Crisis 1/2018

Crisis

Volume 39 / Number 1 / 2018

Editor-in-Chief Jane Pirkis Associate Editors Maria A. Oquendo Ella Arensman

The Journal of Crisis Intervention and Suicide Prevention

Published under the auspices of the International Association for Suicide Prevention (IASP)

Hogrefe OpenMind Open Access Publishing?

It’s Your Choice!

Your Road to Open Access Authors of papers accepted for publication in any Hogrefe journal can now choose to have their paper published as an open access article as part of the Hogrefe OpenMind program. This means that anyone, anywhere in the world will – without charge – be able to read, search, link, send, and use the article for noncommercial purposes, in accordance with the internationally recognized Creative Commons licensing standards.

The Choice Is Yours 1. Open Access Publication: The final “version of record” of the article is published online with full open access. It is freely available online to anyone in electronic form. (It will also be published in the print version of the journal.)

2. Traditional Publishing Model: Your article is published in the traditional manner, available worldwide to journal subscribers online and in print and to anyone by “pay per view.”

Whichever you choose, your article will be peer-reviewed, professionally produced, and published both in print and in electronic versions of the journal. Every article will be given a DOI and registered with CrossRef.

www.hogrefe.com

How Does Hogrefe’s Open Access Program Work?

After submission to the journal, your article will undergo exactly the same steps, no matter which publishing option you choose: peer-review, copy-editing, typesetting, data preparation, online reference linking, printing, hosting, and archiving. In the traditional publishing model, the publication process (including all the services that ensure the scientific and formal quality of your paper) is financed via subscriptions to the journal. Open access publication, by contrast, is financed by means of a one-time article fee (€ 2,500 or US $3,000) payable by you the author, or by your research institute or funding body. Once the article has been accepted for publication, it’s your choice – open access publication or the traditional model. We have an open mind!

Crisis The Journal of Crisis Intervention and Suicide Prevention

Volume 39 / Number 1 / 2018 Published under the Auspices of the International Association for Suicide Prevention (IASP)

Editor-in-Chief

Jane Pirkis, Centre for Mental Health, Melbourne School of Population and Global Health, University of ­Melbourne, Melbourne 3010, VIC, Australia (Tel. +61 3 8344-0647, Fax +61 3 9348-1174, E-Mail j.pirkis@unimelb.edu.au)

Associate Editors

Maria A. Oquendo, MD, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA (Tel. +1 215 662-2818, Fax +1 215 662-6911, E-mail moquendo@pennmedicine.upenn.edu) Ella Arensman, PhD, National Suicide Research Foundation, Department of Epidemiology and Public Health, University College Cork,Western Gateway Building, Room 4.34,Western Road, Cork, Ireland (Tel. +353 21 420-5551 or +353 87 052-2284, E-mail earensman@ucc.ie)

Editorial Board

A. Apter, Tel Aviv, Israel R. Battegay, Basel, Switzerland M. P. Battin, Salt Lake City, UT, USA A. L. Beautrais, Christchurch, New Zealand A. L. Berman, Washington, DC, USA S. S. Canetto, Fort Collins, CO, USA J. Cutcliffe, Amherstview, ON, Canada D. De Leo, Mt. Gravatt, Australia O. Ekeberg, Oslo, Norway A. Erlangsen, Aarhus, Denmark M. Goldblatt, Boston, MA, USA D. Gunnell, Bristol, UK K. Hawton, Oxford, UK L. M. Hayes, Mansfield, MA, USA K. van Heeringen, Gent, Belgium H. Hjelmeland, Trondheim, Norway E. Isometsa, Helsinki, Finland G. Jessen, Odense, Denmark N. Kapur, Manchester, UK N. Kapusta, Vienna, Austria A. J. F. M. Kerkhof, Amsterdam, The Netherlands M. M. Khan, Karachi, Pakistan

Editor Emeritus

Diego De Leo

Past Editors-in-Chief

Raymond Battegay, Annette L. Beautrais, David C. Clark, John F. Connolly, Diego De Leo, Ad J. F. M. Kerkhof, Hermann Pohlmeier

Responsible Organization

Publisher

The journal is published under the auspices of the International Association for Suicide Prevention (IASP) (IASP Central Administrative Office, National Centre for Suicide Research and Prevention, Sogsvannsveien 21, Building 12, N-0372 Oslo, Norway, Tel. +47 22 923715, E-mail admin@iasp.info, www.iasp.info).

Production

Juliane Munson, Hogrefe Publishing, Merkelstr. 3, 37085 Göttingen, Germany Tel. +49 551 99950-0, Fax +49 551 99950-111, E-mail juliane.munson@hogrefe.com

Subscriptions

Hogrefe Publishing, Herbert-Quandt-Str. 4, 37081 Göttingen, Germany Tel. +49 551 99950-900, Fax +49 551 99950-998

Advertising/Inserts

Marketing, Hogrefe Publishing, Merkelstr. 3, 37085 Göttingen, Germany, Tel. +49 551 99950-0 Fax +49 551 99950-111, E-mail marketing@hogrefe.com

ISSN

ISSN-L 0227-5910, ISSN-Print 0227-5910, ISSN-Online 2151-2396

Copyright Information

Publication

© 2018 Hogrefe Publishing. This journal as well as the individual contributions and illustrations contained within it are protected under international copyright law. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without prior written permission from the publisher. All rights, including translation rights, reserved.

Subscription prices

Annual subscription (2018, 6 issues): Institutions US $379.00 / € 298.00; Individuals US $179.00 / € 128.00.

Postage and handling: US $24.00 / € 18.00. Single issues: US $63.00 / € 49.50 (+ postage and handling)

Payment

Payment may be made by check, international money order, or credit card, to Hogrefe Publishing, Merkelstr. 3, D-37085 Göttingen, Germany, or, for North American customers, to Hogrefe Publishing Corp., Journals Department, 7 Bulfinch Place, 2nd floor, Boston, MA 02114, USA.

Electronic Full Text

The full text of Crisis is available online at www.econtent.hogrefe.com

Abstracting Services

The journal is abstracted/indexed in Index Medicus/Medline, Current Contents/Social and Behavioral Sciences (CC/S&BS), Social Sciences Citation Index (SSCI), Social SciSearch, PsycINFO, PSYNDEX, Scopus, and EMCare.

Impact Factor (2016): 1.44.

Crisis (2018), 39(1)

K. Kõlves, Brisbane, Australia A. A. Leenaars, Windsor, Canada J. L. McIntosh, South Bend, IN, USA J. Mann, New York, NY, USA M. Marttunen, Helsinki, Finland L. Mehlum, Oslo, Norway B. L. Mishara, Montreal, Québec, Canada T. Niederkrotenthaler, Vienna, Austria S. Occhipinti, Nathan, Australia (Statistical Advisor) R. O’Connor, Glasgow, UK M. Phillips, Shanghai, China M. Pompili, Rome, Italy A. Preti, Cagliari, Italy M. Silverman, Chicago, IL, USA S. Stack, Detroit, MI, USA L. Vijayakumar, Chennai, India M. Voracek, Vienna, Austria D. Wasserman, Stockholm, Sweden E. J. de Wilde, Rotterdam, Netherlands P. Yip, Hong Kong SAR, China J. Zhang, Buffalo, NY, USA

Hogrefe Publishing GmbH, Merkelstr. 3, 37085 Göttingen, Germany Tel. +49 551 99950-0, Fax +49 551 99950-111, E-mail publishing@hogrefe.com,Web http://www.hogrefe.com North America: Hogrefe Publishing, 7 Bulfinch Place, Suite 202, Boston, MA 02114, USA, Tel. (866) 823-4726, Fax (617) 354-6875, E-mail customerservice@hogrefe.com, Web http://www.hogrefe.com

Published in 6 issues per annual volume.

