ access to their own medical record. Evidence on health outcomes, satisfaction, efficiency, and attitudes [Internet]. Washington (DC): Department of Veterans Affairs; 2012 Jul [cited 2012 Dec 20]. Available from: http://www.hsrd.research.va.gov/publications/esp/myhealthevet-EXEC.pdf [59] Baldry M, Cheal C, Fisher B, Gillett M, Huet V. Giving patients their own records in general practice: experience of patients and staff. BMJ (Clin Res Ed) 1986;292:596–8. [60] Matheny ME, Gandhi TK, Orav EJ, Ladak-Merchant Z, Bates DW, Kuperman GJ, et al. Impact of an automated test results management system on patients’ satisfaction about test result communication. Arch Intern Med 2007;167:2233–9. [61] Baer D. Patient-physician e-mail communication: the Kaiser Permanente experience. J Oncol Pract. 2011;7(4):230–3. [62] Giardina TD, Menon S, Parrish DE, Sittig DF, Singh H. Patient access to medical records and healthcare outcomes: a systematic review. J Am Med Inform Assoc. 2014;21(4):737–41. [63] Walker J, Leveille SG, Ngo L, Vodicka E, Darer JD, Dhanireddy S, et al. Inviting patients to read their doctors’ notes: patients and doctors look ahead: patient and physician surveys. Ann Intern Med. 2011;155(12):811–819. [64] Wagner PJ, Dias J, Howard S, Kintziger KW, Hudson MF, Seol YH, et al. Personal health records and hypertension control: a randomized trial. J Am Med Inform Assoc. 2012;19(4):626–34. [65] Tenforde M, Nowacki A, Jain A, Hickner J. The association between personal health record use and diabetes quality measures. J Gen Intern Med. 2012;27(4):420–4. [66] Bates DW, Wells S. Personal health records and health care utilization. JAMA. 2012;308(19):2034–6.

22   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

[67] Gysels M, Richardson A, Higginson IJ. Does the patient-held record improve continuity and related outcomes in cancer care: a systematic review. Health Expect. 2007;10(1):75–91. [68] Archer N, Fevrier-Thomas U, Lokker C, McKibbon KA, Straus SE. Personal health records: a scoping review. J Am Med Inform Assoc. 2011;18(4):515–522. [69] Rozenblum R, Donzé J, Hockey PM, Guzdar E, Labuzetta MA, Zimlichman E, et al. The impact of medical informatics on patient satisfaction: a USA-based literature review. Int J Med Inform. 2013;82(3):141–58. [70] Rozenblum R, Bates DW. Patient-centred healthcare, social media and the internet: the perfect storm? BMJ Qual Saf. 2013;22(3):183–186. [71] Greaves F, Pape UJ, Lee H, Smith DM, Darzi A, Majeed A, et al. Patients’ ratings of family physician practices on the internet: usage and associations with conventional measures of quality in the English National Health Service. J Med Internet Res. 2012;14(5):e146. [72] Stage 1 vs stage 2 comparison table for eligible professionals. [Internet] Centers for Medicare & Medicaid Services. [cited 2014 Aug 3]. Available from: http://www.cms.gov/Regulations-andGuidance/Legislation/EHRIncentivePrograms/Downloads/Stage1vsStage2CompTablesforEP. pdf [73] Wells S, Rozenblum R, Park A, Dunn M, Bates DW. Personal health records for patients with chronic disease: a major opportunity. Appl Clin Inform. 2014;5(2):416–29. [74] Wells S, Rozenblum R, Park A, Dunn M, Bates DW. Organizational strategies for promoting patient and provider uptake of personal health records. J Am Med Inform Assoc. Forthcoming 2015. [75] Ahern DK, Woods SS, Lightowler MC, Finley SW, Houston TK. Promise of and potential for patient-facing technologies to enable meaningful use. Am J Prev Med. 2011;40(5 Suppl 2):S162–72. [76] Shortliffe EH. Strategic action in health information technology: why the obvious has taken so long. Health Aff (Millwood). 2005;24(5):1222–1233. [77] Nowak M. Evolutionary dynamics: exploring the equations of life. Cambridge: The Belknap Press of Harvard University Press; 2006. [78] Lieberman E, Hauert C, Nowak MA. Evolutionary dynamics on graphs. Nature. 2005;433(7023):312–316. [79] Parker L. The Darwinian take on Facebook. The Manchester Guardian; 2007.

C. Martin Harris and Gene Lazuta

2 Placing patients at the center of patient-centered care: a healthcare provider system perspective of a powerful new technology-enabled “language” The critical role of secure, two-way patient/clinician connectivity tools in the strategic design, planning and growth of provider organizations Abstract: Technology-savvy patients demand online access and convenience in practically every purchase, transaction and decision they make.[1] By including secure, technology-based two-way patient connectivity tools as a core component of their strategic equation, many forward-thinking healthcare provider organizations are working to satisfy consumer demand and use the information patients enter for and about themselves to design services that will more efficiently align clinician skills, facilities and locations with patient needs and preferences in ways that will increase outcomes quality and service affordability. The data generated through online systems that provide self-service insurance and co-payment management, pre-procedure questionnaires aligned to clinician-designed, technology-enabled care plans, demographically and geographically detailed maintenance capabilities, patient satisfaction surveys, and the ability to monitor patient recovery as part of readmission and avoidable complication reduction initiatives, can all be used to focus the perspective and guide the thinking of provider organization leadership as they work to meet the challenges of the coming, change-intensive years.[2] Building healthcare delivery systems that will remain competitive over the long-term requires strategic decisionmaking skills based on both experience and accurate, actionable information. Including the voice of the patient in this process will empower provider organizations to clearly understand the populations they serve as well as envision and build better versions of themselves that will meet and potentially exceed the growing demands of an aging population.[3]

2.1 Introduction In many ways, written language may represent the single most important technology ever developed by human beings because it embodies three critical concepts that fundamentally transformed the way people understand themselves and their place in the world. The first of these concepts is the idea that a sound or symbol can represent something other than itself.[4] This transcendent leap of creativity resulted in

24   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

the construction of words, the world’s first bytes of discrete data. Organizing words according to a commonly understood set of rules, or standards, and thereby capturing and transmitting ideas that are far more complex than any of the construct’s component parts, is the second of our critical concepts. And the third is that writing these data streams down on some type of a medium allows them to be saved for, and used by, other people in other places and at other times. This simple paradigm of organizing discrete packets of data into information that is then recorded in a way that makes it accessible and useful to others changed, quite literally, everything – from the way societies could be organized to the skills and aptitudes individuals and groups needed to compete and succeed.[5] When examined through a lens of core principles, the ongoing digitization of the modern healthcare industry can be seen to contain precisely the same components – discrete data capture, data organization and meaning, and standards-based information storage and exchange – that made language’s impact so all-encompassing and profound. Every day, all around us, incredible amounts of data (much like the unarticulated thoughts of pre-articulate people were once trapped inside their own, individual brains) is moving from the dark paper prison of the traditional medical chart into the bright light of a new day as the discrete, accessible raw material of tomorrow’s new knowledge. Clinicians are more effectively connected. Increasingly integrated workflows and treatment interventions are expressed as a part of ongoing cycles of coordinated care. Additionally, demographic and outcomes data are beginning to extend the reach of medical researchers more deeply,[6] and more quickly, into a richly detailed virtual reflection of reality that begins with the dynamic, realtime documentation clinicians perform directly at the point of care. For the leaders of healthcare provider organizations, the stories that today’s health information technology-based systems can tell about the workings of their organizations, both as clinical environments and as businesses, are also changing, quite literally, everything – from how human skills and physical resources are allocated and distributed, to the transparency and reporting expectations of payers and other contractors, to what it actually means for a patient to “see” a doctor. Medicine has always been a “knowledge business”, and in today’s increasingly digital medical practice environment, information technology-enabled tools are a foundational component of the language the knowledgeable speak.[7] But the defining purpose of language is, and will always be, exchange: of observations, ideas and experience. And even as technology-enabled tools are making it easier for clinicians to capture and collaborate through stories of ever-increasing complexity, the patients about whom these stories are written are beginning to use aspects of these same systems to more actively participate in the particulars of their care – and they are doing it by writing portions of their own stories themselves. For these patients, secure information technology connectivity is the medium through which they are making their voices heard from the places where we all spend the vast majority of our time; not in a doctor’s office, but at home, at work, and at play.

2 Healthcare provider system perspective   

   25

[8] And while other industries might gather information about their customers in the hope of better understanding their behavior for the purpose of increasing sales, the goal of every healthcare professional is, as it has always been, to deliver high quality patient care. So each and every day, healthcare providers across the country are using the emerging language of technology-based connectivity to more thoroughly understand the status and needs of their patients based, in part, on information those patients enter for, and about themselves.[9] Physicians and patients are using their visit time to add depth and color to the health-related stories they are writing together. In addition, leaders of provider organizations who are actively navigating the increasingly complex intricacies of today’s rapidly changing healthcare environment are aggregating these patient stories into knowledge-rich narratives that they can then use to inform the decisions they make about the forms and functions of their organizations in a nation that is moving from a volume-based to a value-based model of care delivery. To illustrate the real-world, transformative impact of two-way, technology-enabled provider/patient connectivity on care quality and the ability of provider organizations to remain viable and competitive in an increasingly demanding marketplace, the following story relates the experiences of a fictional patient moving through an equally fictitious healthcare delivery system. This fictitious provider system is intended to depict the high-level capabilities (not actual vendor-developed or regulator-approved products or systems) of an organization that uses the language of integrated information technology-enabled tools to connect its employed and aligned caregivers, its facilities and locations, and its processes and services to meet the needs of its patients as individuals and, through integrated analytic capabilities, its various regional patient populations. As events progress, we will pause to discuss some of the transformative features that make this story possible; many of them are based, in part, on actual innovations that are presently functioning components of Cleveland Clinic’s “patients first” integrated care model. Finally, we conclude our discussion with a look forward to an envisioned national healthcare environment that embodies some of the improvements that technology-enhanced patient/provider connectivity may help us achieve.

2.2 One patient’s story 2.2.1 The emergency department Michael Smith is a healthy, forty-four year old man who has a passion for staying active. He loves to push himself in any competitive sport, especially cross-country cycling. With spring in the air, he is spending his Sunday afternoon ramping up his outdoor activities with a pick-up game of basketball at the local park. Late in the

26   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

second half of the fast-paced game, he takes a quick corner pass, sees an open lane and charges in for a layup. Stretching over his defender’s head he rolls the ball off his fingertips to score, but as he is coming down he clips the defender’s shoulder, twists, comes down hard on his right leg, feels the knee buckle and lands in a heap on the concrete. Falls are nothing new for Mike, who counts the occasional bump and bruise as the price of admission for the games he loves to play, but from the amount of pain blasting up his leg he knows that, whatever he did to his knee, it was bad. By the time he hops over to a bench the knee is already swelling, and it keeps swelling, quite impressively, as one of his buddies drives him to the closest emergency department (ED), which happens to be part of one of the three large healthcare provider systems serving the region. Since Mike Smith has been treated by clinicians employed by the health system in the past, the attending physician uses the integrated electronic medical record (EMR) system that links every caregiver in every one of the system’s eleven regional hospitals and fourteen community family health center locations to update his demographic information, including his active medications, medication allergies and other pertinent details before going on to document the specifics of the encounter. Through the EMR, the ED physician also orders an X-ray of his knee and, because the system can display the images associated with a study’s results inside an open encounter, goes on to review the results with the patient, right there in the exam room. Turning the computer screen to make the actual X-ray pictures part of his condition education, the doctor says, “Well, Mike, you obviously don’t have any broken bones, which is good. But given the amount of pain you’re experiencing, and that swelling, I’d say the chances are good that you have some soft tissue damage. Since an X-ray doesn’t really capture that kind of injury, I think it would be a good idea for you to follow-up with an orthopedic specialist, who will probably want you to have an MRI. Until then, we’ll get you some crutches and send you home. Keep the leg elevated, and ice the knee to start getting that swelling down. I’ll order an antiinflammatory and some pain medication to help keep you comfortable overnight. Do you still use the pharmacy on Woodland Drive?” Mike indicates that yes, he does. “Okay”, the doctor says, pressing “enter” on his keyboard and using the EMR to transmit the ePrescriptions to the patient’s pharmacy of choice. “You can pick those up on your way home”. A few minutes later, Mike Smith is standing – somewhat awkwardly – on his new crutches when the person at the discharge desk looks at a computer screen and says, “I notice that the doctor recommended a follow-up appointment with an orthopedic specialist. Would you like me to schedule that for you now?” “Yes, please”, Mike replies, and after a few seconds during which the hospital employee uses the integrated scheduling feature of the health system’s EMR to do an appropriate provider availability search based on the information documented by the attending ED physician, she surprises him with the choices she is able to offer.

2 Healthcare provider system perspective   

   27

“I want to make sure that you get to the right specialist, in the right location”, she says, her eyes scanning the illuminated screen. “So, you could see one of our specialty staff physicians, but the first available appointment for tomorrow is all the way on the other side of town. Or, okay, here’s an orthopedic specialist who isn’t on our staff but who is a member of our Quality Choice community physician partnership program. He’s an excellent doctor, and he’s got a spot available tomorrow at ten a.m. He’s located very close to where you live; so, what would you like to do?” Mike chooses the Quality Choice physician, and the scheduler books the appointment, which automatically sends a verification to his health system-provided personal health profile account and, at the same time, activates a timed reminder that will send him an alert through the app on his smartphone one hour before he is scheduled to arrive at his new physician’s office. Then, with his follow-up appointment scheduled, his ED after-visit summary downloaded to the “upcoming appointments” section of his personal health profile, and his encounter closed, Mike heads for home, stopping at his local pharmacy, where his medications are waiting to be picked up.

2.2.2 Transformative Features: Connectivity, Care Continuity, Patient Choice Immediately following his injury, Mike Smith enters an emergency department that happens to be close to the site of his accident. But even though he did not actively choose this location, he does not enter it as a stranger. Instead, he enters as someone who has received care at various locations within this particular health system in the past. The moment he provides his identifying information, he becomes known to everyone he sees – from the person who takes care of his registration, to the financial counselor who verifies his insurance, to the radiologist performing his X-ray study. And by interacting with, and adding to the information that has previously been entered by all the different clinicians who have ever been a part of Mike Smith’s lifetime care team, his ED physician becomes that team’s newest member and the first to begin documenting this latest chapter in the story that is his ongoing health history. But this organization’s connectivity does not stop inside the health system. Instead, the system has extended the continuity of the care it can help provide by using its integrated EMR to connect to clinicians who participate in the community physician Quality Choice partnership program. An outgrowth of the ability of digital information tools to capture, record and report on the discrete information documented by clinicians during the care process, the community physician Quality Choice program is an organization composed of both physicians who are employed by the health system and regional physicians in private practice who either lease a customized version of the system’s EMR that is available as a secure online service, or who use a certified EMR that is interoperable with that system. To participate in the Quality Choice partnership, physician prac-

28   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

tices must agree to meet a number of clearly defined quality metrics and to provide evidence that they are meeting or exceeding these quality scores through a monthly report generated by a third-party from the information they document in the EMR. The benefits of participating in the Quality Choice partnership for physicians in private practice are the right to publicly site their “member in good standing” status as a way of differentiating themselves in their local market based on the quality of their work and inclusion in selected payer contracts negotiated by the health system. The benefits of the Quality Choice partnership program to the health system are increased care continuity, including informed patient “hand-offs” or “transitions of care” based on need and specialty expertise, measureable quality consistency and increased patient satisfaction. And the benefits to the patient are choice, because of the variety and wide physical and specialty distribution of Quality Choice physicians throughout the region, and the peace of mind that comes from knowing that any Quality Choice clinician they choose is consistently and demonstrably delivering high quality care.

2.2.3 The Quality Choice clinician visit, and the surgical evaluation The morning after his basketball mishap, Mike Smith visits a local orthopedic specialist who is a member of his health system’s community physician Quality Choice partnership. The first thing he notices is that, even though he has never visited this physician before, he is not asked to fill out any lengthy forms or questionnaires. Instead, the physician assistant simply asks him to verify certain demographic and incidentrelated information, and then the physician consultation begins. Using the same integrated EMR system that the ED physician used the day before, the Quality Choice clinician reviews the details of the accident, studies Mike’s X-Ray images and results, adds her own observations to the electronic medical record following a brief physical examination and, as predicted, uses the EMR to order an MRI at the imaging center across the street, which uses an information system that is interoperable with the system used by this physician practice. Ninety minutes later, Mike is back in the exam room where he watches his clinician use his on-screen MRI to deliver the news that, as a result of his tumble, he has a medial meniscus tear in his right knee and, as often happens with this type of injury, a torn anterior cruciate ligament, or ACL. While a more conservative treatment approach would be physical therapy and medication, because Mike is a particularly active man, and is very anxious to get back in the game as quickly as he can, a surgical repair performed by a Sports Health orthopedic surgeon sounds like his best course of action. So the Quality Choice physician refers him to a specialist in the health system’s Orthopedics Department, and a member of the office staff uses an online scheduling tool to book his appointment for a surgical evaluation later that afternoon. Deliberately arriving a bit early at the local family health center where his new Sports Health orthopedist practices, Mike takes a seat in the waiting room and fires

2 Healthcare provider system perspective   

   29

up his tablet. Connecting to the health center’s WiFi, he opens his Gmail and finds a notice asking him to, “Please log into your personal health profile account because you have new information to review”. Because e-mail is not considered secure, any e-mail notification he receives about any of his healthcare-related activities contains only non-specific information with the invitation to log into his user ID- and password-protected account where he can view his personal health information in a secure, encrypted and HIPAA-compliant environment. After logging in, Mike spends a few minutes using his personal health profile’s various features to complete a number of pre-surgical questionnaires that were automatically pushed to him when his appointment in the Orthopedics Department was verified in the scheduling system. In preparation for his surgery, he is also prompted to provide his own “personal preferences”, including any hearing or sight impairment he might have, any of his allergy or culture-related dietary restrictions or requirements, or any other special needs about which his caregivers should be aware as they construct his treatment plan. He then completes a series of financial clearance tasks, including insurance verification, co-payments and self-pay options, reviews the procedure-specific patient education information related to his surgery that was recommended for him based on the personalized details contained in his electronic medical record, and verifies that, of the available health system hospital locations where the type of surgery he needs is performed, he prefers the one that is closest to his home. Then a new notice pops up. Clicking on it, he discovers that, because of the specifics of his injury, his age, level of physical activity and several other medical and demographic criteria, he is qualified to participate in a clinical research study designed to evaluate the tear healing of his ACL. The study is being jointly sponsored by his health system and a well-known scientific research organization, and after reading an overview of the purpose and goals of the study, he agrees to participate by e-signing a consent form and completing a very detailed condition survey that includes both discrete and free text patient-entered data fields. Once he is called back to the examination room, Mike meets his surgeon, who reviews all of his pertinent previous treatment encounters and imaging studies, conducts an electronic medication reconciliation, documents that all necessary evaluation processes and materials are complete, and describes, step-by-step, everything that will happen during and after the surgery he is recommending to address Mike’s injury. Mike reads and then e-signs a “consent-to-treatment” form, and his surgery is scheduled for early the next morning.

2.2.4 Transformative features: care coordination, patient-entered-data, care guides In the previous, Emergency Department portion of this story, we saw how integrated health information technology tools facilitated our patient’s ability to actively partici-

30   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

pate in the creation of his own connected, condition-specific care team by allowing him to select his doctors from a list of choices generated to meet his medical needs as well as the location(s) where his care would be delivered as determined first by his medical circumstances and then by proximity, appointment availability and other convenience factors. By first ensuring that every member of the care team – from the Emergency Department, to the office of an affiliated privately practicing physician, to the health system-based Orthopedics Department family health center exam room – were always working from the same up-to-date electronic medical record data set, the health system’s EMR successfully accomplished the critical core functions of an effective language: it captured discrete data about Mike Smith’s treatment journey; it allowed that data, in accordance to a set of commonly understood rules and standards, to be organized and reorganized into new information; and it stored that information in a way that made it accessible, and useful – not only to his caregivers, but also to Mike Smith himself. By using the information recorded about him (diagnostics, treatment plan, ongoing status) and entered by him (demographics, subjective assessments, personal preferences) in the EMR as the measures upon which his options were based, the system substantially reduced the likelihood that Mike, or any of his clinicians, would make anything but a fully-informed choice. Just as one effective way to help minimize the possibility of making a non-optimal choice is to first weigh a group of potential choices against the best available evidence, and to then make a decision by selecting from a remaining list of options that have each met an evidence-based standard, using a best-practice-informed, technology-enabled processes will now become a common feature in the rest of our story. Because, while it may not be immediately obvious to Mike Smith (caught up as he is in the moment and doing his best to remember everything he is being told regarding his experience and recovery), underlying all that happens from this point forward is the active application of the accumulated skill and experience of whole teams of clinicians representing multiple specialties and disciplines. In this particular regional health system, this process is called a “Care Guide” – and it is important. Several years prior to Mike Smith’s basketball mishap, the leadership of the regional health system’s multidisciplinary group practice conducted a careful assessment of the way its clinicians and other caregivers aligned to care for a group of common or particularly complex conditions. What they found was that across the health system, multidisciplinary care teams composed of medical doctors, surgeons, nurses and other allied health professionals were increasingly utilizing the organization’s integrated information technology capabilities to collaborate and communicate. Leadership therefore empowered the organization’s innovative clinician staff to identify care team best-practices specific to selected conditions, disorders or procedures, and to define ways that the EMR could serve not only as a connectivity tool, but to deliver highly-focused clinician decision support options. The resulting initiative allowed some of the most accomplished and experienced clinicians in the world to

2 Healthcare provider system perspective   

   31

come together to identify their group’s best judgment about the most effective path a specific type of patient’s care should follow and configure the integrated information connectivity systems to support the consistent delivery of that specialty-guided care, anywhere across the health system, regardless of how far the organization grew. The final result of these multidisciplinary clinical/technical efforts were what the clinicians called multidisciplinary, technology-enabled Care Guides. The work that went into their creation was illuminating. The first step toward operationalizing a technology-enabled Care Guide is the creation of a care team-approved, condition-specific Care Set. This single document reflects all of the best practices identified by acknowledged experts in all the pertinent disciplines related to the care of a condition such as prostate cancer or a stroke, or the delivery of care related to a particular procedure such as total hip or knee replacement surgery. At the start of the initiative, an initial proof-of-concept Care Set is finalized. Then that Care Set is subjected to a detailed review by a team of technology professionals who map the activities of the envisioned care flow against an information management schematic, plotting each of the required steps to a simple binary yes/no, if this/ then that structure. Following this demanding process, each step in the Care Set that involves a recommended course of action based on documented data points is programmed into the electronic medical record system. Working closely with an existing physician advisory group charged with evaluating ongoing clinician-suggested improvements and modifications to the health system’s EMR, the programming team constructs a technology-enabled Care Guide that presents its alerts and on-screen decision-support opportunities in real-time, and in ways chosen as effective and nonobtrusive by the clinicians themselves. No system-initiated recommendation can ever supersede a clinician’s judgment, but in some cases the workflow asks the physician to provide documentation of the specific clinical reasoning for not incorporating a Care Guide-identified treatment component (which is particularly useful in improving the Care Guide through real-world clinical application). After an initial pilot period that gradually expanded through the health system proved the viability and reliability of the first technology-enabled Care Guide, this same development process became the health system’s standard after it was modified to include a standardized pre-review step designed to address issues and components that were proven to be common to virtually all clinical workflows. So, with a multispecialty-developed, technology-enabled Care Guide working as part of the very infrastructure of the organization through which his surgery and subsequent recovery services would be delivered, Mike Smith enters an environment in which literally every component of the care he receives has been evaluated and measured according to the collective experience of dozens of highly-trained, highlyskilled clinicians while the facilities at which each episode of his care takes place, and the skills of the caregivers delivering those services at each step along the way, are

32   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

being carefully coordinated to ensure that all available assets are utilized in the most efficient and effective manner possible.

2.2.5 Surgery, recovery and additional patient services Following the surgical repair of his torn ACL, which took place at the hospital he selected because it is within a forty-five minute drive of his home, Mike Smith uses the scheduling function in his personal health profile to coordinate his post-surgical physical therapy encounters, which are offered in multiple family health center locations. During each physical therapy session, his progress notes are entered into the EMR by his therapists for review. Also, utilizing secure, remote monitoring technology such as a digital blood pressure/pulse rate monitoring device, Mike transfers data directly into his EMR so that much of his work can be overseen by his therapists without the need for him to actually visit a health system facility. And, all along the way, he receives secure ongoing updates and messages from various caregivers involved in his recovery through his personal health profile account. Once he is fully recovered, Mike’s eye again turns to the outdoors. Though he missed most of the summer’s best weather, there is still plenty of time for him to enjoy his favorite activity, cross-country cycling, before fall’s turning leaves transition to snow, which moves his exercise activities back indoors to the stationary bike and the other equipment he has in his home gym. At his last physical therapy session, he expresses his intention to “get back on my bike”, and his therapist recommends that he consider enrolling in a supervised program offered by the health system’s Sports Health team. Called the “Cyclist Performance Training Plan”, the progressive workout regimen includes both supervised onsite and web-based components that will allow Mike to use his secure online personal health profile to record and track his performance progress (including data entered directly from a set of metabolic monitors that record his heart-rate and other dynamic indicators of his effort and exertion), see data downloaded by the caregivers who are monitoring his activities (including nutritional counseling and other performance enhancement tools and services) and allow him to schedule follow-up appointments with health system and regional Quality Choice private practice physicians if and when they are necessary. The next day, monitors strapped into place, wireless bike computer calibrated to measure his distance and digital timer positioned on his handlebars, Mike sets off on what might feel like a solitary ride, but that actually includes an entire of team of caregivers who will accompany him over every mile, offering their expert guidance to help him achieve new levels of safe, individualized health and fitness. But as up-todate as his virtual training team’s capabilities might be, the one thing that will remain beyond their scope is preventing another accident, whether on the basketball court or the bike trail. Though, if something like that should happen, they will be there to help, once again.

2 Healthcare provider system perspective   

   33

2.2.6 Transformative features: organizing a health system based on need, personalized patient services, eResearch Leading from the understanding that the more often a clinician/care team performs a particular procedure, the better they are likely to become at doing it, various healthcare theorists, including Michal E. Porter, Ph.D. and Elizabeth Teisberg, Ph.D. in their book, Redefining Health Care: Creating Value-Based Competition on Results,[10] have observed that this experience-based approach, if applied to the healthcare industry as a whole, could potentially change the way skills and resources are organized and distributed. Instead of a “full-service” model in which every hospital everywhere offers a full range of services, from delivering babies to heart or spine surgery, hospital-based centers of excellence could be organized in carefully chosen locations, giving caregiver specialists the opportunity to focus on just their area of expertise thereby improving their skills through experience and streamlining the health system’s resource outlay by reducing redundant services. To accomplish this level of organizational efficiency requires access to detailed data, not only of the various quality metrics associated with a practitioner’s performance, but of the demographics and preferences of the patient population a health system serves. In the case of Mike Smith’s health system of choice, the integrated information technology-based EMR that caregivers and patients use to stay connected also connects the health system’s executive leadership to the kind of information that allows them to view their geographic region not as a collection of small, self-contained sub-markets radiating around each of their hospitals, but as a single, dynamic marketplace in which their resources can be organized to maximize clinical quality, minimize cost, and meet patient demand. Several years before Mike Smith’s experience, the health system leadership transformed the composition and distribution of the organization’s resources based on real-world information. The work began with an analysis of data accumulated over several years through which it immediately became clear that the vast majority of patients are perfectly amenable to traveling a reasonable distance to the location where specialized surgery or other therapeutic procedures can be performed, but for follow-up appointments, rehabilitation visits and other post-surgical activities, they much prefer to stay closer to home. This data set was aligned with quality metrics reports that illustrated how clinicians who are allowed to focus all of their attention on their particular clinical specialty achieved demonstrably higher quality outcomes than those who are expected to spread their attention throughout a more general practice environment. In an effort to improve population outcomes and, ultimately, to attempt to lower some of the costs related to delivering care, the health system’s leadership decided to leverage caregiver coordination capabilities that were made possible by the system’s integrated EMR connectivity tools to reorganize their human and physical assets.

34   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

They began by first redefining the regional marketplace into three zones (east, central and west) and creating an inventory of all health system assets and corresponding clinical activities that took place in each of these zones over a quantified period of time. Lessons learned during this analysis showed that there were a large number of duplicative surgical and medical programs operating through multiple, often closely proximal facilities, the locations of which were the result not of any strategic application of resources, but of the organic growth of the organization over decades. Next, an integrated clinical, financial and system operations team conducted a base case analysis of a large number of randomly selected patients, creating a dashboard report that detailed the fixed and variable labor and supply costs associated with a cataloged health system “book of business” as well as all direct and indirect expenses. Almost immediately several factors became apparent. The first was that, across the various sites, there were few, if any standardized approaches to treatment, documentation or, significantly, resource purchasing or allocation. Second, the distribution of medical and surgical specialists and their associated allied support personnel was geographically sub-optimal, resulting in overlap and redundancy that created what was, in essence, an environment of internal competition between various health system-owned sites. And third, that as a direct outgrowth of this non-optimized distribution of human skills, physical assets and substantial savings through bulk purchasing and contracting opportunities going unrealized due to uncoordinated, site-specific management, the health system had a total cost per encounter for similar services that could vary by an extraordinarily wide degree. To address these unacceptable variations, the health system executive leadership worked with department chairs, hospital presidents and administrative and other allied support personnel across the organization to implement a strategic consolidation of facilities and services through which they established several surgical centers of excellence supported by coordinated, zone-aligned outpatient, rehabilitation, extended care and other recovery resources, with only the most complex surgical interventions located on the central, downtown campus. This new distribution meant that the maximum distance a patient would need to drive for routine elective specialty care of any kind would be 30  minutes, 60  minutes for complex interventions, with consolidated rehabilitation and outpatient facilities keeping expected patient drive time down to twenty minutes or less. Across this re-imagined system, specialty-designed Care Guides were applied to the majority of the common and / or complex procedures performed, increasing the efficiency of the continuum of care in optimized clinical spaces that are better utilized by more specialized staff. Initially, this repurposing of facilities and reorganization of staff was met by expressions of disappointment by a community that believed, understandably though incorrectly, that something important was being “taken away” from their local hospitals. But a carefully crafted public relations effort through the health system’s Media Relations and Marketing departments outlined the benefits of the model to the area population, including directly addressing some of the “harder” questions that were

2 Healthcare provider system perspective   

   35

raised by citizens and local newscasters. Within three months the public’s concerns transitioned to an understanding of the ways in which the health system’s changes improved the level and quality of the services delivered, and the organization saw measurable reductions in both cost per case and cost per unit of service accompanied by increases in outcomes metrics and patient satisfaction. Returning to Mike Smith’s personal experience as our example, we can see how these changes aligned to offer him greater provider choice, improved care quality and a patient experience that kept him connected to his caregivers during every step of his treatment and recovery. This connected style of patient experience also provided him with a number of unique patient-centered services, such as the cycling performance program conducted by the Sports Health portion of the Orthopedics Department which had both regional and online connectivity opportunities that allow patients everywhere to benefit from its expertise. Which leaves one very important component of Mike Smith’s care that we have yet to address: the research study in which he agreed to take part. Our earlier description of a paper medical chart as a “dark prison” was deliberately selected because, for all practical purposes, information recorded on paper and then filed away in a chart storage facility all but completely disappears from view. As evidenced by the tremendous amount of time, effort and financial expense associated with clinical trials and practically every other type of medical research conducted over the past hundred years or so, paper charts are an unwieldy medium that only slowly gives up its secrets. But in a practice environment in which patient information is entered as discrete, searchable data points, the power of computer systems to do what they do best – rapidly collate, organize and reorganize selected information – frees human investigators to do what they do best: analyze information and creatively arrive at conclusions that can then inform future clinical behavior. Already, as part of our fictional health system, discrete, de-identified patient data is being parsed and dissected by interested clinicians and endlessly curious physician scientists who are formulating new queries and running increasingly complex reports against a database that contains the continually-updated treatment specifics of over five million individuals. And because this health system is a recognized national and international tertiary care resource for several medical and surgical specialties, the demographic depth and richness of this patient population closely reflects the general population in a range of statistically significant parameters. So, in addition to serving as a guiding light that illuminates the organization’s executive and physician leadership’s efforts to design a more effectively competitive health system structure, the data entered about and by the patients served in every facility of this growing enterprise is also revealing more, with each new query and every passing day, about the clinical efficacy of today’s medical and surgical treatments as the data-driven, outcomes-based prologue that will lead to the innovations and breakthroughs that will result from tomorrow’s new knowledge.

36   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

2.3 Conclusions We began our discussion by drawing a parallel between written language as perhaps the most broadly significant human technology ever developed and the emergence of a contemporary medical practice environment that can leverage the most significant technologic development of the past several centuries – namely machine-enabled computational power made widely accessible to individuals and organizations – to create a new way of understanding the health status and healthcare needs of our fellow human beings. This new technology-enabled language, still in its relative infancy, promises to deliver a transformation no less profound than what language itself initiated across human culture; if, that is, it can be used in ways that will maximize its value. What is true at the moment is that, for the first time in the history of modern medicine, people of differing backgrounds who represent a range of medical, cultural and societal disciplines and perspectives are beginning to “see” into the murky ocean of data that, up until so very recently, had been all but completely obscured behind the covers of countless paper charts. As they gaze into data’s nether reaches, they are discerning the first faint outlines of the underlying structures of some of the most burdensome diseases that have afflicted us and impeded our ability to thrive for a very, very long time. Given further time and study, we can reasonably expect creative and innovative people to see beyond these dim first glimpses and collaboratively discover ways to construct new, working interventions based on real, verifiable evidence that might, one day, relieve our suffering and, potentially, even extend the joy and satisfaction of our useful, healthy lives. And, as we saw over the course of our brief story, each progressive step in this promising process can be fabricated from three simple, language-analogous building blocks: 1. Discrete standards-based information capture: the “words” that form the vocabulary used by care teams to collaborate with one another, and with the patients they treat. 2. Security-protected, privacy-controlled information access: the grammar that allows data to be arranged and rearranged into meaningful statements and new understanding. 3. Widespread adoption and use of interoperable information technology tools: the real-world, meaningful communication between people that is the final goal and driving point of the entire technology-enabled process. Going forward, medical professionals, payers, policy makers and patients everywhere will have the opportunity to help guide the creation of a renaissance in the ways in which people can join together to confront the diseases and physical and cognitive impediments to health and vitality that have been so unavoidably a part of life since life itself began.[11] But as we saw in our fictional patient’s story, for this information-

2 Healthcare provider system perspective   

   37

driven process to truly deliver on its promise, it must be used as a bridge that will link patients and providers, researchers and the manufacturers of medications and medical devices, insurers, employers and their employees, government and other regulators, and the constituents of what may historically have been perceived as competing interests into a functional network of data-driven, success-oriented activity and achievement. Because, as we have implicitly observed, there is work enough for all, and abundant opportunity to succeed and to serve. The model outlined in our fictional healthcare delivery system can be imagined as expanding beyond a single market or region to become a national healthcare resource with centers of excellence specific to different conditions linked by interoperable information systems not only to one another, but to the regional and local provider organizations that will deliver the post-intervention rehabilitation and ongoing care patients will need once they return home from travelling to the right specialist for their condition.[12] Widely available outcomes reporting based on clearly defined, comparable information will not only help guide patient choices and physician referral relationships, but will also contribute to a broader understanding of which interventions and therapies are most effective[13] and therefore most likely reduce a full cycle of care costs by reducing further complications or recurrences. This same data made available to medical researchers bolstered by increasingly sophisticated information management systems could help to significantly reduce the time it will take for a new medication to move from the researcher’s bench to the clinician’s toolset[14] and help population health monitors track emerging illnesses such as flu or other conditions spread by human movement and contact.[15] And finally, this new technology-enabled language could one day bring together not only the clinicians and allied health professionals dedicated to caring for a local, regional or national community, but those individuals and groups around the world dedicated to this kind of work wherever they call home.[16] Because, on this small planet, where the mobility of people, food products, livestock and goods of every kind has all but completely erased even the most tangential barrier between cultures or groups, the ultimate health of any one of us may well rely, to an increasing personal degree, on the health of us all.

References [1] [2]

[3]

Abhishek P. Preparing for the 21st-century patient. JAMA. 2013;309(14):1471–1472. Yusofa MM, Kuljis Y, Papazafeiropouloub A, Stergioulasb LK. An evaluation framework for health information systems: human, organization and technology-fit factors (HOT-fit). IJMedInf. 2007;77(6):377–385. Katz SJ, Moyer CA. The emerging role of online communication between patients and their providers. JGIM. 2004;19(9):978–983.

38   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

[4] Deacon TW. The symbolic species: The co-evolution of language and the brain. New York: W.W. Norton and Company; 1997. [5] Miller CS. Health information technology execution and use: Exchanging patient data – benefits and rewards. In: Hospitals & health care organizations: Management strategies, operational techniques, tools, templates, and case studies. Florida: CRC Press, Taylor & Francis Group; 2013. [6] Buntin MB, Burke MF, Hoaglin MC, Blumenthal, D. The benefits of health information technology: A review of the recent literature shows predominantly positive results. Health Aff. 2011;30(3):464–471. [7] Walker J. Building a learning health system clinicians will use. Digital infrastructure for the learning health system: The foundation for continuous improvement in health and health care. Washington DC: The National Academies Press; 2011. [8] Randeree E. Exploring technology impacts of healthcare 2.0 initiatives. Telemed J E Health. 2009;15(3):255–260. [9] Ahern DK, Woods SS, Lightowler MC, Finley SW, Houston TK. Promise of and Potential for Patient-Facing Technologies to Enable Meaningful Use. Am J Prev Med. 2011;40(5):S162–172. [10] Porter ME, Teisberg EO. Redefining Health Care: Creating Value-Based Competition on Results. Massachusetts: Harvard Business Review Press; 2006. [11] Hoffman A, Montgomery R, Aubry W, Tunis SR. How Best To Engage Patients, Doctors, And Other Stakeholders In Designing Comparative Effectiveness Studies. Health Aff. 2010;29(10):1834–1841. [12] Vest JR, Gamm LD. Health information exchange: Persistent challenges and new strategies. J Am Med Inform Assoc. 2010;17(3):288–294. [13] Courtney PK. Data liquidity in health information systems. Cancer J. 2011;17(4):219–221. [14] Payne PR, Johnson SB, Starren JB, Tilson HH, Dowdy D. Breaking the translational barriers: The value of integrating biomedical informatics and translational research. J Invest Med. 2005;53(4):192–201. [15] Salathe M, Freifeld CC, Mekaru SR, Tomasulo AF, Brownstein JS. Influenza A (H7N9) and the importance of digital epidemiology. N Engl J Med. 2013;369:401–404. [16] Deshpande A, Brown M, Castro L, Daniel WB, Generous EN, Hengartner A, et al. A systematic evaluation of data streams for global disease surveillance. 2013 Online J Public Health Inform. 2013;5(1):e2.

David Lansky and Stephanie Glier

3 Using health IT to engage patients in choosing their doctors, health plans and treatments Abstract: Consumers’ healthcare choices have a major impact on the quality and cost of the care they receive and on the trends of the healthcare market at large. Despite the importance of these choices, consumers face significant barriers in accessing information about the comparative cost and quality of doctors, health plans, healthcare facilities or medical treatments. This chapter examines the availability and usefulness of current information to support these choices, identifies common barriers and discusses future opportunities to inform the healthcare market in a way that enables meaningful consumer choice and encourages providers to deliver high-quality, affordable care.

3.1 Why choosing a doctor or health plan matters Employers provide healthcare coverage to about 160 million Americans, and incur a total cost of about $900 billion every year.[1–3] Workers and their families pay about 20% of that total  – and that number has grown steadily since 2000 as companies find it harder to absorb the steady annual increase in healthcare costs.[1] The consequences of continued growth in healthcare costs for American business and working Americans’ personal finances are sobering.[4, 5] Despite the high cost of healthcare in the U.S., Americans have poorer health than people in almost all other high-income countries.[6] In addition, health outcomes and the quality of health vary widely by state and locality, with no correlation to the cost of healthcare.[7, 8] Unfortunately, this variation means that patients, families, businesses and public agencies cannot feel confident about the quality of care they will receive as they make expensive healthcare decisions. Choosing an insurance plan with its particular provider network, coverage rules and complex procedures; choosing a primary care doctor in the hope of having someone to shepherd the “medical home”; deciding whether to have various tests and procedures performed in an environment with daily news reports of overuse, underuse and error; selecting a specialist or surgeon to entrust with one’s life, at an enormous, imposing facility of unknown reliability… all of these decisions carry anxiety, costs and risks. Yet patients, their loved ones, their employers and even their referring physicians have almost no useful information to guide these decisions. In every other area of our lives, we have robust information resources to help us make these choices. Selecting a book, an appliance, a contractor, a college or a can of soup is supported by more information than selecting a hospital, surgeon or family

40   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

doctor. Information technology – often coupled with either government reporting requirements or crowdsourced ratings – is ubiquitously supporting modern life and, in turn, driving customer-centered business strategies. Employers and other healthcare purchasers are desperate to constrain and even reduce their healthcare spending, and families are increasingly choosing coverage and treatments with cost in mind and are unable to assess the quality of the coverage, provider or treatment. Purchasers believe that transparency and consumer choice are the best ways to manage costs over the long term.[9] They would like to provide consumers with information on price and quality and thereby create a marketplace where health systems and health plans that earn consumer trust will be rewarded. In such an environment, providers and plans will seek continuous improvement to gain market share and patient loyalty. The viability of this model will depend upon major improvements in two areas: the national information technology infrastructure and the development of consumer decision aids that leverage this infrastructure to support value-based choices. This chapter reviews the current state of consumer decision support tools and outlines some key next steps.

3.2 Information to make better healthcare decisions Patients and families – and the purchasers who organize the healthcare marketplace for them – face a variety of decisions: 1. Choosing a primary care or specialist physician 2. Choosing a health insurance plan 3. Choosing a hospital, nursing home or other facility 4. Choosing a medical treatment or test Each of these decisions is unique. Each patient has a unique health situation, family context, economic circumstance, set of competing priorities and level of healthcare knowledge. The good news is that modern health IT can be adaptive and personalized. The bad news is that few tools have yet realized the potential to support personalized decision-making. 3.2.1 Choosing a primary care or specialist physician Purchasers and patients believe that the single most important decision a patient makes is his or her choice of a doctor. Primary care doctors can be lifelong trusted advisors, coaches in personal lifestyle decisions, monitors of important preventive and screening care, managers of long-term chronic care and advisors for navigating more complex care. They may be the most important professional health resource any

3 Healthcare market and purchaser perspectives   

   41

of us has. Purchasers recognize that as much as 70% of total healthcare spending is influenced by the primary care physician’s prescribing and referral decisions; additionally, the physician’s ability to affect patient self-care, adherence and advanced treatment decisions contributes even more. Today, approximately three in four adults have a designated primary care provider, and about 11% choose a new primary care doctor and 28% seek a specialist each year.[10, 11] Patients choose a new physician because of a change in where they live, the insurance coverage or job they have, dissatisfaction or a change in physician practice availability. What information can they use to make these decisions? From consumer research we have some information on what consumers would like to know when choosing a doctor, so we can evaluate the gap between their information needs and today’s information availability.[12] They want to know that a doctor is in their insurance network. They want to know basic information about the doctor and his or her practice – such as training, location, accessibility, languages spoken and board certification. Patients also want to know how well people like themselves have been treated, whether service and access are good and what results patients with similar needs have obtained. They want to know whether doctors follow the recommended best practices for their area. And they want to know what it will cost them to get care from a potential provider. If consumers are interested in patient experience, patient outcomes and patient out-of-pocket costs, purchasers want measures of overall value and recognize that cost and quality are not correlated. They want discrete measures of physician cost and quality performance, and they want to educate consumers that significant variations exist, that high cost is not a proxy for high quality and that the patient’s own decisions about provider, facility and treatment will determine the ultimate quality and cost of care that the patient receives. There are a few public sector, and free, resources to help consumers, but these have incomplete information and seem to be little used – in particular, public sector tools have woefully little price information and mediocre user interface designs. The federal Centers for Medicare and Medicaid Services (CMS) sponsors Physician Compare, which was mandated to address many of these domains by federal statute, but has moved slowly to do so and does not appear to attract much consumer traffic. [13] Some states have posted websites with physician quality data – such as the California Office of the Patient Advocate, which reports summary ratings for physician groups, but not for individual doctors.[14] Many health plans host provider directories that provide the structural data about each doctor’s practice, but virtually no comparable quality data or personalized price data.[15, 16] Several states have used mandated all-payer claims databases to produce physician group quality scorecards; these also appear to be little used.[17] There are a small number of public websites that provide quality data on specialists, such as the New York state reports on coronary artery bypass graft mortality and the Society of Thoracic Surgeons’ collaboration with Consumer Reports.[18, 19]

42   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

But these are not designed to assist in consumer decision-making; they simply report quality scores or ratings for a set of physicians without any navigation tools or integration with other important decision dimensions. Recently, some health insurance plans and provider organizations have moved more aggressively to offer pricing and quality information about their doctors to the public. Two examples: the University of Utah now surveys every patient following a visit and posts the individual doctor ratings and any associated comments for public view; the Surgery Center of Oklahoma provides an easy-to-use consumer interface to obtain estimates of procedure prices.[20, 21] Commercial vendors have attempted to address consumers’ need to select a highperforming doctor by aggregating as much cost and quality data as possible and building a new generation of consumer-friendly web tools that mimic successful approaches of non-healthcare services like Amazon and Orbitz. Castlight Health permits users to navigate through a structured decision process, examine quality and price data for individual doctors and select a personal physician. Castlight acquires its quality data from public sources and re-analysis of claims data; it gets its pricing data by aggregating actual paid claims from contributing employers and health plans and then presenting an “average” or “blended” price that approximates what typical patients can expect to pay. Some commercial products, such as RateMD, Vitals and Angie’s List invite online users to rate their doctor, aggregate those ratings and make them available to subscribers, crowd-sourcing the evaluation of physician performance. A new generation of web applications, such as GrandRounds and Doctor-onDemand, allows patients to select and contact a physician for second opinion or consultation services over the internet or phone. In these approaches, the list of available physicians has been determined by the business relationships undertaken by the software company, and they may or may not offer detailed quality information. Prices tend to be uniform for each particular type of service needed. Two significant barriers confound the market’s desire to populate consumer decision tools with useful comparison data on physician quality and price: there is still no consensus on what to measure, and there is no data infrastructure to acquire the primary information needed to generate measures of interest. Here’s an example driven by interest from purchasers in California to help patients (and referring physicians) choose an orthopedic surgeon based on quality and cost. Purchasers recognize wide variation in the cost and quality of total joint replacement surgery. From international registries and some small-scale regional data sets, there appears to be three-fold variation in short-term outcomes such as complications and reoperations, two-fold variation in long-term outcomes, such as pain relief and achievement of normal function, and eight-fold variation in the prices of the procedures. These high levels of variation could have significant implications for individual patients and provide important opportunities for system improvement. In the case of joint replacement surgery, there is consensus among surgeons on many key markers of quality, and strong consensus internationally about which

3 Healthcare market and purchaser perspectives   

   43

outcomes should be measured. CMS has even codified the recommended outcome domains in developing its federal measurement requirements.[22] These include patient reports of changes in pain, function and stiffness at six to twelve months following surgery. Clearly patients would like to be able to select surgeons who do a better job helping them achieve normal function. In developing such an information system, two particular challenges arise: how to capture data from patients before and after surgery, and how to risk-adjust the measured outcomes to reflect differences in the underlying severity of illness or condition. California surgeons and purchasers have developed a statewide registry that captures and reports the relevant quality indicators on a voluntary basis. Some insurance plans and purchasers are making comparative price information public. These

Figure 1: Generation of California Joint Replacement Registry (CJRR) Submission Files By Participating Sites

44   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

data sources remain separate today and are only crudely linked on some websites. We expect that various publishers of consumer-facing quality information, such as Castlight Health, will capture the available data and present them within their existing physician choice platforms. To acquire the 181 data elements needed to assess the quality of an orthopedic surgeon’s services requires interfaces across a number of health IT platforms. Figure 1 provides a simplified illustration of the data architecture required in every hospital where participating surgeons practice. This cooperative venture between the state’s orthopedic surgeons, purchasers, hospitals, insurance plans and philanthropies has produced a robust IT platform for capturing and reporting physician performance information for public uses. However, there is no underlying business model to justify expansion of this approach to other disciplines or across the nation. Ultimately, hospitals and physicians must recognize that their patient volumes and financial rewards depend upon providing such information to their patients and the community – and performing well in public comparisons.

3.2.2 Choosing a health insurance plan Another major decision consumers regularly face is the selection of a health insurance plan. When consumers choose high-value health plans that fit their cost, coverage and provider choice needs, they improve the efficiency of the healthcare market. In 2012, nearly 85% of the U.S. population had health insurance coverage of some kind, and more than half the population had employer-sponsored insurance.[23] More than 10 million adults gained coverage in 2014 during the major coverage expansions of the Affordable Care Act, including 8  million people who signed up for coverage through the new health insurance marketplaces.[24, 25] Little is yet known about how well the online marketplaces served consumers shopping for high-value coverage; however, previous research shows that consumers dread shopping for health insurance coverage and that they have very little understanding of the available health insurance policies.[26] In particular, consumers are confused by cost-sharing terms and jargon. As a result, consumers are only slightly better than chance at choosing the “best” plan even in a simplified information environment.[27] In addition to poor ability to choose the plan that minimizes total cost, consumers underutilize provider quality information when selecting health plans. Since September 2012, private health plans have been required to provide a uniform summary of benefits to all enrollees and those shopping for coverage. This form must also include coverage examples that illustrate how the health plan might cover medical care and expenses in different situations. Starting in 2014, any qualified health plan – that is, a plan that meets the requirements of the individual mandate –

3 Healthcare market and purchaser perspectives   

   45

fits within standardized parameters designed to improve consumers’ understanding of the health plans and to ensure sufficient consumer protection. In addition to including federally defined essential health benefits, qualified health plans are designated in one of four “metal” tiers (bronze, silver, gold or platinum) that indicate the actuarial value of the plan. For consumers shopping in the health insurance marketplaces, standardized information about plan benefits, cost and quality will help make the options easier to understand and compare. Though this standardization will help reduce consumers’ decision fatigue, the information underlying these standards remains spotty; for example, the marketplaces are not required to provide health plan ratings until 2016, and most consumers shopping for plans in the 2013–2014 open enrollment period had little more than summary scores of customer satisfaction and access to care. Consumers with employer-sponsored insurance often face a more limited set of choices based on what their employers offer. Purchasers know that plan selection will impact employee’s health and satisfaction as well as their own bottom line. Large purchasers’ exploration of the best ways to present information about plan options has yielded best practices for the content of a selection tool as well as for the “choice architecture” – the way the information is displayed to the user and the personalization or filtering options available. Even beyond the ways a selection tool lets a user assess choices, the default presentation affects final choices by establishing a baseline for further comparisons.[28] Best practices for these health plan chooser tools include: – Limiting cognitive tasks; – Personalizing the information display to use information that fits the user’s needs; and – Balancing cost and quality information. These guidelines are designed to help consumers choose the plan that best meets their needs and to reduce barriers that make that choice difficult. Table 1 illustrates a hierarchy for how features of the available plans can be displayed to best serve consumers’ needs, based on research by the Pacific Business Group on Health (PBGH) and others. Health plan chooser tools including the health insurance marketplaces should also feature some specific information to support consumer plan selection, including:[29] – The true value of the plan to the user – comparing plans on total cost (premiums or share of premiums plus out-of-pocket costs) based on average healthcare expenses for similar populations; – How plans compare on care and service quality; – Possible expenses in each plan in very good and very bad years, and the likelihood of having such years;

Tax subsidy amount

Tax credit and cost-sharing reduction eligibility rules

LAYER 2

LAYER 3

Coverage type & rules*

Cost-share amounts – $500 deductible, $25 co-pay, 20% co-insurance, etc.

Top services (user preferences)

Cost at time of care yearly – Your cost (dollar amount) – “Metals” category

Explanations: health plan/ product ratings

CAHPS composites – Getting needed care – Paying claims – Getting cost info., etc.

Health plan ratings – Access – Customer service

QUALITY

Explanations

MD use rules

OON rules

Doctor choice rules

Plan services (wellness, DM, & other)

Plan clinical ratings (HEDIS)

Provider directory search

Provider ratings

Provider network & plan services – Named MD – Number of PCPs in zip

* Includes health plan type, personal account, co-pay, major medical, etc. Also includes cost-sharing reduction eligibility and benefits. Source: Consumer Choice of Health Plan: Decision Support Rules for Exchanges, Installments I, II, and III. Pacific Business Group on Health; 2012 Nov 30. Available from: http://www.pbgh.org/storage/documents/plan_choice_rules_consumer_decision_support_installments_i_ii_and_iii_120312.pdf.

Calculator to adjust subsidy & time period

Total premium yearly

LAYER 1

COST

Table 1: Plan choice dimensions hierarchy example

46       Part I: Information Technology for Patient Empowerment: 360° Perspectives

3 Healthcare market and purchaser perspectives   

– – –

   47

Likely effects on out-of-pocket costs of known expensive future procedures or treatments, such as an expensive operation or pregnancy; A provider directory that spans the available plan options or the full health insurance marketplace; Any coverage gaps and any unusual benefit strengths.

A number of private sector tools to choose health insurance coverage have emerged in recent years, some serving as online brokers or connecting directly with brokers, and some operating as private insurance exchanges separate from the governmentoperated marketplaces. PBGH supports a Health Plan Chooser tool for about 2 million employees of member companies: an easy-to-use, online decision support tool that helps employees determine which plan best fits their needs and make applesto-apples comparisons between plans.[30] By incorporating information about the employer’s plan offerings, the Health Plan Chooser presents customized information to the employee that includes the cost, quality, coverage and services of each plan. Private companies such as eHealth (formerly eHealthInsurance.com) and Stride Health offer decision support tools that include basic information about premiums, cost sharing and provider network. These tools also offer recommendations for plan selection based on user information; for example, Stride Health recommends a “best plan” and “runner up” based on user’s age, preferred doctor, anticipated utilization and prescription drug use. However, these tools offer little or no information about the quality of the health plan or the network of providers and guide most users to choose a least-costly plan. Though many government-run health insurance marketplaces and private exchanges use these best practices and guidelines to help structure consumer choice, the underlying data needed to achieve these tools’ potential is not available – in particular, little information about health plan performance is available beyond plan accreditation status or the Healthcare Effectiveness Data and Information Set (HEDIS) tool and measures. Beginning in 2016, the health insurance marketplaces will include information about enrollee satisfaction with plans as well as cost and quality ratings, but the specific information that will be included is not yet finalized and it may not provide more robust information than is already available. Today, consumers shopping for health plans can see a moderate amount of information including in-network provider directory and out-of-pocket costs, but those who are interested in comparing the value of plans side-by-side have few options.

3.2.3 Choosing a hospital or care facility Today there is more robust comparative quality information about hospitals and nursing homes than about individual physicians or health plans. CMS publishes hospital performance on more than 100 quality measures on the Medicare Hospital

48   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

Compare website, which provides a snapshot of hospital quality and allows a sideby-side comparison of hospitals in a geographic area along with state and national average performance scores.[31] For example, Hospital Compare includes measures of patient experience from the Hospital Consumer Assessment of Healthcare Providers and Systems Survey (HCAHPS) such as how often patients received help as soon as they wanted, how well doctors and nurses communicated with patients and how well patients’ pain was controlled. Hospital Compare also includes measures of timely and effective care such as the average time patients spent in the emergency department before they were seen by a healthcare professional, measures of readmissions, complications and deaths, and other measures of interest to patients. The private sector offers other tools such as Why Not the Best, a comparison tool that combines the Hospital Compare data with additional information about patient safety, population health and utilization and costs.[32] A handful of private sector tools provide other measures of hospital performance: for example, the Leapfrog Group and Consumer Reports each produce their own hospital ratings and benchmarks, and Healthgrades publishes reports of hospital clinical outcomes and top performance. Some similar information is available for nursing homes and long-term care facilities through Nursing Home Compare, the Eldercare Locator and state websites reporting nursing home quality. For these facilities, emphasis is placed on structural measures such as nurse-to-resident ratios and patient outcomes such as pressure ulcers and patientreported pain. However, the need to choose a care facility, such as a hospital or nursing home, is a less commonplace occurrence than the choice of a physician or health insurance plan. A person choosing a hospital for an elective procedure or choosing a facility for long-term care may have some time to compare options, but in many cases people choose the facility closest to them, particularly in a time-sensitive situation like a life-threatening emergency requiring immediate hospital care or in an unexpected hospital discharge to a long-term care facility. Long-term care choices are further complicated in cases where a family member or caregiver is navigating facility options on behalf of the patient and may not know how to prioritize the patient’s preferences. In addition, there is moderately consistent information available about nursing home facilities, but little comparative information about costs of those facilities or about what insurance plans will cover in nursing home care. In addition, there are virtually no consumer-facing tools to help a patient or caregiver choose among alternative options such as a residential nursing home, in-home care or community living and care arrangements.

3.2.4 Choosing a treatment Perhaps the most personal healthcare decision a person faces is treatment choice. A  new diagnosis or a change in a chronic condition is typically accompanied by a

3 Healthcare market and purchaser perspectives   

   49

conversation with a physician to establish a treatment plan, but in many cases this conversation is stunted by insufficient evidence. The current evidence on medical treatments is rarely enough to help an individual understand the best course of action for his or her particular situation. A few initiatives have tried to address the need for granular and specific data through collaborative patient reporting. PatientsLikeMe is an online community where patients with specific conditions can share information about their responses to various treatments and interventions, and can discuss options with others with similar diagnoses or experiences. There are a number of patient registries that provide information about patients’ demographic information, treatments and health outcomes over time but these are typically not accessible to patients and families and the data is not organized in a meaningful and relevant way. The Patient-Centered Outcomes Research Institute (PCORI) was established in 2010 to boost clinical comparative effectiveness research and to redirect research to be more patient-centered, focusing on issues and outcomes critical to patients. PCORI anticipates investing $3.5 billion in research and building the necessary infrastructure to meet patients’ needs by the end of this decade. To date, PCORI has not produced tools that facilitate consumer choice of treatments. Other approaches to deploying health IT in support of treatment decisions include shared decision-making, health coaching and structured patient education. The federal “Meaningful Use” program has created an optional mechanism for providers to earn “credit” towards the electronic health record (EHR) incentive payments by deploying patient-specific education resources.[33] The Digest of Patient Experiences (DiPEX) at Oxford University has a long-standing platform for assisting patients in understanding potential benefits and risks of alternative treatment approaches through stories told by patients like themselves.[34]

3.2.5 Common challenges Purchasers are eager to help patients and families choose physicians, insurance plans, healthcare facilities and treatments in ways that are likely to achieve the best outcomes at an affordable cost. They believe that modern information technology – as deployed in other economic sectors – can support that goal, and that the widespread use of online decision aids by consumers will also drive improved performance by the healthcare system. Nonetheless, the deployment of health IT to support these four decision types continues to be limited by five common challenges. First, the US does not now have adequate source information on quality, outcomes, or price to support meaningful choice. Over the decades, there have been small successes, such as the standardization of hospital-acquired infection and readmission data, the widespread collection of health plan measures of prevention and screening services, and some common primary care process measures. Unfortunately, these

50   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

measures are not of greatest interest to consumers. The national system for collecting hospital patient experience data has been moderately useful, but also fails to speak to people with specific conditions, and current surveys of patients’ experience with medical groups does not address the performance of individual doctors or even practices. Patients are interested in health outcomes and patient experiences for people like them – and particularly when facing higher risk and higher cost services. In general, those developing and endorsing quality measures have focused on counting the frequency that specific care processes are performed, rather than the effectiveness or impact of those services on the patients. The recent expansion of our national health IT infrastructure has done little to correct this deficiency. By focusing on pointof-care information delivery and the ability to capture process measures in discrete settings and specialties, the information infrastructure has neglected the opportunity to capture patient outcomes over time and the ability of providers to coordinate care across settings. Second, the granular data needed to construct measures of outcomes and continuity of care are too often locked in data silos, under the control of healthcare organizations and their vendors. Slow progress toward broad adoption of interoperability standards, inadequate standardization and development of data aggregators and registries (“data intermediaries” as conceptualized by the federal Office of the National Coordinator for Health Information Technology, or ONC), industry uncertainty about the applicability of the Health Information Portability and Accountability Act (HIPAA) and the privacy rule to data sharing and aggregation, and the cost of data extracts and interfaces have all limited the opportunity to construct measures of interest to patients and purchasers. Third, most of us do not have a mental framework that organizes healthcare information for comparison purposes. How do I compare one primary care doctor to another, or one ophthalmologist to another? What are the dimensions of quality that I should consider, and how have I been exposed to comparative information in the past? The frameworks we have learned to compare products – from Consumer Reports displays to real estate listings – do not exist for most of us as patients and caregivers. As the health insurance exchanges created by the Affordable Care Act came into being, the IDEO design group did an extensive user-driven design project to help the new exchanges optimize the consumer decision-making experience.[35] In practice, the urgency of offering a minimally operational website by Fall 2013 meant that most of the state and federal exchanges gave little attention to the carefully developed user experience approach. Fourth, the structure of the healthcare marketplace has limited consumer demand for improved user interface design. Many employees have limited choice of health insurance plans and only a subset is required to select a primary care physician. Many of the most important healthcare decisions must be made quickly, in a context of clinical uncertainty and vulnerability. Few patients faced with the need for heart surgery or an urgent hospital admission will have the impulse or time to “shop” for

3 Healthcare market and purchaser perspectives   

   51

high quality, low cost options. Most continue to rely on the referring physician or other advisors for direction, and those advisors are rarely motivated to undertake unbiased research designed to optimize the patient’s imperfectly articulated values. As a result, the software industry has been slow to develop personalization tools of the caliber we now take for granted at Netflix or Amazon. Finally, in the American private sector healthcare system, cross-cutting and conflicting business interests frequently mitigate broad adoption of consumer decision aids. There are few parties who benefit from seeing most consumers make their care decisions based on unbiased quality and cost data. Many of the most successful health systems have thrived due to reputation and market power, rather than objective comparisons of quality. Furthermore, there is little association between price and quality. A market that rewards high quality and attention to reasonable pricing would disfavor many of today’s market leaders among providers, suppliers and insurers. As a result, many of the instrumental components of a consumer-facing health IT environment have been slow to take shape: interoperable, universal data, measures of health outcomes and patient experience, standard interfaces and open APIs for data exchange, and industry leadership to create a level playing field for consumer choices.

3.3 Future opportunities Much of the current practice in supporting consumer choice in the healthcare market is based on a disappearing model of healthcare organization and information availability. The consumer market is now split into two broad segments: those who are offered a choice of delivery systems or health plans that promise a comprehensive healthcare resource and those who are encouraged to “shop” for individual services, often using an account-based health plan, with high cost-sharing requirements. These two market segments call for differing decision criteria and decision aid software. In the market for comprehensive care solutions, patients can expect to select “products” that emphasize care by teams of multi-disciplinary professionals and can ask to see results that are achieved through coordination across the continuum of time and space and settings. The measures we have to capture these outcomes – both cost and quality – are poor, and the data infrastructure is inadequate. In the market for individual services and products, patients can look for price and quality data specific to individual processes – a blood test, a doctor visit, an imaging test. However, these are generally not associated with a meaningful result. Few patients have interest or understand the technical parameters of these care processes and the metrics associated with those processes. In this market, consumer choice may be focused on price, access and service quality. The decision aids are typically offered by a health plan sponsor, such as the patient’s own insurance company or employer.

