Handbook of Attenuated Psychosis Syndrome Across Cultures: International Perspectives on Early Identification and Intervention [1st ed. 2019] 978-3-030-17335-7, 978-3-030-17336-4

Table of contents :
Front Matter ....Pages i-xxii
Front Matter ....Pages 1-1
Attenuated Psychosis Syndromes Seen Through the Cultural Prism: Relevance, Terminology, and Book Structure (Daniel I. Shapiro, Huijun Li, Larry J. Seidman)....Pages 3-6
Assessment of Risk for Psychosis (Daniel I. Shapiro, Huijun Li, Emily R. Kline, Margaret A. Niznikiewicz)....Pages 7-40
Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence, and Future Directions (Daniel I. Shapiro, Kristen A. Woodberry, Huijun Li, Larry J. Seidman)....Pages 41-63
Front Matter ....Pages 65-65
Attenuated Psychosis Syndromes Among Australian Youth and Young Adults: Early Identification and Intervention (Barnaby Nelson, Patrick D. McGorry)....Pages 67-84
Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk Syndromes (SIPS) and the Scale of Psychosis-Risk Symptoms (SOPS) (Scott W. Woods, Barbara C. Walsh, Albert R. Powers III, Thomas H. McGlashan)....Pages 85-113
Clinical High Risk for Psychosis Syndromes Among Swiss and German Youth and Young Adults: Early Identification and Intervention (Frauke Schultze-Lutter, Nina Schnyder, Chantal Michel, Stefanie J. Schmidt)....Pages 115-142
Front Matter ....Pages 143-143
Early Experiences of Psychotic Illness From a Cross-Cultural Perspective: An Anthropological View From Research in Indonesia (Byron J. Good, Carla R. Marchira, M. A. Subandi, Sandeep Nanwani, Mary-Jo Del Vecchio Good)....Pages 145-159
Medical Causes of Psychosis: Lessons for Individuals with Attenuated Psychosis Syndromes (Ashley N. Matskevich, Matcheri S. Keshavan)....Pages 161-183
Front Matter ....Pages 185-185
Identification and Treatment of Youth with Attenuated Psychosis Syndromes: A Canadian Perspective (Jean Addington, Georgia Carstensen, Danijela Piskulic, Thomas Raedler, Donald Addington)....Pages 187-197
Cultural Considerations in the Treatment of African American Youth with Attenuated Psychosis Syndromes: The Importance of Socio-contextual and Clinical Factors (Derek M. Novacek, Allison M. LoPilato, Katrina B. Goines, Hanan D. Trotman, Michael T. Compton, Elaine F. Walker)....Pages 199-218
Attenuated Psychosis Syndromes Among Asian American Youth and Young Adults: A Culturally Relevant Case Illustration Approach (Huijun Li, Michelle Friedman-Yakoobian, Victoria Choate Hasler, Daniel I. Shapiro, Emily Wu)....Pages 219-236
Attenuated Psychosis Syndromes Among Latino-American Youth and Young Adults in the United States: Early Identification and Intervention (Tracy Alderman, Isabel Domingues)....Pages 237-256
Attenuated Psychosis Syndromes Among Mexican Youth and Young Adults: A Culturally Relevant Case Illustration Approach (Francisco Reyes-Madrigal, Pablo León-Ortiz, Camilo de la Fuente-Sandoval)....Pages 257-277
Attenuated Psychosis Syndromes Among Brazilian Youth and Young Adults: Early Identification and Intervention (Graccielle R. Cunha, Elson M. Asevedo, Elisa Brietzke, Rodrigo A. Bressan, Ary Gadelha)....Pages 279-288
Attenuated Psychosis Syndromes Among Nigerian Youth and Young Adults: Early Identification and Intervention (Adeniran Okewole, Sewanu Awhangansi, Akinloye Akinfala)....Pages 289-300
Attenuated Psychosis Syndromes Among Youth and Young Adults in China: Early Identification and Intervention (Tianhong Zhang, Daniel I. Shapiro, Jijun Wang)....Pages 301-310
Attenuated Psychosis Syndromes Among Japanese Youth and Young Adults: Early Identification and Intervention (Masafumi Mizuno, Naomi Inoue)....Pages 311-322
Attenuated Psychosis in Youth and Adolescents: Clinical and Cultural Considerations from India (Avinash De Sousa, Amresh Shrivastava)....Pages 323-332
Attenuated Psychosis Syndromes Among Danish Youth and Young Adults: Early Identification and Intervention (Louise Birkedal Glenthøj, Nikolai Albert, Freja H. Christensen, Julie Nordgaard, Merete Nordentoft)....Pages 333-348
Early Identification and Interventions of Attenuated Psychosis Syndromes in Spain (Inmaculada Baeza, Clemente García-Rizo, Gisela Sugranyes)....Pages 349-366
Front Matter ....Pages 367-367
Integration of Literature Across Countries: Challenges, Opportunities, and Implications for Future Research (Barbara C. Walsh, Scott W. Woods, Albert R. Powers III)....Pages 369-378
Back Matter ....Pages 379-398

Citation preview

Huijun Li · Daniel I. Shapiro · Larry J. Seidman Editors

Handbook of Attenuated Psychosis Syndrome Across Cultures International Perspectives on Early Identification and Intervention

Handbook of Attenuated Psychosis Syndrome Across Cultures

Huijun Li  •  Daniel I. Shapiro Larry J. Seidman Editors

Handbook of Attenuated Psychosis Syndrome Across Cultures International Perspectives on Early Identification and Intervention

Editors Huijun Li Department of Psychology College of Social Sciences, Arts and Humanities Florida A&M University Tallahassee, FL, USA Larry J. Seidman Beth Israel Deaconess Medical Center Harvard Medical School Boston, MA, USA

Daniel I. Shapiro Department of Psychiatry Early Psychosis Programs University of California-Davis Sacramento, CA, USA Beth Israel Deaconess Medical Center Harvard Medical School Boston, MA, USA

ISBN 978-3-030-17335-7    ISBN 978-3-030-17336-4 (eBook) https://doi.org/10.1007/978-3-030-17336-4 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Dedication

Dr. Larry Seidman unexpectedly passed away in September 2017, when this book was roughly two thirds complete. This volume is dedicated in loving memory to Larry, as well as to his wife, children, and grandchildren, and to the extended family of people spread around the world who have been touched by his kindness, welcoming generativity, intellectual curiosity, and sage foresight. The book is a tribute to Larry’s tireless efforts to better understand and improve the lives of all those touched by psychosis, ADHD, and other conditions affecting neurocognition. But even more broadly, and fitting as a last book, it is a homage to a man who at his core valued people most and relished the opportunity to experience something new—some new ideological horizon or perspective to be pondered or hashed out with others, some shared food or event or sight enjoyable for its novelty and pleasure, or some new tradition or culture experienced together with somebody from somewhere else. He was a man who valued diversity of thought and experience and, among all the things he was good at, had an uncanny sense of how things fit into their larger contexts and a vision for how this might guide the future of his field. Dr. Seidman has been mentor, friend, or colleague to both coeditors and many of the authors who contributed to this book. True to his nature, he was proud of this project and of the collaborations across all conceivable geographical, political, and cultural lines that it represents. This was not as new an endeavor to Larry as it is to our field. In more ways than one, this book would not exist without his influence. And in more ways than one, it is a tribute to him. Daniel I. Shapiro and Huijun Li October 2018 v

Foreword

Prevention and early intervention in psychotic disorders was regarded for many decades as a pipe dream. Those who held that dream were accused by their senior colleagues of the cardinal sin of offering false hope and indulging in “rescue fantasies.” How much the world has changed over the past three decades! Schizophrenia and psychotic disorders, imbued with pessimism and the “soft bigotry of low expectations,” seemed to be the least promising arena for the development of a preventively oriented psychiatry, but a global wave of dynamic translational research has transformed our field. We have finally realized that every other branch of medicine values the power of hope, and to strip hope away from patients and families at the onset of any treatable illness, however potentially disabling, deprives clinicians of one of their most powerful therapeutic tools. However, hope is not enough. We have also learned from cancer and other major noncommunicable diseases that if we diagnose early, treat intensively, and guarantee a secure tenure of expert care for as long as needed, especially in the early years post diagnosis, then outcomes can be dramatically improved, even with existing treatments. The evidence base for psychosis confirms the same is true for these disorders, provided that once we have got people well, we endeavor to keep them well. This approach however is still in mental health care, more honored in the breach than the observance. Despite the progress we have made, only a minority of patients worldwide benefit from implementation of this knowledge. We can do so much more for our patients, and the early intervention paradigm powered by implementation science and advocacy is the key. The origins of what in North America is now increasingly referred to as the “Attenuated Psychosis Syndrome” goes back a long way. Kraepelin and Bleuler both described how dementia praecox and schizophrenia all too often developed imperceptibly and gradually with subtle changes and precursor signs and symptoms which eventually evolved into more florid and acute phases of frank psychosis. Harry Stack Sullivan in the 1920s described carefully the onset of schizophrenia and imagined the day when interventions might forestall the full expression of the illness. More recently, Ainsley Meares and Heinz Hafner described the prodromal stage of illness, and an operational definition of this prodromal stage was even included in the DSM III-R. Ironically, it was dropped from DSM IV because it was regarded as non-specific and unreliable. Yet, non-specificity is an essential feature of the concept of prodrome as adapted from infectious disease. Inspired by the vii

viii

knowledge that over 70% of first episode cases manifested a prodromal stage, our early psychosis research group, from 1991, initially Henry Jackson and I, and, then from 1993, Alison Yung, decided to study this stage of illness, initially retrospectively and then prospectively, using an operational definition that combined known risk factors, such as family history, functional decline, and attenuated or subthreshold symptoms as “warning signs” of fully fledged and sustained psychotic disorder. This later became known as “indicated prevention.” We established a satellite (PACE) clinic of the recently established EPPIC program in Melbourne and studied a prospective cohort of patients who manifested what we termed an “at-risk mental state.” We observed a 40% rate of progression to first episode psychosis within 1 year, despite providing needs-based clinical care to these patients. This led us to coin the term “ultrahigh risk” state to underline the huge elevation of risk that was present. In the USA, Barbara Cornblatt amended this term to “clinical high-risk” state to contrast the approach with the genetic high-risk paradigm that had been pursued to that point. The at-risk or ultrahigh risk approach was an example of the “close-in” research strategy that studied risk factors, including subthreshold symptoms, close to onset of disorder. It was able to enrich the sample for risk and collapse follow-up periods, offering huge advantages. In retrospect, these features, which seem simple and obvious, underpinned a real breakthrough in research strategy and methodology and paved the way for a global wave of research effort in many countries. Tom McGlashan and Barbara Cornblatt introduced these ideas and strategies into the USA, and others followed their lead. The leadership mantle was later taken up by Ty Cannon and, crucially, by Bob Heinssen at the National Institute of Mental Health, who brought the US leaders in this field, including Larry Seidman, together under the North American Prodrome Longitudinal Study (NAPLS) banner. This extension of early intervention added an extra edge to the momentum building around the first episode research and has produced new evidence confirming biological changes around the onset phase of illness and of the efficacy of intervention during the subthreshold stage. It is now possible to at least delay the onset of psychosis in some people who are starring down the barrel of risk. There is now Cochrane level 1 evidence for this, and I have seen countless examples of bullets which appeared to have been dodged (at least temporarily) in my clinical practice. Even if psychosis becomes sustained, duration of untreated psychosis is minimized, and outcomes can be improved. From follow-up studies, we now know so much more about the natural history of these atrisk mental states, which have not only a heightened valence for psychotic disorders but also a heightened risk for a range of other syndromes. In general, they connote risk for poorer functional outcomes, suicidal behavior, and a range of exit syndromes, often comorbidly. This paradigm shift has created controversy, and, while genuine concerns have been raised appropriately, the issue has been exploited by scaremongers of various hues to promote their particular causes. Fortunately, the evidence that has accumulated now speaks for itself. People who meet these

Foreword

Foreword

ix

UHR/APS criteria have an undeniable need for care, without which their prognosis is more guarded. Engagement in care at this stage of illness gives people the best chance of recovery. Treatment must be carefully staged, sequenced, and guided exquisitely by risk-benefit considerations. And now that we know, unequivocally, the importance and demonstrated benefits of such early intervention in those with an APS, the mantle can be taken up by clinicians and researchers around the world to investigate how this engagement, sequencing, and implementation of care are affected by differences in culture and variations in global systems of care. This comprehensive and scholarly edited volume captures the state of play of research in this dynamic field, not only in North America and Europe but, admirably, in other parts of the world too, and even other hemispheres! It is a tour de force, and the editors, with Larry as the generative, modest, and inspiring “coach,” would be so proud of the final product. The next phase of research and reform will likely proceed along three parallel trajectories. Firstly, researchers will seek to enrich samples and enhance prediction of transition not only to psychosis but to other outcomes, notably poor functioning, using sophisticated statistical approaches such as machine learning. Doing so will also require a study of diverse peoples to ensure that such approaches can be appropriately applied across differing places and peoples. Secondly, the clinical staging model reveals that the earliest stages of illness where a need for care exists are not linked purely to one of the traditional syndromal silos. Operational definitions of this stage may have a variable level of valence for different late syndromes like psychosis or mania, but there is overlap. We need to consider transdiagnostic definitions of early stages of illness. Biomarkers may be helpful in refining prediction and guiding treatment; however, it is unlikely they will validate the current DSM/ ICD nosology. The APS or UHR syndromes may come to be seen as a prototype concept on a pathway to a new nosology, one with greater utility and validity and one which may be moderated by culturally linked factors. Larry Seidman would have been 100% behind such an exciting venture. Lastly, understanding how differences in cultural context might affect all of this work represents a different kind of frontier, one that has been seriously underrepresented in the history of psychiatry. The paradigm shift toward early intervention in those with UHR/APS syndromes was catalyzed in primarily Australian, European, and North American centers with primarily Western populations. Much of the precipitous expansion from these hubs has been to translate, validate, and replicate approaches into new cultures and systems, each with differing pathways to care, valued outcomes, and ways of conceptualizing mental constructs, health, and healing. It may be that extant approaches can be modified, but it may also be that homologous interventions, built from unique cultural vantage points, need to be included in our models. This work has only just begun—the volume that follows is an exciting and significant step in this endeavor. It is a huge honor to have been asked to write the foreword for this book, an honor tinged with deep sadness, because the senior editor, Dr. Larry Seidman, a dear friend and one of my most admired and inspiring colleagues, is no

x

l­onger with us. Larry is not only one of the international pioneers of the paradigm shift to which I have already referred but truly one of the most generative and collegial of the array of international leaders in psychiatry research that I have ever known. I am extremely pleased to see this book come to f­ ruition as part of a lasting tribute to his contribution to early intervention in psychosis. Orygen, The National Centre of Excellence Patrick McGorry in Youth Mental Health Centre for Youth Mental Health, The University of Melbourne Patrick McGorry, Parkville, VIC, Australia 

Foreword

Acknowledgments

We thank all the patients and their families who passionately seek change for them and us by engaging in clinical treatment and participating in research. Their drive, curiosity, and willingness to channel and challenge difficult experiences are inspiring and have led to improvements in quality of life for countless families around the world. We appreciate our family members, friends, supportive institutions, and graduate students for their incredible support and understanding during the writing and publication process. Finally, we owe so much to wonderful mentors past and future, who make endeavors like this imaginable.

xi

Contents

Part I Introduction and Overview of Assessment and Intervention in Attenuated Psychosis Syndromes 1 Attenuated Psychosis Syndromes Seen Through the Cultural Prism: Relevance, Terminology, and Book Structure������������������������������������������������������������������������    3 Daniel I. Shapiro, Huijun Li, and Larry J. Seidman 2 Assessment of Risk for Psychosis��������������������������������������������������    7 Daniel I. Shapiro, Huijun Li, Emily R. Kline, and Margaret A. Niznikiewicz 3 Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence, and Future Directions��������������������������������������������������������������������   41 Daniel I. Shapiro, Kristen A. Woodberry, Huijun Li, and Larry J. Seidman Part II Conceptual and Measurement Foundations in Attenuated Psychosis Syndromes 4 Attenuated Psychosis Syndromes Among Australian Youth and Young Adults: Early Identification and Intervention����������������������������������������������������������������������������   67 Barnaby Nelson and Patrick D. McGorry 5 Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk Syndromes (SIPS) and the Scale of Psychosis-­Risk Symptoms (SOPS)��������������������������������������������������������������������������   85 Scott W. Woods, Barbara C. Walsh, Albert R. Powers III, and Thomas H. McGlashan 6 Clinical High Risk for Psychosis Syndromes Among Swiss and German Youth and Young Adults: Early Identification and Intervention����������������������������������������������������  115 Frauke Schultze-Lutter, Nina Schnyder, Chantal Michel, and Stefanie J. Schmidt

xiii

xiv

Part III Borderlands of Cultural and Medical Conceptualizations of Attenuated Psychosis Syndromes 7 Early Experiences of Psychotic Illness From a Cross-Cultural Perspective: An Anthropological View From Research in Indonesia������������������������������������������������������������������������������������  145 Byron J. Good, Carla R. Marchira, M. A. Subandi, Sandeep Nanwani, and Mary-Jo Del Vecchio Good 8 Medical Causes of Psychosis: Lessons for Individuals with Attenuated Psychosis Syndromes����������������������������������������  161 Ashley N. Matskevich and Matcheri S. Keshavan Part IV International Research and Clinical Practice on Attenuated Psychosis Syndromes 9 Identification and Treatment of Youth with Attenuated Psychosis Syndromes: A Canadian Perspective��������������������������  187 Jean Addington, Georgia Carstensen, Danijela Piskulic, Thomas Raedler, and Donald Addington 10 Cultural Considerations in the Treatment of African American Youth with Attenuated Psychosis Syndromes: The Importance of Socio-­contextual and Clinical Factors��������  199 Derek M. Novacek, Allison M. LoPilato, Katrina B. Goines, Hanan D. Trotman, Michael T. Compton, and Elaine F. Walker 11 Attenuated Psychosis Syndromes Among Asian American Youth and Young Adults: A Culturally Relevant Case Illustration Approach��������������������������������������������������������������������  219 Huijun Li, Michelle Friedman-Yakoobian, Victoria Choate Hasler, Daniel I. Shapiro, and Emily Wu 12 Attenuated Psychosis Syndromes Among Latino-American Youth and Young Adults in the United States: Early Identification and Intervention����������������������������������������������������  237 Tracy Alderman and Isabel Domingues 13 Attenuated Psychosis Syndromes Among Mexican Youth and Young Adults: A Culturally Relevant Case Illustration Approach����������������������������������������������������������������������������������������  257 Francisco Reyes-Madrigal, Pablo León-Ortiz, and Camilo de la Fuente-Sandoval 14 Attenuated Psychosis Syndromes Among Brazilian Youth and Young Adults: Early Identification and Intervention����������  279 Graccielle R. Cunha, Elson M. Asevedo, Elisa Brietzke, Rodrigo A. Bressan, and Ary Gadelha 15 Attenuated Psychosis Syndromes Among Nigerian Youth and Young Adults: Early Identification and Intervention����������������������������������������������������������������������������  289 Adeniran Okewole, Sewanu Awhangansi, and Akinloye Akinfala

Contents

Contents

xv

16 Attenuated Psychosis Syndromes Among Youth and Young Adults in China: Early Identification and Intervention����������������������������������������������������������������������������  301 Tianhong Zhang, Daniel I. Shapiro, and Jijun Wang 17 Attenuated Psychosis Syndromes Among Japanese Youth and Young Adults: Early Identification and Intervention����������������������������������������������������������������������������  311 Masafumi Mizuno and Naomi Inoue 18 Attenuated Psychosis in Youth and Adolescents: Clinical and Cultural Considerations from India����������������������  323 Avinash De Sousa and Amresh Shrivastava 19 Attenuated Psychosis Syndromes Among Danish Youth and Young Adults: Early Identification and Intervention����������  333 Louise Birkedal Glenthøj, Nikolai Albert, Freja H. Christensen, Julie Nordgaard, and Merete Nordentoft 20 Early Identification and Interventions of Attenuated Psychosis Syndromes in Spain������������������������������������������������������  349 Inmaculada Baeza, Clemente García-Rizo, and Gisela Sugranyes Part V Directions for Future Research and Clinical Practice with Attenuated Psychosis Syndromes 21 Integration of Literature Across Countries: Challenges, Opportunities, and Implications for Future Research ��������������  369 Barbara C. Walsh, Scott W. Woods, and Albert R. Powers III Epilogue��������������������������������������������������������������������������������������������������  379 Barbara A. Cornblatt Index��������������������������������������������������������������������������������������������������������  385

Contributors

Donald  Addington Department of Psychiatry, University of Calgary, Calgary, AB, Canada Jean Addington  Department of Psychiatry, Hotchkiss Brain Institute & the Mathison Centre for Mental Health Research & Education, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada Akinloye Akinfala  Neuropsychiatric Hospital, Aro Abeokuta, Ogun State, Nigeria Nikolai Albert  Department of Clinical Medicine, University of Copenhagen, København, Denmark Tracy  Alderman  Department of Psychiatry, University of California, San Diego, CA, USA Elson M. Asevedo  Universidade Federal de São Paulo, São Paulo, Brazil Sewanu  Awhangansi Neuropsychiatric Hospital, Aro Abeokuta, Ogun State, Nigeria Inmaculada  Baeza Department of Child and Adolescent Psychiatry and Psychology, Hospital Clínic i Universitari de Barcelona, IDIBAPS, CIBERSAM, Universitat de Barcelona, Barcelona, Spain Rodrigo A. Bressan  Universidade Federal de São Paulo, São Paulo, Brazil Elisa Brietzke  Universidade Federal de São Paulo, São Paulo, Brazil Georgia  Carstensen  The Mathison Centre for Mental Health Research & Education, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada Freja  H.  Christensen Department of Clinical Medicine, University of Copenhagen, København, Denmark Michael T. Compton  Columbia University, New York, NY, USA

xvii

xviii

Barbara  A.  Cornblatt Recognition and Prevention (RAP) Program, Department of Psychiatry, The Zucker Hillside Hospital, New York, NY, USA Psychiatry and Molecular Medicine, Hofstra North Shore-LIJ School of Medicine, Hempstead, NY, USA Graccielle R. Cunha  Universidade Federal de São Paulo, São Paulo, Brazil Avinash  De Sousa  Department of Psychiatry, Lokmanya Tilak Municipal Medical College, Mumbai, India Camilo de la Fuente-Sandoval  Laboratory of Experimental Psychiatry & Neuropsychiatry Department, Instituto Nacional de Neurología y Neurocirugía, Mexico City, Mexico Isabel Domingues  Department of Psychiatry, University of California, San Diego, CA, USA Michelle  Friedman-Yakoobian Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Ary Gadelha  Universidade Federal de São Paulo, São Paulo, Brazil Clemente García-Rizo  Department of Psychiatry and Psychology, Hospital Clínic i Universitari de Barcelona, IDIBAPS, CIBERSAM, Universitat de Barcelona, Barcelona, Spain Louise Birkedal Glenthøj  Department of Clinical Medicine, University of Copenhagen, København, Denmark Katrina B. Goines  Department of Psychology, Emory University, Atlanta, GA, USA Byron J. Good  Department of Global Health and Social Medicine, Harvard Medical School, Boston, MA, USA Mary-Jo  Del  Vecchio  Good Department of Global Health and Social Medicine, Harvard Medical School, Boston, MA, USA Victoria Choate Hasler  Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA Naomi  Inoue  Department of Neuropsychiatry, School of Medicine, Toho University, Tokyo, Japan Matcheri  S.  Keshavan Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Emily R. Kline  Beth Israel Deaconess Medical Center, Boston, MA, USA Harvard Medical School, Boston, MA, USA Pablo León-Ortiz  Laboratory of Experimental Psychiatry & Department of Education, Instituto Nacional de Neurología y Neurocirugía, Mexico City, Mexico Huijun Li  Department of Psychology, College of Social Sciences, Arts and Humanities, Florida A&M University, Tallahassee, FL, USA

Contributors

Contributors

xix

Allison  M.  LoPilato  Emory University School of Medicine, Atlanta, GA, USA Carla R. Marchira  Department of Psychiatry, Faculty of Medicine, Public Health and Nursing, Gadjah Mada University, Yogyakarta, Indonesia Ashley  N.  Matskevich Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Thomas H. McGlashan  PRIME Clinic for Attenuated Psychosis Syndromes, Yale University, New Haven, CT, USA Patrick D. McGorry  Orygen, The National Centre of Excellence in Youth Mental Health, Parkville, VIC, Australia Centre for Youth Mental Health, The University of Melbourne, Parkville, VIC, Australia Chantal  Michel University Hospital of Child and Adolescent Psychiatry and Psychotherapy, University of Bern, Bern, Switzerland Masafumi  Mizuno Department of Neuropsychiatry, School of Medicine, Toho University, Tokyo, Japan Sandeep Nanwani  Yayasan Kebaya, Yogyakarta, Indonesia Barnaby  Nelson Orygen, The National Centre of Excellence in Youth Mental Health, Parkville, VIC, Australia Centre for Youth Mental Health, The University of Melbourne, Parkville, VIC, Australia Margaret A. Niznikiewicz  Beth Israel Deaconess Medical Center, Boston, MA, USA Harvard Medical School, Boston, MA, USA Boston VA Healthcare System, Brockton, MA, USA Merete  Nordentoft Department of Clinical Medicine, University of Copenhagen, København, Denmark Julie  Nordgaard Department of Clinical Medicine, University of Copenhagen, København, Denmark Derek M. Novacek  Department of Psychology, Emory University, Atlanta, GA, USA Adeniran Okewole  Neuropsychiatric Hospital, Aro Abeokuta, Ogun State, Nigeria Danijela  Piskulic The Mathison Centre for Mental Health Research & Education, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada Albert R. Powers III  PRIME Clinic for Attenuated Psychosis Syndromes, Yale University, New Haven, CT, USA Thomas Raedler  Department of Psychiatry, Hotchkiss Brain Institute & the Mathison Centre for Mental Health Research & Education, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada

xx

Francisco  Reyes-Madrigal  Laboratory of Experimental Psychiatry, Instituto Nacional de Neurología y Neurocirugía, Mexico City, Mexico Stefanie J. Schmidt  Department of Clinical Psychology and Psychotherapy, University of Bern, Bern, Switzerland Nina  Schnyder School of Public Health, University of Queensland, Brisbane, Australia Frauke  Schultze-Lutter Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University Düsseldorf, Düsseldorf, Germany Larry  J.  Seidman (Deceased) Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Daniel  I.  Shapiro Department of Psychiatry, Early Psychosis Programs, University of California-Davis, Sacramento, CA, USA Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Amresh  Shrivastava Lawson Health Research Institute, London, ON, Canada M. A. Subandi  Faculty of Psychology, Gadjah Mada University, Yogyakarta, Indonesia Gisela  Sugranyes Department of Child and Adolescent Psychiatry and Psychology, Hospital Clínic i Universitari de Barcelona, IDIBAPS, CIBERSAM, Universitat de Barcelona, Barcelona, Spain Hanan D. Trotman  Department of Psychology, Mercer University, Macon, GA, USA Elaine  F.  Walker Department of Psychology, Emory University, Atlanta, GA, USA Barbara  C.  Walsh PRIME Clinic for Attenuated Psychosis Syndromes, Yale University, New Haven, CT, USA Jijun Wang  Shanghai Mental Health Center, Shanghai Key Laboratory of Psychotic Disorders, Shanghai Jiaotong University School of Medicine, Shanghai, People’s Republic of China Kristen  A.  Woodberry Maine Medical Center, Center for Psychiatric Research, Portland, ME, USA Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Scott W. Woods  PRIME Clinic for Attenuated Psychosis Syndromes, Yale University, New Haven, CT, USA Emily  Wu Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA Tianhong Zhang  Shanghai Mental Health Center, Shanghai Key Laboratory of Psychotic Disorders, Shanghai Jiaotong University School of Medicine, Shanghai, People’s Republic of China

Contributors

About the Editors

Huijun Li  received her Ph.D. in School Psychology from the University of Arizona in 2003. She is a Nationally Certified School Psychologist and Nationally Certified Youth Mental Health First Aid Trainer and an Associate Professor in the Department of Psychology, College of Social Sciences, Arts, and Humanities, Florida A&M University. She is also a Research Collaborator of Psychiatry and served as the Director of Multicultural Research of the Commonwealth Research Center, prior to joining Florida A & M University, at Beth Israel Deaconess Medical Center, Harvard Medical School. She has received federal and foundation grants to study psychosocial factors such as culture-specific beliefs about causes of mental illness, stigma, and barriers to services related to help-seeking behaviors among individuals from diverse backgrounds and actively contributes to local community services by providing presentations and workshops on youth mental health. In addition, she served as Expert Professional on youth mental health on the local ABC News after the Newtown, Connecticut, school shooting incident and is the author or coauthor of peer-reviewed journal articles, book, book chapters, translated books, and conference presentations. Daniel  I.  Shapiro  recently joined the University of California-Davis, Department of Psychiatry, where he serves as Director of Operations of the EDAPT/SacEDAPT Early Psychosis Programs. He is also a Co-Director of the Atlanta Center for Cognitive Therapy certification and training program. Prior to this move, Dr. Shapiro was on the faculty at Harvard Medical School, Department of Psychiatry, where the majority of the work on this volume was completed.  There he served as Project Director of Clinical High-Risk Research and a Project Scientist at the Commonwealth Research Center of the Beth Israel Deaconess Medical Center, as well as a Clinical Psychologist and Supervisor at the Massachusetts Mental Health Center. He is an expert in the identification and treatment of early stages of psychotic illness and has directed the operations of both community-based clinical programs and federally- and privately-funded clinical research grants aimed at better understanding the developmental trajectory, treatment of, and barriers to care in early stages of psychosis and other serious mental illness. Within this work, he is particularly interested in: (1) the role stress, neurocognition, and individual factors play in the development and recovery from mental illness, and (2) development and dissemination of targeted and specialized interventions for individuals and families affected by psychotic illness.  Dr. Shapiro is a xxi

xxii

practicing Clinical Psychologist with a secondary specialization in the practice and teaching of Cognitive Behavioral Therapy, which began during his fellowship at the University of Pennsylvania. He is active as a clinician and clinical supervisor and is passionate about the training of developing clinicians. Larry  J.  Seidman  was Professor of Psychology in the Department of Psychiatry at Harvard Medical School, at the Beth Israel Deaconess Medical Center (BIDMC), and at Massachusetts General Hospital, where he conducted neuroimaging research since 1992. He was Director of the ­ Massachusetts Department of Mental Health sponsored “Center of Excellence in Clinical Neuroscience and Psychopharmacological Research” at BIDMC beginning in 2002 and Vice Chair for Research at BIDMC Public Psychiatry Division at Massachusetts Mental Health Center beginning in 2005. He spent more than 30 years studying the causes of psychotic disorders and mapping the components of neurodevelopmental disorders of prefrontal cortex and executive control in schizophrenia and ADHD.  He focused primarily on cognition in schizophrenia and ADHD and studies of youth “at risk” for psychosis. He was a Licensed Clinical Psychologist who had long worked with teenagers. He published more than 380 peer-reviewed papers and was Principal Investigator of 31 grants, participating in 80-funded grants since 1978. His focus over the past 10 years was investigating the phase of clinical high risk for psychotic illnesses and treatment of psychosis in the early phases. He was involved in teaching and mentoring and mentored more than 50 individuals with faculty appointments around the world in addition to scores of clinicians and many others. In recognition of these efforts, he was awarded the prestigious William Silen Lifetime Achievement Excellence in Mentoring Award in 2016. He also served as Director of Neuropsychological Training and Services at Massachusetts Mental Health Center and President of The Massachusetts Neuropsychological Society. He was recognized in August 2014 by Thomson Reuters Science Watch as one of the “The World’s Most Influential Scientific Minds of 2014” based on his highly cited papers.

About the Editors

Part I Introduction and Overview of Assessment and Intervention in Attenuated Psychosis Syndromes

1

Attenuated Psychosis Syndromes Seen Through the Cultural Prism: Relevance, Terminology, and Book Structure Daniel I. Shapiro, Huijun Li, and Larry J. Seidman

1

Introduction

Psychotic disorders, including schizophrenia, appear to affect a significant proportion of the world’s population. No culture seems to be immune, even though presentation and interpretation of the causes of the illnesses and intervention strategies may vary. The latest meta-analysis reports the pooled median global prevalence of psychotic disorders at 4.6 per 1000 persons; the median point and 12-month prevalence at 3.89 and 4.03 per 1000 persons, respectively; and the median lifetime prevalence at 7.49 per 1000 persons (Moreno-Kustner, Martin, & Pastor, 2018). Conditions involving threshold symptoms of psychosis vary in their severity and range from single D. I. Shapiro Department of Psychiatry, Early Psychosis Programs, University of California-Davis, Sacramento, CA, USA Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA e-mail: [email protected] H. Li (*) Department of Psychology, College of Social Sciences, Arts and Humanities, Florida A&M University, Tallahassee, FL, USA e-mail: [email protected] L. J. Seidman (Deceased) Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA

episode, or easily manageable symptoms, to more chronic and debilitating disorders. However, in all cases psychosis has the potential to disrupt the lives of those who experience it—to disrupt social, cognitive, vocational, and psychological functioning; to impact caregivers, family members, and other supports; and to affect society via need for clinical resources and lost productivity (Vigo, Thornicroft, & Atun, 2016). These disorders typically reach diagnosable threshold in adolescence or young adulthood but are often preceded by nonspecific premorbid cognitive, social, motor, and academic/vocational functioning difficulties that frequently date from early childhood (Tandon, Nasrallah, & Keshavan, 2010). More proximal to the onset of threshold clinical symptoms, most individuals who develop a psychotic disorder experience a prodromal period of worsening symptoms and increasing impact on functioning that can last from a few months to a few years (McGorry & Singh, 1995). This means not only that illness-related mechanisms may begin to exert their effects long before they can currently be identified but that they do so at critical periods of development. They can cause disruptions at times during which people typically build foundational knowledge and critical thinking skills, build social skills and use them to begin navigating relationships with others, start to develop interest in a trade and transition into independent life, learn how to manage emotion and cope with distress, and build important psychological

© Springer Nature Switzerland AG 2019 H. Li et al. (eds.), Handbook of Attenuated Psychosis Syndrome Across Cultures, https://doi.org/10.1007/978-3-030-17336-4_1

3

D. I. Shapiro et al.

4

f­oundations for one’s sense of self and how one fits into the world. Much of the foundation for all aspects of adult functioning is laid during the very times that psychosis typically begins to unfold. Disrup­tions during such critical periods can have long-lasting effects, suggesting that efforts to identify individuals before psychosis develops may provide opportunities to intervene and avoid or mitigate these effects (Woodberry, Shapiro, Bryant, & Seidman, 2016), potentially prevent or delay psychosis, or reduce the morbidity associated with developing a full-­blown psychotic episode. With this aim, the last few decades have seen a groundswell of efforts to develop and refine reliable methods for prospectively identifying those “at elevated risk” for later manifestation of a psychotic disorder. A major focus has been on identifying those who might be in the prodromal phase of illness, a key period for intervention, and methods for identifying possible prodromal syndromes have been developed that have reasonable predictive power (Keith & Matthews, 1991; Loebel et  al., 1992; Yung & McGorry, 1996). This has opened up exciting possibilities for early intervention and to better understand the predictors and the mechanisms of psychosis onset. Indeed, the promise represented by the notion of early intervention has led to a quickening in the pace with which this movement is spreading from the primarily Australian, Western European, and North American contexts in which they were developed into new international sociocultural contexts. New efforts to identify and intervene in those who are “at risk” are now underway in many countries around the world, typically put forth by importing models developed elsewhere and exploring whether they function similarly in new cultures. The approaches and interventions are often modified and tailored to these new contexts, with variations among cultural lines typically examined as secondary outcomes or concerns. These efforts thrust into the spotlight the importance of understanding how differences in culture may impact risk assessment and intervention paradigms as they transport out of the cultural contexts in which they were developed into new ones. The purpose of this book is to present international perspectives on the identification of those at risk for psychosis, with a particular focus on

efforts to intervene in those who may be in the prodromal phase of illness. The goal is to examine how culturally linked factors may affect this endeavor, a topic that has to date been vastly understudied. It is of critical importance as the early intervention paradigm revolutionizes traditional models in psychiatry and catches new roots. This volume gathers the broadest set of authors on this topic to date in order to broaden the scope of the field at a key time in its growth. A second goal is to summarize international research and clinical endeavors that f­ acilitate culturally informed assessment and intervention approaches. Terminology and structure of this edited book are discussed below.

2

Terminology Conventions in the Present Volume

A number of paradigms have been developed for identifying individuals thought to be at incipient risk of psychosis (e.g., in the prodromal phase), each with corresponding terminology and syndrome names, typically attached to the measure or method used for identification. In preparing the current volume, it has become evident that different conventions, terminologies, and names predominate in different areas of the world. Sometimes, similar terms are applied according to different conventions or operationalized in unique ways, due to linguistic differences, necessity, structural limitation, or culturally linked differences in how psychosis and mental health are conceptualized. Sociocontextual systems shape how models developed in one place and time are understood and implemented in another. Sometimes culturally linked factors, including community and healthcare structures that have developed within specific cultural contexts, affect when individuals come to clinical attention, for what concerns, and to whom, all of which may constrain how the same models of high psychosis risk are implemented. However, all systems for operationalizing putative incipient risk tend to share one major overlap in core phenomenology, which is that they involve the presence of certain characteristic signs and symptoms that manifest

1  Attenuated Psychosis Syndromes Seen Through the Cultural Prism: Relevance, Terminology, and Book…

in attenuated forms during the risk phase, relative to the acute phase of psychosis. Major terms that have been used in the field include prodromal (e.g., Keith & Matthews, 1991), ultrahigh risk (Yung 2003), at risk mental state (McGorry & Singh, 1995; McGorry et al., 2005), clinical high risk (CHR) (Correll, Hauser, Auther, & Cornblatt, 2010), psychosis risk syndrome (PRS) (Correll et al., 2010; McGlashan, Walsh, & Woods, 2010), outpost syndrome (Huber, Gross, Schiittler, & Linz, 1980), “prepsychotic,” and basic symptom syndrome or COGDIS (Klosterkötter, Ebel, Schultze-Lutter, & Steinmeyer, 1996; Schultze-­ Lutter et  al., 2012), among others discussed throughout this volume. In an attempt to capture this diversity, we use in this volume the term Attenuated Psychosis Syndromes (APS) to collectively refer to the class of putative prodromal or high-risk syndromes that have been empirically validated. Additional specificity (e.g., CHR, PRS) is used when warranted or in reviewing a particular study or methodology that uses a particular term. We acknowledge that this is not a universally accepted convention but use it here as a general umbrella term that can allow for a discussion and comparison of both similar and different operationalizations from writers around the world.