© 2018 Hogrefe Publishing

Contents Editorial Passing the Baton

Diego De Leo and Jane Pirkis

Research Trends Risk Factors and Mediators of Suicidal Ideation Among Korean ­ Adolescents

Yi Jin Kim, Sung Seek Moon, Jang Hyun Lee, and Joon Kyung Kim

1

4

Telephone Crisis Support Workers’ Psychological Distress and Impairment: A Systematic Review Taneile A. Kitchingman, Coralie J. Wilson, Peter Caputi, Ian Wilson, and Alan Woodward

13

Age at Exposure to Parental Suicide and the Subsequent Risk of Suicide in Young People Kuan-Ying Lee, Chung-Yi Li, Kun-Chia Chang , Tsung-Hsueh Lu, and Ying-Yeh Chen

27

Patient-Identified Priorities Leading to Attempted Suicide: Life Is Lived in Interpersonal Relationships Niklaus Stulz, Urs Hepp, Dominic G. Gosoniu, Leticia Grize, Flavio Muheim, Mitchell G. Weiss, and Anita Riecher-Rössler

37

Effects of Awareness Material on Suicide-Related Knowledge and the Intention to Provide Adequate Help to Suicidal Individuals Florian Arendt, Sebastian Scherr, Thomas Niederkrotenthaler, Sabrina Krallmann, and Benedikt Till

47

Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices: The Impact of Fear of Suicide-Related Outcomes and ­Comfort Working With Suicidal Individuals Jared F. Roush, Sarah L. Brown, Danielle R. Jahn, Sean M. Mitchell, Nathanael J. Taylor, Paul Quinnett, and Richard Ries

55

Short Report Investigating the Dynamics of ­Suicidal Ideation: Preliminary Findings

From a Study Using Ecological ­Momentary Assessments in Psychiatric Inpatients

Nina Hallensleben, Lena Spangenberg, Thomas Forkmann, Dajana Rath,

Ulrich Hegerl, Anette Kersting, Thomas W. Kallert, and Heide Glaesmer

65

Framing Suicide – Investigating the News Media and Public’s Use of the

Problematic Suicide Referents Freitod and Selbstmord in German-Speaking Countries

Florian Arendt

News, Announcements, and IASP

70

© 2018 Hogrefe Publishing

74

Crisis (2018), 39(1)

Editorial Passing the Baton Diego De Leo and Jane Pirkis

Diego De Leo After a renewal of my initial 4-year term, my mandate as Editor-in-Chief of Crisis was supposed to end in July 2016.

I happily continued in my role for an extra 18 months, until the search for a new editor was eventually finalized. It has been a long journey and a major commitment, but I certainly did not walk alone during these years. First of all I would therefore like to express my gratitude to the International Association for Suicide Prevention (IASP), which honored me by appointing me to the editorship of the journal in the first place. It has been a great privilege and responsibility. It goes without saying that suicide, apart from being the worst of all human tragedies, represents a kind of battlefield of different theories, approaches, viewpoints, religious perspectives, and ethical dilemmas. However, my editorship crossed five different IASP Executive Boards and nobody has ever attempted to interfere with my work and the editorial independence of the journal; at the same time I was always supported in the management of the journal and felt appreciated for the efforts I and others were making in its development. Many thanks, then, to all the IASP Executive Board members that I encountered for their impeccable behavior during my editorship.

In nearly 10 years of editorship, the composition of Crisis’ Editorial Board has changed remarkably. A few people retired, others passed away. I cannot but think of Norman Farberow and Terry Maltsberger, great figures in the world of suicidology who are not any longer with us, and would also would like to mention the assistance gently provided by Bela Buda for many years. He too is missed by many of us.

All members of the journal’s Editorial Board have been instrumental in the development of Crisis and I am deeply grateful to all of them for this. A special mention has to go to my Associate Editors, first of all to Ad Kerkhof, who preceded me in the management of Crisis and who certainly did a lot in establishing the journal as one of the leaders in the area of suicide research and prevention. Then, Maria Oquendo and Ella Arensman have been my alter egos for many years and their significance as leading researchers has contributed to the status and reputation of Crisis. © 2018 Hogrefe Publishing

Apart from the members of the Editorial Board, my heartfelt thanks go to the many anonymous reviewers who have helped me so much in choosing and improving the quality of the hundreds of papers published in the past decade by Crisis. Their contributions have been of paramount importance in securing a good balance and appropriate selection of articles – and their rigorous work now means that less than 20% of all submissions are finally selected for publication.

I have found Hogrefe Publishing to be a great partner.

Robert Dimbleby and Juliane Munson, in particular, have been fantastic travel mates throughout these past years.

Timely, precise, open to innovations, they have been a central element in the success of the journal.

Griffith University has been incredibly generous in supporting the editorial responsibilities of Crisis all along my mandate, and I will never forget the unmatchable role that Wendy Iverson has played in assisting me with my editorial duties. Wendy will also assist the new Editor-in-Chief, transferring invaluable knowledge acquired over nearly 10 years and thus securing a very smooth transition.

I am not exaggerating when I say that the role of new Editor-in-Chief could not be placed in any better hands than those of Professor Jane Pirkis. Jane is a well-established and prolific scholar in suicide research and prevention.

With more than 250 peer-reviewed papers published and two decades of active presence in the international arena, she represents an excellent choice by the IASP committee formed to appoint a new editor for Crisis. Currently, Jane is also the IASP 1st Vice President, with previous experience on the board of the association. This is a plus, since it has permitted her to acquire familiarity with virtually all problems related to the field of suicide prevention.

A registered psychologist, Jane is the Director of the Centre for Mental Health in the Melbourne School of Population and Global Health at the University of Melbourne.

She is a robust scientist, with a particular interest in the evaluation of suicide prevention programs and mental health care initiatives. She is also known worldwide for her particular research interest in suicide and the media.

She was the recipient of the Lifetime Research Award by Suicide Prevention Australia in 2010. With Rory O’ConCrisis (2018), 39(1), 1–3 https://doi.org/10.1027/0227-5910/a000530

2

nor, she co-edited the second edition of the International Handbook of Suicide Prevention (O’Connor & Pirkis, 2016).

In addition to her many academic achievements and research projects, at the relationship level Jane is the type of person that can make everybody feel at home. This is a rare quality, and one that fits well with my special wish for her and the future of Crisis: that the journal establishes itself as the common house of scholars and contributors from all parts of the world, and that cultural diversities may also be recognized as a powerful resource for teaching suicide prevention.

Good luck, Jane!

Jane Pirkis I feel very privileged to be taking over as Editor-in-Chief from Diego De Leo. Diego became the Editor-in-Chief of Crisis in 2008, having been on the Editorial Board since 1990. He shared the role with Annette Beautrais until 2010 and then continued solo. He was visionary and was keen to see Crisis become the go-to journal for those working in suicide prevention. He actively set out to ensure that the journal would attract and publish the best studies in the field, ones that would significantly advance knowledge and influence policy and practice.

Diego’s vision has certainly been achieved if standard metrics are anything to go by. On his watch, Crisis has gone from strength to strength. In 2010 the journal moved to six issues per year, as opposed to four, and in 2013 its volume was increased by a further 20% when it expanded to approximately 72 pages per issue. The journal’s impact factor rose from 1.127 in 2008 to 1.440 in 2016 (and its 5-year impact factor increased from 1.597 in 2010 to 2.400 in 2016). In 2016, articles in Crisis were cited a total of 1,318 times.

Perhaps more importantly, there is objective evidence that the journal is having its desired impact at the policy and practice level. By way of example, the World Health Organization (2014) report entitled Preventing Suicide: A Global Imperative – arguably the most influential guide to suicide prevention activity internationally – cites five key articles from Crisis in its reference list (Bohanna & Wang, 2012; Coveney, Pollock, Armstrong, & Moore, 2012; Jansen, Buster, Zuur, & Das, 2009; Luxton, June, & Comtois, 2013; Mishara & Martin, 2011).

Diego is one of the most respected senior suicide prevention researchers in the world. He has an enviable list of research outputs that includes over 350 peer-reviewed articles and 170 book chapters. His epidemiological work has furthered our understanding of suicidal behavior among different cultural groups and among older people, Crisis (2018), 39(1), 1–3

Editorial

and his large-scale, multisite intervention studies have improved our knowledge of effective suicide prevention strategies in different countries. His significant achievements have been recognized by numerous honors, including the International Academy of Suicide Research Morselli Medal (2017), the American Association of Suicidology’s Louis I Dublin Award (2011), Suicide Prevention Australia’s Lifetime Research Award (2007), and the International Association for Suicide Prevention’s Stengel Award (1991). Each of these awards recognizes outstanding contributions to the study of suicidal behavior. With these credentials, it is not surprising that Diego was able to achieve so much success as Editor-in-Chief of Crisis. Diego led by example and was willingly supported by those in his extensive international network of suicide prevention researchers who keenly submitted some of their most outstanding manuscripts to the journal.

For my part, I am very pleased to be taking on the role of Editor-in-Chief because Crisis has a special place in my heart. I sent one of my first ever articles on suicide prevention to the journal as a “green” researcher in 2000, and was over the moon when it was accepted (Pirkis, Burgess, & Dunt, 2000). Ad Kerkhof was the Editor-in-Chief at the time and encouraged me to submit another one, so I felt like I had really made it as a suicide prevention academic!

I’m proud to say that to this day, that first article and two others that I published in Crisis later (Pirkis & Blood, 2001; Vijayakumar, John, Pirkis, & Whiteford, 2005) are among my six most highly cited articles.

My own vision for Crisis builds on Diego’s. I’m keen to consolidate and leverage the firm foundation that Diego has laid and to continue to maintain the journal’s positive trajectory. I’d like it to reflect on where we’re at now in suicide prevention and showcase the cutting-edge research that can help us get to where we want to be. We know a lot more about what works and what doesn’t work in suicide prevention than we did 10 years ago, but there is still a great deal to be learned. We’re overcoming some of the definitional, ethical, methodological, and practical issues that have traditionally beset suicide prevention research, but we haven’t nailed them yet. And we’re better than we used to be about taking into account the perspectives of those with lived experience of suicide, but there is definitely still room for improvement.