52   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

But the healthcare marketplace – where people spend their money and get access to services – is changing. There is no uniform set of “products” in health care today. An insurance premium, a particular imaging test, a procedure or a visit – all of these might come in many different forms and “bundles” which make comparisons difficult. How do I compare the quality and price of a cataract operation if I get my care from a Health Maintenance Organization (HMO), or an Accountable Care Organization (ACO), or as part of a hospital Medicare shared savings program, or from an outpatient ambulatory surgery center or a specialty ophthalmology center? This complexity amid rapid market change means that the market for decision tools will also be segmented and dynamic. To increase patient engagement in these key decisions, it may be less important to focus directly on the consumer-facing tools than on the back-end infrastructure that supports those tools. If sufficient comparative data are available about price and quality, a lively competitive environment will emerge that builds upon those common back end resources. The health IT industry will benefit by rapid alignment and progress in these three back-end domains. First, the healthcare market needs clarity around several sets of standards. These include uniform data coding and extraction tools, common definitions of key marketplace products (e.g., total knee replacement episode, maternity care bundle), quality and price metrics, and the content and interoperability of provider directories. Market progress also depends, critically, on movement by payers and purchasers towards increased financial rewards for providers who perform well on value or engage in services, like shared decision-making, that are valued by patients. Second, market transformation can be accelerated by national agreement about necessary health IT, including the enabling data acquisition and transmission infrastructure. Health information exchange, all-payer claims databases, clinical registries, patient experience surveys and the associated federal guidance around the movement of data through these aggregation platforms will be critical. In practical terms, payment and recognition programs managed by large national payers – particularly CMS and the big carriers – will drive participation in these data intermediaries. A light touch of federal guidance could facilitate growth of intermediaries and will stimulate innovation in such areas as data extraction and translation tools, and crowdsourced patient-generated data. Third, payers need to devote a higher percentage of provider payment to measures of outcomes or comprehensive care processes. If providers are paid for improving health and managing costs, they will make better use of enabling information technology – particularly to influence the relevant patient behaviors and decisions. Fledgling efforts are underway at the National Quality Forum, Institute of Medicine and in the “Meaningful Use” program to develop such measures, but they are not yet in wide or consistent use. Such measures include total cost of care, patient-reported outcomes, conformity to appropriate use criteria, care transitions and continuity, and bundles of clinical outcomes, such as the D5 goals for comprehensive diabetes care in Minnesota.[36]

3 Healthcare market and purchaser perspectives   

   53

This approach – achieving national agreement on the back-end data infrastructure that supports improved patient engagement – has some implications for our current national health IT strategy. – Patients have to be seen as primary customers of health IT. As a result, developers of all health IT systems need to understand the patient and family’s information requirements, even if their immediate users are in the clinical settings. Ultimately, the data captured in the clinical setting by clinical systems need to have value to patients, families and purchasers in the community. – Adhering to national, industry and professional standards is critical. The public wants to compare providers, plans and treatments to each other. If each provider, product or implementation deviates from national standards, comparability is lost. This potential failure harms both the public interest and the opportunities for growth in the software sector. – Health IT source systems need to capture the data of interest to the public. As the market matures in defining products, and as payment methods mature in defining metrics that differentiate performance, we will need to re-engineer EHR and other source systems to capture and make available the necessary data in real time. This focus will, for example, require an ambulatory EHR to capture standardized data about indications of treatment appropriateness or symptom level for low back pain, as well as repeated measures of patient pain and functional levels. The American College of Cardiology’s PCI registry now captures indications of appropriateness for revascularization, particularly use of stents. – As clinical and patient-centered data are used for payment and recognition programs, there will be an increased risk of gaming and fraud. The next generation of EHR and data aggregation products will need edit, validation and audit functions to ensure data reliability for public uses. – It is unlikely that consumers will establish a new and different online behavior to support health decisions. How will healthcare decision-making be “mainstreamed” into normal search and shopping behavior? How might a Google search become a natural platform for selecting a physician? How will broad platforms that resemble Amazon or Yelp emerge for healthcare choices?

3.4 Conclusion What will motivate increased patient engagement? For good or ill, the inexorable increase in healthcare costs and in the patient’s share of costs will drive utilization of these tools, as well as more frequent discussions with doctors about whether to have certain tests or take certain drugs. There is increasing evidence today that higher cost-sharing coupled with tremendous variability in price is leading more patients to choose insurance plans with fewer benefits in order to have lower monthly premiums.

54   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

Will health IT enable more patients to make the best decisions about where to seek care and which care to undergo? Over the forty-year history of EHRs, there has been little attention to how patients and families might use health information – and little attention to how patients contribute data to the source record. Ultimately, of course, almost all of digital health information comes from patients. The medical history they report, the biological specimens they provide and their reports of symptoms and functioning all form the basis of the health IT industry. We have now built a national health information infrastructure to capture, move and apply this data to decisions made in the clinical setting. But we have not yet adjusted that infrastructure to capture and manipulate such data to create value directly for patients and families. The time has come to balance our investment in clinical health IT systems with a corresponding commitment to providing meaningful information to the patients and families who use and pay for the healthcare system.

References [1]

State Health Access Data Assistance Center. State-Level Trends in Employer-Sponsored Health Insurance [Internet]. Minneapolis, MN: State Health Access Data Assistance Center; 2013 Apr [cited 2014 Jul 24]. 78 p. Available from: http://www.shadac.org/files/shadac/publications/ ESI_Report_2013.pdf [2] Kaiser Family Foundation and Health Research & Educational Trust. Employer Health Benefits: 2013 Summary of Findings [Internet]. Washington, DC: Kaiser Family Foundation and Health Research & Educational Trust; 2013 Aug [cited 2014 Jul 24]. 8 p. Report No.: 8466. Available from: http://kff.org/report-section/2013-summary-of-findings [3] Schoenman J, Chockley N. Spending for Private Health Insurance in the United States [Internet]. Washington, DC: National Institute for Health Care Management Research and Educational Foundation; 2013 Jan [cited 2014 Jul 24]. 16 p. Available from: http://www.nihcm.org/images/ stories/Spending_for_Private_Health_Insurance_Rev.pdf [4] Swartz K. Cost-Sharing: Effects on Spending and Outcomes [Internet]. Princeton, NJ: Robert Wood Johnson Foundation; 2010 Dec [cited 2014 Jul 24]. 40 p. Research Synthesis Report No. 20. Available from: http://www.rwjf.org/content/dam/farm/reports/issue_briefs/2010/ rwjf402103/subassets/rwjf402103_1 [5] Chandra A, Gruber J, McKnight R. Patient Cost-Sharing and Hospitalization Offsets in the Elderly. American Economic Review [Internet]. 2010 Mar [cited 2014 Jul 24];100(1):193–213. Available from: http://www.aeaweb.org/articles.php?doi=10.1257/aer.100.1.193 [6] Institute of Medicine. U.S. Health in International Perspective: Shorter Lives, Poorer Health [Internet]. Washington, DC: Institute of Medicine of the National Academies; 2013 Jan [cited 2014 Jul 24]. Available from: http://www.iom.edu/Reports/2013/US-Health-in-InternationalPerspective-Shorter-Lives-Poorer-Health.aspx [7] Fisher E, Goodman D, Skinner J, Bronner K. Health Care Spending, Quality, and Outcomes: More Isn’t Always Better [Internet]. Lebanon, NH: Dartmouth Institute for Health Policy and Clinical Practice, Dartmouth Atlas Project; 2009 Feb 27 [cited 2014 Jul 24]. Available from: http://www.dartmouthatlas.org/downloads/reports/Spending_Brief_022709.pdf

3 Healthcare market and purchaser perspectives   

   55

[8] Radley D, How SKH, Fryer AK, McCarthy D, Shoen C. Rising to the Challenge: Results from a Local Scorecard on Local Health System Performance, 2012 [Internet]. New York, NY: Commonwealth Fund; 2012 Mar 24 [cited 2014 Jul 24]. Available from: http://www.commonwealthfund.org/publications/fund-reports/2012/mar/local-scorecard [9] Catalyst for Payment Reform. Statement by CPR Purchasers on Price and Quality Transparency in Health Care [Internet]. San Francisco, CA: Catalyst for Payment Reform; 2012 Nov 1 [cited 2014 Jul 24]. Available from: http://www.catalyzepaymentreform.org/images/documents/ Price_Transparency_Statement.pdf [10] Blackwell DL, Lucas JW, Clarke TC. Summary Health Statistics for U.S. Adults: National Health Interview Survey, 2012 [Internet]. Hyattsville, MD: National Center for Health Statistics; 2014 Feb [cited 2014 Jul 24]. 171 p. DHHS Publication No. 2014–1588. Vital and Health Statistics. Series 10, No. 260. Available from: http://www.cdc.gov/nchs/data/series/sr_10/sr10_260.pdf [11] Tu HT, Lauer J. Word of Mouth and Physician Referrals Still Drive Health Care Provider Choice [Internet]. Washington, DC: Center for Studying Health System Change; 2008 Dec [cited 2014 Jul 24]. 8 p. HSC Research Brief No. 9. Available from: http://www.hschange.org/ CONTENT/1028 [12] Hanauer DA, Zheng K, Singer DC, Gebremariam A, Davis MM. Public Awareness, Perception, and Use of Online Physician Rating Sites. JAMA. 2014 Feb 19;311(7):734–735. [13] Physician Compare Initiative [Internet]. Baltimore, MD: Centers for Medicare and Medicaid Services; Physician Compare Overview; [modified 2013 Dec 27; cited 2014 Jul 24]; [about 2 screens]. Available from: https://www.cms.gov/Medicare/Quality-Initiatives-PatientAssessment-Instruments/physician-compare-initiative/physician-compare-overview.html [14] California Health Care Quality Report Cards [Internet]. Sacramento, CA: State of California Office of the Patient Advocate; Medical Group Report Card; [cited 2014 Jul 24]. Available from: http://reportcard.opa.ca.gov/rc/medicalgroupcounty.aspx [15] Pacific Business Group on Health [homepage on the Internet]. San Francisco, CA: Pacific Business Group on Health; c2014. Health Plan Shopping Services Evaluation [cited 2014 Jul 24]. Available from: http://www.pbgh.org/news-and-publications/pbgh-articles-apublications/199-health-plan-shopping-services-evaluation [16] Price Transparency: An Essential Building Block for a High-Value, Sustainable Health Care System [Internet]. San Francisco, CA: Catalyst for Payment Reform [cited 2014 Jul 24]. 10 p. Available from: http://www.catalyzepaymentreform.org/images/documents/CPR_Action_Brief_ Price_Transparency.pdf [17] Love D, Custer W, Miller P. All-Payer Claims Databases: State Initiatives to Improve Health Care Transparency [Internet]. New York, NY: Commonwealth Fund; 2010 Sep [cited 2014 Jul 24]. 13 p. Commonwealth Fund pub. 1439, Vol. 99. Available from: http://www.commonwealthfund. org/~/media/Files/Publications/Issue%20Brief/2010/Sep/1439_Love_allpayer_claims_ databases_ib_v2.pdf [18] Adult Cardiac Surgery in New York State, 2009–2011 [Internet]. Albany, NY: New York State Department of Health, 2014 Mar [cited 2014 Jul 24]. 52 p. Available from: http://www.health. ny.gov/statistics/diseases/cardiovascular/heart_disease/docs/2009-2011_adult_cardiac_ surgery.pdf [19] Doctors: ConsumerReports.org [Internet]. Consumer Reports; c2006–2014. Heart Bypass and Surgery Ratings [cited 2014 Jul 24]. Available from: http://www.consumerreports.org/health/ doctors-hospitals/surgeon-ratings/ratings-of-bypass-surgeons.htm [20] Find A Doctor [Internet]. Salt Lake City, UT: University of Utah Health Care; c2014 [cited 2014 Jul 24]. Available from: http://healthcare.utah.edu/fad [21] Pricing [Internet]. Oklahoma City, OK: Surgery Center of Oklahoma; c2011–2014 [cited 2014 Jul 24]. Available from: http://www.surgerycenterok.com/pricing

56   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

[22] Yale New Haven Health Services Corporation, Center for Outcomes Research and Evaluation. Patient-Reported Outcome-Based Hospital Performance Measures Following Total Hip Arthroplasty and/or Total Knee Arthroplasty (THA/TKA) [Internet]. Baltimore, MD: Centers for Medicare and Medicaid Services; 2014 Mar [cited 2014 Apr]. Available from: https://www. cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/MMS/CallforPublicComment.html [23] DeNavas-Walt C, Proctor BD, Smith JC. Income, Poverty, and Health Insurance Coverage in the United States: 2012 [Internet]. Washington, DC: U.S. Census Bureau; 2013 Sep [cited 2014 Jul 24]. 77 p. Current Population Reports, P60–245. Available from: http://www.census.gov/ prod/2013pubs/p60-245.pdf [24] Sommers BD, Musco T, Finegold K, Gunja MZ, Burke A, McDowell AM. Health Reform and Changes in Health Insurance Coverage in 2014. New England Journal of Medicine [Internet]. 2014 Jul 23 [cited 2014 Jul 25]; epub ahead of print. Available from: http://www.nejm.org/doi/ full/10.1056/NEJMsr1406753?query=featured_home#t=article [25] Office of Health Policy. Health Insurance Marketplace: Summary Enrollment Report for the Initial Annual Open Enrollment Period [Internet]. Washington, DC: U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation; 2014 May 1 [cited 2014 Jul 24]. 45 p. Available from: http://aspe.hhs.gov/health/reports/2014/MarketPlaceEnrollment/Apr2014/ib_2014apr_enrollment.pdf [26] What’s Behind the Door: Consumers’ Difficulties Selecting Health Plans [Internet]. Consumers Union; 2012 Jan [cited 2014 Jul 24]. 12 p. Available from: http://consumersunion.org/ wp-content/uploads/2013/03/Consumer_Difficulties_Selecting_Health_Plans_Jan2012.pdf [27] von Glahn T (Pacific Business Group on Health, San Francisco, CA). Consumer Choice of Health Plan Research: Rules to Guide Exchange Decision Support [Internet]. Research findings presented to Minnesota Health Insurance Exchange Measurement and Reporting Technical Work Group; [cited 2012 Jun 11]. Available from: https://www.mnsure.org/images/MR-PBGHPresentation.pdf [28] Kleimann Communication Group and Consumers Union. Choice Architecture: Design Decisions that Affect Consumers’ Health Plan Choices [Internet]. Washington, DC: Consumers Union; 2012 Jun 9 [cited 2014 Jul 24]. 41 p. Available from: http://consumersunion.org/pdf/Choice_ Architecture_Report.pdf [29] Krughoff R, Francis W, Ellis R. Helping Consumers Choose Health Plans In Exchanges: Best Practice Recommendations. Health Affairs Blog [Internet]. 2012 Feb 29 [cited 2014 Jul 24]. Available from: http://healthaffairs.org/blog/2012/02/29/helping-consumers-choose-healthplans-in-exchanges-best-practice-recommendations [30] Helping Employees Choose a Health Plan [Internet]. San Francisco, CA: Pacific Business Group on Health; 2010 Jun [cited 2014 Jul 24]. 9 p. Available from: http://www.pbgh.org/storage/ documents/PlanChooser_IssueBrief_2010.pdf [31] Hospital Compare [Internet]. Centers for Medicare and Medicaid Services. What is Hospital Compare? [cited 2014 Jul 24]. Available from: http://www.medicare.gov/hospitalcompare/ About/What-Is-HOS.html [32] About: WhyNotTheBEST? [Internet]. Commonwealth Fund. c2014 [cited 2014 Jul 24]. Available from: http://www.whynotthebest.org/about [33] Step 5: Achieve Meaningful Use Stage 2 [Internet]. HealthIT.gov. Patient-Specific Education Resources. [cited 2014 Jul 24]. Available from: http://healthit.gov/providers-professionals/ achieve-meaningful-use/core-measures-2/patient-specific-education-resources [34] About: healthtalkonline.org [Internet]. University of Oxford; Our Story; c2014 [cited 2014 Jul 24]. Available from: http://healthtalkonline.org/about/our-story

3 Healthcare market and purchaser perspectives   

[35] Enroll UX 2014: Welcome to Coverage [Internet]. IDEO; [cited 2014 Jul 24]. Available from: http://www.ux2014.org [36] Minnesota HealthScores [Internet]. Minneapolis (MN): Minnesota Community Measurement. Diabetes; c2014 [cited 2014 Jul 24]. Available from: http://www.mnhealthscores.org/?p=our_reports&sf=clinic&search_phrase=&category_ section=category_condition&category=1&name_id=&compare

   57

Michael L. Millenson and Jane Sarasohn-Kahn

4 Old media to new in health: from information to interactivity Abstract: This chapter examines the ways in which new media are altering how individuals engage with the healthcare system and their own health. It examines how traditional print and electronic information (radio/TV) have typically functioned as personal health references; how that has changed; how apps and websites in a Web 2.0 world have assumed some of those functions and other unique capabilities related to availability, scope and personalization of information; and how old and new media interact. It examines provider control of messaging versus online patient communities and specific motivators for behavior change that are independent of medium. It places the evolution of media types within the context of a doctor-patient relationship moving from authoritative to facilitative and, eventually, collaborative. While old media can encourage and inform activation, new media provide tools that enable it, particularly as “triggers” towards behavioral change become more widespread. For providers, the challenge is to deal with a cultural revolution in the relationship with patients at the same time as a technological revolution in communication media. For patients, new media brands have yet to emerge that connote trust and reliability, although some apps carry federal regulatory approval. We believe the interactivity of Web 2.0 will enable a new level of healthcare consumerism, provider-patient collaboration and peer-to-peer learning that will be good for the healthcare system and even better for the health of every individual.

4.1 Introduction The Smokey Bear smartphone app, which helps users safely build and extinguish a campfire, has one flaw – it requires a connection. This, noted one reviewer, “is a little crazy if […] it’s for when you’re camping”. Even with a connection, the instructions are text-heavy and difficult to use on a small screen.[1] The suboptimal Smokey app conveys a message that concerns health care as well as campfires. The effectiveness of any particular medium depends upon the content, design and circumstances under which it is used. Marshall McLuhan notwithstanding, the medium alone is not the message. Effective messaging involves an interaction between medium, situation and user. As one researcher put it: “[B]oth media and methods influence learning, and they frequently do it by influencing each other”.[2] This dynamic is increasingly relevant for health care. This chapter discusses how the changing nature of media is altering the way in which individuals engage with the healthcare system and with their own health. We

60   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

include in our definition the ways in which individuals choose to engage in health on their own, as well as what happens when engagement involves outreach from the healthcare system. Traditionally, healthcare professionals have treated patient engagement as an extension of beneficence. It’s something providers do (when they do it) for patients: we engage you with brochures, videos and other communication tools. Patient engagement using health information technology (HIT) via new media capabilities can function as an extension of the same mode of behavior. The presentation may offer some interactivity, but it is still conveyed by the expert to the recipient to produce compliance. This is part of the Web 1.0 paradigm, where users are consumers of content, not creators.[3] That traditional perspective can be appropriate. A learned profession, after all, is defined by its unique knowledge, and physicians are among the most trusted sources of information in health care.[4] At the same time, individuals live most of their lives outside the healthcare system. Only 10% of a person’s health is attributable to healthcare system inputs, versus the 50% of health bolstered through personal behaviors at home, at work and in everyday life outside of the physician’s exam room.[5] People engage with their health and the healthcare system on their own terms. Nearly 70% of consumers reported that they look up symptoms online rather than going to the doctor first, and 64% try to learn about their condition instead of relying solely on the doctor.[6] New media, in combination with other social forces, fundamentally change how this engagement is carried out. Online information is abundant, cheap, personally oriented and designed for participation.[7] As a result, the care relationship is inexorably moving from authoritative to facilitative, bringing an increase in shared decision-making, patient self-care and proactive patients.[8] Croteau and Hoynes remind us how pervasive media are: The media surround us. Our everyday lives are saturated by the Internet, television, radio, movies, recorded music, newspapers, books, magazines and more. In the 21st century, thanks in part to the proliferation of mobile devices, we navigate through a vast media environment unprecedented in human history.[9]

The ways in which old and new media interact will be crucial to the course of patient empowerment in the future. Both will enable a new model of collaboration that challenges current provider work habits and business models.

4.2 The current situation: needs, gaps and challenges “Media” is the plural of “medium”, from the Latin medius, or “middle”. Communication media, Croteau and Hoynes explain, “are the different technological processes that facilitate communication between (and are in the middle of) the sender of a

4 Perspectives on Old and New Media   

   61

message and the receiver of that message”. The difference between a “reader” (of print or “reading” the sound and pictures of electronic media) and a “user” of media is that users contribute content, be it a review, tweet, photo or tag.[10] A TV ad for an online travel site is an example of old media; using the site to compare travel options and then book a trip is new media. For the purposes of this chapter, “new media” refers to those interactive capabilities. That difference between old and new media is beginning to have an impact.

4.2.1 New media are fundamentally changing health engagement Beginning in the post-war years, the provider-patient relationship has been profoundly altered by patient engagement with health information. The rebellious grandmothers of today’s Millennials rejected pediatricians who told them to feed their babies on a doctor-set timetable and embraced instead the radical notions of The Common Sense Book of Baby and Child Care. Author Dr. Benjamin Spock told mothers they could trust their instincts and feed their child “when he seems hungry, irrespective of the hour”. A quarter-century later, the moms of today’s Millennials banded together to write a booklet, Women and Their Bodies, to show that “women can become their own health experts”. It sold 200,000 copies by word-of-mouth, then morphed into the international bestseller Our Bodies, Ourselves.[11] Books, magazine articles, television and the social activist descendants of the Civil Rights movement have combined over time to help shatter the paradigm of instinctive deference to expert opinion. HIT in its turn is enabling the full realization of new roles in previously unimagined ways unique to the capabilities of new media. Interactions will be different. Some health data will originate with the individual’s doctor, but some will have no connection to the doctor, yet still be reliable and personalized. That data may be generated through wearable devices like digital health trackers or smartphones that bundle heart-monitoring capabilities. (This type of information that starts with the patient, not a clinician, is sometimes called patientgenerated data, user-generated data or consumer-generated data.) Sometimes, this type of data can also come from sophisticated websites where patients crowdsource personal health data. Already, a patient using online sources can develop an informed opinion about a diagnosis, treatment alternatives and those who provide them. This evolving information ecosystem will enable, and may eventually require, patients to take more responsibility for their health. It will certainly inform better choices about using and paying for healthcare services. Still, there will be pitfalls. Some consumers will inevitably overestimate the extent to which the “e” in “e-patient” stands for “expertise”. Doctors, after all, are the ones who say, “A physician who treats himself has a fool for a patient”. That will sometimes apply to self-care, if not quite as much as doctors might believe.

62   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

In addition, new media create opportunities for patient-powered research networks, enabling people with shared conditions to come together for “curing together”. [12] At the same time, new media bring new challenges related to control of data, consent for data to be used for research or other purposes – particularly patient-generated data – and privacy protection. Data reliability will inevitably be an issue, as well. There are issues related both to data provenance (“Is the data source what it’s purported to be?”) and accuracy (“Are patient-generated data coming from similarly calibrated devices or similar understanding of medical terms?”). At the same time, behavioral change ultimately depends not just on information, but upon the individual being sufficiently motivated to act, having the ability to perform the behavior and being triggered to perform the behavior, according to B.J. Fogg, director of Stanford University’s Persuasive Technology Lab. All three factors must be present.[13] Some aspects of the ability to change can be affected by, say, simplicity of web design. Other barriers, particularly for vulnerable populations, may require additional interventions, even though social media in general appear to be disseminating throughout the U.S. population independent of education, race or ethnicity.[14] Certainly, old and new media alike have a part to play in prompting behavioral change, sometimes separately and sometimes complementarily.

4.2.2 Patient engagement and activation – enabled, but not inevitable In contrast to Web 1.0, Web 2.0 creates a dynamic interactivity between user and content[15] that will reinforce the facilitative relationship discussed above. In time, that could lead to an enhanced physician-patient partnership characterized by collaboration. Even now, fully nine in ten U.S. adults say they want at least joint decision-making about their medical treatment, and one in four want to be “completely in charge” of their decisions.[16, 17] Engagement will become an essential characteristic in a collaboration that accommodates patients who wish their doctor to serve mostly as a guide, patients who prefer more traditional ways and those who may vacillate depending upon clinical and personal circumstances.[18] If information is power, digitized information is distributed power. It will be wielded by traditional healthcare sources (providers), less traditional ones (insurers and employers), outsiders (entrepreneurs and non-healthcare companies such as IBM or Google) and by patients themselves (via sites such as CureTogether or SmartPatients). But just as the medium is not really the message, neither the message nor its medium is inevitably linked to action. Financial incentives, for example, can certainly play a part in motivating health behavior change.

4 Perspectives on Old and New Media   

   63

Figure 1: 9 in 10 U.S. Adults (92%) want to share in health decision making, Spring 2014. 1 in 4 wants to be “completely” in charge.

4.2.3 Patients take on greater financial and clinical roles Incentives and behavioral economics have begun to affect financial and clinical behaviors by patients. So-called consumer-directed health care in the form of highdeductible plans and health savings accounts (HSAs) are growing quickly as part of a financial “risk-shift” to employees. In 2014, 79% of employers planned to offer some form of “account-based” health plan.[19] Not surprisingly, employees in these plans were more likely to make cost-conscious decisions.[20] As consumers take on more healthcare financial management responsibilities in the form of HSAs and high-deductible plans, they will evince “shopping” behaviors. Those behaviors, in turn, will be supported by the growing availability of online marketplaces and sites sharing patient satisfaction rankings. Still, challenges remain. A 2014 report on state efforts to collect and publish healthcare price information produced a failing grade for 45 states, with not one getting an “A”. Critics say available tools do not yet provide needed clinical detail.[21] Meanwhile, financial risk is helping prompt some consumers to pay more attention to self-care encouraged by providers who are themselves at greater financial risk for the costs of care in a “value-based payment” world. Tools for self-care and provider-patient collaboration include offerings for managing chronic obstructive pulmonary disease, diabetes and sleep disorders, online patient communities and

64   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

expanding telehealth options for care transitions. Although patient and provider interests may align, issues of control remain.

4.2.4 Healthcare organizations want to control the message New media still make many in the healthcare community nervous. Most corporations long ago learned you can’t control what visitors to your site see elsewhere. As a result, you can find negative as well as positive reviews on the websites of retailers and manufacturers. “Better they hear it here and stay here”, seems to be the philosophy. By contrast, University of Utah Health care garnered national publicity in 2014 by becoming the first health system to post patients’ positive and negative ratings of individual physicians on the hospital’s own website. Although comments were overwhelmingly positive, all are posted.[22] This example of transparency remains rare. In 2007, an affiliate of Blue Cross and Blue Shield of Minnesota launched The Health Care Scoop, promising, “Patient reviews from people like you”. Its candid postings about providers and insurers included negative reactions to Blue Cross.[23] Few, if any, organizations followed that lead, and the site has since been taken down. Allowing negative tweets to remain on a corporate Twitter account, albeit with replies, is the most typical departure from a command-and-control outlook. Engagement seems to mean deploying new media for its persuasive powers. Persuasion is certainly a form of engagement, but it isn’t collaboration.

4.2.5 Whom do people trust? Who should they trust? Individuals’ notions of authority and trust are changing. Respondents to the Edelman Health Engagement Barometer survey indicated that health companies’ websites, magazine articles and TV news, as well as medical brand websites, will become less important sources of information.[24] Meanwhile, other surveys have found that hospitals, pharmacists, nurses and doctors enjoy high levels of trust.[25, 26] Trusted intermediaries include Consumer Reports, publisher of Best Buy Drugs and physician ratings, AARP and medical societies. Pharmaceutical and device companies that wish to engage patients in their care must use authentic clinical evidence and reliable expert sources, since trust is a precursor to engagement.[5] To patients, healthcare has already entered the shared decision-making era. It is less clear whether providers see it the same way or are ready to share “content creation” in a new media world.

4 Perspectives on Old and New Media   

   65

4.2.6 Individuals are earlier adopters of new media than the health industry Individuals as consumers have long been living in a Web 2.0 world when it comes to managing finances (e.g. paying bills and trading stocks), booking travel and consuming media (e.g. newspapers, music, television and movies). Now, individuals in their role as patients and healthcare consumers have begun to participate in a Health 2.0 world as well. Examples include crowdsourcing of therapies on social networks,[27] sharing opinions about medication side effects[28] and creating clinical content through blogs, video logs (“vlogs”) and pictorial sites such as Pinterest and Instagram. This widespread digitization is pushing the healthcare industry to boldly go where everyone else has gone before. As these non-pioneers arrive, they are adopting techniques, strategies and learning from elsewhere.

4.3 Proposed solutions 4.3.1 Embrace and personalize transparency Perhaps the best example of old media blending with new is the combination of investigative journalism and tools provided by ProPublica, a Pulitzer Prize-winning non-profit supported by grant funding. ProPublica only exists online, but its stories are produced to be used, and often localized, by newspapers and broadcast outlets such as National Public Radio. Individuals can go online to their local outlet, or to ProPublica’s site, to employ an array of tools enabling them to act on the information they’ve just read or heard. Topics have included patient safety, drug and device company payments to doctors and doctor prescribing habits.[29] The key is being able to look at one’s personal physician or local hospital. Old and new media have a role to play in bringing transparency and accountability to areas of medicine traditionally cloaked in mystery, such as price, quality and outcomes. One example is using old media credibility to motivate change. Consider the April 2013 “old media” Time magazine cover story, “Bitter Pill: Why Medical Bills are Killing Us”.[30] The article by Steven Brill, of unprecedented length and depth for the iconic magazine, ignited a public furor by opening up the secretive, idiosyncratic and often unfair hospital chargemaster and medical billing processes. That scrutiny prompted public and political pressure for transparency. It is new media, however, which can turn outrage into actionable information. Sites for patients to share information on costs, quality, and service include ClearHealthCosts, OKCopay and PokitDok. Online marketplaces such as ZocDoc provide one-stop shopping to locate providers, book appointments and pre-pay for services online.[31]

66   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

Those who pay the brunt of rising healthcare costs – employers, government and individuals – are bringing unprecedented pressure on those generating those costs. A recent national insurer survey estimated that one of every five dollars paid to providers now falls under a “value-based purchasing” bundled payment arrangement that rewards improving care and lowering costs.[32] A dollar sign is being attached to “the right care for the right person at the right time”. As a 2013 Institute of Medicine roundtable concluded: “Prepared, engaged patients are a fundamental precursor to high-quality care, lower costs and better health”.[33]

4.3.2 Patient communities will support self-care and safer care transitions The leisurely back-and-forth of letters-to-the-editor in old media has evolved to blog comments and to sophisticated online patient communities (OPCs).[34] OPCs can offer emotional support, links to useful resources, tools for caregivers and an interactive environment for patients, caregivers and providers alike. Almost one in four people (23%) among those living with chronic conditions  has gone online to find others dealing with chronic and acute medical conditions,[35] and one in five has joined a health forum or community.[36] With over 140,000 members, PatientsLikeMe [37] is arguably the largest and most prominent OPC. Its registered members share information such as health status and progression of disease in response to various therapies. Patients hope to learn what works for others with the same condition in order to empower self-care as well as, perhaps, gain insights to share with their doctors. Ironically, community members in PatientsLikeMe were part of a study published in a journal behind a “paywall”, meaning the average individual had to pay to see it. One patient reacted to this scenario in a blog post titled, “Can Patients Collaborate Without Open Access?”.[38] Though the ability of OPCs to engage patients may be growing, they exist within a medical and scientific landscape that has not yet embraced open access to information for all.

4.3.3 Entertain to engage, where appropriate and authentic Despite what those in the field may sometimes think, health and healthcare topics do not automatically captivate a potential audience. Successful engagement must be truly engaging. Under the right circumstances, entertainment and information can mix. Talk show hosts like Oprah or her protégé, Dr. Mehmet Oz, have wielded enormous influence through television, radio and print, and added to their influence through newer media. The medium, the message, its actionability and their personal charisma have given them credibility that has been undented by traditional media reports alleging that the medical information they provide is often unreliable and

4 Perspectives on Old and New Media   

   67

even possibly harmful. It is unclear whether “reporting” by the odd hybrid of comedynews shows represented by Last Week Tonight With John Oliver will have a greater impact. Dr. Oz’s questioning by Congress in mid-2014 about questionable weight-loss advice was covered in greater depth by Last Week Tonight than by traditional news media.[39, 40] It was also a lot more entertaining. The “engage with entertainment” approach is so potentially powerful that it is being embraced by traditional organizations. Website and mobile apps use fun as a starting point for behavior change, offering personalization and interactivity that no TV or radio celebrity can match. Gamification can be particularly useful in dealing with diseases in children; e.g. autism[41] or other long-term chronic ills.[42] For example, “Diabesties” pairs college students for support in tracking diabetes with a peer. GeckoCap uses glowing caps on asthma medication as a signal for kids to take a dose; parents set up rewards for compliance. The game Nanobot’s Revenge for kids with cancer uses ChemoBlast 6MP instead of fireballs.

4.3.4 New media can help healthcare move quicker As noted previously, effective messaging with old or new media involves an interaction between medium, situation and user. Thomas Paine helped inspire the American Revolution with a pamphlet. The effect of Facebook postings and tweets on the Egyptian revolution against Hosni Mubarak in Tahrir Square was remarkably similar in why it had impact, while being completely different in the speed of its effect.[43] So, too, in healthcare: old and new media can make an impact, but new media can speed up engagement. Examples include recruiting for clinical trials, interacting with patients in communities, or bolstering compliance to medication through “social adherence”.[44]

4.3.5 New media in healthcare support patient activation The Patient Activation Measure (PAM) assesses self-efficacy and health confidence and assigns a score along a four-point scale.[45] What distinguishes newer HIT tools is that they offer patients an immediate way to translate feelings into action. As noted by Fogg, for information to change behavior, an individual must be sufficiently motivated, have the ability to perform the behavior and be triggered to perform it.[46] New media can place individuals in front of motivational “hot triggers” in the form of cues, prompts and other spurs to healthy actions. For example, the Food and Drug Administration-approved digital platform Propeller Health (formerly Asthmapolis) uses real-time health coaching to help individuals manage respiratory conditions. The device marries a sensor to a mobile app that together monitor a user’s condition if it worsens (exacerbation) and compares

68   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

metered-dose inhaler use to a personal baseline and general guidelines. The user receives alerts via a smartphone app to bolster medication adherence, along with other education and reminders. This type of evidence-based app increasingly has the active backing of clinicians. Similarly, a growing number of clinicians now recommend reliable websites, a practice sometimes referred to as information therapy.[47]

4.4 Discussion Changing inbred behaviors is difficult. Doing so during a technological and a social revolution in the doctor-patient relationship is more difficult still. In 1993, at the dawn of widespread use of the Web, an American Medical Association policy committee declared, “Physician and patient are bound in a partnership that requires both individuals to take an active role in the healing process”.[48] What has changed since then is the ability of one partner, the patient, to access sophisticated, personalized and actionable information outside the control or knowledge of the other partner, the physician. Sometimes that information may even be directed to physicians but available to patients, such as an online calculator of morbidity and mortality risk from the College of American Surgeons or the health app, Epocrates. What has also changed is the patient’s ability to partner with someone other than the doctor, when appropriate, to improve health or obtain better healthcare. OPCs and peer-to-peer healthcare offer tools and advice that an individual clinician often cannot, or will not, match. The evidence about the impact of new media on outcomes is encouraging, whether in general terms or in rigorous examination of the evidence for certain conditions, such as diet and exercise behavior.[49] However, it is still early, and the plethora of interventions (social networking sites, blogs/microblogs, “virtual worlds”, etc.) has made it difficult to definitively compare different tools.[50] Moreover, the growth of new media by no means signifies that the old doctorpatient relationship is defunct. What it does mean, however, is that the relationship needs to evolve as we have suggested, going from authoritative to facilitative and, eventually, to collaborative.[51] The transition could be turbulent, especially as patients adopt new media for clinical use far faster than physicians. Patient control of data may also disrupt long-established medical routines; for example, the “blinding” of clinical trials has become more difficult in an era when patients enrolled in a test of a drug may communicate with others about side effects and symptom amelioration via OSLs. New forms of trials that explicitly take this effect into account will surely begin to arise in response. In this same context, one threat to new media is doubt about the reliability and provenance of information, as mentioned earlier. The New Yorker cartoon captioned,

4 Perspectives on Old and New Media   

   69

“On the Internet, no one knows you’re a dog” became a meme for good reason. There are solutions beyond the scope of this chapter; for our purposes, it’s important to emphasize the role trusted “old media” can play. An ABC News-sponsored tweetchat on blood pressure control, for example, comes with credibility a random dot-com can’t match. Tools from ProPublica inspire trust in a way a site festooned with ads may not. Other tools will win credibility because of FDA or other regulatory approval or the transparency of the evidence supporting them. Trusted new media brands will arise, and they need not be non-profits. Brands like FedEx and Coca-Cola top “most admired” lists because of the emotional bond they’ve built with consumers and because of perceived reliability. Time will tell whether a PatientsLikeMe will become as trusted for a new generation as Our Bodies, Ourselves or a Merck Manual for previous generations. Or whether patients themselves may start earning an economic return for sharing their medical information. It’s easy enough to see the old media struggling; the shape of the new, or of a hybrid, has yet to clearly emerge. Importantly for new entrants to an entrenched industry, many have substantial financial backing. Some come with Old Media legacies, such as CareInSync, acquired by Hearst. The new entrants are challenging deeply embedded cultural, managerial and fiscal habits, and the change will be wrenching for many. Disruptive innovation disrupts, and those on the disrupted end are rarely disposed to passively accept their fate. Nonetheless, disruption is inevitable. “Participative technologies”, as one group of experts put it, will “accelerate demands for information, interactivity, and access to meet public expectations for accountability and consumer expectations for transparency and decision support”.[52]

4.5 Conclusion Old media storytelling can motivate and, on the web, provide modest actionability. For instance, New York Times reviews of activity trackers that produce patient-generated data include links to the manufacturer’s site.[53] Old media and new may be appropriate individually or in concert at different times. We learn by seeing, hearing and doing, by repetition and rote, by intellectual and emotional responses. The digitization of information opens up extraordinary communication capabilities. In healthcare, we stand at the start of the learning curve on how to best use them. The medium and the content must also be seen in a social context. The actions we take based on what we see, hear and feel – the “triggers” discussed in this chapter – are shaped by who we are as well. Age, ethnicity, race, gender and even political leanings can all play a part. So does income and education level. What seems humorous or motivational in a “gamification” app to one person may seem offensive to another

70   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

or simply fail to connect at all. Healthcare is going to have to learn careful research habits from the consumer products field. As new media become mainstream, “navigators” or information concierges will become more prominent; they’ll be the Information Age equivalent of the kindly librarian of old. We live in an emerging ecosystem; the web woven by web-based interactivity has yet to reach its full dimensions. There is little evidence that the old HIT, controlled by providers, has improved patient satisfaction.[54] But as distinctions between the HIT of the hospital and doctor’s office and that of the “rest of world” lessen, just as old/new media distinctions are beginning to, that may change. The path to collaboration may not be smooth, but the destination may yet make both partners in the provider-patient relationship much happier. In any event, given the convergence of patient-centered healthcare, social media and the Internet, a major shift in how patients and healthcare organizations connect is inevitable.[55] The activist patient rallying cry “Nothing about me without me” can be traced back to at least the 1990s. We believe the interactivity of Web 2.0 will enable a new level of healthcare consumerism, provider-patient collaboration and peer-to-peer learning that will be good for the healthcare system and even better for the health of every individual.

References [1]

[2] [3] [4]

[5]

[6]

[7]

[8]

Marks J. Smokey Bear's official app is close, but no campfire [Internet]. NextGov.com [updated 2012 Nov 27; [cited 2014 Jul 14.] Available from: http://www.nextgov.com/mobile/2012/11/ smokey-bears-official-app-close-no-campfire/59751 Kozma RB. A reply: media and methods. Educational Technology Research and Development 1994;42(3):11–4. Cormode G, Krishnamurthy B. Key differences between Web 1.0 and Web 2.0. First Monday.13(6). Edelman. Edelman health barometer 2011. [Internet] Health, activated. [cited 2014 Jul 14] Available from: http://healthbarometer.edelman.com/wp-content/uploads/ downloads/2011/10/Edelman-Health-Barometer-2011-Global-Deck-10.19.11.pdf. p. 19. Institute for the Future. Health and health care 2010. The forecast, the challenge. Second edition. [cited 2014 Jul 14] Available from: http://www.iftf.org/uploads/media/SR-794_ Health_%26_Health_Care_2010.pdf. p. 23. Lynch W, Perosino K, Slover M. Altarum Institute survey of consumer health care opinions. Spring 2014. [cited 2014 Jul 14] Available from: http://altarum.org/our-work/center-forconsumer-choice-in-health-care-research-findings%E2%80%94spring-2014 Rainie L. The rise of the e-patient: trends in the use of digital technology for health purposes. Presentation to Medical Library Association, Atlantic City, NJ. 2009 Oct 7. Available from: http://www.pewinternet.org/2009/10/07/the-rise-of-the-e-patient-2 Nash B. Presentation: Healthcare delivery and the new consumer. 2009. Available from: http://net.acpe.org/Current_Materials/Quality/Quality_PowerPoint/David_Nash/10_ Healthcare_Delivery_and_the_New_Consumer_Nov_2009_.pdf

4 Perspectives on Old and New Media   

   71

[9] Croteau DR, Hoynes WD (2013). Media/society: Industries, images, and audiences. Sage Publications. p. 2. [10] ibid, p. 8. [11] Millenson ML. Spock, feminists, and the fight for participatory medicine: a history. Journal of Participatory Medicine 2011;3:27. [12] Fleurence RL, Beal AC, Sheridan SE, Johnson LB, Selby JV. Patient-powered research networks aim to improve patient care and health research. Health Affairs 2014;33(7):1212–9. [13] Fogg BJ. A behavior model for persuasive design. Proceedings of the 4th international Conference on Persuasive Technology; 2009: ACM. [14] Chou W-YS, Hunt YM, Beckjord EB, Moser RP, Hesse BW. Social media use in the United States: implications for health communication. Journal of medical Internet research 2009;11(4). [15] O'Reilly T. What is Web 2.0: design patterns and business models for the next generation of software. Communications and Strategies 2007;65(1):17–37. [16] op. cit. Lynch, et al. [17] Quill TE, Brody H. Physician recommendations and patient autonomy: Finding a balance between physician power and patient choice. Ann Intern Med 1996;1(125):763–9. [18] Millenson ML. New roles and rules for patient-centered care. J Gen Intern Med. 2014:29(7):979–80. [19] Towers Watson. 2014. The new health care imperative: driving performance, connecting to value. 19th annual Towers Watson/National Business Group on Health employer survey on purchasing value in health care. [20] Fronstin P. Findings from the 2013 EBRI/Greenwald & associates consumer engagement in health care survey. December 2013. Employee Benefit Research Institute. No. 293. Available at: http://www.ebri.org/pdf/briefspdf/EBRI_IB_012-13.No393.CEHCS.pdf [21] Zamosky L. Online health pricing tools growing, but usefulness, patient engagement still low. iHealthBeat. 21 May 2014. Available at: http://www.ihealthbeat.org/insight/2014/onlinehealth-pricing-tools-growing-but-usefulness-patient-engagement-still-low [22] For an explanation of the University of Utah Health Care policy on posting Press Ganey satisfaction survey results, see: http://healthcare.utah.edu/fad/pressganey.php. For an example of patient ratings and opinions posted on the listing page of an individual physician (not all doctors have posted comments) see: http://healthcare.utah.edu/fad/mddetail. php?physicianID=u0440462 [23] Sarasohn-Kahn J. Getting the health care scoop: health opinions meet social media in Minnesota. Health Populi. 2007 Dec. 17. Available from: http://healthpopuli.com/2007/12/17/ getting-health-care-scoop-health [24] Edelman. Edelman health engagement barometer 2008. Health influence in the era of public engagement. Available from: http://www.slideshare.net/EdelmanInsights/edelman-healthbarometer-2008 [25] Gallup. Honesty/ethics in professions. Dec. 5–8, 2013. Available from: http://www.gallup.com/ poll/1654/honesty-ethics-professions.aspx [26] Harris. Americans less likely to say 18 of 19 industries are honest and trustworthy this year. Dec. 12, 2013. Available from: http://www.harrisinteractive.com/NewsRoom/HarrisPolls/ tabid/447/ctl/ReadCustom%20Default/mid/1508/ArticleId/1349/Default.aspx [27] Hamm MP, Chisholm A, Shulhan J, Milne A, Scott SD, Given LM, et al. Social media use among patients and caregivers: a scoping review. BMJ Open [Internet]. [Cited 2013 July 14], 2014; 3(5). Available from: http://bmjopen.bmj.com/content/3/5/e002819.full [28] IMS Institute for Healthcare Informatics. Engaging patients through social media. Is healthcare ready for empowered and digitally demanding patients? January 2014. Available from: http://bit.ly/1ku6SRG

72   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

[29] http://www.propublica.org/tools [30] Brill S. Bitter Pill. Time. 2013;181:16–55. [31] Sarasohn-Kahn J. Help yourself: the rise of online healthcare marketplaces. California HealthCare Foundation. August 2013. [32] Abelson R. Health insurers are trying new payment models, study shows. New York Times. July 9. 2014. Available from: http://www.nytimes.com/2014/07/10/business/health-insurersare-trying-new-payment-models-study-shows.html?_r=0 [33] Institute of Medicine. Partnering with patients to drive shared decisions, better value and care improvement. Workshop Proceedings. Roundtable on value & science-driven health care. Washington DC: National Academies Press; 2013. [34] http://healthcarediy.com/health-and-technology/power-online-patient-communities [35] http://www.pewinternet.org/2011/02/28/peer-to-peer-health-care-2 [36] http://www.pwc.com/us/en/health-industries/publications/health-care-social-media.jhtml [37] See: www.patientslikeme.com [38] Ciccarella A. Can patients collaborate without open access? 9 July 2014. Available from: https://www.linkedin.com/today/post/article/20140709143113-215478825-can-patientscollaborate-without-open-access [39] Christensen J, Wilson J. Congressional hearing investigates Dr. Oz “miracle” weight loss claims. June 19, 2014. Available from: http://www.cnn.com/2014/06/17/health/senate-grills-dr-oz/ http://www.cnn.com/2014/06/17/health/senate-grills-dr-oz [40] Hartsell C. Watch John Oliver verbally pants Dr. Oz over dietary supplements. Huffington Post. June 23, 2014. Available from: http://www.huffingtonpost.com/2014/06/23/john-oliver-dr-ozdietary-supplements_n_5521413.html [41] Ern AM. The use of gamification and serious games within interventions for children with autism spectrum disorder: a systemic review 2014. Available from: http://essay.utwente. nl/64780/1/Ern,%20A.M.%20-%20s1079581%20(verslag).pdf [42] Wilson AS, McDonagh JE. A Gamification model to encourage positive healthcare behaviours in young people with long term conditions. EAI Endorsed Transactions on Serious Games, 04 2013–05 2014;(1)2, e3. [43] Tufekci Z, Wilson C. Social media and the decision to participate in political protest: Observations from Tahrir Square. Journal of Communication 2012;62(2):363–79. [44] Scheurer D, Choudhry N, Swanton KA, Matlin O, Shrank W. Association between different types of social support and medication adherence. American Journal of Managed Care 2012;18(12):e461–7. [45] Hibbard JH, Stockard J, Mahoney ER, Tusler M. Development of the Patient Activation Measure (PAM): conceptualizing and measuring activation in patients and consumers. Health Services Research 2004;39(4p1):1005–26. [46] op. cit., Fogg. [47] Mitchell DJ, editor. Toward a definition of information therapy. Proceedings of the Annual Symposium on Computer Application in Medical Care. American Medical Informatics Association; 2014. [48] AMA Code of Medical Ethics. Opinion 10.2 – patient responsibilities. Issued June 1994. Update June 2001. Available from: http://www.ama-assn.org/ama/pub/physician-resources/medicalethics/code-medical-ethics/opinion1002.page? [49] Williams G, Hamm MP, Shulhan J, Vandermeer B, Hartling L. Social media interventions for diet and exercise behaviours: a systematic review and meta-analysis of randomised controlled trials. BMJ Open 2014;4(2):e003926. [50] op. cit., Hamm, et al. [51] op. cit., Lynch, et al.

4 Perspectives on Old and New Media   

   73

[52] Fordis M, Street Jr RL, Volk RJ, Smith Q. The prospects for web 2.0 technologies for engagement, communication, and dissemination in the era of patient-centered outcomes research: selected articles developed from the Eisenberg conference series 2010 meeting. Journal of Health Communication 2011;16(sup1):3–9. [53] New York Times. The Well guide to activity trackers. [Updated 18 Jun 2014] Available from: http://well.blogs.nytimes.com/projects/activity-trackers [54] Rozenblum R, Donzé J, Hockey PM, Guzdar E, Labuzetta MA, Zimlichman E, et al. The impact of medical informatics on patient satisfaction: a USA-based literature review. International Journal of Medical Informatics 2013;82(3):141–58. [55] Rozenblum R, Bates DW. Patient-centred healthcare, social media, and the internet: the perfect storm? BMJ Qual Saf. 2013;22(3):183–186.

Asaf Bitton, Michael Poku and David W. Bates

5 Policy context and considerations for patient engagement with health information technology Abstract: This chapter lays out major policy issues relevant to patient engagement and health information technology. We discuss how and why patient engagement has become an increasingly large focus of both federal and private initiatives to rein in costs and improve quality, focusing on relevant sections of the Affordable Care Act as well as the rise of consumerism in the healthcare marketplace. We then highlight seven promising tools with policy-relevant dimensions in this area: personal health records; patient-reported outcomes; team-based care; telemedicine; mobile health applications; shared decision-making; and publicly available data. We conclude by examining opportunities for policy development in this area and potential development in the commercial vendor markets.

5.1 Introduction Considerable evidence continues to build demonstrating the benefits of having patients more attuned to and responsible for their health and health care. As a result, the notion of patient engagement has increasingly been recognized as a critical driver of value in the healthcare delivery system. However, studies have demonstrated “a lack of consistency in terminology and definitions” as they relate to the concept of patient engagement.[1] While it’s popularly held that patient engagement is a key component of any strategy to achieve the Triple Aim of better health, better care and lower costs, there are many definitions of this concept, which may vary depending in part on the stakeholder. Health Affairs’ Patient Engagement Health Policy Brief published in 2013 describes the concept of patient engagement as interventions geared toward driving positive patient behaviors and enhancing patient activation. Patient activation is a concept describing a patient’s skills, confidence, self-efficacy, education, and willingness to actively participate in his or own health care and wellness.[2] There is increasing recognition among a broad array of healthcare stakeholders (government, patient representatives, academics, provider organizations and payers) that patient engagement is an imperative for improvement within the healthcare system at both the individual and societal level because of the potential benefits in terms of improved outcomes and lower costs.

76   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

5.1.1 Why do policy makers care about the intersection of patient engagement and Health Information Technology? With spiraling costs, mediocre quality and inadequate coverage for millions of Americans, the U.S. healthcare system has been described as being in a perpetual state of crisis. Nonetheless, major shifts in the underlying dynamics over the last decade have fundamentally begun to change the playing field for this fragmented and complex system. The recession at the end of the first decade of the 21st century catalyzed a reduction in healthcare cost growth still seen to this day. Federal governmental responses to the recession took many forms, including notably the creation of the Electronic Health Record (EHR) incentive program (eventually to become known as meaningful use) meant to catalyze adoption of EHRs. After a slow start, provider adoption of EHRs, especially by hospitals and primary care physicians, has increased rapidly and catalyzed the formation, and consolidation, of a robust EHR vendor community in the U.S. Shortly thereafter, a monumental, once-in-a-generation political battle erupted over healthcare reform that encompassed last-minute congressional votes, Supreme Court challenges, political casualties on both sides, the creation of new state and federal delivery reform innovation centers and mechanisms, a flawed technical rollout (through Healthcare.gov) and eventually a surprisingly robust insurance exchange enrollment uptake. Despite many political and legal machinations, health reform and insurance expansion through the Affordable Care Act (ACA) appear likely to be ensconced as a new reality in American society. Moreover, a consensus is emerging that meaningful engagement of patients with their own health care, through the promotion of more integrated, patient-friendly and information-connected systems of care, is crucial to the entire underpinning of reform in the US. Further, the convergence of funding and policy streams for focusing on increasing the use of information technology within health care only add impetus for critically thinking about how to best promote patient engagement within this context through levers at multiple layers – clinical interactions, community participation, policy focus and governmental funding. Section 3021 of the ACA sets out the creation of the Center for Medicare and Medicaid Innovation (CMMI), which was specifically tasked with the mission to identify, test, evaluate and scale new models of care as supported by CMS. These innovations in payment and service delivery were structured to try to produce better health care, as measured by traditional technical quality measures as well as experiential patient measures, along with better health and reduced costs. The success of various innovations in care such as accountable care organizations, patient-centered medical homes and bundled payments in part hinges upon how patients perceive their impact on their own successful engagement with the healthcare system. Further, CMMI sponsored large multi-stakeholder engagement strategies like the Partnership for Patients to further advance the goal of meaningfully bringing patients into the process of health delivery reform. The Office of the National Coordinator (ONC) for

5 Policy perspective   

   77

Health IT has also deeply focused on the impact of EHR adoption and expansion on patients’ experience and engagement in their care. Thus, the stars finally appear to be aligning for addressing this issue.

5.2 The current situation: needs, gaps and challenges 5.2.1 What is the impact of consumerism on patient engagement? Positives and drawbacks?

Soon after the managed care backlash in the late 1990s, consumerism in health care gained a foothold and has been on the rise ever since. A consumer-driven framework for health care essentially means putting the patient in the driver’s seat by giving them more of the decision-making capacity and cost burden, simultaneously promoting transparency, choice and competition.[3] Surveys have demonstrated that a majority of U.S. consumers want to use health information technology to afford themselves greater participation in their health care and wellness.[4] Many providers and payers are also convinced by evidence that suggests that more engaged patients are more likely to have better health outcomes and exhibit more robust self-management and preventive practices – all potentially at a reduced costs.[5, 6] According to the Kaiser Family Foundation, 20% of covered workers in the U.S. had consumer-driven health plans (i.e. a high deductible health plan paired with a health savings option) in 2013, up from 4% in 2006.[7] Studies suggest that these insurance products can lead to lower utilization and reduced spending.[8] The private sector has also participated in the healthcare consumerism trend; Castlight Health is a San Francisco-based company founded in 2008 with the goal of empowering people to make better healthcare decisions and achieve better outcomes through cost and quality transparency and other mechanisms. In 2014, Castlight held its initial public offering and, after the first day of trading, the market valued the company at more than $3 billion. Despite the potential opportunity consumerism brings to the healthcare industry, this trend also carries significant risks. For one, there is the question of information asymmetry: what health data are patients able to process (and have access to) in ready and actionable ways, and what negative consequences may result if patients have more responsibility for their health and health care without the necessary health literacy, education and activation.[9] When patients are forced to make choices based on costs, evidence demonstrates that they do not always make ones that will be good for their health; for example, when the number of prescriptions covered by insurance was arbitrarily limited, patients were just as likely to discontinue essential medications like insulin as ones with limited or no demonstrated effectiveness.[10, 11] The potential for adverse effects of cost sharing has been demonstrated even with

78   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

modest increases in co-payments.[12] Moreover, given that there arguably is not yet a robust and widely adopted healthcare value framework (i.e. meaningful healthcare outcomes as they relate to costs), consumerism in health care could lead to detrimental outcomes such as worsened disparities by education status or adverse selection. Serious conflicts could arise without a constructive framework that is shared by providers, patients and policymakers to ensure that all stakeholders are aiming to maximize value.[13] Further, a principal enabler of consumerism in health care and care quality improvement is transparency of healthcare costs and outcomes. By providing tools and mechanisms to allow for such transparency, patients can preferentially seek treatment from high-value providers and providers can work to improve their care delivery quality. For example, the Massachusetts Health Quality Partners collaborated with Consumer Reports in order to provide patients with reliable quality data for each primary care practice group in the state. The aim of this program is to equip patients with meaningful information to drive better decision-making when it comes to choosing one’s primary care provider, and to lift the overall quality of care in the state by allowing providers to compare themselves with their peers and competitors. From a health policy perspective, this initiative has forced policymakers to think about what type of data may be meaningful to patients and providers to reach these stated goals (e.g. is patient satisfaction data enough? Are proxy measures needed? Outcome measures?), as well as spur policymakers to think about what level of granularity is necessary to be actionable for patients and meaningful to drive quality improvement (e.g. data at the practice level, health system level and individual provider level). Similar initiatives to drive transparency in outcomes/quality have been implemented with variable success in other states such as New York and Pennsylvania. In the case of New York, the state began requiring hospitals to disclose mortality rates from certain heart procedures, and the results were initially robust. Deaths from heart surgery in the state fell by more than 40% in the first four years of this initiative and at least one hospital chose to fire one of their surgeons who was unable to improve on this metric.[14] These benefits occurred even though patients appeared to be relatively unaware of the disclosed data. This case study is important for health policy makers for a few reasons. First, it demonstrates the powerful impact of reporting in terms of driving providers to manage themselves toward excellence, even if patients pay little attention early on. This case also highlights the point that it may not be possible to “lift all boats” (without eliminating consistent underperformers) when it comes to provider quality of care, and if this is the case, policymakers must consider the ramifications of this reality. In the case of Pennsylvania, the Pennsylvania Health Care Cost Containment Council, a state agency, has published the Consumer Guide to Coronary Artery Bypass Graft Surgery for many years.[15] These reports detail the number of coronary-artery bypass graft (CABG) procedures performed in a calendar year by hospital and by

5 Policy perspective   

   79

surgeon, in addition to the associated risk-adjusted mortality. Early studies have demonstrated that while a majority of cardiologists were aware of the Consumer Guide, only 10% or so discussed the Guide with their patients or said the Guide influenced their referral patterns in any way.[16] Furthermore, surveyed cardiologists also felt that the Guide adversely impacted CABG access to the most severely ill patients, while a majority of surveyed cardiac surgeons reported that the Guide made them less willing to operate on severely ill patients needing CABG procedures. These contradictory findings are likely due, in part, to various social science and monetary policy principles such as Campbell’s Law and Goodhart’s Law; the former states, "The more any quantitative social indicator (or even some qualitative indicator) is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor”.[17] In other words, a key quality measure should not be used for both high-stakes accountability and improvement at the same time. Goodhart’s Law teaches that “any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes”.[18] It can be argued that the CABG Guide ceased to be a reliable measure as soon as mortality was published and cardiac surgeons were deemed to be judged on this metric; the mortality measure was subject to corruption as cardiac surgeons may have sought to deny the sickest patients CABG procedures (even though risk-adjustment was incorporated in the Guide). These case studies serve as an important caution to policy makers before they implement rules and regulations governing quality and outcomes transparency in health care. More generally, data has suggested that patients themselves have paid more attention to reputation of providers than performance, although this may change as more data become available.[19] Increasingly, too, patients are paying a great deal of attention to non-traditional sources such as social media ratings and internet commentaries about providers and healthcare organizations.[20]

5.2.2 What tools can we use to engage patients? How are they furthered or hampered by current policies? Many different health information technology tools exist at various stages of adoption, and each can be influenced by wise policies to promote their future effective use and spread. Given the appropriate policy environment, these HIT tools have the potential to enhance the value of healthcare delivery and population health by promoting sustained patient engagement. The promise of these tools is supported by some encouraging tailwinds: increased wiring of hospitals and clinics; continued advances in broadband connectivity/speed; and increased ownership of smartphones.[21–23] The HITECH Act and meaningful use have already had profound effects, and patient engagement was a key consideration in the development of meaningful use to date, though adoption levels of some of relevant tools such as patient portals is still modest.

80   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

In the following sections, we briefly review these tools and highlight areas of policy intersection and clarification that may be key to their successful implementation and widespread use.

5.2.2.1 Personal health records (PHR)/patient health portals Patient portals are web-based solutions that allow patients to securely and remotely access their personal health information, thus bolstering their involvement in the ongoing care management process and increasing their sense of ownership in their care.[24, 25] Many of these portals also allow patients to communicate securely with members of their care team, request refills for prescriptions, schedule clinic appointments, access educational material and pay medical bills. From a policy perspective, patient portals represent one of the prime avenues to engage and empower patients on an ongoing and sustaining basis. Meaningful Use Stage 2 requires a minimum level of use for patient portals (i.e. 5% of patients must utilize their health data) – and providers pushed back even against this minimal level. But patient demand seems to be far greater than this, and providers will have to catch up – according to one study, more than half of U.S. adults considered access to patient portals important, even though far fewer currently had such access.[23] The intent of Meaningful Use Stage 2 was to get providers “on the escalator” of sequential HIT adoption – the hope is that once this occurs, patient demand will take over, though there are also risks – for example, that the PHRs that providers start with will include inadequate functionality, or that a torrent of incoming messages could ensue that providers would not want to answer. Further, there are a host of sociotechnical issues surrounding the effective use of patient portals.[26] At present, no broad set of national standards exists to govern patient portals’ features, structure or data display; thus, a wide range of variability exists across vendor products. Patients may have to navigate various portals if their providers practice within different health systems, and learning one may not help with the next. There are also substantial variations in portal quality. Some portals have user-friendly interfaces and excellent synchronicity with the patient’s medical records, but others suffer from poor design, clunky interfaces and ineffective transmission of information into the main EHR. Policy developments in this area could include being more prescriptive about certain areas of patient portal development and use. These areas include data standards, health literacy regulations, multiple language display, requirements on ancillary features for portals and interoperability standards to transmit and merge health data among various provider-tethered portals and between a patient-tethered portal and a provider-tethered portal. Certification of portals should be considered, and policies that enable open comparisons among portals that consumers can evaluate may be helpful, though because the most successful architecture so far appears to be portals that are “tethered” to the provider EHR, it is not clear how important patient choice will be.[27]

5 Policy perspective   

   81

Another important challenge from the viewpoint of policymakers is how best to use HIT to engage not just the patient, but also the patient’s family and caregivers. Approximately 66  million people serve as unpaid family caregivers and many are untrained. They can feel unsupported and unrecognized by providers and the health system.[28, 29] No widely recognized standards exist to govern patient portal access on behalf of others and current meaningful use policy have not yet addressed the critical roles caregivers play in the healthcare system.[30, 31] Significant legal and ethical issues arise around the best, and safest, ways to allow family and caregivers to access patient portals on behalf of their loved ones, especially for children and the elderly, although studies are now going on which may help inform this area. Furthermore, while the use of patient portals is growing, there is a dearth of research pertaining to their specific impact on health outcomes, utilization or cost; research is also lacking in terms of the impact of patient portals in different contexts. [32–34] Major issues exist around whether, and how, certain portal features may improve or exacerbate the “digital divide” between certain groups of patients. While recent research has documented that this divide is not inevitable, efforts to ensure language concordance, appropriate literacy levels and highly interactive tools for those with limited visual proficiency are needed to make sure that portal reach is optimized in the community.