3

Book Structure

This edited book includes research and clinical findings on APS and the role culture plays in its symptom presentation, assessment, and treatment in various cultures around the world. It includes contributions from prominent researchers from six continents, each using a similar structure to identify their context and elaborate on their region of focus, composite racial and ethnic groups, health systems, typical APS presentations, help-seeking behaviors, barriers to services, and assessment and intervention strategies in their unique sociocultural contexts. There are six major sections in the book, following a forward by Dr. Patrick McGorry. Section I is comprised of summaries of the theoretical framework that has guided scientific and clinical attempts to

5

identify those likely to develop psychosis later in life and how different paradigms for identifying those at high risk do at prospectively predicting illness (Chap. 2). In Chap. 3, we summarize the rationale for early intervention and provide a brief review of extant treatments and their effects in those at putative risk for psychosis. Section II addresses the conceptual and measurement foundations of APS, with chapters written by key figures in the development of each of the currently predominant APS paradigms. Specifically, Chap. 4 summarizes early identification and intervention programs for APS in Australian youth and the role that the originators of the UHR paradigm have played in shaping early intervention throughout the world. The development and utility of the Structured Interview for Psychosis-Risk Syndromes and the Scale of Psychosis Risk Symptoms is discussed in Chap. 5. In Chap. 6 the basic symptoms approach is presented in the context of a discussion about disorders of the self. This model of psychosis risk is demonstrated via discussion of its application among Swiss and German youth. There are two chapters in Sect. III Borderlands of Cultural and Medical Conceptualizations of APS: Chap. 7 on early psychotic experiences from an anthropological perspective, illustrated via an Indonesian cultural viewpoint, and Chap. 8 on medical causes of APS.  Section IV includes contributions from a diverse collection of international authors, summarizing research and clinical practice on APS in their specific regions and cultures. Each set of authors also provides an overview of their specific cultural and structural context and then some discussion of how culture may affect presentation, pathways and barriers to care, the validity and reliability of methods of identifying APS, and suggestions or guidelines for providing culturally competent care. Furthermore, using a case illustration approach, the authors in Sect. IV provide readers an opportunity to understand the illness and its assessment and intervention within their specific cultural landscape, giving an in-­ depth view of individuals, families, and the mental health systems they interact with. There are 12 chapters in this section, presenting work from North America (Canada, African Americans,

6

Asian Americans, Latino Americans, Mexico), South America (Brazil), Africa (Nigeria), Asia (China, India), and Europe (Denmark, Spain). A summary and directions for the future are presented in Sect. V.  Dr. Barbara Cornblatt concludes the book with an epilogue, highlighting where the next horizon in APS research and clinical practice may be amidst a future of increasing globalization.

References Correll, C.  U., Hauser, M., Auther, M., & Cornblatt, B. (2010). Research in people with the psychosis risk syndrome: A review of the current evidence and future directions. Journal of Child Psychology and Psychiatry, 51(4), 390–431. https://doi. org/10.1111/j.1469-7610.2010.02235.x Huber, G., Gross, G., Schiittler, R., & Linz, M. (1980). Longitudinal studies of schizophrenic patients. Schizophrenia Bulletin, 6(4), 592–605. 1980. Keith, S. J., & Matthews, S. M. (1991). The diagnosis of schizophrenia: A review of onset and duration issues. Schizophrenia Bulletin, 17(1), 51–67. Klosterkötter, J., Ebel, H., Schultze-Lutter, F., & Steinmeyer, E. M. (1996). Diagnostic validity of basic symptoms. European Archives of Psychiatry and Clinical Neuroscience, 246(3), 147–154. Loebel, A. D., Lieberman, J. A., Alvir, J. M. J., Mayerhoff, D.  I., Geisler, S.  H., & Szymanski, S.  R. (1992). Duration of psychosis and outcome in first-episode schizophrenia. American Journal of Psychiatry, 149(9), 1183–1188.

D. I. Shapiro et al. McGlashan, T., Walsh, B., Woods, S. (2010). The psychosis-­risk syndrome: Handbook for diagnosis and follow-up. Oxford Publisher, New York. McGorry, P. D., Phillips, L. J., Kelly, D., Dell’Olio, M., Francey, S.  M., … Buckby, J.  (2005). Mapping the onset of psychosis: The comprehensive assessment of at-risk mental states. Australian New Zealand Journal of Psychiatry, 39, 964–971. McGorry, P.  D., & Singh, B.  S. (1995). Schizophrenia: Risk and possibility. In B. Raphael & G. D. Burrows (Eds.), Handbook of preventive psychiatry (pp.  492– 514). Amsterdam: Elsevier. Moreno-Küstner, B., Martín, C., & Pastor, L. (2018). Prevalence of psychotic disorders and its association with methodological issues: A systematic review and meta-analyses. PLoS One, 13(4), e0195687. Schultze-Lutter, F., Ruhrmann, S., Fusar-Poli, P., Bechdolf, A., Schimmelmann, B. G., & Klosterkötter, J. (2012). Basic symptoms and the prediction of first-­ episode psychosis. Current Pharmacological Design, 18(4), 351–357. Tandon, R., Nasrallah, H. A., & Keshavan, M. S. (2010). Schizophrenia, “just the facts”: Treatment and prevention. Past, present, and future. Schizophrenia Research, 122(1–3), 1–23. Vigo, D., Thornicroft, G., & Atun, R. (2016). Estimating the true global burden of mental illness. The Lancet Psychiatry, 3, 171–178. Woodberry, K. A., Shapiro, I. A., Bryant, C., & Seidman, L. J. (2016). Progress and future directions in research on the psychosis prodrome: A review for clinicians. Harvard Review of Psychiatry, 24, 87–103. Yung, A. R. (2003). Commentary: The schizophrenia prodrome: A high-risk concept. Schizophrenia Bulletin, 29, 859–865. Yung, A.  R., & McGorry, P.  D. (1996). The prodromal phase of first-episode psychosis: Past and current conceptualizations. Schizophrenia Bulletin, 22, 353–370.

2

Assessment of Risk for Psychosis Daniel I. Shapiro, Huijun Li, Emily R. Kline, and Margaret A. Niznikiewicz

Editors’ Note  A number of paradigms exist for prospectively identifying individuals who have elevated risk for developing psychosis due to the presence of syndromes comprised of identifiable risk factors and risk indicators. Conventions for which models are used, how individuals are

D. I. Shapiro (*) Department of Psychiatry, Early Psychosis Programs, University of California-Davis, Sacramento, CA, USA Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA e-mail: [email protected] H. Li Department of Psychology, College of Social Sciences, Arts and Humanities, Florida A&M University, Tallahassee, FL, USA e-mail: [email protected] E. R. Kline Beth Israel Deaconess Medical Center, Boston, MA, USA Harvard Medical School, Boston, MA, USA e-mail: [email protected] M. A. Niznikiewicz Beth Israel Deaconess Medical Center, Boston, MA, USA Harvard Medical School, Boston, MA, USA Boston VA Healthcare System, Brockton, MA, USA e-mail: [email protected]

identified, and which terminologies predominate vary throughout the world, sometimes related to culturally linked factors. In order to capture this diversity within one volume, the term Attenuated Psychosis Syndromes (APS) is used here to collectively refer to the class of putative prodromal or psychosis-risk syndromes that have been empirically validated. We acknowledge that this is not a universally accepted convention but use it as an umbrella term due to its heuristic value.

1

Introduction and Orientation to the Chapter

This chapter aims to summarize methods for assessing risk for psychosis, presented with an eye to how the history of this endeavor has shaped its methodologies. While efforts to identify individual markers of risk for psychosis predate the development of psychosis-risk syndromes, we will start this chapter by introducing the concept of risk in psychiatry, then identify Attenuated Psychosis Syndromes (APS), and briefly discuss the methods and major measures for identifying them (Part I). We also include vignettes that illustrate prototypal presentations of a number of APS. In the second part of the chapter, we attempt to broaden our lens by presenting the theoretical framework that underlies efforts to assess risk for psychosis, both via specific markers

© Springer Nature Switzerland AG 2019 H. Li et al. (eds.), Handbook of Attenuated Psychosis Syndrome Across Cultures, https://doi.org/10.1007/978-3-030-17336-4_2

7

D. I. Shapiro et al.

8

and via methods that attempt to prospectively quantify risk at the individual level. In total, we aim to give a brief overview of approaches to the identification of risk factors and risk indicators and then summarize research devoted to their elucidation. We will close with remarks on how these different sources of knowledge, clinical observations, epidemiology, research on biomarkers, and a newer focus on prediction of psychosis, can be used jointly to maximize the predictive power of available tools to aid in real-­ time clinical decision-making regarding the possible course of unfolding illness.

2

Part I

2.1

Approaches to Psychosis-Risk Identification: An Introduction

From a public health perspective, approaches to preventing psychiatric illness can be characterized as falling into universal, selective, and indicated (Mrazek & Haggerty, 1994) strategies. Universal approaches are applied at the population level and do not relate to any specific person or markers of risk. They are meant to target risk factors thought to affect a whole population. Anti-stigma campaigns that place billboards in busy thoroughfares are a good example. Because lack of understanding and negative beliefs about mental illness are associated with decreased and later utilization of mental health care, these risk factors are thought to affect everybody. Putting fluoride in the common water supply to prevent tooth decay is another. Selective approaches are applied to members of a subgroup because membership in that subgroup conveys risk. A good example is a special education and treatment group for children of substance using parents because membership in this group is known to be associated with higher incidence of substance abuse later in life. In the indicated prevention framework, early signs of illness or risk “indicators” that imply one is in the early stages of illness are identified and lead to intervention (McGorry, 1998).

Currently, the predominant approaches to early, putatively preventive intervention for psychosis use a risk factor and/or indicated prevention framework. Specifically, the broad theory is that all individuals carry some level of risk, or probability, for developing psychosis based on the accumulation of dynamic and interactive biological (e.g., genetic impacting brain structure and function), environmental, and psychological contributants (Woodberry, Shapiro, Bryant, & Seidman, 2016). This risk for psychosis is not dichotomous but rather exists on a continuum; we are all vulnerable to developing psychosis under the right conditions, but some people are more vulnerable than others, and certain stressors or circumstances are more likely to activate this vulnerability or protect against its expression (e.g., Rosenthal, 1970). This degree of risk, however, is difficult to unambiguously and prospectively measure because there are no sufficiently sensitive predictors of psychosis that can portend every case and because none of the risk indicators that have been identified are specific at an individual level—the same risk marker may lead to different end points for different individuals (multifinality). In the absence of such markers, clinicians are left to estimate latent risk by measuring risk/protective factors and indicators and making inferences about the probability of illness. Risk and protective factors are phenomena that have been shown to be associated with a higher or lower likelihood of a specific outcome—here, the development (or not) of psychosis—e.g., having a first-degree relative with psychosis confers a greater risk of psychosis development, while a host of mitigating factors confers protection (Kendler & Diehl, 1993). In a clinical suicide assessment, the presence of a previous suicide attempt is a risk factor for a subsequent attempt. With respect to psychosis, risk indicators are signs and symptoms that, while manifest, signal an increased likelihood of developing psychosis. Insights on risk indicators for psychosis have come from retrospective reports about individuals who already have psychosis (Yung & McGorry, 1996a, 1996b), from studies of family members of those with psychosis who presum-

2  Assessment of Risk for Psychosis

ably share some of the same (epi)genetic risk factors (e.g., Kendler, Neale, & Walsh, 1981), and more recently from prospective research on those with syndromes shown to predict psychosis in a significant proportion of cases (e.g., ultra/clinical high-risk or basic symptom Attenuated Psychosis Syndromes (APS)). Indices that discriminate those with an APS who develop psychosis from those who do not are indicators of risk, when present. Examples of risk indicators are specific syndromes of signs (observable phenomena) and symptoms (subjectively experienced phenomena). With APS, these include decline in social and vocational functioning, subthreshold positive and disorganized symptoms of psychosis, and new or worsening cognitive symptoms (see next section). As discussed in Part II of this chapter, other risk indicators include characteristic cognitive function changes like difficulties with memory and executive functions (Seidman et  al., 2016), disruptions of function like impaired tolerance or experience of stress (e.g., Walker et al., 2013), early social difficulties (Tarbox & Pogue-­ Geile, 2008), changes in brain structure and function (Pantelis et al., 2009), and neuroinflammation (Flatow, Buckley, & Miller, 2013) or environmental factors that indicate chronic stress or adversity (Bentall et al., 2014; Longden & Read, 2016). Detailed review of each of these areas is beyond the scope of this chapter, but we summarize the major areas after discussing Attenuated Psychosis Syndromes (APS). We present APS here because in recent decades they have played a major role in efforts to identify markers of risk.

3

Attenuated Psychosis Syndromes

The question of how to identify individuals who might be at highest risk for developing psychosis or who may already be experiencing its prodromal stages has guided efforts to balance the benefits of intervening at the earliest possible moment with the potential risks or costs of undertaking clinical interventions where they are not warranted. To this end, a number of paradigms have been developed over roughly the past half

9

century to identify individuals with clinical syndromes that (1) resemble what retrospective research has identified as common prodromal presentations or (2) presumably indicate high levels of biological risk. Most typically these syndromes involve the presence of subthreshold forms of psychotic or thought disorder symptoms or decline in functioning in those with a known family history, presumably due to early aspects of illness processes. Because different systems for identifying such syndromes have developed and are utilized around the world, all sharing in common some of these features, we collectively refer to them as Attenuated Psychosis Syndromes (APS). The most prominent examples, discussed separately, are the ultra/clinical high-risk syndromes, basic symptoms syndromes, and syndromes that assess schizotypy or “milder” ends of a psychosis spectrum. Each of these will be briefly discussed, key measurement tools will be presented, and a vignette depicting a prototypical case will be presented.

3.1

Ultra-High-Risk and Clinical High-Risk Syndromes

In the 1990s Alison Yung, Pat McGorry, and colleagues at the University of Melbourne (and the PACE clinic) utilized a “close-in” strategy to develop the concept of “ultra-high-risk” syndromes (see Chap. 4 for thorough discussion by Nelson and McGorry). Building on research on biological relatives of those with schizophrenia or other psychotic disorders, as well as retrospective research in those with extent illness, these authors developed three different syndromes comprised of the most commonly described clinical features and presentations observed during the prodrome (McGorry, Yung, & Phillips, 2003; Yung & McGorry, 1996a, 1996b). These so-­ called “ultra-high-risk” syndromes, characterized by specific attenuated positive symptoms of particular duration and intensity (attenuated psychosis symptoms syndrome/group), outpost syndromes characterized by brief periods of threshold psychotic symptoms (Brief Limited or Intermittent Psychosis Symptoms Syndrome/

10

group) and a trait and state risk factor group comprised of individuals with a recent decrease in functioning plus either a first-degree relative with a psychotic disorder or who meet criteria for schizotypal personality disorder (vulnerability syndrome). While these syndromes are meant to identify individuals who are putatively in the prodrome, these authors collectively refer to the period categorized by these three syndromes as the “at-risk mental state” (ARMS), given that a prodrome can only be defined as such once an endpoint is known. Shortly after these developments in Australia, Tandy Miller, Tom McGlashan, and colleagues in the United States picked up this framework and developed a measurement tool called the SIPS, originally the Structured Interview for Prodromal Syndromes (Miller et al., 2002, 2003) (see Chap. 5 for discussion from the developing group). They referred to syndromes in the ARMS as psychosis-­ risk syndromes and operationalized nearly the same three specific syndromes as Attenuated Psychosis Symptom Syndrome, Brief Intermittent Psychosis Syndrome, and Genetic Risk and Deterioration Syndrome, which they refer to as “clinical high-risk” syndromes.

3.1.1 Assessment of Ultra-High-Risk and Clinical High-Risk Syndromes Two clinician-administered assessment tools have been developed and widely disseminated for assessment of clinical/ultra-high-risk syndrome APS and are the most utilized methods for identifying an APS worldwide: the Comprehensive Assessment of At-Risk Mental States (CAARMS) and the Structured Interview for Prodromal/Psychosis-Risk Syndromes (SIPS). These measures are discussed in depth by the authors involved in their respective development in Chaps. 4 and 5, so they are covered only briefly here. In short, both are semi-structured assessment instruments containing questions for interviewers, rating anchors for symptom severity, and diagnostic criteria and scales for identifying psychosis-risk syndromes. In the SIPS, the scale for identifying clinical high-risk syndromes is called the Summary of

D. I. Shapiro et al.

SIPS Syndrome Criteria (SOPS). A seminal meta-analysis including predictive validity studies of these two tools indicates that approximately 29–36% of those identified with a CHR/ UHR syndrome transition to full psychosis within 2–3 years (Fusar-Poli et al., 2012). The Comprehensive Assessment of At-Risk Mental States (CAARMS; Yung et al., 2005) is a semi-structured interview and assessment tool initially developed at the Personal Assistance and Crisis Evaluation (PACE) clinic in Australia to identify vulnerability, attenuated psychosis, and brief limited intermittent psychosis “ultra-high-­ risk” (UHR) syndromes, and criteria for psychosis is called “psychosis threshold.” The scale is comprised of 28 items organized into seven symptom subscales and rated separately for severity and frequency. On several of the subscales, both subjective and objective observation is rated. The subscales cover the following domains: positive symptoms (unusual thought content, non-bizarre ideas, perceptual abnormalities, disorganized speech), cognitive change (subjective experience, observed cognitive change), emotional disturbance (subjective emotional disturbance, observed blunted affect, observed inappropriate affect), negative symptoms (alogia, avolition/apathy, anhedonia), behavioral change (social isolation, impaired role functioning, disorganizing/odd/stigmatizing behavior, aggression/dangerous behavior), motor/physical change (subjective complaints of impaired motor functioning, informant reported or observed changes in motor functioning, subjective complaints of impaired bodily sensation, subjective complaints of impaired autonomic functioning), and general psychopathology (mania, depression, suicidality and self-harm, mood swings/lability, anxiety, obsessive-­ compulsive disorder symptoms, dissociative symptoms, impaired tolerance to normal stress). Only the positive symptom domain is used to identify UHR syndromes. This instrument has been translated into four languages, and the Youth Psychosis At-Risk Questionnaire (YPARQ; Ord, Myles-Worsely, Blailes, & Ngirlmau, 2004) has been constructed as a screening questionnaire based on the CAARMS.

2  Assessment of Risk for Psychosis

The CAARMS has demonstrated good discriminant validity and excellent inter-rater reliability (ICC in the range of 0.62–0.93 with only one subscale below 0.7). Sensitivity, or the ability to predict psychosis in all tested individuals, was 83% at 6 months, and specificity, or the percentage of all tested individuals who did not meet criteria and did not develop illness, was 74% at 6  months. Construct validity is high since high CAARMS scores in a UHR groups are significantly associated with onset of psychotic disorder (Yung et al., 2005). The Structured Interview of Psychosis-Risk (née Prodromal) Syndromes (SIPS; Miller et al., 2003) is a semi-structured diagnostic interview designed for trained clinicians to identify clinical high-risk syndromes. The SIPS includes five components: a 19-item Scale of Prodromal Syndromes (SOPS), a checklist for the Criteria of Prodromal Symptoms (COPS), Global Assessment of Functioning, DSM-IV Schizotypal Personality Disorder checklist, and a family history of mental illness. The SIPS is used to identify the attenuated positive symptom syndrome, the Brief Intermittent Psychosis Syndrome, the Genetic Risk and Deterioration Syndrome, and criteria for transition to psychosis called the presence of psychosis syndrome (POPS). The SIPS has been translated into 15 different languages and is the basis of the screening instrument, the PRIME screen (Miller, 2004), which has also been translated into at least two other languages (Kobayashi et al., 2008; Mamah et al., 2012). In addition to identifying the presence of clinical high-risk syndromes, newer versions of the SIPS also evaluate progression or remission of these syndromes over time. It is comprised of four subscales—positive, negative, disorganization, and general symptoms scales. The positive subscale is made up of five items: unusual thought content/delusional ideas, suspiciousness/persecutory ideas, grandiose ideas, perceptual abnormalities/hallucination, and disorganized communication. Like the CAARMS, only the positive symptoms subscale is used to identify CHR syndromes, and the other three subscales measure the severity of symptoms once the “diagnosis” is established.

11

Reliability and validity data indicate excellent inter-rater reliability and predictive validity. Specifically, the developers of the SIPS found inter-rater agreement in making diagnostic judgments regarding the presence of psychosis-risk syndromes of 93% (Miller et al., 2003). The SOPS rating scale has shown high reliability (ICC values at 0.95 for the total score and above 0.75 for all four subscales—positive, negative, disorganized, and general symptoms subscales). Sensitivity (percentage of all tested individuals who have developed this illness) was 100% at 6, 12, and 24 months; and specificity (percentage of all tested individuals who did not meet criteria and have not developed illness) was assessed at 71%, 74%, and 73% at 6, 12, and 24 months.

3.1.2 Attenuated Psychosis Vignette The following vignette illustrates a typical APS presentation in North America.

Mark: Attenuated Psychosis Syndrome

“Mark” is a 16 year old who lives with his mother and stepfather in a poor urban neighborhood in Canada. He is in 9th grade and was diagnosed with ADHD at age 11 in the context of academic difficulties. He had typically gotten As, Bs, and Cs in school and had a good group of friends. However, over the past school year, his grades have been trending downward, and he has become a C average student, stating that he has had a lot of difficulty paying attention in class. He enjoys riding his bike and creating art and videos that he posts online. According to Mark’s mother, there is no known family history of mental illness in either her or Mark’s father’s families. He was referred to a specialized early psychosis clinic by his general practitioner where the SIPS was administered in the context of a specialized psychodiagnostic assessment.

D. I. Shapiro et al.

12

About 6 months ago, Mark began to feel that he could predict events in the future. For example, he recently entered the cafeteria at lunchtime and felt strongly that there would be a fight, and then a fight did take place. At other times, he feels he is predicting odd details of daily life, such as random body movements of his classmates, his mother’s car pulling into the driveway, or that an object will fall off a table or wall. Over time, his predictions have become more specific and frequent. He finds this phenomenon weird rather than scary, and his friends generally believe and get excited about his predictions rather than finding them odd. He states that he does not behave any differently due to these predictions but finds them absorbing, and they occupy his attention. Mark also states that “I can never let my guard down. That’s just my opinion.” His mother describes him as “vigilant” without any particular focused concern about his safety. She also stated that she feels their neighborhood is somewhat dangerous, and at times she wishes he was more (rather than less) vigilant. He feels that his peers are not trustworthy because everyone at school gossips. He denies feeling that anyone in particular is talking about him or trying to make things harder for him. He denies feeling watched or singled out. His feelings of mistrustfulness began last year when he was bullied by some classmates. Mark reports hearing three voices that occur on and off every day. He hears them outside his head as actual sounds but is aware that no one else can hear them. They are sometimes helpful and sometimes have violent or sexual content. He is not sure where the voices are coming from; it could be his mind playing tricks on him or maybe spirits. He has noticed that he tends to hear the voices more when he is stressed out or upset and that he rarely hears them when he is relaxed and hanging out with friends.

Based on these experiences and a drop in his functioning at school (but not with friends), Mark met criteria for an attenuated positive symptom CHR syndrome on the SIPS.

A few details about “Mark” make this case a good example of a CHR/UHR APS. First, he has prominent attenuated positive symptoms (auditory hallucinations, a vague sense of paranoia or foreboding, an odd experience of predicting the future). Second, these positive symptoms are new and worsening rather than long-standing or viewed by Mark as part of his typical personality. Third, Mark has experienced some difficulties maintaining attention and follow-through in his schoolwork over the last year which has led to a decline in his academic functioning but in general was functioning fairly well socially and academically before his positive symptoms began. Frequency of experiences was not discussed in the vignette, but as assessed with either the SIPS or the CAARMS, attenuated positive symptoms would need to occur at least once per week over the past month (SIPS) or at least once per month if they persisted for more than an hour and three to six times per week if shorter (CAARMS).

3.2

Basic Symptoms Syndrome

The basic symptoms concept developed prior to and independently of the UHR/CHR model that focuses primarily on attenuated positive symptoms (Huber & Gross, 1989). The “basic symptoms syndrome” has a number of similarities to the attenuated symptoms model of a clinical high-risk state. Like APS, the basic symptoms concept was first described by clinician scientists examining retrospectively the emergence of psychosis among patients prodromal to schizophrenia (Schultze-Lutter & Theodoridou, 2017). These careful retrospective observations were then operationalized in order to study the predictive validity of a psychosis-risk category focus-

2  Assessment of Risk for Psychosis

ing on the emergence of “basic” symptoms. Basic symptoms are “subtle, subjectively experienced disturbances in mental processes including thinking, speech, attention, perception, drive, stress tolerance, and affect” (Schultze-Lutter & Theodoridou, 2017). They are conceptualized as the “basic” responses of the self-perceptual system to neurobiological changes. Thus, this cluster of symptoms may represent a very early prodrome, with more elaborated expressions of delusions and other positive symptoms representing a later stage. Basic symptoms might be described as subtle alterations in one’s sense of self, not necessarily resulting in behavioral changes observable by others. These alterations have been reported in prodromal, acute, and residual phases of psychotic illness but on their own (without other symptoms such as delusions, hallucinations, and negative symptoms) may represent the very early expression of psychosis vulnerability or an “early prodrome” (Schultze-Lutter, 2009), and thus when examined in isolation, the concept has been criticized as overly broad and lacking in specificity (Schultze-Lutter & Theodoridou, 2017). This is theoretically consistent as more common, non-­specific symptoms are thought to characterize the early prodrome which is thus more difficult to prospectively identify (Yung & McGorry, 1996a, 1996b). For example, a patient with no prior mental health difficulties might describe her thoughts as sped up or slowed down, feel that she is especially vulnerable to distraction, or notice that her hearing has become extrasensitive to certain pitches or voices. Some of these experiences would be captured by the “attenuated positive symptoms” concept, while others would not be. The Schizophrenia Proneness Instrument (SPI) (Schultze-Lutter, Addington, Ruhrmann, & Klosterkötter, 2007; Schultze-Lutter & Koch, 2010; Schultze-Lutter, Ruhrmann, Picker, & Klosterkötter, 2006) can be used to assess basic symptoms in children and adults age 8 or older. It was developed from an initial measure, the Bonn Scale for the Assessment of Basic Symptoms (Gross et al., 1987; Gross & Huber, 2010) to specifically capture items that were predictive of

13

psychosis (Klosterkotter, Hellmich, Steinmeyer, & Schultze-Lutter, 2001). There are nine subscales on the SPI, but two are considered “basic symptoms”: the Cognitive-Perceptual Basic Symptoms scale (COPER) and the Cognitive Disturbances scale (COGDIS), the high-risk state operationalized within the SPI.  COGDIS items include inability to divide attention, thought interference, thought pressure, disturbance of receptive speech, disturbance of expressive speech, unstable ideas of reference, disturbances of abstract thinking, captivation of attention by details of the visual field, and thought blockages. COGDIS is met when at least two symptoms are endorsed, occur at least weekly, and are considered a change from the premorbid state. There are child and youth (SPI-CY) and adult (SPI-A) versions of the instrument, which is available in several different languages. The SPI-A measures disturbances in affective functioning, impairments in attention, basic cognitive disturbances, disturbances in self-­ perception and experiences, impaired bodily sensations, and sensitivity to perception. Good inter-rater reliability and construct validity have been established (Schultze-Lutter et  al., 2012; Schultze-Lutter et al., 2007). Furthermore, 34.9% of those experiencing at least one basic symptom experienced psychosis within a 1-year period, suggesting good predictive and construct validity (Schultze-Lutter et  al., 2006, 2012). The SPI-­ Child and Youth version (SPI-CY) (Fux, Walger, Schimmelmann, & Schultze-Lutter, 2013; Schultze-Lutter et al., 2012) is for individuals above age 8 when a certain degree of self-reflection is developed. A few items that require higher level of metacognitive process involving more abstract thinking are for children above age 13. Good to excellent discriminant validity was evidenced, though severity rather than presence of basic symptoms has made the greatest contribution to discriminating groups (Fux et al., 2013). Studies using the basic symptom approach have shown highly variable rates of transition to psychosis. Some studies of the basic symptoms syndrome have demonstrated very high rates (~50%) of conversion to psychosis, including the only study of either the basic symptoms or

D. I. Shapiro et al.

14

CHR/UHR APS that has thus far reported a follow-up time longer than 3–4 years (Klosterkotter et al., 2001). On the other hand, one recent study showed that basic symptoms criteria did not perform better than chance in predicting psychosis onset among a group of patients followed over 3 years (Hengartner 2018). The best results thus far have come when using the basic symptom paradigm in combination with the UHR/CHR paradigm (Schultze-Lutter, Klosterkötter, & Ruhrmann, 2014), prompting the European Psychiatric Association to recommend COGDIS in addition to UHR as models for early detection of individuals at risk for psychosis (Schultze-­ Lutter et al., 2015). See Chap. 19 of this volume for further discussion of the assessment of anomalous self-experience.

3.2.1 V  ignette Illustrating a “COGDIS” Case

Brianna: Basic Symptoms Syndrome

“Brianna” is a 24-year-old woman employed as a waitress and residing with roommates in a large urban center in North America. Although her roommates, coworkers, and family have not noticed any change in her, Brianna has begun to worry about her mental health during the past 3 months. Specifically, she feels that something has shifted in the way her mind and body are working. It is hard for her to put this experience into words. She has noticed that several times per week she has the sensation of being in a tunnel in which lights appear more glaring and sounds seem to echo more than usual. She also has the sense of needing to work harder to find the correct words and phrases to express what she intends to say while interacting with customers during her nightly shift. No one has said anything to indicate that they have noticed a problem with Brianna’s speech, but the subjective difficulty has led her to feel somewhat anxious during these

encounters. She feels that she used to be more outgoing and gregarious with customers and is not her normal self at the moment. She wonders if something might be wrong with the way her brain is working.

3.3

Schizotypy, Schizotaxia, and Psychotic-Like Experiences

As far back as the early twentieth century, the psychiatrist Eugen Bleuler observed that psychotic disorders appeared to constitute a spectrum that encompassed both mild and more debilitating manifestations of psychosis (Bleuler, 1950). Roughly half a century later, Paul Meehl and colleagues developed the concept of schizotypy and schizotaxia. Schizotaxia was thought to be a biological predisposition to psychosis—a “neural integrative defect”— which resulted in certain phenotypic or personality presentations, specifically schizotypy (Meehl, 1962, 1989). He proposed that schizotaxia interacted with other genetic factors, certain personality characteristics, and stressors or environmental risk factors to “potentiate” into clinical schizophrenia or nonpsychotic schizotypal states, depending on interaction with other factors. This early diathesis stress model suggested that schizotypal traits, features, or experiences indicated a biological vulnerability to psychosis dimension, which varies along a continuum in the general population. These manifestations can thus be seen as risk indicators but also exist without psychosis, presumably in those whose schizotaxia is not sufficient to lead to full-blown psychosis or in whom such schizotaxia was not sufficiently shaped by experience to lead to psychosis. Thus, the presence of schizotypy alone is not enough to lead to psychosis. This conceptual theory has broadly been borne out by biological research (Nelson, Seal, Pantelis, & Phillips, 2013) like studies that show neural

2  Assessment of Risk for Psychosis

similarities in those with schizotypy and psychotic disorders (Modenato & Draganski, 2015), as well as in twin studies that show concordance of schizotypy (Jang, Woodward, Lang, Honer, & Livesley, 2005). Construct validity for using schizotypy as a marker of risk for psychosis comes from studies demonstrating association between measures of schizotypy traits and both psychotic disorders, other APS (e.g., UHR syndromes), and specific positive and negative symptoms (BarrantesVidal et al., 2013). Research also consistently finds that specific subclinical psychotic experiences like schizotypy are rather common and in most cases do not lead to an APS. For example, in a systematic review of studies of population rates of such subclinical psychotic experiences, Van Os, Linscott, Myin-­Germeys, Delespaul, and Krabbendam (2009) found that around 5% of the given population likely experiences subclinical psychosis at a given time, and the vast majority (75–90%) of such cases are likely to resolve (i.e., the subclinical psychosis will remit over time) without intervention. Schizotypy is often operationalized as subclinical psychotic experiences. For instance, odd beliefs, unusual perceptual experiences, negative affect, avolition, asociality, and anergia are considered positive and negative schizotypy, respectively (Vollema & van den Bosch, 1995). Long-standing suspiciousness, experiences of seeing auras, or beliefs that some might think are unusual but may occur relatively commonly in a subculture (e.g., ghosts, telepathy, mind reading, unusual interests) are a few specific examples. A number of measures are available to assess schizotypy and are sometimes used as screening measures to identify possible psychosis risk among large numbers of individuals (e.g., college students). Many measures have been developed to assess specific schizotypy traits (see Mason, 2015 for review), and the schizophrenia scale on the Minnesota Multiphasic Personality Inventory (MMPI) was originally developed for this purpose. The Wisconsin Schizotypy Scales (WSS) are among the most popular measures (Kwapil, Barrantes, Vidal, & Silvia, 2007) and

15

include subscales initially developed by others. The WSS shows good predictive and construct validity (Barrantes-Vidal et  al., 2013; Kwapil et  al., 2007, 2012) and is available in multiple languages (e.g., Fonseca-Pedrero, Paino, LemosGiráldez, Sierra-­Baigrie, & Muñiz, 2010). The Structured Interview for Schizotypy-Revised (SIS-R) is another popular interview-based measure with good reliability (Vollema & Ormel, 2000). Schizotypal Personality Disorder (SPD) is another type of APS based on the schizotypy framework. The defining criteria for the SPD were based on research on the biological relatives of those with schizophrenia, based on the notion that they inherited some of the same heritable risk factors (Kendler, Neale, & Walsh, 1995; Kendler et al., 1981). To qualify for a diagnosis of SPD, which is included in both the DSM-5 and ICD-10, an individual must show a pattern of social and interpersonal deficits combined with at least five of the following: ideas of reference, odd beliefs, unusual perceptual experiences, suspiciousness, inappropriate or constricted affect, unusual behavior, odd thinking or speech, social anhedonia, and social anxiety associated with paranoid fears. When identified prospectively, individuals with SPD show a high rate of subsequent schizophrenia and other psychoses, supporting its utility as an APS.  For example, Asarnow et al. (2005) found that 75–92% of her sample developed schizophrenia over 3-year follow-­ ups. Supporting validity, a substantial proportion (50–70%) of individuals who meet diagnostic criteria for SPD also meet CHR criteria (Woods et  al., 2009), and SPD is far more common among relatives of individuals with schizophrenia than in the general population (Kendler et al., 1993). Estimated prevalence rates of SPD range from 0.6% to 4.6% (American Psychiatric Association, 2013). SPD can be assessed via interview measures for DSM-5 or ICD-10 (Structured Clinical Interview for DSM-5—Personality Disorders Version) (First, Williams, Benjamin, & Spitzer, 2015) or self-report measures such as the Schizotypal Personality Questionnaire (Raine, 1991).

16

Finally, similar to the concept of schizotypy are self-limiting and transient subthreshold psychotic-­ spectrum symptoms referred to as “psychosis-like experiences” (PLEs). Examples of PLEs are odd beliefs, dissociative anomalous perceptions, and hallucinatory anomalous perceptions (Unterrassner et  al., 2017) that, like schizotypy, do not lead to psychosis in the vast majority of cases. PLE is a label that conveys an agnosticism regarding both the origins and meaning of unusual thoughts and perceptions and are typically examined as isolated phenomena (e.g., not as part of an APS). However, like schizotypy, they are thought to represent some elevated vulnerability for psychosis and thus seen as indicators of risk, particular when identified in children (Kelleher & Cannon, 2011). PLEs are quite common in the general population, leading many to think of them as either the mild end of a psychosis spectrum, a nonclinical psychosis phenotype, or an extreme variation of normal personality. PLEs have been found to occur in 5–8% of the population at any given time (Van Os et  al., 2009), though with even higher prevalence rates in adolescents (Poulton et al., 2000). PLEs are associated with increased incident of subsequent psychotic disorders. In children, the presence of delusions or hallucinations has been associated with a 5- to 16-fold increase in risk for a psychotic-­spectrum disorder (Poulton et al., 2000; Welham et al., 2009), while a study in adults found that 8% of those who reported a PLE developed psychosis within a 2-year follow-­ up (Hanssen, Krabbendam, Vollema, Delespaul, & Van Os, 2005). PLEs also appear to cluster in families, supporting the idea that they may be linked to a schizotaxia (Hanssen, Krabbendam, Vollema, Delespaul, & Van Os, 2006), and covary with the presence of other known risk factors for psychosis (see Kelleher & Cannon, 2011 for review). Assessing PLEs is not so different from assessing APS or schizotypy. Brief, sensitively worded screening questions can be used to elicit descriptions of PLEs. The content and context of what patients then describe—the duration and frequency of such experiences, whether the experiences seem directly related to other events in the

D. I. Shapiro et al.

person’s life, whether they are distressing or causing impact on behavior, and the patient’s cultural and personal interpretations of the experiences— will guide the clinician in differentiating “APS” (recent, distressing, worsening) from subthreshold psychosis better conceptualized as an expression of schizotypy (long-standing, stable, may or may not cause distress) or “PLE” (self-­limiting or better explained by another disorder). Self-report measures have also been developed, for example, the Exceptional Experiences Questionnaire, referred to as the PAGE-R (Unterrassner et  al., 2017), and measures listed above in the schizotypy section could be conceived of as assessing PLEs. Many studies of PLEs use chart review to identify isolated psychotic-­spectrum symptoms, or screening measures for APS, like the Community Assessment of Psychic Experience (CAPE; Stefanis et  al., 2002) or those most commonly used to identify CHR/UHR cases discussed below.

3.3.1 Vignettes Illustrating Schizotypy and Psychotic-Like Experiences

Nelson: Schizotypal Personality Disorder

“Nelson” is a 25-year-old man residing with his biological parents in an upper-­ class suburb in Scotland. He is seeking psychotherapy for problems with depressed mood, irritability, and being unable to get along with others. He is living with his parents after 2  years of living independently while employed as a pastry chef and baker. He quit his job because he felt that he was disrespected by coworkers and not afforded the recognition he deserved, and he now questions whether the culinary industry is the right place for him. Nelson states that yesterday he slept most of the day and reread part of the Game of Thrones book series, which is fairly typical of his current functioning. Nelson has used marijuana heavily since high school and usually smokes a few times per day.

2  Assessment of Risk for Psychosis

Nelson acknowledges that he has unusual interests and preoccupations, specifically his “obsession” with the J.  R. R.  Tolkien Lord of the Rings series and books by the author Ayn Rand, his strong belief that the Internet should be avoided, and his desire to be an “ecoterrorist” by bombing companies and vehicles that have a negative ecological impact. He identifies with the character John Galt from the novel “Atlas Shrugged” (Ayn Rand) and idealizes a highly individualistic existence in which he can minimize interaction with others. He also admits that he knows his behavior and appearance can be odd and off-putting; he is disdainful of hygiene norms and sees no reason to smile at or engage in small talk with coworkers or family members. He has no close friends and states that he does not believe in friendship. Moreover, he states that he believes that all relationships are essentially exploitative. “I know intellectually that humans are meant to be social, but I see no benefit personally.” He states that he has not dated in 3  years and feels he is “past all that” and does not believe in romantic relationships. Although Nelson comes across as exceptionally bright, his speech is elaborative, vague, and at times hard to follow. Results of an assessment for an autism spectrum diagnosis did not support such a diagnosis. Nelson’s parents are quite concerned about his recent decision to quit his job and say that he seems quite unhappy most of the time. His mother is particularly worried because her brother has schizophrenia and she is worried that Nelson will develop a mental illness too. They state that his tendency to live in the world of novels and fantasy worlds and odd manner of speech have been a part of his personality since he was 9 or 10  years old but that he used to have an active social life with friends and girlfriends and he never seemed so unhappy as he does now.

17

Nelson has several hallmark features of schizotypy. He is socially isolative and has a paranoid way of viewing others; he has unusual interests and overvalued ideas about avoiding the Internet and exacting revenge upon corporations committing ecological harms; and he is socially odd and off-putting. Importantly, all these qualities are “ego-syntonic” in that he views these as essential aspects of his personality. His parents provide helpful context by confirming that these qualities are steady traits rather than representing a recent change in thinking or behavior. Finally, Nelson has a family member with schizophrenia and may have some of the same heritable risk factors for mental illness as his uncle.