Diego has worked tirelessly to make Crisis the success it is today, and I am conscious that I have big (stylish Italian leather) shoes to fill. I am delighted that Diego will continue to play an important role as Editor Emeritus and will remain on the Editorial Board, and would like to thank him sincerely for everything he has done for the journal in the past nine years.

Onward and upward!

© 2018 Hogrefe Publishing

3

Editorial

References

Published online February 14, 2018

Bohanna, I., & Wang, X. (2012). Media guidelines for the responsible reporting of suicide: A review of effectiveness. Crisis, 33(4), 190–198.

Coveney, C., Pollock, K., Armstrong, S., & Moore, J. (2012). Callers’ experiences of contacting a national suicide prevention helpline. Crisis, 33(6), 313–324.

Jansen, E., Buster, M., Zuur, A., & Das, C. (2009). Fatality of suicide attempts in Amsterdam 1996–2005. Crisis, 30(4), 180–185.

Luxton, D., June, J., & Comtois, K. (2013). Can postdischarge follow-up contacts prevent suicide and suicidal behaviour? A review of the evidence. Crisis, 34(1), 32–41.

Mishara, B., & Martin, N. (2011). Effects of a comprehensive police suicide prevention program. Crisis, 33(3), 162–168.

O’Connor, R., & Pirkis, J. (2016). The International Handbook of Suicide Prevention. Chichester, UK: Wiley.

Pirkis, J., & Blood, R. W. (2001). Suicide and the media: Part 1. Reportage in nonfictional media. Crisis, 22(4), 146–154.

Pirkis, J., Burgess, P., & Dunt, D. (2000). Suicidal ideation and suicide attempts among Australian adults. Crisis, 21(1), 16–25.

Vijayakumar, L., John, S., Pirkis, J., & Whiteford, H. (2005). Suicide in developing countries (2): Risk factors. Crisis, 26(3), 112–119.

World Health Organization. (2014). Preventing suicide: A global imperative. Geneva, Switzerland: Author.

Diego De Leo, MD, PhD, DSc, FRANZCP, is Emeritus Professor of Psychiatry at Griffith University, Australia, and Director of the Slovene Centre for Suicide Research, Primorska University, Slovenia. Prof. De Leo was President of the International Association for Suicide Prevention and the International Academy for Suicide Research.

© 2018 Hogrefe Publishing

Professor Jane Pirkis is Director of the Centre for Mental Health at the University of Melbourne, Australia, and is the 1st Vice President of the International Association for Suicide Prevention.

Prof. Diego De Leo Griffith University Mt. Gravatt 4122 QLD Australia d.deleo@griffith.edu.au Prof. Jane Pirkis Centre for Mental Health Melbourne School of Population and Global Health University of Melbourne Melbourne 3010, VIC Australia j.pirkis@unimelb.edu.au

Crisis (2018), 39(1), 1–3

Research Trends

Risk Factors and Mediators of Suicidal Ideation Among Korean ­Adolescents Yi Jin Kim1, Sung Seek Moon2, Jang Hyun Lee3, and Joon Kyung Kim4 University of Mississippi, School of Applied Sciences, Department of Social Work, University, MS, USA University of South Carolina, College of Social Work, Columbia, SC, USA 3 Pyeongtaek University, Department of Child & Youth Welfare, Gyeonggi-do, South Korea 1 2

Namseoul University, School of Child Welfare, Chungcheongnam-do, South Korea

4

Abstract. Background: A significant number of Korean adolescents have suicidal ideations and it is more prevalent among adolescents than any other age group in Korea. Aims: This study was conducted to attain a better understanding of the contributing factors to suicidal ideation among Korean adolescents. Method: We recruited 569 high school students in Grades 10 and 11 in Pyeongtaek, Korea. The Beck Scale for Suicidal Ideation was used to measure suicidal ideation as the outcome variable. The Interpersonal Needs Questionnaire, the Beck Hopelessness Scale, the School Related Stress Scale, the Olweus Bully/Victim Questionnaire, and the Youth Risk Behavior Surveillance questions were used to measure thwarted belongingness and perceived burdensomeness, hopelessness, school-related stress, bullying, and previous suicidal behaviors, respectively. Data analyses included descriptive statistics and structural equation modeling.

Results: The findings suggest that perceived burdensomeness, hopelessness, school-related stress, and previous suicidal behaviors have significant direct effects on suicidal ideation. Hopelessness fully mediated the relation between thwarted belongingness and suicidal ideation, and partially mediated between perceived burdensomeness, school-related stress, and suicidal ideation. Conclusion: These findings provide more specific directions for a multidimensional suicide prevention program in order to be successful in reducing suicide rates among Korean adolescents.

Keywords: suicidal ideation, hopelessness, Korean adolescents, interpersonal theory of suicide

Suicide is the number one leading cause of death among the 10- to 30-year-old age group in Korea (Statistics Korea, 2010). The suicide rate among all adolescents deaths aged 15–19 years has more than doubled from 13.6% in 2000 to 28.2% in 2010 (Korean Statistical Information Service, 2010). Suicide has currently outpaced all other causes and has been the leading cause of death since 2008 (Statistics Korea, 2010). More adolescents die from suicide than from car accidents, cancer, heart disease, drowning accidents, and even homicide (Statistics Korea, 2011a).

It is interesting that Korea has shown the highest suicide rate among Asian countries, even though these countries share a similar cultural background. A majority of studies stressed that Korean adolescents have unique cultural characteristics that may work as triggers for suicide: extreme academic stress (Lee & Jang, 2011), an education system that focuses mainly on the college entrance exam (Yoon, 2010), excessive competition (Yoon, 2010), and bullying victimization (Hong, 2011). Also, these risk factors have been regarded as one of the most significant risk factors for completing suicide among Korean adolescents. Thus, the current study focuses on the relationship Crisis (2018), 39(1), 4–12 https://doi.org/10.1027/0227-5910/a000438

between risk factors and suicidal ideation so as to prevent further suicide behaviors among Korean adolescents.

Literature Review Substance Use and Suicide Studies reporting a significant correlation between substance use and suicide among adolescents have become abundant (Conason, Oquendo, & Sher, 2005a, 2005b). Suicidal behaviors were more likely to occur among people who used illegal drugs (Gould, Greenberg, Velting, & Shaffer, 2003), abused alcohol (Chung, 2011), or smoked cigarettes (Kim, Shim, Noh, Hwang, & Park, 2013). The Substance Abuse and Mental Health Services Administration (US Department of Health and Human Services, 2002) suggested that among adolescents aged 12–17 years who attempted suicide, approximately 20–30% were substance users, compared with 10% who were nonusers. This trend was similar in the Korean adolescent population (Chung, 2011). © 2016 Hogrefe Publishing

5

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

Bullying Victimization and Suicide Several studies examining suicidal behaviors among adolescents included the risk factor of negative relationships with friends instigated by bullying victimization and peer violence (Hong, 2011). Adolescents who were victims of bullying were more likely to have suicidal ideation (Hong, 2011) and to attempt suicide (Brunstein Klomek, Sourander, & Gould, 2010). Roeger, Allison, Korossy-Horwood, Eckert, and Goldney (2010) posited that adolescents with a history of bullying victimization were nearly three times more likely to have suicidal ideation than those who did not have such an experience.

Hopelessness and Suicide Hopelessness has been reported as one of the most significant long-term risk factors for suicidal behaviors in clinical populations (Joiner, Brown, & Wingate, 2005). Specifically, hopelessness has been associated with suicidal ideation, verbally expressed suicidal intent, and past and future suicidal attempts among adolescents (Goldston et al., 2001).

These findings point to a central role for hopelessness in predicting suicidal behaviors.

Aside from its strong predictive relationship to suicidal behaviors, hopelessness also has been shown to mediate the associations between suicidal ideations and other risk factors including thwarted belongingness (Joiner, 2005), perceived burdensomeness (Van Orden et al., 2010), school-related stress (Jang, 2010), and bullying victimization (Bonanno & Hymel, 2010).

Previous Suicidal Behaviors and Suicide The risk factors related to repeated suicide attempts have been studied extensively. Most of the studies consistently indicated that suicide attempts may become a repeated pattern of behavior. Previous research has shown that a history of suicide attempt is a main risk factor for completing suicide attempt (Reid, 2009). Beck claimed that previous suicidal behaviors sensitized suicidal ideation and behaviors such that previous attempters later became more accessible (Teasdale, 1988). The more accessible the modes became, the more easily the previous attempters were triggered (Joiner & Rudd, 2000).