5.2.2.2 Patient-reported outcomes Patient-reported outcomes (PROs) are another opportunity to engage patients through HIT. A PRO has been defined by the FDA as “any report of the status of a patient’s health condition that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else”.[35] As the healthcare system continues to move in the direction of patient-centered care, PROs (which often capture information that give clinicians a more accurate, longitudinal sense of how the patient is managing his/ her illness) have become much more important data points to capture and incorporate in ongoing care management. Bennett and colleagues describe a typology for the use of PROs to include the following areas: (1) inputting patient perceptions of their care or health; (2) longitudinal care monitoring; (3) complex disease management or care coordination; (4) vital sign and symptom tracking outside the clinic; and (5) alerting platform to communicate symptom changes or a need for care.[36] Promoting electronic PROs (e-PROs) are particularly important to policymakers as electronic collection has significant advantages over paper administration including remote, longitudinal access and the minimization of data entry errors.[37, 38] Despite these potential benefits, there are still significant barriers preventing providers from engaging patients with e-PROs, many of which can be addressed, in part, with thoughtful health policy. National quality standard-setting bodies in the U.S. (like NQF, NCQA and CMS) are only in the initial stages of working through standards for PROs. Moreover, much work needs to be done in both the provider and patient

82   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

communities to increase the potential for acceptance of these measures as a new frontier in quality measurement, despite the current dissatisfaction with most HEDIS and challenges with EHR-based quality measures. Another concern with PRO implementation is the current lack of widely adopted standards in terms of how to collect and transmit PROs, in addition to how/when to integrate this information into a patient’s EHR; moreover, all of these activities must be done in a secure and private manner.[39] The degree of provider liability and accountability for PROs must be made plain, and instructions for how clinical teams ought to process, analyze and incorporate PROs must be established. In many cases, core measurement issues (i.e. what constitutes an average baseline for a given population, and what amount represents a clinically significant change) have yet to be worked out for some measures. Policies must be developed by regulatory bodies to clarify these accountability issues in a legal and clinical sense while the measures develop a further evidence and experiential base. There is also much debate among providers as to what specific patient-reported outcomes add value to ongoing care management – and how to best entice providers to participate and contribute to this new stream of data.[40] Data ownership issues also must also be resolved to promote widespread adoption. Implementation research is urgently needed to produce evidence on what types of PROs, with what interfaces, have the best chance for driving end outcomes. Federal funding agencies can provide important impetus and seed funding for these initiatives.

5.2.2.3 Team-based care Team-based care approaches represent another avenue to bolster patient engagement and self-management with smartly deployed HIT. Team-based care models have become increasingly more prominent as the health delivery system of silos continues to recede in lieu of medical homes, medical neighborhoods and accountable care organizations. The appropriate HIT infrastructure will be a powerful enabler of the team-based care model and population health management approaches. Policymakers have already highlighted the importance of team-based care as a key element in the Patient Protection and Affordable Care Act (e.g. Medicare Shared Savings Program, multiple primary care practice transformation demonstrations projects, and Bundled Payment Initiatives). There is overwhelming consensus that the American healthcare system desperately needs to rein in costs and begin delivering consistent value to patients. The marked complexity of delivering health care that yields reliable highquality outcomes means that effective team-based care approaches will continue to evolve over time as we discover which modalities and interventions add the most value. Policymakers will continue to take interest in ensuring that providers use the most efficacious and evidence-based HIT to promote care coordination. There are still gaps in terms of best practices for authenticating members of a care team, defining roles, distributing assignments, creating and managing a common plan of care, and interacting with care team members of disparate health systems.[41]

5 Policy perspective   

   83

Policymakers will also have to take care to balance the need to establish standards in these practices with the need to maintain enough flexibility to allow for innovation. In addition, health policy must continue to drive the most effective payment/reimbursement models for healthcare services as well as work with the Accreditation Council for Graduate Medical Education (ACGME), American Medical Association (AMA), the Association of American Medical Colleges (AAMC) and other stakeholders to reform medical education and training to ensure that future generations of physicians are adequately trained in delivering care in teams and HIT skills. Research is needed to establish what tools will be helpful within teams – while it is clear that the EHRs of today do not deliver them, what they should look like is uncertain and the vendors are too busy today responding to the current demands of meaningful use to innovate in this area.[42] Finally, in order to make these multidisciplinary teams a reality, policy makers will also have to tackle the issue of healthcare workforce shortage and interprofessional training, especially with respect to physicians and nurses.[43]

5.2.2.4 Telemedicine Telemedicine can be defined as providing care at a distance over a digital network. Common telemedicine solutions include: two-way video conferencing, typically used for e-visits and e-consults; store and forward/interpret telecommunication services, often used to make a diagnosis; remote monitoring equipment, utilized to managed chronic diseases; and some forms of secure messaging. Telemedicine technologies bolster patient engagement by affording patients additional opportunities to interact with members of their care team outside of the traditional visit. The expanded access can in some cases lead to improvements in overall care quality and significant cost-savings.[44] Because of the opportunity to seize the rare combination of costefficiencies and expanded access, policymakers have invested more of their energy promoting policies to expand telemedicine implementation. According to the National Conference of State Legislatures, 43 states and the District of Columbia provide reimbursement for telemedicine services for their Medicaid program; nineteen states and the District of Columbia require private insurance plans in the state to reimburse telemedicine services.[45] The Centers for Medicare and Medicaid Services has also made strides in the area of telemedicine, recently expanding the geographic areas where telemedicine services can be covered and agreeing to pay for certain chronic disease management services that are not conducted face-toface.[46] Despite these advances, there are still barriers to realizing the full value of telemedicine services: health insurers fail to reimburse various telemedicine services at parity with traditional face-to-face visits in at least 30 states, fewer than 20 states reimburse for home telemedicine services, and fewer still reimburse for remote monitoring services.[47] Moreover, the state-by-state telemedicine licensure model further impedes the implementation of these solutions. Some states impose certain encumbrances for telemedicine activities, such as requiring special licenses for physicians

84   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

wishing to engage in telemedicine across state boundaries; studies have demonstrated that these measures curb the adoption of telemedicine services.[48] Legislation to decrease the burden of practicing telemedicine, such as the proposed TELEMED Act of 2013 (HR 3077), which legalizes the treatment of Medicare patients across state lines via telemedicine, is necessary to promote adoption of these cost-effective solutions. Health policy research must further elucidate the specific telemedicine services (and associated medical conditions) for which the most value is added, as well as clarify how best to handle issues around provider accountability/liability.

5.2.2.5 Apps/mHealth In her 2011 keynote address at the 2011 Annual mHealth Summit, former US Health and Human Services Secretary, Kathleen Sebelius, called mobile health care (mHealth) "the biggest technology breakthrough of our time [being used] to address our greatest national challenge".[49] There is great opportunity in this area given the deep penetration of smartphones and other mobile platforms. Thus, it is likely that mobile applications geared towards empowering patients will help patients take a larger role in managing their health and wellness. There are already an estimated 40,000 health related apps readily available.[50] Understandably, policy makers have taken an interest in these solutions. In 2013, the FDA issued guidance regarding the agency’s oversight of mobile medical applications (MMAs) focusing on functionality and the risks these devices may present to patients. With these guidelines in place, the FDA will not verify that apps undergo evidence-based peer reviews nor will the FDA ensure that apps meet a certain level of data security.[51] Numerous questions pertaining to the FDA’s role in this space still abound: Does the FDA need additional legislative tools to regulate this industry? Will the FDA be able to keep up with the rapid and ever-changing technology industry? How will privacy/security issues be assuaged? Does purported efficacy of applications need to be verified by the FDA? How will software updates and security issues be addressed? There is also a question of how medical applications will fit into the wider HIT ecosystem. The Office of the National Coordinator for Health Information Technology (ONC) is supporting some research to address this issue. For example, Substitutable Medical Apps – the reusable technologies (SMArt) program – aims to ultimately create an “app store” for electronic health records, i.e. develop an underlying programming interface system allowing for the development and use of downloadable apps, similar to the current smartphone ecosystem. A key will be how such apps will be able to seamlessly interface with electronic health records, both via extracting information through “APIs” or application programming interfaces, and by putting coded information back in. This has historically been difficult, but the major vendors are making more information available via APIs, and predictably, the app developers are jumping at this opportunity.

5 Policy perspective   

   85

One major gap is that the apps of today tend to focus more on wellness than on patients with chronic illnesses who are most expensive and might benefit most. In addition, the functionality available does not necessarily address the needs of the elderly, underserved or those disadvantaged in some other way. More research is also needed in this area to address the potential benefits of apps.

5.2.2.6 Shared decision-making Shared decision-making (SDM) refers to the process of meaningful engagement between providers and patients in making complex decisions where the clinical evidence is either not clear or within preference-sensitive conditions. Emerging evidence suggests that decision aids implemented within certain conditions such as arthritis surgery, prostate cancer and back pain treatment can reduce costs, improve outcomes and improve decision quality as viewed by patients and providers.[52, 53] It is also clear that implementing SDM into clinical practice encompasses serious logistical challenges, especially surrounding provider buy-in.[54] Section 3056 of the ACA calls for the establishment of a shared decision-making resource center to facilitate better implementation of SDM tools into clinical practice. So far, government appropriations to fund this center have not been made. Given the promise of SDM, a logical next step should be to appropriate the funds for this center either through existing agencies like AHRQ, ONC or CMMI, or through a new agency. A number of CMMI demonstration projects such as the Comprehensive Primary Care initiative are actively focused on implementing SDM into clinical practice (including into EHRs for active use). These tests will provide valuable evidence about how, when and who should best implement SDM.

5.2.2.7 Promotion of publicly available quality data Another key is that information technology should eventually enable patients to access a wide array of publicly available data about costs, quality and outcomes of care. An early target of meaningful use has been access to quality data, but the quality metrics available so far are quite limited in scope, as noted earlier. Access to information for patients about the costs of care is still in its infancy. Recent evaluations have demonstrated that physicians have great difficulty answering questions for patients, like how much commonly used orthopedic devices cost.[55] Eventually, though, greater transparency can be expected to have major benefits, especially as access to new types of data, such as patient-reported outcomes after specific surgical procedures, becomes available.

86   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

5.3 Future Opportunities 5.3.1 What does this all mean? Patient engagement will clearly be a key area for innovation, creativity and focus within the health delivery system and the associated HIT industry over the next decade. Many people consider patient engagement to be the “secret sauce” that will promulgate future success toward achieving the triple aim, especially as technology tools to promote and integrate it into existing platforms proliferate. Two potential scenarios can be imagined regarding the trajectory of market and policy forces in this area (as policy levers continue to encourage HIT adoption). First, large vendors (who are gaining market share in a consolidating marketplace) may find themselves unable to rapidly meet the needs of a changing and increasingly wired consumer and provider base in the area of patient engagement. Their platforms may just be too ossified to allow the wide array of creativity and free thinking occurring in this marketplace to be captured and implemented well. As the mode of competition shifts, this could end up opening the marketplace up for a wide array of new approaches and lowercost vendors that can better match tools to needs.[56] Alternately, large vendors may evolve more rapidly and responsively, providing base platforms off of which these new tools can operate and succeed, thus further cementing the position of a few large vendors within the marketplace.

5.3.2 Policy Reform to Date and in the Future Meaningful use has clearly been a success in terms of promoting basal adoption of electronic health records both inside and outside the hospital. However, its impact in terms of improving care and value broadly remains to be defined, and it has notably had only limited success to date with respect to promoting interoperable clinical data exchange. While it has already included incentives to get early-stage functionality to promote patient engagement started, the future directions are sufficiently uncertain that little specificity has been so far around, for example, what structure personal health records should contain. Moreover, even Stage 2 has been controversial, and there have been calls to slow down, pause or narrow the scope. The draft recommendations for Stage 3 of meaningful use are already out for comment, so the broad outlines this is likely to take are clear. While this will be a step forward with respect to patient engagement, it is likely to be a modest one – vendors and providers are already scrambling to keep up. The future of meaningful use – for example whether there will be a Stage 4 or 5 – is still uncertain. If there is, this can be one valuable tool for pushing vendors to include and providers to adopt new changes that will enable patient engagement. Furthermore, future standards may begin to clarify important regulatory and legal questions around patient portals and patient-

5 Policy perspective   

   87

reported outcomes. Certification is likely to persist and is one attractive approach for ensuring that applications include specific functionality interoperability, requirements particularly in the areas of patient portals and bi-directional patient-facing registries for chronic conditions. But from the policy perspective, perhaps the greatest near-term need is much more support for research exploring the potential benefits of personal health records, patient-reported outcomes, tools to enable team-based care, telemedicine, mobile apps, shared decision-making and publicly-available outcome data. Relatively little research funding is currently being directed to these areas, and much more is needed if we are to realize the benefits of the investment in HITECH. If healthcare reform proceeds as expected and accountable care becomes widely prevalent, this need will become even more immediate.

5.4 Conclusion Patient engagement through HIT is an enigma: everyone is talking about it as a lynchpin of healthcare reform, but a lack of consensus on unified definitions, approaches and policy prescriptions for it still exists. In this chapter, we lay out a current state overview of development in the areas of consumerism, patient portals, patientreported outcomes, team-based care tools, telemedicine, mHealth apps and shared decision-making. Promising pilots and tools currently exist to promote the goals of patient engagement and point to the possibility of real value creation. Yet, few of these programs and tools have scaled, and major regulatory and measurement issues exist to be solved before further roll-out. We argue that a policymaker focus on the intricate set of new challenges that patient-facing portals, measures, apps and decision tools offer is urgently needed to promote improvement in this area. Furthermore, interoperability is a common challenge across every data collection and measurement platform; one that urgently needs to be addressed in coming meaningful use and vendor developments. Finally, without the space and resources for large-scale research efforts in the public, private, and academic arenas, the promise of these platforms is unlikely to be fulfilled.

References [1] [2] [3]

Gallivan J, Burns KK, Bellows M, Eigenseher C. The Many Faces of Patient Engagement. J Participat Med. 2012;4. Hibbard JH, Greene J, Overton V. Patients With Lower Activation Associated With Higher Costs; Delivery Systems Should Know Their Patients’ “Scores”. Health Aff. 2013;32(2):216–22. Robinson JC. Managed Consumerism In Health Care. Health Aff. 2005;24(6):1478–89.

88   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

[4] Keckley PH, Coughlin S, Eselius L. 2011 survey of health care consumers in the United States: key findings, strategic implications. Washington (DC): Deloitte Center for Health Solutions; 2011. [5] Hibbard JH, Greene J. What the Evidence Shows about Patient Activation: Better Health Outcomes and Care Experiences; Fewer Data on Costs. Health Affairs 2013;32(2):207–14. [6] Hibbard J, Mahoney E, Stock R, Tusler M. Do increases in patient activation result in improved self-management behaviors? Health Serv Res. 2007;42(4):1443–63. [7] Kaiser Family Foundation. 2013 Employer Health Benefits Survey. August 2013. [cited 2014 Dec 10]. Available from: http://kff.org/report-section/ehbs-2013-section-8 [8] Buntin MB , Damberg C, Haviland A, Kapur K, Lurie N, McDevitt R, et al. Consumer-Directed Health Care: Early Evidence about Effects on Cost and Quality. Health Affairs 2006:w516–w530. [9] Prey JE, Woollen J, Wilcox L, Sackeim AD, Hripcsak G, Bakken S, et al. Patient engagement in the inpatient setting: a systematic review. Am Med Inform Assoc. 2014 Jul–Aug;21(4):742–50. [10] Soumerai SB, Avorn J, Ross-Degnan D, Gortmaker S. Payment Restrictions for Prescription Drugs under Medicaid. New England Journal of Medicine. 1987;317(9):550–6. [11] Goldman DP, Joyce GF, Zheng Y. Prescription drug cost sharing: associations with medication and medical utilization and spending and health. JAMA. 2007;298(1):61–9. [12] Goldman DP, Joyce GF, Escarce JJ, Pace JE, Solomon MD, Laouri M, et al. Pharmacy benefits and the use of drugs by the chronically ill. JAMA. 2004;291(19):2344–50. [13] Sommers R, Goold SD, McGlynn EA, Pearson SD, Danis M. Focus groups highlight that many patients object to clinicians’ focusing on costs. Health Aff. 2013;32:338–346. [14] Lauer C. The Shameful State of Our Hospitals. [cited 2014 Dec 10]. Available from: http://www. beckershospitalreview.com/quality/chuck-lauer-the-shameful-state-of-our-hospitals.html [15] Schneider EC, Epstein AM. Influence of Cardiac-Surgery Performance Reports on Referral Practices and Access to Care — A Survey of Cardiovascular Specialists. New England Journal of Medicine. 1996;335(4):251–6. [16] Leatherman S, McCarthy D. Public disclosure of health care performance reports: experience, evidence and issues for policy. International Journal for Quality in Health Care 1999:11(2):93–105. [17] Campbell DT. Assessing the Impact of Planned Social Change. Hanover: The Public Affairs Center, Dartmouth College; 1976. [18] Chrystal KA, Mizen PD. Goodhart's Law: Its Origins, Meaning and Implications for Monetary Policy. Lecture given 12 November 2001, London. [19] Bates DW, Gawande AA. The impact of the Internet on quality measurement. Health Aff (Millwood). 2000;19(6):104–14. [20] Rozenblum R, Bates DW. Patient-centred healthcare, social media and the internet: the perfect storm? BMJ Qual Saf. 2013;22(3):183–186. [21] Smith A. Smartphone Ownership 2013. Pew Research. [cited 2014 Dec 10]. Available from: http://www.pewinternet.org/2013/06/05/smartphone-ownership-2013 [22] Akamai. 2013 Third Quarter 2013 State of the Internet Report. [cited 2014 Dec 14]. Available from: http://www.akamai.com/dl/akamai/akamai-soti-q313.pdf?WT.mc_id=soti_Q313 [23] Ricciardi L, Mostashari F, Murphy J, Daniel JG, Siminerio EP. A National Action Plan To Support Consumer Engagement Via E-Health. Health Affairs 2013;32(2): 376–384. [24] HealthIT.gov. What is a patient portal? [cited 2014 Dec 10]. Available from: http://www.healthit. gov/providers-professionals/faqs/what-patient-portal [25] Healthcare IT News. Patient Portals. Healthcare IT Index. [cited 2014 Dec 10]. Available from: http://www.healthcareitnews.com/directory/patient-portals

5 Policy perspective   

   89

[26] Ancker JS, Miller MC, Patel V, Kaushal R; HITEC Investigators. Sociotechnical Challenges to Developing Technologies for Patient Access to Health Information Exchange Data. Journal of the American Medical Informatics Association. 2014 Jul-Aug;21(4):664–70. [27] Kaelber DC, Jha AK, Johnston D, Middleton B, Bates DW. A Research Agenda for Personal Health Records (PHRs). J Am Med Inform Assoc. 2008;15(6):729–36. [28] Adelman RD, Tmanova LL, Delgado D, Dion S, Lachs MS. Caregiver Burden: A Clinical Review. JAMA. 2014 Mar 12;311(10):1052–60. [29] Lilly MB, Robinson CA, Holtzman S, Bottorff JL. Can we move beyond burden and burnout to support the health and wellness of family caregivers to persons with dementia? Evidence from British Columbia, Canada. Health Soc Care Community. 2012 Jan;20(1):103–12. [30] Sarkar U, Bates DW. Care Partners and Online Patient Portals. JAMA. 2014;311(4):357–358. [31] Lynn J. Strategies to Ease the Burden of Family Caregivers. JAMA. 2014;311(10):1021–1022. [32] Goldzweig CL, Orshansky G, Paige NM, Towfigh AA, Haggstrom DA, Miake-Lye I, et al. Electronic Patient Portals: Evidence on Health Outcomes, Satisfaction, Efficiency, and Attitudes, A Systematic Review. Annals of Internal Medicine 2013;159(10):677–687. [33] Ammenwerth E, Schnell-Inderst P, Hoerbst A. The impact of electronic patient portals on patient care: a systematic review of controlled trials. J Med Internet Res. 2012;14(6):e162. [34] Otte-Trojel T, de Bont A, Rundall TG, van de Klundert J. How outcomes are achieved through patient portals: a realist review. J Am Med Inform Assoc. 2014 Jul-Aug;21(4):751–7. [35] U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER), Center for Devices and Radiological Health (CDRH). Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims. Guidance for Industry. 2009 Dec. [cited 2014 Dec 10]. Available from: http://www.fda.gov/downloads/Drugs/Guidances/ UCM193282.pdf [36] Bennett AV, Jensen RE, Basch E. Electronic patient-reported outcome systems in oncology clinical practice. CA Cancer J Clin. 2012;62(5):337–347. [37] Deshpande PR, Rajan S, Sudeepthi BL, Abdul Nazir CP. Patient-Reported Outcomes: A New Era in Clinical Research. Perspectives in Clinical Research. 2011;2(4):137. [38] Shapiro M, Johnston D, Wald J, Mon D. Patient-Generated Health Data. RTI International. [cited 2014 Dec 10]. Available from: http://www.healthit.gov/sites/default/files/rti_pghd_ whitepaper_april_2012.pdf [39] National eHealth Collaborative. Patient-Generated Health Information Technical Expert Panel FINAL REPORT. 2013 Dec [cited 2014 Dec 10]. Available from: http://www.healthit.gov/sites/ default/files/pghi_tep_finalreport121713.pdf [40] Bitton A, Onega T, Tosteson A, Haas J. Toward a Better Understanding of Patient-Reported Outcomes in Clinical Practice. Am J Manag Care. 2014;20(4):281–283. [41] Krist AH, Beasley JW, Crosson JC, Kibbe DC, Klinkman MS, Lehmann CU, et al. Electronic Health Record Functionality Needed to Better Support Primary Care. J Am Med Inform Assoc. 2014 Sep-Oct;21(5):764–71. [42] O’Malley AS, Grossman JM, Cohen GR, Kemper NM, Pham HH. Are Electronic Medical Records Helpful for Care Coordination? Experiences of Physician Practices. J Gen Intern Med. 2010;25(3):177–85. [43] Kane GC, Grever M, Kennedy JI, Kuzma MA, Saltzman AR, Wiernik PH, et al. The anticipated physician shortage: meeting the nation’s need for physician services. Am J Med. 2009;122(12):1156–62. [44] Grabowski DC, O’Malley AJ. Use Of Telemedicine Can Reduce Hospitalizations Of Nursing Home Residents And Generate Savings For Medicare. Health Affairs 2014;33(2):244–50.

90   

   Part I: Information Technology for Patient Empowerment: 360° Perspectives

[45] National Conference of State Legislatures. State Coverage for Telehealth Service. [cited 2014 Dec 10]. Available from: http://www.ncsl.org/research/health/state-coverage-for-telehealthservices.aspx [46] Kvedar J, Coye MJ, Everett W. Connected Health: A Review Of Technologies And Strategies To Improve Patient Care With Telemedicine And Telehealth. Health Affairs 2014;33(2):194–99. [47] Johnson J. Expanding Access to Telemedicine. The Hill. 2014 Mar. [cited 2014 Dec 10]. Available from: http://thehill.com/blogs/congress-blog/healthcare/200493-expanding-access-totelemedicine [48] Adler-Milstein J, Kvedar J, Bates DW. Telehealth Among US Hospitals: Several Factors, Including State Reimbursement And Licensure Policies, Influence Adoption. Health Affairs 2014;33(2):207–15. [49] PWC. Emerging mHealth: Paths for Growth. 2012. [cited 2014 Dec 10]. Available from: http:// www.pwc.com/en_GX/gx/healthcare/mhealth/assets/pwc-emerging-mhealth-full.pdf [50] IMS Institute for Healthcare Informatics. Patient Apps for Improved Healthcare: From Novelty to Mainstream. IMS Institute for Healthcare Informatics website. [cited 2014 Dec 10]. Available from: http://www.imshealth.com/deployedfiles/imshealth/Global/Content/Corporate/ IMS%20Health%20Institute/Reports/Patient_Apps/IIHI_Patient_Apps_Report.pdf. October 2013. [51] Powell AC, Landman AB, Bates DW. In Search of a Few Good Apps. JAMA. 2014 May 14;311(18):1851–2. [52] O’Connor AM, Wennberg JE, Legare F, Llewellyn-Thomas HA, Moulton BW, Sepucha KR, et al. Toward The “Tipping Point”: Decision Aids And Informed Patient Choice. Health Aff. 2007;26(3):716–25. [53] Stacey D, Légaré F, Col NF, Bennett CL, Barry MJ, Eden KB, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database of Systematic Reviews. 2014 Jan 28;1:CD001431. [54] Friedberg MW, Busum KV, Wexler R, Bowen M, Schneider EC. A Demonstration Of Shared Decision Making In Primary Care Highlights Barriers To Adoption And Potential Remedies. Health Aff. 2013;32(2):268–75. [55] Okike K, O’Toole RV, Pollak AN, Bishop JA, McAndrew CM, Mehta S, et al. Survey Finds Few Orthopedic Surgeons Know The Costs Of The Devices They Implant. Health Aff. 2014;33(1):103–9. [56] Christensen CM, Raynor ME. The Innovator’s Solution: Creating and Sustaining Successful Growth. Boston: Harvard Business School Press; 2003.

Part II: Current and Future Information Technology Solutions for Patient Empowerment

Mary Jo Deering and Cynthia Baur

6 Patient portals can enable provider-patient collaboration and person-centered care Abstract: Patient portals are multi-function, multi-purpose technologies offered by providers as a means of engaging patients in their care. The technical functions, policies and uses of portals are in flux, and insufficient evidence makes it difficult to predict portals’ effectiveness in improving patient care and health outcomes. Even with some concerns about information privacy and confidentiality as well as portal usability, patients report that portals offer many advantages over the current ways they have of sharing information and communicating with providers. Providers who have experience with patient portals, such as using secure messaging or receiving patient-generated data, are generally pleased with their contributions to care. Optimally, portals should reflect the needs and interests of patients as well as providers, but many portals currently reflect constraints of legacy electronic health record systems and professional perspectives of what patients need. Although patients appreciate some basic features that strengthen their ability to connect with their providers, issues of health literacy and usability pose barriers to portal use across age, race, ethnicity and literacy boundaries. This chapter examines portals’ value for specific purposes and identifies strategies that providers, portal developers, policy makers and other stakeholders, including patients, can use to improve them. No single formula exists for a successful portal, and they are at an early enough stage of development that they may yet mature into tools for collaboration and person-centered care.

6.1 Introduction “Patient portals” are evolving, multi-function and multi-purpose technologies responding to changing conditions in technology, policy and patients’ and providers’ expectations and experiences. Portals comprise a collection of online tools and resources offered by providers as a service to their patients and a means of involving them more directly in the healthcare process. Portals may allow patients and their authorized caregivers to: view personal health information, such as lab results, immunizations or medications; access educational resources; update administrative information, such as insurance provider and contact information; communicate with providers through secure messaging; complete transactions such as requesting prescription refills or scheduling appointments; or a number of other services.[1]

Author’s disclaimer: The findings and conclusions in this chapter are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention (CDC).

94   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Patient portals are responding to two distinct trends. First, healthcare providers increasingly seek to connect with patients outside the clinical visit because of policy and payment changes, an aging population with increasing health management issues and patients’ demand for a more digitally enabled healthcare experience. Second, the dynamic information technology sector is creating new capabilities and expectations on the part of patients, and perhaps providers. With supportive policies, attention to implementation “bumps in the road” and ongoing research and evaluation, portals can be better tools for clear, effective patient-provider communication and collaboration, an important direction for healthcare system change and personcentered health care.[2, 3] Healthcare systems that work collaboratively with patients in the design and adoption of portals will likely have greater success in attracting patients to use the portals in ways that advance overall patient engagement and empowerment. All these changes in technologies, policies, needs and expectations make it likely that, over time, providers and patients will connect in different ways that affect both health goals and outcomes.

6.1.1 Background “Patient portals” and “personal health records” (PHRs) are sometimes used interchangeably, but they differ in purpose and function. The concept of PHRs appeared over a decade ago as a counterpart to provider-controlled electronic health records (EHRs). The purpose of PHRs is to enable consumers and patients to bring together information from multiple providers and share it with whomever they wanted, thereby helping individuals take a greater role in managing their health and care.[4–6] A large number of PHR developers exist. Some sell their PHRs to providers for their portals; others sell or give free access to consumers, patients and caregivers.[7] With the passage of the Health Information Technology for Economic and Clinical Health (HITECH) Act in 2009, federal policy began to incentivize the “meaningful use” of electronic health records. To increase patient and family involvement in health care, Meaningful Use requirements began to promote patient and provider use of secure messaging and electronic access to information. The new Meaningful Use Stage  2 requirements for patients to be able to view, download and transmit their information make these portal functions necessary and hence more widely offered. In 2013, KLAS research found that Meaningful Use Stage 2 has pushed patient portals from a “niceto-have technology to a must-have technology”.[8] Vendors usually include portals as components of their EHR systems, often with a PHR. This integration has the advantage of making it easy for providers to activate portals but also reduces the providers’ ability to select or create portals tailored to their patients’ needs. There are no definitive surveys of patients’ use of provider portals. Anecdotal reports suggest that ongoing usage, not just registering for an account, is very low.[9]

6 Portals to enable provider-patient collaboration   

   95

Lack of utility is often cited as a reason. At the end of 2012, fewer than one in five Americans reported having their own PHR.[10] Some reasons given for PHRs’ failure to diffuse may be applicable to patient portals: inconvenience; design shortcomings; the inability to share information across organizations; users’ computer competency; limited Internet access; and limited health literacy.[11] Additionally, the reluctance of providers to use PHRs, even their own, has been a significant factor in lack of uptake.[12]

6.1.2 Portals and Patient Engagement and Empowerment Provider organizations, employers and policy makers often perceive of portals, as well as “patient engagement” and “patient empowerment” strategies, as means to the end of getting patients to take greater responsibility – both in terms of participating in decision-making and paying a larger share of costs – for their health and health care. As Goldzweig and colleagues note, “Portals are being created as part of a movement to make patients more active participants in their care”.[13] Because “patient engagement” and “empowerment” are often ambiguous or undefined terms in the published literature, the following working definitions inform this chapter. It should be noted that patients may involve their caregivers in any of these activities, or in the case of minors or patients incapacitated in some way, caregivers may be more directly involved than patients. “Engagement” reflects behaviors.[14] Patients are engaged in their health when they are aware of and communicate about their health, make informed health decisions on their own or with their providers, contribute to health transactions and processes, such as refilling prescriptions or monitoring the quality of care and providing feedback, are active in carrying out health activities at home and act in healthpromoting ways, such as stopping smoking or adding physical activity to daily routines. This engagement may, and optimally should, lead to a partnership between the patient and provider who work together toward shared goals and outcomes. Empowerment reflects capacity; it is a health promotion outcome that can alter the basic determinants of health.[15] Patients may feel empowered or able to speak up and take action to promote health or manage a condition, which may lead to engagement and partnerships, or they may become empowered as a result of engagement with providers and health tasks and productive partnerships. Patients may experience engagement without empowerment and empowerment without engagement. Communication, or the interactive process of developing common understanding, is the unifying thread among engagement, empowerment and partnership because they all require patients and providers to be attentive, express themselves effectively, verbally and in writing, and exchange information to achieve health-related goals.

96   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

6.2 Current situation In 2013, about 50% of hospitals and 40% of physicians had some type of patient portal technology.[16] Nearly 2500 companies were identified in the new health technology space at the end of 2013, of which 51% offered consumer-facing products and services. Another 31% offered professional-facing products and services, 8% offered patient-provider communication products and services and 10% provided data, analytics and information exchange products and services.[17] The market segments showing the fastest growth were self-management tools and trackers, clinician workflow tools, data utility layer technology and patient-provider communication. In general, the number of tools blending elements for both the providers and consumers is increasing most rapidly. As of 2014, core portal functions include the ability for patients to view personal health information such as lab results, immunizations or medications; access educational resources; update administrative information such as insurance provider and contact information; communicate through secure messaging; and complete transactions such as requesting prescription refills or scheduling appointments.[1, 13] Less frequently available features allow patients to: enter or update medical and family history; enter or update medications, allergies and conditions; view visit summaries; use tools for tailored health management, including reminders; complete registration forms; view accounts and pay bills; and keep a health journal. Some delivery systems that are consciously trying new ways to engage their patients are giving them the ability to view provider notes, and this trend is expected to accelerate.[18, 19] Some portals incorporate a distinct PHR, where the personal health information and other data are stored separately from the provider’s EHR.[20] Some of these PHRs can receive patient data not only from a single provider, but also from other providers such as laboratories and pharmacies that use the same vendor system.[21] Increasingly, portals accept data from patients in digitally seamless ways, strengthening two-way communication through secure messaging or direct input to questionnaires or to a patient’s personal health record. A small number of provider systems can receive data from monitoring devices – most frequently those that have been requested or prescribed by the physician for key clinical values such as blood pressure and blood sugar. Although there is increasing discussion about the potential value of accepting information from consumer health tracking devices, technical and policy issues still pose barriers to the widespread deployment of this function.

6.2.1 Key portal features: interest and evidence 6.2.1.1 Data access Patients and their caregivers cannot be fully informed without knowing essential clinical data. The federal government is promoting patients’ access to their data through multiple policies and programs. Meaningful Use Stage 2 requires participating provid-

6 Portals to enable provider-patient collaboration   

   97

ers to enable patients to view, download and transmit their information. The federal Office of the National Coordinator for Health IT (ONC) and the White House have jointly sponsored annual Patient Access Summits to promote access and explore how patients can use their own electronic health information to achieve better health and health care.[22] The Blue Button Initiative, a public-private activity, builds on a Veterans Administration tool that allows patients to view and download their data, and an increasing number of providers and payers are implementing it.[23, 24] Some organizations are pioneering access to providers’ notes. The Open Notes study, conducted in 2010 at three major healthcare systems, offered patients access through an online portal to notes written and posted by their providers during recent visits. The findings showed that patients of all ages were enthusiastic about being able to read the notes, a large majority reported increased medication adherence, and very few felt confused, worried or offended by what they read.[19] Providers experienced very low to moderate impact on their workflow, and all chose to continue to post their notes after the study ended.[19] Based on positive results from its own pilot evaluation, the VA is now offering access to all notes to all patients.[25, 26] Providers increasingly support patient access to their data. A 2013 provider survey found that 49% felt that giving patients access to their data is critical for effective care; 65% believed patients should have some access to their records and 31% believed they should have full access.[27] However, the slowness of extending access is reflected in the finding that only 21% currently offered their patients access to their medical summary or clinical chart, which are the most basic clinical data.

6.2.1.2 Receiving patient-generated information There is growing evidence that patient portals may be especially effective in improving patient engagement, care and outcomes when patients can contribute, not just view, data.[28] “Patient generated health data” (PGHD), sometimes called patientreported or patient-initiated information, are data that come from the patient outside the clinical visit. Case studies at three healthcare systems demonstrated the utility of using patient-reported measurement systems to improve healthcare outcomes and value, and illustrated the feasibility of using patient-reported measures in typical clinical settings.[29] Geisinger Health System conducted a pilot that showed patient feedback on gaps, errors or updates to their online medication data improved the accuracy of the medication lists, and medication feedback improved patient engagement, communication and information sharing.[30] Based on evidence from portal research, public policy and programs are encouraging the acceptance of patient initiated data. Stage 3 Meaningful Use draft recommendations include the optional objective of enabling patients to submit provider-requested information through online questionnaires and secure messaging.[31] ONC has prepared guidelines to help providers understand the value and feasibility of accepting PGHD in ways that can address their concerns and those of their patients.[32, 33]

98   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

6.2.1.3 Secure messaging Secure electronic messaging differs from regular e-mail in that messages are encrypted; a user name and password are required to initiate or read the message. Secure messaging is an important tool for two-way communication. Providers can promote care coordination between visits, handle routine health issues, address patient questions and concerns, monitor patient conditions and help patients better track and report their conditions. Patients can communicate with their provider in an unpressured setting that fits their schedule, avoiding the frustration of trying to reach their provider by phone. One survey of “e-mail use”, which did not differentiate secure messaging from plain e-mail, found that, as of March 2013, only 28% of providers were using e-mail. [34] Significantly, the same survey found that 45% of “opinion leader” providers used it. Providers’ age was not a dramatic differentiator. A study of providers from 21 medical groups that were using secure messaging found it was widely perceived to be a safe, effective and efficient means of communication that improves patient satisfaction and saves patients time.[35] The authors concluded, however, that practice redesign and new payment methods will help electronic communication be more widely used in patient care. Secure messaging will be stimulated by policy developments in coming years. Stage  2 Meaningful Use includes an objective that providers use secure messaging to communicate relevant health information to their patients, and the draft stage 3 recommendations expand the use of secure message to include receipt of patientinitiated information.[29, 36]

6.2.1.4 Interactive Tools Many portals include a range of tools and resources to help patients maintain their health, reduce health risks and manage chronic diseases. These tools have the potential to strengthen patients’ self-management of their health, especially in partnership with providers.[37] Health risk assessments or other questionnaires are often promoted as an entry tool to help individuals understand how diet, exercise, smoking, alcohol and drug use, family history and other health-related components of their lives may influence their likelihood of disease or injury. The results may trigger links to tailored educational modules or an individual action plan. Some portals offer online tracking of specific health parameters and may include rewards or points for positive trends.

6.2.1.5 Links to authoritative information Looking for health information is one of the most widespread uses of the Internet. In 2013, 59% of U.S. adults reported they had looked online for information about a range of health topics in the past year; 35% said they have gone online specifically to try to

6 Portals to enable provider-patient collaboration   

   99

figure out what medical condition they or someone else might have.[38] However, health professionals often state concerns about the validity of the information. Offering “curated” information that has been developed and updated by reputable health education companies, non-profit organizations and public agencies is valuable for helping patients find accurate information.

6.2.2 Problems and challenges with patient portals The problems and challenges with portals reflect long-standing and complex issues in health care, policy, technology and social inequalities. Three literature reviews conclude there is insufficient evidence that portals enhance patient engagement, patient satisfaction, care utilization, efficiency or outcomes.[3, 13, 39] However, they found that examples of positive impact tended to occur when patients and providers used multiple functions, such as secure messaging, interactive decision-support, health-related reminders or case management. The reviews noted that the variation in functionality made it hard to link any findings (positive or negative) to the effect of specific functionalities. One of the reviews emphasized that the lack of information about implementation processes and provider context made it hard to determine factors responsible for the results and even more difficult to replicate success.[13] Case management appears to increase the chances of significant positive health outcomes, making it a confounding factor in portal implementation studies. An ongoing digital divide, meaning access to the Internet and associated technologies, has consequences for equitable access to portals.[40] The divide has narrowed in some ways over time but access disparities by age, income and education remain. [41] Some studies show women may be more willing to use portals than men, and whites are more frequent users than black, Hispanics or other racial/ethnic groups. [13, 42–44] Although people with less education or limited health literacy skills may express more concerns about how to use technologies in health care, researchers find that these individuals will value and use them if someone explains the tools to them, they perceive the tools are helpful and they receive encouragement and support registering for portals and completing tasks.[42, 43, 45, 46] Differences in the vocabulary and health concepts that patients and providers use with each other contribute to major health literacy problems in healthcare service delivery.[47] Healthcare organizations often do not recognize and adequately meet the information and communication needs of people with limited health literacy skills, and technology may make this situation worse.[42, 48] In addition to miscommunication, some patients with access to providers’ notes expressed concerns about the derogatory language or apparently contradictory information that providers use in their online notes.[25] Providers used to writing notes for their own use or to share with colleagues may need to recognize that portals provide a larger “audience” for patient information and will have to adjust their language accordingly.

100   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

The ability of healthcare organizations and technology companies to develop and deploy health information technologies that secure personal health information is an ongoing concern. A few studies have found that some patients are worried about portals’ impact on the privacy of their personal health information, but these concerns weren’t significant enough to stop them from using portals.[19, 49] The researchers hypothesized that the benefits patients derived from knowing more about their health issues and interacting with their providers outweighed the perceived risks of having their information shared or exposed in unauthorized ways.[19, 50] However, unauthorized – or even just unexpected – data uses can damage patients’ trust. Other security problems pertain to unauthorized access to portals. Some provider organizations require in-person identification and strong passwords. Although it is important to have policies and procedures for identifying patients initially and for routine log-in requirements, security experts caution against having such strong requirements that they become barriers to patient participation. Recommendations from the federal government’s advisory Health IT Policy Committee urge the Department of Health and Human Services to develop guidance on best practices, observing that “no one size fits all”.[51] The recommendations state that while in-person identification provides the strongest protection, rural and elderly people need remote methods. Though not mentioned, this would also apply to people with mobility disabilities and limited transportation options. The committee also recommends that the government “strongly encourage” providers to use more than just a user name and password, and encourage them to look to methods used in online banking, which are familiar to most people now. Increasingly, these security issues will apply to patients’ caregivers. Caregivers’ portal access is an issue that policymakers and clinical organizations have yet to address fully. The inability to integrate data from different sources is a major barrier to optimum portal use. Patients have multiple providers over time, and many see more than one health professional in any given year. Increasingly, patients will want their information shared with mobile devices and applications, raising additional policy and technology issues, including security and interoperability. Under fee-for-service payment, providers have limited incentives to promote information sharing and connect with consumer devices. The proliferation of proprietary portals will do little to reduce the problems patients and their caregivers experience in trying to share information with multiple providers and create a holistic view of their health. Standards to ensure that health data can be seamlessly shared, integrated and presented in an understandable fashion are not yet fully developed or implemented. Standards are most important in four areas: how applications interact with users (such as the portal interface); how systems communicate with each other (such as messaging standards); how information is processed and managed (including information exchange); and how consumer devices integrate with other systems (such as EHRs) and applications (such as smart phones and tablets).[52] Standards for the transmission of a simple summary document exist. Meaningful Use Stage 2 includes the requirement that providers (profes-

6 Portals to enable provider-patient collaboration   

   101

sionals and hospitals) provide a summary of care when transitioning a patient to another provider, and that providers enable their patients to “view, download, and transmit” [emphasis added] their information. Both policy and technology must evolve further so that sharing of a wider range of information is feasible. Despite these problems and challenges, patients highly value transactional portal functions, such as scheduling appointments and renewing prescriptions, and also perceive portals as means to better relationships with their providers.[19, 25, 53] Qualitative studies of patients’ experiences indicate they want to work with their providers to better understand their care and participate in decisions.[25, 53] Even without interoperability, patients indicate they appreciate portals for making it easier to print out health summaries to hand carry to other providers.[25, 53] In short, patients appear grateful for even the minor improvements portals introduce. There could be many reasons for patients’ optimistic attitudes, but one explanation fits the argument of this chapter. Many patients are eager to know at least some of what their providers know, and portals are the first healthcare innovation that facilitates information sharing and communication, both of which open the door to collaboration. Depending on the available functions, portals can show the same information to both sides, and patients’ own words show they think this information sharing “levels the playing field” with their providers.[25, 53] The preliminary evidence suggests portals are at the early stages of helping patients learn more about their health, increase their interaction with providers and reduce some of the hassle in healthcare transactions. These changes are steps on the pathway toward patient engagement and empowerment. The potential for portals lies with a patient-driven approach with patients contributing their knowledge, experience, values and preferences to the healthcare process, and patients participating in equal and collaborative partnerships with providers.

6.3 Addressing patient portal challenges The rapidly expanding installation of electronic medical records with built-in patient portals means that comprehensive solutions to problems and challenges will likely be applied to next-generation portals, not the current crop in hospitals, clinics and doctors’ offices. If already-installed portals have poor usability, jargon-filled content and not enough features valued by patients, there may be little providers can do until they change their medical records systems. However, the suggestions that follow can help providers understand why patients do and don’t use portals, adjust their policies and procedures to reduce barriers to portal use and prepare for future technology investments. Our suggestions focus on creating “continuous learning loops” for portals so that patients’ and providers’ needs and experiences inform current and future portal iterations. This multi-method approach to continually improve portals includes environ-

102   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

mental assessments, user-centered design, techniques to create easy-to-understand content and routine surveys and other feedback mechanisms. The methods and tools to create these learning loops are available to researchers, policymakers, technology vendors and providers. Researchers and vendors can collaborate on easy-to-read content and easy-to-use interfaces and functions. Additional activities, also described below, are needed to fill technology and policy gaps.

6.3.1 Environmental assessments for portals In the health literacy field, health and education professionals collaborate with actual and potential patients to assess health literacy barriers and helpers in hospital, clinic, pharmacy and doctors’ office environments.[54, 55] These barriers might include jargon-filled signs and directions, staff who don’t communicate clearly and lengthy, technical and confusing forms and policy documents. Less frequently, assessments may find environmental aspects that help patients, such as on-demand interpreter services. Healthcare organizations can extend this method to assess elements of their portal environment that get in the way or help patients. For example, do the policies and procedures for authenticating patients and caregivers make it easy or difficult to sign up? Do providers recommend portal use to their patients? Are human helpers available to show patients how to use the portal? This environmental assessment could be especially effective for those organizations that have fewer patients than expected signing up for or using portals or find that some patient groups use portals less often.

6.3.2 User-centered design User-centered design is a well-established approach based on a body of research and practice that aims to create technology that better fits the needs and capabilities of the people who will use the technology, or “the users”.[56] User-centered design also can reduce development costs because design mistakes are discovered and corrected early in the process. Researchers and technology developers who use user-centered design begin with the end-users’ perspective and keep that perspective foremost throughout the process so that the end result is the best fit between technology and users’ needs. Studies indicate a user-centered design approach and usability testing would greatly improve the quality and appeal of patient portals, as well as discover those design features and functions that cause users to make the most mistakes.[42, 43] Researchers and technology developers can learn a great deal about how to design the most engaging and useful portals when they take patients as partners early in the design and testing processes and use widely accepted methods such as cognitive interviews, focus groups and usability testing.

6 Portals to enable provider-patient collaboration   

   103

6.3.3 Techniques to create easy-to-understand content Limited health literacy is a well-documented problem in the U.S. and has a direct effect on patients’ capacities to understand and use all types of health information. [47] Even though portals contain personal health information, which has high relevance for the patient looking at the information, portals also are filled with jargon, unfamiliar terms and numbers, and difficult concepts such as “normal” and “abnormal”.[42] Patients and providers likely have different vocabularies for the same or similar body parts, body functions, symptoms, medicines, tests and treatments. In the future, genetic risk information will become as standard as blood pressure information is today, but this information will be virtually unusable without a concerted effort to present it in a manner that is easily understood. Teens and young adults, especially those with chronic conditions, may need extra research attention. Many contemporary teens are tech-savvy but have limited health literacy skills. As teen patients near the age of 18, organizations advise them to take more responsibility because they are supposed to take over their own care. If they haven’t been in a pediatric provider relationship that modeled effective communication, or they haven’t had a proactive parent or guardian, then young adults may not have models of what an “empowered patient” is. This combination of full responsibility, limited health literacy and possible over-reliance on technology to deliver a satisfying healthcare experience presents special challenges for helping young adults use portals as a means for building their health literacy skills and productive relationships with a range of caregivers. Many techniques exist to create and test content that patients find easier to understand, and research methods are similar to user-centered design.[57] Researchers, technology developers and health information content creators can use cognitive testing, focus groups, task-based usability tests and online feedback mechanisms such as “What questions do you have?”, “How well did this information answer your questions?”, and “What else are you looking for?” on web pages to get patients’ input on content from concept to finished product. Several tools to create easy-tounderstand content and improve patient-provider communication are free and publicly available.[58, 59] Even with these tools, the ultimate test is to involve patients throughout the content creation and dissemination process and offer opportunities for their continuous feedback and revisions.

6.3.4 Patient surveys and other feedback mechanisms Healthcare organizations already have surveys available to them that allow large numbers of patients to provide feedback on their portal experiences and preferences. The Consumer Assessment of Healthcare Providers and Systems (CAHPS) surveys are widely used in the healthcare sector and include modules for patients to report their

104   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

experiences with providers’ communication skills, the Patient-Centered Medical Home model, hospitals and health information technology use in their doctors’ offices.[60] The health information technology questions ask about secure messaging, medication refills, appointment scheduling and access to personal health information. These survey data can and should inform providers’ portal strategies. In addition, automated random feedback surveys for web sites can be added to portals for ongoing patient input.

6.3.5 Technology and policy needs The overarching need in both technology and policy is a commitment to a truly personcentric perspective in portal design and use. Vendors of EHRs and portal packages can consider research with different patient populations to inform their next generation of products. “Patient-tested” could be an effective marketing strategy. Individual tool developers can take steps to integrate functionality and interoperability with clinical systems. Standards that recognize the patient and caregiver as an integral part of the healthcare team and that support the aggregation and sharing of personal information can be given greater emphasis. These include ways to identify patients and caregivers as authors and receivers of information, patient matching across different systems and interoperability across systems and tools. Data standards that support consumer health vocabularies will help so that user interfaces, documents such as visit summaries and other content are appropriate for patients and caregivers. A number of policy efforts can contribute to person-centered technologies. ONC’s National Action Plan to Support Consumer Engagement Via e-Health lays out approaches to increase access to health information, support the development of tools that help people take action on the information and change attitudes toward the traditional roles of patients and providers.[23] ONC has also laid out a longer-term, person-centered health vision, in which individuals are supported by health information technology in managing their health and partnering in their care.[61] The patient-engagement objectives already in Meaningful Use Stage 2 can be more widely publicized and those in the draft Stage 3 recommendations can be supported. Additional public policies can build support for “consumer-mediated exchange”, already identified by ONC as one of three pillars of the national health information exchange strategy in 2012.[62] National policy work to strengthen privacy and security has been mentioned above. Provider organizations should consider steps to strengthen privacy and security related to their portals. Strong internal policies regarding staff access to and use of patient information gathered through portals and stored in PHRs can help protect patients’ information, along with policies and procedures for identification and passwords, for establishing a relationship to the patient and for patient approval of access to the portal. Providers should, at a minimum, execute a HIPAA Business Associate Agreement or similar document with their portal or PHR vendor to limit of its use or disclosure of the heath information.

6 Portals to enable provider-patient collaboration   

   105

6.4 Discussion Portals’ individual components are likely to evolve separately in response to many different forces. As this occurs, fundamental issues of patients’ and providers’ needs, goals and expectations deserve closer attention. Portals that do a poor job of meeting patients’ needs are unlikely to engage them. Both person-centered care and chronic disease self-management models recognize that individual values, preferences, and approaches to health and health care are essential to good outcomes. The design and deployment of portals can be sensitive to and adjust for the many differences between individual patients and patient populations. If portals are to help patients and their caregivers make informed decisions, it is desirable to have decision-support tools with evidence-based guidelines, risks and benefits, and options consonant with patients’ values and capacity to understand complex information. Different stakeholders can explore two tracks of research and monitoring: user-centered research that informs the development or refinement of the tools themselves, and systematic, ongoing patient input and feedback so that provider organizations hear from their own patients about what might be most effective for them. Policies to promote the aggregation, sharing, portability and continuity of personal health information across providers, states and vendors are under development, but the work is necessarily cautious to avoid unintended consequences in a time when healthcare delivery and technology are both evolving. However, even if current interoperability efforts are successful, their success does not ensure the willingness of various healthcare partners to share patient information. Health reform is helpful as it creates incentives for providers to view patients more holistically and coordinate with others, which in turn can spur vendors to develop portal resources to support patients receiving care from multiple providers. The increasing burdens on patients and caregivers to make healthcare choices and coordinate across providers will create an opportunity for resources and tools to support these complex responsibilities. A “consumer movement” in health care could be a powerful force for change. Both research and real-time commentary from diverse experts underscore that current portal technologies are not optimally effective. We currently know little about what patients expect from using portals, the usability of different portals for diverse patient populations and how to motivate more patients to use portals and sustain that use over time. Consequently, it is too soon to predict with much confidence how useful portals will be for patients and the achievement of their health goals, as well as how much effect portals will have on health-system defined goals for improved health outcomes, lower costs and more efficient service utilization. More research and evaluation are needed to find out what works for what purposes, for whom, and in what circumstances. There is currently a split between vendors of clinical systems who add patientfacing components as an adjunct to their core technology, and non-EHR-based tools

106   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

developers focused on the consumer market. The former have high sunk costs in EHR technology, which would be expensive to retool with patient-identified features, and their clients are understandably reluctant to continually modify their systems. These technologies are specific to a provider group and typically are not interoperable with other systems. Technologies coming from the consumer-oriented market are more end-user oriented, but their products tend to be divided by function and typically are not interoperable with providers’ portals. This sector is driving innovation rapidly, but the unfortunate trend is a proliferation of new stand-alone products. Moreover, the new tools are generally geared to already “wired” populations, who may not be the same as the priority populations of concern to providers and policymakers. In addition, innovative technologies oftentimes are ahead of policy and provider practices. There will likely be an extended period of experiment and iteration, with policy aiming for the “sweet spot” of enabling change and not locking in choices prematurely. Providers will likely struggle to balance policy changes driving technology purchases, their own needs to keep workflow disruptions and costs to a minimum, and the needs of their patients and caregivers. Their established workflows and habits cannot be modified quickly. New technologies may feed a cultural tension between “do no harm” and quality improvement as well as patient engagement requirements. Fortunately, the methods of patient input presented in the “Addressing challenges” section can help providers select, integrate and continually improve tools for patients in a less disruptive manner.

6.5 Conclusions Ensuring that portals support patient engagement and empowerment entails a continuum of related actions – from introducing patients to portals, to promoting collaboration between patients and providers, to using portals for person-centric care and continuously improving portals by assessing and addressing barriers to their use. Providers need to carefully consider how to optimize the portal they are implementing to best suit their patients. They should clearly define their clinical goals and understand their patients’ needs and preferences, and the functionality that would support both. Providers can use checklists and guidelines to select portals and patient education content that fit their practices and communicate the importance of using portals to their patients. They can ensure resources are available in the primary languages spoken and read by their patients. And, patients can speak up, try different portal functions and offer opinions and perspectives that help providers understand what they need. Patient portals are tools for people – patients, caregivers and providers – operating in complex environments and making often difficult decisions with uncertain outcomes. The functions, content and user interface will be most helpful to patients

6 Portals to enable provider-patient collaboration   

   107

when portals are tailored, as already happens for providers, as much as possible to the context of use, such as disease prevention and health promotion, primary care or hospital. Nevertheless, all patient portals should strive for clear communication. More research on portals as tools for communication and collaboration can help portals support a person-centered healthcare system and the best possible state of health beyond clinical walls. In the meantime, current portals enable small successes for some patients and providers that would not have happened otherwise and pave the way for new levels of collaboration and integration. Ultimately, patient tools will achieve their potential when they are as highly valued and developed elements of the electronic health information infrastructure as provider tools. Momentum is building toward holding consumers and patients accountable financially and morally for their health. Before we go further in that direction, those who influence health policy, practice and technology would do well to understand which tools and approaches enable patients and caregivers with many different needs and capabilities to become full partners in the care process.

References [1]

[2]

[3] [4]

[5]

[6]

[7] [8]

[9]

Office of the National Coordinator for Health IT (US). What is a patient portal? [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; [cited 2014 Jun 9]. Available from: http://www.healthit.gov/providersprofessionals/faqs/what-patient-portal Koh HK, Brach D, Harris LM, Parchman ML. A proposed ‘Health Literate Care Model’ would constitute a systems approach to improving patients’ engagement in care. Health Aff. 2013; 32(2): 357–367. Archer N, Fevrier-Thomas U, Looker C, McKibbon KA, Straus SE. Personal health records: a scoping review. J Am Med Inform Assoc 2011; 18:515–522. Tang PC, Ash JS, Bates DW, Overhage JM, Sands DZ. Personal health records: definitions, benefits, and strategies for overcoming barriers to adoption, J Am Med Inform Assoc 2006; 13:121–126. National Committee on Vital and Health Statistics (US). Personal health records and personal health record systems. [Internet]. Washington (DC): U.S. Department of Health and Human Services; 2006 Feb [cited 2014 Jun 9]. Available from: http://ncvhs.hhs.gov/0602nhiirpt.pdf Markle Foundation. Policy guide 9: A common framework for networked health information [Internet]. New York: Markle Foundation; [cited 2014 Jun 9]. Available from: http://www.markle. org/health/markle-common-framework/connecting-professionals/p9?page=7 AHIMA (US). MyPHR [Internet]. Chicago (IL): AHIMA. [cited 2014 Jun 9]. Available from: https:// www.myphr.com/resources/choose.aspx KLAS Research. Patient portals 2013: On track for meaningful use? [Internet]. [Orem (UT)]: KLAS Research; 2013 Aug 29 [cited 2014 Jun 9]. Available from: https://www.klasresearch.com/store/ reportdetail.aspx?ProductID=844 Wilkens S. Kevin. If you build a patient portal, why won’t they come? [Internet]. MedPage Today; 2013 Apr 13 [cited 2014 Jun 9]. Available from: http://www.kevinmd.com/blog/2013/04/ build-patient-portal.html

108   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[10] Wolters Kluwer Health. Wolters Kluwer quarterly poll: Consumerization of healthcare [Internet]. Wolters Kluwer; 2012 Dec 12 [cited 2014 Jun 9]. Available from: http://www.wolterskluwerhealth.com/News/Documents/White%20Papers/Wolters%20Kluwer%20Health%20 Consumerization%20of%20Healthcare%20Executive%20Summary.pdf [11] Kahn JS, Aulakh V, Bosworth A. What it takes: Characteristics of the ideal personal health record. Health Aff. 2009; 28:369–376. [12] Wynia MK, Williams Torres G, Lemieux J. Many physicians are willing to use patients’ electronic personal health records, but doctors differ by location, gender, and practice. Health Aff. 2011; 30:266–273. [13] Goldzweig CL, Orshansky G, Paige NM, Towfigh AA, Haggstrom DA, et al. Electronic patient portals: Evidence on health outcomes, satisfaction, efficiency, and attitudes: a systematic review. Ann Intern Med. 2013; 159(10):677–87. [14] Gruman J, Holmes-Rovner M, French ME, Jeffress D, Sofaer S, Shaller D, et al. From patient education to patient engagement: Implications for the field of patient education. Patient Education and Counseling. 2010; 78(3):350–356. [15] Nutbeam D. Health literacy as a public health goal: a challenge for contemporary health education and communication strategies into the 21st century. Health Promot. Int. 2000;15(3): 259–267. [16] Frost and Sullivan. U.S. patient portal market for hospitals and physicians: Overview and outlook 2012-2017 [Internet]. Mountain View (CA): Frost and Sullivan. 2013 Sep 26 [cited 2014 Jun 9]. Available from: http://www.frost.com/prod/servlet/press-release. pag?docid=285477570 [17] Health 2.0. 2013 Annual Report [Internet]. Health 2.0; 2013 Dec [cited 2013 December 27]. Available from: http://www.health2con.com/shop/market-intel/health-2-0-annual-report [18] OpenNotes. Who is Sharing Notes? [Internet] [Cited 2014 Dec 3]. Available from: http://www.myopennotes.org/who-is-sharing-notes [19] Delbanco T, Walker J, et al. Inviting patients to read their doctors’ notes: A quasi-experimental study and a look ahead. Ann Int Med. 2012;157(7):461–470. [20] EPIC Corporation. Frequently asked questions about MyChart [Internet]. [Cited 2014 Dec 3]. Available from: https://www.mychartweb.com/MyChart/default.asp?mode=stdfile&option=faq [21] Cerner Corporation. What is CernerHealth? [Cited 2014 Dec 3]. Available from: https://www.youtube.com/watch?v=EMOcybM9lKs [22] Ricciardi L.Building momentum: Expanding patient access to medical records [Internet]. Washington (DC) U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; 2013 Jul 1 [cited 2014 Jun 9]. Available from: http://www.healthit.gov/ buzz-blog/electronic-health-and-medical-records/building-momentum-expanding-patientaccess-medical-records [23] Ricciardi, L, Mostashari, F, Murphy J, Daniel J, Siminerio, EP. A national action plan to support consumer engagement via e-Health. Health Aff. 2013; 32(2); 376–384. [24] Office of the National Coordinator for Health IT (US). About Blue Button [Internet]. Washington (DC); U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT. [cited 2014 Jun 9]. Available from: http://www.healthit.gov/patients-families/ blue-button/about-blue-button [25] Woods SS, Schwartz E, Tuepker A, Press NA, Nazi KM, Turvey CL, et al. Patient experiences with full electronic access to health records and clinical notes through the My HealtheVet personal health record pilot: Qualitative study. J Med Internet Res. [Internet]. 2013 [cited 2014 Jun 9]; 15(3) e65. Available from: http://www.jmir.org/2013/3/e65/ doi:10.2196/jmir.2356

6 Portals to enable provider-patient collaboration   

   109

[26] Open Notes. VA introduces new and enhanced features for VA Blue Button [Internet]. 2013 Jan 29. [Cited 2014 Dec 3]. Available from: http://www.myopennotes.org/va-introduces-newand-enhanced-features-for-va-blue-button [27] Accenture. Patient access to electronic health records [Internet]. New Orleans (LA): Accenture; 2013 Mar 4 [cited 2014 Jun 9]. Available from: http://newsroom.accenture.com/article_display. cfm?article_id=5668 [28] National eHealth Collaborative. Patient generated health data technical expert panel final report [Internet]. Washington (DC): National eHealth Collaborative; 2013 December 17 [cited 2014 Jun 9]. Available from: http://www.healthit.gov/sites/default/files/pghi_tep_ finalreport121713.pdf [29] Dartmouth Institute for Health Policy and Clinical Practice. Using patient-reported information to improve health outcomes and health care value: Case studies from Dartmouth, Karolinska and Group Health [Internet]. Lebanon (NH): Dartmouth Institute for Health Policy and Clinical Practice 2012 Jun [cited 2014 Jun 9]. Available from: http://tdi.dartmouth.edu/images/uploads/ tdi_tr_pri_ia_sm.pdf [30] Dullabh, P et al. Demonstrating the effectiveness of patient feedback in improving the accuracy of medical records. Unpublished report presented to the U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; August 2012. [31] HIT Policy Committee (US). Meaningful Use Workgroup Stage 3 Recommendations [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; 2014 Apr 1 [cited 2014 jun 9]. Available from: http://www.healthit. gov/facas/sites/faca/files/HITPC_MUWG_Stage3_Recs_2014-04-01.pdf [32] Deering, MJ. Issue brief: Patient-generated health data and health IT [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; 2013 Dec 20 [cited 2014 Jun 9]. Available from: http://www.healthit.gov/sites/ default/files/pghd_brief_final122013.pdf [33] National Learning Consortium. Patient-generated health data: Fact sheet [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; 2014 Mar [cited 2014 Jun 9]. Available from: http://www.healthit.gov/sites/default/ files/patient_generated_data_factsheet.pdf [34] Cognac B. Doctors remain slow to use email with patients [Internet]. Kantara Media; 2013 May 8 [cited 2014 Jun 9]. Available from http://www.kantarmedia-healthcare.com/doctors-remainslow-to-use-email-with-patients [35] Bishop, TF, Press, MJ, Mendelsohn, JL, Casalino, LP. Electronic communication improves access, but barriers to its widespread adoption remain. Health Aff. 2013; 32(8): 1361–1367. [36] Office of the National Coordinator for Health IT (US). Achieve meaningful use stage 2: Use secure messaging [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; [cited 2014 Jun 9]. Available from: http://www.healthit.gov/providers-professionals/achieve-meaningful-use/core-measures-2/ use-secure-electronic-messaging [37] Office of Disease Prevention and Health Promotion (US). Expanding the Reach and Impact of eHealth Tools [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion; 2006 [cited 2014 Jun 9]. Available from: http://www.health.gov/communication/ehealth/ehealthtools/pdf/ehealthreport.pdf [38] Pew Internet & American Life Project. Health and technology in the U.S, 2013 [Internet]. Washington (DC): Pew Internet & American Life Project; 2013 Dec [cited 2014 Jun 9]. Available from: http://www.pewinternet.org/Commentary/2013/December/Health-Report-Roundup.aspx [39] Ammenwerth E, Schnell-Inderst P, Hoerbst A. The impact of electronic patient portals on patient care: a systematic review of controlled trials. J Med Internet Res [Internet]. 2012 Nov 26 [cited

110   

[40] [41]

[42]

[43] [44] [45]

[46]

[47] [48]

[49]

[50]

[51]

[52]

[53]

[54]

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

2014 Jun 9]; 14(6):e162. Available from: http://www.jmir.org/2012/6/e162/ doi:10.2196/ jmir.2238 Yamin CK, Emani S, Williams DH, Lipsitz SR, Karson AS, Wald JS, et al. The digital divide in adoption and use of a personal health record. Arch Intern Med 2011 Mar 28;171(6):568–574. Pew Internet and American Life Project. User Demographics [Internet]. Washington (DC): Pew Internet and American Life Project; 2014 Jan [cited 2014 June 9]. Available from: http://www.pewinternet.org/data-trend/internet-use/latest-stats Zarcadoolas C, Vaughon WL, Czaja SJ, Levy J, Rockoff ML. Consumers’ Perceptions of PatientAccessible Electronic Medical Records. J Med Internet Res [Internet]. 2013 Aug 26 [cited 2014 Jun 9]; 15(8)e168. Available from: http://www.jmir.org/2013/8/e168/ doi:10.2196/jmir.2507 Taha J, Czaja SJ, Sharit J, Morrow DG. Factors Affecting Usage of a Personal Health Record (PHR) to Manage Health. Psychology and Aging. 2013; 28(4):1124–1139. Lyles CR, Harris LT, Jordan L, Grothaus L, Wehnes L, Reid RJ, et al. Patient race/ethnicity and shared medical record use among diabetes patients. Med Care 2012;50: 434–440. Lu HY, Shaw BR, Gustafson D. Online health consultation: examining uses of an interactive cancer communication tool by low-income women with breast cancer. Int J Med Inform [Internet]. 2011; 80(7):518–28. Dhanireddy S, Walker J, Reisch L, Oster N, Delbanco T, Elmore JG. The urban underserved: attitudes towards gaining full access to electronic medical records. Health Expectations [Internet]. 2012 Jun 28 [cited 2014 Jun 9]. Available from: http://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3469742/doi: 10.1111/j.1369-7625.2012.00799.x Nielsen-Bohlman L, Panzer AM, Kindig DA. Health literacy: A prescription to end confusion. Washington, DC: Institute of Medicine of the National Academies; 2004. Institute of Medicine Roundtable on Health Literacy. How can health care organizations become more health literate? Workshop Summary [Internet]. Washington (DC): Institute of Medicine; 2012 [cited 2014 Jun 9]. Available from: http://www.iom.edu/Reports/2012/How-Can-HealthCare-Organizations-Become-More-Health-Literate.aspx Vodicka E, Mejilla R, Leveille SG, Ralston JD, Darer JD, Delbanco T, et al. Online Access to Doctors’ Notes: Patients Concerns About Privacy. J Med Internet Res [Internet]. 2013 [cited 2014 Jun 9];15(9):e208. Available from: http://www.jmir.org/2013/9/e208/ doi:10.2196/jmir.2670 Angst, CM. Agarwal, R. Adoption of electronic health records in the presence of privacy concerns: The elaboration likelihood model and individual persuasion. MIS Quarterly 2009; 3(2):339–370. HIT Policy Committee (US). Recommendations to the National Coordinator for Health IT on Privacy and Security [Internet].Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; 2013 May 3 [cited 2014 Jun 9]. Available from: http://www.healthit.gov/sites/default/files/hitpc_transmittal_050313_pstt_ recommendations.pdf Office of the National Coordinator for Health IT. EHR Interoperability [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; [cited 2014 Mar 9]. Available from: http://www.healthit.gov/providers-professionals/ ehr-interoperability Osborn CY, Mayberry LS, Wallston KA, Johnson KB, Elasy TA. Understanding patient portal use: Implications for medication management. J Med Internet Res [Internet]. 2013 Mar 7 [cited 2014 Jun 9];15(7):e133. Available from: http://www.jmir.org/2013/7/e133/ doi:10.2196/jmir.2589 Rudd RE, Anderson JE. The Health Literacy Environment of Hospitals and Health Centers. Harvard School of Public Health. 2006. Available from: http://www.hsph.harvard.edu/healthliteracy/files/2012/09/healthliteracyenvironment.pdf

6 Portals to enable provider-patient collaboration   

   111

[55] Agency for Healthcare Research and Quality (US). Health Literacy Universal Precautions Toolkit [Internet]. Rockville (MD): U.S. Department of Health and Human Services, Agency for Healthcare Research and Quality; 2013 Nov [cited 2014 Jun 9]. Available from: http://www.ahrq. gov/professionals/quality-patient-safety/quality-resources/tools/literacy-toolkit/index.html [56] U.S. Department of Health and Human Services. Improving the user experience [Internet]. Washington (DC): U.S. Department of Health and Human Services [cited 2014 Jun 3]. Available from: www.usability.gov [57] National Cancer Institute (US). Making Health Communication Programs Work [Internet]. Bethesda (MD): U.S. Department of Health and Human Services, National Institutes of Health; 2008 Apr 2 [cited 2014 Jun 3]. Available from: www.cancer.gov/cancertopics/cancerlibrary/ pinkbook [58] Plain Language Action and Information Network (US). Federal Plain Language Guidelines [Internet]. Washington (DC) U.S. Department of Health and Human Services; 2011 Mar [cited 2014 Jun 9]. Available from: http://www.plainlanguage.gov/howto/guidelines/FederalPLGuidelines/TOC.cfm [59] Centers for Disease Control and Prevention (US). CDC Clear Communication Index [Internet]. Atlanta (GA): U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2014 March 4; [cited 2014 June 3]. Available from: www.cdc.gov/healthcommunication/clearcommunicationindex/index.html [60] Agency for Healthcare Research and Quality (US). Consumer Assessment of Healthcare Providers and Systems (CAHPS) [Internet]. Rockville (MD): U.S. Department of Health and Human Services, Agency for Healthcare Research and Quality; [cited 2014 Jun 3]. Available from: https://cahps.ahrq.gov [61] Daniel, J, Deering MJ, and Murray M. Issue brief: Using health IT to put the person at the center of their health and care by 2020 [Internet]. Washington (DC): U.S. Department of Health and Human Services, Office of the National Coordinator for Health IT; 2014 Jan [cited 2014 Jun 9]. Available from: http://www.healthit.gov/sites/default/files/person_at_thecenterissuebrief.pdf [62] Williams, C, Mostashari, F, Mertz K, Hogin E, Atwal P. From The Office Of The National Coordinator: The Strategy For Advancing The Exchange Of Health Information. Health Aff. 2012; 31(3): 527–536.