Sarah, David, and Carla: Psychotic-Like Experiences

“Sarah” is a 35-year-old postal worker who loves fairies. She collects “fairy doors” which she places against the walls throughout her home in hopes that these will invite fairy spirits into her life. She attributes coincidences to fairies’ actions and likes to imagine that fairies inhabit her home while she and her family sleep at night. She has been doing this for 10 years and it has never caused problems for her. “David” is a 25-year-old graduate student whose close friend suddenly died from suicide last week. He catches himself having nihilistic thoughts such as “what is the point of studying for exams” and then wonders if his friend is somehow inserting these thoughts into his head. This occurs for about a week and then resolves without intervention. “Carla” is a 14-year-old high school student who experiences intense social anxiety. When she walks into a room, she hears people murmuring her name and feels that people are taking special notice of her. This becomes especially intense and distressing when she returns to her classroom from a

D. I. Shapiro et al.

18

data, leaving clinical methods as currently still the most common and perhaps the most easily disseminated method for observing and identifying risk. However, relying only on assessment in help-seekers to detect APS risks misses a great proportion of those in the general populaEach of these PLEs emerges in a different tion who are at risk, particularly among the most context, and thus we might interpret the meaning disadvantaged populations, who by definition of the experience and whether it indicates risk for have higher risk for developing psychosis. A psychosis differently for each of these individu- number of measures have thus been developed als. Sarah’s belief in sweet supernatural creatures to quickly screen for possible psychosis risk in visiting her home at night is not distressing for large numbers of people, with a growing literaher and causes no problems in her life. It is long-­ ture showing effectiveness in identifying those standing without any increase in intensity or dis- in most need of APS assessment. Major efforts tress; thus we might predict that this experience to universally screen in settings like schools and does not impact her risk for developing psycho- primary care offices that come into contact with sis. David, on the other hand, experienced an large numbers of adolescents and young adults acute onset of a somewhat distressing PLE in the are becoming the norm in cultures with centralcontext of a severe stressor. This experience of ized health or education systems. They are also thought intrusions was brief and self-limiting and being introduced in some that do not. We briefly thus does not meet the criteria for an APS label. discuss the most commonly used screening However, were this to recur in other contexts or tools for identifying APS. persist beyond a few weeks, we may be more concerned. Carla’s PLE is the most distressing and persistent of these examples. For now, the 4.1 Self-Report Measures experience may be better explained by her existing diagnosis of social phobia, but given the Screening tools are meant to identify individuals severity of these experiences, their harmful in need of APS assessment and are thus evaluated impact on her behavior, and her young age, we based on their ability to predict APS “diagnosis,” may have well-placed concern about her ability as well as on their ability to discriminate between to manage these symptoms and her risk for cases likely and not likely to meet criteria. psychosis. Unstructured, verbal screening for psychosis-risk symptoms is useful in a fast-paced clinical environment in which clinicians are routinely assessing patients for a variety of potential concerns, 4 Screening for Psychosis Risk particularly in specialty care settings specific to and APS psychotic disorders. Structured psychosis-risk This chapter has thus far discussed APS, which screening may be useful in other clinical contexts are syndromes combining clinical indicators of and particularly in nonclinical contexts. There elevated risk for psychosis. In Part II, it will also are four widely used screeners available, along identify biological and other markers of risk. with shortened sub-version, primarily for the The goal of prevention, particularly from an UHR/CHR APS paradigm. Kline and Schiffman indicated prevention perspective, requires that (2014) provide a good summary of three of these risk syndromes and biomarker risk factors be screeners and summarize their psychometric identified early. While there is hope that bio- properties. In brief, the PRIME Screen-Revised marker assessment may be more accessible in (Miller 2004) is keyed to the SIPS and contains routine clinical practice in the future, it remains 12 items describing attenuated positive sympexpensive and more heavily reliant on group toms. Individuals use a Likert scale to indicate bathroom break, and she therefore tries never to use the restrooms at school, which has led to urinary tract infections.

2  Assessment of Risk for Psychosis

their level of agreement with each item. The authors recommend a cutoff of five endorsed items, based on strong sensitivity and specificity. The Prodromal Questionnaire has two shortened versions, the Prodromal Questionnaire-Brief version (PQ-B; Loewy, Pearson, Vinogradov, Bearden, & Cannon, 2011) and the Prodromal Questionnaire-16 item (PQ-16; Ising et  al., 2012), which contain a number of items meant to identify positive and negative symptoms. Items are selected as true/false and individuals also rate level of distress. The PQ-B has 21 items, while the PQ-16 has 16, and a cutoff of 6 is recommended for each. The Youth Psychosis At-Risk Questionnaire-Brief (YPARQ-B) is keyed to the CAARMS and includes 28 items scored yes/no (Ord et al., 2004). A score of 11 yeses is recommended as a cutoff. The Community Assessment of Psychic Experience (CAPE) is also keyed to the CAARMS and has a 15-item version (Capra, Kavanagh, Hides, & Scott, 2013) which is comprised of three subscales: persecutory ideation, perceptual abnormalities, and bizarre experiences. Items focus on attenuated positive symptoms and are rated on frequency and distress (1–4). A score is derived by calculating the mean of responses. It has been used in at least 15 different countries (Mark & Toulopoulou, 2015) and has a recommended cutoff of 1.47 for both frequency and distress scores (Bukenaite et  al., 2017). For a discussion of other screeners that have received study but are less frequently used (e.g., ERIraos (Hafner et al., 2004) and Eppendorf Schizophrenia Inventory (Niessen et  al., 2010)) see Kline & Schiffman, 2014. Measures of schizotypy and PLEs are also often used to screen for possible APS.

5

Part II

5.1

 nown Markers of Risk K for Psychosis

A major aim of current research is to understand which individual or combined risk indicators and factors are most indicative of risk for psychosis.

19

In the sections below, we will first summarize the different factors and indicators that have been shown to be associated with subsequent psychosis. Then we discuss approaches for quantifying or stratifying risk in order to determine which individuals might be most at risk and thus in need of early intervention.

5.1.1 Familial Genetic Risk Factors There is strong evidence that schizophrenia and other psychotic disorders have a large genetic component, with heritability estimated at between 64% and 81% (Lichtenstein et al., 2009; Sullivan, Kendler, & Neale, 2003). For example, the approximate chance of developing schizophrenia in a child is 40% if both parents have the illness and 12% if one parent has it (Miller & Mason, 2002), compared to a roughly 1.1% chance in the overall population above age 18. In addition, when a biological child of individuals with a psychotic disorder is adopted, he or she has an elevated risk of developing psychosis, as expected for first-degree relatives. Further, if one of a pair of identical twins has a psychotic disorder, the children of both identical twins have higher rates of schizophrenia (Fatemi & Folsom, 2009). This means that having a family history of schizophrenia or other psychotic disorder in a first-degree relative is a risk factor for developing schizophrenia or other psychoses. Researchers have been trying to identify possible genetic risk factors through genetic studies in those who have developed schizophrenia and other psychoses. Research findings suggest that multiple genetic factors account for many cases of schizophrenia (Nicolson et  al., 2003; Ripke et al., 2013, 2014). Researchers using molecular genetic techniques (such as candidate gene analyses, genome scans, and linkage studies) have identified several specific genes (e.g., serotonin type 2a receptor (5-HT2a) gene responsible for learning and memory and the dopamine D3 receptor gene for cognitive and emotional functions) as contributing to the development of schizophrenia. However, it has become apparent that individual genetic polymorphisms only tend to account for a small amount of variance

20

in psychosis prediction (Badner & Gershon, 2002; Mowry & Nancarrow, 2001) and thus are not useful risk predictors. For example, a specific genotype associated with one gene might yield a 2–3% chance of psychosis, in the absence of any other risks, which is roughly similar to the population base rate of schizophrenia and other psychotic disorders. Thus the focus in recent years has been on genome-wide studies that examine the combined effect of large numbers of genes or attempt to identify specific polygenetic profiles that predict psychosis. These studies typically find between 100 and 200 different single genes that are associated with incidence of schizophrenia and other psychotic disorders (Pardinas et  al., 2018; Ripkes et al., 2014), but individuals with psychosis tend to vary in which specific genetic risk factors they manifest. A current research trend is on developing polygenetic risk scores that use the accumulations of specific risk genes to estimate a risk probably (Padmanabhan, Shah, Tandon, & Keshavan, 2017). However, researchers are still working on understanding how and whether such risk scores can be implemented in clinical settings. Further, genome-wide testing is still expensive and not easily accessible in most clinics. Thus, as of today, a positive family history of schizophrenia seems the only way to identify the presence of genetic risk factors with enough predictive power to influence decisions about whether somebody is at high risk. It is still unknown how often risk genes or gene alterations are inherited, how often they may lead to schizophrenia, how often individuals who possess a genetic vulnerability for schizophrenia pass it on to their offspring, and how environmental and clinical risk factors interact with these genetic predispositions to lead to psychotic disorders.

5.1.2 Environmental Risk Factors Environmental factors appear to confer risk for psychosis. Perhaps the strongest and earliest evidence that environmental factors are related to the development of psychosis comes from studies examining the rates of psychosis within families. Twin studies have found a roughly 50% schizo-

D. I. Shapiro et al.

phrenia concordance rate among monozygotic twins who by definition initially share 100% of their genes (Brown 2011). If heritable factors like genes alone led to psychosis, one would expect this rate to be 100%, meaning that factors other than genes contribute to illness. In fact, a number of environmental factors are associated with higher than normal incidence of psychosis, suggesting they could constitute risk factors. Broadly, these factors fall into two categories: those that affect neural development early including prenatal experiences and early and/or enduring adversity and factors that impact brain function later in life. Among those likely to affect brain development, increased incidence of schizophrenia has been associated with pregnancy complications like maternal bleeding, diabetes, preeclampsia, and rH incompatibility; abnormal fetal development (including malnutrition or exposure to virus), and delivery complications like hypoxia, asphyxia, delivery using forceps, and emergency C-section (Cannon et  al., 2002; Van Os, Kenis, & Rutten, 2010). Elevated risk for psychosis has also been associated with maternal exposure to a number of viruses and stressors (Torrey, Miller, Rawlings, & Yolken, 1997; Torrey et  al., 2009) and advanced paternal age, the latter likely due to increased risk of genetic aberrations (Saha, Chant, Welham, & McGrath, 2005). Thus the presence of these factors can be thought of as risk factors for psychosis. However, these links were identified retrospectively in groups of individuals with a psychotic disorder, so it is not clear by how much they might increase risk in the general population or when observed in one individual. Other risk factors include increased levels of adversity in the environment of an individual, presumably increasing stress and/or decreasing access to resources or to other protective mechanisms during development. Higher incidence of psychosis has been found in developing countries (March et al., 2008), groups raised in urban areas (Cantor-Graae & Selten., 2005; Pedersen & Mortensen, 2001), and minority groups that either migrated to a new area (Boydell et  al., 2001) or live amidst majority groups (Tienari, Wynne, & Sorri, 2006). Family factors that are related to communication (such

2  Assessment of Risk for Psychosis

as communication deviance) and levels of hostility and criticism (e.g., “expressed emotion”) have also been linked to increased risk for relapse or later psychosis (Hooley, Woodberry, & Ferriter, 2005; Krabbendam & Van Os, 2005). The specific mechanisms responsible for the link between these environmental risk factors/indicators and psychosis continue to be debated (Torrey et al., 1997), though most agree that they interact in complex ways with biological factors (Andreasson, Engström, Allebeck, & Rydberg, 1987; Bernardo et  al., 2017; de Castro-Catala et al., 2015; Howes et al., 2004; Tsuang, 2000). Finally, a few environmental factors contribute to risk as illness develops and may play a role in catalyzing pre-existing vulnerability. Stress is a good example (Walker & Diforio, 1997). In addition, substance abuse may precipitate a first or subsequent psychotic episode, particularly cannabis and stimulants (Arseneault, Cannon, Witton, & Murray, 2004; Gururajan, Manning, Klug, & van den Buuse, 2012; Marconi, Di Forti, Lewis, Murray, & Vassos, 2016; Murray and Di Fori (2016). Substance use earlier in life has also been linked with later development of psychosis (Addington et al., 2014). For example, in the first phase of NAPLS (N = 370), individuals with an APS who also used, abused, or met criteria for cannabis dependence had higher rates of conversion and converted sooner than nonusers (Auther et al., 2015). Outcome and transition to psychosis may be more strongly associated with beginning cannabis use earlier in life, more frequent use, and continued use during the APS phase rather than overall lifetime use (Valmaggia et al., 2014). Similar to individual genes, specific environmental risk factors and indicators vary in how strongly they are associated with later psychosis. Like substance use, in isolation they also tend not to be terribly helpful in identifying those who are at highest risk for psychosis. However, when combined with other markers of risk, they may increase probabilities of subsequent illness. Researchers are continuing to work on developing empirically based models that might yield more quantifiable estimates of environmentally mediated risk.

21

5.1.3 Cognitive Risk Factors Cognition is affected both by biological factors and learning/education opportunities. Cognitive studies in the APS population have consistently demonstrated small-to-medium impairments across most cognitive domains—attention/vigilance, working memory, processing speed, visual learning, verbal learning, and reasoning (Fusar-­ Poli et al., 2012; Guiliano et al., 2012). Moreover, individuals with APS who convert to psychosis show more severe cognitive deficits at baseline than non-converters in most domains, particularly attention, executive functions, and memory (Seidman et  al., 2016). Psychiatric interventions (medication and psychotherapy; see Chap. 3 of this volume) and nutritional supplements that protect brain function (omega-3) (Amminger et  al., 2010) show some evidence of improving subthreshold psychotic symptoms or development of psychosis. Notably, cognitive performance in individuals with an APS tends not to improve with these interventions and remains impaired during illness progression, predicting poor occupational and functional outcome and conversion to psychosis (Green, Kern, & Heaton, 2004). As discussed later, cognitive functions, particularly declarative memory and executive function, do strongly predict progression from APS to psychosis when combined with clinical syndromes (Cannon et al., 2016). 5.1.4 Neurobiological Risk Factors Neurobiological factors contributing to schizophrenia have received perhaps the most attention to date. Brain deficits that include both structural and functional brain impairments have been observed in both first-episode and chronic schizophrenia and, relevant for this review, long before psychosis development. This means that such deficits, when present in the at-risk phase, are risk indicators. However, establishing neurobiological risk factors has been a very complex undertaking given brain complexity and its relationship to symptoms. Advances in technology have ushered tremendous progress in unravelling brain-symptoms relationships. Still, their exact nature is a subject of intense research and debate, making the task of utilizing specific neurobio-

D. I. Shapiro et al.

22

logical risk indicators in determining the level of risk at an individual level a complicated undertaking. The rationale for studying APS groups both cross-sectionally and longitudinally is that such studies provide opportunities to identify causal mechanisms of psychosis development. If such mechanisms are present in those who develop psychosis and absent in those who do not, they can be considered risk factors and observable signs. Often, the distinction is made between endophenotypic traits—observable signs that are believed to be genetically mediated—and specific biomarkers that are associated more with illness progression and are not genetically mediated (e.g., Gottesman & Gould, 2003; Pantelis et al., 2009). They would be considered risk indicators and risk factors, respectively. The findings from neurobiological studies have confirmed that structural and functional abnormalities exist before individuals experience the first psychotic break and that these are thus markers of risk. In addition, evidence is accumulating that even individuals who are at risk but do not convert to psychosis often remain clinically symptomatic and impaired in their neurobiological and thus cognitive function. The question then remains what would be appropriate interventions for those deemed at risk. As with cognitive and clinical risk factors, the hope is that if biological risk factors are found, they could be targeted for intervention. Results from studies examining the integrity of gray and white matter (neurons and the axons that interconnect them, respectively) and of brain function using both functional magnetic imaging (fMRI) and event-related ­ potentials (ERPs) have started to identify both risk factors associated with development of psychosis and to describe brain states that accompany enduring subthreshold psychotic symptoms without the conversion to psychosis. The review of brain changes that precipitate the onset of psychosis has been segregated into studies that examined the brain structural integrity, including the connectivity between brain regions provided by white matter fibers, or tracts, and those that examined brain functional integ-

rity. The premise behind all imaging studies of risk factors for psychosis is that differences in brain structure and function manifest as cognitive impairments and clinical symptomatology. Thus, identifying neurobiological early signs is tantamount to identifying underlying factors that lead to psychosis. While complete characterization of overall brain abnormality that produces psychosis is the ultimate goal of neuroimaging studies, brain complexity necessitates somewhat separate research undertakings. The brain can be abnormal in many ways: the volume of neural tissue may be reduced in particular regions, the connections between these regions may be abnormal, and the function of specific brain regions as related to specific cognitive operations may be impaired. As described below, certain specific brain abnormalities confer risk of psychosis. The structural integrity of the brain has been examined with structural magnetic imaging (sMRI) to yield measures of brain region volume and of cortical thickness, as well as white matter tract integrity using diffusion tensor imaging (DTI). Brain functional studies focused on cognition employ functional MRI (fMRI) that probes activation in brain regions involved in a particular cognitive task and EEG/ERP that probe specific cognitive operations related to activity of neuronal assemblies expressed as either “brainwaves” or as oscillations. Most imaging studies reviewed below were conducted with individuals identified with a clinical high-risk and genetic or familial high-risk syndrome APS (CHR/UHR). Structural and Functional Brain Imaging Studies Gray Matter Volumetric Studies

Brain regions identified as abnormal in high-risk individuals include both cortical and subcortical structures. The cortical structures figuring prominently in reports on brain volumetric changes prior to psychosis onset include frontal and temporal lobes. Pantelis et  al. (2009) reviewed the early structural findings distinguishing between endophenotypes and biomarkers and concluded

2  Assessment of Risk for Psychosis

that most regions reported in these early studies were biomarkers rather than endophenotypes. According to Pantelis and coauthors, the frontal regions were likely endophenotypic, while the temporal regions were biomarkers for psychosis. The Edinburgh longitudinal high-risk study (McIntosh et al., 2011) reported that the volumes of the left and right prefrontal lobe and the temporal lobes were smaller in a CHR APS group relative to healthy controls; greater volume reductions were observed in those who converted to frank psychosis, with the reductions positively associated with the severity of psychotic features. Within the prefrontal regions, pars triangularis specifically was reduced in both the CHR and the first-episode schizophrenia groups relative to their healthy comparison subjects with the extent of the reduction correlated with the severity of psychosis symptoms (Iwashiro et  al. (2012). Finally, volume reductions in the anterior cingulate cortex (ACC) have been reported prior to psychosis onset (Fornito, Yücel, Dean, Wood, & Pantelis, 2008). Cortical thickness, another measure of volumetric integrity, was also found to be affected in those with a CHR APS: its reductions were associated with metacognition abnormalities (Buchy, Stowkowy, MacMaster, Nyman, & Addington, 2015) and with the worsening of subthreshold psychotic symptoms (Cannon et al., 2015; Tognin et al., 2014). Studies focusing on limbic structures present a mixed picture with some studies suggesting volume reductions in hippocampal-amygdaloid complex (Bois et al., 2016), while other studies do not find group differences in the hippocampus between those with a CHR APS who convert to psychosis and non-converters (Walter et  al., 2012). The meta-analysis of the hippocampal findings (Walter et al., 2016) suggests a lack of hippocampal volume reduction in CHR.  It is likely that large, longitudinally designed studies are needed to resolve the issue of the possible contribution of abnormal hippocampal development at the prodromal stage for schizophrenia. To summarize, in terms of brain structural abnormalities that appear to confer risk for psychosis, temporal and frontal structures are most

23

consistently reported across studies, while subcortical structural abnormalities including hippocampal-­amygdaloid complex are receiving less consistent support as risk factors for psychosis in studies that examine differences between those with an APS who develop psychosis and those who do not. However, one can see that methods for identifying these risk factors vary and in some cases the specific structures that are implicated differ depending on the methods used to study them. For this reason, researchers have utilized a multitude of methodologies to identify or measure commonly underlying brain mechanisms that constitute risk factors for psychosis. Having altered brain structures may only be a part of the picture. Structural and Functional Connectivity Studies While studies examining changes in the volume of different brain regions in APS have focused on the anatomical architecture of selected brain regions, several studies have focused on the integrity of connections between brain regions to examine whether factors related to how different brain regions communicate with each other might confer risk. Structural Connectivity

Several studies using diffusion tensor imaging (DTI), which allows for the characterization of the integrity of white matter tracks—the structural connections within and between brain regions—found that measures of white matter integrity such as fractional anisotropy (FA) were lower in those with an APS than in healthy control individuals (Bloemen et  al., 2010; Francis et  al., 2013; Hoptman et  al., 2008; Karlsgodt, Niendam, Bearden, & Cannon, 2009; Clemm von Hohenberg et  al., 2013). However, unlike the gray matter findings, the white matter connectivity findings lack strong evidence to indicate that these connectivity changes distinguish between converters and non-converters, i.e., are risk factors for developing psychosis. There is also evidence of reduced cortical-­ subcortical (thalamo-orbitofrontal) connectivity in APS (CHR) as well as in first-episode schizo-

24

phrenia (FE) (Cho et al., 2015). This result suggests that structural thalamocortical abnormalities are a common feature of CHR and FE stages which are differentiated by the degree and not by the type of disconnectivity. Thus, such reductions in connectivity are indicative of risk for psychosis, though no study has examined whether this is still the case when identified in those who do not also meet criteria for a CHR APS. Functional Connectivity

Studies focusing on functional connectivity, or the patterns of co-activation of different brain regions over time, identified several abnormalities between brain regions already discussed as having compromised function in schizophrenia. They include abnormal connectivity between frontal and temporal brain regions (Jung et  al., 2012; Tijms et  al., 2015), as well as abnormal connectivity within and between default mode network (DMN) brain structures (Amico et  al., 2017; Clark et  al., 2018; Whitfield-Gabrieli & Ford, 2012). The core of the DMN network includes midline brain structures including medial prefrontal cortex (mPFC) and anterior and posterior cingulate cortex which are involved in language, self-reflection, and distinctions between self and other. Abnormal connectivity between these regions contributes to positive symptoms (auditory hallucinations and delusions). Hallucinations and delusions are types of positive symptoms. Thus, it appears that not only are structural abnormalities in frontal and temporal regions risk factors for psychosis, but they may lead to dysfunction in how these brain areas interact with each other, which could be m ­ easured by less invasive and expensive techniques like EEG. As in the case of structural connectivity impairment, several studies have found impaired functional connectivity in the limbic and subcortical structures that comprise the amygdala and thalamus (Anticevic et al., 2014, 2015; Bernard, Orr, & Mittal, 2017; Colibazzi et al., 2017). Thus, as was the case of the brain structural abnormalities, there is a compelling evidence that functional connectivity abnormalities within and between the temporal and frontal brain regions,

D. I. Shapiro et al.

as well as in limbic-cortical connections, confer risk for developing psychosis. Functional MRI Studies Finally, functional MRI studies exploring cognitive processes impacted in schizophrenia have generally found compromised brain activation in individuals with an APS, where often a worse clinical outcome was associated with a worse cognitive performance as captured by fMRI signal. This suggests that abnormal fMRI signal associated with these cognitive processes may be an indicator of psychosis risk. Not surprisingly, abnormal fMRI results have been found on tasks involving language (Li et  al., 2016; Natsubori et  al., 2014; Niendam et  al., 2014; Sabb et  al., 2010), working memory (Li et  al., 2016; Thermenos et  al., 2016), theory of mind and social cognition (Marjoram et al., 2006; Takano et a., 2017), as well as emotion processing in APS (CHR), underscoring the fact that poor cognitive performance is intricately tied to brain function. Furthermore, the brain regions involved in these processes, and found abnormal in fMRI studies, are by and large similar brain regions as have been identified as abnormal on structural MRI.  Examples include abnormalities in brain regions implicated in different aspects of language processing that included medial prefrontal cortex bilaterally, left inferior frontal gyrus (lIFG), middle temporal gyrus, and the anterior cingulate gyrus (Sabb et  al., 2010). Those individuals with APS who later transitioned to psychosis had higher activation levels in the STG, caudate, and LIFG.  Furthermore, signal change in the lIFG, SFG, and MTG correlated with the severity of thought disorder at follow-up in the APS group, while social outcome was correlated inversely with the signal change in the lIFG and ACC. This means that abnormal function of these regions may be indicative of risk for psychosis when observed in those with an APS. In assessing the evidence from fMRI studies for predictors of psychosis, two pieces of evidence stand out: the brain regions which are functionally affected predominantly include the temporal and frontal regions. Cognitive functions that are disrupted in APS are higher-order cogni-

2  Assessment of Risk for Psychosis

tion such as language, cognitive control, working memory, theory of mind, declarative memory, and social cognition, tasks which are heavily reliant on the function of these brain regions. Thus, these cognitive functions appear to be indicators of risk, and specific aspects of the structure and function of these regions may be the underlying mechanisms associated with later development of psychosis. These findings are important because they suggest a link between more easily identifiable indicators of risk, namely, cognitive functions that can be identified with paper and pencil tests and specific brain mechanisms that could be the target of interventions. The link between brain mechanisms and cognitive phenotype also has implications for assessment of risk: many factors contribute to poor cognitive performance on testing, similarly to what is found in the general population. Given the cost and complexity of currently available imaging tools, it certainly would not be clinically useful to use brain performance as a litmus test for psychosis risk. However, in the context of other risk factors or indicators for psychosis, poor cognitive performance on behavioral tests could be used as a reason to conduct more intensive and costly assessment of brain structure and function, such as those discussed herein. If those, too, identify characteristic dysfunction, then it may raise confidence that one is at elevated risk for psychosis. EEG and Event-Related Potential (ERP) Studies While structural and functional imaging studies have identified which brain regions and their connections are abnormal in APS and psychotic ­disorders, EEG methodology is focused on abnormal neurocognitive processes. EEG methodology is the only imaging modality that allows for imaging of the underpinnings of neurocognition with millisecond resolution and accuracy. It is also less expensive than (f)MRI and thus more accessible in clinical contexts. As in the study of structural and functional brain abnormalities described above, the strategy for looking at neurocognitive brain abnormalities in APS is to focus on the brain processes which have been demonstrated to be abnormal in frank psychosis. ERP components

25

are deflections in the brain waveform, a recording of electrical activity produced by neurons and is characterized by the latency at which fluctuations appear relative to a stimulus onset and by the amplitude of their electrical output. Different ERP components are associated with different functional properties, i.e., index unique cognitive operations such as sensory processes, short-term sensory memory, working memory, and attention. It has been proposed that abnormalities in these early sensory processes as well as in attention contribute in a significant way to the development of psychosis and that abnormalities identified in these processes are risk factors for psychosis (Javitt & Sweet, 2015). Reductions in ERP components’ amplitude are evidence of an abnormal cognitive process, while latency prolongations are interpreted as evidence of a slower but otherwise unaffected cognitive process. Mismatch Negativity

Mismatch negativity (MMN) is a negative going ERP component associated with pre-attentive, automatic processing (e.g., Paavilainen, 2013). In simplified terms, it occurs when an individual perceives an unexpected item in a sequence, like when a chirp is heard or a square is seen after a long string of buzzes or circles, respectively. It has been used to study the function of both auditory and visual sensory memory functions, and more recently it has been used to examine prediction processes. Abnormalities in MMN.  Reduced auditory MMN amplitude has been found in both chronic and first-episode stages, and the cognitive processes indexed by MMN were also found abnormal in APS. Several studies have used frequency (where the frequency of a rare stimulus is manipulated) and/or duration (where the duration of a rare stimulus is manipulated) auditory MMN to examine pre-attentive processes in APS as predictors of conversion to psychosis (e.g., a marker of risk) (Atkinson et  al., 2017; Bodatsch et  al., 2011; Bruggemann et  al., 2013; Carrion et  al., 2015; Ericson, Ruffle, & Gold, 2016; Hsieh et al., 2012; Kim, Cho, Yoon, Lee, & Kwon, & Kwon, 2017; Light and Swerdlow, 2015; Nagai et  al., 2013; Perez et  al., 2014; Solis-Vivanco et  al.,

D. I. Shapiro et al.

26

2014). In several of these studies, reduced duration MMN was identified in APS (CHR) groups (Bodatsch et al., 2011; Hsieh et al., 2012; Perez et al., 2014; Solis-Vivano et al., 2014), with some of these studies finding group differences between converters and non-converters. Other studies reported group differences between CHR and HC but not differences between converters and non-converters (e.g., Hsieh et  al., 2012; Solis-Vivano et  al., 2014). A meta-analysis of MMN studies suggested that the component is a biomarker for Attenuated Psychosis Syndromes rather than an index of genetic risk for psychosis development (Erickson et al., 2016), with a Light et al. (2015) review and Kim et al.’s (2015) study pointing out its usefulness in tracking clinical treatment progress. On the other hand, the longitudinal Minds in Transition study results (Atkinson et al., 2017) suggested an absence of group differences in the MMN signal. In addition to processes indexed by MMN, early sensory processes both in visual and auditory modalities have been found abnormal in those with an APS. This suggests that such early brain processing difficulties could be used as a marker of risk given their likely involvement in the cognitive deficits associated with psychosis. Early ERP Responses

N1 and P1. N1 and P1 are ERP components whose peaks are maximal in the window that follows roughly 100 ms after a stimulus onset, and they index early sensory processes. Abnormal sensory processes. In normal individuals N1 is smaller while listening to ­self-­generated speech. This attenuation is associated with the corollary discharge mechanism signaling to the brain that the action was self-generated and not coming from outside: this reduction helps distinguish between self- and nonself-speech. This mechanism has been found abnormal in individuals with a CHR syndrome (Gonzales-­Heydrich et  al., 2016; Perez et  al., 2013). A magno-encephalography (MEG) study of the M100 response in APS (CHR) conducted in conjunction with the structural study of the Heschl gyrus and planum temporale found reduced M100 to be related to the thinning of

these two structures (Shin, Jung, & Kim, 2012). Similarly, reductions in the N100 distinguished between CHR and first-episode individuals in the study examining both sensory-driven and attention-­driven processes (Del Re et al., 2015). The degree of N100 reduction was associated with severity of positive symptomatology (Gonzales-Heydrich et al., 2015) in CHR. Given the limited number of studies, the evidence for these two components to be predictors for psychosis is still debated. P300

P300 oddball and P300 novel are two ERP components that are sensitive to attentional and working memory demands (P300 oddball) and to salience and orienting attention and novelty (P300 novel). They arise from temporal, limbic, and parietal cortices but also cingulate (P300 oddball) and from ACC and frontal and parietal cortices (P300 novel) (Volpe et al., 2007), that is, from brain regions found abnormal in several studies of APS as discussed above. Abnormalities in P300. As in the case of pre-­ attentive and sensory processes, attentional processes have been found to be abnormal in schizophrenia. Most studies have found reduced P300 in APS (CHR syndromes). In some studies, P300 was shown to distinguish converters from non-converters. P300 novel was found reduced in individuals genetically related to schizophrenia patients (Turetsky, Cannon, & Gur, 2000). Also, reduced P3 novel in CHR was reported in the Mondragon-Maya et al.’s (2013) study even though no group differences were found between CHR and first-episode schizophrenia patients. On the other hand, Atkinson et al. (2017) did not find P3 novel group differences between a group of CHR individuals and their healthy controls. Frommann et al. (2008) found that late stage prodromal individuals had more widespread P300 oddball abnormalities than early stage individuals. Kim et al. (2017) found that even though there was no group difference in the P300 amplitude between converters and non-converters, the P300 at parietal sites was predictive of clinical improvement. The attenuated P300 oddball was

2  Assessment of Risk for Psychosis

27

also found in first-degree relatives of schizophrenia patients. The number of P300 studies in APS (CHR) individuals is still limited. The available evidence suggests that P300 is a good predictor of which individuals recover from clinical symptoms and which ones remain symptomatic and can distinguish between those who will convert to psychosis and those who will not. Thus, P300 is a promising risk predictor and can be used in the clinic in conjunction with other risk assessment measures.

individuals at high risk for psychosis albeit to a much lower degree. Abnormalities of late but not of early latency gamma band ASSR were found in APS (ultra-high-risk syndromes) (Tada et al., 2014). In spite of the fact that oscillatory processes are essential for normal cognition, research on oscillatory abnormalities in CHR is limited, and thus it is not clear if oscillatory abnormalities can be used as predictors of risk for psychosis at this time.

Oscillatory Processes

Summary of the EEG/ERP Research ERP/EEG studies on APS are somewhat limited in number, and thus strong statements about their findings warrant caution. However, as these studies provide far superior and precise measures of mechanisms impacted by the disease process relative to both traditional neuropsychological and fMRI modalities, they can be important compliments to assessment. This is especially so given accumulating evidence suggesting that these measures can both predict who does and does not convert from an APS to psychosis but also who will and will not be symptomatic (e.g., remit) a year and more later relative to the initial assessment. Implications of such findings are discussed in greater detail below. As new technological advances will make these tools more user friendly, they can be an important indicator of which clinical interventions will be most appropriate.

Neural oscillations play a critical role in cognition by coordinating activity across brain regions and within brain regions (e.g., Ramyead et  al., 2014). For example, it is believed that they contribute to functional connectivity between and within brain regions. They are rhythmic patterns of neural activity generated by neurons forming functional assemblies and are characterized by their frequency of oscillation (measured in Hertz). Alpha, beta, and especially gamma band oscillations (their names related to the magnitude of dominant frequency) have been examined in several studies of individuals with APS (CHR) and suggested impairments in the processes orchestrating cognition. For example, using measures of both spectral power and inter-trial coherence, Koh et  al. (2011) examined alpha band oscillations which are involved in mediating attentional processes in first-episode schizophrenia, CHR, and healthy controls. CHR individuals showed reductions in alpha power synchronization, involved in mediating attentional processes, relative to HC suggesting that deficits in attentional control exist already in the CHR group. Similarly, decreased phase synchronicity in the beta band was identified in individuals with a CHR syndrome who later transitioned to psychosis relative to HC and those who did not transition (Ramyead et  al., 2014). Abnormal gamma band auditory steady-state response (ASSR), especially to 40 Hz auditory stimuli, is one of the best documented oscillatory abnormalities in schizophrenia, suggesting GABAergic interneuron dysfunction. This abnormality exists already in

5.1.5 O  verview of Biological Risk Factors Where We Are Now and Where We Need to Go Imaging studies reviewed above clearly demonstrate that brains of individuals with an APS are different from those of matched comparison individuals. These differences are considered markers of risk for developing psychosis and, as discussed below, have particular utility when these differences are related to incidence of subsequent psychosis. The brain differences are multiple. They manifest as reduced volume of gray matter, reflecting neuronal, axonal, and dendritic loss in the impacted regions. The brain regions affected by these changes are most consistently localized to the frontal and temporal

D. I. Shapiro et al.

28

areas, brain regions that are strongly involved in the common neurocognitive difficulties seen in APS and psychotic disorders, but also subcortical structures such as thalamus are impacted. There are also marked differences in white matter tracts that connect particular brain regions in high-risk individuals. Again, the two tracts most prominently impacted by the disease process are those that connect the frontal and temporal as well as parietal brain regions, particularly those involved in language function. In addition to differences in brain structure identified in research on APS, functional MRI studies in APS point to abnormalities in brain function during cognitive processes affected in psychosis including language, working memory, social cognition, and theory of mind. The brain regions that are primarily involved include again the frontal and temporal regions. ERP and EEG studies add to this evidence by pointing to sensory, pre-attentive, short-term memory and attentional processes impacted in this group of individuals. While the neural sources of these processes are harder to identify, several lines of evidence again suggest the involvement of temporal and frontal brain regions. In addition to the integrity of both gray and white matter and their connections for normal cognition to occur, neurotransmitter levels, again in turn dependent partly on the integrity of the brain tissue, contribute to clinical symptoms and cognitive impairments seen in APS. Considered together, these results suggest interactions between tissue integrity and neurochemical imbalance as well as abnormal structural and functional connectivity in key brain structures in APS.  As reviewed above some of these impairments are sensitive to conversion to psychosis, while others are not, meaning that we can have increased confidence that they are indicative of risk for psychosis when identified in an individual. As reviewed above, some abnormalities appear in all individuals with an APS and are not sensitive to conversion, while others distinguish between those who transition to psychosis and those who did not. Thus, if the focus on putative interventions is on preventing individuals from

transitioning to psychosis, they should be designed to target these brain regions and functions that distinguish between converters and non-converters. If the goal of the intervention is to improve cognition and reduce symptoms both in those who converted and those who did not but remain symptomatic, then a broader goal of targeting all brain regions found abnormal in APS should be pursued. It appears that large, well-powered studies in addition to comprehensive statistical approaches are needed to better establish the clinical significance of these neurobiological indicators of risk and how they might contribute to efforts to quantify or categorize risk in an individual. Such approaches might include developing machine learning algorithms that are capable of predicting group membership based on one or more biological indicators. Currently a diagnosis of Attenuated Psychosis Syndrome is largely rendered based on clinical symptoms. Developing strong associations among clinical and cognitive profiles as well as specific neurobiological indices would allow for a more nuanced understanding of an individual’s unique risk profiles and thus help craft more precise treatments aimed at specific mechanisms that underlie psychosis and risk for its development. Finally, in spite of the overwhelming evidence for the contribution of neurobiological factors to the development of clinical symptoms in APS, whatever trajectory follows will ultimately unfold in the context of family, culture, and society, each of which exerts its own impact on what may develop.

5.2

 aution in Evaluating Risk C Markers in APS Groups

The findings discussed above have slightly different implications for the identification of psychosis risk than for intervention. The following sections will discuss methods for estimating risk quantitatively by combining individual risk markers into groupings that maximize risk estimates and psychosis prediction. In general, these all involve adding other predictors to an identified APS based on

2  Assessment of Risk for Psychosis

research that identifies risk factors within samples already identified as being at high risk (e.g., meeting APS criteria). Any such grouping is by nature heterogeneous. Because risk groupings like APS force a categorical taxonomy (e.g., at high risk or not) on something that is dynamic, continuous, and multidimensional, such groupings are by nature imprecise and may falsely capture some individuals who are on some other developmental trajectory (see van Os & Guloksuz, 2017 for full discussion). Some proportion of cases are true positives, meaning that identified markers of risk truly tap an underlying psychotic illness that is beginning to unfold. Other cases may be false positives who show markers of risk, but perhaps they are due to other underlying causes, are counteracted by unknown (or poorly understood) protective factors, or are at risk but not to a sufficient degree to progress in severity. There is also likely another group of “false false positives” who do not develop psychosis but would have had they not presented for care (Yung et al., 2007). Thus, while research on atypical biological features identified in cohorts of people with APS or other risk groupings may elucidate underlying mechanisms at play before psychosis develops (risk factors), caution is needed before these features can be assumed to confer quantifiable risk. If such groupings are themselves non-specific and heterogeneous, we may be “keying” to a pleiotropic risk state, rather than to psychosis more specifically. The solution is to examine features that specifically predict psychosis (e.g., distinguish between those at-risk individuals who develop psychosis from those who do not), but this is difficult because doing so requires very large samples of individuals with enriched risk and sufficiently long longitudinal follow-ups to know true clinical outcomes. Because these large-scale studies are difficult, time-consuming, and expensive to conduct, we often must fall back on making inferences based on research in smaller APS samples or on characteristics of those with an APS rather than confirmed prodrome. Therefore, we must acknowledge that any specific markers of risk identified therein introduce some additional uncertainty and potential measurement error into our attempts to measure and quantify risk. One solution has been to include all possible predictors

29

and use a statistical method of prediction, in which mathematical models are used to identify which combinations of risk factors yield the algorithm that best predicts outcome, then apply this algorithm to new samples or, potentially, to individuals.