School-Related Stress and Suicide Korean adolescents experience extreme stress due to academic and social demands in school. Entering and graduat© 2016 Hogrefe Publishing

ing from a high-ranking university in Korea means a greater chance of obtaining a good job with high wages, a high social status, and a good spouse (Lee & Larson, 2000). Also, graduating from a tier-one university has been related to saving face or in other words avoiding bringing shame to one’s self and most importantly to one’s family. Koreans have been strongly influenced by Confucianism, and saving face is one of the characteristics of Confucian culture. For these reasons, Korean adolescents and even their parents are under huge pressure for academic achievement and especially for good grade point average (GPA) scores.

Previous studies have shown that the top reason for completing suicide (54%) was the pressure and anxiety related to academic performance including high academic expectations, academic workload, and academic competition among peers (Statistics Korea, 2011b). A number of studies also have found that school-related stress directly and positively predicted suicidal ideation among Korean adolescents (Lee & Jang, 2011).

Gap in the Literature The main gap in the literature is the lack of comprehensive models explaining the complex phenomenon of adolescents’ suicidal ideation. Although some studies adopted theories to explain the relationship between risk factors and suicidal ideation, the theories were not comprehensive enough to elucidate the diverse risk factors of adolescent suicidal ideation. To fulfill the gaps in previous research, the current study examines risk factors for suicidal ideation among Korean adolescents by testing a modified interpersonal theory of suicide developed by Joiner (2005).

The current study is the first attempt to apply Joiner’s theory to a Korean adolescent population. Although Joiner developed the theory based on the results from multiple and diverse samples including young adults, undergraduates, and older adults, the majority of the samples were based on US populations (Joiner, 2005). Therefore, this study could contribute to testing the generalizability of the theory by applying the theory to a Korean adolescent population who may have different risk factors for suicidal ideation.

Theoretical Framework The interpersonal theory of suicide was developed based on the key points of mainstream suicide theories such as the theories of Durkheim, Shneidman, Beck, Baumeister, and Linehan (Joiner, 2005). The theory posits that an individual will carry out lethal suicide attempts when the person simultaneously has the desire and capability to attempt Crisis (2018), 39(1), 4–12

6

suicide (Van Orden et al., 2010). This theory stipulates the relationships among risk factors and suggests a causal pathway from suicidal ideation to engagement in serious suicidal behavior. However, the interpersonal theory of suicide fails to include other significant risk factors, such as previous suicidal behaviors, substance use, school-related stress, and bullying victimization, which have been proven to be significant predictors of suicidal behaviors, especially among Korean adolescents.

To fill in the gap of Joiner’s interpersonal theory of suicide, a modified interpersonal theory of suicide is suggested for better prediction of suicide. We added the following risk factors in the modified theoretical model to attain a comprehensive understanding of complex suicidal ideation among Korean adolescents: previous suicidal behaviors, substance use, school-related stress, and bullying victimization. Based on previous research, it is apparent that there are gender differences in suicidal ideation among Korean adolescents. Although gender is a significant factor that affects suicidal ideation, it is not a main focus of the current research. Therefore, the hypothesized model controlled for gender.

The hypothesized model (see Figure 1) based on a theoretical framework and empirical research evidence includes a specification of the relations between several risk factors in the form of two objectives. The paths proposed in the model are as follows: −− Objective 1: To examine if there is a direct, significant, and positive relationship between risk factors (thwarted belongingness, perceived burdensomeness, hopelessness, school-related stress, bullying victimization,

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

previous suicidal behaviors, substance use) and suicidal ideation when controlling for gender. −− Objective 2: To examine if hopelessness mediates the relationship between risk factors (thwarted belongingness, perceived burdensomeness, school-related stress, bullying victimization) and suicidal ideation when controlling for gender.

Method Sampling Study participants were recruited from six high schools in Pyeongtaek, South Korea. Pyeongtaek city was chosen as a study location because the rate of suicidal ideation among adolescents in Gyeonggi-do (18.5%), a province of Korea that includes Pyeongtaek city, is approximately equal to the national average (18.3%; Statistics Korea, 2012a). A total of 850 students were contacted and 569 students participated in the study. The data were gathered from November 2013 to February 2014.

A two-stage cluster sampling was used to choose participants from all high school students in Pyeongtaek. Among the 20 high schools in Pyeongtaek, six schools were randomly chosen. Considering the fact that there were 35 to 40 students in a class, three classes were randomly chosen from each randomly selected school. The primary data collection method was a written survey. Surveys were administered at the schools to each of the participants. We reFigure 1. Hypothesized model. BU = bullying victimization. HP = hopelessness. PB = perceived burdensomeness. PS = previous suicidal behaviors. SI = suicidal ideation. SS = school-related stress. SU = substance use. TB = thwarted belongingness.

FigureCrisis 1. Hypothesized model.

BU = bullying victimization. HP = hopelessness. PB = perceived (2018), 39(1), 4–12 burdensomeness. PS = previous suicidal behaviors. SI = suicidal ideation. SS = school-related stress. SU = substance use. TB = thwarted belongingness.

© 2016 Hogrefe Publishing

7

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

ceived verbal agreement and letters of support from the six high schools in Pyeongtaek to assist in the implementation of data collection. Students who received permission from parents and who themselves agreed to participate were involved in a single group-testing session (30–40 min) in each classroom.

during the last 30 days was assessed using four items from the Monitoring the Future Survey 2009 ­(Johnston, Bachman, & O’Malley, 2009). These questions were translated by the current authors and back-translated.

Measures

Univariate statistics including frequencies and central tendencies were tested by using the Statistical Package for the Social Sciences Version 21 to describe the distribution of demographic, outcome, and predictor variables. Then, structural equation modeling (SEM) was used (with AMOS Version 21) to test the hypothesized model. Finally, direct and indirect effects of predictor variables on outcome variables were analyzed.

Reliability analyses were examined to determine the internal consistency of each selected variable. Cronbach’s α was 0.64 for suicidal ideation, 0.78 for thwarted belongingness, 0.87 for perceived burdensomeness, 0.66 for hopelessness, 0.88 for school-related stress, 0.94 for bullying victimization, 0.87 for previous suicidal behaviors, and 0.68 for substance use. Overall, these results show acceptable α levels, based on Nunnally’s (1978) minimum criterion of 0.70. All measures were latent constructs derived from multiple indicators.

The current study used the Korean version of the Beck Scale for Suicide Ideation (SSI) translated by Shin, Park, Oh, and Kim (1990) to measure suicidal ideation. According to Beck, Kovacs, and Weissman (1979), the SSI consists of questions related to suicidal ideation, suicide plans, and suicide attitudes. Thus, for the purpose of the current study, only the six questions related to suicidal ideation were used in the analyses. To measure thwarted belongingness and perceived burdensomeness, the Korean version of Interpersonal Needs Questionnaire was used (Jo, 2010): five items measured thwarted belongingness and seven items measured perceived burdensomeness. To measure hopelessness, the short version of the Beck Hopelessness Scale was used, which was translated into Korean by Shin et al. (1990). To measure school-related stress, the School-Related Stress Scale developed by Yoo and Min (1998) and modified by Park (2006) was used. This questionnaire consists of two dimensions: stress from the class environment and stress from the GPA. The Improved Korean Version of the Olweus’ Bully/Victim Questionnaire (Lee, 2007) was used to measure bullying victimization among Korean adolescents. This scale consists of three dimensions: relational, verbal, and physical bullying. Suicide behavior questions from the 2011 Youth Risk Behavior Surveillance questionnaire were used to measure previous suicidal behaviors. Four questions were used to ask about suicidal ideation, plan, attempt, and attempt that resulted in serious physical injury. These questions were translated by the current authors and back-translated to verify the accuracy of the translation because no translated version of the preexisting measurement was found. To measure substance use, adolescents’ own report of cigarette, a ­ lcoholic beverages, hallucinogenic drug, and inhalant drug use © 2016 Hogrefe Publishing

Data Analyses

Results After listwise deletion to deal with the missing data, there were a total of 243 females (55.2%) and 197 males (44.8%) with a mean age of 16.34 years (SD = 0.59, range = 16–19 years) who participated in the study. Out of the total participants, 314 (71.4%) were 10th graders and 126 participants (28.6%) were 11th graders. As far as the participants’ parental information is concerned, most of the participants’ parents (98.2%) were alive and most of the participants (92.7%) lived with their parents; 210 (47.7%) participants reported that their monthly family income was between US $3,000 and US $4,999.

Results indicate that approximately 48% of participants reported suicidal ideation. Among previous suicidal behaviors, the means for suicide planning (M = 0.18, SD = 0.63) were about two times higher than for suicide preparation (M = 0.09, SD = 0.42); suicide attempt (M = 0.07, SD = 0.39) was slightly lower than suicide preparation.