Mary McNamara

7 Data model for integrated patient portals Abstract: The rate at which patients search for health information online continues to grow, as does the type of content available to them. Patients are currently left to pair their personal health information with other information they find online. This work examines three different methods for disseminating filtered information to patients. This chapter illustrates that different modes have benefits and drawbacks in providing content to patients that is of quality and accessible.

7.1 Introduction The phrase “patient empowerment” is common in healthcare rhetoric today. There are numerous efforts to support the shift in the healthcare paradigm from patients being passive recipients of care to active participants. There are a range of definitions for patient empowerment. This multitude of ideas demonstrates that patient empowerment is a multi-faceted concept in terms of its definition. However, one common theme across many definitions is that empowered patients are informed.[1] Informed patients is thought to be able to make more competent decisions regarding their health, and to feel more confident doing so. One growing source of information for patients is patient portals. Patient portals are applications that provide patients with direct access to medical record content. However, this content is harvested from medical records, content designed for physicians by physicians, with little change. Thus, portals often provide patients with access to content that is not written specifically for the patient, requiring the patient to translate his or her record. Patients can also consult publicly available health information, such as MedlinePlus,[2] and the Mayo Clinic website,[3] social networking sites centered around health, such as PatientsLikeMe[4] or other health websites. Patients searching for health information continue to increase,[5] and it currently ranks as the third most popular task completed online.[6] However, the quality of sources is not always readily available.[7] Even when a source is of good quality, its content may not be applicable to the user, meaning it does not meet the user's information needs. Information needs are defined here as "the recognition that their [the patient's] knowledge is inadequate to satisfy a goal, within the context or situation that they find themselves in a specific point in time".[8] Searching for health information requires patients to look through a variety of sources and determine what is relevant to their personal health situation, which has been documented as a difficult task for patients.[2, 9] Much of the difficulty lies in the differences in language used by professionals versus patients. Often quality content contains medical jargon. In order to select appropriate information and apply it to their care, patients must have a level of health

114   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

literacy. Health literacy is the ability to search for, consume, understand, reflect on and apply health information.[10] Hence, low levels of health literacy are seen as a significant barrier to accessing health information.[11] The rate of accessing digital health resources by consumers who have less financial security, have less education, are more culturally diverse and older, will increase. [3] It has been documented that exposing patients to quality health information can lead to improved outcomes. Thus, as the population accessing health resources becomes more diverse, the vocabulary gap between healthcare professionals and consumers becomes more influential.[12] Some of the best informational health contents on the web, professional medical guidelines, are intended for healthcare practitioners. As they attempt to cover the breadth of a diagnosis and treatment, these guidelines document an expanse of symptoms, test, diagnosis, treatments, clinical studies findings and additional information. This expanse of information requires an expert's knowledge to determine what portions are applicable to an individual patient, meaning a patient may not be able to determine which professional guideline content applies to him or her. Additionally, both medical content created for consumers and for professionals lack personalized content reflective of the individual's diagnosis and process of care. Patient information sources have the potential to provide patients with personalized information. Personalized health information for patients is necessary to give consumers supporting information with which to understand their health. Personalized information contrasts with general guidelines and consumer health content in that it is tailored to the individual and based on their information needs and possibly other content of theirs, such as a medical records or profiles. Thus, personalized content should also try to anticipate the patients’ information needs, or information patients would like to see. This chapter demonstrates three different methods that refine health information content to provide patients with information relevant to their diagnosis and their personal information needs and preferences. For this illustration, the domain of breast cancer will be used. However, these modes of dissemination have been used in other diagnosis domains as well.

7.2 The current situation: needs, gaps and challenges 7.2.1 Patient information needs and preferences In general, cancer patients are more positive about the idea of electronic health records and online health information than those patients without cancer, and tend to cope better with a diagnosis when they have accurate information regarding their disease and treatment.[13] Patients use health information based on their records to

7 Data model for integrated patient portals   

   115

review findings, process information and communicate with their healthcare practitioner. [14] found that cancer patients wanted to be more involved in their care, desiring to have an active role. However, [15] found that as a patient’s condition deteriorated, they tended to want less information, although still wanted to know information that was positive in nature. Yet, [16] documented that patients wanted as much information as possible regarding their diagnosis, regardless of whether the nature of it was positive or negative. Patients specifically wanted information on both diagnosis and treatment. Numerous citations show that patients are interested in diagnosis, treatment, symptoms, prognosis, survival information and side effects of treatments.[14–26] Health information preferences can also vary by demographics. Younger patients tend to search for more information on their own than older patients.[5] Those with higher levels of education tend to search more than those with less. Women tend to search more than men.[5] In terms of ethnicities, Hispanics have the lowest rates of searching for information at 44%. In comparison, 51% of African Americans search for health information and 65% of whites access digital health information.

Figure 1: Venn diagram illustrating cross section of concepts of interest

While these rates of access are likely to stay stable in the short term, just as access to health information online has increased across demographics, so has it increased within individual groups and is likely to continue to do so.[5] With the benefits in health outcomes and prevention in an educated population now documented,[27] more public and private institutions are attempting to decrease the digital divide and provide information online geared towards a variety of users. This means that content

116   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

must strive to be beneficial to users across demographic lines and yet anticipate the nuanced variation in needs between groups. As the older, less educated and those of lower socio-economic status are anticipated to increase their rate of using digital health resources, so must those resources make information not simply available but accessible, meaning that they give the user the opportunity to use the content to make health decisions. While no consumer health tool will benefit all patients without complications, digital resources must strive to be as inclusive as possible.

7.2.2 Portrait of the breast cancer patient The domain of diagnosis and the demographics associated with it can also influence information preferences. For this domain of breast cancer, it is worth noting that the disease primarily affects women, although men can also develop breast cancer. The average age for developing breast cancer is 61. White, non-Hispanic women tend to have the highest incident rate of breast cancer. However, women of various ages and all ethnicities can develop breast cancer.[28] Increases in mammogram screenings have led to a higher rate of women being diagnosed while in situ stage, meaning the cancer was less advanced. Being diagnosed at an earlier stage increases chances of survival.[28]

7.2.3 Information sources Breast cancer screening patients have numerous sources of traditional supporting information including support groups, informational sessions and brochures. General supporting health information in electronic format is now also widely available.[2–4, 24] In addition, patients' personal medical information is becoming available electronically via personal health records (PHRs) and patient portals. While both types of content are readily available, there remains a deficit in information integration in that there are few sources that merge the two types. The Health Insurance Portability and Accountability Act (HIPAA) mandates that patients have access to their medical record. While the content of their medical record is available to patients, the next step is to make it comprehensible to them. To ensure that patients understand the information contained within their medical record, patient portals cannot display the data formatted as it was captured. The content within a portal was originally documented by physicians with the intent that it be used by other medical practitioners. Although patients may want to see some of the same information as a physician, they do not necessarily want it displayed in the same way, nor will they apply it in the same context. In order to make the content understandable to consumers, changes are required in the visualization, vocabulary used and information abstraction level.

7 Data model for integrated patient portals   

   117

Usefulness of quality of a source varies by the user's expectation.[29, 30] While no tool will be able to anticipate all user needs, a needs assessment of a population is necessary to help estimate the types of information required by a specific population. Types of needs assessments include: literature reviews, surveys, focus groups and interviews.[30]

7.3 Proposed solutions 7.3.1 Tailored patient guidelines The linkage of supporting content based on medical record data is standard practice for practitioners within numerous institutions. Similar features are now also available for patients via portals and web services (i.e. MedlinePlus Connect). Although these options often use unique concept identifiers provided by the Unified Medical Language System or another ontology, incorrect and inexact matches still occur. Definitions are often context dependent: consider the difference in the definition of diaphragm in a gynecological versus thoracic domain. While a consumer health source may provide correct information on a topic, it does not demonstrate how the topic relates to the patient's screening process or their health record. Here, an existing information model generating process that was used to link guideline content and patient information needs and preferences for non-small cell lung cancer (NSCLC) screening,[31] was applied to the domain of breast cancer screening. This model was created based on the content that falls within the overlap of two domains: professional guidelines and consumer health information needs. The model content must be contained in both spheres of information in order to be simultaneously reflective of patient information needs and guideline content, with the intent that this model will be used to harvest content from professional guidelines that are reflective of patient needs and can be linked to actual concepts found within the patient’s record. First a literature review on cancer patient information needs, particularly breast cancer patient information needs, had to be conducted. Then the guidelines from breast cancer screening had to be reviewed. The literature review of patient information needs was used to define the class set, as the primary objective was to create a model relevant to patient information needs. Themes found within the literature review included the desire to hear information on diagnosis, treatment options, symptoms, diagnostic tests and common side effects of treatment.[15–17, 20–26, 32–35] Professional guidelines were used as the inspiration for candidate concepts that could be used to populate the classes (themes from the literature review). In this information model, the smallest units are concepts; concepts are the normalized instances of terms (e.g. "lump"). To create a list of concepts that would have appropriate contextual explanations for the domain of breast cancer screening, the breast cancer

118   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

screening guidelines from the National Collaborating Center for Cancer and UpToDate [36, 37] were reviewed. All concepts belong to one or more classes, the class being reflective of the common clinical feature or set of features the concepts share (e.g. "Symptoms"). With this candidate list, the information model was organized, with the constraint that concepts be included only if they are reflective of classes made evident by the literature review on patient information needs. In turn, classes could only be included if they were reflective of the steps in the screening process.

Figure 2: Simplified version of practitioner guidelines for breast cancer screening

The information model for breast cancer screening patient guidelines consists of three classes, as seen in Table 1. The Diagnosis class consists of all the possible diagnoses that can be given to a breast cancer screening patient (Tx–T4). The symptoms class does not contain every possible symptom experienced by a patient undergoing breast cancer screening. Rather, it contains just the most common, as documented in [38]. Analysis of the guidelines documented that not all patients experience symptoms, however the symptoms topic is included here, in the event that a patient who is experiencing symptoms should have access to diagnosis relevant information. The diagnostic test contains all tests that a patient may undergo during screening; this

7 Data model for integrated patient portals   

   119

was done in order to assist patients to better understand the screening process. The concepts within this model can now be linked to content from practitioner guidelines and used to annotate patient records. Concepts found within a patient’s record can be used to pull the guideline appropriate to that concept. Table 1: Model of concepts and classes. Class

Number Guideline concepts mapped to class of sources citing information need

Diagnosis

7

Tx, T0, Tis, T1, T2, T3, T4

Symptoms

3

Lump in breast, lump in underarm, thickening or swelling of breast, irritation of breast skin, dimpling of breast skin, redness in nipple area, flaky skin in nipple area, nipple discharge (other than blood), any change in shape of breast, any change in size of breast, pain in any area of breast

Diagnostic tests

4

Mammogram, ultrasound, MRI, biopsy

7.3.2 Tailoring content based on user profiles The BCKOnline portal uses metadata tags to create both metadata for information sources and for patient profiles in the domain of breast cancer,[29, 39] in order to match patients with information. To create these user-centric tags, breast cancer patients and family members were surveyed on their information needs and preferences. Demographic information on participants was also collected. The survey found that information should be tailored by precision based on a user's circumstance (e.g. stage), the user's choice in format (e.g. article, short-length) and provision of resource quality (e.g. government institution). Information on breast cancer patients' information preferences, their demographics, resource content, resource format and resource quality were then used to create specific tags that were then used to index documents. Patients were able to create a user profile based on their information preferences, demographics and specifics regarding resource content in order to retrieve documents that best fit their search, based on the metadata tags. By prioritizing what types of information they prefer, the BCKOnline portal is able to limit information overload, excluding articles that do not fit the user profile. Users create their profile by answering the multiple choice questions presented to them when they first log onto the site. The age group question determines what age range the patient falls into, as information preferences have been shown to vary by age. For instance, younger women have been shown to be more likely to want infor-

120   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

mation on fertility issues.[39] The disease stage question steers users towards information appropriate for a specific stage, while the information preference question allows the user to choose content based on a matrix of types of information (plain and brief, plain and detailed, scientific and brief, scientific and detailed). The user type question filters information by the type of user (patient, family member or friend). User testing demonstrated that the portal was well received. Patients rated the portal as easy to use, indicating that it filtered irrelevant information out and provided information patients perceived as relevant.[39]

7.3.3 Information content via social networking Breast cancer patients who are socially isolated have higher rates of mortality.[40] Alongside providing human contact, social networking sites have been shown to have similar outcomes to other more formal health interventions, discouraging unhealthy behavior and encouraging healthy behavior.[41] Social networking sites, such as PatientsLikeMe, allow for patients to connect with other patients that have the same diagnoses and symptoms.[42] These sites have been shown to have been of particular value to patients with breast cancer.[42] In addition to providing individuals with other individual's information, these sites also can permit crowd sourcing, so that a user can get a summary of many others' experiences. [43] found that breast cancer patients tended to join a social networking site to find information, but stayed as a member even after they had progressed into remission in order to support others. Breast cancer patients who are member of a social networking site for breast cancer tend to be active participants, meaning they both read and publish content.[1] This cycle of information supply among patients makes participants both the source and the recipient of the information. [43] reported that users of the social networking site were able to communicate more freely then they would in face-to-face conversation, having less fear to do so. As members of these sites, patients are allowed to remain as anonymous as they desire, but conversely they can share as much as they want to.

7.4 Discussion This chapter has shown three modes to produce personalized informational content for patients, with all three modes filtering out information based on an anticipation of what types of information will interest the patient. The method to produce tailored content taken from guidelines is based on [31], and applied here to a second domain of breast cancer screening. This method filters out content, based on the overlap that occurs between documented patient information needs and professional guidelines.

7 Data model for integrated patient portals   

   121

[29, 39] demonstrate how a user profile can be used to filter information within a portal for breast cancer. To better anticipate the user's need, the content can be filtered by filling out a short survey after logging on. Users are then linked to content based on the tags they create for themselves via the profile being matched to tags assigned to informational content. Information can also be filtered based on diagnosis in social networking sites such as in [42, 43]. Within these sites information is also filtered by patients both as users and contributors. Using the site to find information, patients can search for other profiles that resemble their own or that contain content they are interested in. As suppliers of information, users filter the type of content available based on what they want to share.

7.5 Conclusion This chapter has examined three different modes of supplying supporting information to patients who are participating in breast cancer screening or treatment. Social networking sites have been rated highly by participants and allow users a significant amount of individual control over what information they access. Content from these cites is written by other patients, making the language accessible to other users. However, concerns regarding the quality and population relevance (e.g., if it's true for one, is it true for all) have been raised. Variations in the content's ability to apply to an individual patient make these sites unsuitable as the sole source of supporting content. A user profile of criteria to match patients to content permits some preferences from individual users to be incorporated into source selection. However, information preference patterns demonstrated over a population may not hold true for an individual patient. E.g., while many women in their twenties being screened for breast cancer may want information on fertility, an individual user may not. Similarly, providing content found in medical guidelines ensures that certain concepts relevant to a diagnosis are further explained, but does nothing to consider individual patient information preferences. Further work must be done to design tools that simultaneously ensure the quality and relevance of information for an individual while accounting for that patient’s individual information needs and preferences.

122   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

References [1]

[2]

[3] [4] [5] [6] [7]

[8] [9]

[10] [11]

[12] [13]

[14] [15] [16] [17]

[18]

[19]

Van Uden-Kraan CF, Drossaert CHC, Taal E, Shaw BR, Seydel ER, Van de Laar MAFJ. Empowering processes and outcomes of participation in online support groups for patients with breast cancer, arthritis, or fibromyalgia. Qualitative Health Research. 2008;18(3). Keselman A, Slaughter L, Smith C, Kim H, Divita G, Browne A, et al., editors. Towards consumer friendly PHRs: patients' experience with reviewing their health records. American Medical Informatics Association Symposium 2007. 399–403. Zielstorff RD. Controlled vocabularies for consumer health. Journal of Biomedical Informatics. 2003;36(4–5):326–33. PatientsLikeMe. PatientsLikeMe Homepage. 2013 [cited 2013 Jan 9]; Available from: http://www.patientslikeme.com Fox S, Jones S. The social life of health information; Pew Research Center; 2009. Smith M, R. Saunders, L. Stuckhardt, J.M. McGinnis. Best Care at Lower Cost: the Path to Continuously Learning Health Care in America. Washington D.C.: Institute of Medicine; 2012. Meric F, Bernstam EV, Mirza NQ, Hunt KK, Ames FC, Ross MI, Kuerer HM, Pollock RE, Musen MA, Singletary SE. Breast cancer on the world wide web: determinants of web site popularity. Proceeds from the American Society of Clinical Oncology. 2001;20(39b). Case DO. Looking for Information: A Survey of Research on Information Seeking, Needs and Behavior. San Diego: Academic Press Elsevier Science; 2002. Tse T, Soergel, D ed. Exploring medical expressions used by consumers and the media: an emerging view of consumer health vocabularies. Proceeds from the American Medical Informatics Association: Symposium; 2003. 674–678. Services USDoHaH. HHS Releases National Plan to Improve Health Literacy. 2010 [cited 2012 Sept 9]; Available from: http://www.hhs.gov/ash/news/20100527.html Zeng QT, T.Tse, J.Crowell, G.Divita, L.Roth, A.C. Browne, editor. Identifying consumer-friendly display names for health concepts. American Medical Informatics Association Symposium; 2005;854–63. Zeng QT, T. Tse. Exploring and developing consumer health vocabularies. Journal of Medical Informatics Association. 2006;13(1):24. Beckjord E, Rechis R, Nutt S, Shulman L, Hesse B. What do people affected by cancer think about electronic health information exchange? Results From the 2010 LIVESTRONG Electronic Health Information Exchange Survey and the 2008 Health Information National Trends Survey. Journal of Oncology Practice. 2011;7(4):237–41. Davidson J, Brundage M, Feldman-Stewart D. Lung cancer treatment decisions: patients’ desire for participation and information. Psycho-Oncology. 1999;8:511–20. Butow P, Maclean M, Dunn S, Tattersall M, Boyer M. The dynamics of change: cancer patients preferences for information, involvement, and support. Annals of Oncology. 1997;8:857–63. Jenkins V, Fallowfield L, Saul J. Information needs of patients with cancer: results from a large study in UK cancer centers. British Journal of Cancer. 2001;84(1):48–51. Leydon G, Boulton M, Moynihan C, Jones A, Mossman J. Cancer patients information needs and information seeking behavior: in-depth interview study. British Medical Journal. 2000;320:909–13. Gore J, Brophy C, Greenstone M. How well do we care for patients with end stage chronic obstructive pulmonary disease (COPD)? A comparison of palliative care and quality of life in COPD and lung cancer. Thorax. 2000;55:1000–6. Murray S, Boyd K, Kendall M, Worth A, Benton T, Clausen H. Dying of lung cancer or cardiac failure: prospective qualitative interview study of patients and their carers in the community. British Medical Journal. 2002;325.

7 Data model for integrated patient portals   

   123

[20] Slaughter L, Ruland C, Rotegard A. Mapping cancer patients' symptoms to UMLS concepts. American Medical Informatics Association Symposium; 2005;699–703. [21] Clauser S, Wagner E, Aiello Bowles E, Tuzzio L, Greene S. Improving modern cancer care through information technology. American Journal of Preventive Medicine. 2011;40(5s2):s198–s207. [22] Grant R, Wald J, Poon E, Schnipper J, Gandhi T, Volk L, et al. Design and implementation of a web-based patient portal linked to an ambulatory care electronic health record: Patient Gateway for diabetes collaborative care. Diabetes Technology and Therapeutics. 2006;8(5):576–86. [23] Hess R, Bryce C, McTigue K, Fitzgerakd K, Zickmund S, Olshansky E, et al. The diabetes patient portal: patient perspectives on structure and delivery. Diabetes Spectrum. 2006;19:106–9. [24] Koch-Weser S, Bradshaw Y, Gualtieri L, Gallagher S. The internet as a health information source: findings from the 2007 health information national trends survey and implications for health communication. Journal of Health Communications. 2010;15(S3):279–93. [25] Bass S, Ruzek S, Gordon T, Fleisher L, McKeown N, Moore D. The relationship of internet health information use with patient behavior and self efficacy: experiences of newly diagnosed cancer patients who contact the National Cancer Institute's Cancer Information Service. Journal of Health Communications. 2006;11:219–36. [26] Sarkar U, Karter A, Liu J, Alder N, Nguyen R, Lopez A, et al. The literacy divide: health literacy and the use of an internet-based patient portal in an integrated health system – results from the diabetes study of northern California (DISTANCE). Journal of Health Communication. 2010;15(S2):183–96. [27] DeWalt D, Berkman N, Lohr K, Pignone M. Literacy and health outcomes. Journal of General Internal Medicine. 2004;19(12):1228–39. [28] American Cancer Society. Breast Cancer Facts and Figures 2011–2012. Atlanta: American Cancer Society; 2012. [29] McKemmish S, Manaszewicz R, Burstein F, Fisher J. Consumer empowerment through metadatabased information quality reporting: The Breast Cancer Knowledge Online Portal. Journal of the American Society for Information Science and Technology. 2009;60(9):1792–807. [30] Eysenbach G. Design and evaluation of consumer health information web sites. In: Lewis D, Eysenbach G, Kukafka R, Stravi Z, Jimison HB, ed. Consumer Health Information: Informing Consumers and Improving Health Care. NY: Springer; 2005. [31] McNamara M, Arnold C, Sarma K, Aberle D, Bui A. Method to produce a data model for personalized guidleines: an exploratory study. American Medical Informatics Association: Washington D.C.; 2014. In press. [32] Hack T, Degner L, Dyck D. Relationship between preferences for decisional control and illness information among women with breast cancer: a quantitative and qualitative analysis. Social Science Medicine. 1994;39(2):279–89. [33] Degner L, Kristjanson L, Bowman D, Sloan J, Carriere K, O'Niel J, et al. Information needs and decisional preferences in women with breast cancer. Journal of American Medical Association. 1997;277(18):1485–92. [34] Rees C. The information needs and source preferences of women with breast cancer and their family members: a review of the literature published between 1988 and 1998. Journal of Advanced Nursing. 1999;31(4):833–41. [35] Wallberg B, Michelson H, Nystedt M, C B, Degner L, Wilking N. Information needs and preferences for participation in treatment decisions among Swedish breast cancer patients. Acta Oncologica. 2000;39(4):467–76. [36] Fletcher SW. Screening for breast cancer: strategies and recommendations. 2014 [cited 2014 Feb 1]; Available from: http://www.uptodate.com/contents/screening-for-breast-cancer-

124   

[37]

[38]

[39]

[40]

[41] [42]

[43]

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

strategies-and-recommendations?source=search_result&search=breast+cancer+screening& selectedTitle=1~49 Network NCC. NCCN clinical pratice guidelines in oncology (NCCN Guidelines) breast cancer. 2013 [cited 2014]; Available from: http://www.nccn.org/professionals/physician_gls/pdf/ breast.pdf Prevention CfDCa. Symptoms of breast cancer. Atlanta: Center for Disease Control and Prevention; 2013 [cited 2014]; Available from: http://www.cdc.gov/cancer/breast/basic_info/ symptoms.htm Burstein F, McKemmish S, Fisher J, Manaszewicz R, Malhotra P. A role for information portals as intelligent decision support systems: breast cancer knowledge online experience. In: N.D. J, Gupta GAF, Manuel Mora T, editors. Intelligent Decision-making Support Systems: Foundations, Applications and Challenges: Springer-Verlag; 2006. Kroenke CH, Kubzansky LD, Schernhammer ES, Holmes MD, Kawachi I. Social networks, social support, and survival after breast cancer diagnosis. Journal of Clinical Oncology. 2006;24(7):1105–11. Johnson K, Jimison HB, Mandl KD. Consumer health informatics and personal health records. In: Shortliffe EH, Cimino JJ, editors. Biomedical Informatics. London: Springer; 2014. Frost J, Massagli M. Social uses of personal health information within PatientsLikeMe, an online patient community: what can happen when patients have access to one another's data. Journal of Medical Internet Research. 2008;10(3):e15. Shaw BR, McTavish F, Hawkins R, Gustafson DH, Pingree S. Experiences of women with breast cancer: exchanging social supprt over the CHESS computer network. Journal of Health Communication: International Perspectives. 2000;5(2):135–59.

Roy Schoenberg

8 Telehealth: connecting patients with providers in the 21st century Abstract: Technology is continuously making significant steps toward becoming a viable platform of healthcare delivery in America. Recent advances in telecommunications and healthcare connectivity have converged to create a telehealth service channel and leverage the same technologies that have revolutionized commerce in other industries. Centered around synchronous technologies that link the patient and provider in real-time through video consults, telehealth is offering both patients and providers new avenues to access and deliver healthcare services. In the wake of substantial and ongoing regulatory reform of the American healthcare industry, telehealth now provides a pivotal solution to help meet future patient demands and alleviate acute provider shortages. By efficiently matching patient demand with provider supply in a convenient and cost-effective online network, telehealth promises to alleviate not only issues of accessibility, but also spiraling costs. As payment systems transition from volume-based incentives to those centered on value, transparency of cost, quality and accessibility all become paramount to a successful healthcare delivery channel. Telehealth vows to meet the evolving needs of all three facets of the triple aim in health care and be an increasingly central avenue for providers to manage panels stressed by high-cost, chronic care patients. This chapter examines the progressing role of technology in healthcare delivery and details the emergence of telehealth as a cornerstone solution to the growing problems plaguing the American healthcare system. A final discussion is made regarding the future of telehealth and its role in making consumer-driven health care more patient-centered.

8.1 Introduction The adoption of technology in the healthcare industry has historically been a protracted process, marked by deliberate provider skepticism of any mechanization that stands between them and the patient in the delivery of care. Contrasting with health care’s slow adoption trends, technology has revolutionized many industries by making common and complex processes more expedient, convenient and less costly. From the mobile commerce conducted on smartphones to higher education programs taught over the internet, technology has connected the world in new and previously unimaginable ways. These connections make accessibility to goods and services a reality, independent of where a person may live. A democratization of the American economy has emerged, whereby consumers can engage in commerce at their fingertips, with online stock trading, auctions and merchandise purchasing all occurring

126   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

in the ubiquitous grounds of the internet. Industries traditionally built on brick-andmortar centers like banking and retail are now conducting a majority of their business online, making them accessible 24 hours a day to customers who may not even be in proximate location to their operations. Such innovations make conducting commerce easier for the consumer, a key facet to fostering greater adoption of supporting technology. Health care, meanwhile, has recently ushered in technological advancements of electronic medical records and physician order systems for pharmacy prescriptions and lab tests as solutions to fix duplicate, illegible or incomplete communications between providers. And while such systems have enhanced patient safety by way of medication contraindication or duplicate test avoidance, these innovations have failed to match the value proposition found in other industries that have similarly adopted technology to improve their operations and outcomes. A primary reason that the value proposition of technology adoption in health care has been muddled is a complex network of stakeholders that are involved with healthcare delivery transactions. While many industries directly connect the consumer and supplier, whether in-person or virtually, health care involves not just the patient and provider, but many third party stakeholders as well. Insurance companies are typically the payer for most healthcare services, while pharmacies and medical supply companies provide ancillary services to those delivered in the physician’s office. Furthermore, the patient has not traditionally been the bearer of healthcare costs for each exam or procedure performed, thus making prices in health care largely inelastic to consumer demand by the patient. More broadly, stakeholders exist within the healthcare delivery networks of providers, the public health regulators that certify their operations and the pharmaceutical and medical device companies that supply the drugs and tools that physicians use to manage and cure diseases. With such a broad array of stakeholders involved in the delivery of health care, the value proposition of technology adoption for the patient, who is the end consumer, has largely been overshadowed by the interests of other stakeholders. In health care, those who drive technology adoption to improve efficiency in care delivery are those who bear the financial risk. For the last half century, health insurance companies have assumed the vast majority of risk by insuring patients through their employers and passing on the increasing costs of health services in the form of higher premiums. Health insurance companies, as a result, were quick to adopt electronic billing systems to better administer claims processing and to oversee the appropriateness of care being delivered. The advent of health maintenance organizations (HMOs) in the 1980s and 1990s grew out of the health insurance companies’ heightened need to better manage risk and served as the first broad-based effort at population health management.[1] But the technology to manage claims and billing did not provide a solid value proposition to the patient, whose interests in better accessibility to providers and ultimately greater health outcomes was sometimes eclipsed by the payer’s desire to boost profitability by rationing expensive care. Furthermore, the technology needed to manage population health to the degree desired by health

8 Telehealth: connecting patients with providers in the 21st century   

   127

maintenance organizations, such as longitudinal predictive and prescriptive analytics, hardly existed decades ago. Today, however, these technologies exist to serve multiple stakeholders and are becoming more refined than the crude predecessors of previous years. Consequently, a new focus on the value proposition offered by these technologies has been made not only to health insurance companies but to patients and providers as well. If previous iterations of technology to support population health management generally failed to incorporate components that would provide value to the patient, recent innovations are largely driven by patient demand to become more engaged in managing their own health and controlling the financial risk associated with it. Industry trends have begun to shift financial risk from the insurance companies to providers and even patients with payment mechanisms of outcomes-based reimbursement and health risk-based premium pricing. While traditional fee-for-service models are certainly still the norm, the winds of change in healthcare payment point towards a greater focus on improving value – specifically, maximizing outcomes and minimizing costs – as the driving force in healthcare innovation today. And although insurance companies play a key role in identifying value-enhancing opportunities from among their trove of claims data, the pivotal setting for improving value in health care lies within the patient-provider relationship; this is where health is impacted most among the network of healthcare delivery stakeholders. It is within this relationship that technology can finally serve as an indispensable tool to connect patients with providers in ways that have long evaded the healthcare industry, even as they revolutionize other important service industries in America.

8.2 The current situation For decades, the patient-provider relationship has been forged and cultivated through visits to the hospital or provider office where care can be delivered. This arrangement put the provider at the center, with hospitals organized into departments around physician specialties rather than patient needs. Long waits to schedule an appointment and to see a provider once at the hospital have become unfortunate, yet expected facets of the broken American healthcare system, due in large part to a mismatch between supply and demand for medical services. As consumers of these services, patients have historically exerted little leverage to drive greater value in healthcare delivery, whether it be for shorter wait times, lower prices or better quality services. Technology, however, is poised to give patients the tools they have long desired in healthcare to improve their experience throughout the care continuum, and to enhance the value of every dollar spent on the American healthcare system today. Healthcare in America is at a crossroads. In the wake of spiraling costs, varying quality and increasingly difficult access, a groundbreaking reform law, the Affordable

128   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Care Act (ACA), was passed in 2010 to address these emerging issues. But despite the landmark legislation being the first of its kind in nearly half a century, challenges for the U.S. healthcare system remain following the law’s rollout. To be sure, America has seen a marked increase in the number of insured patients in recent years through the expansion of Medicaid services and the establishment of insurance exchanges in many states. Yet questions remain as to whether this insurance expansion will ultimately control spiraling costs and improve health outcomes that have made the American healthcare system an underperformer when it comes to value delivered to the patient. The patient, now bearing an increasing share of financial risk in the payment of healthcare services as employers and insurance companies increase deductibles, co-pays and co-insurance, is left to navigate an opaque marketplace devoid of clear notions of value. This is a marketplace where prices are rarely transparent, quality of outcomes is seldom known or even tracked, and access is based on proximity to a provider’s office and willingness to wait for their availability. Cumulatively, these challenges inhibit the patient from understanding the value received for each dollar spent on their health, and each challenge represents a unique pain point that technology can help alleviate. Spiraling healthcare costs have been driven over many decades by a third party payer system that incentivizes volume over value and insulates the patient from consumer sensitivities on price for the services rendered. In the traditional fee-for-service system of reimbursement, health insurance companies paid providers for each test and procedure, largely without regard to the improvement of health gained by the patient. As such, providers have been incentivized to provide more and expensive care, thereby driving costs up without commensurate improvements in patient outcomes. To a large extent, the healthcare system has been built on financial incentives that certainly don’t want patients to die, but don’t promote healing and wellness either. Rather, payment incentives are aligned to have patients keep coming back for more care; even as reimbursement levels for a given set of services may dwindle, providers have increasingly made up for such revenue deficits by boosting utilization and capacity for more care. Still, many indications, such as growing wait times for even primary care visits, point towards a widening mismatch in supply and demand for healthcare services that is certain to keep prices inflated for the foreseeable future. With the introduction of millions of newly insured patients into the American healthcare system, the Affordable Care Act has accentuated the need to find innovative ways to meet the growing demand for health services. Greater compensation structures have lured many newly trained physicians towards specialties in recent years, promising earlier payoff of medical education debt and more secure jobs given the unique skills they bring the health delivery system. Furthermore, physicians have largely flocked to more urban environments with large, prestigious academic medical centers, where the volume of patients can more easily fill patient panels for even specialized practices. However, such trends have led to a shortage of primary care providers across the country. Even in a locale such as Boston, Massachusetts, which boasts

8 Telehealth: connecting patients with providers in the 21st century   

   129

among the greatest physician per capita ratio in the country, residents find it exceedingly difficult to find primary care appointments within a month’s time.[2] Without viable alternatives to help treat even routine health needs, many still see high-cost settings like the emergency department as the front door to accessing the healthcare system in the United States, indicating a great unmet need in the availability of timely patient care. Operating within the confines and incentive structure of a fee-for-service world, healthcare delivery systems have responded with shortened appointments and some integration of non-physician providers to address this issue. Nonetheless, physicians are increasingly becoming burned out by demands to see several dozen patients per day, let alone the administrative burdens that come with this volume of visits. With less time to determine an accurate diagnosis or counsel patients on treatment plans, quality of care delivered in truncated timeframes can be jeopardized by the physician shortage as America puts more pressure on the healthcare system to manage the growing demand for services from patients without parallel increases in provider supply. With physicians and hospitals primarily focused on the quantity of services performed to drive utilization and revenues over a largely fixed-cost business, emphasis on the quality of care delivered has been lost in the equation. Even after years of training and education, medical providers still practice a mix of art and science in delivering care to patients. Diverging beliefs on how best to care for patients with similar diagnoses indicate that while medicine strives to become more evidencebased, in many instances, it still has a long way to go to achieve that status. Years of data compiled within the Dartmouth Atlas demonstrate a distinct variation in the cost of care for similar patients, with little correlation between an increased cost of care and better health outcomes.[3] Furthermore, few patients are even aware of such discrepancies in the cost of services or corresponding clinical outcomes. While payers have increasingly demanded providers adhere to basic measures of quality in order to receive full reimbursement for their services, most requirements are process-, rather than outcome-oriented. Compounding the issue has been the increase in administrative paperwork required to document such adherence by providers to these quality prerequisites, all at a time when providers are burdened with growing patient panels, shorter visits and shrinking reimbursement. Providers today have little time or incentive to reflect on the outcomes of their patients, but as payment trends shift towards greater risk-bearing for providers and patients, increased focus will have to be paid towards delivering care that is proven to maximize patient outcomes while minimizing cost. The drive towards efficient patient care, however, must also consider issues of access alongside counterparts of cost and quality. The Affordable Care Act attempted to address the access component of the triple-aim in health care, largely by increasing insurance coverage to a broad swath of the existing uninsured population in America. And while the number and percentage of uninsured Americans has certainly shrunk since the law’s implementation, lingering access issues remain for patients in the U.S.

130   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Many still find it exceedingly difficult to find physicians who are taking new patients onto their panels. For those that do, the wait time to see even a primary care provider could be weeks or months. Many of the newly insured now have care plans bought through healthcare exchanges or are covered by Medicaid. However, such coverage is often only accepted at a handful of providers, making healthcare access an issue even for the insured. Issues of cost remain as well. Subsidies through the law’s statebased exchanges help alleviate the “sticker shock” of high premiums for those with modest incomes, but steep deductibles often make these same patients think twice about going to the doctor when the first few thousand dollars will come out of their own pocket. The patient, however, isn’t trained to understand what care is and isn’t necessary, or well adept at evaluating the incremental benefit of each service that is delivered. Physicians, for their part, are also ill-equipped to understand the value of even clinically-indicated services given that cost is rarely factored into their evaluation of necessary treatment. This confluence of factors creates a passive buyer’s mentality among patients, who remain disenfranchised from decisions impacting their health while still paying for care that is often of unknown quality and incomprehensibly high cost. Clearly, the need to drive towards efficient delivery in health care has never been greater.

8.3 Proposed solution Technology has historically served a limited role in promoting efficiency within the American healthcare system. In the 1990s, the introduction of enterprise electronic medical records connected provider inputs into one centralized record database, surpassing the archaic paper system that hindered greater coordination of care. Consumer content found new delivery channels via the internet, where websites like WebMD emerged as information repositories for patient search and education.[4] Moreover, early iterations of patient portals allowed patients to view their own health records online and help manage the transfer of information between providers and health systems. Still, these innovations were simply a digitization of existing documents without changing the care delivery system itself. The following decade brought forth developments in care delivery consultation, with clinical decision support, automated reporting, error detection and alerts becoming integrated into the digital world of care documentation.[5] Advanced analytics began to decipher the growing trove of electronic health data. Some analytics engaged the patient in an increasingly mobile world with text message reminders for medication adherence or appointment times, but most innovations still focused on the physician. Care became more comprehensive as physicians could rely on technology to ensure that errors would not be made in the prescribing of drugs or the delivery of services to their patients. And while adoption of these technologies has

8 Telehealth: connecting patients with providers in the 21st century   

   131

varied based on provider willingness to incorporate automated protocols into clinical pathways, technology has begun to uncover the immense opportunity available in making care delivery more efficient. More recent innovations in the current decade have focused on not only ensuring the right care is delivered to the right patient at the right time, but also that the care being delivered is in the right setting. An outmigration of care is occurring across the country, where outpatient clinics are performing minimally invasive surgery and sending patients home the same day. Care access points are emerging in previously untapped arenas. Pharmacies are opening routine care clinics and concierge care clinics promise extended hours and expanded communication options with physicians. Providers and patients alike are realizing that shifting care from the hospital to the home is not only less costly, but also more convenient. New remote monitoring technologies are helping to manage chronic diseases like diabetes, hypertension and heart disease. Patients are also becoming more engaged in their health in their own home. Mobile applications put critical support information at their fingertips, while wearable technology to track daily activity levels is becoming more ubiquitous. In an increasingly connected world, the patient is bringing many of the capabilities previously only available at the hospital into their home. Technology has made this a reality, becoming a new care delivery channel itself. And in doing so, it has fundamentally shifted the delivery of care to where the patient is. This patient-centered transformation impacts not only the access points and delivery methods for care, but also how healthcare resources are distributed, leveraged and paid for. Technology, therefore, plays an integral role in addressing the common pain points ailing the U.S. healthcare system today. Technology’s evolution into a key player to rescue American healthcare has seen many stages. At its core, technology has become a healthcare delivery channel through its ability to connect remote data, stakeholders and services. Technology achieves this merging of telecommunications with the healthcare industry via two distinct routes. First, asynchronous technologies known as telemedicine connect providers with each other for better coordination of care. Innovations like electronic health record sharing, computerized physician order entry and even e-mail communications embody a system that is not real-time, but makes communications more efficient among providers. Telehealth, on the other hand, is defined by synchronous technologies that enable providers to diagnose, monitor, treat and educate patients remotely.[6] Leveraging common technology platforms web browsers, smartphones and tablets, telehealth allows for live care delivery sessions to take place without requiring that the patient travel to the provider. Leading the technology trend in bringing care to the patient, telehealth instantaneously makes healthcare services more accessible from almost any point in the country and embodies the fundamental shift in care delivery from being provider-focused to now patient-centered.[7] And while the value of telehealth for patients in remote and rural areas is obvious, the potential

132   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

impact of telehealth for patients even in close proximity to providers is immense as well. Those with chronic conditions that find traveling short distances to be a challenge can instead access the healthcare system from the comforts of their own home. Patients who have sadly grown accustomed to waiting weeks for a primary care appointment now have the option of immediate access and help when they are sick. The ability to avoid missing days at work to make a doctor’s appointment or to take family members to the hospital is also made possible by telehealth technology. Providers who are increasingly shouldering more financial risk in the health of their patients find telehealth to be a valuable way to maintain contact with their panel of patients without scheduling in-person follow-up appointments. Increasingly incentivized to keep patients out of the hospital, providers can now remotely triage patients and avoid costly trips to the emergency department. Better management of chronic conditions is also afforded by increased connectivity between patient and provider, whereby clinical red flags can be more expediently addressed and care plans altered accordingly. Overall, telehealth epitomizes bringing the healthcare system to the patient by having a provider literally at their fingertips for expedient, convenient and cost-effective care delivery. The transition to a virtual care delivery model is not without its own set of challenges, however. A primary hurdle is the inherent behavioral change that is necessary for both patients and providers to be comfortable in the virtual setting. Physicians are not able to palpate for physical abnormalities or visualize the person’s symptoms at any greater detail than the screen resolution through a video monitor. The patient, on the other hand, must put added faith in the physician’s diagnostic capabilities to correctly identify the particular ailment and appropriate care plan while not in the same room. Reimbursement has been slow to follow suit of new healthcare technologies as well. While providers have been accustomed to verifying their patient encounters through documentation that includes patient signatures and other supporting documentation, payers have been reluctant to pay for encounters that occur over the internet due to the risk for fraud or overpayment for unnecessary services. The growth of internet pharmacies in the last decade that illicitly prescribed and filled counterfeit drugs has fueled much of this payer skepticism. And although payers have largely moved away from online pharmacies, other challenges to reimbursement remain in the regulatory environment, such as stipulations that providers must be credentialed in the same state as the patients they see in order to get paid for their services. Until the laws in many states recognize telehealth as a legitimate alternative to the unregulated internet pharmacies of the past, regulatory hurdles will likely remain a considerable barrier to growth for this care delivery channel. A final hurdle exists within the technology itself. The complex nature of healthcare delivery necessitates an equally sophisticated channel of capabilities to leverage for accurate diagnosis and effective treatment. Comprehensive connectivity with electronic health records, exam results, pharmacy order systems, billing and reim-

8 Telehealth: connecting patients with providers in the 21st century   

   133

bursement schedules, and with the patient itself are all prerequisites of a successful telehealth infrastructure. All such facets must come together and be intelligently integrated for the patient and provider to have a user-friendly and productive experience. Where previous technology advancements have failed to simplify and streamline the delivery of care for major stakeholders, the pressure is on telehealth technology to ease the burdens of healthcare delivery and allow the patient and provider to focus on improving patient health in the most efficient and convenient way possible. With the promise of efficiency and convenience at the forefront of its value proposition, telehealth is poised to alleviate many of the challenges plaguing the American healthcare system today. All major healthcare stakeholders are beginning to realize that the value of telehealth lies in the same phenomenon that has revolutionized other industries, namely the democratization of commerce. Telehealth effectively puts healthcare services at the fingertips of virtually any patient with an internet connection, allowing for the delivery of health care to nearly every corner of the country. Not only does telehealth overcome the hurdle of provider proximity to the patient, but also of provider availability. Telehealth serves as a virtual broker between patient demand and provider supply to surmount the acute physician shortage in most American locales. Providers no longer have to waste time traveling from exam room to exam room, hospital wing to hospital wing, just to see patients that they can connect with virtually while remaining in one place. Patients are likewise afforded the ability to connect with physicians without having to travel to a hospital and promised a more expedient appointment than most in-person availabilities today. With a newfound expectation of seeing a provider within minutes, rather than weeks, telehealth promises greater overall access, with the likely outcomes of better patient triage, more frequent management of chronic conditions and reduced likelihood of disease escalation to a more expensive care setting. Not only is seeing the provider in a timely manner a game-changer for patients, but the ease of connectivity makes significant strides towards improving physicians’ engagement and retention in an increasingly strained workforce. Although the Baby Boom demographics have propelled the retirement of many physicians in recent years, many more physicians are considering an early retirement rather than face increasing administrative burdens typically required to deliver health care in America today. Telehealth, however, provides a viable alternative for physicians that are deeply invested in patient care, but are becoming burned out by the truncated appointments, mountains of paperwork and long hours that are now commonly expected. By making office visits as convenient for the physician as they are for the patient, telehealth engages providers to focus on what they are trained to do – care for patients. Physicians can take visits via an internet connection from the comforts of their own home or office and manage their own schedule of availability and patient encounters as they see fit. Furthermore, this engagement promises to help alleviate the impending physician shortage, particularly as practice offices become increasingly overwhelmed with growing patient demand in the wake of insurance expansions under the Afford-

134   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

able Care Act. Still, the change in setting of medical encounters from physician office to patient and provider homes via an internet connection also positively impacts the cost-efficiency of delivering that care. The ability to offset high cost visits with low-cost settings is another major promise of telehealth. Traditional primary care office visits tend to generate an average billing price of about $150 for a ten minute appointment slot. These in-person visits require such high costs given the brick-and-mortar infrastructure of the facility, as well as administrative overhead to manage the business operations. In contrast, Telehealth encounters cost about one-third as much to the patient or insurance company (depending on coverage), making that same ten minute appointment slot a much lower cost without a commensurate reduction in capabilities in care delivery during that time. If immediacy of care is what the patient values most in having their symptoms examined and triaged, the value proposition for telehealth becomes even greater. Compared to urgent or emergency care settings, where costs routinely balloon to the thousands of dollars, telehealth provides an expedient care alternative and a first line of defense against unnecessary trips to those high-cost access points. Although physicians may not be able to perform every type of primary care exam via a telehealth platform that they otherwise would in person, the vast majority of primary care services have been found to be qualitatively viable in a virtual channel.[8] And for patients with chronic conditions that indeed require frequent interaction with the healthcare system, virtual access to a provider is better than none at all. Telehealth’s cost-efficiency value proposition also comes from the ability to connect high frequency users of health care to their providers in a more economical manner. The vast majority of healthcare service costs in America are attributed to only a small fraction of patients, typically those with complex chronic conditions and multiple co-morbidities. Medication adherence and regular follow-up with providers are necessary to manage these patients’ conditions and to prevent costly hospital admissions. But disease management for these patients largely takes place in their own homes, making telehealth uniquely qualified to administer needed healthcare services in a timely manner to monitor patient conditions, intervene at earlier stages of disease progression when necessary, confirm compliance with care plans and avert a preventable hospitalization. The promise to “bend the cost curve” in healthcare spending primarily lies in the management of these high cost patients; telehealth serves a critical role in accomplishing this important task in healthcare delivery.

8.4 Discussion More and more, physicians are bearing the financial risk in managing these highcost patients in an era of accountable care and value-based payment models. Transitions to risk-based contracting among providers are commonly being made through

8 Telehealth: connecting patients with providers in the 21st century   

   135

shared savings arrangements within Accountable Care Organizations under the ACA, as well as bundled payment schedules for discrete care cycles. These payment models put greater control in the hands of the physician to determine the most cost-effective ways to care for their patients, but also necessitate that physicians understand the cost implications to their care pathways. Administratively, converting from a fee-forservice model is a complicated process, with great uncertainty regarding future revenues compared to traditional payment schemes. Clinically, physicians can provide better clarity to the future of risk-based contracts by delivering care that is more likely to improve patient outcomes and reduce downstream costs. Telehealth is a key delivery channel in this process, as the platform promises an easier and more cost-effective means to continuously manage high-cost patients. Making physicians aware of patient status when they are outside of the four walls of a hospital fosters greater and more immediate ability to intervene and proactively manage patient diseases without requiring an in-person follow-up visit. Continuous monitoring of high-cost patients, let alone easier triage of more routine care needs, makes providers more willing to accept the risk involved with value-based payment models and offers a unique opportunity to compete on patient outcomes in an economical manner by leveraging telehealth capabilities. Attracting patients (and favorable insurance reimbursement) based on demonstrable outcome measures is not the only patient-centered feature of telehealth. For many years, accessing healthcare services has been a largely inconvenient and painful process in America. Long waiting room times, impersonal service by providers rushing between exam rooms and dehumanizing experiences of ill-fitting gowns and uninviting exam rooms have made health care an industry devoid of good customer service. Telehealth, by contrast, promises shorter waits to see a physician, and when a patient does, it is from the convenience and comfort of their own home or preferred place with an internet connection. Making the access point as easy as an internet video conference session offers an unparalleled convenience for patients that is often a challenge, rather than a feature, of accessing healthcare services. To unlock the obvious benefits of greater convenience, cost-efficiency and continuous care management, an increasing number of providers are leveraging telehealth to make these promises a reality for millions of patients in the U.S. Telehealth is a paragon of the online transaction revolution, harnessing the power of live medical encounters to meet the growing demand for not just health content, but to access health services as well. Previous decades built patient engagement technologies focused on patient portals and online content that allowed an individual to view their medical record or associated content, but not interact with the system and derive greater value from a more digital and connected healthcare landscape. Telehealth looks to change just that. Value for the patient is most directly derived from the live interaction with trained medical providers to give real-time feedback by way of diagnosis and treatment plans to help cure or manage diseases. Following other industries like banking and retail that have allowed consumers to not only access

136   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

content like account balances or merchandise inventory, but also make transactions like online bill payment or purchasing goods, telehealth is the first major step towards bringing transactional healthcare services, and not merely static information, to the patient consumer. In making health care transactional and dynamically suited to the immediate and particular patient needs, telehealth serves as a brokerage between patient demand and provider supply of services. Providing not only an on-demand portal to connect patient and providers, but also to do so on the time and terms of each party, telehealth bridges the gap between supply and demand for healthcare services. Just as other on-demand services like Uber or OpenTable have made the transportation and restaurant industries more efficient by matching demand with supply, telehealth promises to do the same in an incredibly important industry that impacts every individual in America. But even if this “switchboard” concept of connecting available providers with patients demanding their services seems simplistic, the supporting infrastructure to make a telehealth system work seamlessly for patients and providers alike is not. The complexity of the healthcare industry dictates many advanced requirements be met for a telehealth system to deliver its promising benefits to both patients and providers. Compliance with stringent health information privacy and technological security standards under the Health Insurance Portability and Accountability Act (HIPAA) is a foremost obligation of any viable telehealth platform, as is compliance with both federal and state licensure and credentialing regulations.[9] Integrating various electronic medical records ensures that providers are well-informed of a patient’s medical history, though the interconnectivity of medical records can sometimes pose technical challenges. Furthermore, advanced billing and payment systems are needed to ensure that providers are quickly compensated for their accessibility and services. Claims, coding and co-payments all add complexity to the reimbursement question, but none are insurmountable. Together, these technical requirements, largely administrative rather than clinical in nature, define the complex network of components that innovative companies like American Well have brought together to streamline the delivery of health care as a natural extension of its traditional practice.[10] American Well’s journey to provide telehealth services mirrors the larger industry trend towards a full-bore service offering rather than a mere software solution for healthcare providers. While software to integrate disparate databases and information systems is certainly part of the value proposition offered by telehealth services, the transformation from software integration to software as a service, and ultimately as a full-service entity, is what makes telehealth companies like American Well a viable healthcare delivery channel. When in previous years many healthcare stakeholders were skeptical or ignorant of the kind of service that telehealth could deliver, it has taken pioneering companies like American Well to aggregate all the necessary pieces of a successful telehealth service and deliver them in one integrated platform

8 Telehealth: connecting patients with providers in the 21st century   

   137

where both patients and providers could come to expect a high level of service that rivals an in-person medical encounter. Telehealth’s unique online platform of merging telecommunications and healthcare technologies into a viable channel for care delivery has recently found supporters of its value proposition in large insurance payers. Being the historic brokers between patient demand and provider supply of healthcare services, health insurance companies are not only leveraging telehealth as a first line triage tool to avoid costly trips to the emergency department for immediate care, but also are demanding a high level of service for an agreement to pay for it. Insurance companies like Blue Cross Blue Shield and others are tapping into their claims databases to cross-reference quality providers and allow for dynamic practice patterns between in-person and virtual care with the same physician. Regulators and professional organizations are also following suit, recently adopting standards for acceptable telehealth delivery when in previous years, the platform’s legitimacy was ignored completely. Now, with growing patient demand, demonstrably high patient outcomes and extraordinary patient satisfaction scores, telehealth cannot be overlooked by any major healthcare industry stakeholder. But the most important healthcare industry stakeholder – the patient – has driven more advancements in the telehealth space than any other player. Just as consumers have demanded more convenient access to services that impact their daily lives, like online banking and merchandise purchasing, so too have they demanded easier access to healthcare services through telehealth platforms. The platform that began through a web browser is now accessible through mobile phones and tablet computers, making telehealth a truly portable technology service, with the immediate gateway to providers available wherever the patient may be. Employer customers have likewise demanded workplace telehealth stations to give their employees convenient access to a provider consultation without having to leave the office for a time-consuming visit that detracts from their productivity on the job. Retail centers like supermarkets and pharmacies have found value in driving greater store traffic by setting up telehealth kiosks for provider access and integration with medication dispensaries without the cost of full-time provider staffing. Being a true direct-to-consumer product, telehealth is riding the wave of the mobile consumer revolution, with phone and tablet apps gaining immense popularity, and consequently, telehealth services becoming democratized for all types of patient users. As more and more patients turn to telehealth as a first-line access point to the healthcare system, a number of positive results are beginning to be seen. With each costly visit to the emergency department, urgent care clinic or even primary care office averted, telehealth saves patients hundreds, if not thousands of dollars in services that can largely be rendered via an online encounter instead. Recent surveys estimate that about 85% of such in-person visits can instead be adequately handled through a telehealth visit, making for a much more efficient use of increasingly constrained healthcare resources.[11] Furthermore, as the physician shortage becomes more acute

138   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

in the coming years with the tidal wave of newly insured patients and growing ranks of retiring physicians, telehealth leverages such valuable resources as physicians for the greatest possible use. Greater interaction with providers makes patient adherence to medication and care plans more likely, allowing for better chronic care management and reducing the likelihood of costly readmissions to the hospital. In addition to better risk management, physicians appreciate a drastic reduction in administrative paperwork with telehealth, which otherwise accounts for almost a quarter of their time while delivering care through in-person venues.[12] Patients also appreciate the innate convenience of online encounters versus in-person visits, as evidenced by a satisfaction rating of over 90% among patient respondents.[13] Despite the positive impact telehealth has had among early adopters of the technology, further advancements will need to be achieved to make telehealth a more widespread and common care delivery channel in America. Obtaining legitimacy as a viable alternative to in-person healthcare delivery will require telehealth to further refine its inherent benefits as a service channel. Making virtual visits between patient and provider even easier will be a perpetual mission of any telehealth platform, as convenience remains central to the technology’s value proposition. Furthermore, backing from payers and regulators will likely depend on assuring adequate patient safety throughout the telehealth care continuum. Holding physicians to the same safety standards that they have been accustomed to for inperson visits must also apply to telehealth encounters. After ensuring its safety and convenience, making telehealth continuously affordable and cost-efficient for all stakeholders involved is yet another ongoing goal. Obviating the costs of office rent or expensive capital equipment has helped make telehealth an economical channel of care delivery already, but further opportunities for cost-efficiencies are necessary to bring this technology to even more patient populations. Such opportunities may lie in the areas of behavioral health that are not typically considered traditional care delivery channels. Integrating home-based technologies like wearable pedometers, heart monitors or glucose meters that help support healthier exercise and nutritional habits would further diversify telehealth capabilities to take a more holistic approach to improving a patient’s wellbeing. Bridging the current divide between healthcare services delivered in a traditional setting like a hospital and care delivered in the home through telehealth puts greater focus on the needs of the patient and the most efficient ways to meet them, regardless of care setting. Telehealth thereby becomes not an alternative to traditional medicine, but a new access point with the same legitimacy as in-person venues, though serving a different patient need niche. Collectively, these changes will help usher in a greater adoption of telehealth technology as a common channel of healthcare delivery, creating a tidal wave of users, from patient consumers, to employers, retailers and even providers, all clamoring for a seamless ecosystem of service platform, payment schedule, accepting regulations and integrated care pathways that will make telehealth a normalized facet of the care delivery system in the United States.

8 Telehealth: connecting patients with providers in the 21st century   

   139

8.5 Conclusion Guiding telehealth down the road of improvement, adoption and normalization into the American healthcare system will take a concerted leadership effort from all major stakeholders. The novelty of telehealth, while exciting, can only propel the platform forward so far before a nexus of stakeholders must come together to support widespread implementation. Telehealth service companies like American Well are taking the pioneering step of defining the technologies central to the service platform, though further standardization of information technology standards will be necessary for widespread adoption across vendors and adopting audiences. Regulators can help usher this process along by codifying IT and other standards necessary for more broadbased telehealth adoption, with guidance from professional societies and patient care advocates as key points of reference in the regulatory formulations. The Federation of State Medical Boards recently took a critical step forward in acknowledging telehealth as an acceptable channel to establish a medical relationship between patient and provider.[14] This watershed moment for the telehealth industry has allowed stakeholders to now accelerate regulatory change with the support of a major professional organization, while also calling on regulators to tailor statutes and accommodate needed changes in telehealth system credentialing, multi-state provider licensing and the process of matching patient needs with appropriate and available telehealth providers. Together, these nodes of support for growing America’s national telehealth infrastructure will help foster a new era in healthcare delivery where the patient is truly at the center of this critical service industry and technology serves a heightened and indispensable role in bringing care to the patient’s fingertips. In an increasingly consumer-driven economy, the healthcare industry is not insulated from the same online commerce trends that have revolutionized other services. As millions of previously uninsured Americans purchase healthcare coverage for the first time, increased awareness and interest is invested into how and where those healthcare dollars are spent. The growth in high-deductible health plans and self-insured employers have heightened the emphasis on gaining greater value for those out-of-pocket investments in health and wellbeing. Furthermore, the impact that health and wellbeing has on other facets of a patient’s life, from productivity at work to lifestyle habits, are being felt throughout society. To keep up with the evolving patient consumer who demands increasing transparency in price, quality and access to care, the healthcare industry must embrace technologies like telehealth in order to provide cost-effective delivery channels that accommodate a mobile and connected life that patients lead. With the option to connect with a provider in a timely manner, triage health needs and develop or refine a care plan that avoids costly trips to the hospital and better manages the burden of chronic disease, telehealth finally brings forth a notion of patient-centered convenience that has positively impacted so many other service industries. All major stakeholders stand to win with the further adoption of telehealth, from patients who save time and out of pocket costs, providers

140   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

who reduce administrative paperwork while focusing more on patient care, employers who enhance worker productivity through reductions in absenteeism, and payers who cover a new and proven first-line of care triage that most economically matches patient needs with provider services. Greater awareness of such benefits will certainly be needed before telehealth is adopted as a mainstream healthcare delivery channel, but the promise it holds to transform the way health care is accessed, provided and paid for is among the most impactful upshots to the growth of consumer-driven health care in the United States today.