5.3

 utting Markers of Risk P Together to Assess Level of Risk

5.3.1 Guiding Principles This chapter has thus far discussed individual markers of risk for psychosis by summarizing research identifying specific environmental, genetic, biological, or clinical factors that are associated with the development of psychosis. In contrast to what is implied by a “diagnosis” of APS, risk for psychosis is not dichotomous. Individuals likely vary along a continuum that balances environmental, heritable, and other acquired risk factors with protective factors that may vary or interact differentially at different points in time. The field has thus attempted to use research conducted in large groups to estimate risk in individuals—quantifying some latent risk into probabilities based on rates of psychosis in others who possess the same markers and indicators of risk. In evaluating the utility of markers of risk at the individual level, the goal is to maximize statistical sensitivity—here the ability to identify future psychosis when it occurs—and specificity—the ability to rule out future psychosis when it does not subsequently occur. These are used to compute a positive predictive value (PPV), which can be expressed as the proportion of individuals identified as at risk for psychosis who go on to develop a psychotic disorder (see Fusar-Poli et al., 2016, for a more thorough review of prediction statistics and terminology). PPVs of 0.8 and above are generally considered very strong. Statistics such as these rely on the comparison group, meaning that they compute differently in studies that focus on conversion in high-risk groups vs. in the general population, and often this work is iterative as factors identified in more

30

homogeneous samples are applied to broader populations or new samples. To date, this research has suggested that gross measures of latent risk that capture more downstream effects of what are likely many causal factors, like APS or significant family history of psychosis, perform far better than any specific risk factors or indicators, both statistically and in their feasible application in routine clinical practice. Secondly, as illustrated most clearly in our review of brain mechanisms and genetics, efforts to identify specific biomechanism-based risk factors have yielded mixed results, suggesting that the same mechanisms may not be involved in psychosis in every case. Indeed, disorders involving psychosis are likely characterized by equifinality. Therefore, major efforts in prediction of psychosis have narrowed in on ways to combine individual markers of risk to maximize PPVs. Thirdly, as shown below, the best predictive results come when APS is combined with risk markers, though such models are not yet ready for clinical use.

5.3.2 Methods of Combining Psychosis-Risk Markers: Maximizing PPV APS as a Statistical Predictor A comprehensive meta-analysis conducted by Fusar-Poli et  al. (2012) examined the predictive ability of predominant UHR, CHR, and basic symptom APS, including genetic risk groups. Collapsing across APS type across 27 studies (and roughly 2500 APS cases), these authors found a rate of psychosis of 18, 22, 29, and 36% after 6-, 12-, 24-, and 36-month follow-ups, with an average conversion rate of 29.2% across all studies. Basic symptoms syndromes had a 48.5% rate, but outcomes in specific studies varied widely. This means that absent other indicators, having an APS can be said to confer a roughly 29% probability of developing psychosis within 2–3 years of identification, with the greatest proportion of conversions occurring within the 1st year after identification. However, it is important to note that very little research has examined conversion rates beyond roughly 3 years (Klosterkotter et al., 2001; Reichler-Rossler et al., 2009). A sub-

D. I. Shapiro et al.

sequent meta-analysis by the same group found that within the CHR APS, those with a Brief Intermittent Psychosis Syndrome (BIPS) APS had the highest rates of conversion (39% by 24  months) followed by Attenuated Positive Symptom Syndrome (APSS) (19%). Those with a Genetic Risk and Deterioration Syndrome had risk no greater than those evaluated for a suspected APS but who did not meet criteria upon assessment (Fusar-Poli et al., 2016). No individual genetic or biomarker has achieved better PPV than estimated rate of conversion derived from the presence of an APS alone. Detailed summary of predictive metrics for studies examining conversion in CHR can be found elsewhere (Ruhrmann et al., 2010; Schmidt et  al., 2017), and a useful guide for how to use these in APS assessment is summarized by Fusar-­ Poli and Schultze-Lutter (2016).

5.3.3 Combining APS with Biomarkers and Other Predictors of Outcomes As noted, the APS label is a probabilistic designation that is likely given to people with different developmental trajectories, which may themselves be affected as a result of the labeling process. While CHR syndromes remain the primary categorical determinants of significant risk, the above results indicate that a majority of those thusly labeled will likely not develop a psychotic disorder and a substantial proportion will improve with no treatment at all (Simon et  al., 2011). Thus, identification of who among this heterogeneous population is most at risk is imperative for targeted prevention strategies. To this end, major efforts are underway to develop risk stratification algorithms that could indicate level of risk at the individual level based on the presence or absence of different combinations of risk factors known to be associated with psychosis onset (Ruhrmann et al., 2014). Such algorithms could be the basis for staged treatment methods, as discussed in Chap. 3 of this volume, while also providing differentially targeted interventions for those less likely to develop psychosis but who still experience symptoms and distress. Thus, future identification of APS may be followed by additional

2  Assessment of Risk for Psychosis

clinical assessment, neuropsychological testing, EEG, bloodwork, and/or MRI, with indicators from each contributing to a quantified estimation of risk (Fusar-Poli et al., 2013). Many of the studies that have examined conversion rates have also reported clinical predictors of psychosis within their APS samples (see summary in Ruhrmann et al., 2010). Algorithms derived from these studies to optimize positive prediction of conversion to a psychotic disorder have typically demonstrated PPVs around 80% (Cannon et  al., 2008; Schultze-Lutter et  al., 2014; Thompson, Nelson, & Yung, 2011). For example, investigators in the North America Prodrome Longitudinal Study found that a combination of CHR criteria plus three to five additional risk indicators (genetic risk with recent deterioration, elevated unusual thought content, elevated suspiciousness/paranoia, social impairment, substance abuse history) increased PPV to 68–80% (Cannon et al., 2008) from that of APS (CHR) alone. At Orygen in Australia, APS (UHR) plus genetic risk and deterioration, symptoms present for over 5 years, poor global functioning, and inattention increased PPVs to roughly 81% (Yung, Phillips, Yuen, & McGorry, 2004). Similarly, in the European Prediction of Psychosis Study, PPV was increased to 83.3% in those with either CHR or COGDIS when overall subthreshold positive symptoms, bizarre thinking, sleep disturbance, schizotypal personality disorder, level of recent functioning, and/or lower education were added (Ruhrmann, Schultze-Lutter, & Klosterkötter, 2010). In only one case has an algorithm developed in one large sample been replicated by a different sample in a different clinic or research center (see below). This suggests that such algorithms are not yet ready for use in clinical settings but that the presence of some additional markers from a second domain might increase confidence that somebody with an identified APS is truly at risk. To explore this possibility, researchers have also explored the possibility of including biomarkers to improve risk algorithms. In the FePsy study, Koutsouleris et  al. (2012) found that neuroanatomical patterns derived from an APS sample, along with clinical risk indicators, improved pre-

31

diction of psychosis in two novel samples, correctly identifying roughly 80% of converters. Neurocognition has also been explored. Multiple research groups have shown that verbal learning/ memory and executive functioning add incremental predictive power to APS indicators, again resulting in PPVs around 80% (Cannon et  al., 2016; Koustouleris et al., 2012; Reicher-Rossler et al., 2009; Seidman et al., 2016). A risk stratification algorithm based on combinations of neurocognitive, clinical, premorbid adjustment, and ERP measures (parietal P300 amplitudes) increased PPV to 70% (Nieman et  al., 2013). Other promising biomarkers that may indicate greater risk when combined use with APS status include markers of neuroinflammation, oxidative stress, and dysregulation of the biological stress response system (Perkins et  al., 2014; Walker et al., 2013). Summarizing this literature, Schmidt et  al. (2017) found 25 studies published through 2015 that combined APS status with other clinical, biological, neurocognitive, or environmental predictors to develop a predictive algorithm. They limited their review to studies in which algorithms were also validated in a new sample. Ten studies used clinical factors in their models, five studies examined biomarkers, five used neurocognitive factors, and five investigated environmental predictors with APS. Only eight combined factors in multiple domains. Schmidt and coauthors then attempted to amalgamate these findings into a cohesive, stepwise model of risk prognostication that would maximize predictive values. They aggregated data and developed competing statistical models and found that the best performing one yielded a PPV of 98% when three positive tests were met including APS plus EEG markers, structural MRI, and blood markers. PPV = 71–82% was achieved for two positive tests. This algorithm has not been replicated, but it demonstrates what an algorithm approach might look like. Machine learning techniques are another promising arena for developing predictive models. The field is likely heading toward the development of risk calculators that use these algorithms to yield individualized quantitative risk estimates,

D. I. Shapiro et al.

32

such as have been created for multiple cancers, heart attacks, strokes, and poor surgical outcomes. Attempts thus far are encouraging, but such systems are closely tied to the characteristics of the samples from which they were derived, meaning replication, particularly in new cultures is imperative. One promising example is the risk calculator developed by the North American Prodromal Longitudinal Study (NAPLS) consortium and validated in two additional samples of patients meeting CHR APS criteria (Cannon et  al., 2016; Carrion et  al., 2016; Zhang et  al., 2018). This calculator, available at http://riskcalc. org/napls/, was based on retrospectively identified predictors of conversion in the NAPLS 2 study and provides psychosis probability scores for those meeting SIPS criteria. Clinicians can input age, neuropsychological test performance from the widely available and easily a­ dministered BACS symbol coding (executive function) and HVLT-R (verbal memory and strategy use), severity of unusual thoughts and paranoid ideas from the SIPS, number of stressful life events from the PERI Life Events Scale, change in GAF over the last year, and history of psychosis in a first-degree relative and then get an exact prediction estimate based on the NAPLS sample. Two attempts to replicate this calculator in new samples have been published. The first, also completed in North America with data from the multi-site EDIPPP study, found that the same predictors yielded a PPV of 0.79 and that the NAPLS-2 algorithm performed particularly well in predicting conversion when it yielded an estimated prediction likelihood of 55% or greater (Carrion et al., 2016). The Shanghai At Risk for Psychosis Program (SHARP) also attempted to replicate the NAPLS risk calculator in a Chinese sample in Shanghai (Zhang et  al., 2018) but found that it did not perform as well, resulting at best in PPVs of 41.3 when the NAPLS risk calculator indicated a conversion likelihood of 50% or more—still better than APS alone. Such calculators clearly need more work, which will come with increasing power and data points. These limited but promising results are leading to new large-scale efforts like EPINET, which aim to pool data across large numbers of early psychosis programs.

6

Conclusion

The goal of reducing or even preventing the burden associated with psychotic disorders is a public health endeavor. The more we learn about these illnesses, the more we recognize that hope of bringing this goal to fruition relies on our ability to recognize and identify those affected before illness has settled in. Indicated prevention approaches focused on identifying those at risk for illness or disability and intervening with increasingly intensive interventions are realistic and viable; their potential has ushered in a new era of research dedicated to identifying psychosis risk. At this moment in time, methods involving measurement of clinical indicators of risk—most prominently APS—are perhaps the most feasible approach and outperform any other prospective risk marker identification strategy. However, this paradigm has its limitations. Chief among them is a lack of precision in prognostication. Some imprecision comes from uncertainty innate to the process of categorizing or quantifying risk, when what actually underlies it is likely a dynamic interaction of ever-changing biological, psychological, and environmental factors that both increase and decrease the likelihood that clinical symptoms manifest. However, even if risk assessment is accurate, illness is not destiny. The process of labeling one as at risk— of naming a possible adversary—leads to alterations in underlying and hardly inevitable trajectories. This points to a second major limitation, which is a poor understanding of protective factors. While it is tempting to view protective factors as the opposite or absence of risk factors, surely social, cultural, and environmental phenomena play a part here. What we do in response to a health challenge (here the probability of having a mental illness) and who supports us in doing so affect outcome. The future of preventive efforts likely relies on two key factors. First, advances in technology are giving us new tools for understanding biological mechanisms. Researchers are already working to improve the accessibility of these tools and are exploring ways to combine all known markers of risk to yield individualized “calculators” of risk and, eventually, tailored interventions—pursuing

2  Assessment of Risk for Psychosis

a model of precision medicine. The second factor is the need to learn how different ways of understanding mental well-being and illness affect the ultimate outcome. This includes the roles of those affected by the illness and those who endeavor to help in different social and cultural contexts; how change in any one of these domains takes place may affect prevention efforts (in fact there is mounting evidence on how changes in social environment affect brain function directly). The APS paradigm originated in predominantly western cultures but is rapidly spreading to new cultural contexts. It is not clear which aspects of “risk” are universal and which may be more inherently shaped by culture. Regardless of future outcome, be it psychosis or something else, what is clear is that there is need for some intervention now for those who show signs of imminent mental difficulties. Differences in how we see ourselves, our world, and those around us most definitely shape who we see as in such need and what might be done about it.

References Addington, J., Case, N., Saleem, M.  M., Auther, A.  M., Cornblatt, B. A., & Cadenhead, K. S. (2014). Substance use in clinical high risk for psychosis: A review of the literature. Early Intervention in Psychiatry, 8(2), 104–112. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®). American Psychiatric Pub. Arlington, VA Amico, F., O’Hanlon, E., Kraft, D., Oertel-Knöchel, V., Clarke, M., Kelleher, I., … Power, E. (2017). Functional connectivity anomalies in adolescents with psychotic symptoms. PloS One, 12(1), e0169364. Amminger, G. P., Schäfer, M. R., Papageorgiou, K., Klier, C. M., Cotton, S. M., Harrigan, S. M., … Berger, G. E. (2010). Long-chain ω-3 fatty acids for indicated prevention of psychotic disorders: a randomized, placebo-­ controlled trial. Archives of General Psychiatry, 67(2), 146–154. Andréasson, S., Engström, A., Allebeck, P., & Rydberg, U. (1987). Cannabis and schizophrenia A longitudinal study of swedish conscripts. The Lancet, 330(8574), 1483–1486. Anticevic, A., Haut, K., Murray, J.  D., Repovs, G., Yang, G.  J., Diehl, C., … Cadenhead, K.  S. (2015). Association of thalamic dysconnectivity and conversion to psychosis in youth and young adults at elevated clinical risk. JAMA Psychiatry, 72(9), 882–891.

33 Anticevic, A., Tang, Y., Cho, Y. T., et al. (2014). Amygdala connectivity differs among chronic, early course, and individuals at risk for developing schizophrenia. Schizophrenia Bulletin, 40(5), 1105–1116. Arseneault, L., Cannon, M., Witton, J., & Murray, R. M. (2004). Causal association between cannabis and psychosis: Examination of the evidence. The British Journal of Psychiatry, 184(2), 110–117. Asarnow, J. R., Jaycox, L. H., Duan, N., LaBorde, A. P., Rea, M.  M., Murray, P., … Wells, K.  B. (2005). Effectiveness of a quality improvement intervention for adolescent depression in primary care clinics: A randomized controlled trial. JAMA, 293(3), 311–319. Atkinson, R.  J., Fulham, W.  R., Michie, P.  T., Ward, P.  B., Todd, J., Stain, H., … Schall, U. (2017). Electrophysiological, cognitive and clinical profiles of at-risk mental state: The longitudinal Minds in Transition (MinT) study. PloS One, 12(2), e0171657. Auther, A.  M., Cadenhead, K.  S., Carrión, R.  E., Addington, J., Bearden, C.  E., Cannon, T.  D., … Walker, E. F. (2015). Alcohol confounds relationship between cannabis misuse and psychosis conversion in a high-risk sample. Acta Psychiatrica Scandinavica, 132(1), 60–68. Badner, J. A., & Gershon, E. S. (2002). Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Molecular Psychiatry, 7(4), 405. Barrantes-Vidal, N., Gross, G.  M., Sheinbaum, T., Mitjavila, M., Ballespí, S., & Kwapil, T.  R. (2013). Positive and negative schizotypy are associated with prodromal and schizophrenia-spectrum symptoms. Schizophrenia Research, 145(1–3), 50–55. Bentall, R. P., de Sousa, P., Varese, F., Wickham, S., Sitko, K., Haarmans, M., & Read, J. (2014). From adversity to psychosis: Pathways and mechanisms from specific adversities to specific symptoms. Social Psychiatry and Psychiatric Epidemiology, 49(7), 1011–1022. Bernard, J.  A., Orr, J.  M., & Mittal, V.  A. (2017). Cerebello-thalamo-cortical networks predict positive symptom progression in individuals at ultra-high risk for psychosis. NeuroImage: Clinical, 14, 622–628. Bernardo, M., Bioque, M., Cabrera, B., Lobo, A., González-Pinto, A., Pina, L., … Vieta, E. (2017). Modelling gene-environment interaction in first episodes of psychosis. Schizophrenia Research, 189, 181–189. Bleuler, E. (1950). Dementia praecox or the group of schizophrenias.  Oxford, England: International Universities Press. Bloemen, O.  J. N., De Koning, M.  B., Schmitz, N., Nieman, D.  H., Becker, H.  E., de Haan, L., … van Amelsvoort, T. A. M. J. (2010). White-matter markers for psychosis in a prospective ultra-high-risk cohort. Psychological Medicine, 40(8), 1297–1304. Bodatsch, M., Ruhrmann, S., Wagner, M., Müller, R., Schultze-Lutter, F., Frommann, I., … Brockhaus-­ Dumke, A. (2011). Prediction of psychosis by mismatch negativity. Biological Psychiatry, 69(10), 959–966.

34 Bois, C., Levita, L., Ripp, I., Owens, D. C. G., Johnstone, E.  C., Whalley, H.  C., & Lawrie, S.  M. (2016). Longitudinal changes in hippocampal volume in the Edinburgh high risk study of schizophrenia. Schizophrenia Research, 173(3), 146–151. Boydell, J., Van Os, J., McKenzie, K., Allardyce, J., Goel, R., McCreadie, R. G., & Murray, R. M. (2001). Incidence of schizophrenia in ethnic minorities in London: Ecological study into interactions with environment. BMJ, 323(7325), 1336. Brown, A.  S. (2011). The environment and susceptibility to schizophrenia. Progress in Neurobiology, 93(1), 23–58. Bruggemann, J. M., Stockill, H. V., Lenroot, R. K., & Laurens, K. R. (2013). Mismatch negativity (MMN) and sensory auditory processing in children aged 9–12 years presenting with putative antecedents of schizophrenia. International Journal of Psychophysiology, 89(3), 374–380. Buchy, L., Stowkowy, J., MacMaster, F.  P., Nyman, K., & Addington, J. (2015). Meta-cognition is associated with cortical thickness in youth at clinical high risk of psychosis. Psychiatry Research: Neuroimaging, 233(3), 418–423. Bukenaite, A., Stochl, J., Mossaheb, N., Schäfer, M.  R., Klier, C.  M., Becker, J., … Jones, P.  B. (2017). Usefulness of the CAPE-P15 for detecting people at ultra-high risk for psychosis: Psychometric properties and cut-off values. Schizophrenia Research, 189, 69–74. Cannon, M., Caspi, A., Moffitt, T.  E., Harrington, H., Taylor, A., Murray, R.  M., & Poulton, R. (2002). Evidence for early-childhood, pan-developmental impairment specific to schizophreniform disorder: Results from a longitudinal birth cohort. Archives of General Psychiatry, 59(5), 449–456. Cannon, T.  D., Cadenhead, K., Cornblatt, B., Woods, S.  W., Addington, J., Walker, E., … Heinssen, R. (2008). Prediction of psychosis in youth at high clinical risk: a multisite longitudinal study in North America. Archives of General Psychiatry, 65(1), 28–37. Cannon, T. D., Chung, Y., He, G., Sun, D., Jacobson, A., Van Erp, T. G., … Cornblatt, B. (2015). North American Prodrome Longitudinal Study C.  Progressive reduction in cortical thickness as psychosis develops: A multisite longitudinal neuroimaging study of youth at elevated clinical risk. Biological Psychiatry, 77, 147–157. Cannon, T.  D., Yu, C., Addington, J., Bearden, C.  E., Cadenhead, K.  S., Cornblatt, B.  A., … Perkins, D.  O. (2016). An individualized risk calculator for research in prodromal psychosis. American Journal of Psychiatry, 173(10), 980–988. Capra, C., Kavanagh, D. J., Hides, L., & Scott, J. (2013). Brief screening for psychosis-like experiences. Schizophrenia Research, 149(1–3), 104–107. Carrión, R. E., Cornblatt, B. A., Burton, C. Z., Tso, I. F., Auther, A. M., Adelsheim, S., … Taylor, S. F. (2016). Personalized prediction of psychosis: External validation of the NAPLS-2 psychosis risk calculator with

D. I. Shapiro et al. the EDIPPP project. American Journal of Psychiatry, 173(10), 989–996. Carrión, R. E., Cornblatt, B. A., McLaughlin, D., Chang, J., Auther, A. M., Olsen, R. H., & Javitt, D. C. (2015). Contributions of early cortical processing and reading ability to functional status in individuals at clinical high risk for psychosis. Schizophrenia Research, 164(1–3), 1–7. Cho, K.  I. K., Shenton, M.  E., Kubicki, M., Jung, W. H., Lee, T. Y., Yun, J. Y., … Kwon, J. S. (2015). Altered thalamo-cortical white matter connectivity: Probabilistic tractography study in clinical-­ high risk for psychosis and first-episode psychosis. Schizophrenia Bulletin, 42(3), 723–731. Clark, S.  V., Mittal, V.  A., Bernard, J.  A., Ahmadi, A., King, T. Z., & Turner, J. A. (2018). Stronger default mode network connectivity is associated with poorer clinical insight in youth at ultra high-risk for psychotic disorders. Schizophrenia Research, 193, 244–250. Clemm von Hohenberg, C., Pasternak, O., Kubicki, M., Ballinger, T., Vu, M.  A., Swisher, T., … Woo, T.  U. W. (2013). White matter microstructure in individuals at clinical high risk of psychosis: a whole-brain diffusion tensor imaging study. Schizophrenia Bulletin, 40(4), 895–903. Colibazzi, T., Yang, Z., Horga, G., Yan, C. G., Corcoran, C. M., Klahr, K., … Peterson, B. S. (2017). Aberrant Temporal Connectivity in Persons at Clinical High Risk for Psychosis. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 2(8), 696–705. de Castro-Catala, M., Barrantes-Vidal, N., Sheinbaum, T., Moreno-Fortuny, A., Kwapil, T. R., & Rosa, A. (2015). COMT-by-sex interaction effect on psychosis proneness. BioMed Research International, 2015, 1–7. Del Re, E.  C., Spencer, K.  M., Oribe, N., Mesholam-­ Gately, R.  I., Goldstein, J., Shenton, M.  E., … Niznikiewicz, M.  A. (2015). Clinical high risk and first episode schizophrenia: Auditory event-related potentials. Psychiatry Research: Neuroimaging, 231(2), 126–133. Erickson, M. A., Ruffle, A., & Gold, J. M. (2016). A meta-­ analysis of mismatch negativity in schizophrenia: From clinical risk to disease specificity and progression. Biological Psychiatry, 79(12), 980–987. First, M. B., Williams, J. B. W., Benjamin, L. S., & Spitzer, R. L. (2015). User’s guide for the structured clinical interview for Dsm-5 personality disorders (scid-5-pd). American Psychiatric. Flatow, J., Buckley, P., & Miller, B.  J. (2013). Meta-­ analysis of oxidative stress in schizophrenia. Biological Psychiatry, 74(6), 400–409. Fonseca-Pedrero, E., Paino, M., Lemos-Giráldez, S., Sierra-Baigrie, S., & Muñiz, J.  (2010). Factor structure and measurement invariance of the Wisconsin Schizotypy scales across gender and age. The Spanish Journal of Psychology, 13(2), 941–950. Fornito, A., Yücel, M., Dean, B., Wood, S. J., & Pantelis, C. (2008). Anatomical abnormalities of the anterior cingulate cortex in schizophrenia: Bridging the

2  Assessment of Risk for Psychosis gap between neuroimaging and neuropathology. Schizophrenia Bulletin, 35(5), 973–993. Francis, A.  N., Bhojraj, T.  S., Prasad, K.  M., Montrose, D., Eack, S.  M., Rajarethinam, R., … Keshavan, M.  S. (2013). Alterations in the cerebral white matter of genetic high risk offspring of patients with schizophrenia spectrum disorder. Progress in Neuro-­ Psychopharmacology and Biological Psychiatry, 40, 187–192. Frommann, I., Brinkmeyer, J., Ruhrmann, S., Hack, E., Brockhaus-Dumke, A., Bechdolf, A., … Wagner, M. (2008). Auditory P300  in individuals clinically at risk for psychosis. International Journal of Psychophysiology, 70(3), 192–205. Fusar-Poli, P., Borgwardt, S., Bechdolf, A., Addington, J., Riecher-Rössler, A., Schultze-Lutter, F., … Valmaggia, L. (2013). The psychosis high-risk state: A comprehensive state-of-the-art review. JAMA Psychiatry, 70(1), 107–120. Fusar-Poli, P., Cappucciati, M., Borgwardt, S., Woods, S.  W., Addington, J., Nelson, B., … Simon, A.  E. (2016). Heterogeneity of psychosis risk within individuals at clinical high risk: a meta-analytical stratification. JAMA Psychiatry, 73(2), 113–120. Fusar-Poli, P., Deste, G., Smieskova, R., Barlati, S., Yung, A. R., Howes, O., … Borgwardt, S. (2012). Cognitive functioning in prodromal psychosis: a meta-analysis. Archives of General Psychiatry, 69(6), 562–571. Fusar-Poli, P., & Schultze-Lutter, F. (2016). Predicting the onset of psychosis in patients at clinical high risk: Practical guide to probabilistic prognostic reasoning. Evidence-Based Mental Health, 19(1), 10–15. Fux, L., Walger, P., Schimmelmann, B.  G., & Schultze-­ Lutter, F. (2013). The schizophrenia proneness instrument, child and youth version (SPI-CY): Practicability and discriminative validity. Schizophrenia Research, 146(1–3), 69–78. Fatemi, S. H., & Folsom, T. D. (2009). The neurodevelopmental hypothesis of schizophrenia, revisited. Schizophrenia Bulletin, 35(3), 528–548. Giuliano, A., Li, H., Mesholam-Gately, R.  I., Sorenson, S.  M., Woodberry, K.  A., & Seidman, L.  J. (2012). Neurocognition in the psychosis risk syndrome: A quantitative and qualitative review. Current Pharmaceutical Design, 18(4), 399–415. Gonzalez-Heydrich, J., Enlow, M.  B., D’Angelo, E., Seidman, L.  J., Gumlak, S., Kim, A., … O’Donnell, K. (2015). Early auditory processing evoked potentials (N100) show a continuum of blunting from clinical high risk to psychosis in a pediatric sample. Schizophrenia Research, 169(1–3), 340–345. Gonzalez-Heydrich, J., Enlow, M. B., D’Angelo, E., Seidman, L. J., Gumlak, S., Kim, A., … Duffy, F. H. (2016). N100 repetition suppression indexes neuroplastic defects in clinical high risk and psychotic youth. Neural Plasticity, 2016, 1–11. Gottesman, I. I., & Gould, T. D. (2003). The endophenotype concept in psychiatry: Etymology and strategic intentions. American Journal of Psychiatry, 160(4), 636–645.

35 Green, M.  F., Kern, R.  S., & Heaton, R.  K. (2004). Longitudinal studies of cognition and functional outcome in schizophrenia: Implications for MATRICS. Schizophrenia Research, 72(1), 41–51. Gross, G., Huber, G., Klosterkötter, J., & Linz M. (1987). BSABS.  Bonner Skala für die Beurteilung von Basissymptomen - Bonn scale for the assessment of basic symptoms. Berlin/Heidelberg/New York: Springer. Gross, G., & Huber, G. (2010). The history of the basic symptom concept. Acta Clinica Croatica, 49(Supplement 2), 47–59. Gururajan, A., Manning, E.  E., Klug, M., & van den Buuse, M. (2012). Drugs of abuse and increased risk of psychosis development. Australian & New Zealand Journal of Psychiatry, 46(12), 1120–1135. Häfner, H., Maurer, K., Ruhrmann, S., Bechdolf, A., Klosterkötter, J., Wagner, M., … Wölwer, W. (2004). Early detection and secondary prevention of psychosis: Facts and visions. European Archives of Psychiatry and Clinical Neuroscience, 254(2), 117–128. Hanssen, M., Bak, M., Bijl, R., Vollebergh, W., & Van Os, J.  (2005). The incidence and outcome of subclinical psychotic experiences in the general population. British Journal of Clinical Psychology, 44(2), 181–191. Hanssen, M., Krabbendam, L., Vollema, M., Delespaul, P., & Van Os, J.  (2006). Evidence for instrument and family-specific variation of subclinical psychosis dimensions in the general population. Journal of Abnormal Psychology, 115(1), 5. Hengartner, M. P. (2018). Raising awareness for the replication crisis in clinical psychology by focusing on inconsistencies in psychotherapy research: How much can we rely on published findings from efficacy trials? Frontiers in Psychology, 9, 256. Hooley, J.  M., Woodberry, K.  A., Ferriter, C. (2005). Family factors in schizophrenia and bipolar disorder. In Psychopathology and the family (pp.  205–223). Elsevier, Inc: San Diego, CA Hoptman, M. J., Nierenberg, J., Bertisch, H. C., Catalano, D., Ardekani, B.  A., Branch, C.  A., & DeLisi, L.  E. (2008). A DTI study of white matter microstructure in individuals at high genetic risk for schizophrenia. Schizophrenia Research, 106(2–3), 115–124. Howes, O. D., McDonald, C., Cannon, M., Arseneault, L., Boydell, J., Murray, R. M. (2004). Pathways to schizophrenia: The impact of environmental factors. The International Journal of Neuropsychopharmacology, 7(5), S7–S13. Hsieh, M. H., Shan, J. C., Huang, W. L., Cheng, W. C., Chiu, M. J., Jaw, F. S., … Liu, C. C. (2012). Auditory event-related potential of subjects with suspected pre-psychotic state and first-episode psychosis. Schizophrenia Research, 140(1–3), 243–249. Huber, G., & Gross, G. (1989). The concept of basic symptoms in schizophrenic and schizoaffective psychoses. Recenti Progressi in Medicina, 80(12), 646–652. Ising, H. K., Veling, W., Loewy, R. L., Rietveld, M. W., Rietdijk, J., Dragt, S., … van der Gaag, M. (2012).

36 The validity of the 16-item version of the Prodromal Questionnaire (PQ-16) to screen for ultra high risk of developing psychosis in the general help-seeking population. Schizophrenia Bulletin, 38(6), 1288–1296. Iwashiro, N., Suga, M., Takano, Y., Inoue, H., Natsubori, T., Satomura, Y., … Gonoi, W. (2012). Localized gray matter volume reductions in the pars triangularis of the inferior frontal gyrus in individuals at clinical high-­ risk for psychosis and first episode for schizophrenia. Schizophrenia Research, 137(1–3), 124–131. Jang, K. L., Woodward, T. S., Lang, D., Honer, W. G., & Livesley, W. J. (2005). The genetic and environmental basis of the relationship between schizotypy and personality: A twin study. The Journal of Nervous and Mental Disease, 193(3), 153–159. Javitt, D. C., & Sweet, R. A. (2015). Auditory dysfunction in schizophrenia: Integrating clinical and basic features. Nature Reviews Neuroscience, 16(9), 535. Jung, W.  H., Jang, J.  H., Shin, N.  Y., Kim, S.  N., Choi, C. H., An, S. K., & Kwon, J. S. (2012). Regional brain atrophy and functional disconnection in Broca’s area in individuals at ultra-high risk for psychosis and schizophrenia. PLoS One, 7(12), e51975. Karlsgodt, K.  H., Niendam, T.  A., Bearden, C.  E., & Cannon, T. D. (2009). White matter integrity and prediction of social and role functioning in subjects at ultra-high risk for psychosis. Biological Psychiatry, 66(6), 562–569. Kelleher, I., & Cannon, M. (2011). Psychotic-like experiences in the general population: Characterizing a high-­ risk group for psychosis. Psychological Medicine, 41(1), 1–6. Kendler, K.  S., & Diehl, S.  R. (1993). The genetics of schizophrenia: A current, genetic-epidemiologic perspective. Schizophrenia Bulletin, 19(2), 261–285. Kendler, K. S., Gruenberg, A. M., & Strauss, J. S. (1981). An independent analysis of the Copenhagen sample of the Danish Adoption Study of Schizophrenia: II. The relationship between schizotypal personality disorder and schizophrenia. Archives of General Psychiatry, 38(9), 982–984. Kendler, K.  S., McGuire, M., Gruenberg, A.  M., O’Hare, A., Spellman, M., & Walsh, D. (1993). The Roscommon Family Study: III. Schizophrenia-related personality disorders in relatives. Archives of General Psychiatry, 50(10), 781–788. Kendler, K.  S., Neale, M.  C., & Walsh, D. (1995). Evaluating the spectrum concept of schizophrenia in the Roscommon Family Study. American Journal of Psychiatry, 152(5), 749–754. Kim, M., Cho, K. I. K., Yoon, Y. B., Lee, T. Y., & Kwon, J. S. (2017). Aberrant temporal behavior of mismatch negativity generators in schizophrenia patients and subjects at clinical high risk for psychosis. Clinical Neurophysiology, 128(2), 331–339. Kim, M., Lee, T. Y., Lee, S., Kim, S. N., & Kwon, J. S. (2015). Auditory P300 as a predictor of short-term prognosis in subjects at clinical high risk for psychosis. Schizophrenia Research, 165(2–3), 138–144.

D. I. Shapiro et al. Kline, E., & Schiffman, J. (2014). Psychosis risk screening: A systematic review. Schizophrenia Research, 158(1–3), 11–18. Klosterkötter, J., Hellmich, M., Steinmeyer, E.  M., & Schultze-Lutter, F. (2001). Diagnosing schizophrenia in the initial prodromal phase. Archives of General Psychiatry, 58(2), 158–164. Kobayashi, H., Nemoto, T., Koshikawa, H., Osono, Y., Yamazawa, R., Murakami, M., … Mizuno, M. (2008). A self-reported instrument for prodromal symptoms of psychosis: Testing the clinical validity of the PRIME Screen—Revised (PS-R) in a Japanese population. Schizophrenia Research, 106(2–3), 356–362. Koh, Y., Shin, K. S., Kim, J. S., Choi, J. S., Kang, D. H., Jang, J.  H., … Kwon, J.  S. (2011). An MEG study of alpha modulation in patients with schizophrenia and in subjects at high risk of developing psychosis. Schizophrenia Research, 126(1–3), 36–42. Koutsouleris, N., Gaser, C., Patschurek-Kliche, K., Scheuerecker, J., Bottlender, R., Decker, P., … Meisenzahl, E.  M. (2012). Multivariate patterns of brain–cognition associations relating to vulnerability and clinical outcome in the at-risk mental states for psychosis. Human Brain Mapping, 33(9), 2104–2124. Krabbendam, L., & Van Os, J.  (2005). Schizophrenia and urbanicity: A major environmental influence— Conditional on genetic risk. Schizophrenia Bulletin, 31(4), 795–799. Kwapil, T. R., Barrantes-Vidal, N., & Silvia, P. J. (2007). The dimensional structure of the Wisconsin schizotypy scales: Factor identification and construct validity. Schizophrenia Bulletin, 34(3), 444–457. Kwapil, T. R., Brown, L. H., Silvia, P. J., Myin-Germeys, I., & Barrantes-Vidal, N. (2012). The expression of positive and negative schizotypy in daily life: An experience sampling study. Psychological Medicine, 42(12), 2555–2566. Li, X., Thermenos, H. W., Wu, Z., Momura, Y., Wu, K., Keshavan, M., … DeLisi, L.  E. (2016). Abnormal interactions of verbal-and spatial-memory networks in young people at familial high-risk for schizophrenia. Schizophrenia Research, 176(2–3), 100–105. Lichtenstein, P., Yip, B. H., Björk, C., Pawitan, Y., Cannon, T.  D., Sullivan, P.  F., & Hultman, C.  M. (2009). Common genetic determinants of schizophrenia and bipolar disorder in Swedish families: A population-­ based study. The Lancet, 373(9659), 234–239. Light, G.  A., & Swerdlow, N.  R. (2015). Future clinical uses of neurophysiological biomarkers to predict and monitor treatment response for schizophrenia. Annals of the New York Academy of Sciences, 1344(1), 105–119. Loewy, R. L., Pearson, R., Vinogradov, S., Bearden, C. E., & Cannon, T. D. (2011). Psychosis risk screening with the Prodromal Questionnaire—brief version (PQ-B). Schizophrenia Research, 129(1), 42–46. Longden, E., & Read, J.  (2016). Social adversity in the etiology of psychosis: A review of the evidence. American Journal of Psychotherapy, 70(1), 5–33.

2  Assessment of Risk for Psychosis Mamah, D., Mbwayo, A., Mutiso, V., Barch, D.  M., Constantino, J.  N., Nsofor, T., … Ndetei, D.  M. (2012). A survey of psychosis risk symptoms in Kenya. Comprehensive Psychiatry, 53(5), 516–524. March, D., Hatch, S.  L., Morgan, C., Kirkbride, J.  B., Bresnahan, M., Fearon, P., & Susser, E. (2008). Psychosis and place. Epidemiologic Reviews, 30(1), 84–100. Marconi, A., Di Forti, M., Lewis, C. M., Murray, R. M., & Vassos, E. (2016). Meta-analysis of the association between the level of cannabis use and risk of psychosis. Schizophrenia Bulletin, 42(5), 1262–1269. Mark, W., & Toulopoulou, T. (2015). Psychometric properties of “community assessment of psychic experiences”: Review and meta-analyses. Schizophrenia Bulletin, 42(1), 34–44. Marjoram, D., Job, D. E., Whalley, H. C., et al. (2006). A visual joke fMRI investigation into Theory of Mind and enhanced risk of schizophrenia. NeuroImage, 31(4), 1850–1858. Mason, O.  J. (2015). The assessment of schizotypy and its clinical relevance. Schizophrenia Bulletin, 41(suppl_2), S374–S385. McGorry, P.  D. (1998). “A stitch in time”… the scope for preventive strategies in early psychosis. European Archives of Psychiatry and Clinical Neuroscience, 248(1), 22–31. McGorry, P. D., Yung, A. R., & Phillips, L. J. (2003). The “close-in” or ultra high-risk model: A safe and effective strategy for research and clinical intervention in prepsychotic mental disorder. Schizophrenia Bulletin, 29(4), 771–790. McIntosh, A. M., Owens, D. C., Moorhead, W. J., Whalley, H.  C., Stanfield, A.  C., Hall, J., … Lawrie, S.  M. (2011). Longitudinal volume reductions in people at high genetic risk of schizophrenia as they develop psychosis. Biological Psychiatry, 69(10), 953–958. Meehl, P. E. (1962). Schizotaxia, schizotypy, schizophrenia. American Psychologist, 17(12), 827–838. Meehl, P.  E. (1989). Schizotaxia revisited. Archives of General Psychiatry, 46(10), 935–944. Miller, R., Mason, S. E. (2002). Diagnosis: Schizophrenia: A comprehensive resource. Columbia University Press. New York, NY Miller, T. (2004). The SIPS-Screen: A brief self-­ report screen to detect the schizophrenia prodrome. Schizophrenia Research, 70, 78. Miller, T. J., McGlashan, T. H., Rosen, J. L., Cadenhead, K., Ventura, J., McFarlane, W., … Woods, S.  W. (2003). Prodromal assessment with the structured interview for prodromal syndromes and the scale of prodromal symptoms: Predictive validity, inter-rater reliability, and training to reliability. Schizophrenia Bulletin, 29(4), 703–715. Miller, T. J., McGlashan, T. H., Rosen, J. L., Somjee, L., Markovich, P.  J., Stein, K., & Woods, S.  W. (2002). Prospective diagnosis of the initial prodrome for schizophrenia based on the structured interview for prodromal syndromes: Preliminary evidence of inter-­

37 rater reliability and predictive validity. American Journal of Psychiatry, 159(5), 863–865. Modenato, C., & Draganski, B. (2015). The concept of schizotypy—A computational anatomy perspective. Schizophrenia Research: Cognition, 2(2), 89–92. Mondragón-Maya, A., Solís-Vivanco, R., León-Ortiz, P., Rodríguez-Agudelo, Y., Yáñez-Téllez, G., Bernal-­ Hernández, J., … de la Fuente-Sandoval, C. (2013). Reduced P3a amplitudes in antipsychotic naive first-­ episode psychosis patients and individuals at clinical high-risk for psychosis. Journal of Psychiatric Research, 47(6), 755–761. Mowry, B.  J., & Nancarrow, D.  J. (2001). Proceedings of the Australian neuroscience society symposium: Schizophrenia molecular genetics of schizophrenia. Clinical and Experimental Pharmacology and Physiology, 28(1–2), 66–69. Mrazek, P. J., & Haggerty, R. J. (Eds.). (1994). Reducing risk for mental disorders: Frontiers for preventive intervention research. Washington, D.C.: National Academic Press. Murray, R. M., & Di Fori, M. (2016). Cannabis and psychosis: What degree of proof do we require? Biological Psychiatry, 79, 514–515. Nagai, T., Tada, M., Kirihara, K., Araki, T., Jinde, S., & Kasai, K. (2013). Mismatch negativity as a “translatable” brain marker toward early intervention for psychosis: A review. Frontiers in Psychiatry, 4, 115. Natsubori, T., Hashimoto, R.  I., Yahata, N., Inoue, H., Takano, Y., Iwashiro, N., … Abe, O. (2014). An fMRI study of visual lexical decision in patients with schizophrenia and clinical high-risk individuals. Schizophrenia Research, 157(1–3), 218–224. Nelson, M.  T., Seal, M.  L., Pantelis, C., & Phillips, L.  J. (2013). Evidence of a dimensional relationship between schizotypy and schizophrenia: A systematic review. Neuroscience & Biobehavioral Reviews, 37(3), 317–327. Nicolson, R., Lenane, M., Hamburger, S. D., Fernandez, T., Bedwell, J., & Rapoport, J.  L. (2003). Brain Research Reviews, 31, 147–156. Nieman, D.  H., Ruhrmann, S., Dragt, S., Soen, F., van Tricht, M.  J., Koelman, J.  H., … Linszen, D.  H. (2013). Psychosis prediction: Stratification of risk estimation with information-processing and premorbid functioning variables. Schizophrenia Bulletin, 40(6), 1482–1490. Niendam, T.  A., Lesh, T.  A., Yoon, J., Westphal, A.  J., Hutchison, N., Ragland, J.  D., … Carter, C.  S. (2014). Impaired context processing as a potential marker of psychosis risk state. Psychiatry Research: Neuroimaging, 221(1), 13–20. Niessen, M.  A., Dingemans, P.  M., van de Fliert, R., Becker, H. E., Nieman, D. H., & Linszen, D. (2010). Diagnostic validity of the Eppendorf schizophrenia inventory (ESI): A self-report screen for ultra-high risk and acute psychosis. Psychological Assessment, 22(4), 935.