The mean score of suicide attempt resulting in medical treatments (M = 0.04, SD = 0.34) was the lowest, as expected. Regarding bullying victimization, verbal bullying (M = 1.00, SD = 2.90) was more than two times prevalent than relational bullying (M = 0.45, SD = 1.77). Among subcategories of bullying victimization, physical bullying (M = 0.28, SD = 1.84) showed the lowest mean score.

The means for tobacco (M = 1.29, SD = 0.92) and alcohol use (M = 1.30, SD = 0.81) were higher than for hallucinogen (M = 1.03, SD = 0.81) and inhalant use (M = 1.03, SD = 0.41). There were no significant gaps between school-related stress from GPA (M = 13.75, SD = 3.85) and class environment (M = 12.46, SD = 3.90).

The hypothesized model had an adequate goodness of fit. The χ2 statistic (df = 574, N = 440) was 1,382.29 with Crisis (2018), 39(1), 4–12

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

8

p < .001. However, since the χ2 statistic is sensitive to sample size, it could incorrectly imply a poor model fit (Schumacker & Lomax, 2004). To address the limitations of the χ2 statistic, GFI, CFI, and RMSEA were used as alternative goodness-of-fit indices. The statistics indicated good fit with GFI = .86, CFI = .92, and RMSEA = .05. Based on the results of the goodness-of-fit indices, it can be concluded that the proposed structural model is consistent with the observed data.

ly significant path was found in the relationship between thwarted belongingness and suicidal ideation.

Indirect Parameter Estimates for the ­Hypothesized Structural Model Further, a mediation test was conducted by using bootstrapping methods with AMOS. In the current study, a total of 1,000 bootstrap samples were generated and bias-corrected bootstrap CIs at the p < .05 level were used. The results are presented in Table 2.

In the relationship between thwarted belongingness and suicidal ideation through hopelessness, the direct effect of thwarted belongingness on suicidal ideation without the mediator (β = .11) was significant (p < .05), but the same relationship with the mediator was marginally significant (β = .09). The indirect effect (β = .08) between thwarted belongingness and suicidal behavior was significant based on the results of the 95% bias-corrected bootstrap CIs that did not include zero (95% CI = 0.08–0.09). These results indicated that there were full mediation effects of hopelessness between thwarted belongingness and suicidal ideation.

Direct Parameter Estimates for the ­Hypothesized Structural Model A total of 11 direct paths were estimated and the results showed eight paths to be statistically significant at the p < .05 level (see Table 1). These paths were: perceived ­burdensomeness to suicidal ideation; hopelessness to suicidal ideation; school-related stress to suicidal ideation; previous suicidal behaviors to suicidal ideation; thwarted belongingness to hopelessness; perceived burdensomeness to hopelessness; school-related stress to hopelessness; bullying victimization to hopelessness. Also, a marginal-

Table 1. Results of direct effects for the hypothesized structural model Independent variable

Dependent variable

β

b

SE

Thwarted belongingness

→

Suicidal ideation

0.08

0.04

0.09+

Perceived burdensomeness

→

Suicidal ideation

0.12

0.03

0.33***

Hopelessness

→

Suicidal ideation

1.04

0.30

0.29***

School related stress

→

Suicidal ideation

0.03

0.01

0.09**

Bullying victimization

→

Suicidal ideation

−0.02

0.02

Previous suicidal behaviors

→

Suicidal ideation

0.48

0.10

Substance use

→

Suicidal ideation

0.06

0.05

−0.06 .39*** 0.06

Thwarted belongingness

→

Hopelessness

0.06

0.02

0.25**

Perceived burdensomeness

→

Hopelessness

0.03

0.01

0.16*

School related stress

→

Hopelessness

0.01

0.00

0.27**

Bullying victimization

→

Hopelessness

0.01

0.01

0.13*

Note. +p < .10. *p < .05. **p < .01. ***p < .001.

Table 2. Results of mediation effects for the hypothesized structural model Independent variable

Mediating ­variable

Dependent variable

Direct effect without mediator

Direct effect with mediator

Indirect effect

95% CI

TB

→

HP

→

SI

0.11*

0.09+

0.08**

0.08–0.09

PB

→

HP

→

SI

0.34***

0.33***

0.04**

0.04–0.05

SS

→

HP

→

SI

BU

→

HP

→

SI

0.11* −0.03

0.09** −0.04

0.08**

0.08–0.09

0.06**

0.04–0.05

Note. CI = confidence interval. BU = bullying victimization. HP = hopelessness. PB = perceived burdensomeness. SI = suicidal ideation. SS = school-­ related stress. TB = thwarted belongingness. +p < .10. *p < .05. **p < .01. ***p < .001.

Crisis (2018), 39(1), 4–12

© 2016 Hogrefe Publishing

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

The direct effect of perceived ­burdensomeness on suicidal ideation without the mediator (β = .34) was significant (p < .001). Also, the same relationship with the mediator (β = .33) was significant (p < .001). The indirect effect (β = .04) between perceived burdensomeness and suicidal behavior was significant based on the range of CIs that did not include zero (95% CI = 0.04–0.05). These results proved that there were partial mediation effects of hopelessness between perceived burdensomeness and suicidal ideation.

Next, the direct effect of school-related stress on suicidal ideation without the mediator (β = .11) was significant (p < .05). After the addition of the mediator, the relationship between school-related stress and suicidal ideation (β = .09) was still significant (p < .01). The indirect effect (β = .08) between school-related stress and suicidal behavior was significant based on the range of CIs (95% CI = 0.08–0.09). These results indicate that hopelessness partially mediates the effects of school-related stress on suicidal ideation.

No mediation effect was found in the relationship between bullying victimization and suicidal ideation through hopelessness because the direct effect without the mediator was not significant. Baron and Kenny (1986) suggested that a mediation model can be tested when there is a significant relationship between independent and dependent variables.

Discussion and Implications The main focus of this study was to test a theoretical model of adolescents’ suicidal ideation based on the modified interpersonal theory of suicide. The central objectives were verified: The risk factors, including thwarted belongingness, perceived burdensomeness, hopelessness, school-­ related stress, and previous suicidal behaviors, positively influence suicidal ideation among Korean adolescents; the risk factors, including thwarted belongingness, perceived burdensomeness, and school-related stress, influence suicidal ideation indirectly through hopelessness.

The finding that perceived burdensomeness has a positive relationship with suicidal ideation is consistent with other empirical research findings. For example, Van Orden, Cukrowicz, Witte, and Joiner (2012) sampled 912 undergraduate students where the level of suicidal ideation was associated with a higher level of perceived burdensomeness. Thus, the result of this study supports that perceived burdensomeness functions to increase the level of suicidal ideation.

The expected influences of hopelessness on suicidal ideation are confirmed in the model. This finding adds to © 2016 Hogrefe Publishing

9

the body of knowledge about the positive effect of hopelessness on suicidal ideation (Reinecke, DuBois, & Schultz, 2001). Hopelessness also mediated the relationship between several risk factors and suicidal ideation in the current study. This finding is discussed later in detail.

The objective of examining whether there is a direct, significant, and positive relationship between school-related stress and suicidal ideation was verified. The findings are consistent with prior studies reporting a relationship between school-related stress and suicidal ideation (Lee & Jang, 2011). Importantly, the results of the current study provide additional evidence that stress from class environment and GPA are significantly linked to school-related stress among Korean adolescents and lead to increased suicidal ideation.

Previous suicidal behaviors have a strong positive impact on suicidal ideation, thereby increasing suicidal ideation.

This supports the findings from a number of studies that showed a significant association between previous suicidal ideation with an attempt and current suicidal ideation (Joiner, 2005). The findings of the current study prove that the more accessible the modes become, the more easily the previous attempters are triggered (Joiner & Rudd, 2000).

The most unexpected finding of this study is the relationship between thwarted belongingness and suicidal ideation. Contrary to expectations, the results indicate a direct, marginally significant positive relationship between them. Our results indicate thwarted belongingness as a less strong predictor of suicidal ideation among Korean adolescents than other predictor variables tested in the model. According to the results, hopelessness works as a full mediator in the relationship between thwarted belongingness and suicidal ideation, indicating thwarted belongingness affects suicidal ideation through hopelessness. In this regard, the effects of thwarted belongingness on suicidal ideation are explained well when hopelessness is included as a mediator between the relationship.

Our results also indicate there is an insignificant relationship between bullying victimization and suicidal ideation. This finding is not congruent with the previous studies that maintained a significant association between bullying victimization and suicidal ideation (Fisher et al., 2012). One plausible reason for the discrepancy in findings could be that there is the possibility for participants to under-report their experiences of bullying victimization, leading to unexpected and inaccurate results. Adolescents who are bullied may not report their bullying victimization issues to others for two reasons: fear of reprisal and distrust in the hope of being rescued by family, friends, or teachers (Ahn, 1998). In addition, there can be other factors that could cause these unexpected results: the use of unreliable scales, social desirability, and careless responses to the survey.