References [1]

[2] [3]

[4] [5]

[6] [7] [8]

[9]

[10] [11] [12]

[13] [14]

Cutler DM, Zeckhauser, RJ. The anatomy of health insurance. Handbook of Health Economics. National Bureau of Economic Research. Working Paper 7176. June 1999. Available from: http:// economics-files.pomona.edu/marvasti/data/HealthCareClassArticles/Cutler,%201999.pdf Ghorob A, Bodenheimer T. Sharing the care to improve access to primary care. New England Journal of Medicine 2012;366(12):1955–1957. Newhouse JP, Garber AM. Geographic variation in healthcare spending in the United States: Insights from an Institute of Medicine Report. Journal of the American Medical Association 2013;310(12):1227–1228. WebMD. [Cited 2014 July 5]. http://www.webmd.com Chaudhry B, Wang J, Wu S, Maglione M, Mojica W, Roth E, et al. Systematic review: impact of health information technology on quality, efficiency, and costs of medical care. Annals of Internal Medicine 2006;144(10):742–752. Spivack R. Innovation in telehealth and a role for the government. Future of Intelligent and Extelligent Health Environment. IOS Press, 2005. Rozenblum R, Bates DW. Patient-centred healthcare, social media, and the internet: the perfect storm? BMJ Qual Saf. 2013;22(3):183–186. Palen T, Price D, Shetterly S, Wallace KB. Comparing virtual consults to traditional consults using an electronic health record: an observational case–control study. BMC Medical Informatics and Decision Making 2012;12:65. Health Insurance Portability and Accountability Act of 1996. Public Law 104–191. 104th Congress. Available from: http://www.cms.gov/Regulations-and-Guidance/HIPAA-Administrative-Simplification/HIPAAGenInfo/Downloads/HIPAALaw.pdf American Well. [Cited 2014 July 5]. http://www.americanwell.com Mercer, Employer Cost Savings Study (2012). Online Care for Employers: Case Study of Health Services Company. Castro D, Miller B, Nager A. Unlocking the potential of physician-to-patient telehealth services. The Information Technology and Innovation Foundation. May 2014. Available from: http://www2.itif.org/2014-unlocking-potential-physician-patient-telehealth.pdf Gustke SS, Balch DC, West VL, Rogers LO. Patient Satisfaction with Telemedicine. Telemedicine Journal 2000;6(1):5013. Steinbrook R. Major reform of licensing to encourage medical practice in multiple states. Journal of the American Medical Association. 28 June 2014. Available from: http://jama.jamanetwork.com/pdfaccess.ashx?ResourceID=7007944&PDFSource=13

Yaorong Ge and J. Jeffrey Carr

9 Patient-controlled sharing of medical imaging data Abstract: Sharing medical imaging data is currently carried out in two basic modes: (1) healthcare facilities form an organization-level sharing network (e.g. a Health Information Exchange) so that providers within the network can retrieve a patient’s imaging data from other facilities as needed; (2) and for healthcare facilities that do not belong to a common sharing network, the patients must manually retrieve imaging data from one facility and transport them to another facility. The first mode poses patient privacy concerns, especially as the sharing network grows. The second mode places a significant burden on both patients and providers. This chapter reviews image sharing and patient control’s needs, gaps and challenges. It then analyzes a number of solutions that have been proposed to address image-sharing needs in different scenarios. Finally, it describes an image-sharing framework that involves patients as an integral part of, and with full control of, the image-sharing process. Central to this framework is the Patient Controlled Access-key REgistry (PCARE), which manages the access keys issued by image source facilities that uniquely identify patients’ imaging data. When digitally signed by patients, the access keys can be used by any requesting facility to retrieve the associated imaging data from the source facility. This approach allows patients to control the image-sharing process with a minimal burden. It enables healthcare facilities to exchange imaging data with direct consent from patients in a manner that protects privacy and confidentiality. At the same time, it also affords healthcare facilities maximal control over the image sharing process; minimizing liability concerns as well as burden on the IT infrastructure and network bandwidth utilization.

9.1 Introduction The ability to share patient medical records, including imaging data, across providers of different organizations and locations has tremendous benefits for patients and the healthcare system.[1–3] As the population becomes more mobile and healthcare providers become more specialized, the need for medical data sharing becomes increasingly important. This importance is exemplified in the recent Health Information Technology for Economic and Clinical Health (HITECH) act for promoting Electronic Medical Record (EMR) adoption across the country where the interoperability of EMR’s for data sharing is a central requirement in certifying meaningful use of the EMR implementations.[1, 4]

142   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Medical imaging is an important category of medical data with unique characteristics. It is among the most expensive and fastest growing procedures due to continued improvement of imaging technologies. Its growing data volume is typically significantly larger than other clinical data. It is widely expected that sharing medical imaging data across healthcare enterprises will improve quality of care and reduce healthcare cost.[5] Prior imaging studies from other facilities are often a critical factor for physicians in making better decisions for patients under their care. Timely access to this information can reduce duplicative testing and patient exposure to ionizing radiation and contrast agents associated with imaging. Furthermore, if images are easily accessible from both rural hospitals and tertiary care medical centers, physicians from rural hospitals can provide timely diagnosis and treatment for rural patients by remotely obtaining specialty consultation from tertiary hospitals while eliminating the cost of physically transferring patients.

9.2 Current situation – needs, gaps, and challenges Since 2005, Regional Health Information Organizations (RHIO’s) have been established to enable sharing of essential medical information, such as medication lists and discharge summaries, across healthcare organizations. Technically, these organizations are referred to as Health Information Exchanges (HIE’s). To date, the majority of RHIO/HIE’s is sharing only summary medical data. Only a few are capable of sharing medical imaging data. Healthcare institutions in a RHIO/HIE usually have similar interests in patient care, business, insurance and/or public health. They form a central exchange network to share essential information about the collective patient base. Patients are required to provide broad consent to join this network with a wholly opt-in or opt-out policy. Once a patient consents, all providers and relevant personnel in the network can access that patient’s medical data if that access can be justified based on the need to provide care, and if that access satisfies regulatory requirements and organizational policies. Patients do not have control or knowledge of how their data is being accessed within the network. This design has recently provoked criticism and debate among patient advocates and privacy communities.[6–10] The lack of consideration for privacy protection is becoming more prominent as the networks grow in number and size. The organization-driven approach creates significant challenges when patient identities must be linked and reconciled based on sometimes conflicting records at different facilities.[11] There is also resistance to sharing data at a broader scale due to competing business and financial interests and differing regulatory concerns among healthcare providers.[8, 12] Beyond the established RHIO/HIE networks, most healthcare facilities are still relying on patients to manually transport medical imaging data to other facilities

9 Patient-controlled sharing of medical imaging data   

   143

using physical media such as CDs and DVDs. Typically, a patient goes to the source imaging facility, requests a copy of the imaging data, signs necessary paperwork for consent and other agreements, receives a copy of the imaging data on CD or DVD and then carries these media to the new facility at the next appointment. The new facility retrieves and views the imaging data and may keep part or all the imaging data for future reference. Experience in the past few years has shown that this manual process is burdensome and error-prone for both patients and providers.[5] On the other hand, the consent and regulatory procedures of this manual process are well established. Patients are in full control of the sharing process and are empowered to protect their privacy. Furthermore, the sharing facilities do not need prior agreements to exchange data: it is the patient who binds the facilities together in terms of identity matching, consent and other regulatory requirements. How can information technologies remove the burden of image sharing while ensuring patient privacy and control? This is the central question we will be examining in this chapter.

9.3 Proposed solutions Information technologies for medical data sharing can be categorized by who coordinates the sharing – organization-coordinated vs. patient-coordinated – and how data are organized – centralized vs. distributed data storage. In the organizationcoordinated sharing approach, healthcare organizations form a sharing network with pre-negotiated policies and methods. Users of the sharing network can access patient data from any participating organization at any time as long as they follow the network’s access policies. Patients provide broad consent but are, thereafter, not involved in the sharing process. In contrast, the patient-coordinated approach does not require sharing organizations to have direct relationships. It is the patient who coordinates the transfer of data from the source institutions to the destination institution and, in doing so, is able to directly control who has access to what data and, thus, can more carefully protect his or her privacy. In terms of data storage, most sharing approaches adopt the centralized data strategy that replicates patient data from all participating institutions to a central data repository. Data sharing is achieved by accessing the central data repository without the involvement of source institutions. In the distributed data strategy, the source institutions maintain the original datasets without such replication. Data sharing is achieved by querying and requesting data directly from each source institution at the time it is needed. In order to enhance the performance of this strategy, a central repository of metadata that indicates the location, type, and other characteristics of the actual patient data are usually maintained so that the record locator services can efficiently identify all the source institutions with the desired datasets.

144   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Most approaches for sharing non-imaging medical data are organization-coordinated and use a centralized data strategy.[13] Imaging data is a special type of medical data; both organization- and patient-coordinated methods for general medical data sharing also apply to imaging data. However, the distributed storage strategy is preferred due to the large size of imaging data.

9.3.1 The IHE approach The architecture proposed by the industry standards group Integrating Healthcare Enterprises (IHE) (www.ihe.net) provides a standards-based solution for health data exchange within an affinity domain defined by participating institutions. The IHE technical framework defines an integration profile [14] for image sharing, which employs a centralized metadata registry for record location while maintaining the actual imaging data at each participating institution. After records are identified from the central registry, the imaging data are exchanged directly between the source and requesting institutions. The IHE profile is designed to support organization-coordinated sharing using a distributed storage strategy and has been adopted by many RHIOs and HIEs. This approach completely eliminates the burden on patients and physicians. As long as the source and requesting institutions are linked with the IHE profile, physicians can access the images whenever needed. The biggest challenges are how to protect patient privacy, ensure patient safety and comply with federal and state laws and institutional policies.[1, 2, 8–11, 15–22] This issue is looming large as the number and scale of IHE-based networks grow. Related to these issues are the institutional trust and liability concerns.[8] Another major challenge is linking patient records from multiple institutions – the Master Patient Index problem.[11] Without direct verification by patients, even the best algorithm will have potentially devastating errors.

9.3.2 Cloud-based approaches More recently, the centralized storage strategy has been adopted for organizationcoordinated and patient-coordinated image sharing by some exchanges and vendors using cloud-based technologies.[23] These cloud-based approaches are mostly ad-hoc and transient: a provider uploads an image to the cloud, sends the universal resource link (URL) to the other provider, then the other provider uses the URL to view or download the image. Due to the large size of the imaging data, they are usually deleted from the cloud after a limited timeframe (e.g. one month). The cloud-based approach has also been proposed to enable patient-coordinated image sharing using patient health record (PHR) systems.[24–26] The PHRs are deployed by payers, healthcare organizations, employers and information tech-

9 Patient-controlled sharing of medical imaging data   

   145

nology vendors, to give patients the ability to collect their own health and medical data using internet technologies.[27, 28] Mendelson described a recent effort by the Radiological Society of North America to enable image sharing using a cloud-based PHR.[29] In this approach, a patient with a PHR account is given a unique code after imaging is complete, the images are sent to an edge server at the facility, encrypted with the unique code and then sent by the edge server to a clearinghouse. The images are temporarily stored at the clearinghouse until the patient presents the unique code to the PHR and instructs the PHR to retrieve the images. After being retrieved to the PHR and decrypted with the unique code, the images can be shared with physicians by e-mailing them a URL. The PHR-based approach can theoretically overcome many of the patient privacy and safety challenges facing the organization-coordinated approaches.[30] When patients are involved in the actual sharing of specific images between specific organizations, their ability to protect privacy is dramatically increased and the chance of mismatched identities is also significantly reduced. Furthermore, with explicit consent from the patients, the exchanging institutions do not require complex legal agreements to cover liability concerns such as hypothetical use and breach scenarios. However, the PHR-based design also introduces additional burdens for both patients and providers: 1. Patient burden: Existing implementations of PHR-based systems require patients to be well versed in internet technologies, which is a challenge for significant proportions of older, rural and disadvantaged populations. 2. Provider burden: – PHR systems are typically not integrated with internal EMR systems, requiring physicians to sign in to an external system and search for information that is disconnected from internal records, breaking their already busy workflow. – Imaging data are not simply texts. They require sophisticated viewing and analytical capabilities. Analyzing prior images on one platform, provided by an external PHR, and viewing new images using another platform, provided by an internal image management system, can be inefficient and potentially error prone. Downloading prior images from an external PHR is also inefficient, especially when images are large.

9.3.3 The PCARE approach We propose a framework for patient-coordinated, distributed image sharing that enables patient control and privacy protection while minimizing burdens on both patients and providers. This framework is especially suitable for enabling sharing among unaffiliated healthcare organizations, with each organization being either a single healthcare enterprise or an organization-coordinated sharing network (e.g. RHIO/HIE). Our framework aims to provide an image-sharing network that is scalable nationwide and is effective in protecting patient privacy and safety.

146   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Conceptually, the proposed framework is a network of networks, similar to the designs of proposed NHIN,[31] but with a fundamental difference: while NHIN assumes organization-coordinated sharing in both levels of the network, our framework adopts patient-coordinated sharing at the higher levels while allowing the lower levels to be either independent facilities or organization-coordinated sharing networks (see Figure 1). This hybrid sharing strategy achieves the best balance between physician workflow efficiency and patient privacy protection. That is, within regional sharing networks naturally formed by organizations with common patient care and business interests, physicians can exchange images directly based on broad policybased patient consent. Beyond regional sharing networks, among unaffiliated organizations or crossing state boundaries, patients must explicitly give permission to move specific datasets from a specific source imaging facility to the specific requesting organization. By facilitating patient consent and authorization for each specific sharing of imaging data across organization, network or even state boundaries, this framework dramatically simplifies the problem of reconciling differing regulations, policies and laws governing data sharing.[2, 8, 18, 20]

9.3.3.1 Architecture As shown in Figure 1, the proposed image-sharing framework is logically composed of a PCARE Master Server and a network of PCARE Facility Servers, one at each participating facility or organization (note that each server may require multiple physical or virtual servers). A core component of the PCARE Master Server is the PCARE registry. This is the critical design feature that sets our framework apart from existing patient-coordinated sharing frameworks such as PHRs. Instead of dealing with actual clinical data as in a PHR, the PCARE registry is a collection of access keys that uniquely represent clinical datasets. These unique access keys are generated by a healthcare imaging facility upon patient authorization to provide a secure electronic conduit to the actual dataset. Each access key is a token that can be used to redeem the corresponding dataset. It contains a limited and specified number of attributes such as the patient’s name, date of birth and the patient’s unique identifier at that facility (commonly referred to as a “medical record number”) in addition to metadata describing the dataset. The access key and accompanying patient metadata are encrypted and digitally signed so that only the patient and this facility are able to decrypt and authenticate the content. Furthermore, each access key also contains Universal Resource Locators (URLs) that specify links to the facility that issued the key. These facility-specific URLs provide the links to services where the actual clinical data, in this case medical imaging data, can be obtained. To ensure strong security, the facility must update the access key periodically or whenever relevant information changes, for example, when facility’s URLs change. For finer control of access, multiple access keys may be generated for each patient to allow access to different parts of a patient’s health record.

9 Patient-controlled sharing of medical imaging data   

   147

Figure 1: Overall architecture of the PCARE patient-controlled image sharing network. Each node of this network may be an independent healthcare enterprise or an organization-coordinated image sharing network such as a HIE.

Another critical component of the PCARE Master Server is the Master Patient Index (MPI). As discussed previously, the MPI links together patient identities that may be different at different healthcare facilities. In organization-coordinated sharing networks, MPI is established by comparing demographic information in existing records. In some PHRs, MPI can be established by patients logging into the PHR account and then into personal accounts at each healthcare facility. In the PCARE framework, we fully leverage patient participation in their healthcare process by engaging patients to establish the linkage between their local identity and the PCARE identity as a part of their normal registration or check-in process. The MPI created through such physical verification processes eliminates major sources of error in conventional MPI linkage systems. The third major component of the Master Server is the Patient Control Portal. We note the difference between a Patient Control Portal and a Patient Access Portal. Most patient portals, including PHRs, are access portals in the sense that these portals give patients direct access to their health data. However, we believe that access to the actual data is not necessary for patients to control their privacy in a data-sharing network. The access keys in the PCARE registry provide metadata information about the type of studies that are sufficient for patients to make their sharing decisions. Therefore, our patient portal is called Patient Control Portal to emphasize the fact

148   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

that what the patients have access to in the portal are access keys rather than actual health data, and what patients do in the portal is to control the exchange of health data rather than to view or manipulate the actual content of health data. The Facility Server is the gateway for each participating institution. Conceptually, each Facility Server corresponds to one node in the PCARE network that represents one institution. The Facility Server includes modules for access key management and local EMR integration.

9.3.3.2 Card-based user interface To further minimize patient burden, a card-based user interface is developed for patients to easily authorize data sharing in a majority of clinical scenarios. Presently credit cards, bank cards, insurance cards and various other cards permeate our everyday lives. Most people are accustomed to using cards for making transactions at gas stations, supermarkets and hospitals. The design of our image-sharing approach makes the card mechanism an ideal interface for patients to control and coordinate image sharing. The PCARE card is similar to a credit card. It stores identifying information that allows a patient to initiate image-sharing control at a participating facility by swiping the card at a kiosk. After the user is identified, potentially with an additional personal identification number (PIN), the kiosk displays the records of prior imaging data from other facilities and allows the patient to determine which images to share with the current facility.

9.3.3.3 User assessment With approval from the Wake Forest University Health Sciences Institutional Review Board, a small user assessment of the prototype PCARE system was performed. Twenty-one patients or family members and nineteen medical center staff members were randomly selected. Each subject provided verbal consent after a brief introduction to the test and individually received a short demo of the prototype system. After given sufficient time to assess the system and ask questions, the subject filled out a short questionnaire. The outcomes of this assessment are summarized in Figure 2. The top row shows results from patients (n=21); the bottom row is from staff (n=19). As seen in this Figure 2 (a), more than half of the patients found it difficult to use CDs for exchanging images. All patients found the PCARE system easy or very easy to use. As a comparison, most of these patients were comfortable with using cell phones and ATM machines. The question for “View all images” (Figure 2 (b)) asked the subjects whether they believed their physician should be able to view all images without specific consent. Significant number of patients disagreed with this statement, while almost all patients agreed that the “PCARE privacy” mechanism provided them with control over their privacy. The staff’s sentiment was similar to that of the patients in general, except that they were more comfortable with technologies.

9 Patient-controlled sharing of medical imaging data   

   149

Figure 2: User assessment results. The top row shows results from patients. The bottom row shows results from staff members.

9.4 Discussion The need for image sharing occurs in many different scenarios that require different technological solutions. While physician workflow, patient care quality and healthcare costs are important considerations in designing image-sharing solutions, the concerns for patient privacy and patient empowerment must be addressed as an integral part of the solution. In fact, as shown in the proposed solutions, engaging patients in the image sharing process can not only enable patient control but also lead to more accurate results. It is important to realize that some image sharing scenarios are best served by an organization-coordinated approach because patient involvement will break physician workflow and interfere with patient treatment. For example, in emergency transfer or sub-specialty consultation scenarios, physicians need efficient and direct means to access all relevant patient data without delay. Since the healthcare providers are known to each other in these scenarios and are usually in a small number of organizations, proper policies and procedures can be easily established to protect patient privacy and safety in such well-defined “affinity domains”. On the other hand, for patients seeking care in a different organization due to travel or relocation, the new organization may be in a completely different geographic location and have no reason to establish routine sharing relationships with prior organizations. In this and many other scenarios, patient-coordinated sharing becomes the compelling choice. The PCARE framework allows both types of image sharing to be handled effectively and securely. It is complementary to the existing image-sharing efforts in HIE’s and provides an effective way to link all the existing organization-coordinated sharing networks together to enable larger scale sharing. Image sharing is a complex process involving many people and various factors. Selection of a particular image sharing solution must balance concerns of patients and physicians, business and regulation, as well as efficiency and cost. The PCARE

150   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

approach not only makes the most technical and economical sense, but also addresses a number of legal, regulatory and business concerns. In the PCARE framework, imaging data remains in the repositories of the performing healthcare facilities where that data was originally archived. Only the secure access keys, not the images, are stored in a central repository through which patients can directly control their imaging data. The patient controls and manages access while the facilities manage the direct transfer of the full-fidelity images. In contrast to the PHR-based solutions, the PCARE approach takes full advantage of the patient-controlled nature while overcoming many of its pitfalls including patient and physician burdens. For example, the fact that the images remain at the source imaging facility and are exchanged only between healthcare facilities significantly reduces the scope of failure points and makes it possible to minimize liability for the exchanging facilities. Furthermore, the PCARE framework distinguishes data access across and within enterprises, and accordingly adopts a two-level consent mechanism: consent for sharing data between two organizations vs. consent for accessing data within an organization. This strategy is critical for managing the complexity of legal and policy differences across enterprises and local governments.

9.5 Conclusions Patient engagement is the key to enable large-scale sharing of medical imaging data while ensuring patient privacy and control. Solutions for patient-controlled image sharing must address the need to minimize the burden on patients, providers and infrastructure, and maximize patient as well as institutional control over the sharing process. The PCARE framework demonstrates that such a general solution is feasible and welcomed by both patients and clinical staff.

Acknowledgements The authors would like to acknowledge the contribution of David Ahn and Shree Unde for developing and testing the PCARE system. This work was supported in part by grant number RC2  EB011406 from the National Institute of Biomedical Imaging and Bioengineering.

9 Patient-controlled sharing of medical imaging data   

   151

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25]

Kuperman, GJ. Health-information exchange: why are we doing it, and what are we doing? J Am Med Inform Assoc 2011;18(5):678–82. Vest JR,Gamm LD. Health information exchange: persistent challenges and new strategies. J Am Med Inform Assoc 2010;17(3):288–94. Unertl KM, Johnson KB, Lorenzi NM. Health information exchange technology on the front lines of healthcare: workflow factors and patterns of use. J Am Med Inform Assoc 2011. Morrissey J. Health information exchange. Hosp Health Netw 2011;85(2):22–7. Flanders AE. Medical image and data sharing: are we there yet? Radiographics 2009;29(5):1247–51. Sweeney L. Privacy Issues Overlooked. Mod Healthc 2010. Rode D. Building trust into the NHIN: Key legislation can ensure the privacy of personal health information. J AHIMA 2005;76(8):18,20. Gravely S, Whaley ES. The next step in health data exchanges: trust and privacy in exchange networks. J Healthc Inf Manag 2009;23(2):33–7. McGraw D, et al. Privacy as an enabler, not an impediment: building trust into health information exchange. Health Aff (Millwood) 2009;28(2):416–27. Connors B, Leipold J. The 42 CFR Part 2 and NHIN conundrum. Behav Healthc 2009. 29(7):52–3. Just BH, et al. Managing the integrity of patient identity in health information exchange. J AHIMA 2009;80(7):62–9. Goldstein MM. Health information technology and the idea of informed consent. J Law Med Ethics 2010;38(1):27–35. Donnelly J, et al. Building an interoperable regional health information network today with IHE integration profiles. J Healthc Inf Manag 2006;20(3):29–38. Mendelson DS, et al. Informatics in radiology: image exchange: IHE and the evolution of image sharing. Radiographics 2008;28(7):1817–33. Gadd CS, et al. User perspectives on the usability of a regional health information exchange. J Am Med Inform Assoc 2011;18(5):711–6. Adler-Milstein J, Bates DW, Jha AK. A survey of health information exchange organizations in the United States: implications for meaningful use. Ann Intern Med 2011. 154(10):666–71. Hincapie AL, et al. Physicians' opinions of a health information exchange. J Am Med Inform Assoc 2011;18(1):60–5. Stevenson C, et al. Share, don't hoard: The importance of information exchange in 21st century health-criminal justice partnerships. Crim Behav Ment Health 2011;21(3):157–62. Edwards A, et al. Barriers to cross-institutional health information exchange: a literature review. J Healthc Inf Manag 2010;24(3):22–34. Page D. Information exchange. Health information exchanges hold promise, pose perils. Hosp Health Netw 2010;84(1):12. Adler-Milstein J, Bates DW, Jha AK. U.S. Regional health information organizations: progress and challenges. Health Aff (Millwood) 2009;28(2):483–92. McDonald C. Protecting patients in health information exchange: a defense of the HIPAA privacy rule. Health Aff (Millwood) 2009;28(2):447–9. Prestigiacomo J. Taking it to the clouds. The image movement of Montana starts sharing images via cloud-based solution. Healthc Inform 2010;27(9):30,32,55. Adida B, et al. Indivo x: developing a fully substitutable personally controlled health record platform. AMIA Annu Symp Proc 2010;2010:6–10. Mandl KD, et al., Indivo: a personally controlled health record for health information exchange and communication. BMC Med Inform Decis Mak 2007;7:25.

152   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[26] Mandl KD, Szolovits P, Kohane IS. Public standards and patients' control: how to keep electronic medical records accessible but private. BMJ 2001;322(7281):283–7. [27] Archer N, et al. Personal health records: a scoping review. J Am Med Inform Assoc 2011;18(4):515–22. [28] Do NV, et al. The military health system's personal health record pilot with Microsoft HealthVault and Google Health. J Am Med Inform Assoc 2011;18(2):118–24. [29] Mendelson DS. Image sharing: Where we’ve been, where we’re going. Applied Radiology 2011;40(11). [30] Halamka JD, Mandl KD, Tang PC. Early experiences with personal health records. J Am Med Inform Assoc 2008;15(1):1–7. [31] ONC-HIT. NHIN Architecture Overview. 2010. [cited 2012 March 1] Available from: http://healthit. hhs.gov/portal/server.pt/gateway/PTARGS_0_11113_911643_0_0_18/NHIN_Architecture_ Overview_Draft_20100421.pdf.

Mehnaz Adnan, Jim Warren and Hanna Suominen

10 Patient empowerment via technologies for patient-friendly personalized language Abstract: Free-text reports are used in health care to transfer information between working shifts and sites. This text, written by a physician, nurse, specialist, ward secretary or other healthcare worker, is full of jargon, idioms and shorthand that patients find difficult to understand. If patients are to be empowered to take an active role and make informed decisions in their health care, they need support for understanding these reports. This chapter discusses language technologies as a way to provide support for patients to better understand free-text reports with difficult clinical language. This includes expanding shorthand, replacing words with patient-centric terms, providing term definitions, hyperlinking to further information on patientfriendly and reliable sites on the internet, and personalizing medication advice and other content. To conclude, statistical evaluations and benchmarks in shared tasks give evidence of language technologies being successful in making text easier to understand and better personalized. Moreover, electronic health records that both patients and clinicians use to read, write and share information are becoming more commonplace and provide a platform for language technologies to assist patients in reading free-text reports.

10.1 Introduction Free-text reports transfer information between working shifts and sites in health care. They include, for example, case sheets, discharge summaries, medical records and nursing notes. This clinical text is written by a physician, nurse, therapist, specialist, ward secretary or other healthcare worker. Such reports can be full of ward-specific jargon, idioms and shorthand that both patients and healthcare workers from other specialties or sites find difficult to understand (Figure 1). These challenges are further documented for a Thai medical-surgical ward,[1] Norwegian medicine and cardiopulmonary units,[2] a US hospital,[3] New Zealand hospitals,[4] and Finnish and Swedish intensive care.[5] Patients can be empowered to have an active role in their healthcare and make informed decisions concerning their health and care through supportive patientfriendly, personalized language on their reports. Patient empowerment refers to patients having partial control and mastery over their own health and care, leading to patients making better health/ care decisions, being more independent from healthcare services and decreasing healthcare costs.[6] Making the right decisions depends on having the right information at the right time; therefore, it is necessary to provide

154   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

patients with personalized and readable information about their health conditions for their empowerment. On their reports, this could mean expanding shorthand, replacing words with patient-centric synonyms, providing term-definitions, hyperlinking to further information on patient-friendly and reliable sites on the internet, personalizing medication advice and other content and an option to see the original text (Figure 1). “80 y/o male with 2 yr h/o SOB and GERD. Had Echo that revealed AS. Has had serial Echo’s since, with the last done on […] which showed worsening AS.” 80 year old male with 2 years history of shortness of breath (treatment guidelines) and heartburn (also known as gastroesophageal reflux disease; (treatment guidelines). Had an echocardiogram ultrasound test that revealed aortic stenosis (that is, the aortic valve not opening fully which decreases blood flow from the heart: (treatment guidelines; guidelines on separating the symptoms of heart conditions from heartburn). Has had serial echocardiograms since, with the last done on […] which showed worsening aortic stenosis.

Figure 1: Example of supporting patient-friendly, personalized language on record text (bottom). The original text (top) is a style-preserving anonymized example from a U.S. discharge summary, originating from the CLEFeHealth2013 (goo.gl/it3zE) data.[31] More information about these data and related human subjects training, ethics clearance, research permission, registration, user access and contact people is available at http://goo.gl/STHsnS.

This process of identifying entities in a text and assigning semantic metadata to them that a computer can process is known as semantic annotation.[7] In this context, the relevant entities may include shorthand, difficult words and definitions or further information in external resources; the metadata may include expansions, patient-centric synonyms, definitions and document hyperlinks. Semantic annotations on freetext reports have the potential to provide an ‘interpretative layer’ to facilitate patients in understanding their health information. This provides a promising alternative to support patient readability of complex reports as compared to the de facto approach of the patients themselves searching the disconnected resources scattered across the Internet in search of expansions for shorthand, definitions of difficult terms and to find further information on key clinical concepts.

10.2 The current situation: need, gaps and challenges The requirement to provide patients with valid, readable information about their health and health care is stipulated by international and national policies and laws. For example, A Declaration on the Promotion of Patients’ Rights in Europe by the World Health Organization in 1994 states that patients have the right to be fully informed about their health status, prognosis, medical conditions, diagnoses, pro-

10 Technologies for patient-friendly personalized language   

   155

posed and alternative care procedures with their potential risks and benefits, effects of non-treatment, treatment progress and discharge guidelines as well as to receive a written summary of this information from their healthcare providers. It also obligates healthcare workers to communicate this information to patients in a way appropriate to their capacity for understanding, minimizing the use of unfamiliar jargon. Similarly, the Swedish Patient’s Right to Information 06:So559/2005 and Patient Data Law 255/2008 and the Finnish Act on the Status and Rights of Patients 785/1992 and Statute 298/2009 on Patient Documents state that to ensure good care, patient reports must cover all necessary information, adequately detail the patient’s conditions, care, and recovery, be explicit and comprehensive, and include only generally well-known, accepted terms and shorthand. In order to meet these requirements for valid, readable information, document and terminology standards have been developed and adopted in health care, but the conformance of reports with these standards is poor.[8–9] Key standards include HL7  Clinical Document Architecture (goo.gl/ulfPa), ICD International Classification of Diseases (goo.gl/s2HgYa), MeSH Medical Subject Headings (goo.gl/ nqnF), SNOMED CT Systematized Nomenclature of Medicine – Clinical Terms (goo. gl/e2ZBZ8) and the UMLS Unified Medical Language System (goo.gl/ZhqQ). The aim in adopting such standards is improved interoperability and understanding of synonymous terms and term hierarchies. For example, heartburn and its synonym of pyrosis are assigned in MeSH to the unique identification code of D006356, related to the concept of Gastroesophageal Reflux (D005764), classified in the term hierarchy as C23.888.821.525, which describes that heartburn belongs to Digestive Signs and Symptoms (C23.888.821), which are Signs and Symptoms (C23.888) under Pathological Conditions, Signs and Symptoms (C23). However, assuring by hand that the standards are followed is difficult and time-consuming. The power of the internet as a tool for patient empowerment is based on its wide accessibility and range of information as well as patients’ abilities to choose and control the content and amount of information.[10–13] Most fundamental in this is searching the internet for general health information: nearly half of Europeans consider the Internet as an important source of health information;[14] over two fifths of Australian searches relate to health and medical information;[15] and nearly 70% of search engine users in the U.S. want information about health conditions.[16] A further developing approach to patient empowerment via the availability of valid, readable information has been through personally controlled electronic health records (PCeHR) on the internet for both patients and healthcare workers to use for reading, writing and sharing information nationally. PCeHRs are internet-based and promote information accessibility, patient engagement and interaction. For example, the Estonian PCeHR (goo.gl/cOasWJ) has been open since 2008 for patients to read their own record, prescriptions and prescription medications purchases, monitor and manage their data privacy, contact details, healthcare bookings and reminders, and submit organ donor will or refusal from certain treatment. The gradual develop-

156   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

ment of this PCeHR is ongoing and in 2013 almost all physicians and half of patients use it. The Finnish PCeHR (goo.gl/dmuoZe) has been open since 2010 and in 2013 it encompassed prescribing, report archiving, access by patients to their prescriptions and reports, and access by healthcare workers to the national pharmaceutical database. The Australian PCeHR (http://goo.gl/e15e9) has been open since 2012. It enables registered patients to read a summary of their health information, control its content and manage their data privacy. Thus, PCeHRs are becoming increasingly common, although their development and uptake is still ongoing in many countries.

10.3 Proposed solutions For nearly fifty years, language technologies, that is, technologies based on natural language processing (NLP), have been recognized as a way to support producing and using text. An example NLP use is generating semantic annotations. PubMed search for NLP on 1. August 2013 returned over 2,900 results, including a pioneering study from the 1970s,[17] a recent methodological introduction[18] and reviews.[19–24] Statistical evaluations and benchmarks in shared tasks give evidence of NLP being successful in making text easier to understand and better personalized with their quality gradually improving to over 90% correctness.[25] Successful examples include the MedLEE Medical Language Extraction and Encoding System in the New York Presbyterian Hospital to semantically annotate text in accordance with the UMLS terminology standard and Autocoder at the Minnesota Mayo Clinic which adds diagnosis codes to free-text reports.[26–27] To drive patient empowerment via the availability of valid, readable information, NLP is a necessity; about 40% of clinical reports is free text and during one inpatient period in intensive care alone, this can translate to up to 37,000 words or 64 pages.[28] However, not many NLP methods have progressed from research and development to use in practice.[29] Consequently, most PCeHRs are not equipped with even the basic proofing and searching tools that typical word processors and search engines include. The internet is widely used among patients to look for health information,[14–16] and this affects patient decisions.[30] Applying NLP on PCeHRs to support patient-friendly, personalized language in connection with resources on the internet has the potential to facilitate patients in seeking, retrieving and understanding health information. Next, we present some efforts for transforming PCeHR data into meaningful information for patients through semantic annotations.

10 Technologies for patient-friendly personalized language   

   157

10.3.1 The CLEFeHealth 2012–2014 shared tasks for patient-friendly, personalized reports Shared tasks, that is competitions or comparisons, where all participants solve the same problem using the same data but their own methods, are one way to encourage the flow of NLP to PCeHRs.[29] They support access to data, conformance with standards, collaboration between participants, abilities to reproduce and scale the results and user-centricity such as patient-friendly language. Typically, such competitions have focused on the tasks of healthcare workers and/or researchers: finding patient cohorts for clinical trials; de-identification to enable use of clinical text for research purposes; diagnosis coding; extraction of chief complaint or medication information; identification of smoking status; and recognition of obesity and comorbidities, among others.[31] In contrast to previous shared tasks, the CLEFeHealth workshops in 2012–2014 have focused on shared tasks for NLP to address patients’ needs related to their clinical reports. CLEFeHealth2012 (goo.gl/YsyGkM) identified the task of using language technologies to make text easier to understand for patients.[32] This was the common denominator of the three key-note talks, panel and eleven peer-reviewed extended abstracts. CLEFeHealth2013 (goo.gl/it3zE) introduced three shared tasks to enrich text with semantic annotations that make PCeHRs easier to understand for patients. [31] It attracted nearly 200 participants from 20 countries and released data and tools for future research. The three tasks were: (1) identification and mapping of disorders to standardized terminologies; (2) expansion/ mapping of shorthand to standardized terminologies; and (3) patient-friendly search for further information on the internet. The participants were provided access to 300 clinical reports in English, authored in U.S. intensive care, and approximately one million documents from health and medicine sites on the internet. The documents were mainly in English. 34  participating teams developed a total of 113 systems to simplify the vocabulary, expand shorthand and link text to further information. In the first task, when compared to expert annotations, the best system was able to identify over 70% of the disorder boundaries (i.e. recall = 0.71) and of the system-generated disorders, 75% were correct (i.e. precision = 0.75). The best systems for mapping of disorders and shorthand to standardized terminologies were 59% and 72% accurate, respectively. The best search system in the third task had the precision of 52% within the top-10 documents (i.e. precision@10 = 0.52). CLEFeHealth2014 (goo.gl/nCIkTr) continues this development towards patientfriendly, personalized language on PCeHRs. Its shared tasks include the design of a visual-interactive search and exploration tool to connect the 2013 technologies with information systems, extension of the 2013 task  1 to address words that determine disorders and elaboration of the 2013 task 3 to better utilize the discharge summary in the search for further information on the internet.

158   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

10.3.2 The SemLink system for patient-friendly, personalized discharge summaries SemLink is an NLP system that maps difficult terms in discharge summaries with their patient-friendly names and inserts dynamic hyperlinks to the most relevant resources in the internet.[33] Similarly to the CLEFeHealth2013 tasks, its goal is to empower patients by supporting them in understanding their own PCeHR, health and health care. The system uses standards (i.e. UMLS, Open Access Collaboratives’ Consumer Health Vocabulary (CHV, http://goo.gl/dnxngM) and locally approved shorthand expansions). It annotates UMLS concepts with their preferred names, CHV concepts with easier terms, based on CHV scores, shorthand with expansions and difficult terms with hyperlinks to specific, patient-oriented resources on the internet (Figure 2). Its method for generating the hyperlinks leverages a set of patient health portals and dictionaries on the internet to provide topically relevant and readable resources or definitions. For example, if a patient uses the SemLink interface to read his/her discharge summary and sees the annotated term haemoptysis, she/he can click it. This opens a relevant hyperlink to define and support understanding of the clinical concept (in this case, a consumer-oriented page from the US National Library of Medicine, goo.gl/vRR9w) in a new internet browser window.

Figure 2: An example phrase in which annotated terms are underlined in a PDF document. In this document, when the user hovers on the underlined term ‘haemoptysis’, the interface shows its preferred name ‘coughing up blood’ in the tooltip.

In an expert evaluation, SemLink, generated 65% of topically relevant hyperlinks from trusted patient resources on the internet for difficult terms in discharge summaries. The automated hyperlinking showed good performance for providing quality information from trusted online consumer health portals for annotated terms having UMLS semantic types Diagnostic Procedures and Pharmacologic substance. However, the performance decreased when hyperlinks were provided from online dictionary resources.

10 Technologies for patient-friendly personalized language   

   159

10.3.3 The SemAssist system for patient-friendly, personalized discharge summaries The SemAssist system is in line with the goals of CLEFeHealth and SemLink to support patients’ understanding of their own PCeHR, health and health care.[34] However, instead of semantically annotating PCeHRs, SemAssist has been developed to help healthcare workers in writing the patient advice section of the discharge summary. The goal of this NLP system is to reduce medical errors and improve physicians’ efficiency, focusing on the fact that patients’ limited knowledge about medications is a key factor in post-discharge adverse drug events.[35] SemAssist generates model text for the patient advice section; if the text in the advice section currently is missing important elements, it provides a critique so that appropriate corrective actions can be taken before signing off the discharge summary. SemAssist uses a model for a set of medications known to account for a high percentage of readmissions to hospital (warfarin, ACE inhibitors, amoxicillin, acetaminophen and prednisone) and their patient-related advice in an ontology that includes patients’ actions (things to do or to avoid), as well as common side-effects and reminders of required follow-up. First, it annotates medication instances and the patient advice section with information specific to each medication class. For example, for a 75-year-old female patient with discharge medications including Marevan (a brand of warfarin) and prednisone, SemAssist generates text sections for the headings of the required patient action, medication side effects and follow-up (Figure 3). Second, it offers the generated sections to the author for proofing and insertion to the patient advice section. Authors can also enter any advice from scratch. Third, before the sign off, SemAssist computes missing advice and prompts the author with a list of missing Patient_Information instances as per the ontology (e.g. if a patient on warfa-

Figure 3: Example advice for high-risk medications Marevan and prednisone for a 75-year-old female patient.

160   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

rin was not reminded to seek a follow-up INR test, this omission would appear as an alert). Through the provision of model text and NLP-based critique, SemAssist serves to assure that patients get the information they need to be empowered in avoiding adverse events from their discharge medications.

10.3.4 Related work The need to produce patient-friendly, personalized clinical language that is directly hyperlinked with the relevant resources on the internet is growing as more patients gain access to their reports through PCeHRs on the internet. NLP on PCeHRs can play a key role in supporting patient-centric and information-driven health care. These technologies can automatically process free text in reports to produce customizable views and tools for patients. In this chapter, we have discussed NLP research, development and evaluation related to the CLEFeHealth 2012–2014 shared tasks and the SemLink and SemAssist systems for transforming PCeHRs into meaningful information for patients. Document and terminology standards together with NLP have been used in freetext reports to address the goal of supporting patients with health-related information. Another effort similar to the CLEFeHealth2013 Tasks 1–2 and SemLink in reducing the vocabulary difficulties met by patients using PCeHRs is a prototype system implemented to identify difficult terms in clinical text and automatically replace them with easier synonyms and add explanations for these terms in lay vocabulary.[36] This system extends the already existing Health Information Text Extraction (HITEx) system for concept identification to synonym mapping, explanation generation and explanation insertion to the documents.[37] The majority of text replacements (68.8%) generated by the prototype system were found to be correct and helpful by expert review; however, a user evaluation of the system has not shown a statistically significant trend (p = 0.15) towards better comprehension when translation is provided. Similar to the CLEFeHealth2013 Task 3 and SemLink, the Infobuttons system is developed to improve access to health information on the internet by embedding context-specific hyperlinks.[38] Infobuttons have generally been used to target healthcare workers’ rather than patients’ information needs; however, a prototype application of Infobuttons has been implemented to facilitate patients’ understanding of their Pap smear reports.[39] This application provides explanation of terms, links to findings-specific resources on the internet and general resources related to Pap smears. The Intelligent Consumer-Centric Electronic Medical Record (iCEMR)[40] is another system that supports patient empowerment as an extension of their PCeHR. iCEMR recommends pages on the Internet that are relevant to home medical products based on an expert system and internet-search technologies. First, it extracts the preliminary set of user topics (i.e. disease, symptoms and other medical conditions) automat-

10 Technologies for patient-friendly personalized language   

   161

ically from the record. Second, it uses this information to retrieve personalized recommendations and explain difficult medical concepts. In a user evaluation, iCEMR was easier to use and able to find a larger number of desired home medical products very quickly (in less than two seconds) as compared to keyword-based searches. NLP has also been used to support patients in accessing and reading health information on the internet independently from PCeHRs. The Health Topic Overview (HTO) system uses NLP techniques to facilitate the readability and navigation of health information by organizing and grouping the information in the internet pages into patient-preferred categories.[41] This takes the form of a topic overview. HTO uses the CHV and UMLS standards to categorize the most commonly used UMLS concepts semantically, generate topic overviews for these concepts and present them in an internet page. The system has been found to categorize health phrases in health pages on the internet with a 82% and 75% recall. A text summarization approach is used in the Technical Article Summarizer to automatically generate tailored summaries of the relevant medical literature for both healthcare workers and patients.[42] In a user evaluation, this system enabled lay people to quickly identify information as compared to results returned by the Yahoo search engine. This system has also been extended to support patients’ comprehension of medical literature by providing definitions for difficult terms automatically.[43] This system resulted in a 40% increase in mean comprehensibility rating when the sentences were provided with automatic definitions as compared to sentences not augmented with definitions. NLP has gained attention for extracting critical information from free-text reports to facilitate a clinical decision support system. Similarly to SemAssist, the Active Semantic Electronic Medical Record (ASEMR) is designed to use semantic annotations to facilitate documentation in health care.[35] Both SemAssist and ACEMR use NLP to produce semantic annotations that are leveraged by a clinical decision support system to alert healthcare workers if a rule is broken. These systems aim to reduce the time and effort for healthcare workers in providing necessary information to patients. For example, ASEMR resulted in both expediting and enhancing the patient documentation process by enabling clinicians to complete all of their patient documentation while the patient is still in the room.

10.4 Conclusion PCeHRs equipped with NLP capabilities hold the potential for empowering patients via patient-friendly, personalized language that is accessible and editable by both patients and healthcare workers. Statistical evaluations and benchmarks in shared tasks give evidence of NLP being successful in making text easier to understand and better personalized. However, more work is needed to strengthen the flow of these technologies to use in practice.

162   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Acknowledgement NICTA is funded by the Australian Government through the Department of Communications and the Australian Research Council through the ICT Centre of Excellence Program.

Contributor Statement All co-authors defined the chapter content. HS and MA wrote the manuscript and all co-authors reviewed and significantly edited the final version.

References [1] [2] [3] [4]

[5]

[6]

[7] [8] [9] [10] [11] [12] [13]

Cheevakasemsook A, Chapman Y, Francis K, and Davies C. The study of nursing documentation complexities. Int J Nurs Pract 2006;12(6):366–374. Hellesø R. Information handling in the nursing discharge note. J Clin Nurs 2006;15(1):11–21. Hyun S and Bakken S. Towards the creation of an ontology for nursing document sections: mapping section headings to the LOINC semantic model. AMIA Annu Symp Proc 2006:364–368. Adnan M, Warren J, Orr M. Assessing text characteristics of electronic discharge summaries and their implications for patient readability. In: Proceedings of the Fourth Australasian Workshop on Health Informatics and Knowledge Management. Brisbane: Australian Computer Society, Inc. 2010. p. 77–84. Allvin H, Carlsson E, Dalianis H, Danielsson-Ojala, R, Daudaravicius V, Hassel M, et al. Characteristics of Finnish and Swedish intensive care nursing narratives: a comparative analysis to support the development of clinical language technologies. J Biomed Sem 2011;2(Suppl 3):S1. McAllister M, Dunn G, Payne K, Davies L, and Todd C. Patient empowerment: The need to consider it as a measurable patient-reported outcome for chronic conditions. BMC Health Serv Res 2012;12:157. Bechhofer S, Carr L, Carole AG, Simon K, Timothy M. The semantics of semantic annotation. LNCS 2002;2519:1152–1167. Cimino JJ. Review paper: coding systems in health care. Methods Inf Med 1996;35(4–5):273–284. Cimino JJ, Zhu X. The practical impact of ontologies on biomedical informatics.Yearb Med Inform 2006:124–135. Clement WA, Wilson S, and Bingham BJ. A guide to creating your own patient–oriented web-site. J R Soc Med 2002;95(2):64–67. Hassling L, Babic A, Lonn U, and Casimir-Ahn H. A web-based patient information system – identification of patients’ information needs. J Med Syst 2003;27(3):247–257. Lemire M, Sicotte C, and Pare G. Internet use and the logics of personal empowerment in health. Health Policy 2008;88(1):130–140. Ilic D. The role of the Internet on patient knowledge management, education, and decisionmaking. Telemed J E Health 2010;16(6):664–669.

10 Technologies for patient-friendly personalized language   

   163

[14] Kummervold PE, Chronaki CE, Lausen B, Prokosch H-U, Rasmussen J, Santana S et al. eHealth trends in Europe 2005–2007: A population-based survey. J Med Internet Res 2008;10(4):e42. [15] Google receives 87.81 percent of Australian searches in June. Experian Hitwise. 2008 Jul 17. [16] Fox S. Health topics: 80 % of internet users look for health information online. [Internet]. DC, Washignton: Pew Research Center; 2011 [cited Dec 2014]. Available from: http://www.pewinternet.org/files/old-media/Files/Reports/2011/PIP_Health_Topics.pdf [17] Becker H. Computerization of patho-histological findings in natural language. Pathologia Europaea 1972. 7, 2, 193–200. [18] Nadkarni PM, Ohno-Machado L, and Chapman WW. Natural language processing: an introduction. J Am Med Inform Assoc 2011;18(5):544–551. [19] Demner-Fushman D, Chapman WW, and McDonald CJ.What can natural language processing do for clinical decision support? J Biomed Inform 2009;42(5):760–772. [20] Reiner B.Uncovering and improving upon the inherent deficiencies of radiology reporting through data mining J Digit Imaging 2010;23(2):109–118. [21] Sarkar IN.Biomedical informatics and translational medicine. J Transl Med 2010;8:22. [22] Liu K, Hogan WR, and Crowley RS. Natural Language Processing methods and systems for biomedical ontology learning. J Biomed Inform 2011;44(1):163–179. [23] Fernandez-Luque L, Karlsen R, Bonander J. Review of extracting information from the Social Web for health personalization. J Med Internet Res 2011;13(1):e15. [24] Uzuner O, Bodnari A, Shen S, Forbush T, Pestian J, and South BR.Evaluating the state of the art in coreference resolution for electronic medical records. J Am Med Inform Assoc 2012;19(5):786–791. [25] Meystre SM, Savova GK, Kipper-Schuler KC, Hurdle JF. Extracting information from textual documents in the electronic health record: a review of recent research. Yearb Med Inform 2008:128–144. [26] Mendonça EA, Haas J, Shagina L, Larson E, Friedman C. Extracting information on pneumonia in infants using natural language processing of radiology reports. J Biomed Inform 2005;38(4):314–321 [27] Pakhomov SV, Buntrock JD, Chute CG. Automating the assignment of diagnosis codes to patient encounters using example-based and machine learning techniques. J Am Med Inform Assoc 2006;13(5):516–525. [28] Suominen HJ, Salakoski TI. Supporting communication and decision making in Finnish intensive care with language technology. J Healthcare Eng 2010;1(4):595–614. [29] Chapman WW, Nadkarni PM, Hirschman L, D'Avolio LW, Savova GK, Uzuner O. Overcoming barriers to NLP for clinical text: the role of shared tasks and the need for additional creative solutions. J Am Med Inform Assoc 2011;18(5):540–543. [30] Warner D, Procaccino JD. Women seeking health information: distinguishing the web user. J Health Commun 2007;12(8):787–814. [31] Suominen H, Salanterä S, Velupillai S, Chapman W, Savova G, Elhadad N, et al. Overview of the ShARe/CLEF eHealth Evaluation Lab 2013. LNCS 2013;8138:212–231. [32] Suominen H, editor. The Proceedings of the CLEFeHealth2012 – the CLEF 2012 Workshop on Cross-Language Evaluation of Methods, Applications, and Resources for eHealth Document Analysis; 2012 Sep 17–20; Rome, Italy. Canberra, ACT, Australia: NICTA; 2012. [33] Adnan M, Warren J, Orr M. Iterative refinement of SemLink to enhance patient readability of discharge summaries. Stud Health Technol Inform 2013;188:128–134. [34] Adnan M, Warren J, Orr M. Ontology based semantic recommendations for discharge summary medication information for patients. In: The 23rd IEEE International Symposium on ComputerBased Medical Systems. Perth, WA, Australia: The Institute of Electrical and Electronics Engineers; 2010. p. 456–461.

164   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[35] Sheth A, Agrawal S, Lathem J, Oldham N, Wingate H, Yadav P, et al. Active semantic electronic medical record. LNCS 2006;4273:913–926. [36] Zeng-Treitler Q, Goryachev S, Kim H, Keselman A, Rosendale D. Making texts in electronic health records comprehensible to consumers: a prototype translator. AMIA Annu Symp Proc 2007:846–850. [37] Zeng-Treitler Q, Goryachev S, Weiss S, Sordo M, Murphy SN, Lazarus R. Extracting principal diagnosis, comorbidity and smoking status for asthma research: evaluation of a natural language processing system. BMC Med Inform Decis Mak 2006;6:30. [38] Cimino JJ, Del Fiol G. Infobuttons and point of care access to knowledge. In: RA G, editor. Clinical Decision Support: The Road Ahead. Amsterdam, the Netherlands: Elsevier; 2007. p. 345–371. [39] Baorto DM, Cimino JJ. An “infobutton” for enabling patients to interpret on-line Pap smear reports. AMIA Annu Symp Proc 2000:47–50. [40] Luo G, Thomas SB, Tang C. Intelligent consumer-centric electronic medical record. Stud Health Technol Inform 2009;150:120–124. [41] Miller T, Leroy G. Dynamic generation of a health topics overview from consumer health information documents. Int J Biomed Eng Technol 2008;1(4):395–414. [42] Elhadad N, Kan MY, Klavans JL, McKeown KR. Customization in a unified framework for summarizing medical literature. Artif Intell Med 2005;33(2):179–198. [43] Elhadad N. Comprehending technical texts: predicting and defining unfamiliar terms. AMIA Annu Symp Proc. 2006:239–243.

Jeff Harwell, Christopher Pentoney and Gondy Leroy

11 Finding and understanding medical information online The internet provides a wealth of information brought together by millions of people. Searching for information is one of the most frequent online activities. As a result, search engines have become tremendously efficient in gathering more information and presenting it quickly and efficiently. The information is available at the click of a button and presented in bite-size text snippets. There are two potential problems with today’s use of search engines that have an especially disturbing impact in health care. First of all, not all information is trustworthy, much is incomplete, biased or subjective, and finding a complete answer in response to a query is not facilitated by the search engine interface or the results presentation. Second, searching objectively is difficult. Only a few words are used to search among millions of documents, those are used by the search engine to guess, estimate, or decide what is relevant, and only matching information is returned even if the stated query is erroneous. Users, in turn, are often unaware of missing information. Many, especially younger generations, do not read entire texts but limit themselves to scanning snippets provided in the results list.[1] These limitations are problematic since these search engines inform and influence entire families and communities. This chapter will review query options for searching and presenting results and how these impact searching for and understanding of medical information by patients, caregivers and medical professionals.

11.1 Search in Medicine – Current State and Future Developments 11.1.1 Introduction The internet is the main provider of information for many people, and it continues to grow both in content and in reach with an estimated 2.7 billion people now having access.[2] Within this sea of information is a large amount of medical, healthcare or fitness information providing advice, background knowledge and opinions on treatments, diseases and conditions. Unfortunately, not all information is trustworthy, not all sources provide complete comprehensive and objective information, and many readers do not have the necessary skills to understand it. Inadequate health literacy, defined as the “limited ability to obtain, process, and understand basic health information and services needed to make appropriate health decisions and follow instructions for treatment”,[3] affects an estimated 89 million people in the United

166   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

States alone; this impedes their ability to understand and follow preventative care or treatments.[1] The cost of this lack of health literacy is estimated to be between $50 and $73 billion per year.[4] Comparable results are found internationally. In particular, low health literacy correlated with less understanding of their own condition[5, 6] makes people less likely to search for information[7] and less able to learn from information that they do read.[8] In the medical field, much attention has been given to understanding information and less to searching for it. However, searching for information is one of the most frequent online activities. We rely on search engines to find most of our information, with an estimated 19.3 billion searches conducted in the United States in October 2013 alone.[9] Non-mobile searching consists of navigational, transactional, but mostly (50–80%) informational queries,[4, 10, 11] while a recent study on European mobile search showed a much higher percentage (60%) of transactional queries.[12] 35% of U.S. adults, representing 72% of internet users, have gone online to research a medical concern, and 77% of those searchers started at a search engine.[13] The majority of medical searches by laymen are focused on finding information on a specific medical problem (55%) or a specific medical treatment (43%).[13] Overall, the quality of the medical information available on the internet is high; however, quality does vary widely with more general queries concerning preventative and social health issues returning lower quality information.[14] Additionally, Web 2.0 and the social web have introduced new challenges to the transparency and disclosure that are required to evaluate the trustworthiness of medical information available on the internet.[15] There are both dangers and advantages of these developments for patients and providers alike. The growing influence of search engines in how people gather and understand medical information makes it more of a concern that consumers have a limited grasp of healthcare vocabulary,[16] and younger people in particular often do not have the necessary skills to effectively use the internet to find health information. [17] Furthermore, consumer satisfaction with the internet as a source of medical information does not seem to be strongly related to their success in searching it.[18, 19] Finally, internet searches can cause unnecessary anxiety about common symptoms, cyberchondria, in individuals with little medical training.[20] More positively, there is evidence that minorities with internet access are able to engage in information searching to overcome inequality in levels of access to healthcare information.[21] As usage of the internet among laymen has grown, patients have been bringing the information that they have found into their relationship with their medical provider. Although medical professionals continue to be the most sought after and most trusted sources of medical information,[22] the availability of health related information on the internet has fundamentally changed the dynamic between patients and doctors; changing patients from more passive consumers to more active partners.[23] This in turn requires that doctors become more “net-friendly” and be able to counsel these engaged patients to contextualize the information that they bring from the internet and then guide them towards more reliable and useful sources

11 Finding and understanding medical information online   

   167

of information.[24] While the perceived quality and helpfulness of the patient’s communication with their doctor has no impact on how frequently patients engage in medical information seeking,[25] engaging in searching shows a strong positive impact on demand for healthcare services.[26]

11.1.2 Searching for health and medical information on the internet: needs, gaps and challenges In the context of the importance of online health information as discussed above, there is a need to understand, support and improve online medical information gathering and facilitate searcher understanding. The design of the search user interface is a crucial element in achieving this goal.

11.1.2.1 The need for better query expression In traditional search engines, queries are limited to a few keywords, often guided by auto-completion suggestions provided by the search engine. Hitwise reports that in the Spring of 2012 about 50% of US searches were only two words long and a little over 80% of searches were four words or fewer.[27] This limits the expressiveness and precision of queries as well as the diversity of information returned. Furthermore, the traditional approach to search interfaces forces users to search for matching (confirming) information, while ignoring alternative and contradictory information. For example, the query “salty diet autism” would retrieve documents on the causal relationship between salt and autism, but would ignore documents discussing the age of the parents as a potential causal factor. Query expansion, both interactive (IQE) and automatic (AQE) is a proposed solution to the limited amount of information users supply to the search engine. Although useful, most people do not like these automated methods[4] and thus many researchers try to coach more terms from them directly.[2, 11–13] IQE has been shown to be effective only if the user is making good query expansion decisions, which most users are unable to do.[28] Most IQE interfaces do not provide the user with enough context to enable them to make good decisions about the query expansion being offered. Without context, users tend to react conservatively and not choose expansion terms that they don’t fully understand.[29] In health informatics, this problem is compounded by the consumers’ limited grasp of medical vocabulary.[16] The application of semantic search techniques is a cornerstone technology in many efforts to improve the user’s query and thus both the accuracy and completeness of the information being returned by the search engine. The term “semantic search” is quite broad; it most generally means that the search engine is aware of some or multiple aspects of the context of a search and is able to use that context to increase the accuracy of the results. In many ways semantic search can be thought of

168   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

as a subset of the query expansion problem in which the search engine uses an ontology or other outside information to either modify the keywords in a search, expand or modify the search, or directly find documents using an ontological index. Mangold[30] lays out several criteria for classifying semantic search engines. These criteria include whether the system is designed as stand alone or as a meta search engine (architecture), the methods used to modify the user’s query (query modification), and the way in which the ontological data is stored and represented internally (ontology technology). Another criterion is coupling, or how explicitly the concepts in the source documentation must be tied to the ontology. Tightly coupled systems require that the source documents be semantically annotated so that it becomes very easy to resolve homonymies. Loosely coupled systems do not require semantic annotation but this limits the search capabilities of the system. An additional important criterion is transparency or how much the user is required to interact with the ontology. A transparent system does not expose the ontological categories to the user whereas a hybrid or interactive system might require the user to validate or clarify the semantic tagging of the query or the results. Regardless of the quality of terms, query expansion has a dramatic effect on the output of a search engine. For example, with Google, the number of distinct domains in the results as well as the ranking of results and domains changes significantly with the addition of suggested terms. Depending on the types of terms, expanded queries can become more specific or broader. A comparison of the social method and content-based method of query expansion, both using Google data, which involves expanding user’s queries with keywords from other successful searchers (social) or with words from the underlying available text corpus itself (content), shows that the social approach returns a smaller and significantly fewer number of distinct domains. [31] This suggests that the popular query suggestion tools may be narrowing the information being provided to the user.

11.1.2.2 The need for better information presentation Search result presentation has changed little over the last few decades. Early improvements included the use of text snippets, highlighting matching keywords in the snippets and, more recently, clearer distinctions between sponsored and other results. Most search engines limit themselves to providing a listing of results with highlighting in the text snippets of matching words. A few exceptions exist in which results are further processed to help organize and digest the information. For example, some search engines cluster results. Yippy.com (formerly clusty.com) shows a listing of results and the labeled clusters can be used to focus on subcategories. This uniform list-like format for reporting search results provides very little context for the results being returned. The search engine interface is not giving the user any feedback as to the appropriateness of her query in matching documents, quality of returned documents or missing information. In particular, no information

11 Finding and understanding medical information online   

   169

is presented on documents that contradict the query or suggestions of topics that are related to the query but would lead to an alternate result set. The cyberchondria phenomenon, which is the unfounded escalation of concerns over a common symptom after reading search results and material from the web, underscores the importance of providing context for medical searches.[20] Specific information given to those with little medical expertise can lead them to form incorrect conclusions as to their condition or the conditions of others.

11.1.2.3 Modern search challenge The modern search engine needs to address these deficiencies both in query formation and in the display of results. It needs to provide a user interface that encourages users to provide more information to the search engine, and the output of the search engine needs to communicate both specific results and the context of those results. The interface should help the searcher gain a broad and complete understanding of the topic in which they are searching in addition to finding the specific answer they are looking for. It is important that this context be communicated in a visual way, as opposed to a purely textual representation, as searchers are no longer reading long sections of text; instead, they are making judgments based on small snippets.[10] It is also important that the search engine take a more active and intelligent role in filtering out misinformation and assisting the searcher in reasoning with the information that is returned. With the rapid and increasing penetration of mobile platforms it becomes vital that the search interface be appropriate for the mobile experience; in particular, this means that the interface is even less dependent on keyboard usage and that it allows the query to be manipulated through touching or dragging. The interface must enable much more interactivity between the query and the search results all while remaining usable on the smaller screens of mobile devices.

11.1.3 Proposed solutions There is a great deal of energy focused on addressing the above challenges. The efforts range from fairly narrow solutions aimed at improving information access for patients and caregivers to ambitious question-answering systems designed to enhance the ability of experts to reason through the ever-growing amounts of medical information.