38 Ord, L.  M., Myles-Worsley, M., Blailes, F., Ngiralmau, H. (2004). Screening for prodromal adolescents in an isolated high-risk population. Schizophrenia Research 71(2–3), 507–508. Paavilainen, P. (2013). The mismatch-negativity (MMN) component of the auditory event-related ­ potential to violations of abstract regularities: A review. International Journal of Psychophysiology, 88(2), 109–123. Padmanabhan, J. L., Shah, J. L., Tandon, N., & Keshavan, M.  S. (2017). The “polyenviromic risk score”: Aggregating environmental risk factors predicts conversion to psychosis in familial high-risk subjects. Schizophrenia Research, 181, 17–22. Pantelis, C., Yücel, M., Bora, E., Fornito, A., Testa, R., Brewer, W. J., … Wood, S. J. (2009). Neurobiological markers of illness onset in psychosis and schizophrenia: the search for a moving target. Neuropsychology Review, 19(3), 385. Pardiñas, A.  F., Holmans, P., Pocklington, A.  J., Escott-­ Price, V., Ripke, S., Carrera, N., … Han, J.  (2018). Common schizophrenia alleles are enriched in mutation-­intolerant genes and in regions under strong background selection. Nature Genetics, 50(3), 381. Pedersen, C. B., & Mortensen, P. B. (2001). Evidence of a dose-response relationship between urbanicity during upbringing and schizophrenia risk. Archives of General Psychiatry, 58(11), 1039–1046. Perez, V.  B., Roach, B.  J., Woods, S.  W., Srihari, V.  H., McGlashan, T.  H., Ford, J.  M., & Mathalon, D.  H. (2013). Early auditory gamma-band responses in patients at clinical high risk for schizophrenia. Supplements to Clinical Neurophysiology,  62, 147–162. Perez, V.  B., Woods, S.  W., Roach, B.  J., Ford, J.  M., TH, M. G., Srihari, V. H., & Mathalon, D. H. (2014). Automatic auditory processing deficits in schizophrenia and clinical high-risk patients: Forecasting psychosis risk with mismatch negativity. Biological Psychiatry, 75(6), 459–469. https://doi.org/10.1016/j. biopsych.2013.07.038 Perkins, D.  O., Jeffries, C.  D., Addington, J., Bearden, C.  E., Cadenhead, K.  S., Cannon, T.  D., … Tsuang, M.  T. (2014). Towards a psychosis risk blood diagnostic for persons experiencing high-risk symptoms: preliminary results from the NAPLS project. Schizophrenia Bulletin, 41(2), 419–428. Poulton, R., Caspi, A., Moffitt, T. E., Cannon, M., Murray, R., & Harrington, H. (2000). Children's self-reported psychotic symptoms and adult schizophreniform disorder: A 15-year longitudinal study. Archives of General Psychiatry, 57(11), 1053–1058. Raine, A. (1991). The SPQ: A scale for the assessment of schizotypal personality based on DSM-III-R criteria. Schizophrenia Bulletin, 17(4), 555–564. Ramyead, A., Kometer, M., Studerus, E., Koranyi, S., Ittig, S., Gschwandtner, U., … Riecher-Rössler, A. (2014). Aberrant current source-density and lagged phase synchronization of neural oscillations as markers for emerging psychosis. Schizophrenia Bulletin, 41(4), 919–929.

D. I. Shapiro et al. Riecher-Rössler, A., Pflueger, M. O., Aston, J., Borgwardt, S.  J., Brewer, W.  J., Gschwandtner, U., & Stieglitz, R.  D. (2009). Efficacy of using cognitive status in predicting psychosis: A 7-year follow-up. Biological Psychiatry, 66(11), 1023–1030. Ripke, S., Neale, B.  M., Corvin, A., Walters, J.  T., Farh, K.  H., Holmans, P.  A., … Pers, T.  H. (2014). Biological insights from 108 schizophrenia-associated genetic loci. Nature, 511(7510), 421. Ripke, S., O’Dushlaine, C., Chambert, K., Moran, J. L., Kähler, A. K., Akterin, S., … Kim, Y. (2013). Genome-­ wide association analysis identifies 13 new risk loci for schizophrenia. Nature Genetics, 45(10), 1150. Rosenthal, D. (1970). Genetic theory and abnormal behavior. New York, NY: McGraw-Hill. Ruhrmann, S., Schultze-Lutter, F., & Klosterkötter, J. (2010). Probably at-risk, but certainly ill—advocating the introduction of a psychosis spectrum disorder in DSM-V. Schizophrenia Research, 120(1–3), 23–37. Ruhrmann, S., Schultze-Lutter, F., Salokangas, R.  K., Heinimaa, M., Linszen, D., Dingemans, P., … Morrison, A. (2010). Prediction of psychosis in adolescents and young adults at high risk: Results from the prospective European prediction of psychosis study. Archives of General Psychiatry, 67(3), 241–251. Ruhrmann, S., Schultze-Lutter, F., Schmidt, S. J., Kaiser, N., Klosterkotter. J. (2014). Prediction and prevention of psychosis: current progress and future tasks. Eur Arch Psychiatry Clin Neurosci, 264(Suppl), S9-S16. Sabb, F. W., van Erp, T. G., Hardt, M. E., Dapretto, M., Caplan, R., Cannon, T. D., & Bearden, C. E. (2010). Language network dysfunction as a predictor of outcome in youth at clinical high risk for psychosis. Schizophrenia Research, 116(2–3), 173–183. Saha, S., Chant, D., Welham, J., & McGrath, J. (2005). A systematic review of the prevalence of schizophrenia. PLoS Medicine, 2(5), e141. Schizophrenia.com (n.d.). Schizophrenia facts and statistics. Retrieved July 12, 2018 from http://www.schizophrenia.com/szfacts.htm# Schmidt, A., Cappucciati, M., Radua, J., Rutigliano, G., Rocchetti, M., Dell’Osso, L., … McGuire, P. (2017). Improving prognostic accuracy in subjects at clinical high risk for psychosis: Systematic review of predictive models and meta-analytical sequential testing simulation. Schizophrenia Bulletin, 43(2), 375–388. Schultze-Lutter, F. (2009). Subjective symptoms of schizophrenia in research and the clinic: The basic symptom concept. Schizophrenia Bulletin, 35(1), 5–8. Schultze-Lutter, F., Addington, J., Ruhrmann, S., & Klosterkötter, J.  (2007). Schizophrenia proneness instrument, adult version (SPI-A). Rome, Italy: Giovanni Fioriti. Schultze-Lutter, F., Klosterkötter, J., & Ruhrmann, S. (2014). Improving the clinical prediction of psychosis by combining ultra-high risk criteria and cognitive basic symptoms. Schizophrenia Research, 154(1–3), 100–106. Schultze-Lutter, F., & Koch, E. (2010). Schizophrenia proneness instrument: Child and youth version (SPI-CY). Rome, Italy: Fioriti.

2  Assessment of Risk for Psychosis Schultze-Lutter, F., Michel, C., Schmidt, S.  J., Schimmelmann, B. G., Maric, N. P., Salokangas, R. K. R., … Meneghelli, A. (2015). EPA guidance on the early detection of clinical high risk states of psychoses. European Psychiatry, 30(3), 405–416. Schultze-Lutter, F., Ruhrmann, S., Fusar-Poli, P., Bechdolf, A., Schimmelmann, B. G., & Klosterkotter, J. (2012). Basic symptoms and the prediction of first-­ episode psychosis. Current Pharmaceutical Design, 18(4), 351–357. Schultze-Lutter, F., Ruhrmann, S., Picker, H., & Klosterkooter, J. (2006). Development and evaluation of the schizophrenia proneness instrument, adult version (SPI-A). Schizophrenia Research, 86, S4–S5. Schultze-Lutter, F., & Theodoridou, A. (2017). The concept of basic symptoms: Its scientific and clinical relevance. World Psychiatry, 16(1), 104–105. Seidman, L. J., Shapiro, D. I., Stone, W. S., Woodberry, K. A., Ronzio, A., Cornblatt, B. A., … Woods, S. W. (2016). Neurocognition and transition to psychosis: Baseline functioning in the second phase of the North American Prodrome Longitudinal study. JAMA Psychiatry, 73(12), 1239–1248. Shin, K.  S., Jung, W.  H., Kim, J.  S., Jang, J.  H., Hwang, J.  Y., Chung, C.  K., & Kwon, J.  S. (2012). Neuromagnetic auditory response and its relation to cortical thickness in ultra-high-risk for psychosis. Schizophrenia Research, 140(1–3), 93–98. Simon, A. E., Velthorst, E., Nieman, D. H., Linszen, D., Umbricht, D., & de Haan, L. (2011). Ultra high-risk state for psychosis and non-transition: A systematic review. Schizophrenia Research, 132(1), 8–17. Solís-Vivanco, R., Mondragón-Maya, A., León-Ortiz, P., Rodríguez-Agudelo, Y., Cadenhead, K.  S., & de la Fuente-Sandoval, C. (2014). Mismatch negativity reduction in the left cortical regions in first-episode psychosis and in individuals at ultra high-risk for psychosis. Schizophrenia Research, 158(1–3), 58–63. Stefanis, N.  C., Hanssen, M., Smirnis, N.  K., Avramopoulos, D.  A., Evdokimidis, I.  K., Stefanis, C. N., … Van Os, J. (2002). Evidence that three dimensions of psychosis have a distribution in the general population. Psychological Medicine, 32(2), 347–358. Sullivan, P.  F., Kendler, K.  S., & Neale, M.  C. (2003). Schizophrenia as a complex trait: Evidence from a meta-analysis of twin studies. Archives of General Psychiatry, 60(12), 1187–1192. Tada, M., Nagai, T., Kirihara, K., Koike, S., Suga, M., Araki, T., … Kasai, K. (2014). Differential alterations of auditory gamma oscillatory responses between pre-­ onset high-risk individuals and first-episode schizophrenia. Cerebral Cortex, 26(3), 1027–1035. Takano, Y., Aoki, Y., Yahata, N., Kawakubo, Y., Inoue, H., Iwashiro, N., …, & Takao, H. (2017). Neural basis for inferring false beliefs and social emotions in others among individuals with schizophrenia and those at ultra high risk for psychosis. Psychiatry Research: Neuroimaging 259:34-41 Tarbox, S. I., & Pogue-Geile, M. F. (2008). Development of social functioning in preschizophrenia children

39 and adolescents: A systematic review. Psychological Bulletin, 134(4), 561. Thermenos, H.  W., Juelich, R.  J., DiChiara, S.  R., Mesholam-Gately, R.  I., Woodberry, K.  A., Wojcik, J., … Petryshen, T.  L. (2016). Hyperactivity of caudate, parahippocampal, and prefrontal regions during working memory in never-medicated persons at clinical high-risk for psychosis. Schizophrenia Research, 173(1–2), 1–12. Thompson, A., Nelson, B., & Yung, A. (2011). Predictive validity of clinical variables in the “at risk” for psychosis population: International comparison with results from the North American prodrome longitudinal study. Schizophrenia Research, 126(1–3), 51–57. Tienari, P., Wynne, L. C., & Sorri, A. (2006). Genotype-­ environment interaction in schizophrenia-spectrum disorder: Long-term follow-up study of finish adoptees. Year Book of Psychiatry & Applied Mental Health, 2006, 273–274. Tijms, B.  M., Sprooten, E., Job, D., Johnstone, E.  C., Owens, D. G., Willshaw, D., … Lawrie, S. M. (2015). Grey matter networks in people at increased familial risk for schizophrenia. Schizophrenia Research, 168(1–2), 1–8. Tognin, S., Pettersson-Yeo, W., Valli, I., Hutton, C., Woolley, J., Allen, P., … Mechelli, A. (2014). Using structural neuroimaging to make quantitative predictions of symptom progression in individuals at ultra-­high risk for psychosis. Frontiers in Psychiatry, 4, 187. Torrey, E. F., Buka, S., Cannon, T. D., Goldstein, J. M., Seidman, L.  J., Liu, T., … Yolken, R.  H. (2009). Paternal age as a risk factor for schizophrenia: how important is it? Schizophrenia Research, 114(1–3), 1–5. Torrey, E.  F., Miller, J., Rawlings, R., & Yolken, R.  H. (1997). Seasonality of births in schizophrenia and bipolar disorder: A review of the literature. Schizophrenia Research, 28(1), 1–38. Tsuang, M. (2000). Schizophrenia: Genes and environment. Biological Psychiatry, 47, 210–220. Turetsky, B.  I., Cannon, T.  D., & Gur, R.  E. (2000). P300 subcomponent abnormalities in schizophrenia: III.  Deficits in unaffected siblings of schizophrenic probands. Biological Psychiatry, 47(5), 380–390. Unterrassner, L., Wyss, T.  A., Wotruba, D., Ajdacic-­ Gross, V., Haker, H., & Rössler, W. (2017). Psychotic-­ like experiences at the healthy end of the psychosis continuum. Frontiers in Psychology, 8, 775. Valmaggia, L. R., Day, F. L., Jones, C., Bissoli, S., Pugh, C., Hall, D., … Byrne, M. (2014). Cannabis use and transition to psychosis in people at ultra-high risk. Psychological Medicine, 44(12), 2503–2512. Van Os, J., & Guloksuz, S. (2017). A critique of the “ultra-high risk” and “transition” paradigm. World Psychiatry, 16(2), 200–206. Van Os, J., Kenis, G., & Rutten, B. P. (2010). The environment and schizophrenia. Nature, 468(7321), 203. Van Os, J., Linscott, R. J., Myin-Germeys, I., Delespaul, P., & Krabbendam, L. (2009). A systematic review and

40 meta-analysis of the psychosis continuum: Evidence for a psychosis proneness–persistence–impairment model of psychotic disorder. Psychological Medicine, 39(2), 179–195. Vollema, M.  G., & Ormel, J.  (2000). The reliability of the structured interview for schizotypy-revised. Schizophrenia Bulletin, 26(3), 619–629. Vollema, M. G., & van den Bosch, R. J. (1995). The multidimensionality of schizotypy. Schizophrenia Bulletin, 21(1), 19–31. Volpe, U., Mucci, A., Bucci, P., Merlotti, E., Galderisi, S., & Maj, M. (2007). The cortical generators of P3a and P3b: A LORETA study. Brain Research Bulletin, 73(4–6), 220–230. Walker, E.  F., & Diforio, D. (1997). Schizophrenia: A neural diathesis-stress model. Psychological Review, 104(4), 667. Walker, E. F., Trotman, H. D., Pearce, B. D., Addington, J., Cadenhead, K.  S., Cornblatt, B.  A., … Tsuang, M.  T. (2013). Cortisol levels and risk for psychosis: initial findings from the North American prodrome longitudinal study. Biological Psychiatry, 74(6), 410–417. Walter, A., Studerus, E., Smieskova, R., Kuster, P., Aston, J., Lang, U. E., … Borgwardt, S. (2012). Hippocampal volume in subjects at high risk of psychosis: a longitudinal MRI study. Schizophrenia Research, 142(1–3), 217–222. Walter, A., Suenderhauf, C., Harrisberger, F., Lenz, C., Smieskova, R., Chung, Y., … Borgwardt, S. (2016). Hippocampal volume in subjects at clinical highrisk for psychosis: A systematic review and metaanalysis. Neuroscience & Biobehavioral Reviews, 71, 680–690. Welham, J., Scott, J., Williams, G., Najman, J., Bor, W., O’callaghan, M., & McGrath, J. (2009). Emotional and behavioural antecedents of young adults who screen positive for non-affective psychosis: A 21-year birth cohort study. Psychological Medicine, 39(4), 625–634.

D. I. Shapiro et al. Whitfield-Gabrieli, S., & Ford, J. M. (2012). Default mode network activity and connectivity in psychopathology. Annual Review of Clinical Psychology, 8, 49–76. Woodberry, K. A., Shapiro, D. I., Bryant, C., & Seidman, L. J. (2016). Progress and future directions in research on the psychosis prodrome: A review for clinicians. Harvard Review of Psychiatry, 24(2), 87. Woods, S. W., Addington, J., Cadenhead, K. S., Cannon, T. D., Cornblatt, B. A., Heinssen, R., … McGlashan, T. H. (2009). Validity of the prodromal risk syndrome for first psychosis: Findings from the North American Prodrome Longitudinal Study. Schizophrenia Bulletin, 35(5), 894–908. Yung, A.  R., & McGorry, P.  D. (1996a). The prodromal phase of first-episode psychosis: Past and current conceptualizations. Schizophrenia Bulletin, 22(2), 353–370. Yung, A.  R., & McGorry, P.  D. (1996b). The initial prodrome in psychosis: Descriptive and qualitative aspects. Australian and New Zealand Journal of Psychiatry, 30(5), 587–599. Yung, A. R., Phillips, L. J., Yuen, H. P., & McGorry, P. D. (2004). Risk factors for psychosis in an ultra high-­ risk group: Psychopathology and clinical features. Schizophrenia Research, 67(2–3), 131–142. Yung, A. R., Yuen, H. P., Berger, G., Francey, S., Hung, T.  C., Nelson, B., … McGorry, P. (2007). Declining transition rate in ultra high risk (prodromal) services: Dilution or reduction of risk? Schizophrenia Bulletin, 33(3), 673–681. Yung, A. R., Yung, A. R., Pan Yuen, H., Mcgorry, P. D., Phillips, L.  J., Kelly, D., … Stanford, C. (2005). Mapping the onset of psychosis: the comprehensive assessment of at-risk mental states. Australian and New Zealand Journal of Psychiatry, 39(11–12), 964–971. Zhang, T., Xu, L., Tang, Y., Li, H., Tang, X., Cui, H., … Li, C. (2018). Prediction of psychosis in prodrome: Development and validation of a simple, personalized risk calculator. Psychological Medicine, 1–9.

3

Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence, and Future Directions Daniel I. Shapiro, Kristen A. Woodberry, Huijun Li, and Larry J. Seidman

Editors’ Note  A number of paradigms exist for prospectively identifying individuals who have elevated risk for developing psychosis due to the presence of syndromes comprised of identifiable risk factors and risk indicators. Conventions for which models are used, how individuals are identified, and which terminologies predominate vary throughout the world,

D. I. Shapiro (*) Department of Psychiatry, Early Psychosis Programs, University of California-Davis, Sacramento, CA, USA Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA e-mail: [email protected] K. A. Woodberry Maine Medical Center, Center for Psychiatric Research, Portland, ME, USA Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA e-mail: [email protected] H. Li Department of Psychology, College of Social Sciences, Arts and Humanities, Florida A&M University, Tallahassee, FL, USA e-mail: [email protected] L. J. Seidman (Deceased) Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA

sometimes related to culturally linked factors. In order to capture this diversity within one volume, the term Attenuated Psychosis Syndromes (APS) is used here to collectively refer to the class of putative prodromal or Psychosis-Risk Syndromes that have been empirically validated. We acknowledge that this is not a universally accepted convention, but use it as an umbrella term due to its heuristic value.

1

Introduction: Rationale and Approach for Treatment in Attenuated Psychosis Syndromes

Advances in science and technology over the last century have revolutionized medicine and the way at least Western cultures think about chronic and severe mental illness. Once thought virtually hopeless, the treatment of schizophrenia and other psychotic illnesses is moving from efforts to isolate, institutionalize, and as a result, dehumanize already highly stigmatized individuals to a precision medicine approach where diverse pathological biological, environmental, and psychosocial mechanisms are identified early and treated with targeted interventions. Key discoveries relevant to the treatment of individuals with

© Springer Nature Switzerland AG 2019 H. Li et al. (eds.), Handbook of Attenuated Psychosis Syndrome Across Cultures, https://doi.org/10.1007/978-3-030-17336-4_3

41

42

schizophrenia and other psychotic disorders include:

D. I. Shapiro et al.

apparent (DeLisi, Szulc, Bertisch, Majcher, & Brown, 2006; Liu, Keshavan, Tronick, & Seidman, 2015; Seidman et al., 2016). A detailed 1. That these are a heterogeneous class of condi- review of such early abnormalities is beyond the tions that result from an ever-expanding list of scope of this chapter (see, e.g., Fusar-Poli et al., dynamically interacting heritable and nonher- 2013); to summarize, differences in motor itable factors which vary somewhat from per- (Walker, Savoie, & Davis, 1994), cognitive son to person—there is no “smoking gun” (Seidman et  al., 2016), social and vocational (Schizophrenia Working Group of the (Addington, Penn, Woods, Addington, & Psychiatric Genomics Consortium, 2014; Perkins, 2008), brain structural/functional Tsuang, 2000) (Brent, Thermenos, Keshavan, & Seidman, 2. That many of the most serious pathological 2013; Lieberman, Perkins, et al., 2001), and horchanges that take place in psychotic illness monal (Walker et al., 2013) domains are already occur before and in the first few years after apparent before psychosis is clinically diagclinical criteria are met (Cannon et al., 2002; nosed. In some domains, trajectories of decline Lieberman et al., 2001) appear to slow after the first few years of illness 3. That people who will later develop a psychotic with larger changes associated with worse longdisorder can be reliably identified via the pres- term outcomes (Arongo et al., 2012; Cahn et al., ence of psychosis risk (PR) syndromes (Fusar-­ 2006; Fusar-Poli et al., 2013; Lieberman et al., Poli et al., 2012; Yung & McGorry, 1996) 2001), raising the question of whether the onset 4. That treatment in those with recognizable risk of illness and accompanying biological, social, syndromes has the potential to delay and and functional impacts can be interrupted or potentially prevent psychosis (Fusar-Poli slowed if identified early. In short, intervention et al., 2016; Van der Gaag et al., 2013) in those at risk who do not yet have a psychotic disorder is motivated by research suggesting that This chapter will discuss the advances, nature of, it may have a stronger impact earlier in the and evidence for intervention in those with syn- course of illness when it can disrupt these trajecdromes known to confer risk for psychosis, most tories, rather than later, once the majority of their prominently ultra/clinical high-risk (including impact has already occurred. genetic risk with deterioration) and basic An increasingly consistent literature investi­symptom syndromes. Per convention in this vol- gating links between duration of untreated psyume, the term Attenuated Psychosis Syndromes chosis (DUP) and subsequent outcomes and (APS) will be used as an umbrella term to clinical features supports the suggestion that eardescribe syndromes that have been shown in the lier treatment—even before people have historiliterature to convey elevated risk of imminent cally presented—may be more impactful. A conversion to psychosis, rather than solely the recent meta-analysis found that longer DUP was specifically defined DSM-V syndrome listed as a associated with worse symptoms (positive, negacondition for study. tive, and overall), social functioning, employThe past two decades have seen a flourish of ment, and quality of life, as well as more research on the prodromal period of psychosis. rehospitalizations and less time spent in remisAlthough schizophrenia was once thought of as a sion over an average 8.1  year follow-up (Mean disease of lifelong degeneration, a growing body DUP  =  61.3; Penttilä, Jääskeläinen, Hirvonen, of research has shown that many of the differ- Isohanni, & Miettunen, 2014). Similar associaences seen between those with and without tions were also found in a large cohort (n = 786) major psychotic disorders emerge long before of Australian youths whose median DUP of these illnesses can be diagnosed, many even 8.7 weeks was much shorter and in samples from before initial psychotic-like symptoms are low- and middle-income countries (Farooq,

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

Large, Nielssen, & Waheed, 2009), indicating consistency of this link across cultures and duration of symptoms. Presenting for treatment earlier, during the high-risk state, is associated with even shorter DUP in those who go on to develop psychosis and in fewer and shorter subsequent hospitalizations (Fusar-Poli et al., 2016). In short, the earlier the intervention, the better long-term outcomes appear to be.

43

Thankfully, recent characterization of APS samples can inform the response to this dilemma. The first, and perhaps most critical, finding is that the vast majority (~80%) of APS, even in samples from different countries and cultures, are not treatment naïve, many having sought mental health treatment well before the onset of a recognizable APS (Woodberry et  al., 2018). Second, they are more likely than not to meet both current and past criteria for at least one nonpsychotic disorder (Solakangas et al., 2012) and by definition 1.1 Weighing the Risks to report significant distress and/or impairment. and Benefits of Early This includes impairments in quality of life Intervention (Bechdolf et al., 2005), social and role functioning (Addington et al., 2008), and other sources of Of course, intervening prior to the onset of a disability (Velthorst et  al., 2010). Third, those clearly recognizable psychotic disorder conveys with APS are more likely to develop diagnosadditional risks. The prodrome may represent an able nonpsychotic than psychotic diagnoses ideal window for intervention given that methods (Addington et  al., 2011), reflecting the multifiof prospectively identifying putatively in this nality of APS (Ruhrmann, Schultze-Lutter, & period via APS syndromes have fairly good sensi- Klosterkotter, 2010). These findings are all reletivity and specificity (Fusar-Poli et  al., 2012). vant to determining what treatments should be However, rates of transition to a psychotic disor- offered and when (Heinssen & Insel, 2015), how der in these samples average around 36% (Fusar-­ specific to psychosis prevention vs. current Poli et al., 2012) and may be dropping worldwide symptoms and impairment treatments should be (Fusar-Poli, Borgwardt, et al., 2013; Yung et al., and have contributed to the substantial advocacy 2007), meaning that over 60% of these samples for a stepped or staged model of care (McGorry do not transition to psychosis, at least in the short 2015; McGorry, Killackey, & Yung, 2008). In this term. This presents a dilemma: How should the model, individuals with more general indicators risks of illness be weighed against the risks of of psychopathology or risk are given more genintervention in an at-risk group as a whole eral and benign treatments to treat existing symp(Heinssen & Insel, 2015)? Psychotic illnesses can toms, prevent progression, and improve be severely disabling and chronic, p­articularly functional outcomes. Only those with the most when treatment is delayed. Yet intervening with serious symptoms or most imminent indicators individuals who will never develop a psychotic are given intensive and psychosis-specific treatillness, even in the absence of intervention, carries ments, in particular, antipsychotics. Thus, early the potential for completely unjustified iatrogenic treatment may target and reduce an individual’s or side effects. Psychiatric, particularly antipsy- specific risk factors, symptoms, and distress and chotic, medications may affect the brain during prevent serious mood or anxiety disorders and ongoing development (e.g., Ho, Andreasen, other pathological outcomes as more focal tarZiebell, Pierson, & Magnotta, 2011) and intro- gets of intervention than psychosis. Moreover, duce metabolic and extrapyramidal side effects to the empirical base for treating these co-occurring a young population particularly vulnerable to conditions is often far more established than for them (Correll et al., 2009; Francey et al., 2010). APS and has received more validation and exten“Labeling” and/or entry into the mental health sion across cultures (e.g., de Souza et al., 2013). system may increase the experience of stigma and The key questions remaining are (1) at what its negative consequences (Yang, Wonpat-Borja, threshold of risk (e.g., symptom severity, impairOpler, & Corcoran, 2010). ment, rate of increase, or syndromal pattern)

D. I. Shapiro et al.

44

should the intervention be specifically oriented toward the reduction of APS or prevention of acute psychosis? (2) What should this treatment entail?

1.2

Guiding Theoretical Strategy of Intervention for APS

Follow-up studies of APS suggest that most individuals who meet psychosis-risk syndrome criteria based on clinical or familial risk criteria and go on to develop psychosis do so within 1–2 years (Fusar-Poli et al., 2012). Though a few studies have included follow-up periods longer than 2–3  years, those that do find diminishing rates of new psychosis as time goes on (e.g., Klosterkötter, Hellmich, Steinmeyer, & Schultze-Lutter, 2001; Riecher-Rössler et  al., 2009). This confirms that the current syndromes identify imminent risk for transition to a psychotic disorder (during the first year or two after identification), with a short window for engagement and treatment. Quick onset of services is a key guiding principle. The question then becomes what should be treated and with which interventions. Some studies have sought to prevent psychosis, largely by directly treating subthreshold psychotic symptoms, while others have broadened their scope to focus more on current mood, anxiety, attention, and other nonspecific symptoms or behaviors, along with functional difficulties. Evidence for interventions with APS is currently based on ­relatively small samples and, as previously stated, has focused on follow-up periods of typically less than 2 years, meaning that more work is needed to truly understand long-term outcomes in adolescents and young adults, many of whom have not yet passed through the age of peak risk. This makes it difficult to know what targets are key and at what developmental time points because studies have by and large not had the power or longitudinal follow-ups for such analyses. Larger sample sizes are also needed to compare multiple active treatments against each other on rates of conversion, quality of functioning, and other

outcomes. Longer follow-ups are also needed to test whether interventions prevent rather than just delay conversion (although a delay may still have significant clinical impact). Thus, more work is also needed to better understand which interventions yield better results (and in whom). In spite of the immaturity of this science, a number of generally convergent practice guidelines for preventive assessment and intervention have been proffered by professional organizations and panels of leaders in the field (e.g., Deutsche Geselleschaft fur Psychiatrie, 2006; International Early Psychosis Association Writing Group, 2005; National Institute for Health and Clinical Excellence, 2014; Schmidt et  al., 2015; see Woodberry et  al., 2016 for review). The theory behind these recommendations is that preventive treatment aims to ameliorate malleable factors and symptoms while buffering against the impact of less malleable ones. The assumption is that altering the balance of risk and protective factors can alter the diathesis-­stress equation and “illness” trajectory. Most commonly, treatments aim to reduce sources of stress that might interact with extant vulnerabilities to lead to psychosis. Examples include treatments that target environmental stressors (e.g., family therapy, educational interventions, case management) and psychological therapies that target stress-producing appraisals (e.g., CBT). The various treatment guidelines from around the world show substantial similarity. Differences tend to be in the specific treatments endorsed as new results emerge. For example, after a quantitative review, The European Psychiatric Association (Schmidt et al., 2015) provided a set of recommendations for assessment and intervention. Their recommendations are summarized as follows: • Identification of APS requires assessment by trained clinicians using established measures and methods prior to intervention. • Interventions in the APS state aim not only to prevent psychosis but also to address functional and social deficits already present, using interventions with established evidence for

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

•

•

those deficits. Functional outcomes should be a focus of the overall treatment plan, rather than symptoms or diagnosis. Comorbid, non-specific conditions like mood, anxiety, and substance use concerns should be treated first based on evidence and guidelines for those conditions. A staged intervention should be applied such that the least restrictive or potentially risky interventions be given first (e.g., psychological treatments). In the case that these are ineffective, they should be complemented by low-dose second-generation antipsychotics in order to stabilize symptoms enough for psychological interventions to work. Current evidence does provide some support for use of psychotherapeutic and pharmacological interventions, but this evidence (discussed below) is not sufficient to justify primarily preventive intervention. Thus, treatments during this phase should be considered useful, in itself, and target the deficits and concerns that are present. Continued monitoring during the APS phase by trained, expert clinicians is also recommended.

45

Vancampfort, De Herdt, Yu, & De Hert, 2013). Examples include diet, exercise, substance use, and smoking cessation interventions. Discussed below is the available science which informs these recommendations.

2

 o Treatments for APS Make D a Difference?

2.1

 vidence for the Efficacy E of Early Intervention

What we know about APS states is based in large part on research in help-seekers; individuals who • present for participation in research or to a clinician due to their own or a family member’s desire to better understand and find relief from distressing symptoms. Indeed, subjective distress is a diagnostic requirement for clinical risk (other than that based on genetic/state risk) and basic symptom syndromes. But which treatments lead to the best outcomes? Answering this question • requires attention to the effects of interventions, not only on rates of subsequent transition to psychosis but also on distress and functioning. The evidence base currently is largest and strongest Other guidelines (e.g., McGorry et al., 2009; for the former (rates of transition), even as NICE, 2014; Singh, DeJoseph, & Cadenhead, researchers and clinicians alike recognize the 2012; Woodberry et  al., 2016) more strongly greater importance of the latter (distress and emphasize the use of more benign treatments functioning, Yung, 2017). (e.g., CBT, omega-3 fatty acids) as a first-line At the time this chapter was written, there treatment, given their more positive benefit-risk were 14 published randomized controlled trials ratio, higher acceptability and tolerance, lower of pharmacological and/or psychosocial interrisk of stigma, and reduced risk of exposing ventions in reliably identified APS groups with at nonpsychotic youth to pharmacological side least 12-month follow-ups, along with a larger ­ effects (McGorry et  al., 2009). It is also worth number of often small feasibility and efficacy trinoting that treatments aimed at improving gen- als. Several meta-analyses and reviews of these eral mental and physical well-being are excluded studies have been published with varying results, from EPA guidelines but included elsewhere depending on the number and quality of interven(e.g., Addington et al., 2014; Singh et al., 2012; tions included. Nearly all of the studies included Woodberry et  al., 2016), given their stress-­ in these meta-analyses use SIPS and CAARMS inoculation effects, and the need to actively pre- criteria (Clinical/Ultra-High Risk) or Basic vent poor health outcomes associated with Symptoms to identify APS.  Van der Gaag and chronic illness (Mueser, Deavers, Penn, & colleagues (2013) conducted what is often seen Cassisi, 2013) and address common side effects as the most definitive quantitative review of interof antipsychotics when given (Mitchell, ventions to date. Pooling across ten randomized

D. I. Shapiro et al.

46

prevention trials (three of antipsychotics, one of omega-3 fatty acids, two of integrated psychotherapy programs, five of CBT), these authors found that early intervention has a significant effect on transition to psychosis but that the quality of the interventions trials needs improvement. Collapsed across treatment type, the authors found that the risk of developing psychosis over 12 months was reduced, on average, by 52–54%, with a number needed to treat (NNT) of nine. This means that, statistically, nine individuals with APS need to be treated to delay or prevent acute psychosis in one individual. Effects of interventions diminished some over the longer term but were still associated with an overall 35–37% reduction in risk of psychosis (NNT = 12) over 2–4-year follow-up. Three studies of antipsychotic medications were associated with a reduction of risk of 55% (NNT = 7), while pooled effects across five studies of CBT yielded a risk reduction of 52% (NNT = 13). Given concerns about the side effect profiles of most medications and small number of high-quality studies of other interventions, CBT has the most support to date but cannot be said to be more efficacious than any other treatments in delaying or preventing psychosis in the short term. This conclusion is echoed by other meta-analyses (e.g., Schmidt et  al., 2015; Stafford, Jackson, Mayo-Wilson, Morrison, & Kendall, 2013). Unfortunately, data on other outcomes, like functioning, quality of life, and distress, are more scant. Although this may be changing, these have tended to be considered “secondary” outcomes in treatment trials and thus more rarely examined. In the van der Gaag et al. review (2013), only 6 of 13 studies examined functioning. Collapsed across intervention type, social functioning showed a nonsignificant improvement at 12 months. It is possible that the results might be significant with more statistical power, but effects are likely to be small. Stafford and co-authors (2013) performed a meta-analysis on 11 randomized trials (7 were also included in the van der Gaag study). Four were CBT trials, two examined antipsychotic medications, two tested combined CBT and antipsychotics, two tested integrated therapies, and one tested omega-3

fatty acid. These authors found no effect across psychosocial or antipsychotic treatments on depression, mania, or quality of life. Similarly, Schmidt et al. (2015) meta-analyzed 15 treatment trials (although not all were RCTs), including the majority of those included in the above reviews. They did report superior improvements in functioning over 2–6 months across seven experimental interventions compared with control treatments, but superiority disappeared at 12 and 18 months, when both control and experimental groups showed improved global functioning. To summarize, evidence is emerging that early intervention may reduce rates of conversion in those with APS, at least within the 1–3-year range. The size of this reduction is modest and decreases over time. There is, unfortunately, a relative paucity of research on the effects of intervention on other outcomes, like social and role functioning, quality of life, and comorbid symptoms. All groups appear to show some improvement over time, suggesting the need for more effective or powerful treatments targeting these domains to show a differential effect. Given the importance of these domains, expert clinical recommendation favors the use of low-risk interventions or treatments with established evidence bases for the treatment of co-occurring conditions in other populations, including individuals with established psychotic disorders.

2.2

 vidence Base for Individual E Interventions

Results of RCTs in those with an APS are briefly summarized below. For a detailed description of those published prior to 2016, please refer to Woodberry et al. (2016).

2.2.1 Psychotropic Agents As discussed throughout this volume, psychotropic treatment in those with APS is common worldwide. Antipsychotics are typically less commonly prescribed than antidepressants, anxiolytics, and other such classes of medications (Woods et al., 2013). Five published RCTs have examined pharmacological treatments either

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

alone or in conjunction with psychosocial interventions. McGlashan et al. (2006) found a nearly significant difference in conversion rates (16.1% vs. 37.9%) between an APS group that received 1 year of relatively low-dose olanzapine (n = 29) and a placebo group (n = 31), but this difference disappeared at 1-year follow-up. Those in the experimental group (EG) had a reduction in positive symptoms but also gained significantly more weight (8.49  kg/18.7  lbs). Functioning improved in both groups. McGorry and coinvestigators published two trials of risperidone in conjunction with a psychosocial treatment. In the first (McGorry et al., 2002), they compared combined CBT, low-dose risperidone, and needs-based supportive psychotherapy/case management with the needs-based intervention alone. After 6  months of treatment, conversion rates were lower in the EG (3 of 31 = 9.7%) than in the needs-based group (10/28  =  35.7%). At 6-month follow-up, conversion rates were no longer different overall, but when analyses were restricted to only those compliant with EG treatment, this group outperformed the control group. The groups both showed improvements in quality of life, global functioning, APS symptoms, anxiety, depression, mania, and negative symptoms, but there were no between treatment group differences. A similar pattern of results was found when these authors compared 12 months of conjunctive cognitive therapy and low-dose risperidone (n = 43) with cognitive therapy (CT) and placebo (n = 44), supportive therapy and placebo (n  =  28), or clinical monitoring alone (n  =  78) (McGorry et  al., 2013; Yung et  al., 2011). Rates of transition to psychosis did not differ between groups over 12  months of treatment and an a­ dditional 12-month follow-up, and all groups showed comparable improvements in APS symptoms, anxiety, quality of life, and global functioning. In neither study was it possible to tease apart treatment effects due to CT or medications. Two RCTs have examined the effects of Omega3 polyunsaturated fatty acid (PUFA) in the APS state. The first, by Amminger et  al. (2010), created enormous optimism in the field given the relatively innocuous nature of Omega3.