Crisis (2018), 39(1), 4–12

10

Another new finding is the role of hopelessness as a mediator between several risk factors and suicidal ideation. Although it was expected, according to the interpersonal theory of suicide, that thwarted belongingness and perceived burdensomeness have an effect on suicidal ideation through hopelessness, the results of this study show that hopelessness also mediates the effects of school-related stress on suicidal ideation among Korean adolescents. Given these findings, it can be concluded that hopelessness itself not only is a significant predictor of suicidal ideation, but also a key factor that mediates other risk factors including thwarted belongingness, perceived burdensomeness, and school-related stress.

Although the mediation effect of hopelessness between bullying victimization and suicidal ideation was assumed, no mediation effect was found because the direct effect without a mediator was not significant.

In addition, the results indicate that substance use is not a significant risk factor for suicidal ideation. This finding is not consistent with prior studies that reported a significant association between substance use and suicidal ideation (Kim et al., 2013). Some may argue that the unexpected result is because of employing hallucinogenic and inhalant drug use as indicators in the substance use variable. Compared with cigarette and alcohol use, the rates of hallucinogenic and inhalant drug use among Korean adolescents were relatively low (Korea CDC, 2013). This phenomenon may have affected the relationship between substance use and suicidal ideation in the current study.

This study has strengths that set it apart from other research. It is the first attempt to apply Joiner’s interpersonal theory of suicide to the Korean adolescent population. Also, by adding school-related stress and bullying victimization, which are significant risk factors for Korean adolescents, to Joiner’s model, the current study provides more comprehensive explanations about the relationship between risk factors and suicidal ideation among Korean adolescents.

Despite these strengths, a number of limitations should be discussed to clarify the generalizability of the findings.

First, only high school students were recruited. Second, all the collected data are self-reported, resulting in an increased possibility for participants to under- or over-report their behaviors in sensitive questions. Third, the current study used data gathered only from six high schools in Pyeongtaek, Korea. These limitations could decrease the generalizability of the study findings.

Our findings suggest that it may be beneficial to modify the interpersonal theory of suicide to make the model more applicable to other cultures, races/ethnicities, or populations. Because Joiner developed the interpersonal theory of suicide mainly based on US residents, his theoretical framework has particular characteristics geared toward US perspectives. Although there have been studies Crisis (2018), 39(1), 4–12

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

that examined the applicability of Joiner’s theory, most of the target population comprised Whites (Stewart, Eaddy, Horton, Hughes, & Kennard, 2015). In this regard, future studies are recommended to add other significant risk factors that are unique for the specific population group or to modify the interpersonal theory of suicide to develop a more precisely delineated model that is specialized for the population.

This study also suggests designing a multiple-factors-based suicide prevention program. Suicide prevention programs in Korea mainly focus on individual or group counseling and general suicide education (Nam, 2011).

Based on the finding of the current study, it is recommended to develop a suicide prevention program based on multiple factors and that monitors and deals with the risk factors included in the current study.

In Korea, the Suicide Prevention Law was enacted in April 2012 (Kim, Oh, & Lee, 2013). The basic policy direction is to focus on multidimensional suicide prevention guidelines that take into consideration the specific circumstances of people who are at risk for suicide (National Law Information Center, 2012). Although the Suicide Prevention Law has good intentions, the direction of the guidelines is not clear and specific. The Suicide Prevention Law, therefore, needs to be modified so that it provides a more specific direction based on the findings of the current study.

For example, it should include the risk factors thwarted belongingness, perceived burdensomeness, hopelessness, academic stress, and previous suicidal behaviors.

Some suggestions can be made for future studies. First, based on the previous research that indicated gender differences in suicidal ideation (Statistics Korea, 2012a, 2012b), it is strongly recommended to examine gender differences in risk factors for male and female adolescents.

Second, given the fact that there have been significant gaps in suicide rates between ordinary school and vocational school students (Statistics Korea, 2012a), it would be worthwhile to include students from both types of schools to compare the risk factors of suicidal ideation.

References Ahn, S. W. (1998). 학교폭력의 실태와 대책 [A study on school violence and measures]. 청소년행동연구 [Studies on Adolescent Behaviors], 3, 19–36.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.

Beck, A. T., Kovacs, M., & Weissman, A. (1979). Assessment of suicidal intention: The scale for suicide ideation. Journal of Consulting and Clinical Psychology, 47, 343–352. doi:10.1037/0022006X.47.2.343 © 2016 Hogrefe Publishing

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

Bonanno, R. A., & Hymel, S. (2010). Beyond hurt feelings: Investigating why some victims of bullying are at greater risk for suicidal ideation. Merrill-Palmer Quarterly, 56(3), 420–440. doi:10.1353/ mpq.0.0051 Brunstein Klomek, A., Sourander, A., & Gould, M. (2010). The association of suicide and bullying in childhood to young adulthood: A review of cross-sectional and longitudinal research findings.

Canadian Journal of Psychiatry. Revue Canadienne de Psychiatrie, 55(5), 282–288.

Conason, A. H., Oquendo, M. A., & Sher, L. (2005a). Psychotherapy in the treatment of alcohol and substance abusing adolescents with suicidal behavior. International Journal of Adolescent Medicine and Health, 18, 9–13.

Conason, A. H., Oquendo, M. A., & Sher, L. (2005b). Psychotherapy of alcohol adolescents with suicidal behavior. In J. Merrick & G. Zalsman (Eds.), Suicidal behavior in adolescents: An international perspective (pp. 325–330). London, UK: Freund Publishing House.

Chung, S. K. (2011). 음주시작연령 및 음주문제가 청소년의 자살 생각과 시도에 미치는 영향 [Influence of early onset of dinking and problem drinking on suicide ideation and attempt among Korean adolescents]. Journal of Korean Alcohol Science, 12(1), 15–27.

Fisher, H., Moffitt, T., Houts, R., Belsky, D., Arseneault, L., & Caspi, A. (2012). Bullying victimisation and risk of self harm in early adolescence: Longitudinal cohort study. British Medical Journal (Clinical Research Ed.), 344, e2683. doi:10.1136/bmj.e2683 Goldston, D., Daniel, S., Reboussin, B., Reboussin, D., Frazier, P., & Harris, A. (2001). Cognitive risk factors and suicide attempts among formerly hospitalized adolescents: A prospective naturalistic study. Journal of the American Academy of Child and Adolescent Psychiatry, 40(1), 91–99. doi:10.1097/00004583200101000-00021 Gould, M. S., Greenberg, T., Velting, D. M., & Shaffer, D. (2003). Youth suicide risk and preventive interventions: A review of the past 10 years. American Academy of Child and Adolescent Psychiatry, 42, 386–405. doi:10.1097/01.CHI.0000046821.95464.CF Hong, N. M. (2011). 부모학대와 또래괴롭힘이 청소년 자살생각 에 미치는 영향: 대인관계내재화와 절망감의 매개경로를 중심 으로 [The effects of parental abuse and peer victimization on adolescent’s suicidal ideation: The mediating pathway of interpersonal needs and hopelessness] (Master’s thesis). Retrieved from RISS. Jang, J. H. (2010). 청소년의 스트레스, 절망감, 사회적 지지와 자 살생각과의 관계 [Relationship between suicidal ideation and stress, despair, and social support in adolescence] (Master’s thesis). Retrieved from RISS. Jo, M. H. (2010). 대인관계 심리학적 모형을 통해서 본 한국군의 자 살 [Evaluation of Interpersonal Psychological Model of Suicide in Korean Army] (Master’s thesis). Retrieved from RISS. Johnston, L. D., Bachman, J. G., & O’Malley, P. M. (2009). Monitoring the future questionnaire responses from the nation’s high school seniors. Retrieved from http://www.monitoringthefuture.org/ Joiner, T. E. (2005). Why people die by suicide. Cambridge, MA: Harvard University Press.

Joiner, T. E., Brown, J. S. & Wingate, L. R. (2005). The psychology and neurobiology of suicidal behavior. Annual Review of Psychology, 56, 287–314. doi:10.1146/annurev.psych.56.091103.070320 Joiner, T. E., & Rudd, M. D. (2000). Intensity and duration of suicidal crisis vary as a function of previous suicide attempts and negative life events. Journal of Consulting and Clinical Psychology, 68(5), 909–916.