11.1.3.1 Patient centric solutions There are a number of efforts, many of which are in the commercial space, which have the explicit goal of helping patients and caregivers access, organize and understand medical information. They address many of the issues discussed above, particularly

170   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

problems of presenting only relevant medical information, by being very narrow in their scope and/or employing social networking techniques to enable users to work together to discover, filter and aggregate applicable content. Cake Health is focused on helping patients manage insurance and payment of their health care and alerting them to available health services that they are eligible for.[32] Meddik is a social networking startup that allows patients to organize a health timeline and then collaborate with one another around shared symptoms.[33] Smart Patients is another social networking startup that seeks to connect patients to each other and to information about medical trials. It includes a clinical trial search engine that allows patients to search for relevant clinical trials and then start discussions with others patients about the trial.[34] One of the largest patient-oriented services is Microsoft's Healthvault. It is a service designed to allow individuals to centralize all of their health information electronically and then share it with caregivers and medical professionals. Microsoft has put significant effort into making Healthvault the center of an electronic health record ecosystem. It launched in 2007 with partnerships that include the American Heart Association, Johnson & Johnson and the Mayo Clinic.[35] Microsoft has continued to expand this ecosystem as various vendors, such as Fitbit, introduce internet-enabled medical devices such as pedometers, scales, blood pressure monitors and blood glucose monitors. Spil and Klein[36] express doubts in the model and predict that Healthvault will succumb to the same factors that led to the failure of Google Health: primarily a lack of relevance to consumers, a lack of consumer trust and a perception of high risk.

11.1.3.2 General search engines The general web search engines, in particular Google and Bing, continue to make efforts to improve the relevance and accuracy of their results; this has an effect on medical information-seeking for many users. Bing acquired semantic search startup Powerset in 2007 and in 2009 announced integration of a semantically indexed version of Wikipedia and the ability to search using full questions.[37] In 2012, Google announced their Knowledge Graph which, as of 2012, was a 500  million object, 3.5 billion fact ontology.[38, 39] When Google announced the Hummingbird search algorithm, they implied that Google had begun to algorithmically expand the Knowledge Graph with web content in order to optimize the search engine for session-based conversational search.[40] Both Bing and Google are using transparent semantic search, meaning that the user interacts with the system just as they would a typical keyword-based search engine; all of the semantic searching then occurs with no further user interaction, and the user is unaware of the ontology or ontologies being utilized. Neither Bing nor Google release detailed descriptions of the technology driving their respective search engines. This, as well as other confounding factors, makes it difficult to determine

11 Finding and understanding medical information online   

   171

how these changes are impacting general medical search. However, as these search engines are the most common starting point for search on the internet, even small changes can potentially have a large impact on medical search on the web.

11.1.3.3 Research IQ There is significant ongoing work in academia to capitalize on the wealth of medical information now available in order to assist doctors and researchers. Research Integrative Query (Research-IQ), being developed at Ohio State University, is an attempt to construct a semantic search system that will allow non-technical medical practitioners to harness and reason with the large heterogeneous data sources that are now available. As described in 2011,[41] Research-IQ can be classified as a tightly coupled interactive semantic search system. An experienced knowledge engineer modeled the data dictionary from the Osteoarthritis Initiative (OAI) as a Unified Modeling Language (UML) class diagram. The data dictionary was then mapped to SNOMED-CT concepts and the resulting semantically annotated dictionary was modeled in the Apelon Distributed Terminology System. The Research-IQ system allows users to put in plain text queries that it then algorithmically annotates using MetaMap and presents the annotations to the users. The user then selects the valid annotations, at which point Research-IQ will return a list of OAI dictionary elements that were mapped to the user selected annotations. An evaluation of the relevance of the returned OAI elements by two subject matter experts showed that the majority of the returned elements were considered relevant to the expert's queries although the relevance was uneven.[42]

11.1.3.4 IBM Watson In 2011 Watson, a question and answer system created by IBM, defeated the two highest scoring players of all time in the game show Jeopardy! Soon after, IBM announced its intention to leverage the technology developed to create Watson in order to assist medical doctors in diagnosing and treating patients. In March of 2014, IBM and the New York Genome Center announced a partnership to test a prototype of Watson designed for genomic research to assist oncologists in treating cancer patients.[43] The system underlying Watson, DeepQA, is an abductive reasoning system. Abductive reasoning is the essential process of creating a hypothesis that would explain the observed facts if it is true.[44] DeepQA uses various natural language processing (NLP) techniques to create a hypothesis from a corpus of answer sources and then uses hundreds of NLP algorithms to either support or refute each hypothesis using a variety of content sources. The evidence is weighed using an ensemble of machine-learning algorithms that then present a ranked set of hypotheses to the user. A claimed advantage of Watson is that it is designed to do automated semantic reasoning on natural language content giving it the capabilities of a tightly coupled semantic search engine without requiring semantic annotation of the source docu-

172   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

ments. Another unique ability of DeepQA is that it can alert users to facts that are missing from the current case which, if they were present, would allow the system to better diagnose the symptoms in question.[45] As of 2012, many modifications have been made and evaluated so that Watson would work more effectively in the medical environment. These adaptations include adding a number of medical content sources, including standard medical texts such as the Merck Manual of Diagnosis and Therapy and the Medical Knowledge Self-Assessment Program (MKSAP) which were analyzed for medical concepts and semantic type using the Unified Medical Language System (UMLS). There were also numerous functional adaptations including adding the Medical Subject Heading (MeSH) system from U.S National Library of Medicine and SNOMED medical taxonomies from the International Health Terminology Standards Development Organization (IHTSDO), creating a rule-based annotator to recognize measurements and map them to concepts, and changing the way that Watson extracted supporting passages from the medical texts. DeepQA was trained on a random selection of 1322 questions from the American College of Physicians (ACP) Doctor's Dilemma competition. The system was then evaluated on 188 unseen Doctor's Dilemma questions. It selected the correct answer with an accuracy of 50.5% and the correct answer was in the top ten suggestions 77% of the time.[45]

11.1.3.5 Diagram-based search Research on search engines and improving search and the understanding of results continues. We are proposing a search engine that leverages a new diagram-based interface for both search and depicting results. A basic node of the diagram consists of a box containing a word or short phrase. Two nodes can be linked together and the relationship between them described with a third term. In this fashion, each node can be related to one or more other nodes in the diagram. The diagram is a visual way of describing the relationship between multiple subject, predicate and object entities. When displaying results, the interface provides a list of documents from the corpus that best match the query, and it will modify the diagram by suggesting additional terms that the corpus suggests are relevant to the search near some nodes in the diagram. Additionally, the interface will thicken and/or color diagram lines to indicate the strength of the relationship between nodes as suggested by the corpus. The interface will also allow the user to highlight one or more of the nodes and dynamically change the ranking of the list of results being returned.

Backend Index The diagram-based search interface allows the user to describe relationships between terms – but to be effective, the interface requires a document index that contains relationships extracted from documents, not just keywords. This index is built by a

11 Finding and understanding medical information online   

   173

triple parser that parses documents and then uses a combination of Support Vector Machines and Finite State Automata to identify the meaningful triples in the documents. This parser is able to achieve more than 91% precision and recall against a gold standard when parsing biomedical abstracts.[46–48]

Query formation In order to search using the diagram interface, users will be able to touch/ click to add a new node or drag additional nodes. As shown in Figure 1, each node can contain a search term, and the links between the nodes are directional and can contain additional search terms. In this example, the user is searching for information on salt as the cause of autism. Queries formed by cancer researchers using query diagrams had significantly higher precision and recall than a keyword search.[48] A pilot study performed with laymen showed that users can easily understand the search diagram interface and that – when using the interface – they include more keywords and relationship than when they perform a keyword search.[49]

Figure 1: Screenshot of Query Diagram Interface

11.2 Discussion Watson and other sophisticated semantic search tools are aimed at helping professionals reason over the large amounts of medical information available. While there is a great deal of excitement and potential in these systems, proving their efficacy in the clinical setting is ongoing work. At present, systems like Watson or ResearchIQ are targeted almost exclusively at medical professionals; due to challenges in healthcare literacy, these systems would be unsuitable for many patients and caregivers. It is notable that systems like Watson also have a strong reliance on electronic health records (EMR), as it would be impractical to type in all of a patient’s symptoms and history before using Watson in a clinical setting. EMR continues to struggle with issues of standardization and consistency within the healthcare profession and challenges with consumer perceptions of relevance from laymen. Health Informatics systems designed to resource patients and caregivers tend to have a very strong social web component and often take the form of utilities that assist with the organization of

174   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

personal health care data. Systems such as ours attempt to assist patient and patient communities in reasoning over the data available to them in a comprehensible way while avoiding problems such as cyberchondria – but much work remains to be done to prove effectiveness for the general population. The large public search engines continue to improve their ability to provide more sophisticated answers and insights into user queries. However, given both the domain specificity of health informatics and the broad focus of the public search engines, it is reasonable to assume, in the short term, that the public search engines will continue to function primarily as methods of connecting patients and caregivers with social communities and domain-specific systems that address their specific context. Where information is often incomplete or simply incorrect, systems designed specifically for presenting valid medical knowledge in a contextualized manner are necessary. This factor is the primary difference between systems designed for use by medical professionals in searching professional, vetted literature and broad systems used by laymen for searching broad heterogeneous data sources. Inefficient, incorrect or misguided queries created by users with a limited understanding of the medical field make information quality problems more damaging and difficulty to address. Retrieved health information needs to be trustworthy, and the technology should be helpful in displaying knowledge and guiding users to the information that is going to help them. This is a constantly evolving area that requires ongoing assessment. Careful consideration must be made regarding the limitations of these technologies especially with the hype and aggressive marketing that often accompanies new medical technology initiatives. One danger is that people will become overly reliant upon the technology. For example, in earlier work, visual tables of contents led users to rely on the visualization while they ignored the underlying text, leading to reduced understanding of the information.[8] Especially with health information, a focus on safety must be maintained. Care must be taken in the presentation of results so that users are easily able to understand what has been displayed. Some of the dangers can be mitigated with proper evaluation, but the proper evaluation of information systems is a difficult, time-consuming and expensive task. As with any up-and-coming technology, mobile implementation is a major consideration. Fast and convenient mobile access to information is quickly moving from a luxury to a need as consumer expectations change. It is already a key factor in user’s adoption of an information system. The proliferation of low cost mobile devices also represents an opportunity to provide additional healthcare information and resources to historically underserved populations, as minorities and low socioeconomic status groups are able to leverage online resources to compensate for their lack of access to healthcare resources.[21] This underscores the importance of providing a powerful but intuitive search interface for mobile users that enhances their ability to search for and understand medical information.

11 Finding and understanding medical information online   

   175

11.3 Conclusion Given the correlation between health and health literacy, the ever-increasing amount of health and medical information available on the internet, and the impact of this information on both the patient’s understanding of their own health and the doctor/ patient relationship, the importance of medical search has never been greater. And with increasing use and reliance on mobile technology, no solution should be put forward that cannot be used on different types of devices and by different consumers. The systems discussed in this chapter use a wide variety of techniques to address issues such as poor term selection in queries and the user’s unwillingness to read large sections of text as they search. Through innovations in presentation, query assistance, semantic search, data storage and indexing, visualization and social networking technology, these systems attempt to minimize misunderstanding and misrepresentation and amplify the ability of professionals and laymen to understand and reason with the large amounts of information now available to them. Although much work remains to be done, particularly in evaluating the effectiveness of these systems, medical search technology continues to make strides towards being both easier to use and providing a richer context. Aided by these innovations patients can increase their health literacy and become more empowered healthcare consumers, and medical professionals can more effectively leverage the ever-growing body of research and professional literature in treating patients and moving the medical field forward.

References [1] [2]

[3] [4] [5]

[6] [7] [8]

Surowiecki J. The wisdom of crowds: Why the many are smarter and how collective wisdom shapes business, economies, societies, and nations. New York: Random House; 2004. The World in 2013, ICT Facts and Figures [Internet]. International Telecommunication Union [cited 2014 Nov 15]; Available from: http://www.itu.int/en/ITU-D/Statistics/Documents/facts/ ICTFactsFigures2013.pdf Bodner RC, Chignell MH. ClickIR: Text retrieval using a dynamic hypertext interface. TREC. Citeseer; 1998. p. 506–15. Jansen BJ, Booth DL, Spink A. Determining the user intent of web search engine queries. Proceedings of the 16th international conference on World Wide Web. ACM; 2007. p. 1149–50. Finkelstein L, Gabrilovich E, Matias Y, Rivlin E, Solan Z, Wolfman G, et al. Placing search in context: The concept revisited. Proceedings of the 10th international conference on World Wide Web. ACM; 2001. p. 406–14. Jansen BJ, Spink A, Saracevic T. Real life, real users, and real needs: a study and analysis of user queries on the web. Inf Process Manag 2000;36(2):207–27. McCray AT, Tse T. Understanding search failures in consumer health information systems. AMIA annual symposium proceedings. American Medical Informatics Association; 2003. p. 430. Leroy G, Miller T. Perils of providing visual health information overviews for consumers with low health literacy or high stress. J Am Med Inform Assoc 2010;17(2):220–3.

176   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[9] comScore Releases October 2013 U.S. Search Engine Rankings [Internet]. comScore; 2013. Available from: http://www.comscore.com/Insights/Press_Releases/2013/11/comScore_ Releases_October_2013_US_Search_Engine_Rankings [10] Broder A. A taxonomy of web search. ACM Sigir forum. ACM; 2002. p. 3–10. [11] Purcell K, Brenner J, Rainie L. Search engine use 2012 [Internet] Washington: Pew Research Center; 2012 [cited 2014 Nov 15 ]. Available from: http://www.pewinternet.org/2012/03/09/ search-engine-use-2012 [12] Church K, Smyth B, Bradley K, Cotter P. A large scale study of European mobile search behaviour. Proceedings of the 10th international conference on Human computer interaction with mobile devices and services. ACM; 2008. p. 13–22. [13] Fox S, Duggan M. Health Online 2013 [Internet]. Pew Research Center; 2013 Jan [cited 2014 Nov 15]. Available from: http://www.pewinternet.org/2013/01/15/health-online-2013 [14] Kitchens B, Harle CA, Li S. Quality of health-related online search results. Decis Support Syst. 2014;57(0):454–62. [15] Adams SA. Revisiting the online health information reliability debate in the wake of “web 2.0”: An inter-disciplinary literature and website review. Spec Issue Inf Technol Health Care Socio-Tech Approaches 2010 Jun;79(6):391–400. [16] Zeng QT, Tse T. Exploring and developing consumer health vocabularies. J Am Med Inform Assoc 2006;13(1):24–9. [17] Kim H, Park S-Y, Bozeman I. Online health information search and evaluation: observations and semi-structured interviews with college students and maternal health experts. Health Inf Libr J 2011;28(3):188–99. [18] Zeng QT, Kogan S, Plovnick RM, Crowell J, Lacroix E-M, Greenes RA. Positive attitudes and failed queries: an exploration of the conundrums of consumer health information retrieval. Int J Med Inf 2004;73(1):45–55. [19] Zeng QT, Crowell J, Plovnick RM, Kim E, Ngo L, Dibble E. Assisting consumer health information retrieval with query recommendations. J Am Med Inform Assoc 2006;13(1):80–90. [20] White RW, Horvitz E. Cyberchondria: studies of the escalation of medical concerns in web search. ACM Trans Inf Syst TOIS 2009;27(4):23. [21] Mesch G, Mano R, Tsamir J. Minority status and health information search: A test of the social diversification hypothesis. Soc Sci Med 2012;75(5):854–8. [22] Mayer DK, Terrin NC, Kreps GL, Menon U, McCance K, Parsons SK, et al. Cancer survivors information seeking behaviors: A comparison of survivors who do and do not seek information about cancer. Patient Educ Couns 2007;65(3):342–50. [23] McMullan M. Patients using the Internet to obtain health information: How this affects the patient–health professional relationship. Patient Educ Couns 2006;63(1–2):24–8. [24] Wald HS, Dube CE, Anthony DC. Untangling the Web—The impact of Internet use on health care and the physician–patient relationship. Patient Educ Couns 2007;68(3):218–24. [25] Xiao N, Sharman R, Rao HR, Upadhyaya S. Factors influencing online health information search: An empirical analysis of a national cancer-related survey. Decis Support Syst 2014;57(0):417–27. [26] Suziedelyte A. How does searching for health information on the Internet affect individuals’ demand for health care services? Soc Sci Med 2012;75(10):1828–35. [27] Experian Hitwise, Bing, Google, Powered Searches, April 2012 [Internet]. [cited 2014 Mar 5]. Available from: http://press.experian.com/United-States/Press-Release/experian-hitwisereports-bing-powered-share-of-searches-at-30-percent-in-april-2012.aspx [28] Magennis M, van Rijsbergen CJ. The potential and actual effectiveness of interactive query expansion. ACM SIGIR Forum. ACM; 1997. p. 324–32.

11 Finding and understanding medical information online   

   177

[29] Ruthven I. Re-examining the potential effectiveness of interactive query expansion. Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval. ACM; 2003. p. 213–20. [30] Mangold C. A survey and classification of semantic search approaches. Int J Metadata Semant Ontol 2007;2(1):23–34. [31] Harwell J, Pentony C, Leroy G. Big Data for Query Expansion: A Comparison of Contentbased versus Socail-based Keywords. Snowbird, Utah; 2014. Available from: http://gkmc. utah.edu/winter/sites/default/files/webform/abstracts/131202_WCBI_Query_ExpansionAbstract-20140226.pdf [32] Kincaid J. Cake Health Wants To Be The “Mint For Health Insurance” (Beta Invites) | TechCrunch [Internet]. Techcrunch. 2011 [cited 2014 May 1]. Available from: http://techcrunch. com/2011/05/13/cake-health-wants-to-be-the-mint-for-health-insurance-beta-invites [33] Learn and share with people who’ve been there | Meddik [Internet]. Meddik. [cited 2014 May 1]. Available from: https://www.meddik.com/# [34] Smart Patients [Internet]. Smart Patients. [cited 2014 May 1]. Available from: https://www. smartpatients.com [35] Lohr S. Microsoft Rolls Out Personal Health Records – New York Times [Internet]. The New York Times. 2007 [cited 2014 May 1]. Available from: http://www.nytimes.com/2007/10/04/ technology/04nd-soft.html?_r=0 [36] Spil T, Klein R. Personal Health Records Success: Why Google Health Failed and What Does that Mean for Microsoft HealthVault? System Sciences (HICSS), 2014 47th Hawaii International Conference on. IEEE; 2014. p. 2818–27. [37] Johnson M. Researching With Bing Reference – Search Blog [Internet]. Bing Blogs. 2009 [cited 2014 May 1]. Available from: http://www.bing.com/blogs/site_blogs/b/search/ archive/2009/07/27/researching-with-bing-reference.aspx [38] Singhal A. Introducing the Knowledge Graph: things, not strings – Inside Search [Internet]. Insite Search – The Official Google Search Blog. 2012 [cited 2014 May 1]. Available from: http://insidesearch.blogspot.com/2012/05/introducing-knowledge-graph-things-not.html [39] Stewart D. Google’s Knowledge Graph: Yeah, that’s the Semantic Web (sort of) [Internet]. Gartner Blogs. 2012 [cited 2014 May 1]. Available from: http://blogs.gartner.com/darinstewart/2012/05/17/googles-knowledge-graph-yeah-thats-the-semantic-web-sort-of [40] Sullivan D. FAQ: All About The New Google “Hummingbird” Algorithm [Internet]. Search Engine Land. 2013 [cited 2014 May 1]. Available from: http://searchengineland.com/googlehummingbird-172816 [41] Osteoarthritis Initiative [Internet]. OAIOnline. [cited 2014 May 2]. Available from: http://oai.epiucsf.org/datarelease/default.asp [42] Borlawsky TB, Lele O, Payne PRO. Research-IQ: Development and evaluation of an ontologyanchored integrative query tool. J Biomed Inform 2011 Dec 1;44:S56–S62. [43] IBM News room – 2014-03-19 The New York Genome Center and IBM Watson Group Announce Collaboration to Advance Genomic Medicine – United States [Internet]. 2014 [cited 2014 Apr 30]. Available from: http://www-03.ibm.com/press/us/en/pressrelease/43444.wss [44] Pople HE. On the Mechanization of Abductive Logic. IJCAI. 1973. p. 147–52. [45] Ferrucci D, Levas A, Bagchi S, Gondek D, Mueller ET. Watson: Beyond Jeopardy! Artif Intell 2013;199:93–105. [46] Kwak M, Leroy G, Kim M. Development and evaluation of a triple parser to enable visual searching with a biomedical search engine. Int J Biomed Eng Technol 2012;10(4):351–67. [47] Kwak M, Leroy G, Martinez JD. A pilot study of a predicate-based vector space model for a biomedical search engine. Bioinformatics and Biomedicine Workshops (BIBMW), 2011 IEEE International Conference on. IEEE; 2011. p. 1001–3.

178   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[48] Kwak M, Leroy G, Martinez JD, Harwell J. Development and evaluation of a biomedical search engine using a predicate-based vector space model. J Biomed Inform 2013;46(5):929–39. [49] Leroy G. Persuading consumers to form precise search engine queries. AMIA Annual Symposium Proceedings. American Medical Informatics Association; 2009. p. 354. [50] Fernandez-Luque L, Karlsen R, Krogstad T, Burkow TM, Vognild LK. Personalized health applications in the Web 2.0: the emergence of a new approach. Engineering in Medicine and Biology Society (EMBC), 2010 Annual International Conference of the IEEE. IEEE; 2010. p. 1053–6. [51] Polikar R. Ensemble based systems in decision making. Circuits Syst Mag IEEE 2006;6(3):21–45.

Yukari Schneider, Maria Adela Grando, Jihad S. Obeid and Wajeeh Bajwa

12 Electronic media for engaging patients in the research consent decision process Abstract: Informed consent has long been one of the central ethical tenets in human subjects’ research, and it remains a complex challenge for investigators and regulatory authorities today. Those difficulties include meeting the ethical principles outlined in the Belmont Report (respect for persons, beneficence and justice) while simultaneously creating an effective consent process that is informative, non-coercive and comprehensible to every potential research subject. Given rapid technological advances and the omnipresent internet in today’s world, the use of electronic mechanisms to obtain informed consent (eConsent) has become a subject of great interest and is rapidly being adopted. While eConsent may provide some new solutions to traditional informed consent issues, it also creates a host of new problems and challenges. This chapter discusses the various issues involved with the utilization of eConsent in human subject research and offers recommendations to improve procedural effectiveness and mitigate ethical challenges. Topics covered include the forms and media through which information is presented to the potential participants in a research study as well as the wide range of factors that may affect their participation including comprehension, voluntariness, literacy and length of the consent document.

12.1 Introduction The topics of electronic consent (eConsent) and the use of multimedia aids in the informed consent process have been under discussion since the late 1990s. The demand for an efficient eConsent process is increasingly significant given the rapid development and ubiquitous availability of electronic devices such as laptop computers, tablets and smartphones combined with widespread internet connectivity. In today’s online world, sensitive and private information can be delivered to end users (research subjects) in a secure and efficient manner, although such delivery can often involve new and unique information security challenges. Nonetheless, the continuous evolution and worldwide adoption of such technologies are already having a clear impact on human subject research and informed consent procedures. This impact will require constant review and relevant adaptability by both researchers and regulatory agencies. There are multiple challenges and benefits of using eConsent for both participants and investigators that need to be considered. In addition to the challenges in traditional informed consents including participants’ comprehension and literacy,

180   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

voluntariness, coercion, confidentiality and the need to strike a balance between full disclosure and overwhelming information, there is the issue of access to electronic media. Portions of the population remain, particularly among the elderly and socioeconomically-disadvantaged, that may not have access to or fluency with electronic media, making an eConsent process more difficult for them. The benefits of eConsent on the other hand for both participants and research coordinators include convenience and the potential for improved comprehension and compliance. This chapter will discuss these issues in detail and provide suggestions for appropriate adoption of the new technology into existing informed consent workflow.

12.2 The current situation: needs, gaps and challenges In as early as 1900, Prussia instated the practice of securing informed consent from any persons participating in medical research.[1] This practice was initiated after a physician injected an experimental syphilis vaccine into his patients who were being treated for unrelated conditions. Tragically, the vaccine failed and infected his unwitting patients, altering the quality of their lives forever. Subsequently, informed consent has been one of the central tenants of all modern discussions about ethical research practices. Regulatory highlights concerning the protection of human subjects in medical research include the Nuremberg Code in 1947, which served as a standard for building criminal cases against physicians who participated in the Nazi human experiments, and the Belmont Report in 1974, which was developed in reaction to the Tuskegee Syphilis Studies and also led to the establishment of the Institutional Review Board in the United States.[2] Sadly, each of these ethical milestones has come about as reactions to failure in previous systems to elicit voluntary consents from human participants. Informed consent is defined as “the person’s voluntary agreement, based upon adequate knowledge and understanding, to participate in human subjects’ research or undergo a medical procedure”.[3] Regarding, specifically, informed consent for participation in human subjects’ research, The Belmont Report, which continues to guide the policies of the National Institutes of Health (NIH) Office of Human Subject and Research (OHSR), such is bound to three ethical principles: (1) respect for persons; (2) beneficence; and (3) justice.[4] The first principle, respect for persons, requires researchers to recognize each individual as autonomous agents, whose participation is solely determined by their voluntary and correctly informed decision to do so. The second principle, beneficence, holds researchers to, first, not doing harm and working towards maximizing potential benefits while minimizing potential harm when outcomes are less clear. Finally, the third principle, justice, requires researchers to carefully weigh and disclose groups of people who are predicted to bear the benefits and burdens of their research, primarily because medical studies typically

12 Electronic media for engaging patients in the consent process   

   181

benefit future patients and not participants, and benefits for researchers could potentially lead to conflicts of interests. These three tenants of medical research encompass the concept of informed consent. Given the importance of obtaining informed consent in accordance with these principles of respect for persons, beneficence and justice, an effective and welldesigned consent process is absolutely necessary. Essential components of the informed consent process include: (a) information, (b) comprehension and (c) voluntariness. (a) Information Current national guidelines enforced by the NIH OHSR require the following minimum components in an informed consent form:[4, 5] – Description of the research procedure; – The purpose of the procedure; – Associated risks and benefits; and – Alternative solutions. Consent procedures must also provide the opportunity for participants to ask questions and gain more knowledge related to proposed studies. The specific information that is to be communicated is outlined in the informed consent form that is regulated at local levels by Institutional Review Boards (IRB). The IRB is an administrative body charged with protecting the rights and welfare of research subjects through a formalized process of reviewing, approving and monitoring any biomedical and behavioral research involving human subjects.[3] Common criticisms of research consent forms during IRB review include excessive length, which challenges the ability of potential subject to process it in a reasonable amount of time, and unnecessarily complex language and legal jargon that make it difficult to read and comprehend. Issues such as these significantly limit a specific consent form’s ability to accomplish the primary objective of securing truly informed consent.[6, 7] (b) Comprehension Researchers must adapt the information they hope to convey into a format that is appropriate for people’s capacity to understand. Moreover, researchers are obligated to confirm their participants’ comprehension of consent information. However, guidelines note that an evaluation of participants’ understanding is appropriate but not mandatory. Confirming comprehension of pertinent study information remains an operational challenge because of its abstract, value-laden nature.[8, 9] Additional difficulties are posed by the likelihood of significant variation in literacy and educational level among potential subjects. The use of electronic media has the potential to enhance comprehension by improving the educational content within the informed consent. Improved comprehension empowers patients to make better decisions regarding participating or volunteering for research.

182   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

(c) Voluntariness Voluntariness refers to the fact that people have the right to equally refuse or join research studies without consequence. They may also leave the studies at any time, again, without consequence.[4] Consent to participate is not valid unless decisions to join are based on personal choice free of undue pressure or inducement. Ambiguous delineation between ethical and unethical forms of persuasion complicate the issue of securing voluntary consent, especially in light of the power unbalance between stakeholders. Next, we elaborate on the main challenges in the informed consent process.

12.2.1 Participant’s low literacy and form’s length According to mainstream discussions of the consent process, full disclosure of medical information through traditional consent forms remains a challenge because of the complex medical concepts being communicated coupled with low literacy rates among American adults.[7] The current average reading literacy among American adults is at the eighth grade level.[10] Notably, 75% of adults with long-term illnesses also fall below this national average.[11] So, medical researchers are assigned with a difficult task of communicating complex information to populations with significant barriers to comprehension. In a review of the informed consent templates of 114  medical IRB, researchers found that most templates were written at an average of the tenth grade level. This average is approximately three grades above the IRB’s self-recommended standard. [7] This disparity between consent forms and literacy levels of potential study participants has fueled a growing body of research suggesting that participants frequently do not comprehend consent information. For example, one study reported that 70% of participants in an oncological clinical trial did not know that the experimental treatment had not been proven to be effective, and 25% did not know that the trial findings would mostly benefit future patients, not study participants.[12] Even in a clinical trial that largely enrolled participants with college-level educations, only 12% of participants (n = 82) could name the three trial drugs and only 17% could recall three or more out of 23 possible risks.[13] A recent study found that the length of US and international consent forms for NIH-sponsored HIV trials were an average of 22 pages, which exceeded recommendations for how much information could be reasonably processed.[14] The authors suggested highlighting key information to enhance the reading process, but also noted the ethical dilemma of identifying the appropriate information to highlight. Additionally, a study evaluating how different stakeholders value information on a consent form found that the IRB officials identified most of the form as important (72 % of the sentences). But in a telling contrast, participants thought that less than half of the form was vital (40% of the sentences).[6] Importantly, informed consent forms have

12 Electronic media for engaging patients in the consent process   

   183

also been described by patients as symbolic rituals for protecting medical doctors and institutions rather than vehicles for ensuring patient autonomy.[15] Therefore, not only are medical consent forms lengthy and difficult to understand, but a discrepancy exists between what different stakeholders perceive to be the true objective of the form and the way they distinguish meaningful and necessary information.

12.2.2 Participant’s desire to have more control over what is shared Recent studies suggest that subjects want more granular control over how their clinical data is shared for research. It seems that subjects’ attitudes toward sharing information for research could be influenced by multiple factors:[16] 1. Type of information: subjects are less willing to share information that becomes more personal, for instance sensitive information related to drug abuse, sexualrelated diseases, etc.[16] 2. The identity of the recipient of the information: support for sharing medical record data decreases when suggested users included commercial, for-profit recipients.[17, 18] 3. Level of anonymity: subjects have privacy concerns and are more prepared to share anonymous information.[19] 4. Demographics of the subject: subjects with chronic or progressive illnesses seem to be more willing to share.[17, 20] 5. Perceived value: the emergence of highly active virtual research communities of persons affected by similar conditions seems to be based on subjects’ perceived value of peer-based sharing models and their hopes of accelerating the prevention, treatment and cure of those conditions.[21] Though better understanding of attitudes and willingness to engage in public health research is needed, the mentioned studies suggest that the use of broad consent models should be rethought to favor tiered consent models that provide subjects greater choice in information sharing. Furthermore, support for the idea of giving patients granular control is also encouraged by the U.S. Office of the National Coordinator for Health Information Technology, which indicated that patients should have a “greater degree of choice to determine, at a granular level, which personal health information should be shared with whom, and for what purpose”.[22]

12.2.3 Participant’s interest to know research results There is an open debate on the scope and limits of investigators’ responsibilities on sharing research results. Some advocate for routinely offering results to research participants,[23, 24] and there are national and international policies and guidelines on

184   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

the duty to return results.[25] Furthermore, the review by Shalowitz and Miller has shown that of studies that reported desire to receive results as a percentage of respondents, a median of 90% (range 20%–100%) wished to receive study results.[26] In addition, it was reported that those who were responsible for clinical trials considered it was important to return those results.[26] Nevertheless, it has been reported that clinical trial participants are often not offered the option of receiving the trial results, or are not provided those results when offered,[27] though there is evidence that the process of preparing and disseminating clinical trial results could be feasible and executable.[28] To address the problem that clinical trial participants are not offered access to trial results, in 2011, the U.S. FDA required informed consent forms to include a statement informing the participants that the results of the trial will be uploaded in the ClinicalTrials.gov registry. While the adoption of the registry significantly increased the number of returned trial results,[29] it did not help to address the problem of facilitating participant’s access to those results.[30] Even if the subjects were able to access to the registry and found the results returned by the trial they participated in, they could not necessarily interpret and understand those results. While the registry links clinical trials with associated scientific publications, those publications are meant for the scientific community and therefore cannot be easily understood by the subjects. By not disclosing results in ways that are meaningful for the trial participants, the research community could be losing the opportunity to build trust with participants.

12.2.4 Attitudes of both medical researchers and participants toward informed consent processes A study in Austria interviewed stakeholders of a bio-repository study that collected human tissue for research in an effort to take into account broader factors that influenced patient decisions. They found that medical researchers viewed informed consent primarily as a process to secure legal research rather than an issue of assuring patients’ freedom of choice.[15] Comparatively, researchers who also entertained ethically ideal concepts of patient autonomy appeared to justify perfunctory consents by holding to assumptions that patients were not willing to receive in-depth information: “Often…it [the information] challenges patients very much intellectually. Why should they be interested in something that doesn’t really interest them… 10, 15 pages, no patient wants to read that, definitely not [sic.]“.(p. 94)[15] In the same study by Felt and colleagues (2009), patients were interviewed following consent procedures. Researchers found that many patients had low interest in learning from the consent process because they were more interested in broader notions of contributing to medical innovation than the study directly at hand. Some patients also avoided asking extra questions about the study because they were concerned about putting further demands on their medical provider or that refusing a

12 Electronic media for engaging patients in the consent process   

   185

study could cause conflict and complicate their medical care. Another study examining patient preferences for autonomy found that, although they wanted to be informed, patients preferred medical decisions to be made primarily by their medical providers, especially as the severity of the illness increased.[31] Such findings demonstrate how informed consent goes beyond the scope of providing accurate and comprehensible information but also delves into realms of social obligation and influence.

12.2.5 Patient autonomy and social forces of influences The informed consent process for medical research is not normatively accepted as a persuasive process in medical literature but as a method for ensuring patient autonomy and understanding of pertinent study features.[32, 15] Indeed, the Belmont Report clearly states, “Informed consent requires conditions free of coercion and undue influence” (Part C, paragraph 11).[4] However, it later brings to light a paradoxical issue by noting that “unjustifiable pressures usually occur when persons in positions of authority or commanding influence – especially where possible sanctions are involved – urge a course of action for a subject” (Part C, paragraph 12). This is a paradoxical issue because medical experts exercise significant influence de facto as they habitually care for and enroll people in vulnerable states.[32, 15] Consequently, persuasive forces are inherent to the informed consent process as Felt et al. (2009) aptly described: “power relationships between medical practitioners and patients lead to unequal encounters wherein personal autonomy cannot be realized in the way it is imagined by the dominant bioethical ideal”. (p. 89) The following discussion further demonstrates how “coercion” and “undue influence” are difficult concepts to define and may be overlooked when consent procedures are examined through the broader context of medical research and recruitment. [15] For example, this statement by a medical investigator interviewed in a qualitative examination of consent processes demonstrates how recruitment and consent communication are not only intentional but also framed to elicit more favorable attitudes: “You want to actively recruit people for a study, which means that you make things attractive. You don’t stand there and tell them that it might be dangerous for them. Then, maybe, you wouldn’t find anybody.” (p. 94)[15] This statement does not represent the views of all medical researchers, but it brings to light the fundamental motivation for researchers to employ persuasive strategies in order to actively recruit participants and to move medical investigations forward. Moreover, for persuasion to take place, message recipients must be free to adopt attitudes and behaviors that are contrary to the advocated positions.[33, 34] Otherwise, the change in attitude cannot be attributed to persuasive appeals of the communication. Nevertheless, no one is truly free of influence by virtue of the social nature of humanity. For example, a medical student was asked to reflect on the issue of freedom of choice and a patient’s ability to refuse a physician. The student stated, “You won’t

186   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

refuse a request from somebody who is helping you”.(p. 94)[15] Inadvertently, the student touched on a critical force of social influence. Indeed, Robert Cialdini (1984), an eminent scholar in persuasion research, expounds that mutual exchange or reciprocity is a fundamental driver of human behavior that defies all social and cultural boundaries.[35] Essentially, people are “programmed” to feel obligated to return favors. Thus, patients may feel obligated to join medical studies for helpful medical practitioners that they otherwise would not. Such instances are not necessarily problematic, but it shows how the “atmosphere of free choice” is difficult to construe and actualize in the research recruitment context and should be taken into consideration in effective informed consent processes.

12.3 Proposed solutions 12.3.1 Participant’s low literacy and form’s length Beyond recall-testing interventions for informed consent, several approaches for improving comprehension of consent materials have been investigated in the literature. Improving the readability and length of forms – a popular approach – has been shown to significantly boost comprehension of consent information.[36] Readability refers to how easily consent forms could be read, and it is intentionally enhanced by using simpler words, writing in active voice and organizing the form in a more appealing manner (i.e. using bullets, increasing white space, etc.). Simply adding pictures to spoken instructions has also been shown to improve recall of medical information.[37] Moreover, a large body of research supports the usefulness of multimedia for enhancing comprehension; however, its adoption rate appears to be hampered by its involving development process.[38, 39] Despite the existence of these helpful strategies to improve comprehension, a 2004 systematic review of consent interventions determined that the most effective way to secure consent was to continue using faceto-face discussion between neutral educators and participants.[8] Therefore, consent interventions including electronic systems, eConsents and multimedia should be viewed as complimentary components to the human-mediated consent process rather than a stand-alone program.

12.3.2 Participant’s desire to have more control over what is shared Medical records provide a rich source of information for advancing clinical care through research, particularly with the advent of electronic health records (EHR).[40] As a result, secondary use of EHR for health outcomes research has become prevalent. However, patient control over sharing of their personal medical records is rare. Many healthcare organizations use broad consent forms for future research involving a patient’s medical

12 Electronic media for engaging patients in the consent process   

   187

data. Biological specimens extracted from patients during clinical care also provide an important resource for research especially in the genomic era. However, the use of these specimens for research requires explicit consent from research participants. The University of California San Diego (UCSD) has developed a tool called iCONS (informed CONsent for clinical records and Sample use in research) that allows for inclusion of educational materials in informed consent documents by using an electronic consent form. This form is web-browser compatible and can be used on tablet

Figure 1: Screenshot of the iCONCUR tool, being developed by the University of California, San Diego

188   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

computers. This tool was built for a study at the Moores Cancer Center for patients who were consenting to donate biosamples and clinical data to the biorepository. In the tool, patients can view the regular content of the consent form as well as additional multimedia resources that were designed to offer more information. Besides, UCSD has developed a Permission Ontology that allows expressing subject’s consent permissions in a precise and uniform formalism that supports machine-based reasoning. [41] This Permission Ontology could be linked to the iCONS system to capture participant’s choices on sharing data and biosamples during the informed consent process. Furthermore, in order to be able to maximize researchers’ access to clinical data and biosamples obtained through informed consent and available at Moores Cancer Center Biorepository, UCSD has built a Resource Locator prototype.[42] The Resource Locator prototype accesses repositories of consent permissions expressed in the Permission Ontology and determine if researchers’ request for resources from the biorepository could be granted in compliance with subjects’ permissions. In addition, UCSD is currently developing an online tool called iCONCUR (informed CONsent for clinical data use for research) that uses taxonomies of choices to give patients control over how the information in their medical record is shared for research. Patients can decide what to share and with whom. For instance, a patient wants to share all the clinical information with non-profit researchers, but only diagnostic data with for-profit organizations. The tool will be used in a two-year clinical trial. During the duration of the study, the UCSD Clinical Data Warehouse will honor participants’ choices when sharing clinical data for research. Other academic medical centers have developed and piloted electronic consent systems.[43–45] At the Medical University of South Carolina (MUSC), a home-grown electronic system for capturing permissions related to research was piloted over a 12-month period. For example, patients could opt out of being contacted for research during the review of the electronic consent during patient registration on a tablet device. MUSC is exploring the use of a new version of this system for informed consent for participation in clinical trials. Commercial approaches have been developed too to make it safe for individuals to allow granular control over their confidential information, such as personal medical records, and to share that information over the internet. For instance, this is the case of Private Access Inc. Through their web-based application, “[…] patients are able to make their information available to trusted researchers or research groups, and to decide who should have the ability to search and view this information on the basis of anonymous or personally identifiable records. Once established, these private access rights open the ability for researchers to locate patients who match their particular criteria and to inform these patients of clinical trials that may be of interest.” A patient’s expressed permission is required to release highly granular data and personal contact information.

12 Electronic media for engaging patients in the consent process   

   189

12.3.3 Participant’s interest to know research results With internet access being widespread, patients have more opportunities than ever to share information about medical procedures, pharmaceutical drugs and research studies. Health-related forums give a chance to review doctors or medicines, find others suffering from the same disease and lend moral support. This patient-centered approach uses social media and personal communication to disseminate information, unlike more traditional organization-centered approaches of publishing data and presenting to peers at conferences.[21] One example of a patient-centered approach is 23andme.com. This company gives the opportunity for an evaluation of your DNA for $99. Results are presented in such way as to highlight ancestry, potential risks for diseases and physical traits such as hair color. Participants who elect to give DNA sample also consent to have their genetic information used for research studies. Comparisons with peers through the web portal 23andme.com allow for dissemination of results to others who have similar results or shared ancestry. Another example is PatientsLikeMe (http://www.patientslikeme.com/) which is an online patient network that allows subjects to connect with others with similar medical conditions (from prevalent conditions like diabetes, to rare conditions like ALS) to share experiences, learn, support others and contribute to research. As it is stated in their website: “The value of this open, community-driven approach to healthcare research was first demonstrated in 2011, when PatientsLikeMe revealed the results of a patient-initiated observational study refuting a 2008 publication that claimed lithium carbonate could slow the progression of ALS. The study, published in the scientific journal Nature Biotechnology, marked the first time a peer-to-peer network was used to evaluate a treatment in a patient population in real time. It was the first of a number of patient-reported outcome studies that have increased our understanding of diseases.” Companies like 23andMe, PatientsLikeMe and other research networks[21, 46] seem to be filling subjects’ need to be more involved and informed on research studies, while strengthening the trust of subjects on research. They perform research that could be relevant for the subject, publish and share the results with the scientific community and explain the research results to the members of the research network in an easy-to-understand way.

12.3.4 Online health information and its implications for patient autonomy The Web 2.0 has transformed the way people process and formulate health attitudes in a way that could help enhance informed consent. Web 2.0 refers to the newly evolved way to use the internet platform “where content and applications are not created and published by individuals, but instead are continuously modified by all

190   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

users in participatory and collaborative fashion”.(p. 61)[47] Consequently, according to Pew Internet’s 2010 Internet and American Life Project Survey, 80% of internet users look for health information online, making it the third most popular online activity.[48] Although most individuals continue to prefer receiving health information from their healthcare providers, a growing number of people are actively seeking health information online and other laypeople or online users.[49] Indeed, 60% of the online health information seekers discussed in the Pew report claimed that their discoveries had some influence on their health decision. With unrestricted numbers of people creating and sharing content online, the consequence of excess information appearing on the Web is inevitable. This deluge of information has led people to employ different strategies for filtering through meaningful information. One way for people to retrieve meaningful information is by relying on other users and consensus heuristics, such as the belief that “the truth lies in numbers”. In assessing how individuals search for and appraise health information online, Eysenbach and Kohler (2002) found that despite participant claims that they critically evaluate online health information, they instead bypassed these assessments and relied on search engine rankings to retrieve health information.[50] People’s preference for information recommended by search engine rankings over other expert sources has also been demonstrated in other studies.[51] Search engines are presumably credible because they are promoted by the collective “others” and their collective intelligence.[47, 51] Another decision tool that employs consensus heuristics is consumer ranking systems. According to Pew Internet’s 2009 report discussing the social characteristics of online health information: – 24% of e-patients have consulted rankings or reviews online for doctors or other providers; – 24% of e-patients have consulted rankings or reviews online of hospitals or other medical facilities.(p. 3)[49] Little is known about how individuals process rankings or reviews of medical information; however, evidence from consumer research provides valuable insights into how they influence decisions. For example, consumer review systems are becoming increasingly popular thanks to Web 2.0. Although there is no standard model, consumer rating systems generally take in user feedback to generate various symbols intended to communicate a specified item’s level of popularity.[52] Examples of consumer rating systems include Amazon.com’s customer reviews, which allow consumers to rate products and sellers, and Kindle’s Popular Phrases application, which underlines phrases that have been highlighted most frequently by other Kindle users, along with the number of people who highlighted each phrase (see Figure 2). According to current findings in consumer research, user ratings significantly influence behavioral intentions. This was best highlighted in an experimental study investigating the influence of online reviews on customer purchase intentions, Park,

12 Electronic media for engaging patients in the consent process   

   191

Figure 2: Consumer rating systems: (a) customer reviews from Amazon.com; (b) Kindle’s Popular Phrases application.

Lee and Han (2007), which demonstrated that the quality and number of reviews have a direct relationship to consumer purchase intentions.[53] Participants in the group with a low number of poor-quality reviews demonstrated the least intentions to purchase. In contrast, the group with a high number of high-quality reviews demonstrated the highest intentions to purchase. Most importantly, the number of reviews, regardless of their quality, significantly predicted purchase intentions and perceived trustworthiness of the reviews. Similar studies by both Chevalier and Mayzlin (2006) looking at consumer ratings and book sales, and Zhang and Dellarocas (2006) looking at the movie industry, have obtained similar results.[54, 55] This reliance upon customer reviews comes with an important caveat: there are reports of user reviews being commissioned by companies so some perceived “others” may be commercial sources, not similar individuals.[56, 57] The manipulative nature of these practices is underscored by the fact that current literature suggests that consumers prefer usergenerated reviews to seller reviews because they are less biased, more credible and are consumer-centered.[58, 59] In summary, Web 2.0 technologies and associated new ways of thought about information-sharing and meaning-making may suggest a solution to the issue of securing informed consent, especially to help alleviate or counteract the power imbalance between healthcare providers and patients. Specifically, participants could figuratively construct road signs on existing consent forms to filter through meaningful information while also preserving the breadth of information that is necessary to consent forms. Indeed, an experimental study by Schneider (2013) revealed that participants given an informed consent form that was highlighted by patients were significantly more likely to expend greater effort in processing the overall consent form than participants who were given an informed consent form that was highlighted by researchers.[60]

192   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

12.4 Discussion In today’s environment of an increasingly complex medical literature and rapidly advancing biotechnology and medical instrumentation, patients are faced with the daunting task of making health and research-related decisions based on very limited lay knowledge. Moreover, as mentioned above, there are several inherent problems in the current informed consent process including lengthy forms with overwhelming information, limited literacy, lack of adequate comprehension and social forces of influence. As a result, there are several associated ethical concerns about whether patients and research participants understand the nature of the research, the fact that it is voluntary and that there are other options. Complicating this issue are socioeconomic disparities and geographic limitations. On the other hand, participation in research is crucial for the continued advancement in medical knowledge, therapeutic options and ultimately improved health for the whole population. Recent advances in information technology and the advent of the era of big data adds complexity to traditional research and clinical trials. As a result, consents today are typically multifaceted and include several subcomponents that require individual decisions. For example, several research protocols not only require consent for participation in a clinical trial, but may also include options for participation in a genomic database, research registry, biospecimen repository, use of electronic medical record data for research or any combination of the above. In compliance with research participants’ wishes, these intricate decisions, hereby referred to as consent data, have to be diligently managed and tracked. It is also imperative that investigators use the latest version of an informed consent given that research protocols often undergo amendments to keep up with new research discoveries, changes in personnel or changes in workflow. To address these issues, an electronic system for presenting, tracking and managing consents has become an imperative for several large clinical research organizations and academic institutions. Collection, storage and reporting of informed consent documentation will be greatly facilitated by the use of electronic systems. In addition to facilitating the tracking of crucial consent data and versions of informed consents, an electronic system allows researchers to leverage rich media capabilities of modern electronic devices such as tablet computers. From a consumer perspective, electronic media empowers research participants to peruse the consent document content on their own, explore related explanatory content in the form of linked material, video, graphics and audio material, which results in improved comprehension and retention of the information. They can also look to other people and sources of information – outside the normal consent process – to help form decisions. Consequently, research participants are empowered to make better decisions. The use of electronic systems opens new avenues for remote consenting using eConsent with teleconferencing aids. Remote subject recruitment enables access to

12 Electronic media for engaging patients in the consent process   

   193

larger populations. Participants can also leverage modern media such as chat, text and e-mail to seek input from family, friends and peers. There are several challenges that have to be considered by research organizations when implementing eConsent systems. These challenges include: high initial expense for infrastructure and technology to manage online documents; establishing systems to validate electronic consent; process of verification of electronic consent; legitimacy of the electronic signatures; confidentiality and protection of personal health information (PHI); and compliance with applicable regulations (21 CFR Part 11). However, the benefits of such a system may outweigh these challenges.

12.5 Conclusion While traditional ethical and procedural issues involved in research consent remain valid and must not be ignored, eConsent and the utilization of electronic media to facilitate participant recruitment brings with it both new benefits and new challenges that must be carefully considered by investigators, regulatory authorities and potential research participants.

Acknowledgements Yukari Schneider would like to acknowledge Debbie Treise and H. Robert Kolb from the University of Florida for their guidance and wisdom Adela Grando would like to acknowledge Elizabeth Bell from University of California San Diego for initial chapter feedback and comments. Jihad Obeid’s work was supported by Health Sciences South Carolina and funded by Grant number RC2 LM010796 from the National Library of Medicine and the South Carolina Clinical & Translational Research (SCTR) Institute, with an academic home at the Medical University of South Carolina, National Institutes of Health Grant number UL1 TR000062. Wajeeh Bajwa would like to acknowledge assistance and help of H. Robert Kolb (University of Florida) for introducing authors to each other, and Jennifer Ann Lyon (University of Florida) for reviewing and providing comments on an early draft of the chapter. His work was supported in part by the NIH/NCATS Clinical and Translational Science Award to the University of Florida, UL1 TR000064.

194   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

References [1] [2] [3]

[4]

[5]

[6] [7] [8] [9] [10]

[11]

[12] [13] [14] [15] [16] [17]

[18]

[19]

Vollmann J, Winau R. Informed consent in human experimentation before the Nuremberg code. BMJ 1996;313(7070):1445. Nelson-Marten P, Rich BA. A historical perspective of informed consent in clinical practice and research. Seminars in oncology nursing 1999;15:81–88. National Institute of Health (NIH) [Internet]. Informed consent, Glossary of NIH Terms [cited November 2014]. Available from: http://www.nia.nih.gov/research/dgcg/clinical-researchstudy-investigators-toolbox/nia-glossary-clinical-research-terms N. C. for the P. of H. S. of Biomedical and B. Behavioral Research, The Belmont report: Ethical principles and guidelines for the protection of human subjects of research. ERIC Clearinghouse 1978. Gottesman MM, Sandler AL. Guidelines for the conduct of research involving human subjects at the National Institutes of Health. Washington, DC: US Department of Health, Public Health Services, National Institutes of Health. 2004: 00–4783. Beskow LM, Friedman JY, Hardy NC, Lin L, Weinfurt KP. Simplifying informed consent for biorepositories: stakeholder perspectives. Genet. Med. 2010;12(9):567–572. Paasche-Orlow MK, Taylor HA, Brancati FL. Readability standards for informed-consent forms as compared with actual readability N. Engl. J. Med. 2003;348(8):721–726. Flory J, Emanuel E. Interventions to improve research participants’ understanding in informed consent for research: a systematic review. J. Am. Med. Assoc. 2004;292(13):1593–1601. Sugarman J, McCrory DC, Hubal RC. Getting meaningful informed consent from older adults: a structured literature review of empirical research. J. Am. Geriatr. Soc. 1998;46(4):517–524. Ridpath JR, Greene SM, Wiese CJ, PRISM readability toolkit. Seattle: Group Health Center for Health Studies, 2007. Available from: http://www.nhlbi.nih.gov/files/docs/ghchs_readability_ toolkit.pdf Kirsch IS, Jungeblut A, Jenkins L, Kolstad A. Adult Literacy in America: A first look at the findings of the National Adult Literacy Survey. Washington DC: U.S. Department of Education, Office of Education Research and Improvement; 2002. April, Report No.: NCES 1993–275. Joffe S, Cook EF, Cleary PD, Clark JW, Weeks JC. Quality of informed consent: a new measure of understanding among research subjects. J. Natl. Cancer Inst. 2001;93(2):139–147. Fortun P, West J, Chalkley L, Shonde A, Hawkey C. Recall of informed consent information by healthy volunteers in clinical trials. QJM 2008;101(8):625–629. Kass NE, Chaisson L, Taylor HA, Lohse J. Length and complexity of US and international HIV consent forms from federal HIV network trials. J. Gen. Intern. Med. 2011;26(11):1324–1328. Felt U, Bister MD, Strassnig M, Wagner U. Refusing the information paradigm: informed consent, medical research, and patient participation. Health (N.Y.) 2009; 13(1):87–106. Whiddett R, Hunter I, Engelbrecht J, Handy J. Patients’ attitudes towards sharing their health information. Int. J. Med. Inf. 2006;75(7):530–541. Willison DJ, Steeves V, Charles C, Schwartz L, Ranford J, Agarwal G, Cheng J, Thabane L. Consent for use of personal information for health research: Do people with potentially stigmatizing health conditions and the general public differ in their opinions? BMC Med. Ethics 2009;10(1):10. Willison DJ, Swinton M, Schwartz L, Abelson J, Charles C, Northrup D, Cheng J, Thabane L. Alternatives to project-specific consent for access to personal information for health research: Insights from a public dialogue. BMC Med. Ethics 2008;9(1):18. Weitzman ER, Kaci L, Mandl KD. Sharing Medical Data for Health Research: The Early Personal Health Record Experience. J. Med. Internet Res. 2010;12(2):e14.

12 Electronic media for engaging patients in the consent process   

   195

[20] Barrett G, Cassell JA, Peacock JL, Coleman MP, National survey of British public’s views on use of identifiable medical data by the National Cancer Registry. BMJ 2006;332(7549):1068–1072. [21] Kaye J, Curren L, Anderson N, Edwards K, Fullerton SM, Kanellopoulou N, Lund D, MacArthur DG, Mascalzoni D, Shepherd J, Taylor PL, Terry SF, Winter SF. From patients to partners: participantcentric initiatives in biomedical research. Nat. Rev. Genet. 2012;13(5):371–376. [22] Health IT Policy Committee, Privacy and Security Tiger Team, Letter to David Blumenthal, Chairman of the Office of the National Coordinator for Health IT [Internet] 2014 [cited November 2014]. Available from: http://www.healthit.gov/sites/default/files/hitpc_transmittal_p_s_ tt_9_1_10_0.pdf [23] Fernandez CV, Kodish E, Weijer C. Informing study participants of research results: an ethical imperative. IRB 2003;25(3):12–19. [24] Shalowitz DI, Miller FG. Disclosing individual results of clinical research: implications of respect for participants. JAMA J. Am. Med. Assoc. 2005; 294(6): 737–740. [25] Knoppers BM, Joly Y, Simard J, Durocher F. The emergence of an ethical duty to disclose genetic research results: international perspectives. Eur. J. Hum. Genet. 2006;14(12):1322–1322. [26] Shalowitz DI, Miller FG. Communicating the Results of Clinical Research to Participants: Attitudes, Practices, and Future Directions. PLoS Med 2008;5(5):e91. [27] Cox K, Moghaddam N, Bird L, Elkan R. Feedback of trial results to participants: A survey of clinicians’ and patients’ attitudes and experiences. Eur. J. Oncol. Nurs. 2011;15(2–4):124–129. [28] Getz K, Hallinan Z, Simmons D, Brickman MJ, Jumadilova Z, Pauer L, Wilenzick M, Morrison B. Meeting the obligation to communicate clinical trial results to study volunteers. Expert Rev. Clin. Pharmacol. 2012;5(2):149–156. [29] Fernandez CV, Kodish E, Shurin S, Weijer C, Children’s Oncology Group. Offering to return results to research participants: attitudes and needs of principal investigators in the Children’s Oncology Group. J. Pediatr. Hematol. Oncol. 2003;25(9):704–708. [30] Zarin DA, Tse T, Williams RJ, Califf RM, Ide NC. The ClinicalTrials.gov Results Database – Update and Key Issues. N. Engl. J. Med. 2011;364(9):852–860. [31] Ende J, Kazis L, Ash A, Moskowitz MA, Measuring patients’ desire for autonomy. J. Gen. Intern. Med. 1989; 4(1):23–30. [32] Faden RR, Beauchamp TL, King NM. A history and theory of informed consent. NY: Oxford University Press; 1986. [33] Brock TC, Green MC. Persuasion: Psychological Insights and Perspectives. SAGE Publications; 2005. [34] Perloff RM. The Dynamics of Persuasion: Communication and Attitudes in the Twenty-First Century. Routledge; 2010. [35] Cialdini RB. Influence: The Psychology of Persuasion. William Morrow, New York, 1984. [36] Paris A, Nogueira da Gama Chaves D, Cornu C, Maison P, Salvat-Mélis M, Ribuot C, Brandt C, Bosson JL, Hommel M, Cracowski JL. Improvement of the comprehension of written information given to healthy volunteers in biomedical research: a single-blind randomized controlled study. Fundam. Clin. Pharmacol. 2007;21(2):207–214. [37] Houts PS, Witmer JT, Egeth HE, Loscalzo MJ, Zabora JR. Using pictographs to enhance recall of spoken medical instructions II. Patient Educ. Couns. 2001;43(3):231–242. [38] Jackson CL, Bolen S, Brancati FL, Batts-Turner ML, Gary TL. A Systematic Review of Interactive Computer-assisted Technology in Diabetes Care. J. Gen. Intern. Med. 2006;21(2):105–110. [39] Miller RH, Sim I. Physicians’ use of electronic medical records: barriers and solutions. Health Aff. (Millwood) 2004;23(2):116–126. [40] Lieberman MI, Embi P, Ricciardi TN, Tabb K. Accelerating Biopharmaceutical Development in the Decade of Health Information Technology: Applications of EHRs for outcomes research and clinical trials recruitment. Biotechnol. Healtch. 2005;2(4): 52–57.

196   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[41] Grando MA, Boxwala A, Schwab R, Alipanah N. Ontological Approach for the Management of Informed Consent Permissions. 2012 IEEE Second International Conference on Healthcare Informatics, Imaging and Systems Biology (HISB) 2012:51–60. [42] Grando MA, Schwab R. Building and Evaluating an Ontology-based Tool for Reasoning about Consent Permission. American Medical Informatics Association Annu. Symp. Proc., 2013: 514–523. [43] Chalil Madathil K, Koikkara R, Obeid J, Greenstein JS, Sanderson IC, Fryar K, Moskowitz J, Gramopadhye AK. An investigation of the efficacy of electronic consenting interfaces of research permissions management system in a hospital setting. Int. J. Med. Inf. 2013;82(9):854–863. [44] Obeid JS, Gerken K, Madathil KC, Rugg D, Alstad CE, Fryar K, Alexander R, Gramopadhye AK, Moskowitz J, Sanderson IC. Development of an Electronic Research Permissions Management System to Enhance Informed Consents and Capture Research Authorizations Data. AMIA Joint Summits Transl. Sci. Proc. 2013:189–193. [45] Sanderson IC, Obeid JS, Madathil KC, Gerken K, Fryar K, Rugg D, Alstad CE, Alexander R, Brady KT, Gramopadhye AK, Moskowitz J. Managing clinical research permissions electronically: A novel approach to enhancing recruitment and managing consents. Clin. Trials Lond. Engl. 2013;10(4):604–611. [46] Ohno-Machado L, Alipanah N, Day ME, El-Kareh R, Farzaneh S, Freeland P, Grando A, Kim H, Meeker D, Kim K. Comprehensive Inventory of Research Networks. Clinical Data Research Networks, Patient-Powered Research Networks, and Patient Registries. Patient-Centered Outcomes Research Institute (PCORI) Report, 2013. [47] Kaplan AM, Haenlein M. Users of the world, unite! The challenges and opportunities of Social Media. Bus. Horiz. 2010; 53(1): 59–68. [48] Fox S, Health Topics: 80% of internet users loo for health information online [Internet]. DC Washington: Pew Internet and American Life Project; 2011 [cited November 2014]. Available from: http://www.pewinternet.org/files/old-media/Files/Reports/2011/PIP_Health_Topics.pdf [49] Fox S, Jones S. The social life of health information [Internet]. DC, Washington: Pew Internet and American Life Project; 2009 [cited November 2014]. Available from: http://www.pewinternet. org/2009/06/11/the-social-life-of-health-information [50] Eysenbach G, Köhler C. How do consumers search for and appraise health information on the world wide web? Qualitative study using focus groups, usability tests, and in-depth interviews. BMJ. 2002; 324(7337): 573–577. [51] Sundar SS, Nass C. Conceptualizing sources in online news. J. Commun. 2001; 51(1): 52–72. [52] Lohmann S, Ziegler J, Tetzlaff L. Comparison of tag cloud layouts: Task-related performance and visual exploration. Human-Computer-INTERACT 2009, Springer 2009: 392–404. [53] Park D. H, Lee J, Han I. The effect of on-line consumer reviews on consumer purchasing intention: The moderating role of involvement. Int. J. Electron. Commer. 2007; 11(4): 125–148. [54] Chevalier JA, Mayzlin D. The effect of word of mouth on sales: Online book reviews. J. Mark. Res. 2006;43(3):345–354. [55] Zhang X, Dellarocas C. The Lord Of The Ratings: Is A Movie’s Fate is Influenced by Reviews? ICIS 2006 Proceedings. Paper 117; 2006. [56] Parsa A. Exclusive: Belkin’s development rep is hiring people to write fake positive amazon reviews. The Daily Background Weblog [Accessed Apr 2013]. [57] Jøsang A, Ismail R, Boyd C. A survey of trust and reputation systems for online service provision, Decis. Support Syst. 2007;43(2):618–644. [58] Dellarocas C. The digitization of word of mouth: Promise and challenges of online feedback mechanisms. Manag. Sci. 2003;49(10):1407–1424.

12 Electronic media for engaging patients in the consent process   

   197

[59] Bickart B, Schindler RM. Internet forums as influential sources of consumer information, J. Interact. Mark. 2001;15(3):31–40. [60] Schneider Y. Studying the effects of need for cognition and elaboration appeals on participant evaluation of informed consent information. Dissertation. Gainesville: University of Florida, 2013.

Catalina Danis, Martha Jean (Marty) Minniti, Sasha Ballen, Marion Ball, Scott Cashon, Margaret Piscitelli, Marjorie Miller and Robert Farrell

13 Patient engagement at the point of care: technology as an enabler Abstract: There is a long-standing interest among clinicians and researchers in the healthcare community in the role that patients’ involvement in their healthcare management might play in improving health outcomes and healthcare quality, as well as in controlling the costs of healthcare provision. Recent advances in mobile computing technology make it feasible to scale successful patient engagement programs first delivered in limited face-to-face trials to larger patient populations. However, comparatively little is known about how technology-enabled patient engagement systems might fare in deployments in clinical contexts involved in the treatment of patients with chronic diseases. We initiated a six-month trial with 25  patients to explore patient and provider interactions with one commercially available patient engagement system. We deployed the system comprised of a kiosk, mobile phone and web user interfaces to patients with a primary diagnosis of Diabetes Mellitus, type  2 or Hypertension who were receiving care at a large urban medical practice that emphasizes patient engagement. We used a mixed-methods methodology to collect qualitative and quantitative feedback on the use of the technology. We found a large range among patients in their ability to engage through the technology. Physicians were generally interested and positive about the use of the technology. We are currently exploring ways to help both stakeholders improve in incorporating the newly available data into their work practices.

13.1 Introduction Healthcare systems are stretched beyond capacity while the incidence of chronic disease continues to grow;[1] how can healthcare systems afford to offer assistance to the rapidly growing population of individuals who need it most? Against this backdrop, patient engagement has engendered a great deal of enthusiasm over the past two decades for its professed ability to improve health outcomes and control healthcare costs[2–5] through enabling the patient to play a significant role in her health care. With roots in the Chronic Care Model,[6–8] the concept of patient engagement (sometimes discussed as patient empowerment¹) has two fundamental tenets. The 1 While some authors may distinguish between these two terms, we treat them inter-changeably in this chapter.