47

These authors found that 12  weeks of omega3 PUFA resulted in lower transition to psychosis rates at 12 months in the EG (2/41 = 4.9%) than in the placebo group (11/40 = 27.5%), associated with a 22.6% reduction in risk. Positive, negative, and general symptoms, as well as global functioning, also improved relative to the control group. Given the promise of these results, the NEURAPRO study (McGorry et  al., 2017) sought to replicate them. Investigators randomized youth with APS to 6  months of cognitive behavioral case management and either conjunctive omega-3 PUFA (n  =  153) or placebo (n  =  139). Rates of psychosis transition did not differ across groups after 12  months but were relatively low (11.5% and 11.2%, respectively). Both groups improved over time on APS symptoms, negative symptoms, mania, depression, social and occupational functioning, and global functioning, and it was not possible to determine the relative effects of the psychosocial vs. PUFA treatment. Adherence was also lower in this study than in the Amminger trial.

2.2.2 Psychosocial Treatments: Cognitive Behavioral Therapy No psychosocial treatment has been tested more in psychosis or the APS state than Cognitive/ Cognitive Behavioral Therapy (CT/CBT). Results are promising, but as in established psychotic disorders, the proliferation has presented some complications in interpretation (and debate among investigators). Different CBT “packages” are comprised of different components and measure outcomes with different metrics (McKenna & Kingdon, 2014; Thomas, 2015). This makes direct comparison of results difficult. Morrison and Barratt (2010), for example, conducted a Delphi study in which CBT for psychosis (CBTp) experts identified 77 different elements of CBTp, encompassing aspects of the therapist’s approach, structure of therapy, change strategies, formulation, and so on. Even this extensive list neglects the issue of how improvement through therapy should be measured (e.g., symptom severity, change in targets thought to underlie symptom maintenance, cognitions and behaviors, functioning).

48

CBT models that have been modified for the APS population and evaluated in RCTs have varied in their components. To date, seven RCTs have examined such treatments. Two conjunctive trials were discussed above. In the CBT/risperidone trial conducted by McGorry and colleagues (2002), one of the first to examine cognitive therapy in this population, treatment consisted of 6 months of a manualized therapy combining strategies used in comorbid nonpsychotic conditions, including those for addressing stress management, depression/negative symptoms, positive symptoms, and other comorbid conditions. As stated previously, it was not possible to determine the degree to which beneficial effects were due to medication, CBT, “needs-­ based interventions” (treatment for social, vocational, and family issues, as well as case management), or some combination. No differences were found across three groups in a subsequent trial using the same CBT package over 12 months (McGorry et al., 2013). CBT with risperidone or placebo was compared with supportive therapy and placebo. Other RCTs have examined CT/CBT alone, though control conditions vary. In the EDIE trial, Morrison et al. (2004) evaluated a different cognitive therapy package (French & Morrison, 2004) in a group of 37 individuals with APS, compared with 23 who received no treatment. Both groups received clinical monitoring and some case management, but those in the CT group also received up to 26 sessions of CT over 6  months. CT was based on a manualized approach that prioritizes normalization of symptoms, development/evaluation of alternative explanations, decatastrophizing, and behavioral experiments to test alternative beliefs. It also incorporated more traditional treatment approaches for comorbid anxiety, mood, and other concerns. Those in the EG showed a significantly lower rate of transition to psychosis (2/37 = 6% vs. 5/23 = 22%; NNT = 6), a lower rate of prescription of antipsychotic medications (2/37 = 6% vs. 7/23 = 30%; NNT = 5), and reduction in positive symptoms after 12 months, but no group differences in functioning. In a 3-year follow-­up of 47% of these participants (Morrison

D. I. Shapiro et al.

et  al., 2007), the lower rate of antipsychotics remained significant. After controlling for baseline levels of positive symptoms and cognitions targeted by CT (negative beliefs about uncontrollability of unwanted thoughts and fear of rejection/criticism), those in the CT group had a lower rate of conversion (OR = 0.03). In a larger multisite trial, this same CBT package was tested across five sites in the United Kingdom. One hundred forty-four youths with APS received up to 6 months of weekly therapy plus monitoring, while a control group (n  =  144) received only monitoring, described as regular face-to-face visits with symptom assessment and referral to appropriate community services. An average of 9.1 CT sessions did not result in fewer conversions to psychosis over 24 months from baseline (10/144 = 6.9% in the CT group v 13/144 = 9%). Similarly, treatment did not lead to significantly better change in global functioning, depression, social anxiety, or quality of life, but it did result in a reduction in the severity of APS symptoms. Other groups have tested the same manualized CBT package developed by French and Morrison (2004). Addington and co-investigators (2011) conducted an RCT in Toronto. Study staff randomized 51 youth with APS to up to 6 months of CBT (n  =  27; mean 12 sessions) or supportive therapy (n = 24) and then followed them for an additional 12  months. Conversions occurred in the supportive therapy control group only, though the difference in conversion rate (n  =  0 vs. 3/24  =  12.5%) was not statistically significant. Both groups experienced a drop in anxiety, depression, and positive APS symptoms, with the latter occurring more rapidly in the experimental group. Neither treatment affected social functioning or negative symptoms. van der Gaag et  al. (2012) conducted a RCT on a modified version of this CBT model in the EDIE-NL trial; they added psychoeducation on dopamine super-sensitivity and its effects on sensation and thinking, as well as common cognitive biases. All participants across four sites in the Netherlands met criteria for APS and received evidence-based treatment (TAU) for the nonpsychotic spectrum condition that led to their referral. In the experimental group, 97 youths with APS also received the

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

modified CBT for APS. The control group consisted of 104 youths with APS who only received TAU but were told they were at risk for developing other mental health concerns. Those in the EG attended on average ten sessions over 6 months. Over 18 months of follow-up, those in the EG (10/98  =  10.2%) were less likely to develop psychosis than those in the control group (22/103 = 21.4%) with a NNT of nine and were also more likely to no longer meet APS criteria. Those in the CBT group also reported less distress and feelings of entrapment by their symptoms at 6 months, but no group differences were found in depression or anxiety. Both groups generally saw improvement in symptoms over time. Effectiveness persisted in a subsequent 4-year follow-up of 113 of the original participants (Ising et al., 2016), with transition rates still significantly lower in the EG (two additional conversions, n = 12) than in the controls (OR = 0.44, NNT-8) and higher rates of remission from APS status (76.3% vs. 58.7%). In contrast, in a smaller RCT completed in Australia, Stain et al. (2016) compared this CBT package + standard psychiatry and case management to an active control treatment called non-directive reflective listening. The control condition was designed to contain non-specific treatment factors like genuineness, unconditional positive regard, and empathic reflection that have been shown to account for some treatment effects across different theoretical orientations. After 6  months of treatment (Mean number of CBT sessions = 9.2, Mean number of NDRL sessions  =  10.1), 3/30 (10%), in the CBT group developed psychosis, but none in the NDRL did. No additional conversions occurred over 6 more months of follow-up, after which only two in the CBT group and one in the NDRL group still met APS criteria. Those in the NDRL group reported less distress than those in the CBT group, but there were no group differences in rate or intensity of APS symptoms, anxiety, depression, global, social, or role functioning. The authors note that because low functioning was not a criterion for inclusion (as it is in many RCTs), their sample was comprised of slightly younger, higher-functioning individuals with lower baseline severity of APS symptoms. This

49

may explain the nearly total remission and low conversion rates, which may confound the results. Bechdolf et  al. (2012) developed their own CBT package and tested it as part of a 12-month Integrated Psychological Intervention (IPI) across four sites in Germany. These authors based their CBT model on a basic symptom conceptualization, in which these cognitive symptoms precede negative affect, withdrawal, and functional decline and, via the resultant flawed appraisals of experiences and beliefs, contribute to and maintain APS symptoms when they develop. Those in the experimental group were allotted 25 sessions of structured CBT (average sessions not reported), comprised of assessment and engagement, psychoeducation, stress management, symptom management, and crisis management modules. Results are discussed in more depth in the integrated treatments section. In summary, results of CBT trials in those at APS are mixed but promising, with nearly all trials showing some benefit but in different domains.

2.2.3 Psychosocial Treatments: Family Therapies Because psychosis typically develops when individuals are in their teens and 20s, APS youth often live with and are still dependent on their families of origin. Consistent with the robust literature validating the role of stress on psychosis onset and relapse, family factors can both exacerbate and buffer against a psychosis diathesis (Tienari et  al., 1994; Vaughn & Leff, 1976). Single and multifamily therapies have a very strong evidence base for schizophrenia and other serious mental illnesses (e.g., McFarlane, 2016). Adapting one such treatment, Miklowitz et  al. (2014) tested family-focused therapy (FFT) in APS youth and their family member(s) across eight North American Prodrome Longitudinal Study (NAPLS) sites. FFT, conducted with individual families, provides psychoeducation on positive, negative, depressive, and anxiety symptoms, as well as stress, risk for psychosis, and the diathesis-stress model. Families develop an individualized prevention plan, learn to track stressors and warning signs, and identify coping strategies and response options. Specific skill

50

building focuses on formal problem solving and communication enhancement training, teaching people to express positive feelings, use active listening, communicate more clearly, make effective requests, and say no. Up to 6 months of FFT (18 sessions, average 11) was compared with a three-session enhanced care condition (average 2.4 sessions), in which participants received psychoeducation and developed a personal prevention plan. Both groups could access crisis management sessions, if needed. Participants in both FFT (n  =  66) and enhanced care (n  =  63) saw improvement in positive symptoms, but those in the FFT condition had significantly greater improvements by 6  month follow-up. Although both groups also experienced improvement in negative symptoms, FFT had no significant effect on this domain. Similarly, conversion rates at 6 months were low in both groups: 5/47 (10.6%) in the EC group and 1/55 (1.8%) in the FFT group. Both groups improved on global and role functioning, but group effects were moderated by age. Older participants (20+) benefitted more from FFT, while those aged 16–19 benefitted more from enhanced care. Multifamily group psychoeducation (MFG-P) treatment is another family therapy that is a mainstay of first episode psychosis treatment and most often follows a model developed by Bill McFarlane and colleagues that has been evolving since the 1980s (McFarlane, 2002). MFG-P was tested in the quasi-experimental EDIPP trial, discussed further in the integrative treatments section. MFG-P begins with three individual family joining sessions and a multifamily psychoeducation workshop followed by bi-weekly structured groups in which families socialize and engage in formal problem solving to prevent relapse. In later meetings, problem solving focuses on resuming educational and vocational functioning.

2.2.4 Psychosocial Treatments: Cognitive Enhancement/ Remediation Cognitive deficits, particularly in memory, social cognition, attention, problem solving, and other executive functions, are a core feature of schizo-

D. I. Shapiro et al.

phrenia that are present before the onset of illness (Piskulic et al., 2016; Seidman et al., 2016), and are strongly linked to functional impairment (Green, Kern, Braff, & Mintz, 2000). Thus, some treatment aims to improve these cognitive functions by repeated training and practice with the goal of improving functioning. Commercially available programs like Lumosity, COGPACK, SocialVille, and PositScience have been tested. Four pilot studies have examined such programs in those at APS with promising results (Holzer et  al., 2014; Hooker et  al., 2014; Piskulic, Barbato, Liu, & Addington, 2015; Rauchensteiner et al., 2011). For example, in a small RCT, Loewy et  al. (2016) had 50 youths with APS complete 40  h of PositScience exercises over 5  weeks. Exercises were designed to target verbal learning and memory. A control group (n  =  33) played specific computer games for an equal amount of time. Both groups improved in a few cognitive domains, as well as global functioning and positive, negative, and disorganized symptoms. Those in the EG showed greater improvements in verbal memory and disorganized symptoms after training. The Integrated Psychological Intervention described previously (Bechdolf et al., 2012) also included a cognitive enhancement treatment. These authors used six COGPACK tasks to target attention, memory, and executive functioning. While they did not report effects on cognition or functioning, they did find lower conversion rates in the IPI group.

2.2.5 Psychosocial Treatments: Case Management (CM) and Assertive Community Treatment (ACT) CM and ACT treatments refer to a class of similar services. Both involve some of the following: linkage with community resources, support of independent living skills, efforts to improve engagement or enhance practice of new skills, support of family members, and psychoeducation. ACT is typically carried out in the home or community and can include pharmaco- and psychotherapy. Neither component has been tested alone for effectiveness or efficacy in APS sam-

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

ples, but they have been included as components in many of the RCTs discussed here or in treatment-­as-usual conditions.

2.2.6 Social Skills Training (SST) Social skills training has a robust effect on functioning in established psychosis (Dixon et  al., 2010). It is typically conducted in groups and begins with explicit teaching of skills, procedures, and social conventions necessary for a range of interpersonal scenarios. Some packages also train participants how to recognize nonverbal cues. Participants then practice skills through role plays, often of situations from their own lives. SST has not been tested alone in an RCT of APS, but aspects of the formal SST models tested in established psychoses are included as a component in the integrative Family-Aided Assertive Community Treatment (FACT) (McFarlane et al., 2015), IPI (Bechdolf et  al., 2012), and OPUS (Petersen et al., 2005). 2.2.7 Psychosocial Treatments: Supported Employment/ Education (SEE) Another robust treatment in established psychoses (Dixon et al., 2010), SEE, focuses directly on improving work and school performance. Individuals work with a vocational or educational specialist to identify jobs or academic skills and complete the steps necessary to achieve them. Thus, individuals might receive tutoring, work on a resume, identify and attend job interviews together, etc. Aspects of SEE are a component of FACT.

2.3

Integrated Treatment

A few integrated treatment programs have been discussed thus far (McGorry et al., 2002, 2017; Yung et  al., 2011). Bechdolf and colleagues (2012) tested their Integrated Psychological Intervention (IPI), comprised of 25 CBT sessions, 15 sessions of group skills training (behavioral activation, social skills training, and problem solving training), computerized cognitive remediation, and three sessions of a psychoeduca-

51

tional multifamily group (manuals: Bechdolf, Veith, & Klosterkotter, 2006; Bechdolf, Puetzfeld, Gross, & Guettgemanns, 2010). IPI was compared with supportive counseling, consisting of psychoeducation and supportive therapy. The authors found that IPI resulted in lower conversion rates after the 12-month treatment period (2/63  =  3.2% vs. 11/65  =  16.9% in the control group), as well as 12 months later (4/63 = 6.3% vs. 13/65  =  20%). Those who did convert also converted more quickly in the supportive therapy group (887.1 vs. 784.2 days). McFarlane and colleagues (2015) recently reported 2-year outcomes of the six-site EDIPPP study, a quasi-experimental design study of the integrative Family-Aided Assertive Community Treatment (FACT) package. These authors compared the outcomes of 1 year of FACT, comprised of multifamily psychoeducation groups, pharmacotherapy, and features of ACT, and supported employment/education to monitoring and TAU in the community. This study was not an RCT, having used a regression discontinuity model in which participants were stratified into groups based on baseline APS severity (high and low estimated risk). Tests of regression discontinuity examined whether the clinical and functional trajectories of these groups differed from their respective predicted trajectories. This study is included in the current discussion because the MFG-P component of FACT is a commonly implemented treatment for early psychosis, particularly in the United States and Europe. The two groups included one with APS symptoms (n = 250) suggesting high risk (but not strict SIPS or CAARMS syndromes criteria) or recent onset (30  days or less) of psychotic-level symptoms and another group with less severe APS presumed to reflect lower risk (n  =  87). The high-­ risk group received FACT, and the low-risk group received monthly monitoring assessments and the option of community treatment. After 2 years and relative to the low-risk and TAU group, the APS group that received FACT treatment had statistically superior improvements in positive symptoms, work and school participation, and a clinical-functional composite. Conversion (6.3% APS, 2.3% control), negative event (25%, 22%)

D. I. Shapiro et al.

52

rates, and quality of life did not differ, in spite of baseline differences in risk. The 2-year attrition rate was 34%. In the OPUS trial, Nordentoft et  al. (2006) tested in the Danish Copenhagen and Aarhus counties an integrated intervention for preventing conversion from schizotypal disorder. Forty-two youths received up to 2 years of integrated treatment, consisting of aspects of ACT, SST, and MFG-P.  ACT was provided by a multidisciplinary team consisting of a psychiatrist, psychologist, psychiatric nurse, occupational therapist, and social worker, thus incorporated many elements of the treatments discussed in this section. Thirty-seven in the control condition received standard care in the same hospitals. After 2 years, 9/36 who received OPUS treatment (25%) converted to psychosis, compared with 14/29 (48.3%) of controls. Integrated treatment resulted in fewer antipsychotics, greater decline in negative symptoms at 1 year, but no other differences in symptom severity.

2.4

I nterventions Currently Being Tested

A number of RCTs for APS are currently being conducted. Interventions under investigation include Cognitive Behavioral Social Skills Training (Addington 2014), Social Recovery CBT (Gee et  al., 2016), Goal Management Training (clinicaltrials21), and a mindfulness cognition/well-being intervention (Langer et al., 2017); multiple cognitive remediation packages (Clinicaltrials2; clinicaltrials3; clinicaltrails20; Addington, 2012; Glenthøj et al., 2015; Hooker et al., 2014), including one developed ­specifically for Latino youth (clinicaltrials5); integrated treatment packages like PRIME (clinicaltrials4), a staged integrated treatment program (clinicaltrials.gov1), and transcranial direct current stimulation with virtual reality motivation training for negative symptoms (clinicaltrials8); pharmacological agents including aspirin (clinicaltrials6), gabapentin (clinicaltrials7), sertraline and risperidone combination (clinicaltrials10), omega-3 fatty acids (clinicaltrials11; 12; 13; Armando

et  al., 2016), fluoxetine vs. ariproprizole (clinicaltrials17), minocylcine (clinicaltrials16), pomaglumetad methionil to reduce glutamate (clinicaltrials19), N-Acetylcysteine (clinicaltrials22), and galantamine for cognition (clinicaltrials15); exercise (Mittal, 2015); and automatic clinical alerts for assessment of clinical high-risk patients in a standard care setting (clinicaltrials14). A group in Norway is testing the preventive effects of community-­ wide education on signs and symptoms of the psychosis prodrome (Joa, Gisselgard, Bronnick, McGlashan, & Johannessen, 2015). Many feasibility, pilot, or smaller trials have also been conducted or are underway. These, and the interventions described above, may represent the future of care for APS.  Some examples of innovative treatments include new pharmacological agents like D-serine for negative symptoms (Kantrowitz et al., 2015), Glycine (Woods et al., 2013), and lithium (Woods et al., 2007); modifications of established treatments into new mediums like CBTp-guided self-help (Naeem et  al., 2016), group-based CBT (Landa et  al., 2016), and cognitive enhancement therapy (clinicaltrials18); and technologically innovative interventions like video game-based biofeedback (Woodberry et al., 2014).

2.4.1 Limitations of RCTs While often considered the gold standard for empirically validating the relative effects of a specific treatment, RCTs have their limitations. Given the typically high cost in time, money, and sample size, RCTs often are not conducted until a treatment has shown strong evidence elsewhere, particularly in pharmacological trials. Rigorous testing of innovative treatments takes time, and their absence from RCTs does not necessarily mean that they are less effective. Methodological rigors can be hard to implement in real-world settings with complex and heterogeneous samples. Thus, the generalizability of RCT results to different individuals, settings, and presentations is often unknown (Westen, Novotny, & Thompson-­Brenner, 2004). At this point, there is no research to inform use of available treatments in currently understudied partici-

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

pants (e.g., individuals with low IQ, neuromedical complications, or significant suicide risk) or in very different cultural contexts or mental healthcare systems. Given some evidence that individuals with schizophrenia in low- and middle-income countries may actually do better than those in higher income countries, there is a need for careful consideration of both treatment and study designs in different sociocultural contexts (Padma, 2014). There may be a lot that Western medicine can learn from more socially oriented and less medically oriented models of care, particularly for social and functional outcomes.

3

Implementation

Understanding which treatments work, for whom, and how to deliver them in the community are different issues. No one implementation model can work across all geographical, community, or institutional borders, but cost-­effectiveness studies are showing that early intervention programs can sustain themselves in the short term and that APS programs save money in the long term, at least in societies with structured mental health systems.

3.1

Implementation Models

APS implementation models vary in scope. Those on the broader end, predominantly extant in countries or states with centralized public healthcare systems (e.g., Australia, Denmark, the National Health Services in the United Kingdom; see later chapters in this volume), rely on a clinical staging model where tiered levels of care are made available to the entire population. Increasingly specialized and focused services are provided by increasingly specialized clinicians as apparent specificity of and distress due to psychosis risk indicators increases. Other models use a specialty clinic model, where individuals are referred to psychosis risk centers that assess, treat, and interface with the community. Such clinics, themselves, vary some in how they sup-

53

port the services they provide, with some operating within academic frameworks and some in hospital or community health centers.

3.1.1 Stepped Care McGorry, Yung, and colleagues in Australia were among the first to develop systematic programs for identifying and treating those at psychosis risk (McGorry et  al., 2008; Yung & McGorry, 1996). Their efforts evolved from a secondary prevention specialty care center model (EPPIC, ORYGEN) to perhaps the clearest example of a nationalized stepped-care, or clinical staging model, that targets all youth mental health (Headspace; McGorry et  al., 2007). These authors describe their services in detail later in this book. In brief, the aim is to provide catered care for youth in whatever state of risk or distress they present at any moment, rather than exclusively to prevent specific outcomes like psychosis. This care begins with education initiatives targeted to the general population (lowest statistical risk). When difficulties develop, youth present to general adolescent mental health services which provide a broad spectrum of interventions. As they worsen or become more specific—as individuals “move” along the prototypical stages of the prodrome— they receive increasingly intensive and specialized treatment from increasingly specialized clinicians and centers. However, since most of the earliest manifestations of unfolding psychosis (e.g., anxiety, depression, attention problems) are non-specific, and most who present with such symptoms never develop psychosis, this model also allows for specialized care should other diagnoses develop, or step-down services as difficulties remit. Not only does this model allow for quicker monitoring and intervention; it negates issues related to the low precision of categorical psychosis risk labels. Treatment for current distress is not based on the presence of an APS, nor is it withheld because of its absence. On the other hand, such comprehensive models can be costly in terms of resources, trading higher preventive expenses for lower long-term healthcare and lost productivity costs. Thus, they also require a healthcare system in which current

54

payers (e.g., individuals, governments, insurance agencies) are the same entities that benefit financially from improved long-term outcomes.

D. I. Shapiro et al.

community typically offer psychiatry, individual, group, and/or family treatment. Both are typically also involved in some manner of community outreach to teach potential referrers (e.g., 3.1.2 Specialty Care Centers anybody in contact with teen and young adults) and Clinics the signs and symptoms of APS or to build a Specialty care services are typically cohesive referral base for the larger agency. Such educateams that target a particular condition or popula- tional campaigns have been shown to lead to bettion. APS specialty care was initially developed ter long-term outcomes (Heglstad et al., 2012). In in academic centers and tied to worldwide some stepped-care systems, like the UK trust sysresearch on APS. In recent years some have been tem, such specialty care teams are regionally developed in nonacademic clinical settings as located and may take on clients who “graduate” well. Specialized centers vary in the scope of ser- out of less intensive treatments because of vices they provide, largely related to the source increasing need or specificity of PR. and extent of funding, but also clinician experProliferation of specialized first episode psytise, ways of conceptualizing psychosis and its chosis programs outside of academia has been treatment targets, and agency mission. Services catalyzed by the large-scale Treatment and typically include community or provider educa- Intervention in Psychosis Study (TIPS; Heglstad tion, consultation, and/or screening to facilitate et al., 2012; Larsen et al., 2001), OPUS (Petersen accurate referrals and engagement of youth and et  al., 2005), and Recovery After an Initial families, structured assessment to determine psy- Schizophrenia Episode (RAISE; Kane et  al., chosis risk and diagnostic comorbidities, and 2016) trials. Each of these influential studies clinical feedback (standardized or individual- independently showed that early identification ized) regarding risk for psychosis and treatment (in this case soon after an initial episode) and/or indicators and options. Treatment planning is care provided in such programs led to better outlargely dependent on available treatments but comes than standard clinical care. While one cantypically includes collaborative and sometimes not assume that such results generalize to the gradual engagement initially focused on present- APS state, this early intervention movement has ing concerns and individualized goals over psy- led to the opening of similar APS clinics (e.g., chosis risk or prevention, per se. Adjunctive National Health Service in the United Kingdom, treatment services, such as psychiatry, family Headspace in Australia, California’s Mental therapy, residential or inpatient care, school-­ Health Services Act in the United States) or the based treatment, traditional cultural approaches, expansion of first episode specialty care services and case management, may be available through to include APS teams (e.g., EASA, OPUS, PIER). specialized teams, community providers, or clini- Some countries or systems with national healthcal trainees, even to participants of treatment tri- care systems or mixed private-public payer sysals. In some countries or regions, clinics are tems have developed networks of such specialty comprised solely of psychiatric services with early intervention centers. Australia’s Headspace assessment, education of the individual and fam- centers are a good model. These centers are ily, and medication serving as the central service. spread throughout the country and provide multiIn many western countries, university-linked disciplinary services catered to adolescents who clinics typically build around a core of research present with early mental health concerns. They studies that give youth access to continued moni- incorporate specialized services for APS but are toring and any experimental treatments being designed to be broader and more encompassing investigated. Often clinician trainees or study so that individuals at risk for or currently showstaff provide adjunctive treatment or coordina- ing a broad range of psychopathology can get tion of treatment with outside providers. In addi- expert care. This includes not only specialty care tion to assessment, specialty clinics in the services described above but general practice

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

doctors, substance use treatment, online services and social forums, and extra help with schoolwork. In the United States, the federal government recently passed legislation to “set aside” a certain portion of each of the 50 states’ mental health budgets to develop such teams, which typically include some combination of assessment and outcome tracking, individual therapy, case management, psychiatry, family treatment, supported education/employment, and often additional group therapy (Bello et al., 2016; Heinssen, Goldstein, & Azrin, 2014). Finally, as discussed in various chapters in the current volume, many other countries across Asia, Africa, Europe, and the Americas are opening clinics as academic, privately, or state-funded pilot projects, with the hope of expanding or duplicating such early psychosis teams. As such teams become sustainable, some have chosen to expand their services to include APS populations. Future research is needed to better understand effectiveness of such programs for APS.

3.2

Risk Stratification

Risk stratification models have been born out of necessity as well as caution. It is neither possible nor appropriate to provide treatment to all individuals in a population regardless of degree of risk. These resources and ethics suggest that careful consideration be given to delivering the right intensity of treatment to the right individuals at the right time. One approach to intervening in APS is to direct resources to those who appear to be at greatest risk for psychosis. Most centers conduct some type of initial screening prior to inviting referrals for a formal assessment. Screening thresholds are available for a number of self-report screens for individuals seeking help from specialized centers (e.g., Kline & Schiffman, 2014). However, they are not yet established for screening in general population samples such as schools, primary care, or online surveys or in diverse cultural or linguistic contexts. Empirical methods for risk enrichment, like probability calculators based off known predictors of conversion, are described elsewhere in this volume

55

(Chap. 2). Some examples include combining basic symptoms and CHR criteria (Schultze-­ Lutter, Klosterkotter, & Ruhrmann, 2014) and the NAPLS risk calculator combining demographics with select symptoms and cognitive test results (http://riskcalc.org:3838/napls/; Cannon et al., 2016). Using such enhanced sampling, specialized treatment can be prioritized for individuals who meet both sets of criteria. Over time, additional risk factors, including biological assays, may facilitate better allocation of treatment resources. Such stratification methods are currently in the research and development stages.

3.3

Cost-Effectiveness of Specialized Aps Programs

The movement to intervene early in the course of psychosis was born out of the recognition of the enormous burden major psychotic disorders can place on individuals, families, and communities—the emotional, financial, and economic costs. Economically, the World Health Organization estimates that care in Western countries for those with schizophrenia accounts for an amount equal to 7–12% of these nations’ gross national products (Barbato, 1998). This does not include losses in potential productivity/earnings in both probands and caretakers or increased reliance on other resources (e.g., social welfare) that could be saved. Moreover, delays in treatment for mental illness over the lifespan are associated with increased legal involvement, more hospitalizations, increased rates of substance use, homelessness, and comorbid medical illness (Insel, 2008). Thus, one may think that avoiding or lessening such costs via early intervention would be enough reason to support such services. But treatment, itself, is costly in terms of time and economic resources. Without unequivocal evidence that specialized interventions prevent or mitigate such outcomes, it is difficult to say whether investment in such intervention services will be less than the full costs of the status quo. This is particularly true given the vast variability in how psychosis is treated around the world, in the financial costs of illness and treatment, in

D. I. Shapiro et al.

56

who bears responsibility for these costs, and in perceptions of relative burden (e.g., treatment in a long-term inpatient hospital vs. lifelong care in the home by family and community members). A number of studies have attempted to evaluate whether treatment for acute and chronic psychosis is cost-effective in the long-term, from an individual and programmatic perspective. While the vast majority of these studies have focused on high-income countries, they have generally established that, globally, treatment of schizophrenia decreases economic burden on the individual and at a societal level (Chisholm et  al., 2008; Chong et al., 2016). Less has been written about early intervention for psychosis. Work in predominantly Western countries does suggest that investment in early psychosis intervention services like EPPIC in Australia (Mihalopoulos, Harris, Henry, Harrigan, & McGorry, 2009), OPUS in Denmark (Hastrup et al., 2013) and the trusts in England (Park, McCrone, & Knapp, 2016; Tsiachristas, Thomas, Leal, & Lennox, 2016) are associated with lower yearly treatment costs and higher levels of employment and well-­ being over roughly 10  years. Researchers in Hong Kong found that their program also resulted in fewer hospitalization costs (Wong et al., 2011). It is tempting to extrapolate from specialized treatment for first episode psychosis into the APS population. Unfortunately, little research is available to evaluate whether specialized services for APS are cost-effective in the long-term; only two studies have been published. Valmaggia and colleagues (2009) compared the costs of treatment of 114 individuals with APS treated in the Outreach and Support in South London (OASIS) program with costs of treatment in a first episode psychosis specialty care service, including treatment, hospitalization, assessment, lost employment, and primary care costs. Using a statistical model to compare, they found that OASIS services for APS were less costly (4396  lbs) after 2 years than were costs for those seen in the first episode clinic (5357), supporting the cost-­ effectiveness of APS intervention for the payer. In a similar study, Ising et  al. (2017) compared costs in the Dutch Early Detection and Intervention Evaluation (EDIE) study between

their CBT experimental group (n  =  95) and an APS group who received only routine care (n = 101). When computing costs associated with treatment, travel expenditure, and lost productivity over 4  years, the experimental group spent less on treatment (a difference of roughly US$5777), due in large part to a greater number presumed to have averted psychosis. Their model found a strong probability of services being more cost-effective if implemented on a large scale. Thus, evidence for the cost-effectiveness of APS services is promising, but more data and analyses are needed. It is likely to come as specialty services and youth-friendly stepped-care models continue to grow and expand around the world.

4

Conclusions and Future Directions

Work thus far, including both scientific research and community efforts to identify and treat youth with APS, has shown that intervention for APS is possible and clinically promising and can be financially viable. Fortunately or unfortunately, the current body of evidence does not yet point to a gold standard treatment or protocol. Taken together, existing experimental treatments do decrease conversion rates in the short term, a clinically meaningful outcome for youth in a critical developmental transition to adulthood. The ultimate goal is to have interventions that alter pathogenetic processes and improve long-term outcomes, including the possible prevention of psychosis. Much work is still needed to discover or demonstrate this. One important next step will be to understand not just which treatments reduce conversion rates or improve quality of life but the mechanisms by which these treatments work. Another piece will be to understand the role of different clinical presentations, trajectories, cultures and contexts, and outcomes. We need to better understand age, stage, and timing. Needs, abilities, and support systems differ through development, and it may be that different illness mechanisms are differentially at-play at different epochs of development. Cultural variation in what is valued or expected at different ages, too,

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence…

will affect fit between case and intervention. These pieces will inform personalized medicine approaches fit to specific individuals at specific ages and stages of illness. In part this will depend on a more nuanced science of individual differences. Which demographic, psychological, biological, and psychosocial factors moderate and mediate treatment response? One reason for only modest effects in treatment studies may be that treatments work for only subgroups of people or different treatments work for different people. Effects may be hidden in heterogenous samples. It may be that understanding responders and nonresponders will be most informative. Understanding the role of cultural factors in the presentation of illness, such as the meaning created in response to aberrant experiences, the impact of social context, and how more traditional and culturally linked family/community responses may affect them, may yield new insights into mechanisms, moderators, or mediators. Almost no work has been done to test the generality of treatments developed in mostly western populations to non-Western populations and settings or to experiment with community-­based interventions in Western societies. Similarly, little is known about “traditional” treatment techniques or how responding to new problems in life by increasing family and community involvement helps or does not. This is a more common response in some cultures than others but is often lumped in as “no treatment” or “treatment as usual,” if it is studied at all. Both psychological and sociological interventions may be useful in adapting to specific cultural variations in what is “normal” or “abnormal” or to changing perceptions of experience. We have a long way to go in understanding which factors are key to reducing risk. Variations in ­individualism/collectivism, spirituality, acculturation, and other factors are all worthy of investigation. Finally, we need to better understand protective mechanisms and factors that specifically improve well-being in those with APS. Indeed, it may be that mechanisms that decrease disability are not the same as those that produce it. This, too, will vary with culture; harnessing ethnocentric views on what it means to be “healthy” and what fosters well-being is integral.

57

Because intervention science is relatively young in APS samples, a number of methodological factors will need to be addressed to move the field forward. Very few studies have followed participants for longer than 1–2 years after an APS is identified. While it appears that most who develop psychosis do so within this window, longer follow-up and lower attrition are needed to identify predictors of long-term outcome and steps needed to maintain treatment gains. Because risk is not a static construct (and those at APS are young), the relative lack of long-term data limits our understanding about what happens after people remit. Related, prevention of psychosis is an important metric for success of an intervention, but from a clinical perspective, others seem more important. Quality of life, relationships, and functional ability are far more important to the individual and families affected by psychosis than whether or not a psychotic episode ever occurs. For example, one could maintain distressing sub-threshold symptoms for a lifetime, but it would be hard to argue that this is a more favorable outcome than one episode of psychotic-level symptoms followed by remission and a return to one’s normal life. Very few studies systematically measure or report such outcomes and those that do appear to have small effects. Testing cultural and other moderators of long-term outcomes will both require larger, potentially stratified samples. These will require increasing cooperation across otherwise competing institutions and networks. They will also require consistency in how those at risk are identified and how interventions are structured (e.g., what is included in CBT) across sites, countries, and cultures. It may be that help-­ seekers fare differently than non-help-seeking APS who are rarely studied. Strong characterizations of control conditions are just as critical to comparing findings across studies so differences do not obfuscate interpretation of findings and comparisons. Eventually, component analysis of packaged treatments will be needed to identify the relative contribution of different treatment components to outcomes of interest. Finally, balancing the need for both methodological rigor and ecological validity will be an important, even if challenging, aspect of this work going forward.

58

References

D. I. Shapiro et al.

Morrison (Eds.), Working with people at ultra-high risk of developing psychosis (pp. 169–181). New York: Wiley. Addington, J.  (2012). Effects of cognitive remediaBechdolf, A., Puetzfeld, V., Gross, S., Guettgemanns, J. tion on cognition in young people at clinical high (2010). Cognitive behavioural therapy in people at risk risk of psychosis. Identification No. NCT01619319. of psychosis. Huber. Retrieved from https://clinicaltrials.gov/ct2/show/ Bechdolf, A., Wagner, M., Ruhrmann, S., Harrigan, S., NCT01619319. Putzfeld, V., Pukrop, R., … Bottlender, R. (2012). Addington, J.  (2014). Cognitive behavioral social skills Preventing progression to first-episode psychosis training for youth at risk of psychosis. Identification in early initial prodromal states. British Journal of No. NCT02234258. Retrieved from https://clinicaltriPsychiatry, 200, 22–29. als.gov/ct2/show/NCT02234258. Bello, I., Lee, R., Malinovsky, I., Watkins, L., Nossel, I., Addington, J., Case, N., Saleem, M.  M., Auther, A.  M., Smith, T., … Radigan, M. (2016). OnTrackNY: The Cornblatt, B. A., & Cadenhead, K. S. (2014). Substance development of a coordinated specialty care prouse in clinical high risk for psychosis: A review of gram for individuals experiencing early psychosis. the literature. Early Intervention in Psychiatry, 8(2), Psychiatric Services, 68(4), 318–320. 104–112. Brent, B.  K., Thermenos, H.  W., Keshavan, M.  S., & Addington, J., Cornblatt, B.  A., Cadenhead, K.  S., Seidman, L.  J. (2013). Gray matter alterations in Cannon, T. D., McGlashan, T. H., Perkins, D. O., … schizophrenia high-risk youth and early-onset schizoHeinssen, R. (2011). At clinical high risk for psychophrenia: A review of structural MRI findings. Child sis: Outcome for nonconverters. American Journal of and Adolescent Psychiatric Clinics, 22, 689–714. Psychiatry, 168(8), 800–805. Cahn, W., van Haren, N. E., Hulshoff Pol, H. E., Schnack, Addington, J., Epstein, I., Liu, L., French, P., Boydell, H.  G., Caspers, E., Laponder, D.  A., & Kahn, R.  S. K. M., & Zipursky, R. B. (2011). A randomized con(2006). Brain volume changes in the first year of illtrolled trial of cognitive- behavioral therapy for indiness and 5-year outcome of schizophrenia. British viduals at clinical high risk of psychosis. Schizophrenia Journal of Psychiatry, 189(4), 381–382. Research, 125, 54–61. Cannon, M., Caspi, A., Moffit, T.  E., Harrington, H., Addington, J., Penn, D., Woods, S. W., Addington, D., & Taylor, A., Murray, R.  M., & Poulton, R. (2002). Perkins, D. O. (2008). Social functioning in individuEvidence for early-childhood, pan-developmental als at clinical high risk for psychosis. Schizophrenia impairment specific to schizophreniform disorder. Research, 99(1–3), 119–124. Archives of Geenral Psychiatry, 59(5), 449–456. Amminger, G. P., Schafer, M. R., Papageorgiou, K., Klier, Cannon, T.  D., Yu, C., Addington, J., Bearden, C.  E., C.  M., Cotton, S.  M., Harrigan, S.  M., … Berger, Cadenhead, K.  S., Cornblatt, B.  A., … Perkins, G. E. (2010). Long-chain omega-3 fatty acids for indiD.  O. (2016). An individualized risk calculator for cated prevention of psychotic disorders: A randomresearch in prodromal psychosis. American Journal of ized, placebo-­ controlled trial. Archives of General Psychiatry, 173(10), 980–988. Psychiatry, 67, 146–154. Chisholm, D., Gureje, O., Saldivia, S., Villalon Calderon, Armando, M., De Crescenzo, F., Vicari, S., Digilio, M. C., M., Wickremasinghe, R., Mendis, N., … Saxena, Pontillo, M., Papaleo, F., & Amminger, G. P. (2016). S. (2008). Schizophrenia treatment in the developIndicated prevention with long-chain polyunsaturated ing world: An interregional and multinational cost-­ omega-3 fatty acids in patients with 22q11DS genetieffectiveness analysis. Bulletin of World Health cally at high risk for psychosis. Protocol of a randomOrganization, 86(7), 542–551. ized, double-blind, placebo-controlled treatment trial. Chong, H.  Y., Teoh, S.  K., Wu, D.  B.-C., Kotirum, S., Early Intervention in Psychiatry, 10(5), 390–396. Chiou, S.-F., & Chaiyakunapruk. (2016). Global Arongo, C., Rapado-Castro, M., Reig, S., Castro-­ economic burden of schizophrenia: A systematic Fornieles, J., González-Pinto, A., Otero, S., … review. Neuropsychiatric Disease and Treatment, 12, Parellada, M. (2012). Progressive brain changes in 357–373. children and adolescents with first-episode psychosis. Correll, C.  U., Manu, P., Olshanskiy, V., Napolitano, Archives of General Psychiatry, 69(1), 16–26. B., Kane, J.  M., & Malhotra, A.  K. (2009). Barbato, A. (1998). Schizophrenia and public health. Cardiometabolic risk of second-generation antipsyGeneva, Switzerland: World Health Organization. chotic medications during first-time use in children Bechdolf, A., Pukrop, R., Köhn, D., Tschinkel, S., Veith, and adolescents. Journal of the American Medical V., Schultze-Lutter, F., … Klosterkötter, J.  (2005). Association, 302(16), 1765–1773. Subjective quality of life in subjects at risk for a first de Souza, M. A., Salum, G. A., Jarros, R. B., Isolan, L., episode of psychosis: A comparison with first epiDavis, R., Knijnik, D., … Heldt, E. (2013). Cognitive-­ sode schizophrenia patients and healthy controls. behavioral group therapy for youth with anxiety disSchizophrenia Research, 79(1), 137–143. orders in the community: Effectiveness in low and Bechdolf, A., Veith, V., & Klosterkotter, J. (2006). Group middle income countries. Behavioral and Cognitive intervention for people at high risk of developing Psychotherapy, 41(3), 255–264. psychosis. In J.  Addington, S.  H. Francey, & A.  P.