Kim, K. H., Oh, B. D., & Lee, K. S. (2013). 청소년 자살예방 정책 연 구 [A study of adolescents’ suicide prevention policy]. National Youth Policy Institute. Retrieved from http://dbpia.co.kr Kim, S., Shim, J., Noh, H., Hwang, H., & Park, H. (2013). The relation-

© 2016 Hogrefe Publishing

11

ship between smoking status and suicidal behavior in Korean adults: The 4th Korea National Health and Nutrition Examination Survey (2007-2009). Korean Journal of Family Medicine, 34(3), 178–189. doi:10.4082/kjfm.2013.34.3.178 Korea CDC. (2013). 제9차 (2013년) 청소년건강행태온라인조사 통계 [9th (2013) Adolescent health status online survey]. Retrieved from http://www.cdc.go.kr/ Korean Statistical Information Service. (2010). 15-19살 사망자 가 운데 자살 비율: 2000-2010년 [The suicide rates among the people died between 15-19 years old: 2000–2010]. Retrieved from http://kosis.kr Lee, C. (2007). An ecological prediction model of bullying behaviors among South Korean middle school students (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses.

Lee, M., & Larson, R. (2000). The Korean “examination hell”: Long hours of studying, distress, and depression. Journal of Youth and Adolescence, 29(2), 249–271. doi:10.1023/A:1005160717081 Lee, S. W., & Jang, Y. E. (2011). 학업스트레스가 청소년의 자살생각 에 미치는 영향: 가족응집성의 조절효과 [A study on the effect of adolescents’ academic stress to suicidal ideation: Moderating effect of family cohesion]. 청소년학연구 [Research on Youth Studies], 18(11), 111–136.

Nam, M. S. (2011). 청소년 자살예방 교육 프로그램이 생명존중의식, 자살태도, 자살위험성에 미치는 효과 [The effects of suicide preventive program for adolescent on their life-esteem, attitude toward suicide and suicidal risk] (Master’s thesis). Retrieved from RISS. National Law Information Center. (2012). 자살예방 및 생명존중문 화 조성을 위한 법률 [The Suicide Prevention Law]. Retrieved from http://law.go.kr Nunnally, J. C. (1978). Psychometric theory. New York, NY: McGraw Hill.

Park, S. H. (2006). 아동의 학교학업 스트레스 및 과외학업 스트레 스와 무력감간의 관계 [A study on the relationship between academic stress and children’s helplessness] (Master’s thesis).

Retrieved from RISS. Reid, W. H. (2009). Prognosis after suicide attempt: Standard of care and the consequences of not meeting it. Journal of Psychiatric Practice, 15(2), 141–144. doi:10.1097/01. pra.0000348367.42042.06 Reinecke, M. A., DuBois, D. L., & Schultz, T. M. (2001). Social problem solving, mood, and suicidality among inpatient adolescents. Cognitive Therapy & Research, 25(6), 743–756. doi:10.1023/A:1012971423547 Roeger, L., Allison, S., Korossy-Horwood, R., Eckert, K., & Goldney, R. (2010). Is a history of school bullying victimization associated with adult suicidal ideation?: A South Australian population-based observational study. The Journal of Nervous and Mental Disease, 198(10), 728–733. doi:10.1097/NMD.0b013e3181f4aece Schumacker, R. E., & Lomax, R. G. (2004). A beginner’s guide to structural equation modeling. Mahwah, NJ: Lawrence Erlbaum Associates.

Shin, M. S., Park, K. B., Oh, K. J., & Kim, J. S. (1990). 고등학생의 자살 성향에 관한 연구: 우울, 절망, 자살간의 구조적 관계에 대한 분석 [The structural relation among depression, hopelessness, and suicidal ideation: A study of suicidal ideation among high school students]. The Korean Journal of Clinical Psychology, 9(1), 1–19.

Statistics Korea. (2010). 청소년 (15-24세) 사망원인 [The leading causes of death among adolescents (15–24 years old)]. Retrieved from http://kostat.go.kr/ Statistics Korea. (2011a). 사망원인통계 [The leading causes of death]. Retrieved from http://kostat.go.kr/ Statistics Korea. (2011b). Youth statistics; 2011. Retrieved from http://kostat.go.kr Statistics Korea (2012a). 청소년 건강행태 온라인조사 [Adolescents health status online survey]. Retrieved from http://kostat.go.kr/ Crisis (2018), 39(1), 4–12

12

Statistics Korea (2012b). 아동 청소년 인권 실태조사 [Children and adolescents human rights survey]. Retrieved from http://kostat. go.kr/ Stewart, S. M., Eaddy, M., Horton, S., Hughes, J., & Kennard, B. (2015). The validity of the interpersonal theory of suicide in adolescence: A review. Journal of Clinical Child and Adolescent Psychology, 53, 1–13. doi:10.1080/15374416.2015.1020542 Teasdale, J. D. (1988). Cognitive vulnerability to persistent depression. Cognition and Emotion, 2(3), 247–274. doi:10.1080/ 02699938808410927 US Department of Health and Human Services. (2002). The national household survey on drug use abuse report: Substance use and the risk of suicide among youths. Retrieved from http:// www.oas.samhsa.gov/2k2/suicide/suicide.pdf Van Orden, K. A., Cukrowicz, K., Witte, T., & Joiner, T. (2012). Thwarted belongingness and perceived burdensomeness: Construct validity and psychometric properties of the Interpersonal Needs Questionnaire. Psychological Assessment, 24(1), 197– 215. doi:10.1037/a0025358 Van Orden, K. A., Witte, T. K., Cukrowicz, K. C., Braithwaite, S. R., Selby, E. A., & Joiner, T. E. (2010). The interpersonal theory of suicide.

Psychological Review, 117(2), 575–600. doi:10.1037/a0018697 Yoo, A. J., & Min, H. Y. (1998). 학령기 아동의 일상적 생활 스트레스 척도 개발 [Development of a Daily Hassles Scale for school age children]. Korean Journal of Child Studies, 19(2), 77–96.

Yoon, S. H. (2010). 자아존중감이 청소년 자살생각에 미치는 영향 [Effects of self-efficacy on suicidal ideation among adolescents] (Master’s thesis). Retrieved from RISS.

Crisis (2018), 39(1), 4–12

Y. J. Kim et al., Suicidal Ideation Among Korean Adolescents

Received April 1, 2015 Revision received July 6, 2016 Accepted August 4, 2016 Published online November 21, 2016 About the authors Dr. Yi Jin Kim is an assistant professor in the Department of Social Work University of Mississippi, MS. Dr. Kim received his PhD degree in social work from the University of Texas at Arlington in 2014. His research areas include suicide and substance use.

Dr. Sung Seek Moon is an associate professor at the College of Social Work, University of South Carolina, SC. Dr. Moon received his degree in social work from the University of Georgia in 2002. His major research interest is in immigrant children and youth’s risky health behaviors.

Dr. Jang Hyun Lee is a professor in the Department of Child & Youth Welfare of Pyeongtaek University, South Korea.

Dr. Joon Kyung Kim is a professor at the School of Child Welfare of Namseoul University, South Korea.

Yi Jin Kim Assistant Professor Department of Social Work School of Applied Sciences The University of Mississippi 5197 W E Ross Pkwy Southaven, MS 38671 USA yjkim@olemiss.edu

© 2016 Hogrefe Publishing

Research Trends

Telephone Crisis Support Workers’ Psychological Distress and Impairment A Systematic Review Taneile A. Kitchingman1, 2, Coralie J. Wilson2, 3, Peter Caputi1, Ian Wilson3, and Alan Woodward4, 5 School of Psychology, University of Wollongong, Australia Illawarra Health and Medical Research Institute, University of Wollongong, Australia 3 School of Medicine, University of Wollongong, Australia 4 Lifeline Research Foundation, Lifeline Australia, Canberra, Australia 5 Centre for Mental Health, Melbourne School of Population and Global Health, University of Melbourne, Australia 1 2

Abstract. Background: In order to respond to crises with appropriate intervention, crisis workers are required to manage their own needs as well as the needs of those they respond to. Aims: A systematic review of the literature was conducted to examine whether telephone crisis support workers experience elevated symptoms of psychological distress and are impaired by elevated symptoms. Method: Studies were identified in April 2015 by searching three databases, conducting a gray literature search, and forward and backward citation chaining. Results: Of 113 identified studies, seven were included in the review. Results suggest that that telephone crisis support workers experience symptoms of vicarious traumatization, stress, burnout, and psychiatric disorders, and that they may not respond optimally to callers when experiencing elevated symptoms of distress. However, definitive conclusions cannot be drawn due to the paucity and methodological limitations of available data.

Limitations: While the most comprehensive search strategy possible was adopted, resource constraints meant that conference abstracts were not searched and authors were not contacted for additional unpublished information. Conclusion: There is an urgent need to identify the impact of telephone crisis support workers’ role on their well-being, the determinants of worker well-being in the telephone crisis support context, and the extent to which well-being impacts their performance and caller outcomes. This will help inform strategies to optimize telephone crisis support workers’ well-being and their delivery of support to callers.

Keywords: hotline, helpline, crisis intervention, distress, impairment

Crisis intervention strategies that include readily accessible support without referral requirement are a fundamental suicide prevention strategy. Within Australia, nonclinical telephone and online crisis support services provide essential suicide prevention and intervention for individuals in crisis.