200   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

first, called “collaborative care”, posits that health care for an individual should be determined collaboratively through a partnership between the patient and her healthcare provider. The second, “self-management education”, asserts the importance of teaching patients health-related problem-solving skills, rather than simply educating them about their disease states, to enable them to make good decisions regarding health-relevant behaviors in their daily lives.[8] We agree with others who have concluded that engagement cannot be scaled to the required levels without leveraging information technology to reach the individual and enable their greater role.[9, 10] We acknowledge at the outset that the cost-saving and promised effectiveness of health information technology is itself a complicated topic and many system roll-outs have failed to meet deployment goals.[11, 12] Nevertheless, the affordances[13] of mobile phone technology, including anytime/ anywhere access, personal information storage and ready-at-hand communication, raise the possibility that IT-based enablement of patient engagement could be quite effective. The high rate of penetration of mobile phones and tablets today has created the opportunity to develop a personalized patient engagement platform with greater reach than ever before. In combination with back-end systems that integrate with clinical systems and manage the additional communication load with already overburdened providers, the ideal platform will adapt the critical (and high-cost) components of successful programs and introduce them to a much broader audience. However, relatively little is known about the use of technology-enabled patient engagement systems in chronic care management. Published studies typically involve the use of prototypes rather than actual products, and therefore are not typically deployed in actual clinical situations. Other sources of information are reports from clinical trials of products that typically report results very narrowly focused on efficacy, for example impact on health indicators such as H1AC levels in Diabetes Mellitus type 2 (DM type 2). As such, we know little about the issues raised by integrating as disruptive a technology as patient engagement support into a real-world practice. It will require many studies to articulate the potential and the problems with an electronic engagement platform. It is in this spirit that we present our contribution in which we describe our experiences with deploying one particular mobile patient engagement system to patients for managing their chronic diseases over several months.

13.2 The current situation: needs, gaps and challenges An important step in the study of patient engagement is the clinical randomized trial (RCT) methodology as applied to comparing the efficacy of treatment options (e.g. pharmacologic to non-pharmacologic interventions) in various chronic disease conditions. A number of important studies utilizing the RCT methodology – the “gold standard”

13 Patient engagement at the point of care: technology as an enabler   

   201

of clinical research because it is based on random assignment of study patients to the treatments under investigation – demonstrated better outcomes for patients treated with behavioral interventions relative to pharmacologic agents.[14, 15] These studies strived for successful interventions at costs that were competitive with usual care. Nonetheless, the studies were human resource intensive in many ways that limited their patient reach. For example, they utilized lifestyle coaches, dieticians or monitored exercise sessions that were typically offered to a small number of patients. One such RCT study focused on the design and evaluation of a self-management education intervention – the second of the two tenets of patient engagement – that included typical care topics such as medication use and diet, but also included engagementtype topics like communication with health professionals, problem-solving related to implementing lifestyle changes, decision-making and self-management techniques for disease symptoms.[e.g. 4] In spite of the costs of training educators and delivering face-to-face interventions, the study was able to show a long-term positive return on investment due to avoidance of hospitalization for patients with heart and lung disease co-morbidities. However, scaling such interventions under current real world cost constraints with shorter payback time horizons requires the use of mobile and/ or computer technology to send input to and collect output from patients and automated decision support tools to mitigate the additional workload on health professionals. Defining the capabilities that an IT-enabled patient engagement system might include is aided by an understanding of the research literature on two questions. First, what are the critical patient skills and actions that should be supported through technology? In part, it is a matter of “translating” well-understood patient and provider support onto a new technology base. Some of the skills that have been included under the umbrella of partnering between providers and patients and enabling patients to problem solve with respect to their health have been shown to be supportable through technology. However, it would be a missed opportunity to simply “clone”[16] existing human-based processes and embody them as IT mediated ones. Thus, the second area where published literature is helpful concerns understanding the capabilities of the new technology and how these modify and expand what may and should be supported through IT to enable patient engagement. Whatever the exact definition of patient engagement that one adopts, there is general agreement that broad-based and more frequent involvement of patients in managing their health through technology represents a culture change with respect to the roles, responsibilities and expectations of the primary stakeholders in the healthcare system. It also requires a re-alignment of financial costs and benefits such that new reimbursement rules rewarding improved health outcomes through bonuses and shared savings while penalizing failures to meet quality criteria create an environment in which data from the engaged patient becomes critical to and consumable by relevant stakeholders. Field trials, such as the one we report on in this chapter, enable the observation of both the technical and the organizational aspects of deployment of patient enablement technology.

202   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

13.2.1 Fundamental needs in patient engagement Faced with hundreds of studies carried out under highly varied conditions, the teams led by DiMatteo[17, 18] and Michie[19, 20] adopted meta-analytic techniques to distill the repeatable findings and patterns that facilitate behavioral change and adherence which are critical components of patient engagement.[21] DiMatteo and colleagues produced a model they call the Information-Motivation-Strategy Model[18] named after the three factors that emerged from their meta-analysis. The first factor – information – which they couch in the context of the provider relationship, maps to aspects included in both the collaborative and education tenets in the patient engagement literature.[22, 23] They articulate the need to provide the patient with information concerning their disease, treatment options and required behaviors and place it in the context of a patient-provider relationship built on empathy and trust that enables open communication with two-way feedback. The second factor – motivation – refers to the underlying importance of patients’ belief in the efficacy of their treatment and consequent commitment to participating in it.[24] Their final factor – strategy – refers to factors that might prevent the patient from following through on treatment recommendations, ranging from psychodynamic and cognitive factors that might lead to forgetting appointments to social and pragmatic factors that might produce an unsupportive home environment. A large research and design literature exists that has translated the above constructs into techniques and embodied them in various software artifacts. Michie et al’s meta-analysis of 122 studies identified “self-monitoring” techniques as the single most effective type of technique for bringing about behavior change. In the reviewed studies people commit to tracking a behavior identified as instrumental for their goals (e.g. food intake or exercise to support weight loss) and create records that makes possible an evaluation of their performance possible. While the effect size for this technique is relatively small over all studies that examined behavior change in the context of healthy eating and physical activity (i.e. accounting for only 13% of the variance in the analysis), the impact is bigger when combined with techniques from a set that includes prompting individuals for intention formation, prompting them for specific goal-setting, providing them feedback on their performance, prompting the self-monitoring of behavior and prompting the review of behavioral goals.[19] For example, two studies embedded mobile record keeping in the context of goal setting for bringing about behavior change.[25, 26] In these studies, individuals were asked to specify a goal, such as exercising for a certain amount of time or limiting calorie intake, and track the relevant behavior on a mobile device. In other studies, individuals also share their data with their care team. In combination with these other behaviors, behavior tracking may promote reflection on the part of the individual concerning mechanisms leading to goal achievement in addition to accountability. A number of threads of research have focused on using technology to support aspects of “self-management education”, the second tenet of patient engagement,

13 Patient engagement at the point of care: technology as an enabler   

   203

which goes beyond mere information provision to supporting sense-making activities. For example, work investigating the role of online communities for supporting information needs of patients is particularly important as resource limitations restrict the time available during patient/ provider appointments. This stream of work is longstanding [e.g., 27], with solid evidence that patients with chronic diseases gain valuable insights for disease management strategies as well as emotional support.[27, 28] More recent work that has examined the bridging work of moderators as checks on the validity of information that patients glean from online sources[29] is especially promising because it addresses health professionals’ information quality concerns. Research has also begun to focus on providing a software-based “scaffolding” to promote the development of insights about one’s behavior with respect to chronic disease.[30] In one study, two individuals with chronic DM type 2 used a smart phone application to track “breakdowns” in their experiences with caring for their disease (e.g. questions about food intake) and to discuss them with a diabetic educator. Mobile technology is also widely used to provide “just-in-time” support for individuals endeavoring to make a behavioral change, for example individuals attempting to become tobacco-free[31] or to recover from alcohol abuse.[32] In summary, the meta-analyses of Michie et al.[19] and DiMatteo et al.[17] highlight some of the skills that comprise the partnership and problem-solving goals of patient engagement programs. The literature we have sampled demonstrates that capabilities of newer technologies, both web-based but especially mobile phone based, can be integrated into the individual’s everyday routine and thus extend the reach of previous face-to-face methods.

13.2.2 Gaps & challenges The literature we have sampled above is sourced from a variety of fields. The foundational conceptual work comes from clinical investigations of the care of patients with chronic diseases done by primary care physicians. Subsequent work on interventions is frequently based on point programs – those directed at specific goals like weight loss or smoking cessation. Researchers in the fields of Human Factors and Human Computer Interaction have contributed many of the novel application design efforts based on early, typically brief, usage of system prototypes. While all of the contributing fields have advanced the communal understanding of the role of engagement in health management and the use of technology to promote it, a significant gap exists in our understanding of the use of systems for extended periods of time in active health delivery settings. Part of that gap includes the assessment of attitudes about patient engagement held by the patients themselves but also by the care providers. The former is often discussed in the context of studies of barriers that patients encounter in managing their medical conditions,[33] or the importance of patient readiness[34, 35] in disease management and health outcomes.

204   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

The relationship of the patient and physician, or more broadly, the care team, has long been recognized for its critical role in improving outcomes in chronic care. The attitude of providers is critical because it defines the reception that the patient’s efforts at engagement will receive. This impacts patient satisfaction and fuels continued engagement,[22] but also has important pragmatic implications for the integration of the results of patient’s engagement efforts into the healthcare system, particularly into the Medical Practice. The successful integration of the data and insights produced through the patient’s involvement is what will determine whether the promise of patient engagement for taming chronic disease, introducing accountability in health care and reigning-in costs will be realized on a large scale. It is with this goal in mind that our team undertook an investigation into the effectiveness of patient engagement technology for enhancing base levels of engagement in patients with chronic diseases. We wanted to see how the technology would complement the engagement that already existed within the primary care medical practice (physicians and other caregivers, the organization and individual patient differences). Finally, we were interested in understanding what would be required to integrate the technology into the practice of medicine, from the point of view of the care providers and from that of the patients.

13.3 Proposed solutions The status of IT-enabled patient engagement in real-world practice in the United States must be viewed in the context of the broader push towards IT enablement of healthcare delivery driven by various healthcare reform initiatives championed by government, public interest groups and industry. One important initiative that has had implications for greater involvement of the patient in her health care is the mandate under the HITECH Act of 2009[36] for the adoption of electronic medical record (EMR) systems at the point of care by the end of 2014. Follow-on initiatives that directly impact the patient include the Meaningful Use initiative[37] from the Office of the National Coordinator (ONC) and the Centers for Medicare and Medicaid Services (CMS). Stage 2 of the MU initiative, which became effective in 2013, includes guidelines for providing patients access to their Personal Medical Health records (derived from the provider’s EMR) and mandates a 10% utilization rate of the online resources. These are typically made available through web portal interfaces and frequently include some process support, such as requesting refills, scheduling appointments and communicating with the provider, in addition to giving patients access to their historical healthcare records. Success of engaging patients through MU is currently widely debated.[38] Some large hospital systems have developed their own versions of online portals that try to involve the patient in preparing for their appointments and assessing their knowledge and capability for being involved in their health care.

13 Patient engagement at the point of care: technology as an enabler   

   205

[39] These top-down driven mandates make functionality available online and largely depend on the patient to initiate interaction. In contrast, many hospital systems and vendors are beginning to develop targeted, time and condition-focused interventions under the umbrella of patient engagement that are presented in the context of structured programs. One such study utilized a multi-media interactive, web-based program from Emmi Solutions[40] for preparing first-time colonoscopy patients for the procedure. An evaluation conducted at a research hospital found that relative to paper-based resources, the web program resulted in lowered anxiety scores and sedation medication requirements, and shortened procedure durations for first-time colonoscopy patients.[41] With respect to a mobile platform, EMR vendors and others are creating extensions of their systems to support anytime/ anywhere access, though many of these are initially aimed at the physician or care provider rather than the patient.[42] Many vendors are creating disease-specific systems that provide tracking and support functions for the patient with a chronic disease (frequently for controlling DM type 2) and an associated mobile or web-based application for providers. While these are widely advertised and sold (e.g. Tactio for DM type 2 management[43], DiabetesManager from At&T[44]), we are not aware of research studies that have followed patients except in narrowly defined clinical trials that report efficacy in impacting health indicators such as HA1C (e.g. [44]). Our field trial came about as the result of informal discussion about the applicability of technology to support patient engagement among a subset of the authors of this publication who are employees of IBM (CD, MB, RF) and of CarePartners Plus (MM). We complemented the synergy offered by a robust patient engagement platform available from CPP and the research capabilities offered by the Research Division of IBM with a field site provided by a private internal medicine practice and research assistants from the Department of Nursing Informatics at the Jefferson School of Nursing in Philadelphia, PA. The team began collecting data for a planned six-month study in December of 2013. Our study, which is on-going, is an all-volunteer effort; this is unusual in the area of clinical research where the typical effort involves funding for the study site, the research staff required and often the study participants as well. One major limitation that resulted from the research team member in-kind contributions model is that we were not an integrated part of the clinical team, but operated as a parallel research effort.

13.3.1 Method Our study site, hereafter called the Medical Practice, is a thirty thousand annual visit practice in a large urban center in the United States. The Medical Practice was an early adopter of the Patient-Centered Medical Home (PCMH) model, achieving level 3 rec-

206   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

ognition from NCQA.[45] PCMH stresses care coordination and communication and embraces the role of patients engaged in their own health management. The WellbyTM patient engagement platform[46] from CarePartners Plus[47] provides multiple user access points, including a point-of-care (PoC) kiosk, patient mobile support, in-patient tablet (not used in this study) and web portal access for supporting patients in their ambulatory context.² The PoC kiosk assesses gaps in care immediately following a visit before a patient leaves the provider office and allows the patient to document shared goals agreed to during the visit. The mobile experience can be customized to the specific requirements of particular diseases (e.g. incorporating a symptom checklist for Congestive Heart Failure or one for Diabetes Mellitus type 2). The patient portal, available over the web, provides full patient health record (PHR) functions, general support for health-related education and support for management of chronic conditions through goal setting, tracking and coordinating patient support from family and friends. The platform also supplies providers with extensive report capabilities and rule-based integration with a practice’s standards of care. Prior to patient recruitment, a Wellby point-of-care kiosk was installed near the checkout desks in the Medical Practice. We complemented the data collected through the Wellby platform with 30-minute patient interviews addressing barriers to adherence.³ The results of these interviews are used to characterize the patient’s level of engagement in their care at the start of the study. We also distributed a two-page questionnaire addressing physician attitudes about care to all of the healthcare providers at the Medical Practice. Both are provided in the Appendix. In order to participate in the study, a patient had to: have a relevant medical diagnosis (DM type 2 or Hypertension); express interest in participating; meet the technology requirements of having a personal smart phone and web access from home; and own appropriate screening devices (glucomoter or blood pressure cuff). When a patient met the criteria, they were enrolled in the study and the RA would guide the patient through the consent process and install the mobile application on the patient’s smart phone. The RA would also educate the patient about their account name and password for the study website. The RA was to perform as many of these steps as possible in person with the patient, however allowances were made to accommodate phone contact with patients should they need additional assistance. Once the patient was enrolled and had granted consent, the patient received an e-mail with a link to the study website, where he/ she would complete the first of several AboutMe assessments. After the patient completed the final assessment, a member of the study team would perform the “barrier” interview with the patient. Patient anonymity was maintained during the interview. Finally, the patient status 2 Though not utilized in this study, the platform also provides in-patient engagement support. 3 All active study participants also completed six brief patient characteristics assessments (AboutMe) that we do not discuss in this report.

13 Patient engagement at the point of care: technology as an enabler   

   207

would change from “onboarding” to “active” and the patient would begin to receive daily messages on their smart phone. Patient progress through the onboarding stage was tracked by a team member who made calls or sent e-mails as necessary to move the patient along the process.

13.3.2 Recruitment, onboarding and interview results The main data we present is from the patient stream – the results from recruiting patients into the study and bringing them to “active” status in the study, interview results with a sample of “active” study patients concerning the obstacles they reported to adhering to their physician’s directions and experiences of “active” patients with the engagement support technology. We complement the main focus on patients with a brief characterization of physician attitudes regarding patient engagement that serves to describe the context for the patient experiences in the study.

13.3.2.1 Patient recruitment & onboarding process The study participant pool was comprised of 1200 patients who had one of the originally targeted three diagnoses (diabetes, pre-diabetes and hypertension) and were scheduled to have an in-office appointment from December 1–February 28. Soon after beginning participant recruitment, we found that patients who had been identified as being pre-diabetic did not have glucometers and would not be able to report daily blood glucose levels as required by the study protocol; we thus eliminated them from further consideration. After screening by the RAs, 87  patients met the study criteria and agreed to participate in the study. We added 11 patients who had scheduled appointments after the original recruitment letters had been sent out, for a total of 98  participants. These patients signed informed consent forms in the presence of one of the RAs and thus became enrolled in the study. The step-by-step results of the recruitment process are summarized in Table 1. The onboarding process was begun after enrollment was completed. It was designed to be a two-week process (as described in a patient handout) during which time patients would complete the six-part AboutMe assessment on a secured website associated with the study and would install the smartphone application needed to enable daily mobile check-ins. Table 2 shows the onboarding status of the 98 enrolled patients. There are 59 enrolled patients who have HTN as their primary diagnosis and 39 who have DM type II; this reflects the population at the Medical Practice, where there are nearly twice as many patients with HTN as with DM type II. So far only about a quarter of the enrolled patients in each diagnostic category have achieved the “active” status. Another quarter in each diagnostic category is still in the onboarding process. The onboarding process has taken on average over twice as long as we had expected, averaging 42 days for patients who have achieved the active status. A research team

208   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Table 1: Step-by-step results of recruitment process: starting from mailing to 1200 patients and resulting in 87 enrolled patients to which 11 were later added. Step-by-step results of recruitment process Sent letters

Approached by RAs

Eligible diagnosis

Lacked technology

Not interested

Enrolled

Added

1200

412

407

235

85

87

11

member called patients when their progress deviated from the expected timeline. The most frequently reason given for slower than anticipated responses was forgetting the credentials enrolled participants needed to log on to the secure website to complete the AboutMe onboarding surveys. A further approximately 50% of enrolled patients have been dropped from the study, most commonly because they failed to complete the onboarding process in spite of multiple calls from the study staff. In many cases the staff resolved the problem presented by the enrolled participant (most frequently, lost credentials) but the patient nevertheless failed to complete the onboarding tasks. We report data below data that the active participants are generating daily through responses to the mobile check-in questions as well as providing qualitative feedback through e-mails and phone calls to the research team. Table 2: Patient study status in the two diagnostic conditions expressed as number of participants and percent in each status category. Study Status Total enrolled

Active

Onbaording

Withdrawn/Dropped

Hypertension

59 (100%)

15 (25.4%)

15 (25.4%)

29 (49.2%)

Diabetes Mellitus, type II

39 (100%)

10 (25.6%)

9 (23.1%)

20 (51.3%)

13.3.2.2 Physician interviews: their views on patient engagement The physician’s (or more broadly, the primary caregiver’s) point of view is critical to any investigation of patient engagement. In the context of current primary care practice, the patient-physician relationship is a key determinant of patient engagement through its effect on patient satisfaction and subsequently on the patient’s involvement in his/her healthcare management.[20, 22] And, in the envisioned future scenario of widespread patient-reported data as input to treatment, the physician is a primary consumer of the data produced by the engaged patient. Below we briefly summarize the physicians’ point of view about patient engagement that we ascertained through a 15-item survey we developed (see Appendix) to address the physi-

13 Patient engagement at the point of care: technology as an enabler   

   209

cians’ role as care providers and their attitudes about patient engagement. One of our team members distributed the hard-copy questionnaire at the weekly staff meeting near the start of the study; nine of 11 of the physicians completed the survey. The physicians identified patient engagement along with three other factors as most important in contributing to overall positive health outcomes. Using a scale of one to five, from least to most important contributors, the physicians identified “following medical orders”, “scheduling and attending recommended medical visits”, “nutrition” and “patient engagement” (mean scores of 4.7, 4.6, 4.4 and 4.4, respectively) as being between “extremely” and “very” important to positive health outcomes. Identified as being somewhat less important were the factors of “receiving regular feedback (from providers)” and having a “supportive home/ family situation”, with mean ratings of 4.1 and 3.9, respectively. Judged to be least important for their contributions to overall patient health were “educational handouts” with a mean of 2.8 and “web-based resources” with a mean score of 2.7. With respect to the information state of their patients, the physicians judged that approximately 2 out of 3 understand the ramifications of their conditions. This is germane because they take their patient’s status into account when formulating treatment, with 7 of 9  respondents saying they adjust their treatment or treatment plan to accommodate the level of engagement evidenced by the patient. Examples of activities they initiate in response to patients who they suspect of low engagement include education (“… understand glycemic index…”, “discuss complications of noncompliance…”) and various ways of engaging the patient by requiring reporting of a variety of home-based measurements such as blood pressure readings and weight measurements. While patient engagement was judged to be as important to patient health as more “traditional” factors related to medical directives, the physicians showed less agreement on what factors constitute engagement. The most highly rated behavior, with a mean rating of only 3.7 on the same five point scale (a rating between “very important” and “important”) is “communicating social and family problems”. This factor is frequently identified for its potential to interfere with an individual’s health management in general, and with adherence to physician directives in particular. [25] The two factors that can be facilitated by the patient engagement tools used in our field trial, “communicating information between visits” and “keeping records of testing as directed”, were rated even lower with a mean level of 3.4 and 3.3, respectively, placing them closer to the “important” category, which is the neutral point on the five-point scale. To summarize what we learned about physician attitudes about patient engagement, our results indicate that they believe it to be important but they do not evidence a strong appreciation of the types of patient behaviors that can be facilitated through mobile technology. Thus they appear to value medical directives more strongly than access to additional data from in between medical appointments. This is not surpris-

210   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

ing since they are familiar with the former, but lack experience with the types of data that new engagement technology will be able to deliver in the future.

13.3.2.3 Patient interview results: evidence of engagement at study baseline We report on the results from “barriers” interviews (see Appendix) done with the first 13 patients to reach “active” study status, including eight with a primary diagnosis of hypertension and five with a primary diagnosis of DM type 2. The focus of the interview was on the experiences of patients in trying to follow through on physician directives. Patient-reported medication and lifestyle physician directives were treated as “opportunities for adherence”, with patients’ behavior coded as demonstrating complete, partial or no adherence. In addition, free-form comments were collected to assess achievement of the goals articulated in the two component skills of patient engagement, namely partnering with the healthcare provider and evidence of selfmanagement skills. In the aggregate, the interview participants demonstrated the well-documented adherence pattern whereby adherence to medication is higher than adherence to lifestyle directives.[23] The results, in Table 3, show that patients reported being completely adherent to medication in over 80% of the cases and adherent to lifestyle directives in 65% of the opportunities. The reported absolute rates are significantly higher than the average rates reported in the literature (50% and 35% respectively for medication and lifestyle directives); however, average population rates are aggregated over many factors that significantly affect absolute levels. For example, one might expect higher levels among patients like the current sample for whom the average level of “activation”, or readiness to engage in personal healthcare management, as measured on the PAM scale was 67.4 (out of 100), corresponding to level 3 (out of 4). In addition, one might expect the physicians to be encouraging patient engagement based on the survey results reported above and by the PCMH-level 3 designation held by the Medical Practice. Table 3: Number of interview participants reporting complete, partial or lack of adherence to medication and lifestyle directives. Type of adherence

Level of adherence Complete

Partial

None

Medication

10 (83%)

2 (17%)

0

Lifestyle directives

17 (65%)

7 (27%)

2 (8%)

The free-form comments are on the whole supportive of an assessment of an engaged sample of patients. Several of the patients, especially those who received their diagnosis recently, reported educating themselves about their primary health condition

13 Patient engagement at the point of care: technology as an enabler   

   211

through reading on the web or through following up with clinical educators. Congruent with the self-management education tenet of patient engagement, these patients reported engaging in education efforts in order to support problem-solving activities to manage their medical condition. A common problem raised by a new diagnosis that several patients mentioned was figuring out how to modify one’s eating habits. One recently diagnosed hypertensive patient reported applying her newly acquired nutrition knowledge when shopping for snacks as many of her favorites, like olives, are high in salt. A recently diagnosed diabetic reported experimenting with her lunch options until she settled on having protein shakes because she found that they satisfied her unlike many of the other alternatives she had tried. Adjustments reported by patients extended to behaviors other than eating, for example, a very sick patient with chronic heart failure who adjusted by making large positional changes slowly to minimize the side effect caused by the critical medication. Not all patients reported being successful in applying their disease-related knowledge to managing their conditions. For example, while one long-term diabetic patient reported that his wife routinely obtains menus ahead of time from restaurants they are considering visiting, another diabetic reported that she follows food guidelines for managing her diabetes except when she eats out. She noted that she finds it hard not to succumb to “peer pressure” with regards to eating and drinking when she goes out with her friends. Focusing on partnership with the physician, the second tenet of patient engagement, we obtained a similar picture of engaged patients through our interviews. Patients who experienced presumed side effects from prescribed medications collaborated with their health provider to identify the problem, report it and try out alternative medications. This collaboration required vigilance on their part to detect the potential side effects and then follow through to bring them to the physician’s attention and finally achieve agreement to try an alternative. While some patients knew what to be on the look out for (“it’s a well-known side effect”), others noted that they were unprepared for the symptoms (“…my doctor is good, but sometimes they don’t tell you everything”). Others reported being mindful of the potential for problems, for example, when the patient is taking multiple medications. Sometimes collaboration occurred in a context of partial non-adherence. For example, one patient reported that taking the twice-per-day medication only once per day (therefore, at half the daily prescribed dose) produced a blood pressure reading “within normal range” and she was unwilling, because of cost, to switch to the timerelease formulation prescribed by her physician as an alternative to the twice-daily regimen. In this case, the patient’s feedback to the physician, in the form of results from her self directed “experiment”, is valuable for establishing the effectiveness of the prescribed treatment. In spite of an overall high level of engagement, there were several notable cases of “gaps” between the physician’s directives and the patient’s actions. The most common case is what one might call “intermittent adherence”. This sometimes takes

212   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

the form of unintentional non-compliance, in other words, forgetting to take medication, which sometime happens when one’s routine is broken as it was for some of our sample during business travel. Other cases of intermittent adherence were intentional, for example, patients who reported adherence over some period of time only to lapse at some point due to competing factors. In some cases, the lapse is associated with the physical context, for example the patient who follows diet restrictions except when she is travelling for work or when on vacation. Others reported their pattern being governed by their medical condition, interpreting a temporary remission or lessening of disease symptoms as indicating that behavior change is not required. Others convince themselves that since their condition may still be considered borderline, they can continue as they are but promise themselves to change if “I fall off the edge”. We were also interested in ascertaining the role of the physician in the patient’s adherence to lifestyle directives. The physicians were unanimous in responding “no” to the question “Do you think that patients accept prescribed treatments pertaining to lifestyle changes like modifying their diet or increasing exercise in the same way that they accept a prescribed medication?” For their part, some patients reported that their physicians have not been as direct about lifestyle factors as they were with respect to medications. We cannot determine from our data whether patient perceptions that physicians are less directive about lifestyle prescriptions are correct. What does appear to be the case, however, is that at least some of the patients would like more direct involvement from physicians about lifestyle directives. We believe the above examples illustrate that the interview patients on the whole evidence a fairly high degree of engagement. So, one might ask, is there any need to introduce a technology into a situation such as this? Are there any unmet needs on the part of the patients? As we pointed out in the text above, the gaps in patient engagement fall into two broad, interconnected categories: 1) intermittent patient adherence; and 2) insufficient feedback to and from the provider. The first issue, intermittent patient engagement, is well documented[48] and presents an excellent opportunity to use technology to “nudge” the patient back towards engaged behavior prior to their next office visit, which could be months in the future. Using technology to assist the patient in making connections between health states and lifestyle choices is another immediate benefit of engagement policies and technology. For example, allowing a patient to see the real-time correlation between their “feel good” score (one of the daily check-in questions) and their medication adherence may trigger them to adhere to medical directives. The second gap that is handily addressed by technology is providing the clinician with real or near-real time data in a manageable format. Enabling the flow of information from patient to provider allows the practice to “close the loop” and interact with the patient when adherence seems at risk.

13 Patient engagement at the point of care: technology as an enabler   

   213

13.3.3 Data collected through Wellby technology 13.3.3.1 The mobile check-in: daily data reporting by patients Once patients complete the onboarding process, they move to the “active” status and start to receive three “mobile check-in” questions on their smartphone each morning. The first question asks whether the patient took all of their medication on the previous day. The second asks the patient to provide their blood glucose measurement (if their primary condition is DM type 2) or their blood pressure (if their primary condition is hypertension). And finally, they are asked to report how well they feel (their “feeling good” score) on a 1 to 5 scale, with 5 being the highest or best health. Unless their responses are outside of the expected range (as defined at the Medical Practice for their population of patients, not for individuals), the check-in is completed in less than one minute, provided the patient has taken their health indicator reading prior to initiating the response. Patients continue to receive the mobile check-in questions daily for the duration of the study. Our intention, in designing the study, was not just to gather data, but also to provide the patient with a daily trigger to consider their health and wellness. The mobile check-in process was designed to be as minimally intrusive as possible, from the patient’s perspective. This included streamlined questions and a convenient vehicle for asking them.

13.3.3.2 Value to patients of mobile data check-in One of the main values has been that the patients in the two diagnostic groups receive daily reminders to track the health indicator relevant to their primary health condition. In many cases, patients reported through the interviews that they had been instructed to take regular measurements but they were neither reminded nor held accountable for low adherence. Some patients had been instructed to take their measurements “periodically” during the week and it was up to them to define what this meant in practice. Others, particularly those with DM type 2, were counseled to take their blood glucose daily, but there was no one tracking it. With the mobile check-in they receive a daily reminder for a request to action. We have refined our process to follow up with patients who fail to respond for two days. Patients appear to take their responsibility in daily reporting seriously. We had one patient notify us that he was going to be travelling and because he was unable to pack his blood pressure cuff, he would be unable to complete the daily reporting temporarily. Patients occasionally fail to reply to a check-in message. In most cases the reasons are that the patient has been travelling or felt too busy to respond. One patient reported that the reminders were particularly useful because she had recently begun a study under her physician’s directive to determine if she indeed requires medication to maintain blood pressure levels within normal guidelines. She was directed to stop the medication and to take her blood pressure daily. At the

214   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

close of our interview, she told us “I want to thank you for being able to participate in the study. The main reason I am as engaged with the investigation of my blood pressure status is because I have the mobile app – it makes it so easy to remember and to report the data!” Value to the patients also derives from the opportunity to view aggregated data, including the opportunity to view their data trends over time. In the Wellby system, the patient’s daily reports are aggregated into his or her individual history that can be inspected through the web interface. Some patients contacted us through e-mail to inquire about the capability. In addition, we have been able to see from log reports when patients have gone to the website to view their trending reports. This sometimes happens after the patient reported poor results (i.e. missed taking their medication) or missed reporting their data. We will have to wait until we carry out study debrief interviews to learn their explanation for this behavior; we will be interested to determine whether such experiences are occasions for reflection. Others who reported their values regularly viewed their trends apparently just to determine if they experienced any significant variations. Another value derives from the ability to apprehend correlations among the responses to the three questions asked daily. For example, reporting a less than typical “feeling good score”, which is accompanied by either a low or high blood glucose level, could trigger a chain of inferences on the part of the patient. For example, one patient reported forgetting to take her dose of insulin and then noticed that her “feeling good” score had dropped on the subsequent two days. She noted the impact and then reflected on how her normal routine had been interrupted and reported that routines were important for her to follow through on for her care directives. The value seems to come from the explicit action and the opportunity to correlate the behavior with wellbeing. These types of experiences are consistent with those reported in the study that created scaffolding to promote reflection on the part of patients.[30]

13.3.3.3 Value to the providers We have collected examples of value to physicians both from the data obtained through the aggregated anonymous responses from the kiosk system and through the individual data reported on the mobile system. The kiosk data consists of anonymous patient responses to brief surveys that the patients are asked to complete at the kiosk’s touch screen interface after each appointment with a care provider. They are aimed at monitoring service quality, whether treatment guidelines are followed by the provider, whether patient engagement best practices are followed, et cetera. Very early in the study, the Medical Practice received a valuable insight into a gap in their procedures as a result of the questions in the medication reconciliation section of the kiosk survey. The Medical Practice has protocols in place to perform medication reconciliation at every visit. Early data from the kiosk indicated that nearly 20% of patients did not think that anyone had reviewed over the counter (OTC) medications,

13 Patient engagement at the point of care: technology as an enabler   

   215

just prescription medications. According to practice protocol, both types of medication are to be explicitly reviewed with the patient. The data from the kiosk presented an opportunity to revisit the Medical Practice’s policy and perform additional training of key staff. It is unlikely the practice would have become aware of this gap in care without a tool such as the kiosk. The kiosk does not just provide bad news. The practice was happy to find that in categories such as Professionalism and Patient Satisfaction, they received near-perfect scores. This serves to reinforce practice behaviors and uplift morale. We are also collecting valuable feedback on how the collected patient information needs to be packaged in order to fit in well with the workflow of the Medical Practice. In some cases we are finding that the granularity of the information is not suitable for the goals of the person reviewing the information. And, in other cases, we are discovering that a role player other than whom we had assumed will be consuming the information. We are engaging in prototyping efforts to explore how to meet the information needs that the newly available data must satisfy in order for it to become a useful resource for providers. One of the other values of the data provided by patients in response to the daily check-in questions is for what it can help the physician comprehend about problems at the individual patient level. Besides being entered into a database for subsequent interrogation, a health indicator value (blood pressure or blood glucose level) submitted by a patient is also evaluated with respect to threshold values set by the Medical Practice. Any value which is either too low or too high is tagged and generates an “alert” that is routed to the designated “triage nurse” at the Medical Practice for clinical follow-up. This filtering action is necessary to prevent the Medical Practice from being overwhelmed by data. The goal is to titrate the alert-triggering levels so that only clinically meaningful events are surfaced to the clinical staff. The observation that two months into the study we have yet to generate any alerts is causing us to reflect on the threshold levels (are they set correctly?) and also to ask the deeper question of what the thresholds should be set to detect critical situations or opportunities to influence behavior before a critical situation is reached).

13.4 Discussion Our report is based on a study still in progress. It will run for another 4 months, which will give the most recently enrolled participants a full five-month experience with the patient engagement technology. We think there have already been a number of interesting findings, however, which we categorize into three topic areas: pragmatic issues related to running a study such as our field trial, feedback on the value of a mobile patient engagement system and pointers to future work. The main difference between this study and previously published literature on the use of IT-enabled

216   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

patient engagement support is the rich first-hand account of use that we have collected. Through qualitative and quantitative methods, we have been able to observe the multi-faceted reactions of the two main stakeholders involved in the deployment of patient engagement tools.

13.4.1 Integrating technology-supported patient engagement into a practice Because we were interested in exploring the use of patient engagement technology under “real” conditions, we elected to do a field trial at a busy independent primary care facility and to enroll patients who are actively receiving treatments for chronic diseases. We were aware that this would be harder than running a laboratory study (in which we would collect feedback on the functionality and usability of the mobile system from various potential user groups such as patients, physicians or administrators), but we did not realize how much harder it was going to be. The most significant problem was that as much as the clinic welcomed the research team and as much commitment as the staff of the clinic showed towards the study, we were still “outsiders”. As such, we did not have the authority to make requests of either patients or staff and when day-to-day time pressures arose, the research suffered. One example of this detrimental separation between research and clinical efforts was the kiosk patient response rate. Overall, we have a 20% response rate – which is very good, given the fact that the first two months of the study took place during the holiday season and during one of the region’s snowiest winters in recent memory, but a previous research study[46] had led us to expect it to be higher. One of the major differences between our study and the aforementioned comparison study, where the response rate exceeded 80%, was that the comparison study was done with a full-time paid research staff that was integrated into the clinical practice, and the practice was part of a large healthcare academic health system. That is, the research and the clinical missions were fused. In spite of our study’s disappointing recruitment rate (for a variety of reasons, some of which are discussed below), or perhaps because of it, we think that our study is a very important data point for researchers doing “live” case studies – it highlights some of the very significant problems that need to be overcome to obtain the necessary data. Another surprising finding concerns the low rate of smartphone ownership by the patients at our study site. Given mobile phone penetration rates exceeding 80% in the US and in the urban location of the study site, we did not expect this to be such a problem. We consciously decided to require that patients use personal phones, rather than providing one to them as other studies have done (e.g. [30]), because we did not want to introduce a potentially confounding variable – inexperience with the technology. In spite of our results, we stand by our decision but, nevertheless, believe it raises significant issues related to the reach that technology such as ours can have. As patient recruitment progressed, it became clear that the smartphone ownership requirement was a significant obstacle for otherwise interested patients. The study

13 Patient engagement at the point of care: technology as an enabler   

   217

did not provide an option for patients who had access to a computer but did not possess a smartphone or tablet, and this additionally decreased our sample size of suitable patients. This issue illuminated the importance of providing as much flexibility as possible for patient access to solutions. Smartphones are ever-present among those that own them and thus an appealing vehicle for patient contact, but healthcare solutions must address the entire population, not just the convenient population. Of course, trade-offs would have to be considered as a result of losing the affordances provided by a mobile device. A final pragmatic factor to mention is the limited technology expertise of our study participants. In addition to requiring that a participant own his or her own smartphone, we required that they have used e-mail, and have used applications on their smartphone. We were thus surprised at the large percentage of study participants who encountered a variety of technology roadblocks such as forgetting logon credentials or not being able to recover a link to the study website. These were made worse by the general lack of full-time project staff to help participants with study processes.

13.4.2 Patient use of reporting technology Once patients achieved “active” status, they were willing to report on a daily, or nearly daily basis. This is a very positive finding. It was interesting to find that though people were not necessarily 100% adherent – they occasionally missed for a variety of reasons – they typically resumed replying to the check-in messages. We have found that people have been consistent for over two months, though their patterns of responding differ. Almost 75% of mobile responses take place on the same day as the request, but there are a small number of participants whom we consider to be active, even though they only respond once a week. One of the questions that research like ours will have to answer is, what is the optimal frequency of reporting? Is it important to have patients check-in every day or nearly so, or is some other schedule sufficient? What additional functionality or external motivation needs to be provided to keep patients reporting for the long-term? We also have preliminary evidence that patients are using the collected data to reflect on their health-relevant habits. This is exciting, in part because of the importance that previous research[26, 30] has attributed to self-reflection in chronic care management. In addition, it may set up a “virtuous cycle” in which gaining insight from use motivates further use. This could lead to the patient generating additional data that is valuable to a variety of stakeholders. Based on our experiences, we would characterize the patients we encountered into three groups based on the degree of engagement they demonstrated. The first group – “bucket A” – includes patients who were very engaged with the technology. Most of the patients in the “active” group fit here. They regularly respond to the mobile

218   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

check-in, and they explore additional tools such as representations of historical data. The second group includes patients who were initially enthusiastic about participating but for one reason or another fell short of completing the onboarding process or rarely responded to the daily check in requests after achieving active status. Preliminarily, many of their obstacles seem to be technological; we will interview a sample of these patients at the end of the study to gain more definite insights and determine whether lowering the technological requirements might result in engaging them more effectively. The final group were not willing to engage, stating they did not have time or would not be willing to make a commitment to participate for the duration of the study. We will also attempt to interview some of this sample to determine whether they might have been engaged by more flexibly drawn study participation requirements.

13.4.3 Future explorations into the value of patient-reported data At the most basic level, the data collected through the patient reporting technology can be used to provide the healthcare team with a glimpse into the patient’s life between appointments. If the patient’s health indicator measure exceeds a threshold, it can be programmed to send an alert and set in motion a clinically defined response. Thus, the data generation behavior contributes towards the patient’s partnership with the physician. It is the goal of trials such as ours to try to explore various options with both the patients and the physicians to learn how the data can benefit overall care management and quality reporting. With respect to the lack of alerts we have thus far observed, we note that it is possible to detect levels that are indicative of possible bad trends though not requiring immediate attention. This in fact is one of the long-range goals of patient engagement: detect patterns before they are full-blown problems. Clearly this will require explorations with other stakeholders in the process to determine whether and under what circumstances such a result might be useful. Collecting patient-generated data can also be directed at providing additional support for the patient to view trends in their measures, detect correlations among them and thus support reflection on their health-relevant behavior. Recall that the patient interviews revealed that in spite of being a relatively highly engaged group of patients, some of them nevertheless reported that their engagement waxed and waned over time as a function of clinical status, competing demands on their time and changing motivation. We did not address patient incentives or payer incentives here as a possible motivator. It would be valuable if one were able to detect when a patient’s adherence is waning and use the mobile system or, perhaps even a clinical staff member, to intervene at that point in time to renew the patient’s commitment. We believe this use of technology may be particularly valuable with respect to lifestyle changes. As we heard both in the patient interviews and the physician surveys, the physicians appear to be less involved with helping patients manage the “process”;

13 Patient engagement at the point of care: technology as an enabler   

   219

they are more focused on “outcomes”. Technology-based support may be particularly useful in cases, such as this one, in which patients expressed a need for greater process support.

13.5 Conclusion Our ongoing deployment of a technology system to support patient engagement to a group of patients in a PCMH-level 3 Medical Practice has produced a number of promising results regarding the use of daily tracking functionality. Patients who manage their chronic diseases derived a variety of values from daily reporting of their relevant health indicators and the Medical Practice gained important insights into their care delivery practices from collecting anonymous feedback from patients. As a case study, our contribution is limited to a particular group of patients, healthcare providers and researchers and thus the reader must exercise caution in generalizing from our findings. As we noted at the beginning of our report, we hope to see similar case study contributions from other researchers who may have deployed other systems in order to understand the value of patient engagement tools more broadly. However, we believe that one of the additional values to be derived from our report is the description of the difficulties that we encountered in both identifying patients who were suitable candidates for the study (i.e. had the requisite mobile phone and biometric technology) and in supporting patients’ ongoing participation during the trial. Much of the evaluation of interventions in health technology is based on efficacy trials – tests of the technology under controlled conditions – and such results often do not easily translate into expected effectiveness of systems or performance under real world conditions. Our findings highlight the importance of testing systems in the ecosystems that define their expected context of use. Such studies may help future groups who undertake deployments to be aware of potential problems and thus hopefully anticipate them, thereby increasing their chances of successful technology adoption.

Acknowledgements We thank all of the patients and staff at the Medical Practice without whom this study could not have been carried out. We also thank our colleagues on study for their contributions: Allan, Crimm, M.D., FACP; Rachel Slutsky, BSN, RN-BC; Rosemary Kennedy, PhD, RN, MBA, FAAN; Zuleika C. Font, M.D.; Cara Dolan, BSN, RN, MA; Bency Thomas, BSN, RN, MA; Sue Yeo, BSN, RN, MA; Diane Freed, MSN, RN; Ingrid Hilghman; and Thomas Blue, PhD.

220   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

13.6 Appendix 13.6.1 Patient barriers interview 1. 2.

Can you tell us what you normally see the doctor for? Please tell us the three most important things that you are supposed to do to care for your primary health condition. Follow up with: Are there other things that you are also supposed to do? 3. Which of the things your primary physician has recommended you do have you been focusing on the most in the past two weeks? Why did you decide to focus on them in particular? 4. Do you have other health goals that you are working on that did not originate with your physician, that is, goals you set for yourself? 5. What problems have you encountered in carrying out what your physician asked you to do? (Prompt if necessary to cover all treatment categories). And how have you tried to overcome the challenges you have faced? Is there anything you can think of that would help you to deal with these challenges better? 6. How about the goals you set for yourself – prompt if necessary to cover all goal types enumerated in Q.4 – what challenges have you encountered in regards to meeting these goals? How have you tried to overcome them? What might be helpful to you? Now, finally, two summary questions on a 1 to 5 scale – with 1 being the best and 5 the worst: 7. Can you rate yourself on how engaged or involved in your health care you are? 8. And how successful would you say you have been in taking good care of yourself – again 1 is the best and 5 is the worst?

13 Patient engagement at the point of care: technology as an enabler   

   221

13.6.2 Provider questionnaire 13.6.2.1 Provider name/title 1. In your opinion, what is the approximate percentage of your patients with chronic illness who you feel understand the ramifications of their condition(s)?

____%

2. Do you think that your patients accept prescribed treatments pertaining to lifestyle changes like modifying their diet or increasing exercise in the same way that they accept a prescribed medication?

Yes

No

3. Do you believe that your patients generally leave your office with a clear under- Yes standing of what you are directing them to do?

No

4. Do you believe your patients feel free to ask the right/ important/ necessary questions during an office visit?

Yes

No

5. Do you often feel rushed during appointments and therefore don’t get to cover important issues with patients?

Yes

No

6. Do you believe your patients often feel rushed and they go away with unanswered questions?

Yes

No

7. What percentage of your patients is generally satisfied with the physicianpatient relationship when they leave your office?

____ %

Often

Sometimes Never

8. Do you generally feel that information technology is an: intrusion/ barrier to your relationship with your patient adjunct to your relationship with your patient, either during or after the time of a visit 9. Do you believe your patients generally feel that information technology is an: intrusion/ barrier to your relationship with your patient adjunct to your relationship with your patient, either during or after the time of a visit

10. Do you believe it is the role of the physician to help patients understand how technology may be helpful to them?

Yes

No

222   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

11. Please rate the importance of these factors in contributing to the patient's positive health outcomes

Extremely important

Very important

Important

Not very important

Not important

12. Please rate the following barriers Extremely to patient engagement as you see important them in your patient population:

Very important

Important

Not very important

Not important

Following medical orders Scheduling and attending recommended medical visits Proper nutrition Level of education Income level Web-based resources Patient engagement Educational handouts Targeted interventions (e.g. CDE) Supportive home/ family situation Receiving feedback on a regular basis Other – please specify below

Age Access to internet and other technology Income Education Social issues Family problems Other (please describe):

13 Patient engagement at the point of care: technology as an enabler   

13. Please rate the following indicators of patient engagement as predicators of success in management of chronic illness

Extremely important

Very important

Important

Not very important

   223

Not important

Asking for educational resources Keeping records of specific testing as directed, i.e. blood pressure, blood sugar Attending office visits with a list of questions Use of web/ computer-based tools Communicating information between visits Communicating social and family problems Other (please describe):

14.Do you adjust your treatment/ treatment plan to accommodate the level of patient engagement in his/ her own care?

Yes

No

If you answered “yes”, please provide an example: 15. What do you believe is the number one thing that would improve your patients’ adherence to their care plan?

References [1] [2] [3] [4]

[5] [6] [7] [8]

AARP. Washington, D.C.: AARP Public Policy Institute; 2009 [cited 1 Nov 2006] Available from: http://www.assets.aarp.org/rgcenter/health/beyond_50_hcr.pdf Berwick DM. Promising Care: How we can rescue health care by improving it. San Francisco: Jossey-Bass Publishers; 2014. James J. Health Policy Brief: Patient Engagement. Health Affairs. February 14, 2013. Lorig KR, Sobel DS, Stewart AL, Brown BW, Bandura A, Ritter P, et al. Evidence suggesting that a chronic disease self-management program can improve health status while reducing hospitalization. Medical Care 1999;37(1):5–14. Coulter A. Patient engagement-what works? J Ambul Care Manage. 2012; 35(2):80–89. vonKorf M, Gruman J, Schaefer JK, Curry SJ, Wagner EH. Collaborative management of chronic illness. Ann Internal Medicine 1997;127:1097–1102. Anderson, RM, Funnell, MM. Patient Empowerment: Reflections on the challenge of fostering the adoption of a new paradigm. Patient Education and Counseling 2005;57:153–157. Bodenheimer T, Wagner, EH, Grumbach, K. Improving primary care for patients with chronic illness. JAMA 2002;288(14):1775–1779.

224   

[9| [10]

[11] [12] [13] [14] [15]

[16]

[17] [18] [19] [20] [21]

[22]

[23] [24]

[25]

[26]

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

CapGemini Consulting. Patient Adherence: The Next Frontier in Patient Care. [Internet]. 2011. Available from: http://www.capgemini.com/consulting HIMSS. Personal health information technology: paradigm for providers and patient to transform healthcare through patient engagement. [Internet]. 2013. Available from: http://blog.himss.org/2013/09/13/the-need-for-personal-health-it-partnering-with-patientsto-change-a-paradigm Rand Corporation. More changes in health care needed to fulfill promise of health information technology. [Internet]. 2013. Available from: //www.rand.org/news/press/2013/01/07.html Kellerman AL, Jones SS. What will it take to achieve the as-yet unfulfilled promises of health information technology? Health Affairs 2013;32(1):63–68. Soegaard M. Affordances. [Internet]. 2013. Available from: http://www.interaction-design.org/ encyclopedia/affordances The Diabetes Prevention Program Research Group (Biostatistics Center, George Washington University). Description of lifestyle intervention. Diabetes Care 2002;25(12):2165–2171. Whelton PK, Appel LJ, Espeland MA, Applegate WB, Ettinger WH, Kostis JB, et al. Sodium reduction and weight loss in the treatment of hypertension in older persons: a randomized controlled trial of nonpharmacologic interventions in the elderly (TONE). JAMA 1998;279(11):839–46. Erratum in: JAMA 1998;279(24):1954. Jones L, Danis CM, Boies SJ. Avoiding the mistake of cloning: a case for user-centered design methods to reengineer documents. In: Sprague RH, editor. HICSS-32. Proceedings of the 32nd Annual Hawaii International Conference On System Sciences; 1999, January 5–8; Maui, Hawaii. New York: IEEE; 1999. Vol 5:Track 2. DiMatteo MR, Giordani PJ, Lepper HS, Croghan TW. Patient adherence and medical treatment outcomes: a meta-analysis. Medical Care 2002;40(9):794–811. DiMatteo MR, Haskard-Zolnierek KB, Martin LR. Improving patient adherence: a three-factor model to guide practice. Health Psychology Review 2012;6(1):74–91. Michie S, Abraham C, Whittington C, McAteer J. Effective techniques in healthy eating and physical activity interventions: a meta-regression. Health Psychology. 2009;28(6):690–701. Michie S, vanStralen MM, West R. The behaviour change wheel: a new method for characterizing and designing behaviour change interventions. Implementation Science 2011;6(42). Danis CM, Kellogg WA, Farrell R, Christensen J, Levine D. Reframing adherence: collaborative enactment of healthcare. Paper presented at: the the Workshop on Bridging Clinical and Non-clinical Practice at the Conference on Human Factors in Computing Systems; 2012 May 5–10; Austin, Texas. Harris Interactive & GlaxoSmithKline. Chronic Care in America: Improving the PatientPhysician Interaction. [Internet]. 2003. Available from: http://www.policyarchive.org/ handle/10207/95528 Goold SD, Lipkin M. The doctor–patient relationship: challenges, opportunities, and strategies. J Gen Intern Med 1999;14(Suppl 1):S26–S33. Hibbard JH, Mahoney ER, Stock R, Tisler M. Self-management and health care utilization: do increases in patient activation result in imposed self-management behaviors. Health Serv Res 2007;42(4):1443–1463. Consolvo S, Klasnja P, McDonald DW, Avrahami D, Froelich J, LeGrand L, et al. Flowers or a robot army: encouraging awareness and activity with personal mobile displays. In: McCarthy J, Scott J, Woo W, editors. UbiComp’08. Proceedings of the 10th International Conference on Ubiquitous Computing; 2008, September 21–24, Seoul, Korea. New York: ACM, 2008. P. 54–63. Munson SA, Consolvo, S. Exploring goal-setting, self-monitoring and sharing to motivate physical activity. In: Bardram J, Mihailidis A, Tentori M, editors. Pervasive Health ‘12.

13 Patient engagement at the point of care: technology as an enabler   

[27] [28]

[29]

[30]

[31] [32]

[33] [34]

[35] [36] [37] [38]

[39] [40] [41]

[42] [43]

   225

Proceedings of the 6th International Conference on Pervasive Computing Technologies of Healthcare; 2012, May 21–24; San Diego, CA. New York: ACM; 2012. P. 25–32. Preece J. Empathic communities: reaching out across the Web. Interactions Magazine 1998:2(2):32–43. Huh J, Ackerman MS. Collaborative help in chronic disease management: supporting individual problems. In: Grudin J, Mark G, Riedel J, editors. CSCW ‘12. Proceedings on the computer Support Cooperative Work Conference; 2012, February 11–15; Seattle, WA. New York: ACM Press. P. 853–62. Huh J, McDonald DW, Hartzler A, Pratt W. Patient moderator interaction in online health communities. In: AMIA ‘13. Proceedings of the American Medical Informatics Association Annual Symposium; 2013, November 16–20; Washington, DC. Washington DC: American Medical Informatics Association; 2013. P. 627–36. Mamykina L, Mynatt ED, Davidson PR, Greenblatt D. MAHI: investigation of social scaffolding for reflective thinking in Diabetes management. In: Tan D, editor. CHI ‘08. Proceedings of the Conference on Human Factors in Computing Systems; 2008, April 5–10; Florence, Italy. New York: ACM; 2008. P. 477–86. Whittaker R, Borland R, Bullen C, Lin RB, McRobbie H Rodgers A. Mobile phone-based interventions for smoking cessation. Cochrane Databse Syst Rev 2009;(4):CD006611. Wang KC, Hong MT, Kao CHL, Lin AC, Wang CA, You CW, et al. A phone-based support system to assist alcohol recovery. In: Brewster S, Bodker S, editors. CHI ‘13. Proceedings of the Conference on Human Factors in Computing Systems; 2013, April 27–May 3; Paris, France. New York: ACM; 2013. P. 529–34. Olson JM. Psychological barriers to behavior change. Canadian Family Physician 1992;38:309–319. Hibbard JH, Stockard J, Mahoney ER, Tusler M. Development of the Patient Activation Measure (PAM): conceptualizing and measuring activation in patients and consumers. Health Serv Res 2004;39(4):1005–1026. Prochaska JO, Fava JL, Norman GJ, Redding CA. Detailed overview of the Transtheoretical Model. [Internet]. Available from: http://www.uri.edu/research/cprc/TTM/detailedoverview.htm HITECH Programs and Advisory Committees. [Internet]; 2014. Available from: http://www. healthit.gov/policy-researchers-implementers/hitech-programs-advisory-committees 2014 Definition Stage 1 of Meaningful Use. [Internet]; 2014. Available from: http://www.cms. gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Meaningful_Use.html Santa Rosa Consulting. Meaningful Use Stage 2 Patient Portals Must Focus on the Patient and Engagement. [Internet]; 2013. Available from: https://www.santarosaconsulting.com/ SantaRosaTeamBlog/post/2013/01/03/Meaningful-Use-Stage-2-%E2%80%93-Patient-PortalsMust-Focus-on-the-Patient-and-Engagement.aspx HowsYourHealth.org [homepage on the Internet]. Corporation and trustees of Dartmouth College; 2014. Available from: http://howsyourhealth.org/hc EmmiSoltuions.com [homepage on the Internet]. Emmi Solutions. Available from: http://www.emmisolutions.com Perna G. Web-based patient engagement lowers anxiety for first-time colonoscopy patients. [Internet]. 2014. Available from: http://www.healthcare-informatics.com/news-item/ web-based-patient-engagement-lowers-anxiety-first-time-colonoscopy-patients Epic.com [homepage on the Internet]. Epic; 2014. Available from: https://www.epic.com/ software-phr.php Tactiosoft.com [homepage on the Internet]. Tactio Health Group; 2014. Available from: http://www.tactiosoft.com/tactiorpm

226   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[44] AT&T mHealth Solutions Presents Diabetes Manager. [Internet]; 2014. Available from: http:// www.corp.att.com/healthcare/wfm/docs/mhealth.pdf [45] NCQA.org [homepage on the Internet]. Available from: www.ncqa.org [46] DiRocco DN, Day SC. Obtaining patient feedback at point of service using electronic kiosks. American J Managed Care 2011;17(7):270–276. [47] CarePartnersPlus.com [homepage on the Internet]. CarePartners Plus; 2014. Available from: https://www.carepartnersplus.com [48] DiMatteo MR. Social support and patient adherence to medical treatment: a meta-analysis. Health Psychology 2004;23(2):207–218.

Maarten van der Heijden, Marina Velikova and Peter J.F. Lucas

14 Supporting active patient self-care Abstract: We are currently confronted with a trend of increased pressure on health care, with associated increasing financial costs, due to an aging society and the expected increase in the prevalence of disability and chronic disease. Finding measures for cost reduction, without sacrificing quality of care, is a significant healthcare challenge. Computing technology offers promising solutions in this respect. In this article, we review contributions made by mobile computing technology in supporting the care process. In particular, we consider mobile decision support with a view to enable patient self-management, transferring part of the clinical care from healthcare professionals to the patient. Mobile computing can play an important role in giving patients an active, decisive role in managing their own disease.

14.1 Introduction Current challenges in health care include cost reduction without sacrificing quality of care, and improving the quality of life. A reduction in the number of face-to-face consultations in outpatient clinics and the prevention of costly hospitalization through early risk-detection are typical examples of cost reduction indicators. At the same time, patients’ involvement in making decisions about their health and the management of their diseases has become more and more important. Measures taken to realize this have been referred to as patient empowerment. A particularly important step to achieve this is adapting decision-making to the individual’s characteristics, usually referred to as personalization. In this chapter, we consider the computing aspects of turning a mobile computing and communication platform into a personalized smart care-assistant. The architecture and the typical set-up of user interaction modules of such a smartphone-based monitoring system is presented in Figure 1. Such care assistants allow part of the clinical decision-making process to be moved from the clinic to anywhere where the patient resides: at home; at work; or on holiday. In particular, we consider how such a care assistant can be constructed and what the important characteristics are in order to support active decision-making by patients. Although we focus on self-management of one disease in this chapter, the aim is to deliver optimal health care, offering support to both patients and physicians. In the current state of affairs there is little to no computing research addressing the problem of managing multiple diseases at the same time. Mobile computing is expected to offer an important contribution to cost reduction, quality improvement and larger patient participation in health care.[1, 2]

228   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Figure 1: Modules and user interactions in the smart care-assistant’s framework.

14.2 The current situation: needs, gaps and challenges 14.2.1 Health care at distance: a brief history Figure  2 offers a visual summary of the development of the field. Starting as early as 1905 and leading up to the present, we see an increase in patient involvement in tandem with technological advances. From the mid-1990s on, eHealth emerged as a promising field for better and more efficient healthcare delivery using web-enabled services. In comparison to telehealth, eHealth is a broader term that encompasses health services, information, education and research. Subsequently, many eHealth systems and tools were developed, including commercial ones such as the Bosch Health Buddy and Intel Health, to monitor a patient’s condition, usually in the home environment, and enable earlier diagnosis and more effective treatment. The availability of modern mobile computing and communication technology offers new avenues to move part of the medical decision-making tasks, such as diagnosis, selection of appropriate treatment and prognosis from the doctor to the patient at any place and any time. This technology has given rise to the field of mobile health or mHealth, a term coined in 2004.[4] Currently, the most commonly used mHealth

14 Supporting active patient self-care   

   229

Figure 2: Timeline development of health care at a distance. *Numbers as reported in [3].

technologies include smartphones, wireless tablet computers, wearable wireless biosensors and disease-monitoring devices. A large number of mobile health applications are already available, focusing primarily on healthy-lifestyle tracking, or on the remote monitoring of a few basic vital signs and symptoms in combination with the transmission of data to the clinician for interpretation. These systems simply collect data and provide no, or only very limited, user feedback. They lack the clinical interpretation capabilities of a medical doctor, who takes into account the patient’s characteristics, history and environment, including the nature of the underlying disease. Within the context of disease management, this implies that the patient is passive and mostly dependent on the caregiver, while the clinicians may be overwhelmed by real-time incoming patient data.

14.2.2 Clinical decision-making Part of clinical decision-making concerns the process of finding a diagnosis. A clinician has to decide, based on the patient’s history and current clinical observations, on the set of possible explanations of the presented symptoms and signs. The relevant decisions are whether to perform tests, and if so, which tests will distinguish the diagnoses. Based on the results, a further choice has to be made on what treatment is appropriate, taking into account disease prognosis. A challenging aspect of these processes is dealing with uncertainty. Clinical decision support systems emerged as early as the 1970s to assist medical doctors with difficult tasks such as diagnosis and treatment selection by

230   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

using artificial intelligence techniques.[5, 6] The basic knowledge for the development of clinical decision support systems comes from evidence-based clinical guidelines, treatment protocols, research practices and input from clinical experts. Providing assistance in the management of one or two disorders that are limited in nature is now feasible given the current state of the art. However, reliably dealing with a broad range of diseases is still not feasible, due to the inherent complexity of disease interaction. A care assistant primarily developed for use by patients by definition operates within the confines of a disease that is already diagnosed. The important aspect of chronic disease care is managing the progress of the disease and responding adequately to deteriorating health. Although this does not change the fundamental characteristics of the problem, the domain is reduced to more manageable proportions that can be captured in a computerized decision-support system. Requirements for smart mHealth systems have been formulated in the literature. [7] However, actual systems embedding clinical decision-support were either implemented on devices with limited computational capabilities and therefore unable to deal with complex clinical problems;[8, 9] or rely on transmitting data to a healthcare professional for interpretation (see e.g. [10] for examples concerning chronic obstructive pulmonary disease (COPD)). Decision aids that include contextualized data interpretation are not in wide use yet, and progressing from specifications and prototypes to practical systems appears to be the current challenge.

14.3 Proposed solutions: personal smart care-assistants 14.3.1 Remote patient data collection The continuous monitoring and assessment of the individual’s health status requires the collection of biomedical data including symptoms (e.g. fatigue, headache), signs (e.g. blood pressure, temperature) and biosignals (e.g. electrocardiography (ECG), obstetric ultrasonography). Such data collection outside of the hospital and performed by the patient has been facilitated by recent technological advances in sensors and measuring devices that are compact, cheap, easy to use and wirelessly connected. Compact readers for performing stand-alone biochemical tests such as glucose, hemoglobin and urine analysis are already widely used in home-based practice. Recent developments include usage of the built-in camera for automated image analysis of biochemical test results on the mobile device.[11, 12] Current mobile technology is capable of complex processing of biosignals, including filtering noise, sampling and quantification. This is exemplified by products such as mobile ECG devices allowing real-time heart monitoring and wireless transmission of data to a smartphone or tablet. Furthermore, body area networks – small, intelli-

14 Supporting active patient self-care   

   231

gent devices attached to or implanted in the body – provide real-time feedback to the user or medical staff via wireless communication.[13] Despite the wide availability of measurement devices, using them to provide raw data to mobile devices for further analysis sometimes remains a challenge, because the manufacturer often restricts access to the communication protocols or requires the user to get external software for reading the measurements. Currently this poses limitations on the development of flexible and personalized smart care assistants and restricts both patients and doctors in choosing from the sensors and devices available on the market. A further limitation is to ensure that the measurements are reliably taken by the patient, for which particular procedures, skills or training may be required. See also [14] for a review of wireless technology in disease management.

Personalization For data collection, one aspect that could be personalized is the nature of alerts. Whereas a user who often interacts with smartphone apps possibly would not mind being interrupted by relevant warnings, such as an alert for medication intake, another user might dislike such behavior. Preferred days and times should therefore be taken into account when presenting reminders concerning measurements, medication intakes or filling out questionnaires. Depending on the individual’s health status, the rate of data acquisition can also be varied. If the model predicts a low current risk, monitoring can take place on a weekly basis, keeping the time investment at a minimum. When the risk increases, the system check-in can be automatically adjusted and scheduled daily to ensure the possible worsening is detected and acted upon appropriately. Whether this is appropriate requires an assessment of the cost of data acquisition versus the risk of missing a clinically relevant event.