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence… DeLisi, L.  E., Szulc, K.  U., Bertisch, H.  C., Majcher, M., & Brown, K. (2006). Understanding structural brain changes in schizophrenia. Dialogues in Clinical Neuroscience, 8(1), 71–78. Deutsche Geselleschaft fur Psychiatrie, Psychotherapie und Nervenheilkunde. (2006). Behandlungsleitlinie Schizophrenie. Darmstadt, Germany: Steinkopff Verlag. Dixon, L. B., Dickerson, F., Bellack, A. S., Bennett, M., Dickinson, D., Goldberg, R.  W., … Peer, J.  (2010). The 2009 Schizophrenia PORT treatment recommendations and summary statements. Schizophrenia Buletin, 36(1), 48–70. Farooq, S., Large, M., Nielssen, O., & Waheed, W. (2009). The relationship between the duration of untreated psychosis and outcome in low-and-middle income countries: A systematic review and meta-analysis. Schizophrenia Research, 109(1–3), 15–23. Francey, S. M., Nelson, B., Thompson, A., Parker, A. G., Kerr, M., Macneil, C., … McGorry, P. D. (2010). Who needs antipsychotic medication in the earliest stages of psychosis? A reconsideration of benefits, risks, neurobiology and ethics in the era of early intervention. Schizophrenia Research, 119(1–3), 1–10. French, P., & Morrison, A. P. (2004). Early detection and cognitive therapy for people at high risk of developing psychosis. Chichester, UK: Wiley. Fusar-Poli, P., Bonaldi, I., Yung, A.  R., Borgwardt, S., Kempton, M. J., Valmaggia, L., … McGuire, P. (2012). Predicting psychosis; meta-analysis of transition outcomes in individuals at high clinical risk. Archives of General Psychiatry, 69(3), 220–229. Fusar-Poli, P., Borgwardt, S., Bechdolf, A., Addington, J., Riecher-Rössler, A., Schultze-Lutter, F., … Valmaggia, L. (2013). The psychosis high- risk state. A comprehensive state-of-the-art review. JAMA Psychiatry, 70(1), 107–120. Fusar-Poli, P., Diaz-Caneja, C. M., Patel, R., Valmaggia, L., Byrne, M., Garety, P., … McGuire, P. (2016). Services for people at high risk improve outcomes in patients with first episode psychosis. Acta Psychiatry Scandinavica, 133(1), 76–85. Fusar-Poli, P., Smieskova, R., Kempton, M.  J., Ho, B.  C., Andreasen, N.  C., & Borgwardt, S. (2013). Progressive brain changes in schizophrenia related to antipsychotic treatment? A meta-analysis of longitudinal MRI studies. Neuroscience and Biobehavioral Reviews, 37(8), 1680–1691. Gee, B., Notley, C., Byrne, R., Clarke, T., Hodgekins, J., French, P., & Fowler, D. (2016). Young people’s experiences of social recovery cognitive behavioural therapy and treatment as usual in the PRODIGY trial. Early Intervention in Psychiatry, 12, 879–885. Glenthøj, L.  B., Fagerlund, B., Randers, L., Hjorthøj, C.  R., Wenneberg, C., Krakauer, K., … Nordentoft, M. (2015). The FOCUS trial: Cognitive remediation plus standard treatment versus standard treatment for patients at ultra-high risk for psychosis: Study protocol for a randomised controlled trial. Trials, 16, 25.

59

Green, M. F., Kern, R. S., Braff, D. L., & Mintz, J. (2000). Neurocognitive deficits and functional outcome in schizophrenia. Schizophrenia Bulletin, 26, 119–136. Hastrup, L. H., Kronborg, C., Bertelsen, M., Jeppesen, P., Jorgensen, P., Petersen, L., … Nordentoft, M. (2013). Cost-effectiveness of early intervention in first-­ episode psychosis: Economic evaluation of a randomized controlled trial (the OPUS study). British Journal of Psychiatry, 202(1), 35–41. Heglstad, W., Larsen, T.  K., Auestad, B., Evensen, J., Haahr, U., Joa, I., … Rossberg, J. I. (2012). Long-term follow-up of the TIPS early detection in psychosis study: Effects on 10-year outcome. American Journal of Psychiatry, 169(4), 374–380. Heinssen, R. K., Goldstein, A. B., & Azrin, S. T. (2014). Evidence-based treatments for first episode psychosis: Components of coordinated specialty care. Bethesda, MD: National Institute of Mental Health. Heinssen, R.  K., & Insel, T.  R. (2015). Preventing the onset of psychosis: Not quite there yet. Schizophrenia Bulletin, 41, 28–29. Ho, B.  C., Andreasen, N.  C., Ziebell, Pierson, R., & Magnotta, V. (2011). Long-term antipsychotic treatment and brain volumes; a longitudinal study of first-episode schizophrenia. Archives of General Psychiatry, 68(2), 128–137. Holzer, L., Urben, S., Passini, C. M., Jaugey, L., Herzog, M.  H., Halfon, O., & Pihet, S. (2014). A randomized controlled trial of the effectiveness of computer-­ assisted cognitive remediation (CACR) in adolescents with psychosis or at high risk of psychosis. Behavioral and Cognitive Psychotherapy, 42(4), 421–434. Hooker, C.  I., Carol, E.  E., Eisenstein, T.  J., Yin, H., Lincoln, S. H., Tully, L. M., … Seidman, L. J. (2014). A pilot study of cognitive training in clinical high risk for psychosis: Initial evidence of cognitive benefit. Schizophrenia Research, 157, 314–316. Insel, T. R. (2008). Assessing the economic costs of serious mental illness. American Journal of Psychiatry, 165(6), 663–665. International Early Psychosis Association Writing Group. (2005). International clinical practice guidelines for early psychosis. British Journal of Psychiatry Supplement, 48, s120–s124. Ising, H. K., Kraan, T. C., Rietdijk, J., Dragt, S., Klaasen, R.  M. C., Boonstra, N., … Wunderink, L. (2016). Four-year follow-up of Cognitive Behavioral Therapy in persons at ultra-high risk for developing psychosis: The Dutch Early Detection Intervention Evaluation (EDIE-NL) trial. Schizophrenia Bulletin, 42(5), 1243–1252. Ising, H. K., Lokkerbol, J., Rietdijk, J., Dragt, S., Klaassen, R. M., Kraan, T., … Wunderink, L. (2017). Four-year cost-effectiveness of cognitive behavioral therapy for prevention first-episode psychosis: The Dutch Early Detection Intervention Evaluation (EDIE-NL) trial. Schizophrenia Bulletin, 43(2), 365–374. Joa, I., Gisselgard, J., Bronnick, K., McGlashan, T., & Johannessen, J. O. (2015). Primary prevention of psy-

60 chosis through interventions in the symptomatic prodromal phase, a pragmatic Norwegian ultra high risk study. BMC Psychiatry, 15, 89. Kane, J. M., Robinson, M. D., Schooler, N. R., Mueser, K.  T., Penn, D.  L., Rosenheck, R.  A., … Marcy, P. (2016). Comprehensive versus usual community care for first episode psychosis: Two-year outcomes from the NIMH RAISE early treatment program. American Journal of Psychiatry, 173(4), 362–372. Kantrowitz, J.  T., Woods, S.  W., Petkova, E., Cornblatt, B., Corcoran, C. M., Chen, H., … Javitt, D. C. (2015). D-serine for the treatment of negative symptoms in individuals at clinical high risk of schizophrenia: A pilot, double-blind, placebo-controlled, randomised parallel group mechanistic proof-of-concept trial. Lancet Psychiatry, 2, 403–412. Kline, E., & Schiffman, J. (2014). Psychosis risk screening: A systematic review. Schizophrenia Research, 158, 11–18. Klosterkötter, J., Hellmich, M., Steinmeyer, E.  M., & Schultze-Lutter, F. (2001). Diagnosing schizophrenia in the initial prodromal phase. Archives of General Psychiatry, 58(2), 158–164. Landa, Y., Mueser, K.  T., Wyka, K.  E., Shreck, E., Jespersen, R., Jacobs, M. A., … Walkup, J. T. (2016). Development of a group and family-based cognitive behavioural therapy program for youth at risk for psychosis. Early Intervention in Psychiatry, 10(6), 511–521. Langer, A. I., Schmidt, C., Mayol, R., Diaz, M., Lecaros, J., Krogh, E., … Villar, M. J. (2017). The effect of a mindfulness-based intervention in cognitive functions and psychological well-being applied as an early intervention in schizophrenia and high-risk mental state in a Chilean sample: Study protocol for a randomized controlled trial. Trials, 18(1), 233. Larsen, T. K., McGlashan, T. H., Johannessen, J. O., Friis, S., Guldberg, C., Haahr, U., … Simonsen, E. (2001). Shortened duration of untreated first episode of psychosis: Changes in patient characteristics at treatment. American Journal of Psychiatry, 158, 1917–1919. Lieberman, J., Chakos, M., Wu, H., Alvir, J., Hoffman, E., Robinson, D., & Bilder, R. (2001). Longitudinal study of brain morphology in first episode schizophrenia. Biological Psychiatry, 49(6), 487–499. Lieberman, J.  A., Perkins, D., Belger, A., Chakos, M., Jarskog, F., Boteva, K., & Gilmore, J.  (2001). The early stages of schizophrenia: Speculations on pathogenesis, pathophysiology, and therapeutic approaches. Biological Psychiatry, 50(11), 884–897. Liu, C. H., Keshavan, M. S., Tronick, E., & Seidman, L. J. (2015). Perinatal risks and childhood premorbid indicators of later psychosis: Next steps for early psychosocial interventions. Schizophrenia Bulletin, 41(4), 801–816. Loewy, R., Fisher, M., Schlosser, D.  A., Biagianti, B., Stuart, B., Mathalon, D. H., & Vinogradov, S. (2016). Intensive auditory cognitive training improves verbal memory in adolescents and young adults at clini-

D. I. Shapiro et al. cal high risk for psychosis. Schizophrenia Bulletin, 42(S1), S118–S126. McFarlane, W.  R. (2002). Multifamily groups in the treatment of severe psychiatric disorders. New York: Guildford Press. McFarlane, W. R. (2016). Family interventions for schizophrenia and the psychoses: A review. Family Process, 55(3), 460–482. McFarlane, W.  R., Levin, B., Travis, L., Lucas, F.  L., Lynch, S., Verdi, M., … Cornblatt, B. (2015). Clinical and functional outcomes after 2 years in the early detection and intervention for the prevention of psychosis multisite effectiveness trial. Schizophrenia Bulletin, 41(1), 30–43. McGlashan, T. H., Zipursky, R. B., Perkins, D., Addington, J., Miller, T., Woods, S. W., … Addington, D. (2006). Randomized, double-blind trial of olanzapine versus placebo in patients prodromally symptomatic for psychosis. American Journal of Psychiatry, 163, 790–799. McGorry, P. (2015). Staged treatment in early psychosis (STEP): A sequential multistage randomized clinical trial (SMART) of interventions for ultra high risk (UHR) of psychosis patients. Identification No. NCT02751632. Retrieved from https://clinicaltrials. gov/ct2/show/NCT02751632. McGorry, P. D., Killackey, E., & Yung, A. (2008). Early intervention in psychosis: Concepts, evidence and future directions. World Psychiatry, 7(3), 148–156. McGorry, P. D., Nelson, B., Amminger, G. P., Bechdolf, A., Francey, S.  M., Berger, G., … Nordentoft, M. (2009). Intervention in individuals at ultra-high risk for psychosis: A review and future directions. Journal of Clinical Psychiatry, 70(9), 1206–1212. McGorry, P.  D., Nelson, B., Markulev, C., Yuen, H.  P., Schäfer, M. R., Mossaheb, N., … Chen, E. Y. (2017). Effect of ω-3 polyunsaturated fatty acids in young people at ultrahigh risk for psychotic disorders. JAMA Psychiatry, 74(1), 19–27. McGorry, P.  D., Nelson, B., Phillips, L.  J., Yuen, H.  P., Francey, S.  M., Thampi, A., … Thompson, A.  D. (2013). Randomized controlled trial of interventions for young people at ultra-high risk of psychosis: Twelve-month outcome. Journal of Clinical Psychiatry, 74(4), 349–356. McGorry, P.  D., Tanti, C., Stokes, R., Hickie, I.  B., Carnell, K., Littlefield, L.  K., & Moran, J.  (2007). Headspace: Australia’s national youth mental health foundation—Where young minds come first. Medical Journal of Australia, 187(7S), S68–S70. McGorry, P. D., Yung, A. R., Phillips, L. J., Yuen, H. P., Francey, S., Cosgrave, E.  M., … Adlard, S. (2002). Randomized controlled trial of interventions designed to reduce the risk of progression to first-episode psychosis in a clinical sample with subthreshold symptoms. Archives of General Psychiatry, 59, 921–928. McKenna, P., & Kingdon, D. (2014). Has cognitive behavioural therapy for psychosis been oversold? BMJ, 348, g2295.

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence… Mihalopoulos, C., Harris, M., Henry, L., Harrigan, S., & McGorry, P. (2009). Is early intervention in psychosis cost-effective over the long term? Schizophrenia Bulletin, 35(5), 909–918. Miklowitz, D.  J., O’Brien, M.  P., Schlosser, D.  A., Addington, J., Candan, K.  A., Marshall, C., … Friedman-Yakoobian, M. (2014). Family-focused treatment for adolescents and young adults at high risk for psychosis: Results of a randomized trial. Journal of American Academic Child and Adolescent Psychiatry, 53, 848–858. Mitchell, A. J., Vancampfort, D., De Herdt, A., Yu, W., & De Hert, M. (2013). Is the prevalence of metabolic syndrome and metabolic abnormalities increased in early schizophrenia? A comparative meta-analysis of first episode, untreated and treated patients. Schizophrenia Bulletin, 39(2), 295–305. Mittal, V.  A. (2015). Exercise-based cognitive remediation for youth at ultra high-risk for psychosis. Schizophrenia Bulletin, 41(Suppl 1), S324–S325. Morrison, A.  P., & Barratt, S. (2010). What are the components of CBT for psychosis? A Delphi study. Schizophrenia Bulletin, 36(1), 136–142. Morrison, A.  P., French, P., Parker, S., Roberts, M., Stevens, H., Bentall, R.  P., & Lewis, S.  W. (2007). Three-year follow-up of a randomized controlled trial of cognitive therapy for the prevention of psychosis in people at ultrahigh risk. Schizophrenia Bulletin, 33(3), 682–687. Morrison, A.  P., French, P., Walford, L., Lewis, S.  W., Kilcommons, A., Green, J., … Bentall, R. P. (2004). Cognitive therapy for the prevention of psychosis in people at ultra-high risk: Randomized controlled trial. British Journal of Psychiatry, 185, 291–297. Mueser, K. T., Deavers, F., Penn, D. L., & Cassisi, J. E. (2013). Psycho-social treatments for schizophrenia. Annual Review of Clinical Psychology, 9, 465. Naeem, F., Johal, R., McKenna, C., Rathod, S., Ayub, M., Lecomte, T., … Farooq, X. (2016). Cognitive behavior therapy for psychosis based guided self-help (CBTp-­ GSH) delivered by frontline mental health professionals: Results of a feasibility study. Schizophrenia Research, 173(1–2), 69–74. National Institute for Health and Clinical Excellence (NICE). (2014). Psychosis and schizophrenia in adults: Prevention and management. Retrieved from NICE guidance https://www.nice.org.uk/Guidance/CG178. Nordentoft, M., Thorup, A., Petersen, L., Øhlenschlæger, J., Melau, M., Christensen, T. Ø., … Jeppesen, P. (2006). Transition rates from schizotypal disorder to psychotic disorder for first- contact patients included in the OPUS trial. A randomized clinical trial of integrated treatment and standard treatment. Schizophrenia Research, 83, 29–40. Padma, T. V. (2014). Developing countries: The outcomes paradox. Nature, 508, S14–S15. Park, A.  L., McCrone, P., & Knapp, M. (2016). Early intervention for first-episode psychosis: Broadening

61

the scope of economic estimates. Early Intervention in Psychiatry, 10(2), 144–151. Penttilä, M., Jääskeläinen, E., Hirvonen, N., Isohanni, M., & Miettunen, J.  (2014). Duration of untreated psychosis as predictor of long-term outcome in schizophrenia: Systematic review and meta-analysis. British Journal of Psychiatry, 205(2), 88–94. Petersen, L., Nordentoft, M., Jeppesen, P., Ohlenschlaeger, J., Thorup, A., & Ostergaard, T. (2005). Improving 1-year outcome in first-episode psychosis; OPUS trial. British Journal of Psychiatry, 187(48), s98–s103. Piskulic, D., Barbato, M., Liu, L., & Addington, J. (2015). Pilot study of cognitive remediation therapy on cognition in young people at clinical high risk of psychosis. Psychiatry Research, 225, 93–98. Piskulic, D., Liu, L., Cadenhead, K.  S., Cannon, T.  D., Cornblatt, B. A., McGlashan, T. H., … Woods, S. W. (2016). Social cognition over time in individuals at clinical high risk for psychosis: Findings from the NAPLS-2 cohort. Schizophrenia Research, 171(1–3), 176–181. Rauchensteiner, S., Kawohl, W., Ozgurdal, S., Littmann, E., Gudlowski, Y., Witthaus, H., … Juckel, G. (2011). Test performance after cognitive training in persons at risk mental state of schizophrenia and patients with schizophrenia. Psychiatry Research, 185, 334–339. Riecher-Rössler, A., Pflueger, M. O., Aston, J., Borgwardt, S.  J., Brewer, W.  J., Gschwandtner, U., & Stieglitz, R.  D. (2009). Efficacy of using cognitive status in predicting psychosis: A 7-year follow-up. Biological Psychiatry, 66(11), 1023–1030. Ruhrmann, S., Schultze-Lutter, F., & Klosterkotter, J.  (2010). Probably at-risk, but certainly ill— Advocating the introduction of a psychosis spectrum disorder in DSM-V. Schizophrenia Research, 120(1– 3), 23–37. Schizophrenia Working Group of the Psychiatric Genomics Consortium. (2014). Biological insights from 108 schizophrenia-associated genetic loci. Nature, 511, 421–427. Schmidt, S. J., Schultze-Lutter, F., Schimmelmann, B. G., Maric, N. P., Salokangas, R. K., Riecher-Rössler, A., … Morrison, A. (2015). EPA guidance on the early intervention in clinical high risk states of psychoses. European Psychiatry, 30, 388–404. Schultze-Lutter, F., Klosterkotter, J., & Ruhrmann, S. (2014). Improving the clinical prediction of psychosis by combining ultra-high risk criteria and cognitive basic symptoms. Schizophrenia Research, 154(1–3), 100–106. Seidman, L. J., Shapiro, D. I., Stone, W. S., Woodberry, K. A., Ronzio, A., Cornblatt, B. A., … Woods, S. W. (2016). Neurocognition and transition to psychosis: Baseline functioning in the second phase of the North American Prodrome Longitudinal Study. JAMA Psychiatry, 73(12), 1239–1248. Singh, F., DeJoseph, M., & Cadenhead, K.  S. (2012). Therapeutic considerations in individuals at clini-

62 cal risk for developing psychosis. Fortschritte de Neurologie- Psychiatrie, 80, 570–579. Solakangas, R. K., Ruhrmann, S., von Reventlow, H. G., Heinimaa, M., Svirskis, T., From, T., … Birchwood, M. (2012). Axis I diagnoses and transition to psychosis in clinical high-risk patients EPOS project: Prospective follow-up of 245 clinical high-risk outpatients in four countries. Schizophrenia Resesarch, 138(2–3), 192–197. Stafford, M. R., Jackson, H., Mayo-Wilson, E., Morrison, A. P., & Kendall, T. (2013). Early interventions to prevent psychosis: Systematic review and meta-analysis. BMJ, 346, f185. Stain, H. J., Bucci, S., Baker, A. L., Carr, V., Emsley, R., Halpin, S., … Startup, M. (2016). A randomised controlled trial of cognitive behaviour therapy versus non-­ directive reflective listening for young people at ultra high risk of developing psychosis: The detection and evaluation of psychological therapy (DEPTh) trial. Schizophrenia Research, 176(2–3), 212–219. Thomas, N. (2015). What’s really wrong with cognitive behavioral therapy for psychosis? Frontiers in Psychology, 6, 323. Tienari, P., Wynne, L.  C., Moring, J., Lahti, I., Naarala, M., Sorri, A., … Läksy, K. (1994). The Finnish adoptive family study of schizophrenia. Implications for family research. British Journal of Psychiatry Suppl, 23, 20–26. Tsiachristas, A., Thomas, T., Leal, J., & Lennox, B.  R. (2016). Economic impact of early intervention in psychosis services: Results from a longitudinal retrospective controlled study in England. BMJ Open, 6(10), e012611. Tsuang, M. (2000). Schizophrenia: Genes and environment. Biological Psychiatry, 47(3), 210–220. Valmaggia, L. R., McCrone, P., Knapp, M., Woolley, J. B., Broome, M.  R., Tabraham, P., … Power, P. (2009). Economic impact of early intervention in people at high risk of psychosis. Psychological Medicine, 39, 1617–1626. van der Gaag, M., Nieman, D.  H., Rietdijk, J., Dragt, S., Ising, H.  K., Klaassen, R.  M., … Linszen, D.  H. (2012). Cognitive- behavioral therapy for subjects at ultra high risk for developing psychosis: A randomized controlled clinical trial. Schizophrenia Bulletin, 38, 1180–1188. Van der Gaag, M., Smit, F., Bechdolf, A., French, P., Linszen, D.  H., Yung, A.  R., … Cuijpers, P. (2013). Preventing a first episode of psychosis: Meta-analysis of randomized controlled prevention trials of 12 months and longer-term follow-ups. Schizophrenia Research, 149, 56–62. Vaughn, C. E., & Leff, J. P. (1976). The influence of family and social factors on the course of psychiatric illness: A comparison of schizophrenic and depressed neurotic patients. British Journal of Psychiatry, 129, 125–137. Velthorst, E., Nieman, D.  H., Linszen, D., Becker, H., de Haan, L., Dingemans, P. M., … Heinz, A. (2010).

D. I. Shapiro et al. Disability in people clinically at high risk of psychosis. British Journal of Psychiatry, 197(4), 278–284. Walker, E., Trotman, H., Pearce, B., Addington, J., Cadenhead, K. S., Cornblatt, B. A., … Tsuang, M. T. (2013). Cortisol levels and risk for psychosis: Initial findings from the North American Longitudinal Study. Biological Psychiatry, 74(6), 410–417. Walker, E. F., Savoie, T., & Davis, D. (1994). Neuromotor precursors of schizophrenia. Schizophrenia Bulletin, 20(3), 441–451. Westen, D., Novotny, C.  M., & Thompson-Brenner, H. (2004). The empirical status of empirically supported psychotherapies: Assumptions, findings, and reporting in controlled clinical trials. Psychological Bulletin, 130(4), 631–663. Wong, K. K., Chan, S. K., Lam, M. M., Hui, C. L., Hung, S. F., Tay, M., … Chen, E. Y. (2011). Cost-effectiveness of an early assessment service for young people with early psychosis in Hong Kong. Australian and New Zealand Journal of Psychiatry, 45(8), 673–680. Woodberry, K. A., Seidman, L. J., Bryant, C., Addington, J., Bearden, C.  E., Cadenhead, K., … Woods, S.  W. (2018). Treatment precedes positive symptoms in North American adolescent and young adult clinical high risk cohort. Journal of Clinical Child and Adolescent Psychology, 47(1), 69–78. Woodberry KA, Serur RA, Hallinan S, Mesholam-Gately RI, Giuliano AJ, Wojcik JD, …, Seidman LJ (2014). Integrating biofeed- back games and family therapy to reduce risk for psychosis. Poster presented at the Harvard Medical School Department of Psychiatry Research Day, Boston. Woodberry, K. A., Shapiro, D. I., Bryant, C., & Seidman, L. J. (2016). Progress and future directions in research on the psychosis prodrome: A review for clinicians. Harvard Review of Psychiatry, 24(2), 87–103. Woods, S. J., Berger, G. E., Dell’Olio, M., Hamer, C. A., Wellard, R.  M., Pell, G.  S., … Pantelis, C. (2007). Effects of low dose lithium on hippocampal neuropathology in people at ultra-high risk for psychosis. Biological Psychiatry, 61, 259s–260s. Woods, S. W., Addington, J., Bearden, C. E., Cadenhead, K. S., Cannon, T. D., Cornblatt, B. A., … Walker, E. F. (2013). Psychotropic medication use in youth at high risk for psychosis: Comparison of baseline data from two research cohorts 1998–2005 and 2008–2011. Schizophrenia Research, 148(1–3), 99–104. Woods, S.  W., Walsh, B., Hawkins, K.  A., Miller, T.  J., Saksa, J.  R., D’Souza, D.  C., … Krystal, J.  H. (2013). Glycine treatment of the risk syndrome for psychosis: Report of two pilot studies. European Neuropsychopharacology, 23(8), 931–940. Yang, L.  H., Wonpat-Borja, A.  J., Opler, M.  G., & Corcoran, C.  M. (2010). Potential stigma associated with inclusion of psychosis risk syndrome in the DSM-V: An empirical question. Schizophrenia Research, 120, 42–48. Yung, A. R. (2017). Treatment of people at ultra-high risk for psychosis. World Psychiatry, 16(2), 207–208.

3  Intervention Strategies for Attenuated Psychosis Syndromes: A Review of Current Practice, Evidence… Yung, A.  R., & McGorry, P.  D. (1996). The prodromal phase of first-episode psychosis: Past and current conceptualizations. Schizophrenia Bulletin, 22(2), 353–370. Yung, A.  R., Phillips, L.  J., Nelson, B., Francey, S.  M., PanYuen, H., Simmons, M. B., … Berger, G. (2011). Randomized controlled trial of interventions for young

63

people at ultra high risk for psychosis: 6-month analysis. Journal of Clinical Psychiatry, 72(4), 430–440. Yung, A. R., Yuen, H. P., Berger, G., Francey, S., Hung, T.  C., Nelson, B., … McGorry, P. (2007). Declining transition rate in ultra-high risk (prodromal) services: Dilution or reduction of risk? Schizophrenia Bulletin, 33(6), 73–81.

Part II Conceptual and Measurement Foundations in Attenuated Psychosis Syndromes

4

Attenuated Psychosis Syndromes Among Australian Youth and Young Adults: Early Identification and Intervention Barnaby Nelson and Patrick D. McGorry

Editors’ Note  A number of paradigms exist for prospectively identifying individuals who have elevated risk for developing psychosis due to the presence of syndromes comprised of identifiable risk factors and risk indicators. Conventions for which models are used, how individuals are identified, and which terminologies predominate vary throughout the world, sometimes related to culturally linked factors. In order to capture this diversity within one volume, the term Attenuated Psychosis Syndromes (APS) is used here to collectively refer to the class of putative prodromal or psychosis risk syndromes that have been empirically validated. We acknowledge that this is not a universally accepted convention, but use it as an umbrella term due to its heuristic value.

1

Introduction

Australia has been a pioneer and leader in the early detection and intervention for psychosis. Clinical and research endeavors into the early treatment of psychotic disorders have existed in Australia since the early 1980s. At that point, B. Nelson (*) · P. D. McGorry Orygen, The National Centre of Excellence in Youth Mental Health, Parkville, VIC, Australia Centre for Youth Mental Health, The University of Melbourne, Parkville, VIC, Australia e-mail: [email protected]; [email protected]

McGorry and colleagues focused their attention on comprehensively treating people in the early phase of illness in an effort to reduce disability and potentially prevent progression to more chronic states (McGorry, Edwards, Mihalopoulos, Harrigan, & Jackson, 1996). This early work led to the establishment in 1992 of the first episode psychosis service, the Early Psychosis Prevention and Intervention Centre (EPPIC) (Killackey & Yung, 2007). From this start, and with like-­ minded colleagues around the world (Edwards & McGorry, 2002), the early intervention paradigm has grown rapidly from a revolutionary idea in mental health (Killackey, Yung, & McGorry, 2007) to an accepted part of mainstream service provision in many countries. This chapter will consider the progress of early detection and intervention in Australia since those first efforts, focusing on the pre-onset phase of disorder.

2

History

2.1

 he Development of Early T Psychosis Services in Australia

McGorry and colleagues began clinical and research work specifically with people with first episode and recent-onset psychosis in the early to mid-1980s in the Aubrey Lewis Unit at Royal Park Hospital in Melbourne (McGorry et  al., 1996). From their work in this unit, they developed an

© Springer Nature Switzerland AG 2019 H. Li et al. (eds.), Handbook of Attenuated Psychosis Syndrome Across Cultures, https://doi.org/10.1007/978-3-030-17336-4_4

67

B. Nelson and P. D. McGorry

68

understanding of what elements would be necessary for establishing a fully fledged specialized service for first episode psychosis. Such a unit would require not just a focus on identification, but the development of recovery-focused interventions tailored to this early stage of illness, including both pharmaceutical and psychosocial elements. A focus specifically on young people was another element. This involved recognition of the importance of the developmental challenges of adolescence and young adulthood as well as the critical role of the family. The lessons learned in the Aubrey Lewis Unit paved the way for the development of Early Psychosis and Prevention and Intervention Centre (EPPIC). EPPIC was established as a clinical research center where clinical insights would drive research which in turn would lead to improvement in treatments and service delivery. From the outset, the clinic’s model of care was based on providing community-based rather than inpatient-­ based treatment delivered via a mobile outreach team able to offer support to patients and their families in the community. Each patient had a case manager who was also their primary therapist. In addition, medical management and group programs were provided. The goals of the service were early detection of psychosis, prevention of secondary morbidity, and maintenance of social and occupational functioning during the early critical period of a psychotic disorder (McGorry et al., 1996). In order to promote early identification of first episode psychosis beyond northwestern Melbourne, an education and knowledge transfer component of EPPIC (“EPPIC Statewide Services”) was established. EPPIC Statewide Services aimed to provide education about early detection and intervention across the state of Victoria, including workers in adult and children mental health services, general practitioners, school counsellors, and so on. This strategy ensured that the knowledge that was being generated at EPPIC was disseminated to a wide range of stakeholders. A further knowledge transfer activity in conjunction with the University of Melbourne was the development of a Graduate Diploma in Young People’s Mental Health. This course was offered via distance learning and

p­rovided videotaped lectures by experts on a wide range of topics, allowing clinicians across Australia to develop knowledge and skills in this area. Many of the graduates of this course have become leaders of early identification and intervention in their local health services. Statewide Services evolved over time into Orygen’s Skills and Knowledge Division (https://orygen.org.au/ Skills-Knowledge). The EPPIC model was found to reduce the duration of untreated psychosis (DUP) from over a year to less than 9 weeks (Schimmelmann et al., 2008). This was a significant advance, as DUP has been shown to independently predict clinical and functional outcome (Marshall et  al., 2005; Perkins, Gu, Boteva, & Lieberman, 2005). In addition, the early intervention provided at EPPIC has been found to be an effective intervention from an economic point of view. Consistent with findings elsewhere, the intervention provided for first episode psychosis at EPPIC has been shown to provide a better value for money intervention than standard care (Economics, 2008; Mihalopoulos, McGorry, & Carter, 1999; Mihalopoulos, Vos, Pirkis, Smit, & Carter, 2011). A report by an independent economics firm in Australia found that the early intervention approach was cheaper than “treatment as usual” both during the treatment phase and in terms of incremental cost effectiveness (Economics, 2008). The report found that if early intervention was routinely available, the potential savings to the health system in Australia would be AUD$210,000,000 per year. This saving does not include the saving that would also accrue if some of the more recent aspects of the early intervention model, such as vocational intervention, were also included (Killackey, Jackson, & McGorry, 2008).

2.2

Identification and Intervention in the Pre-­ onset Phase

Initially, the early psychosis movement focused on timely recognition and phase-specific treatment of first episode psychosis. However, it was also recognized that for most patients a prolonged

4  Attenuated Psychosis Syndromes Among Australian Youth and Young Adults: Early Identification…

period of non-specific psychiatric symptoms, attenuated psychotic symptoms, and impaired functioning precedes the first psychotic episode (Häfner, Maurer, Loffler, & Riecher-Rossler, 1993; Yung et al., 1996; Yung & McGorry, 1996). Much of the disability associated with psychotic disorders, particularly schizophrenia, develops long before the onset of frank psychosis and is difficult to reverse even if the first psychotic episode is successfully treated (Häfner et al., 2003). This pre-onset period of illness has been termed the prodromal phase (Huber, Gross, & Schuttler, 1979; Yung, 2003). Within the context of the early intervention paradigm, EPPIC researchers suspected that pushing the point of intervention even further back from the first episode of psychosis to the prodromal phase may result in even better outcomes (McGorry, Phillips, & Yung, 2001; McGorry & Singh, 1995; McGorry, Yung, & Phillips, 2001; Yung et al., 1998). The rationale here was that intervening during this phase may ameliorate, delay, or even prevent onset of a fully fledged disorder (Yung, 2003), thereby reducing the burden of disability, prevalence, and possibly even the incidence of psychotic disorders. One of the challenges in this field is that the non-specific nature of the most common prodromal features means that prodromal intervention may be provided for a substantial proportion of “false-positive” cases, i.e., people who were falsely identified as being at risk of progressing to a psychotic disorder. This does not necessarily mean that intervention is not required, however, as there is typically a genuine need for care in terms of distress and impairment (Fusar-Poli et  al., 2015). Research indicates that approximately two-thirds of cases do not go on to develop psychosis (Fusar-Poli et  al., 2013). Indeed, the term “prodrome” should strictly only be used once the full-blown syndrome has developed (Yung et al., 1996). The original thinking by our group was that prior to diagnosis with a psychotic disorder, the prodrome should be thought of as a risk state for psychosis, not as a specific disease entity (i.e., the presence of the syndrome implies that the affected person is at that time more likely to develop psychosis in the near future than someone without the syndrome). However, if the symptoms resolve, then this degree of increased

69

risk may remit as well. The terms and criteria that we introduced in the mid-1990s – the “ultrahigh-­ risk” (UHR) criteria and “at risk mental state” (ARMS) (McGorry & Singh, 1995; Yung et al., 2003; Yung, Phillips, Yuen, & McGorry, 2004) – reflected these concepts. The term “ultra” (and more recently “clinical”) was used in order to differentiate this approach from the traditional “high-risk” approach based purely on familial risk. The UHR criteria were an attempt to identify people with a high likelihood of developing a psychotic disorder within the near future (e.g., within 12  months). However, in more recent years, particularly driven by the issue of whether to include attenuated psychotic syndrome (APS) in DSM-5, arguments have been made that the attenuated psychotic symptoms captured in the ARMS concept should be thought of as a disease entity in their own right (Carpenter, Regier, & Tandon, 2014; Fusar-Poli et al., 2015; Fusar-Poli, Carpenter, Woods, & McGlashan, 2014). Rather than a “disease entity,” a more flexible perspective is that this is a diverse phenotype with diffuse comorbidity which justifies a “need for care” and as such qualifies for being seen as a “disorder.” It is an earlier stage of evolution than other traditional disorders and has a valence of risk for subsequent psychosis but also for persistent mood and anxiety disorders. In the formulation of the UHR criteria, it was recognized that the non-specific nature of prodromal symptoms makes it problematic to use these features alone to identify people at imminent risk of psychotic disorder. Psychotic-like experiences have been found to occur commonly in the general population, especially among adolescents and young adults (Johns et al., 2004; Tien, 1991; van Os, Hanssen, Bijl, & Vollebergh, 2001; Verdoux & van Os, 2002). Therefore, symptoms alone would result in a high false-positive rate and poor sensitivity. A “close-in” strategy (McGorry, Yung, & Phillips, 2003) was used to maximize the possibility of identifying people who may truly be in the prodromal phase of a psychotic disorder. This strategy included the risk factor of age, as the age of highest incidence of psychotic disorder is adolescence and young adulthood (Häfner et al., 1993). Clinical need for care was another factor. The criteria required that

B. Nelson and P. D. McGorry

70

a young person must be seeking help or be identified by someone, such as a parent or teacher, as needing help. Although the UHR criteria have undergone minor changes over the years (see Table 4.1), they are based around three groups: Attenuated Psychotic Symptoms (APS) group: patients who have experienced subthreshold positive psychotic symptoms (e.g., overvalued ideas, perceptual disturbances) during the past year. Brief Limited Intermittent Psychotic Symptoms (BLIPS) group: patients who have experienced an episode of frank psychotic symptoms that have lasted less than a week and resolved without treatment. Trait and State Risk Factor Group: patients with a schizotypal personality disorder or have a first-degree relative with a psychotic disorder and have experienced chronic poor functioning or a significant decrease in functioning during the previous year. Our research group hypothesized that individuals meeting these criteria would have a high likelihood of developing a psychotic disorder in the near future. This has proven to be the case, with meta-analytic data indicating that 22%

progress to psychotic disorder over a 1-year period and 36% over a 3-year period, which is several thousand-fold greater than the expected incidence rate for first episode psychosis in the general population. The UHR criteria have been adopted by a number of other centers around the world, sometimes with minor modifications (Olsen & Rosenbaum, 2006). Work in this area led to the proposal to include the diagnosis “Attenuated Psychosis Syndrome” in DSM-5. This proposal was vigorously debated in the development of DSM-5 with it ultimately being decided to include the diagnosis in the section “Conditions Requiring Further Research” based primarily on concerns regarding reliability of assessment (Carpenter, 2014; Nelson, 2014; Woods, Walsh, Saksa, & McGlashan, 2010).