14.3.2 Embedded intelligent models For patient self-management, decision aids should be embedded within the mobile devices – enabling direct data interpretation – to provide accurate information about options and outcomes, and help patients become involved in decisions concerning their health. Given the cause-effect relationships and the uncertainty inherent in the medical domain, we use a special class of probabilistic graphical models – Bayesian networks – as appropriate technique for building clinical decision models for smart care-assistants. A Bayesian network is an acyclic-directed graph consisting of vertices representing random variables of interest and arcs representing dependencies between variables.[15] Each random variable has a quantitative part, denoting conditional probabilities of the type P (Y |pa(Y)); that is, the probability that Y takes on a specific value given the values of its parent variables pa(Y). Probabilities of inter-

232   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

est can be computed from the joint probability of all variables, e.g. the probability of health deterioration given the evidence obtained from monitoring. An important observation is that although the model describes general relations between the variables of interest, all predictions are personalized by entering patient-specific data. Bayesian network models can be built manually using domain knowledge, learned from data or constructed through a combination of both approaches. An important advantage of Bayesian networks is that they allow modeling both state and temporal processes, which are inherent in the medical domain. In the state model, the functioning of a particular organ X at a specific moment t determines the outcome of laboratory tests and the presence or absence of symptoms. The set of associated symptoms, abnormal signs or conditions are indicators for the dysfunction of one or more body organs and can be used to establish a diagnosis. The relation is not necessarily directly causal, and it is possible that an underlying problem or risk factor such as existing diseases, age and genetics explains the association. Moreover, treatment decisions affect the status of organ functioning, aiming to reduce risks or further deterioration. The schematic representation of these relationships is depicted in Figure 3a, where the arrows indicate the cause-effect direction. The health monitoring of a patient usually includes observations made at different time points 1, … , T . Taking into account the functioning of X at the previous observation t − 1 allows the medical practitioners to observe relative change in order to establish a diagnosis at t, and if needed, adjust treatment. Furthermore, given the history and the current status, one can attempt to predict the development at the next time point t + 1. This chain of temporal dependencies is depicted in Figure 3b.

Figure 3: Modeling the development of a syndrome. These systems are discussed in section 3.5.

14 Supporting active patient self-care   

   233

We followed these state and dynamic modeling principles in developing the diseasespecific clinical models for COPD and preeclampsia, embedded in the smart careassistants.[16, 17]

Personalization For some diseases, patients may be divided in groups sharing particular characteristics. For example, which symptoms are likely may differ between patients, but this information is usually not readily available from patient records. Therefore, it may be useful to adapt the interpretation model to individual patients. Personalization then entails model updating, sometimes also referred to as online learning, using the initial model as a prior and the data gathered from this patient as new data to change the probability parameters. The interpretation model learns which relations are different for this particular patient compared to the group for which the model was constructed. Clearly, model updating is only useful when patients use the system for an extended period of time.

14.3.3 Decision-making support The success of a smart care-assistant depends for a large part on correct clinical decision-making. This requires a mapping from data and model outcomes to choices. Since we focus on Bayesian networks as prediction models, it appears natural to use influence diagrams (see e.g. [18]), which can be seen as an extension of Bayesian networks with decisions. An influence diagram is a graph G = with N a set of nodes partitioned into chance nodes, decision nodes and value nodes, and A as a set of arcs denoting dependencies. The probabilistic semantics of a network without decisions is the same as for a Bayesian network. The value nodes assign utilities, cost or benefit, to states of the parent variables. Decision nodes can have arrows to all types of nodes, indicating influence on a random variable or on a value node, or precedence on further decisions. Techniques for solving influence diagrams try to find the optimal policy, that is, the decisions such that the expected utility is maximized. One of the advantages of using influence diagrams stems from their graphical nature, which facilitates interpretation. In a clinical context this is an important feature because decisions, especially when made automatically, should be traceable. Preferences are encoded in an influence diagram via utilities, which means that the value of possible outcomes should be established. Determining utilities is crucial for correct decision-making, but is not a trivial task, which, in a healthcare context, was noted already by Torrance and Feeny in 1989: “Whatever method is used to elicit utilities […] We cannot overemphasize the importance of care and precision in preparing, testing and using these instruments”.[19] For mHealth, this poses a problem

234   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

because ideally the possible outcomes are judged by the individual users. Due to the difficult and time-consuming nature of the elicitation process, it does not appear feasible to do so. Instead, decision policies would have to be constructed either from the preferences of a few characteristic patients, or by healthcare professionals. The latter seems reasonable for clinical decisions on treatment, however, decisions on what information to present and in what fashion should be made based on patient preferences. Luckily, the latter preferences are likely to be easier to elicit and to adjust over time, as they do not directly impact health status.

Personalization Personalization of various parts of the smart care-assistant influences decision-making. For instance, the decision to register data can be left to the patient. This is useful as it reduces the burden on patients to provide data regularly, but requires a higher degree of self-management. Whether this is a realistic option will depend on the domain and the type of patients that will be using the system. Further, personalizing the models, as discussed above, has a direct impact on automatic decision making by changing the probability parameters.

14.3.4 User interaction Besides purely clinical decisions, for example about treatment or medication dosage, the system or system developer also has to make decisions about what kind of information to present. A care assistant can help patients with understanding their disease by presenting, in a meaningful way, the data that have been gathered. This implies visualization of history, trends and previous clinical decisions; possibly in different formats – like easily understandable graphics or more detailed data views – depending on the user’s preference. This is a direct way to support patients to make their own informed decisions, which is an important goal of self-management. Similarly, prognostics can be supported by providing the interpretation and predictions from the model. An important part of supporting decisions is making the right knowledge available, but if the decisions are complicated, a smart care-assistant might provide more direct guidance on the appropriate course of action. This leads to output patterns ranging from only presenting information, as described above, to automated decisions to generate an alert directly to a care provider. Depending on the domain, it may be appropriate to let the system make autonomous decisions, for example in life-threatening situations. However, self-management should be supported in other domains by informing the patient of appropriate actions based on the available data, or by advising to take a particular action, but leaving the final decision to the patient. For example, it is a design choice whether to advise the patient to contact a care pro-

14 Supporting active patient self-care   

   235

vider when a deterioration of health has been predicted, or whether an automatic alert will be generated. A typical example of the kind of advice that could be provided is “Please repeat the measurements in 6 hours”.

Personalization The interpretation capabilities of the assistant also require an appropriate interface for displaying and communicating the current health status to the user. It is expected, for example, that most well-educated people would in general be interested in obtaining insight into their health condition, possibly in terms of an explanation of how well the condition is under control. If the system is used long enough, it should also be possible to learn when it is opportune to directly contact a caregiver, or when it should only be advised. This depends for a large part on the severity of the situation and the self-management capability of the patient. The type of feedback that is provided can be customized to obtain a comfortable and safe level of self-management in consultation with both the patient and caregivers.

14.3.5 Application of personal smart care-assistants We next present the application of the smart care-assistant in the context of two disorders: (i) chronic obstructive pulmonary disease (COPD), a lung disease; and (ii) preeclampsia, a pregnancy-related disorder. Each assistant consists of: – a mobile application (app) for (i) automatic and manual collection of patient data, (ii) automated interpretation of the data based on an embedded Bayesian network model and (iii) reporting the patient’s status to the patient and the healthcare team; and – a number of sensors and tests to measure the patient’s condition. In the following we report mainly on the intelligent components of the two care-assistants in terms of the models built and their capabilities obtained from evaluation on patient data and from pilot studies. In particular, we evaluated the abilities of the models for (i) diagnosis – accurate detection of the clinical condition, and (ii) prognosis – predicting the development of the disorder to allow timely intervention. The diagnostic ability of the data interpretation was tested with ROC-analysis using the area under the curve (AUC) as a standard performance measure in clinical practice. In addition, we conducted pilot studies for both disorders to ascertain the technical feasibility of the systems and to obtain early feedback from end users on usability. The general setup of the pilot studies was as follows: patients were provided with a smart care-assistant (smartphone, app and sensors) that was used in a home setting for a period of two to four weeks. Every day at a predefined time, the assistant gave an alert to the patient to take measurements and fill-in a questionnaire. At the end of the

236   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

evaluation period every user filled in a feedback form and took part in an interview with the case manager of the pilot study. The feedback concerned various usability issues such as ease of use, user-friendliness, technical operation and overall usefulness of the system.

14.3.5.1 Home monitoring of COPD patients COPD is a progressive lung disease, currently affecting some 64 million people worldwide. It is one of the major chronic diseases in terms of both morbidity and mortality. The main cause of COPD is exposure to tobacco smoke. COPD is currently not curable, but treatment does reduce the burden considerably. Exacerbations – episodes of acute deterioration of the patient’s condition, usually caused by an airway infection – have a profound impact on patient well-being and on health-care costs. Patients with frequent exacerbations usually have faster disease progression, which makes exacerbation prevention a particularly relevant goal. Additionally, a faster treatment response to exacerbations appears to lead to better recovery. The state of the respiratory system is observable via symptoms including dyspnea, productive cough and decreased activity due to breathlessness and a number of physiological signs such as the forced expiratory volume in one second (FEV1) and blood oxygen saturation. Our COPD care-assistant consists of the AERIAL app and two Bluetooth-enabled sensors – a pulse-oximeter and a micro-spirometer. We studied different types of Bayesian network models – constructed by hand in close cooperation with lung physicians and by using machine-learning techniques – for a timely prediction of exacerbations. The model input consisted of self-reported symptoms using a questionnaire on the phone and the measurements from the sensors. We performed preliminary evaluation studies with both retrospective data and pilot scale prospective data.[16] The model based on an expert opinion yielded an AUC of 0.97 for detecting exacerbation events. This indicates that the model can detect exacerbations as they are happening, which is a useful baseline and required for further model development. In order to predict exacerbations, we constructed temporal models, based on the time series data resulting from monitoring patients with our system. It turns out that it is possible to predict at least part of the exacerbations a day in advance with the best model resulting in an AUC of 0.82.[20] The results from the pilot studies with respect to the usability goal revealed that the patients’ impression of the system after using it is fairly positive. This early feedback from actual COPD patients is important for these kinds of systems because acceptance is often a concern. The next step involves assigning an advice to the predictions. Relevant decisions for COPD care include rescheduling monitoring because of increased risk, dosage adjustment of bronchodilator drugs and contacting healthcare professionals. A deci-

14 Supporting active patient self-care   

   237

sion analysis will be carried out with family doctors as well as lung specialists to establish what decisions the system should make given a particular clinical situation.

14.3.5.2 Home monitoring of high-risk pregnant women Preeclampsia is a pregnancy-related syndrome associated with a high blood pressure (>  140/90  mm  Hg) and a leakage of protein into urine (so-called proteinuria). It is the most important cause of death among pregnant women and a leading cause of fetal complications such as low birth weight. As a pregnancy-related condition, the only way to cure preeclampsia is to deliver the baby. To allow a timely detection of preeclampsia in the current clinical practice, frequent outpatient visits are required, leading to high pressure on the pregnant woman as well as on the healthcare centers. In addition, pregnant women are relative young with an interest in applying modern technology to the management of their condition. Despite these facts, smart solutions for supporting self-monitoring during pregnancy are lacking. The eMomCare smart care-assistant we developed is the first of its kind. The care-assistant consists of the eMomCare app, a Bluetooth-enabled blood pressure meter and automated analyzer of urine reagent strips to measure proteinto-creatinine ratio.[12] Given the dynamic nature of a pregnancy, we manually developed a temporal Bayesian network model for preeclampsia in cooperation with gynecologists. The model, embedded in the eMomCare app, includes 13 risk factors and signs such as blood pressure measured at 10 time points (routine check-ups) during the pregnancy to estimate the current risk for preeclampsia and give a prognosis of the risk until the end of the pregnancy. The model has been evaluated with retrospective pregnancy data and showed high accuracy at the week of the diagnosis – AUCs ranged between 0.80 and 0.99 for the weeks 28–38 of the pregnancy. In addition, in 60% of the preeclamptic women, the model was capable of making the prognosis for the syndrome at least 4 weeks before the diagnosis was actually made (with 10% false positive rate). For the remaining patients, this early prediction was not possible mostly due to inaccurate or missing measurements in the data. The pilot study we conducted with the eMomCare system included 6  pregnant women from Maastricht University Medical Center. The overall feedback obtained was positive in terms of relatively easy system operation (smartphone, app and sensors). Some technical problems and design issues encountered during the pilot have been used to improve the system. The participants (completely) agreed that the eMomCare system has added a value to the usual care, and they are willing to use the system very often, showing the potential of the system for use in practice.

238   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

14.4 Discussion Given the framework for smart care-assistants and the first results from applying careassistants in practice, what remains is to examine their overall clinical impact and to make a comparison with existing solutions. Our experience with pilot studies shows that it is challenging to obtain reliable and objective outcomes in a home environment. A careful design, organization and control throughout the study are critical for the successful evaluation of mHealth systems. Training programs (e-Learning) for patients and caregivers are also expected to alleviate certain problems concerning, for example, the technical operation of the system. Such programs can facilitate the mind-shift towards sharing responsibilities between the patient and the caregiver during the care process. Acceptance of smart care-assistants depends on the extent of integration within existing healthcare structures. It is important to consider at an early stage of development how the system will be used: patient initiative (complete self-management) or prescribed and monitored by a physician (shared-care). In the latter case, integration of the smart care-assistant and healthcare information systems is essential, allowing for communication between the patient and caregiver in the context of the patientcare process. Establishing the public’s trust in smart care-assistants for personalized healthcare management will be key for ensuring the public’s acceptance of this new medical technology. Current concerns relate to the privacy and security of collecting and transmitting patient data. These are valid concerns that should be addressed, for example, by using electronic signatures, authentication and multi-application smart card solutions; access to healthcare data should be under control of the patient in negotiation with caregivers. Furthermore, a smart care-assistant can, in principle, function without access to remote data, and so whether or not the patient’s data are transmitted can be controlled by the patient. Given the current capabilities of the smart care-assistants, we expect that supporting people’s decisions about their health will stimulate a more pro-active attitude toward health concerns, thus increasing patient involvement in the care process and reducing the burden on the healthcare system. This can help detect health deterioration at early stages, thus allowing timely treatment interventions and preventing hospitalizations. As a result, costs and work pressure within healthcare centers are to be reduced while maintaining or even improving health outcomes. Furthermore, smart care-assistants can play an educational role. Through better information provision to patients, the system can assist in teaching them to cope with their diseases. Compared to existing systems, improvements consist of the use of clinically sensible data interpretation models and the improved availability of support. Many existing systems do not offer autonomous operation, relying on interpretation by care providers, or are purely technology-driven and do not provide clinically validated advice. This even appears to be the case in recent systems, e.g. [21] on using HIT for diabetes

14 Supporting active patient self-care   

   239

management where mobile technology is only used as an additional communication channel to facilitate self-management support. Furthermore, personalization appears to be an important aspect of data interpretation that is gaining attention in the literature. For instance in [22], where a personalized system for lifestyle advice is described.

14.5 Conclusion In conclusion, the rapid penetration of mobile devices in our daily lives and the deployment of mHealth systems is expected to provide greater access to health care to larger segments of the population (see expected number of mHealth users in 2017 in Figure 2). This will be beneficial for the healthcare system as a whole and for its ability to provide qualitative, safe and cost-effective care through improved early risk assessment, with appropriate referral and consultation among providers of basic and specialty care.

Acknowledgement This work has been partially supported by the Dutch organizations STW, ZonMw and STITPRO.

References [1]

Force ET. e-Health: Redesigning health in Europe for 2020 [Internet]; 2012. [cited 10 December 2014] Available from: http://www.epractice.eu/en/library/5362646 [2] Blackburn M, Pitts J, Walden G, Bilbray B, Burgess M, Gingrey P. US Congres Letter to FDA and FCC on mHealth; 2012. http://blackburn.house.gov/uploadedfiles/letter_from_congress_to_ fda_and_fcc_-_3apr2012.pdf [3] Research2guidance. Mobile Health Market Report 2013–2017: The commercialization of mHealth applications [Internet]; 2013;(3). [cited 10 Dec 2014]. Available from: http:// mhealtheconomics.com/mhealth-developer-economics-report [4] Istepanian RSH, Jovanov E, Zhang YT. M-Health: Beyond seamless mobility for global wireless healthcare connectivity. IEEE Trans Information Technology in Biomedicine 2004;8(4):405–412. [5] Buchanan BG, Shortliffe EH. Rule-based Expert Systems: The MYCIN Experiments of the Stanford Heuristic Programming Project. Reading: Addison-Wesley; 1984. [6] Miller RA, Jr HEP, Myers JD. INTERNIST-1: An Experimental Computer-Based Diagnostic Consultant for General Internal Medicine. New England Journal of Medicine 1982;307(8):468–476. [7] Kumar S, Nilsen W, Pavel M, Srivastava M. Mobile health: revolutionizing healthcare through transdisciplinary research. IEEE Computer 2013;46(1):28–35.

240   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[8] Rubin MA, Bateman K, Donnelly S, Stoddard GJ, Stevenson K, Gardner RM, et al. Use of a Personal Digital Assistant for Managing Antibiotic Prescribing for Outpatient Respiratory Tract Infections in Rural Communities. J Am Med Inform Assoc 2006;13:627–634. [9] Lee NJ, Bakken S. Development of a prototype personal digital assistant-decision support system for the management of adult obesity. Int J Med Inform 2007;76:S281–S292. [10] McLean S, Nurmatov U, Liu J, Pagliari C, Car J, Sheikh A. Telehealthcare for chronic obstructive pulmonary disease. Cochrane Database of Systematic Reviews 2011;(7). [11] Mudanyali O, Dimitrov S, Sikora U, Padmanabhan S, Navruz I, Ozcan A. Integrated rapiddiagnostic-test reader platform on a cellphone. Lab on Chip 2012;12(15):2678–2686. [12] Velikova M, Lucas PJF, Smeets R, van Scheltinga JT. Fully-automated interpretation of biochemical tests for decision support by smartphones; 25th IEEE International Symposium on Computer-Based Medical Systems (CMBS 2012), Rome, Italy, 20–22 June, 2012, pp. 320–325. [13] Chen M, Gonzalez S, Vasilakos A, Cao H, Leung V. Body Area Networks: a survey. Mobile Netw Appl 2011;16:171–193. [14] Clifford G, Clifton D. Wireless technology in disease management and medicine. Annual Review of Medicine 2012;63:479–492. [15] Pearl J. Probabilistic reasoning in intelligent systems: networks of plausible inference. Morgan Kaufmann; 1988. [16] van der Heijden M, Lucas PJF, Lijnse B, Heijdra Y, Schermer T. An autonomous mobile system for the management of COPD. Journal of Biomedical Informatics 2013;46:458–469. [17] Velikova M, van Scheltinga JT, Lucas PJF, Spaanderman M. Exploiting causal functional relationships in Bayesian network modelling for personalised healthcare. International Journal of Approximate Reasoning 2014;55:59–73. [18] Shachter R. Evaluating Influence Diagrams. Operations Research 1986;34(6):871–882. [19] Torrance G, Feeny D. Utilities and quality-adjusted life years. International Journal of Technology Assessment in Health Care 1989;5:559–575. [20] van der Heijden M, Velikova M, Lucas PJF. Learning Bayesian networks for clinical time series analysis. Journal of Biomedical Informatics 2014; 48:94–105. [21] Nundy S, Lu C, Hogan P, Mishra A, Peek M. Using Patient-Generated Health Data From Mobile Technologies for Diabetes Self-Management Support. Journal of Diabetes Science and Technology 2014;8(1):74–82. [22] Hu B, Naseer A, Fukuda K. A personalised lifestyle advisory system. In: Healthcom 2012: IEEE 14th International Conference on e-Health Networking, Applications and Services; 2012.

Eyal Zimlichman

15 Using patient-reported outcomes to drive patientcentered care Abstract: With health care becoming more patient-centered and outcome and valuedriven, providers and payers need to be able to measure, report and improve outcomes that are meaningful to patients. These outcomes can only be provided by the patients, and thus systems are needed to be able to capture patient-reported outcomes (PROs) and allow both patients and providers to use the data both on an individual patient level as well as on a population level. Information technology is already playing a central role in collecting, applying analytics to and reporting the PRO data. This includes collection of patient-generated data onsite through tablets/kiosks and offsite through patient portals and internet, smartphone applications and telephones through interactive voice-response applications. Reporting of analyzed data through electronic medical records will allow clinicians to discuss the information with patients. For such a system to work, several challenges need to be overcome, such as attaining high response rates for data reporting from patients, integrating PROs into the provider-patient encounters, allowing for real-time data collection and reporting, enabling data reporting by proxy (family members/caretakers) and addressing methodological issues such as case-mix adjustments. Although PROs are new to the clinical landscape, some successful case studies exist and need to be studied as more local attempts are made and policy starts to emerge. In this chapter I will describe three case studies of successful PRO implementation and will attempt to emphasize lessons learned from each, emphasizing the important role that health information technology plays.

15.1 Introduction The term “patient-reported outcomes” is relatively new in the healthcare services world. While used for many years now as part of clinical research, it needed the convergence of new priorities to bring it front and center, and hopefully to every patient’s doorstep in the upcoming years. The healthcare crisis and recent policy reform environment has focused attention on the following new priorities: maximizing healthcare value (more quality for less cost); becoming more patient- and outcome-centered; and finally focusing on the continuum of care – considering the care for a patient across providers. Providers have started redesigning health care to what the reformed market demands: care that is valuable to the patient at a cost that is appropriate. Yet they were soon confronted by the current healthcare system’s inability to accurately and efficiently assess whether healthcare services are valuable to the patient. Or in other words, how can one measure outcomes that matter to patients?

242   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

Within this new reform environment, scholars and opinion leaders, such as Prof. Michael Porter from the Harvard Business School, have preached the importance of patients taking on the role of consumers in the healthcare market so that competition focuses on the real consumers instead of other stakeholders (providers, payers and industry).[1, 2] According to Porter, this would shift the healthcare market to respond as a healthy consumer market, thus driving up value for patients while driving costs down. Yet for this to happen, patients would need to be able to compare meaningful outcomes between services and providers. In the last two decades, public reporting of quality metrics has set out to do just that, yet has not brought about the anticipated change. Online consumer-directed report cards aiming to encourage consumers to use quality data to help inform their healthcare-related decisions have become common. The hope was that this would subsequently lead to quality improvement among providers. Yet, research has shown consistently that consumer-directed quality reports have been difficult to understand, presenting data that patients often find irrelevant to their needs; therefore, these reports have had a minimal impact on patient choices for providers.[3, 4] All this has brought about a growing interest in the development of new outcome measures that would be more meaningful for patients and allow for a better assessment of patient-centered healthcare value across the continuum of care. Within patient-centered outcome measures, patient-reported outcomes will need to play a central role in any care redesign process and have been the focus of interest from policy makers, healthcare researchers, providers, payers and patient advocacy groups. Patient-reported outcomes (PROs) are defined as “any report of the status of a patient’s (or person’s) health condition, health behavior or experience with health care that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else”.[5] The use of PROs has been common for some time in clinical care trials and specifically in comparative and cost-effectiveness studies, as numerous tools, both generic and disease-specific, have been designed for this purpose throughout the last three decades. PRO measures (or PROMs) describe the use of PRO aggregated data as performance measures that can potentially assess clinician or organizational quality of care. For PROs to transform the value of clinical care, it is not enough to only collect patient-provided data. There is a need to also feed the patient-specific data forward to clinicians so that it can impact health care provided, feed the data back to patients and providers on an individual and aggregate level, and report performance measures on a population level.[6] Health information technology plays and will play a major role in enabling real-time flow of patient and population level data to all stakeholders, thus allowing PROs to drive healthcare value. This emerging field has just begun taking its first implementation steps in the British National Health Service (NHS) PROMs program;[7] additionally, the National Quality Forum (NQF) in the United States has begun developing PROMs for both clinical practice and performance measurement.[8] In this chapter I will elaborate on

15 Using patient-reported outcomes to drive patient-centered care   

   243

current attempts and best practices in implementing PROMs specifically for patient engagement, consider current gaps and challenges, and propose solutions based on technology applications.

15.2 The current situation: needs, gaps and challenges A successful clinical service PROMs program would allow providers to redesign services so that outcomes that matter to patients can be optimized. While there have been a few case studies of programs that have successfully achieved this (as described later in this chapter), these success stories are still rare. Previous qualitative research we performed has surfaced three main challenges and barriers for a successful implementation that relate to the patient’s perspective:[9] 1. For PROMs to represent a patient cohort for a specific physician, clinic, hospital or population, response rate needs to be in the 60–80% range so that managers and clinicians can leverage the information to improve healthcare delivery; 2. Clinicians need to be able to act proactively, in real-time, on trends of PROMs data so that care of the individual patient is improved; 3. Patients need to be engaged through PROMs as an aid in shared decision-making and ultimately be able to choose providers based on performance. One of the first challenges faced when designing a PROMs program is collecting the data from the patients, allowing for the needed response rate as mentioned above. For this, technology will need to play a central role. Most data collected from patients today are patient experience/satisfaction data traditionally collected through paper questionnaires. U.S. hospitals conduct the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) mostly through vendors and report results to Centers for Medicare and Medicaid Services. However, these 21 patient perspectives on care and patient rating items (32 total questions including demographic items) are predominantly conducted through regular mail, post-hospitalization, with response rates that traditionally have been in the 30–40% range at most, raising doubts about potential response and selection bias in the results. Similarly, data collected from patients participating in clinical trials, whether quality of life, disease activity, views and experience of care received, or other patient-reported outcomes, is also traditionally collected through pen and paper either on site or through mail. In an attempt to streamline data collected and improve interface with patients, recent years have seen tablet computers take a central role in data collection for both research and clinical applications. When using tablets to register PRO tools for research purposes, the main challenge has been establishing validation that would allow technology to replace traditional paper questionnaires and tools.[10] For

244   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

clinical practice there were different challenges: initially, patient acceptance of the technology and, later, information capture and integration into existing interfaces and database.[10, 11] Nonetheless, technology allows us to advance beyond on-site patient-generated data collection. The internet, as well as mobile technology, has provided health care with the tools to interact with and collect data from patients remotely. Patients are now able to access their health-related information through provider and payer-sponsored patient portals or patient health records (PHR). With the rapidly growing internet use rate, a growing number of early adopting centers are reporting patient portal enrollment as high as 50–60% of patient population. Yet even these centers are reporting lower rates for lower income, elderly and lower literacy level patients.[12] Current trends show a gradual continuous increase in internet use even for these populations.[13] With time, patient portal penetration is sure to increase, bringing about the ability for patients to not only view their health records but also to provide data to their clinicians in a secure fashion. Patient portals are considered the preferred way to allow patients to communicate remotely with their healthcare provider – enabling both data capture and access to patient record data. Yet, other modes for data capture are emerging (Table 1). Some make use of newer, rapidly growing platforms such as smartphones. With the expanding smartphone market,[14] these devices are considered the ultimate platform for engaging consumers in most non-healthcare markets and are now making their first steps into the e-health arena. Using smartphones to collect patient-generated data (while introducing new behavior/social related biases to data analysis which will need to be better understood) will soon become the dominant mode for collecting data from patients. With time, personal mobile health will also expand to include not just actively reporting PROMs but also passive patient-generated data originating from usage of mobile phones and mobile web applications.[15] As the central interface for wearable devices and health trackers collecting physiologic and behavioral data from the patient, the smartphone will allow for a whole new realm of personalized medicine. Aggregating this data with more traditional PROMs (and other clinical data residing within electronic medical records (EMRs)/PHRs) will enable the patient and his provider to better understand his health status; introducing predictive analytics will enable the recommendation of the most appropriate individually tailored therapy. As promising as smartphones are to opening new opportunities for patient engagement, there is still the need to engage patients who are computer or technology-illiterate. These are mostly elderly populations or patients with very low socioeconomic status. Unfortunately, chronic and multiple morbidity patients tend to be from these demographic groups. As we design our clinical PROMs program, as mentioned before, we will need to include a large part of our index disease population, and so would need to make use of low-tech modalities such as telephones or mail-in surveys to reach these populations. While calling patients and completing questionnaires on the phone provides an easy solution to reach all patients, this raises two concerns: one being the introduction of an acquiescence bias;[16] and the other,

15 Using patient-reported outcomes to drive patient-centered care   

   245

Table 1: Methods for collection of patient generated data. Application

Strength

Weakness

Patient portal (web based)

– Patient fills out from home – Only 15% adoption rate – Patient can fill out when – Relevant only to tech-literate convenient patients with a personal – Simple data integration computer – Promotes adoption of patient portals thus improves patient engagement

Interactive voice response (IVR) – Widely used in other industries, rapidly growing – Can have patient call-back option – Cost sensitive, easy to scale up for large populations – Simple data integration

– May be off-putting to patients (esp. elderly) – Limitations to voice recognition capabilities

Human operator phone calls

– Widely used – Human response

– Acquiescence bias – More expensive than IVR – Cannot feed data back to provider in real-time

Mobile phone apps

– Patient can fill out from anywhere – Simple data integration

– Limited to tech-literate patients

Kiosks/tablets

– Patient fills out while waiting – Collects data at single point for appointment of care – Simple data integration – Potential for provider to intro– Can be leveraged for other duce a selection bias purposes

Paper

– Tried-and-tested method

– Want to move into electronic data gathering – Cannot control (or keep track of) the point of time at which patient answers – Traditionally turns up low compliance rate – Cannot feed data back to provider in real-time

perhaps more serious, is the challenge and costs associated to scale up to large populations. A solution that needs to be considered, which would answer the challenge of reaching all patients while being more cost-sensitive, is the use of Interactive Voice Response (IVR). This relatively new technology, based on the interaction between

246   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

telephony and computers, has the potential to play a major role in patient interaction and engagement in the near future. As these applications need to be particularly relevant to older adults, research has already been done to assess attitudes towards this technology specifically for the older populations[17] and has shown how to improve the experience by using algorithms specific to sub-populations and even adapted to each individual.[18] Furthermore, reliability and validity of PROMs tools were maintained when using IVR compared to other modes of administration.[19] With voice recognition technology improving at a rapid pace, these solutions will become more friendly and approachable specifically for those populations from which we currently struggle to collect patient-generated data. In order to engage all patients within a specific patient population, another challenge we need to face is the ability to work with proxy submitted data. This is extremely relevant to populations at both ends of the age groups – both pediatrics and elderly but also for patient populations with extreme disabilities (e.g. stroke, dementia, multiple sclerosis). To solve this, one needs to study the effect of proxy reporting and the associated bias involved. For pediatric populations, especially when dealing with smaller children, there has been more experience in building PRO tools that are intended for parents or guardians.[20] While for other conditions where some patients use family members/guardians to fill out the PRO tools, statistical adjustments need to be made so that data is comparable.[21] When considering the need to adjust for data collection mode (mode adjustment) as well as for proxy involvement, this demonstrates just some of the complexities that are needed to be addressed so that the PRO data is valid.

15.3 Proposed solutions The functionalities of PROMs need to reach beyond patients to also include providers and healthcare managers to allow for the value for patients to be generated. The following cases illustrate, both from a technical and a methodological point of view, how PROMs can play a central role in engaging patients and improving outcomes that matter to patients. While all three are successful implementations, I will also try to highlight what challenges still need to be worked out, based on the three challenges discussed before.

15.3.1 Dartmouth-Hitchcock Spine Center: a pioneer in PROMs The Spine Center at Dartmouth-Hitchcock Medical Center (DHMC) in Lebanon, NH was launched in 1998 with the aim to provide a one-stop comprehensive care facility for patients with back and neck problems. As per design of the center, providers were

15 Using patient-reported outcomes to drive patient-centered care   

   247

to measure the health status of patients in real-time as structured data with the aim to provide individual patient-tailored care, activate the patient for shared decision-making, provide longitudinal health data to monitor treatment impact on the individual patient over time and finally, to use aggregated sub-population data for comparative effectiveness purposes and public reporting. Patients are approached prior to their visit at the center through a secured patient portal and asked to complete a 30-minute health survey. The clinicians have chosen to collect outcome data relevant and meaningful to patients with back pain such as functional and mental status, disability status and pain levels. Patients that do not complete the survey prior to the visit are asked to do this in the waiting room using a tablet before seeing the physician. The DHMC experience with tablets was one of the earliest attempts with this technology and has paved the way for much wider adoption with the proliferation of tablets. Collected data is analyzed in real time and displayed as a patient summary report for both the patient, through the secured portal, and the clinician to base his treatment plan on real-time provided data (this is the “feedforward”, as coined by the DHMC researchers[22]). With data collected from patients on each return visit, follow-up longitudinal data is gathered that displays changes in health outcomes. These changes can be associated with a specific treatment plan on an individual level and serve as a basis on which the clinician and patient discuss changes in symptoms as well as future treatment plans. Furthermore, as data is collected, it is aggregated for specific conditions and treatment modalities, thus providing reports to the public that offer accurate information on health outcomes. This data can be used by patients, families, consumers and referring clinicians as a decision aid when determining a preferable treatment plan (such as making a decision between surgical treatment or non-surgical for spinal disease). As reported by Hvitfeldt et al. who studied this implementation, over 80% of patients treated at the Spine Center rated the system as “excellent to good” and about a third indicated that the PRO system had led to positive changes in their health compared to other clinics.[22] In the same study, clinicians reported that the PRO system was extremely important for both follow-up and feedback, and that, in their view, most patients found the system useful and positive. However, both patients and clinicians expressed concerns relating to the length and complexity of the survey (especially for more elderly or low literacy patients). The DHMC Spine Clinic remains one of the early and most known examples of a PROMs program. Its main success is in its ability to engage both patients and clinicians and bring the PROMs data center stage to the discussions between them. In that regard, the program has managed to answer to the three challenges that determine a successful PROMs program (as outlined before) and has served as a model for other systems that followed (one of which is the Partners Healthcare system described later in this chapter). Yet, some of the concerns this program had to deal with were with patient interfaces, complexity of the survey tools and, later on, integration into electronic clinical systems.

248   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

The Spine Center basically only offered the patients a digital method to input patient-reported data (through internet or on site using a tablet). The fact that the patients were engaged around a clinic visit allowed the clinicians to approach the patient face-to-face and ensure the data was collected, even from computer illiterate patients who held a tablet for the first time in their lives. Yet, encounter-based PROMs carry disadvantages that time-based PROMs are free of. This includes the ability to compare between patients and cohorts around specific time points (such as six months post-surgery). Obviously, to ensure reports are valid and represent the patient’s symptoms and conditions, they have to be simple, aim at a lower level literacy crowd and be short enough so it will not deter patients from completing. From interviews with patients, the usual upper limit patients are likely to spend completing a survey is around 10–12 min, which on average sums up to around 20–25  multiple choice questions.[9] The Spine Center survey was considerably longer than that and also included repetitions arising from the integration of different tools into one survey. As was noted by patients in the study by Hvitfeldt et al.,[22] this certainly affected patient response rates and contributed to bias related to social and education factors. Lastly, the PROMs system at the DHMC Spine Center was designed as a stand-alone information technology system not integrated into other clinical information systems. When eventually the medical center decided to implement Epic as a system-wide EMR, the original program had to be abandoned as all clinical systems needed to operate within the new EMR. Similar challenges were faced by other providers who developed a stand-alone PROMs electronic system, while those that originally integrated with the EMR and the patient portals enjoyed the benefits of improved integration and usability.

15.3.2 Memorial Sloan-Kettering Cancer Center Urology Clinic Another successful PROMs program was developed at Memorial Sloan Kettering Cancer Center (MSKCC) (New York, NY) where a system for electronic symptom tracking and reporting (STAR) was developed with the aim of initially collecting symptoms of outpatient chemotherapy patients.[23] The STAR system provides a web interface for patients to complete online questionnaires either from home or from a clinic based computer/tablet. In 2009, the system was adapted for use by the Urology Clinic to record data on functional recovery after radical prostatectomy. Considered a mainstay of treatment for prostate cancer, yet associated with potential long-term complications in the form of urinary incontinence and erectile dysfunction, radical prostatectomy is a procedure deemed classic for collecting PROMs. The large variability in the incidence of these complications suggests that surgeon experience and technique might be a major factor in improving outcomes. In attempting to better assess post-operative morbidities allowing for performance improvement on one side and promoting shared decision-making through

15 Using patient-reported outcomes to drive patient-centered care   

   249

better patient awareness of possible complications on the other hand, Vickers et al. have developed a PROMs instrument assessing incontinence, erectile dysfunction and health-related quality of life. In an effort to allow for a maximum response rate, the developers have managed to keep the tool to 15  questions long while retaining validity.[24] Furthermore, they have also used branching logic (or computer adaptive testing) to allow patients to skip non-relevant questions based on prior answers. Patients who provide an e-mail address are sent an e-mail inviting them to complete the questionnaire at 3, 6, 9, 12, 18, 24, 36 and 48 months post-op. Unlike the program at DHMC’s Spine Center, the follow up here is time-oriented rather than visit-oriented. This was done with the intention to be better able to group patient cohorts longitudinally. This approach enables a creation of an outcomes registry that can support process improvement initiatives and comparative effectiveness research as well as providing access to outcomes that matter to patients for them to view on patient portals. Indeed, the Urology Clinic at MSKCC carries out all three features. The main weakness of the MSKCC Urology Clinic PROMs program was again rooted in the fact that it was only available to patients who are using the internet. Specifically, only 50% of patients provided an e-mail address and were thus eligible to participate.[24] Yet, through successful patient engagement and a “feed forward” and feedback approach, as well as keeping the PROMs tool short, response rate from those patients approached 80%. Still, the PROMs data probably does not fully include such socio-demographic populations as the very elderly, lower socioeconomic classes and the illiterate. It is also unclear what the patient’s views regarding this program were and how much added value they believe it provided.

15.3.3 Partners Healthcare System PROMs Program The most recent of these three PROMs program cases described here was developed at Partners Healthcare System (Boston, MA). The program at Partners was built based on learning from past experience and PROMs success stories, such as the other two described here. Since 2010, Partners was engaged in a system-wide strategic care redesign initiative aimed at redesigning health care according to what the healthcarereformed market demands: care that is valuable to the patient at a cost that is appropriate.[25] While seeking to drive care to be patient-centered and to assess outcomes that matter to patients, it was evident to its leaders that Partners would need to establish a PROMs program that eventually would be able to scale up system-wide, crossing through conditions and the continuum of care. The Partners PROMs program was designed conceptually to be time-oriented across all conditions, where time points were chosen to fit the nature of the condition. For example, while surgical procedures, such as cardiac surgery, had a preoperative assessment to register baseline status and then re-assessment at one, three, six and 12 months post-operative, for chronic conditions, such as diabetes, patients

250   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

were assessed once every six months continuously. The PROMs questionnaires were developed so that they are meaningful and relevant for the three main stakeholders: a) patients – questions portrayed what patients were actually concerned about with their medical conditions, and focus groups of patients with the target condition were put together and all questions were discussed for relevancy and coherency; b) clinicians – so that clinicians deem the reports meaningful for clinical decision making; and c) population health managers – to allow comparison between patient populations for population health needs, the tools included both a disease-specific component as well as a generic tool for measurement of functional status and health-related quality of life. We have learned that while clinicians were mostly interested in tracking disease-specific symptoms such as chest pain for ischemic heart disease, headaches for migraine and shortness of breath for asthma, it was also important for the primary care provider as well as for tracking population health parameters to follow both physical and mental status. Questionnaires were kept as short as possible, while still retaining their validity and functionality. The aim was to keep tools to be 20–25 questions at most; this translated to an average time of 8–10 min to fill out. While baseline outcomes and symptoms are collected at the clinic using a tablet interface, the main challenge in designing the Partners Healthcare PROMs program was to maximize the population covered and response rates. This demanded a data collection method that could reach all patients, regardless of internet access, socioeconomic status or literacy levels. It was clear that this would demand a multi-modal data collection approach that would also have to include a telephone option to reach those patients that are not using internet or might be illiterate. To achieve that, patients are offered a few options to provide their answers. Preferably, patients are signed up to the patient portal (Partners Patient Gateway) and are then sent periodical e-mails directing them to the portal where the tools are administered. This is featured as the preferred method as it allows the patients to view their results, as well as how they compare to patients like them across time. Yet, patients who do not have e-mail, access to a computer interface or are computer illiterate are offered the option to report by phone. This can be done either by phone operators who call the patients at the predefined time intervals or through IVR (described before); this is considered more probable and manageable to scale up across large patient populations. In the works, a smartphone application would be a third method through which patients can report outcomes (Figure 1). Throughout all time points, should the patient visit the clinic within a set time window, the data is collected using the tablet and no attempt to reach out to the patient is made until the next time period. Another challenge was working the administration of PROMs into the clinic workflow so as to create the least burden on the staff and refrain from creating more waiting times for patients. For this to happen, a careful assessment of current workflows at the clinic as well as involving the clinic administrators and staff together with the PROMs implementation specialists who have experience implementing at multiple sites, was found to be critical for a successful implementation.

15 Using patient-reported outcomes to drive patient-centered care   

   251

Figure 1: Design of a PROMs program at Partners Healthcare demonstrating the multi-modal approach to capture data from patients.

The critical element, as we have seen before, would be the successful interaction with both patients and clinicians so that PROMs are being used in a meaningful way to enhance value as seen by the patient. At Partners, in an attempt to create a common language for both patients and providers, a patient report card (or PROMs “snapshot”) is created based on the survey answers (current and historic). This report card summarizes the different domains collected (e.g. physical function, mental status, anxiety, shortness of breath, etc.), providing a trend as well as a comparison to similar patients for each time. Furthermore, actionable “next steps to better health” are provided which translate the scores and changes across time into suggestions (such as advising to contact the provider, a social worker, or encouraging to “keep up the good work”). These report cards have gone through rigorous development, including condition-specific patient focus groups, patient interviews and clinician interviews. Conceptually, it was decided that the same report cards would be accessible to both clinicians and patients in real time (updated as soon as the patient completes the PROMs tools). While patients get to view these through the patient portal, providers have access to the same report cards through their EMR. This was designed to facilitate patient-clinician discussions around a common report and language. While the Partners Healthcare PROMs program is still in its infancy after launching its pilot in 2012 and later scaling up and upgrading to a more complete, wide-

252   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

spread program in 2014, it is probably considered one of the more comprehensive attempts to integrate PROMs strategically into routine care. Yet, when implementing across multiple conditions and sites, as opposed to the other two examples where the clinicians are more of a captured audience, the main challenge is engaging the clinicians. It still remains to be seen how well this is going at the Partners program. Initial results, as might be expected, are showing variability among the different sites, conditions and clinicians. While some “got it” from the start, and have started using PROMs data routinely in their conversations with patients and decision-making, others have been slower to do so. Yet experience is already starting to gather that would build a conceptual framework on how to better engage clinicians.

15.4 Discussion We have seen here how three different institutions tackle the challenges of building a successful PROMs program: two on a clinic level treating a condition and one on a larger scale across conditions. For a PROMs program to achieve the objective of improving healthcare value, three stakeholders need to be engaged in a meaningful way: patients, providers (clinicians) and quality and population health managers. From the patients’ perspective, the PRO tools first need to collect valuable information that is meaningful for patients. Usually patients would like to track specific symptoms that trouble them, whether joint pain for rheumatoid arthritis or incontinence following prostate surgery. Furthermore, tools need to be tailored for specific sub-populations and even on an individual patient level. This can be achieved using computer-adaptive testing when administering the questionnaires on digital media. Feedback through patient-specific report cards needs to happen as close to when the questionnaire was administered as possible, and needs to be clear and comprehensible to patients, even from lower literacy level populations. Some questions for which patients will want to have answers for include: how am I doing today compared to how I was before? How am I doing compared to other patients like me? How do patients that go through a specific treatment modality fare (as a way to enhance shared decision-making)? Which provider/clinic/hospital should I go to for my treatment? Lastly, a multi-modal data-capture approach needs to be implemented so that all patients in a clinic panel can provide their PROs – regardless of age and technology literacy. Providers need to play a central role in any PROMs program. Most importantly, they need to use PROMs on a day-to-day basis when treating individual patients and bring this into the conversation with the patient. Currently, as PROMs are new to the clinical environment, we are seeing providers asking questions such as, “What do I do with this? How does this data effect my clinical decisions?”. With more experience, we will be seeing multidisciplinary protocols and clinical care pathways that involve

15 Using patient-reported outcomes to drive patient-centered care   

   253

PROMs. Since PROMs assess not just specific symptoms but rather the full spectrum that affects health-related quality of life, they would be actionable to the multidisciplinary clinical team. In that effect, Patient Centered Medical Homes would be the classic setting to embrace PROMs as part of the clinical process.[26] Furthermore, for clinicians to be more engaged, PROMs report cards need to be meaningful to them. Reports need to be designed in a fashion that would provide data both on the individual patient level and on a cohort level, sharing data from patients with a similar condition. In the future, with more evidence surfacing, decision support tools based on PROMs need to be integrated into the electronic record. Finally, as seen in the MSKCC urology case, within a clinic/hospital level, comparison of outcomes across providers eventually needs to happen so that learning and innovation can be promoted. This gives rise to identification of best practices through comparative effectiveness and performance management. Once this occurs, this is where most of the value improvement will happen. The third stakeholder group, one that we often tend not to focus on, is the quality and population health managers. For PROMs to potentially improve value, they need to provide a view of outcomes achieved across institutions on all levels. The creation of measures based on PROMs will need to happen in the next few years as use of PROMs is expanded. A road map for the development of measures based on PROMs has been recently made public by the NQF.[8] One example of a PRO metric could be rate of patients who went into a Coronary Artery Bypass Graft (CABG) surgery with chest pain who are chest pain-free at six months post-operation. Benchmarking across institutions using such measures would allow learning of best practices, promoting excellence, clinical integration and driving value that centers on patients’ needs. Once measures that use PROs are well established, the next phase would be to publicly report those, thus promoting providers’ choices that are based on outcomes that patients care about – the desirable outcomes of treatments and not just the adverse ones. Furthermore, public and private insurance, aiming to improve outcomes that matter to patients, would need to include PROMs in pay for performance programs, thus driving provider organizations to focus on promoting these measures. An example might be reduced payments for patients who went through joint replacement but with no substantial improvement in function following the procedure. Another issue that will need to be addressed prior to meaningfully using PROMs for benchmarking is case-mix adjustment that, of course, is critical to allow for accurately drawing comparisons on all levels. Although not entirely tackled yet, there have been several attempts to do this.[27, 28] It would make sense that in the near future federal funding would be provided for the development of case-mix methods that would allow measuring PROMs. Perhaps as part of the Patient-Centered Outcomes Research Institute (PCORI), a federal funded program that was established with a mission in mind to promote informed decision-making by patients based on patient-centered outcome data.[29]

254   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

15.5 Conclusion With health care finally becoming more patient-centered, and as patients take on the true role of consumers in the healthcare market, care outcomes that matter to patients will need to play a central role. These outcomes, mostly unused today, would need to be collected directly from the patient in a fashion that would allow the data to be aggregated and used both on a personal patient level and on the population level. As outlined in this chapter, designing a successful PROMs program carries many challenges that need to be confronted. Basically, both patients and providers need to be engaged and need to see the value PROMs play in promoting patient-sensitive value. Most importantly, patients need to report the data, and response rates need to approach 70% of the target population. For this to happen, technology must be appropriately implemented while keeping costs controlled. Furthermore, clinicians need to actively use PROMs in interactions with patients and in their decision-making process so that real value can be generated with the provider system geared toward improving outcomes that matter to patients. Designing report cards to serve both patients and clinicians in real time will allow both to use PROMs for shared decision-making. Finally, looking at cohorts of patients and at populations, and following proper casemix adjustment will allow for identifying high performance organizations and best practices, and for designing clinical care pathways that promote outcome and efficiency. PROMs are situated on the convergence of some of the critical elements that are believed to improve U.S. and international healthcare systems: patient-centered care, outcome-based registries, comparative effectiveness and care across the continuum and population health, while potentially impacting pay for performance, public reporting and use of health information technology. It will not be surprising to find PROMs playing a critical role in all of these central elements in the near future. Yet, the road is still long and bumpy. While successful small-scale implementations have occurred, larger-scale ones are still few and it is too early to assess how successful they are. Yet these early successes can teach us much about what works and what doesn’t, and can help in creating a conceptual framework and a road map towards the development of larger-scale programs.

References [1] [2] [3]

Porter ME. A strategy for health care reform – toward a value-based system. N Engl J Med 2009;361(2):109–12. Porter ME. What is value in health care? N Engl J Med 2010;363(26):2477–81. Fung CH, Lim YW, Mattke S, Damberg C, Shekelle P. Systematic review: the evidence that publishing patient care performance data improves quality of care. Ann Intern Med 2008;148(2):111–23.

15 Using patient-reported outcomes to drive patient-centered care   

   255

[4] Hibbard JH, Greene J, Daniel D. What is quality anyway? Performance reports that clearly communicate to consumers the meaning of quality of care. Med Care Res Rev 2010;67(3):275–93. [5] U.S. Food and Drug Administration, Guidance for Industry, Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims, Fed Regist 2009;74(35):65132–133. [Cited October 2012]. Available from: http://www.fda.gov/downloads/ Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM193282.pdf [6] Hvitfeldt H, Carli C, Nelson EC, Mortenson DM, Ruppert BA, Lindblad S. Feed Forward Systems for Patient Participation and Provider Support; Adoption Results from the Original US Context to Sweden and Beyond. Quality and Safety in Health Care, 18(4); 247–256, 2010. [7] NHS England. Patient Reported Outcomes Measures 2014. [Cited July 2 2014] Available from: http://www.england.nhs.uk/statistics/statistical-work-areas/proms [8] National Quality Forum. Patient Reported Outcomes in Performance Measurement. Washington, DC: National Quality Forum. Dec 2012. [Cited July 2 2014] Available from: http://www. qualityforum.org/Publications/2012/12/Patient-Reported_Outcomes_in_Performance_ Measurement.aspx [9] Zimlichman E. Integrating Patient Reported Outcomes across the continuum of care. International Society of Quality in Health Care Annual Meeting 2012. Geneva, Switzerland. [Cited July 2 2014] Available from: http://www.isqua.org/docs/geneva-presentations/a5-eyalzimlichman.pdf?sfvrsn=2 [10] Abernethy AP, Herndon JE 2nd, Wheeler JL, Patwardhan M, Shaw H, Lyerly HK, Weinfurt K. Improving health care efficiency and quality using tablet personal computers to collect research-quality, patient-reported data. Health Serv Res 2008;43(6):1975–91. [11] Gurland B, Alves-Ferreira PC, Sobol T, Kiran RP. Using technology to improve data capture and integration of patient-reported outcomes into clinical care: pilot results in a busy colorectal unit. Dis Colon Rectum 2010;53(8):1168–75. [12] Yamin CK, Emani S, Williams DH, Lipsitz SR, Karson AS, Wald JS, Bates DW. The digital divide in adoption and use of a personal health record. Arch Intern Med 2011;171(6):568–74. [13] Greysen SR, Chin Garcia C, Sudore RL, Cenzer IS, Covinsky KE. Functional Impairment and Internet Use Among Older Adults: Implications for Meaningful Use of Patient Portals. JAMA Intern Med 2014 May 16. [14] Statista. Smartphone penetration in the United States from 2010 to 2017. [Cited July 2 2014] Available from: http://www.statista.com/statistics/201183/forecast-of-smartphonepenetration-in-the-us [15] Madan A., Cebrian M., Lazer D. and Pentland A., "Social Sensing to Model Epidemiological Behavior Change", Proceedings of ACM Ubicomp 2010, Copenhagen. [16] Hall MF. Patient satisfaction or acquiescence? Comparing mail and telephone survey results. J Health Care Mark. 1995 Spring;15(1):54–61. [17] Miller DI, Bruce H, Gagnon M, Talbot V, Messier C. Improving older adults’ experience with interactive voice response systems. Telemed J E Health 2011;17(6):452–5. [18] Miller DI, Aubé F, Talbot V, Gagnon M, Messier C. Older people’s attitudes toward interactive voice response systems. Telemed J E Health 2014;20(2):152–6. [19] Bjorner JB, Rose M, Gandek B, Stone AA, Junghaenel DU, Ware JE Jr. Difference in method of administration did not significantly impact item response: an IRT-based analysis from the Patient-Reported Outcomes Measurement Information System (PROMIS) initiative. Qual Life Res 2014;23(1):217–27. [20] Irwin DE, Gross HE, Stucky BD, Thissen D, DeWitt EM, Lai JS, Amtmann D, Khastou L, Varni JW, DeWalt DA. Development of six PROMIS pediatrics proxy-report item banks. Health Qual Life Outcomes 2012;10:22.

256   

   Part II: Current and Future Information Technology Solutions for Patient Empowerment

[21] Sonder JM, Holman R, Knol DL, Bosma LV, Polman CH, Uitdehaag BM. Analyzing differences between patient and proxy on Patient Reported Outcomes in multiple sclerosis. J Neurol Sci 2013;334(1–2):143–7. [22] Hvitfeldt H, Carli C, Nelson EC, Mortenson DM, Ruppert BA, Lindblad S. Feed forward systems for patient participation and provider support: adoption results from the original US context to Sweden and beyond. Qual Manag Health Care 2009;18(4):247–56. [23] Basch E, Artz D, Dulko D, Scher K, Sabbatini P, Hensley M, Mitra N, Speakman J, McCabe M, Schrag D. Patient online self-reporting of toxicity symptoms during chemotherapy. J Clin Oncol 2005;23(15):3552–3561. [24] Vickers AJ, Savage CJ, Shouery M, Eastham JA, Scardino PT, Basch EM. Validation study of a web-based assessment of functional recovery after radical prostatectomy. Health Qual Life Outcomes 2010;8:82. [25] Lee TH. Care redesign – a path forward for providers. N Engl J Med 2012;367(5):466–72. [26] Glasgow RE, Kaplan RM, Ockene JK, Fisher EB, Emmons KM. Patient-reported measures of psychosocial issues and health behavior should be added to electronic health records. Health Aff (Millwood) 2012;31(3):497–504. [27] United Kingdom, Department of Health. The case-mix adjustment for the national Patient Reported Outcome Measures (PROMs) programme. April 10, 2012. [Cited July 2, 2014] Available from: https://www.gov.uk/government/news/case-mix-adjustment-methodology-publishedfor-patient-reported-outcome-measures [28] Nuttall D, Parkin D, Devlin N. Inter-provider comparison of patient-reported outcomes: developing an adjustment to account for differences in patient case mix. Health Econ 2013 Sep 30. [29] Washington AE, Lipstein SH. The Patient-Centered Outcomes Research Institute – promoting better information, decisions, and health. N Engl J Med 2011;365(15):e31.

Index Adult Congenital Heart Association (ACHA) X, XIX Affordable Care Act (ACA) 10, 44, 50, 75, 76, 82, 85, 127–129, 133–135 Agency for Healthcare Research and Quality (AHRQ) XIX, 11, 85 Alert 27, 31, 68, 130, 160, 161, 172, 215, 218, 231, 234, 235 Algorithm 144, 170, 171, 246 American Medical Informatics Association (AMIA) IX, XI, XIX Artificial intelligence IX, 230 Bayesian network 231–233, 235–237 Behavioral IX, 6, 59, 62, 63, 132, 138, 181, 190, 201–203, 244 Blog 65, 66, 68 Care coordination 8, 29, 81, 82, 98, 206 Center of Medicare and Medicaid Services (CMS) 11, 41, 43, 47, 52, 76, 81, 204 Clinical encounter 3, 5–9, 18 Consumer 6, 12, 14, 23, 39, 40–42, 44–53, 60, 61, 63–65, 69, 70, 77–80, 86, 87, 94, 96, 100, 104–106, 114, 116, 117, 125–128, 130, 135–140, 158, 166, 167, 170, 173–175, 190–192, 208, 242, 244, 247, 254 Consumerism 6, 10, 59, 70, 75, 77, 78, 87 Decision support 30, 31, 40, 46, 47, 69, 99, 105, 130, 161, 201, 227, 229, 230, 253 Doctor-patient relationship 7, 59, 68, 127, 175, 185, 204, 208, 221 E-health 7, 12, 14, 15, 47, 104, 228, 244 Electronic Health Record (EHR) 14, 49, 53, 76, 77, 80, 82, 84, 86, 93, 94, 96, 105, 106, 114, 131, 132, 153, 155, 170, 173, 186 Electronic informed consent (eConsent) IX, XIV, 179, 180, 186, 192, 193 Electronic Medical Record (EMR) 26–33, 101, 126, 130, 136, 141, 145, 148, 160, 161, 173, 192, 204, 205, 241, 244, 248, 251 Ethics 81, 154, 179, 180, 182, 184, 185, 192, 193

Facebook 67 Fee-for-service VII, 100, 127–129, 135 Google 53, 62, 168, 170 Health 2.0 65 Health apps IX, 27, 59, 67–70, 84, 85, 87, 137, 214, 231, 235–237, 245 Healthcare VII, VIII, IX–XI, XIII, XIV, XVI, XIX, 3–5, 7–18, 23–26, 28, 33, 36, 37, 39–42, 44, 45, 47–54, 59–68, 70, 75–79, 81–84, 87, 93–95, 97, 99, 100–105, 107, 113–115, 125–127, 129–150, 153–161, 165–167, 170, 173–175, 186, 189–191, 199, 200, 204, 206, 208, 210, 216–220, 227–231, 234–239, 241–244, 246, 247, 249–252, 254 Healthcare market VII, 39, 40, 44, 50–52, 75, 242, 254 Healthcare provider VII, VIII, XIII, XIV, 5, 8, 10, 11–14, 16, 23–26, 48, 94, 103, 136, 141, 142, 149, 155, 190, 191, 200, 206, 210, 219, 242–244 Healthcare quality 8, 10–12, 14, 16, 78, 199 Health Information Exchange (HIE) 52, 104, 141, 142, 144, 145, 147–149 Health Information Technology for Economic and Clinical Health Act (HITECH) 79, 87, 94, 141, 204 Health Information Technology (HIT) X, XIII, XIV, XIX, 3–5, 7, 12–18, 24, 29, 50, 60, 61, 67, 70, 75–77, 79–84, 86, 87, 94, 100, 104, 141, 183, 200, 238, 241, 242, 254 Health Insurance Portability and Accountability Act (HIPAA) 29, 50, 104, 116, 136 Health-Level 7 (HL7) 155 Health literacy XIV, 15, 77, 80, 93, 95, 99, 102, 103, 113, 114, 165, 166, 175 Health policy XI, 75, 78, 81, 83, 84, 107 Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) 10, 48, 243 Image sharing 141, 143–150 Imaging 28, 29, 51, 52, 141–146, 148, 150

258   

   Index

Information retrieval XIV, 4 Information storage 4, 24, 143, 144, 175, 192, 200 Information Technology (IT) IX–XI, XIII, XIV, XIX, 3–6, 12–14, 18, 24, 25, 29, 30, 33, 36, 39, 40, 44, 49, 50, 51–54, 60, 75–77, 79, 84, 85, 94, 97, 100, 104, 139, 141, 143, 183, 192, 200, 201, 204, 215, 221, 241, 242, 248, 254 Informed consent process 179, 181, 182, 184–186, 188, 192 Institute for Healthcare Improvement (IHI) 11 Institute for Patient- and Family-Centered Care 11 Institute of Medicine (IOM) VII, XI, 8, 9, 11–13, 52, 66 Insurance plan 39, 40, 42–44, 48–50, 53, 83 Integration 42, 94, 107, 116, 129, 136, 137, 144, 148, 170, 204, 206, 238, 244, 245, 247, 248, 253 Interactivity 59, 60, 62, 67, 69, 70, 169 Internet 7, 12, 16–18, 42, 60, 69, 70, 79, 95, 98, 99, 125, 126, 130, 132–135, 145, 153–158, 160, 161, 165–167, 170, 171, 175, 179, 188–190, 222, 241, 244, 248–250 Joint Commission on Accreditation of Healthcare Organizations (JCAHO) 11 Meaningful use 83, 85–87, 94, 96–98, 100, 104, 141, 204 Medical information 66, 69, 116, 142, 155, 165–167, 169–171, 173–175, 182, 186, 190 Mobile XIII, XIV, 12, 16, 60, 67, 75, 84, 87, 100, 125, 130, 131, 137, 139, 141, 166, 169, 174, 175, 199–203, 205–209, 213–219, 227–231, 235, 239, 244, 245 Mobile Health (mHealth) 12, 16, 75, 84, 87, 228–230, 233, 238, 239, 244 Mobile technology 175, 203, 209, 230, 239, 244 National Committee for Quality Assurance (NCQA) 11, 81, 206 National Health Services (NHS) 10, 11, 242 Natural Language Processing (NLP) XIV, 156–161, 171 New media 59–62, 64, 65, 67–70

Office of the National Coordinator (ONC) 14, 50, 76, 84, 85, 97, 104, 183, 204 Old media 59, 61, 65, 66, 69 Ontology 117, 159, 168, 170, 188 Outcome measures 78, 135, 242 Patient activation 9–11, 18, 62, 67, 75, 77, 210 Patient and Family Advisory Council (PFAC) 11 Patient-Centered Care (PCC) X, XIV, XIX, 3, 7–12, 14–18, 23, 81, 241, 254 Patient-Centered Outcomes Research Institute (PCORI) XIX, 49, 253 Patient control 68, 141, 145, 147, 149, 150, 186 Patient education IX, 29, 49, 106, 131, 206 Patient empowerment XIII, XIV, XIX, 3, 7, 9, 12, 18, 60, 94, 95, 101, 106, 113, 149, 153–156, 160, 199, 227 Patient engagement X, XIII, XIV, XIX, 3, 5, 7–18, 52, 53, 60–62, 75–77, 79, 82, 83, 86, 87, 94, 95, 97, 99, 101, 104, 106, 135, 150, 155, 199–212, 214–216, 218, 219, 222, 223, 243–246, 249 Patient partnership 7, 9, 10, 13, 62, 68, 76, 95, 200, 211, 218, Patient portals XIV, 12, 14, 79–81, 86, 87, 93–102, 107, 113, 116, 130, 135, 147, 148, 158, 241, 244, 245, 247–251 Patient privacy 141, 143–146, 149, 150 Patient-provider communication 13, 94, 96, 103 Patient Reported Outcomes (PROs) VI, 52, 75, 81, 82, 85–87, 189, 241–243, 246, 247, 252, 253 Patient self-care 41, 60, 61, 66, 227 PatientsLikeMe 49, 66, 69, 113, 120, 189 Personal Health Records (PHR) XIII, 12, 16, 75, 80, 86, 87, 94–96, 104, 116, 144–147, 150, 206, 244 Personalization 45, 51, 59, 67, 227, 231, 233–235, 239 Person-centered care 93, 105 Planetree 11 Query-expansion 167, 168 Self-management IX, 10, 13, 16, 77, 82, 96, 98, 105, 200–202, 210, 211, 227, 231, 234, 235, 238, 239 Sensor 67, 229–231, 235–237

Index   

   259

Shared Decision Making (SDM) 5, 10, 49, 52, 60, 64, 75, 85, 87, 243, 247, 248, 252, 254 Social media 3, 12–14, 16–18, 62, 70, 79, 189

Transparency 24, 40, 64, 65, 69, 77–79, 85, 125, 139, 166, 168 Twitter 14, 64

Tablet IX, 16, 29, 100, 131, 137, 179, 187, 188, 192, 200, 206, 217, 229, 230, 241, 243, 245, 247, 248, 250 Telehealth 15, 64, 125, 131–140, 228

World Health Organization (WHO) XI, 11, 154