2.3

 he Personal Assessment T and Crisis Evaluation (PACE) Clinic

Our research group established a specialized service for the UHR group, the PACE Clinic, in Melbourne in 1994. This service was the first clinical and research clinic internationally for individuals considered to be at incipient risk of

Table 4.1  Personal Assessment and Crisis Evaluation (PACE) criteria for assessing ultrahigh risk for psychosis status (1994–2016) Criteria Symptom/trait requirement

Drop in functioning/ sustained low functioning requirement

Age range

Attenuated Psychotic Symptoms (APS) The presence of attenuated (subthreshold) positive psychotic symptoms within past 12 months

Brief Limited Intermittent Psychotic Symptoms (BLIPS) Presence of frank psychotic symptoms for 80%, regardless of sample size or psychotic symptoms that began in the past method of handling syndromal comorbidity. 3 months but are insufficiently frequent or disor- Across the 22 samples reporting rates of any ganizing/dangerous to meet criteria for frank APSS, the median rate was 95.3% (range 83.7– psychosis. For GRD patients must have a per- 100%, interquartile range 89.6–99.6%). The high sonal history of schizotypal personality or a first-­ rates of APSS in CHR samples led the DSM-5 to degree family history of psychosis along with a define its CHR criteria based solely on the most 30% drop in global functioning in the past year. common subsyndrome (American Psychiatric Rates of endorsement of the component syn- Association, 2013). dromes with the SIPS were included in a recent In the 19 studies that fully accounted for meta-analysis (Fusar-Poli et  al., 2016) and are comorbidity within CHR syndromes, rates of only briefly updated here, as the findings are BIPS without APSS were very low (median similar. The previous review included 33 sam- 0.0%, range 0.0–8.9%, interquartile range 0.0– ples, 16 of which used the SIPS, and reported 1.5%). BIPS rates were somewhat higher in 12 that 85% of CHR cases met criteria for the studies not permitting comorbidity (median APSS subgroup, even though cases meeting 4.6%, range 0.0–14.9%, interquartile range 2.9– BIPS along with APSS were excluded from the 6.4%), the difference presumably reflecting APSS accounting. comorbidity with the more common APSS. These The current review located 34 samples that low rates of BIPS are also similar to those used the SIPS criteria, omitting 5 from the previ- reported in the 2016 Fusar-Poli and colleagues’

5  Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk…

meta-analysis. The low rates of BIPS diagnoses with the SIPS contrast somewhat with higher rates observed with the corresponding CAARMS subsyndrome (Fusar-Poli et  al., 2016), ­presumably due to criterion variance (Miller et  al., 2003; Schultze-Lutter, Schimmelmann, Ruhrmann, & Michel, 2013). With regard to GRD, rates overall were also low but somewhat higher than for BIPS. In the studies that accounted for comorbidity, the median rate of GRD without APSS was 3.9% (range 0.0–10.8%, interquartile range 1.0–6.8%). Rates were similar in studies that did not permit comorbidity (median 1.8%, range 0.0–9.8%, interquartile range 0.0–6.4%). Overall the distribution of CHR component syndromes seemed quite consistent across studies. The quartile coefficient of deviation (QCD, the interquartile range divided by the median) for any APSS was only 9%. This consistency contributes to evidence that the SIPS is being used reliably across research centers. The distribution did not suggest a cultural pattern in the distribution of CHR syndromes across samples. Among the 19 studies that fully accounted for syndromal comorbidity, the 5 samples with highest rates of any APSS come from North America (1 of 6 samples), Asia (3 of 6 samples), and Europe (1 of 7 samples). The five samples with lowest rates of any APSS come from North America (one of six samples), Asia (one of six samples), and Europe (three of seven samples).

2.2

 ate of Endorsement of SOPS R Symptoms in CHR Subjects

The 19 items on the SOPS are divided into 4 groups: positive symptoms, negative symptoms, disorganization symptoms, and general symptoms, each scored 0–6. The five positive items are P1 unusual thought content, P2 suspiciousness, P3 grandiosity, P4 perceptual abnormalities, and P5 disorganized communication. The six negative items are N1 social anhedonia, N2 avolition, N3 expression of emotion, N4 experience of emotions and self, N5 ideational richness, and N6 occupational functioning. The four disorgani-

89

zation symptoms are D1 odd behavior or appearance, D2 bizarre thinking, D3 focus and attention, and D4 personal hygiene. The four general symptoms are G1 sleep disturbance, G2 dysphoric mood, G3 motor disturbance, and G4 intolerance to normal stress. Studies eligible for this section required a report of a SIPS-defined CHR sample and a report of specific item scores, yielding 40 samples (Table  5.1). Median sample size was 48 (range 5–744, interquartile range 32–87). Median sample mean age was 19.6 (range 14.6– 34.7, interquartile range 18.0–20.6). Outliers were defined as a point which falls more than 1.5 times the interquartile range above the third quartile or below the first quartile (Tukey, 1977). There were three outlier samples with regard to age, all at the high end, including both of the two samples collected in Germany and one from China. The SOPS total score was reported in 35 samples, and the SOPS positive symptom subscale score in 38. Samples were collected in 13 countries: from North America (3 countries), Europe (7 countries), and Asia (3 countries). The correlation across samples between age and SOPS score was not statistically significant either for the total score (r = 0.052, p = 0.768) or for the positive symptom subscale score (r = 0.087, p = 0.601). Baseline SOPS total and positive symptom scores in SIPS-defined CHR patients both appeared reasonably consistent across sample. For the total score, the median across the 35 reporting samples was 35.9 (range 19.4–58.6, interquartile range 32.8–41.5). SIPS items are scores on a 0–6 point scale with higher scores indicating greater severity. The SIPS positive symptom scale has five items (total range 0–30), the negative symptom scale has six items (total range 0–36), the disorganization symptom scale has four items (total range 0–24), and the general symptom scale has four items (total range 0–24). There are a total of 19 items on the SOPS, resulting in total scores that range from 0 to 114. The distribution of SOPS total scores was somewhat broader than that of the proportion of APSS diagnoses (see Sect. 2.1): the QCD was 24%. Two studies were identified as outliers, one

S. W. Woods et al.

90 Table 5.1  SOPS scores in SIPS-defined CHR samples Citation Morita et al. (2014) Brucato et al. (2017) Lemos-Giraldez et al. (2009) Ribolsi et al. (2017)a Lindgren et al. (2010) Fresán et al. (2015) Bakker et al. (2016) Carrión et al. (2018) Melton (2012) Thermenos et al. (2016) Lunsford-Avery, Dean, and Mittal (2017) Bernard, Orr, Dean, and Mittal (2018) Miyake et al. (2016) Comparelli et al. (2016) Comparelli et al. (2014) Woods et al. (2009) Addington et al. (2015) Chan et al. (2018) Schlosser et al. (2015) Loewy et al. (2016) Zhang et al. (2018) Wang et al. (2016) Da Silva et al. (2018) Bang et al. (2017) Gerstenberg et al. (2015) Lee et al. (2014) Lo Cascio et al. (2016) Aase et al. (2018) Simon et al. (2012) Kraan et al. (2015) Solé-Padullés et al. (2017) He et al. (2018) Addington et al. (2012) Wilson et al. (2016) Rausch et al. (2016) Zhang et al. (2017) Chen et al. (2016) Shi et al. (2017) Koike et al. (2017) Lincoln et al. (2018) Fulford et al. (2013) Strauss, Ruiz, Visser, Crespo, and Dickinson (2018) Epstein et al. (2014)

Country Japan USA Spain Italy Finland Mexico Netherlands USA USA USA USA

N 46 200 61 94 62 50 23 205 8 37b 63 27 Japan 5 Italy 45 32 USA, Canada 377 USA, Canada 744 China 39 USA 85 83 China 391 China 34 Canada 30 South Korea 74 Switzerland 21c South Korea 75 USA/Italy 22 Norway 46 Switzerland 73 Netherlands 125e Spain 44 China 19 USA, Canada 172 USA 64 Germany 132 China 117 China 63 China 32 Japan 47 Germany 25 USA 33 USA 23 USA 21

Age 23.5 ± 6.6 20.0 ± 3.8 21.7 ± 3.8 14.6 ± 2.0 15.0–18.8 19.6 ± 3.3 24.3 ± 3.1 16.5 ± 3.3 19.6 ± 3.3 19.6 ± 4.0 18.9 ± 1.7

SOPS total 58.6 ± 15.7 53.3 52.2 45.9 43.9 43.0 42.9 42.8 41.7 41.3 nr 40.4 19.8 ± 5.7 39.6 ± 13.6 21.0 ± 0.4 nr 39.3 18.2 38.4 18.4 ± 4.2 38.2 18.0 ± 5.0 37.0 18.7 ± 4.5 nr 36.9 ± 12.1 20.4 ± 6.1 35.9 ± 11.1 21.0 ± 5.7 35.6 ± 11.0 20.3 ± 1.7 35.2 ± 11.0 19.9 ± 3.5 35.2 15.0 ± 1.4 35.0d 20.0 ± 3.8 34.6 14.7 ± 1.4 34.2d 17.9 ± 4.9 33.2 20.4 ± 5.2 32.9 17.7 ± 3.9 32.7 15.2 ± 1.6 31.3 20.5 ± 4.6 31.0 19.8 ± 4.5 30.8 16.7 ± 3.0 30.3 24.8 ± 5.7 29.9 ± 12.5 24.7 ± 7.6 28.0 ± 10.2 21.9 ± 4.5 23.7 18.8 19.4 21.0 ± 3.8 nr 34.7 ± 14.2 nr 18.5 ± 4.4 nr 19.6 ± 1.8 nr 16.1 ± 3.3 nr

nr not reported, sta smaller sample at same site than above a Subjects met DSM-5 criteria for Attenuated Psychosis Syndrome based on SIPS interview b Four subjects met modified SIPS criteria for CHR c Met DSM-5 criteria for Attenuated Psychosis Syndrome based on SIPS d Median e Two subjects did not meet SIPS criteria for CHR

SOPS positive 18.9 ± 4.8 14.4 ± 4.1 11.6 11.9 ± 4.6 11.2 11.6 ± 4.5 9.9 ± 3.7 12.1 ± 4.2 18.0 ± 3.5 12.2 11.7 ± 5.0 sta nr 7.9 sta 11.9 11.9 9.6 ± 3.9 10.4 ± 4.2 sta 9.5 ± 3.6 9.6 ± 3.1 11.0 ± 3.4 10.6 ± 4.1 9.0d 9.7 ± 3.7 7.5d 10.0 ± 3.8 7.4 ± 3.7 9.9 ± 3.9 7.8 ± 3.8 11.5 ± 6.8 11.0 ± 3.2 8.4 ± 5.9 nr 7.1 ± 4.5 8.2 7.8 ± 3.5 9.7 ± 4.4 9.5 ± 4.4 9.1 ± 4.2 9.0 ± 3.8 6.8

5  Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk…

each for high and low SOPS total scores. The high SOPS total score outlier study enrolled helpseeking outpatients at a university hospital in Japan with a mean age of 23.5 (Morita et al., 2014). The low SOPS total score outlier sample (Shi et al., 2017) identified CHR subjects by universal screening of a Chinese university population (Wang et al., 2015) whose mean age was 18.8. Of the five CHR samples with the highest SOPS total scores, one was reported from North America, three from Europe, and one from Asia. Two of these studies screened clinical populations, one of inpatients (Raballo et  al., 2018; Ribolsi et  al., 2017) and one of outpatients (Lindgren et  al., 2010, 2014). Of the five CHR samples with the lowest SOPS total score, one was reported from North America, one from Europe, and three from Asia. Interestingly, the sample with the second lowest SOPS total score (Chen et al., 2016) was identified earlier by the same group also using universal screening of the college population (Shi et  al., 2017). The low SOPS scores among CHR participants identified by general population screening are interesting in view of recent meta-analytic evidence that intensive outreach campaigns targeting the general public and resulting in high proportions of self-­ referrals may also yield samples with lower risk for psychosis despite meeting criteria (Fusar-Poli et  al., 2016). Quantitative meta-analysis would be useful on this point. For the SOPS positive symptom subscale score, the median across the 38 reporting samples was 9.9 (range 6.8–18.9, interquartile range 9.0– 11.6). Distribution was similar to that of the total score (see above): the QCD was 26%. Two studies were identified as outliers, both for the high side. The high SOPS positive symptom subscale outlier study was also the total score outlier that identified help-seeking outpatients at a university hospital in Japan with a mean age of 23.5 (Morita et al., 2014). The other SOPS high positive symptom subscale outlier reported on a small sample for a thesis project from Oregon in the USA (Melton, 2012). The two studies with high SOPS total scores where participants were recruited by screening upon initiation of clinical services

91

were not among the samples with the highest SOPS positive symptom subscale scores. The ranking of samples on SOPS positive symptom subscale scores overlapped somewhat with the ranking for the total score as expected, but not completely. Of the five CHR samples with the highest SOPS positive symptom subscale scores, four were reported from North America and one from Asia. The two samples with high SOPS total scores that recruited participants by screening on those already receiving clinical services were not among the five highest-­ scoring CHR samples for positive symptoms. Of the six CHR samples with the lowest SOPS total scores (two were tied for fifth lowest), one was reported from North America, two from Europe, two from Asia, and one jointly from the USA and Italy. Of the samples with low SOPS total scores that identified participants using universal screening of the college population, one was among the lowest six on positive symptoms (Shi et  al., 2017). None of the other low-scoring samples on positive symptoms identified participants using universal screening. Among the five SOPS positive symptoms, three tend to be endorsed more frequently than the other two. The three commonly endorsed items are P1 (unusual thought content), P2 (suspiciousness), and P4 (perceptual abnormalities). P3 (grandiosity) and P5 (disorganized communication) are less frequently endorsed, especially P3. This pattern was quite consistent across studies (Addington et  al., 2015; DeVylder et  al., 2014; Jung et  al., 2010; Lindgren et  al., 2010; Woods et al., 2009; Zhang et al., 2014), with one exception. In a small study (N = 30), the mean P5 score was slightly higher than the mean P4 score (Lemos et al., 2006).

2.3

 ate of Endorsement of SOPS R Symptoms in Healthy Subjects

Studies eligible for this section required a report of a SOPS total or positive symptom subscale scores in healthy volunteer samples, yielding 21 samples (Andersen et  al., 2016; Antshel et  al.,

S. W. Woods et al.

92

2010; Bonner-Jackson, Csernansky, & Barch, 2007; Calkins et al., 2017; Carrión et al., 2018; Dean, Orr, Newberry, & Mittal, 2016; Dodell-­ Feder, Delisi, & Hooker, 2013; Gur et al., 2015; Jalbrzikowski et  al., 2014; Larsen et  al., 2017; Lincoln & Hooker 2014; Roman-Urrestarazu et  al., 2014; Sugranyes et  al., 2015; Thermenos et  al., 2016; Velthorst et  al., 2018; Weinberger et  al., 2018; Woods et  al., 2009; Zhang et  al., 2016, 2018), including 2 that were each split into ­separate reports by symptom type (Kayser et al., 2014; Korver et  al., 2010; Poe et  al., 2017; Velthorst et  al., 2013). The median sample size was 47 (range 18–212, interquartile range 32–83). The SOPS total score was reported in 18 samples, and the SOPS positive symptom subscale score in all 21. Samples were collected in eight countries: from North America (the USA and Canada), Europe (four countries), China, and Israel. For the total score, the median in control groups across the 18 reporting samples was 2.3 (range 0.8–5.2, interquartile range 1.7–3.2). No studies were identified as outliers. For the SOPS positive symptom subscale score, the median across the 21 reporting samples was 0.8 (range 0.1–1.6, interquartile range 0.4–1.0). Again, no studies were identified as outliers. Scores were dramatically and consistently lower than in CHR samples; for example, the median CHR sample mean total score (35.9) was >15-fold higher than the median healthy subject sample mean of 2.3. The lowest CHR sample mean total score (19.4) was nearly fourfold higher than the highest healthy subject sample mean of 5.2.

3

Reliability

The reliability of an instrument refers to the degree to which it is susceptible to error of measurement. Inter-rater reliability concerns the level of agreement between two or more raters evaluating the same patient. Two kinds of inter-rater reliability studies were available: those focusing on diagnostic agreement using the SIPS and those focusing on severity of illness agreement using the SOPS. Several authors have provided guidelines for interpreting numeric reliability values. For exam-

ple, kappa (Cohen, 1960) values >0.75 have been described as indicating “excellent” reliability (Cicchetti & Sparrow 1981; Fleiss, 1981), and percent concordance >90% has also been considered “excellent” (Cicchetti, 2001). In practice, many clinical units or CHR research studies, while continuing to stress rater training and monitoring for rater drift, safeguard against the impact of any residual unreliability by also holding clinical team diagnostic consensus meetings or conferences. Typically each case and its ratings are presented in a team meeting or, for research, on a conference call attended by one or more raters from each site. Cases are discussed until the attendees reach consensus.

3.1

I nter-rater Reliability of SIPS Diagnosis

We identified 21 studies of the inter-rater relia­ bility of a CHR diagnosis using the SIPS (Table  5.2). In all the available studies of the inter-rater reliability of the SIPS diagnostic interview, two or more raters viewed the same patient interview, or videotaped interview, and made ratings independently. Agreement for diagnosis (categorical measure) was expressed by use of the kappa statistic or as percent concordance. Studies were reported from North America (11 samples), Europe (7 samples), and Asia (3 samples). Sixteen of these studies defined CHR according to the SIPS criteria, and five used modified criteria. Some studies reported agreement as an exact value and others as greater than a specified value. Since the SIPS generates several diagnoses (psychosis, APSS, BIPS, GRD, any CHR, neither psychosis nor CHR), different studies expressed binary categorical agreement on diagnosis in different ways that were not always fully explicit. The studies reporting only on reliability of a SIPS diagnosis (CHR vs non-CHR) generally did not detail how often the non-CHR patients rated were non-CHR because they were psychotic vs non-­CHR because they were neither CHR nor psychotic. Most of the papers did not emphasize reliability but instead mentioned the reliability findings briefly in support of other aims. As a

5  Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk…

93

Table 5.2  Inter-rater reliability of SIPS diagnosis Study Site(s) Subjects (a) Not specifically CHR vs psychosis – kappa Kimhy et al. (2007) USA nr Kline et al. (2015) USA 10 Wang et al. (2015) China 3 Loewy et al. (2012) Finland 8 Loewy et al. (2016) USA nr Comparelli et al. (2011) Italy 15 Meyer et al. (2005) USA 12

Raters 1 nr nr nr nr 2 nr

Judgment

nr SIPS diagnosis SIPS diagnoses SIPS diagnosis SIPS diagnosis nr APSS vs not BIPS vs not GRD vs not Addington et al. (2007) USA, Canada 2 40 CHR vs not Zhang et al. (2014) China nr 4 nr Miller et al. (2003) USA, Canada 2 17 CHR vs not Fusar-Poli et al. (2016) UK 21 6 SIPS diagnosis Italy nr 5 CHR syndromes Kotzalidis et al. (2017)c Miller et al. (2002) USA 18 2–4 per S SIPS diagnosis Carol & Mittal (2015) USA nr nr nr USA nr nr SIPS diagnosis Walker et al. (2010)c Keshavan et al. (2009) USA 5 nr nr (b) Not specifically CHR vs psychosis – pairwise concordance Netherlands 2 nr CHR vs not Velthorst et al. (2013)c Europe 2 nr CHR vs not Ruhrmann et al. (2010)c Nieman et al. (2013) Netherlands nr nr nr (c) Not specifically CHR vs psychosis – statistic not reported Kobayashi et al. (2008) Japan nr nr nr d. Specifically CHR vs psychosis nr 37 Psychosis vs McFarlane et al. (2015)c USA CHR

Standard

Statistic Value

vs gold standard Across raters nr Across raters vs gold standard Across raters vs gold standard

Kappa Kappa Kappa Kappa Kappa Kappa Kappa

vs gold standard Rater pairs vs gold standard Across raters Across raters Across raters nr nr Across raters

Kappa Kappaa Kappab Kappa Kappa Kappa Kappa Kappa Kappa

vs gold standard PC vs gold standard PC Across raters PC nr

nr

vs gold standard Kappa PC

1.00 1.00 0.97 0.97 0.95 0.91 >0.90 >0.90 >0.90 0.90 0.88 0.88 0.85 0.82 0.81 >0.80 >0.80 >0.7 0.89 0.77 0.77 exc 0.68 0.93

PC pairwise concordance, nr not reported, exc reported qualitatively as excellent Mean of reported pairwise range b Mean across three sites not reported in Addington et al. (2007), weighted by number of raters c Modified SIPS criteria a

result, the number of raters and number of ­subjects were not always stated. Among the 16 studies reporting agreement as kappa and not specifically focused on CHR vs frank psychosis diagnostic agreement, median kappa was 0.89 (range >0.7–1.00, interquartile range 0.82–0.96). Thus, with the possible exception of the study reporting kappa as >0.7, ­reliability was excellent in all studies. There was no discernible pattern across cultures, with the highest and lowest values being reported from the USA. No studies were identified as outliers. One additional study reported reliability only qualitatively, as “excellent” (Kobayashi et  al., 2008).

Agreement in the three studies reporting percent concordance was somewhat lower, however, ranging from 0.77 to 0.89, all below the excellent range for this statistic. Thus overall, a fair summary is that diagnostic inter-rater reliability with the SIPS was “excellent” in the large majority of studies (14 or 15 of 18). Only three studies specifically address the issue of CHR vs frank psychosis diagnostic agreement. In the original study by the SIPS developers (Miller et al., 2002), 2 of the 11 non-­CHR cases were rated as psychotic by any rater. The raw data from that study show perfect agreement on the psychosis judgment: both of two raters agreed on one case and three raters all agreed in the

S. W. Woods et al.

94

other. Another study focused specifically on the reliability of the SIPS CHR vs psychosis judgment, in 37 raters compared to criterion ratings by an experienced psychiatric researcher (McFarlane et  al., 2015). The percent concordance was in the excellent range at 93%; however, reliability as expressed by kappa was somewhat below the excellent range at 0.68. A disparity between the percent concordance and kappa values can occur when the subject pool contains a substantial preponderance of one ­diagnosis or the other. Lastly, one study reported categorical agreement with the gold standard on whether SOPS positive symptoms were in the CHR (scale values 3–5) vs psychotic (scale value 6) range (Addington et al., 2011; Marshall et al., 2012). Kappa was excellent at 0.90.

3.2

I nter-rater Reliability of SOPS Scores

We identified 23 studies of the inter-rater reliability of the SOPS total score on one or more subscale subtotals (Table 5.3). Agreement was expressed by use of the intraclass correlation (ICC), kappa, and Pearson’s r or Spearman’s rho. These values are referred to as “reliability coefficients.” Studies were reported from North America (13 samples), Europe (4 samples), Asia (2 samples), Africa (2 samples), and Israel (1 sample). Reliability coefficients overall were very high. Among the 23 studies, 9 reported on the reliability of the SOPS total score. The median reliability coefficient was 0.90 (range >0.75–0.96, interquartile range 0.84–0.95). Reliability was similarly high for the positive symptom subscale (21 samples, median 0.88, range >0.75–0.99, interquartile range 0.81–0.92), the negative symptom subscale (12 samples, median 0.86, range >0.75–0.98, interquartile range 0.80–0.93), the disorganization symptom subscale (11 samples, median 0.80, range 0.71–0.95, interquartile range 0.75–0.95), and the general symptom subscale (9 samples, median 0.88, range >0.75–0.95, interquartile range 0.80–0.92). No outlier samples were identified for the any subscale or the total score.

We did not discern a pattern across cultures, represented here by location of the population under study. Of the two samples from Africa, one showed the second highest reliability for positive symptoms and the other the second lowest reliability for the total score. The two samples from Asia were more or less in the middle across subscale, as were two of the three samples from Europe. The sample from Denmark showed the third lowest reliability for positive symptoms at ICC = 0.77.

3.3

Inter-rater Reliability in Clinical Practice

Much less work has been done in reliability of the SIPS or DSM-5 APS diagnosis in clinical practice, using unstructured interviews conducted by clinicians with a prescribed amount of training. The DSM-5 Field Trials on APS (Clarke et  al., 2013; Narrow et  al., 2013; Regier et  al., 2013) were inconclusive due to the small number of subjects evaluated. The kappa achieved for diagnostic agreement was nevertheless in the middle of the range of the coded diagnoses assessed (Fusar-Poli, Carpenter, Woods, & McGlashan, 2014). An additional study, reported only in abstract form (Woods, Walsh, & McGlashan, 2012), provided 30 min of training in DSM-5 APS diagnosis to 11 community clinicians and evaluated 34 patients. Clinicians were asked to rate whether patients were psychotic, APS progressive, APS persistent, APS remitted, or never psychotic or APS. Reliability was somewhat lower than in the DSM-5 Field Trial, suggesting the need for longer training than 30 min, especially if the more complex course specifier judgment is required. This lack of variation or pattern in CHR syndromes, SOPS symptoms, and SIPS inter-rater reliability by country suggests that the reliability of the SIPS is not biased by country of origin and support its use for identifying CHR syndromes in various cultural contexts around the world, provided that clinicians receive adequate training.

USA USA

McFarlane et al. (2015)

11 21

USA UK S. Korea Finland USA USA

Lencz, Smith, Auther, Correll, and Cornblatt (2004) Fusar-Poli et al. (2016)

Jung et al. (2010)

Veijola et al. (2013)

Kline et al. (2015)

Melton and Dykeman (2016)

10

10

14–15

20

nr

Canada

Marshall et al. (2012)

15

Italy

Comparelli et al. (2011)

Tso et al. (2017)

nr

nr

nr

Jalbrzikowski et al. (2012)

Zhang et al. (2014)

nr

USA, Canada China

Addington et al. (2015)

8

6

nr

Kenya

Owoso et al. (2014)

Standard Between site 12 Across raters nr Across raters 2 Across raters nr Gold standard 4 Across raters nr Gold standard 37 Gold standard 35 pairs Across raters 2 Across raters nr Gold standard 2 Across raters 6 Across raters 2 Across raters 5 Across raters nr Across raters nr Across raters

Subjects Raters nr nr

Loewy, Therman, Manninen, Huttunen, and USA Cannon (2012) Karcher, Martin, and Kerns (2015) USA

Site USA/Israel

Study Weisman et al. (2017)

Table 5.3  Inter-rater reliability of SOPS severity of illness

ICC

ICC

ICC

ICC

Kappa

ICC

Kappa

rho

nr

ICC

Kappa

r

ICC

nr

ICC

ICC

nr

0.84

nr

0.96

nr

nr

nr

>0.90

nr

nr

nr

0.96

0.87e

nr

nr

nr

SOPS Statistic total ICC nr

0.81

0.82

0.83h

0.84

0.86g

0.88b

0.90

>0.90

sta

0.91

0.92f

0.92b

0.94e

0.94d

0.95c

0.98b

SOPS positive 0.99a

0.81

nr

nr

0.97

nr

0.83b

nr

>0.90

0.92

nr

0.92f

nr

nr

nr

0.95c

nr

SOPS negative 0.98a

nr

nr

nr

0.86

nr

0.75b

nr

>0.90

0.71

nr

0.92f

nr

nr

nr

0.95c

nr

SOPS disorganized 0.95a

(continued)

nr

nr

nr

0.92

nr

0.85b

nr

>0.90

nr

nr

0.92f

nr

nr

nr

0.95c

nr

SOPS general nr

5  Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk… 95

USA Denmark USA USA USA Nigeria

Kimhy et al. (2007)

Mondrup & Rosenbaum (2010)

Miller et al. (2003)

Meyer et al. (2005)

Antshel et al. (2010)

Okewole et al. (2015)

nr

5

12

14

12

nr

nr

nr

2–3 per S nr

2

1

Subjects Raters nr nr

Standard Gold standard Gold standard Across raters Across raters Across raters Across raters Across raters ICC

ICC

ICC

ICC

ICC

ICC

0.78l

0.90k

>0.75j

0.95

nr

nr

SOPS Statistic total r nr

nr

nr

>0.75j

>0.75

0.77g

0.80i

SOPS positive >0.80

nr

nr

>0.75j

>0.75

nr

0.80i

SOPS negative >0.80

nr

nr

>0.75j

>0.75

nr

0.80i

SOPS disorganized >0.80

nr

nr

>0.75j

>0.75

nr

0.80i

SOPS general >0.80

a

r Pearson correlation coefficient, rho Spearman correlation coefficient, ICC intraclass correlation coefficient, nr not reported, sta smaller sample at same site than above Mean of reported individual items, excluding ideational richness in this 22q11 deletion study b Midpoint of individual item ranges c Reported as 0.92–0.99 across subscale d Reported as 0.93 for P1 and 0.95 for P4 e Midpoint of reported ranges across four annual evaluations f Midpoint of reported range 0.85–1.0, which appears to range across raters and subscales g Mean of 5 item ICCs h Reported as ICC for highest lifetime P1–P5 score i Reported as ~0.7 to 0.9 across subscale j For each individual symptom k Appears to be reliability of total of positive, negative, and disorganization subscales l Reported as 0.79–0.93 across P1–P4 and 0.48 for P5

Site USA

Study Walker et al. (2010)

Table 5.3  (continued)

96 S. W. Woods et al.

5  Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk…

4

Validity

97

subjects and just 1.56% (95% CI 0.7–2.42%) in clinical subjects referred for CHR evaluation who We reviewed two types of validity of the SIPS did not meet CHR criteria, yielding a relative risk diagnosis, predictive and construct. In the CHR of 16.7. This indicates that among subjects field, the future event of most interest for pre- referred for CHR evaluation, those who meet dictive validity thus far has been the subsequent CHR criteria were 16.7 times more likely than occurrence of frank psychosis. Other future those who did not to develop frank psychosis, outcomes of interest have included subsequent substantial evidence of predictive validity. No remission and subsequent functional status. In effect of instrument was observed. Sensitivity for general, the most informative predictive valid- predicting psychosis in the five SIPS samples was ity analysis compares subjects who meet CHR 0.95 (95% CI 0.88–0.99), and specificity was 0.39 criteria to those from the same referral popula- (95% CI 0.32–0.46). The rate of conversion to psychosis in the CHR group was significantly tion who did not. Construct validity is usually held to describe higher than in the non-CHR group in each of the the degree to which a diagnosis or rating scale five studies using the SIPS. In addition, the current review revealed four score measures the same construct as some other diagnosis or scale whose validity is considered other studies (Kline et al., 2015; Kobayashi et al., established and is measured at the same time. 2008; Lindgren et al., 2014; Manninen et al., 2014) Two types of construct validity are distinguished: from the USA, Japan, and Finland employing the one where similarity is expected on the measure, SIPS and a similar design, for a total of nine. Of termed convergent validity, and one where differ- the four additional studies, two also found that the ences are expected, termed discriminant validity. rate of conversion to psychosis in the CHR group Predictive and construct validity of the SOPS was significantly higher than in the non-CHR ratings was beyond the scope of the present review. group, bringing the total of studies making this report to seven of nine. Of the two that did not, one had only seven CHR participants and restricted 4.1 Predictive Validity of SIPS enrollment to adolescents who resided in a reform Diagnosis school (Manninen et al., 2014), while the second (Lindgren et al., 2014) also restricted enrollment The predictive validity of a SIPS diagnosis for to adolescents and had a particularly low converdevelopment of subsequent frank psychosis has sion rate (2/51  =  3.9% over 1  year) in the CHR been evaluated in a recent meta-analysis (Fusar-­ group. In both studies the rate of conversion was Poli et  al., 2015). That meta-analysis located a numerically higher in the CHRs. total of 11 longitudinal studies from Germany, Overall the literature reviewed continues to Australia, Switzerland, North America, Taiwan, provide strong support for the predictive validity Singapore, Poland, and Italy comparing psychosis of the CHR diagnosis using the SIPS, regardless outcomes in CHR subjects with those from the of the country in which this work was conducted. optimal ecological control group, subjects who Interestingly, the two studies that did not find prewere referred for CHR evaluation but did not dictive validity statistically significant were meet CHR criteria. Of these, two used the SIPS among the three studies with the youngest ages at definition of CHR (Addington et al., 2012; Woods baseline. et al., 2009), three the SIPS but with a modified CHR definition (Liu et al., 2011; Schultze-Lutter et  al., 2014; Simon et  al., 2012), four the 4.2 Construct Validity of SIPS Diagnosis CAARMS, and two other instruments to make the initial CHR diagnosis. The overall findings for the meta-analysis showed that the 38-month risk of Two types of comparisons are included: those psychosis was 26% (95% CI 23–30%) in CHR between SIPS-defined CHR cases and patients

S. W. Woods et al.

98

insufficiently ill to meet SIPS criteria and those between SIPS-defined CHR cases and patients who did not meet SIPS criteria due to frank psychosis. Most CHR research clinics screen out cases of chronic or clear-cut psychosis before inviting patients to SIPS interview. With such a procedure, the cases that actually meet criteria for frank psychosis at baseline are a minority. For example, the Portland (Maine) Identification and Early Referral service accounted for all referrals from 2001 to 2007 (McFarlane et al., 2010). Of 274 SIPS interviews conducted during that time, 32 (11.6%) met SIPS criteria for psychosis. Similarly Addington et  al. (2008) conducted SIPS evaluations in 365 referrals, of whom 86 (24.3%) met SIPS criteria for psychosis. Additional examples coming from epidemiologic studies are listed in Tables 5.4 and 5.5. Both of these comparisons speak to the construct validity of the SIPS-defined CHR syndrome. The CHR vs psychosis comparison also speaks to the validity of the SIPS determination of psychosis, both at baseline and at conversion evaluation. It should be emphasized that in the papers selected, this later comparison is restricted to studies that used a SIPS evaluation in previ-

ously undiagnosed cases to distinguish between CHR and psychosis. A large number of studies that collected a first episode sample as a control group, having previously established the FEP diagnosis by separate means, are not included. We identified 21 qualifying studies of the construct validity of a SIPS CHR diagnosis versus patients coming to SIPS interview in the same sample who did not qualify either for CHR or for psychosis (Table  5.6). The median sample size was 110 (range 47–692). Studies originated from North America (seven samples), Europe (eight samples), Asia (two samples), and Israel (two samples), with two additional samples collected internationally. Six of these studies employed modified SIPS criteria for making the CHR vs non-CHR distinction. Eleven studies reported on psychosis-related scales, including two on BPRS scales and ten on SOPS subscales other than the positive symptom subscale that contributed to the diagnostic distinction. CHR patients scored significantly higher than healthy subject controls (HSC) on all three BPRS comparisons, on 6/10 comparisons of the SOPS negative symptom subscale, on 7/10 ­comparisons of the SOPS disorganized symptom

Table 5.4  General population studies using the SIPS or SOPS Citation Country N a. Studies reporting CHR rates according to SIPS criteria Salokangas et al. (2004) Finland 34 Switzerland 1229 Schultze-Lutter et al. (2014)a Wang et al. (2015) China 2800 Schimmelmann et al. (2011) Switzerland 58 Woods, Walsh, Saksa, and McGlashan USA 30 (2010) Chen et al. (2014) China 579 Schimmelmann et al. (2015)a Switzerland 76 Kelleher et al. (2012) Ireland 212 Koren et al. (2016) Israel 100 b. Studies reporting CHR rates only by modified SIPS criteria Finland 9156 Veijola et al. (2013)c

Age

% meeting psychosis

% meeting CHR

14–40 16–40 18.8 ± 1.1 16–35 25 ± 3

0% Screened out Screened out Screened out 0%

0% 0.4% 1.6%b 1.7% 3.3%

16–22 8–15 11–13 13–16

nr Screened out nr Screened out

3.5% 7.9% 9.0% 12.0%

21–23

2.1%

5.4%

prop proposed, plan planned, nr not reported, na not available Did not assess for GRD, which is relatively rare unless comorbid with APSS (Sect. 2.1) b Assuming the screened sample is representative for the whole and that zero screen negative subjects would have met CHR criteria c Modified criteria did not require recent onset of worsening. Calculations assume the interviewed sample is representative for the whole a

5  Reliability, Validity, Epidemiology, and Cultural Variation of the Structured Interview for Psychosis-Risk…

99

Table 5.5  Clinical epidemiologic studies using the SIPS or SOPS Citation Country Zhang et al. (2014) China Salokangas et al. (2004) Manninen et al. (2014) Kobayashi et al. (2008) Masillo et al. (2018) Comparelli et al. (2010) Lindgren et al. (2014) Gerstenberg et al. (2015) Koren et al. (2013) Comparelli et al. (2016) Raballo et al. (2018)

N Age 1461 15–45

Finland

76 14–40

Finland

52 15–18

Setting 1st MHC visit

% psychosis 6.6%a

% CHR 5.5%a

CHR conversion rate 14/53 (26.4%) over 2 years nr

Japan

115 16–30

Entering OP or IP treatment Reform school residents 1st MHC visit

Italy

338 12–35

OPs in 12 clinics

Italy

128 15–30

20.3%

Finland

683

22.9%a 2/51 (3.9%) over 1 year 23.6%c nr

Switzerland

89

Israel

87

Italy

115

Italy

96

8.8%a

11.7%a

Screened out 1.7%

13.5%

3.0%b

18.9%b

1/7 (14.3%) over 5 years 4/16 (25%) over 6 months nr nr

1st IP or OP MH 5.5% contact 15–18 1st public MH clinic 5.7%a visit 12–17 Consecutive IP Screened out 14–18 New OP Screened help-seekers out 15–25 New OP 8.7%d help-seekers 15.5 ± 1.2 Consecutive IP 13.5%

16.5%

32%b

nr

35.7%d nr 38.3%

nr

MHC Mental Health Center, MH mental health, nr not reported, IP inpatient, OP outpatient Assumes that the interview acceptance rates and SIPS diagnosis rates in the sample of screen-negative patients interviewed are representative of the whole b Assumes that none of the subjects who screened negative and were not interviewed with the SIPS would have met criteria c Met DSM-5 criteria for Attenuated Psychosis Syndrome based on SIPS d Assumes the sample not excluded for substance abuse etc. is representative of the whole a

subscale, and on 7/10 comparisons of the SOPS general symptom subscale. These findings support the convergent validity of the SIPS CHR vs HSC distinction, independent of country. Eleven studies reported on the GAF, a scale in which both symptoms and functioning contribute to severity. CHR patients scored significantly higher than HSCs on 9/11 comparisons, supporting the measure’s construct validity. The studies made 33 comparisons on other measures of functioning, from 4 different instruments (Global Functioning Scale-Social, Global Functioning Scale-Role, Behavior Assessment Scale for Children, Child Behavior Checklist). Of these, CHR patients scored significantly higher than HSCs on more than half of the comparisons (18/33). These findings generally support the validity of the SIPS CHR construct.

Six studies reported CHR vs HSC comparisons on one or more type of nonpsychotic symptoms. A certain amount of comorbidity with nonpsychotic symptoms is expected, but since the SIPS/SOPS is intended to identify and assess CHR syndromes, low correlations with measures designed to assess other constructs would support construct validity. One study used a mania scale (Young Mania Rating Scale), three depression scales (Beck Depression Inventory and Montgomery-Asberg Depression Rating Scale), one an anxiety scale (Beck Anxiety Inventory), and two a scale of affective and anxiety symptoms (Mood and Anxiety Symptom Questionnaire). An additional comparison employed a general symptom scale (the Symptom Check List-90). Results on the mania severity comparison were not significant. In two of the three samples, CHR patients scored significantly

Study Gerstenberg et al. (2015) Comparelli et al. (2011) Simon et al. (2012) Woods et al. (2018) Carrión et al. (2018) Tso et al. (2017)e Raballo et al. (2018) Woods et al. (2009) Addington et al. (2012) Liu et al. (2011)e Millman et al. (2018)e Lo Cascio et al. (2016) Raballo et al. (2016) Weisman et al. (2017)g Thompson et al. (2015) Lo Cascio et al. (2017) ns 0.009d

0.056d

 92

294

290  83

USA, Canada USA

Italy