Service operational data show that up to one third of callers to crisis lines and half of all visitors to crisis chat services may be suicidal at the time of contact (Lifeline Australia Research Foundation, 2013). Research indicates that contact with technology-based crisis services significantly reduces suicidality, improves mental state, and facilitates engagement with necessary intervention (Gould, Kalafat, Munfakh, & Kleinman, 2007; Kalafat, Gould, Munfakh, & Kleinman, 2007; King, Nurcombe, Bickman, Hides, & Reid, 2003).

The National Coalition for Suicide Prevention (2014) has identified that our ability to ensure that communities © 2017 Hogrefe Publishing

have the capacity to respond to crises with appropriate interventions requires serious discussion and action. While a number of established services provide free support to individuals in crisis, including those in geographically isolated areas, and are beginning to be integrated with national treatment initiatives (e.g., ATAPS Suicide Prevention Program), a number of recommendations need to be enacted to ensure that individuals in a crisis situation have access to emergency care, including through telephone helplines and the Internet. Among these is the recommendation to train crisis support workers to manage their own needs as well as the needs of those they respond to.

Research identifies a number of occupational hazards associated with helping roles (e.g., medical practitioner, psychologist). The greatest hazard is the impact of the distress of others on the helper. As a consequence of conCrisis (2018), 39(1), 13–26 https://doi.org/10.1027/0227-5910/a000454

14

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

stantly providing care to others in distress, helpers self-report a variety of problems including elevated symptoms of anxiety, depression, and suicidal ideation (Gilroy, Carroll, & Murra, 2002; Pope & Tabachnick, 1994). Helpers also report symptoms of burnout (Rupert & Kent, 2007; Rupert & Morgan, 2005), vicarious traumatization (McCann & Pearlman, 1990; Pearlman & MacIan, 1995), and compassion fatigue (Shapiro, Brown, & Biegel, 2007).

A significant proportion of helpers report impaired functioning associated with elevated symptoms of psychological distress. Impaired personal functioning is commonly clustered around emotional exhaustion, fatigue, irritability, problems with interpersonal relationships, and sleep disturbance (Mahoney, 1997; Thoreson, Miller, & Krauskopf, 1989). Large-scale surveys also suggest helpers who experience elevated personal symptoms of psychological distress are unable to communicate or use their key skills effectively (Sherman & Thelen, 1998; West & Shanafelt, 2007), and are likely to deliver suboptimal or even inadequate care (American Psychological Association, 2010; Guy, Poelstra, & Stark, 1989; Pope, Tabachnick, & Keith-Spiegel, 1987). By doing so, these helpers forego their ethical responsibilities to practice within the limits of their competence and prevent harm that might occur as a result of their impaired conduct (Wise, Hersh, & Gibson, 2012).

To date, research investigating helper distress and impairment has focused on professionals providing face-toface support to those in distress. As a result, helpers who may be at risk of experiencing elevated symptoms of distress and impairment but do not fit this description have been overlooked.

Helpers who deliver support by telephone, particularly within the crisis sector, are frequently exposed to others in distress. Telephone crisis support workers routinely speak with callers who are experiencing severe mental health issues or are at imminent risk of suicide (Gould et al., 2007; Kalafat et al., 2007; Mishara et al., 2007), and hear graphic descriptions of horrific life events and the consequences (Kinzel & Nanson, 2000). Their role is to listen and support the caller to manage his or her own crisis, providing information and referrals to help-services where appropriate (Burgess, Christensen, Leach, Farrer, & Griffiths, 2008; Hall & Schlosar, 1995). Emphasis is placed on listening empathically and on containing the caller’s distress (Kinzel & Nanson, 2000).

A number of unique stressors are present for helpers working within the telephone crisis support context. These stressors include the absence of nonverbal communication cues, the inability to anticipate or control the types of calls received, and the inability to track changes in the caller due to the one-shot nature of crisis line contact (Coman, Burrows, & Evans, 2001). Turnover and resignation Crisis (2018), 39(1), 13–26

rates are high, increasing the workload of remaining staff, the majority of whom are not paid for their time (Kinzel & Nanson, 2000). Telephone crisis support workers also generally have less education, and are less likely to have access to supervision and ongoing professional development than helpers in a registered professional role (Baird & Jenkins, 2003). These stressors may increase telephone crisis support workers’ vulnerability to occupational hazards and impaired performance.

In order to optimize telephone crisis support workers’ well-being and service provision, service providers need to know (a) whether workers experience elevated symptoms of psychological distress, (b) whether elevated symptoms impair workers’ personal functioning and service delivery, and( c) the mechanisms by which workers become distressed and impaired. The current study is a systematic review of the literature to uncover what is currently known about telephone crisis support workers’ distress and impairment. The review sought to answer the following questions: 1. Do telephone crisis support workers experience elevated symptoms of psychological distress?

2. Do telephone crisis support workers experience functional impairment associated with elevated symptoms of psychological distress?

Method The review was planned, conducted, and is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement (Moher, Liberati, Tetzlaff, & Altman, 2009; see Figure 1 and Appendix A).

Databases and Search Methodology The databases CINAHL (1937), Medline (1946), PsycINFO (1806), and Scopus (1995) were systematically searched from inception through to March 2016. Search terms were generated by examining terminology used in the literature (see Appendix B). Wildcard and truncation symbols were used to increase chances of selecting studies that reported symptoms of psychological distress or impairment in telephone crisis support workers. Google and Google Scholar were also searched, forward and backward citation chaining was conducted from each included study, and experts were consulted in order to identify additional potentially relevant studies. Only full-text, Englishlanguage publications that had been subjected to the peerreview process were sought. © 2017 Hogrefe Publishing

15

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

Study Selection

Data Extraction

Figure 1 presents a flow chart for the selection of included studies. The initial database search returned 207 studies.

Five additional studies were identified through gray literature search, forward and backward citation chaining. After the removal of duplicates, 191 records were screened, of which 97 were excluded. The full text of 94 potentially relevant articles was screened. To be included in the review, studies were required to: (a) use a measure of psychological distress and/or impairment; (b) examine telephone crisis support workers (i.e., not crisis callers, not telephone line workers who do not deliver crisis support [e.g., workers on lines for smoking cessation/monitoring health-care treatment compliance], or crisis workers who do not deliver support by telephone [e.g., mobile crisis teams]). In all, 87 studies did not meet the selection criteria and were excluded. Seven studies met the selection criteria and were included in the review. Eligibility assessment was performed independently in an unblinded standardized manner by two reviewers. Disagreements between reviewers were resolved by consensus.

The seven studies included in the review were coded using a pre-formulated rating sheet according to the following characteristics: −− Author(s) and date of publication; −− Location of study; −− Sample characteristics (N, age, % female, population, exclusion criteria); −− Study design (data collection method, participant selection method); −− Outcome variable; −− Study variables; and −− Findings.

Data were examined on participants’ symptoms of psychological distress, impairment related to symptoms of psychological distress, and associated variables.

Quality Assessment The methodological quality of each included study was determined using a quality rating list of 15 criteria (Van Uffelen et al., 2010). Criteria had a yes (1 point), no (0

Figure 1. PRISMA flow diagramm.

© 2017 Hogrefe Publishing

Crisis (2018), 39(1), 13–26

16

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

points), or unclear (0 points) answer format. All criteria had the same weight, and a quality score ranging from 0 to 15 points was calculated for each study (see Appendix C). The number and proportion of studies scoring a point for each quality criterion are reported in Appendix D. The median quality score for the included studies was 10 (25th–75th percentiles = 9–13) points of 15. Hypotheses, variables, data sources, and data collection methods were reported for all studies. Most studies scored a point for presenting the study design (n = 4), reporting the validity of study measures (n = 4), reporting the target population (n = 6), describing participant recruitment (n = 6) and selection processes (n = 4), describing the study sample (n =5), and reporting the number of participants who provided data at each stage of the study (n = 4). However, very few studies reported participant inclusion and/or exclusion criteria (n = 2) or described the reliability of measures used in the study (n = 3). No studies used random sampling. Appendix D provides the quality assessment of each study. Criteria for quality assessment and the number and proportion of included studies scoring a point for each criterion are reported in Appendix C.

Results

80 and 131 participants. The remaining studies (n = 2) had 54 and 62 participants, respectively. One study used random selection. The remaining studies (n = 5) used self-selection sampling. Data reported on the age of participants varied. Five studies provided an age range in years and the mean age of participants. The remaining two studies did not report any data on participants’ ages. Six studies included both males and females. The percentage of female participants in these studies ranged from 49.0 to 88.7%.

One study did not report any data on participants’ sex.

Population Studies included telephone crisis support workers from suicide prevention services, sexual assault and trauma-related agencies, and youth and parent support lines. One study recruited volunteers from a helpline run by university students and a control group of students from the same university who had never worked on a crisis line. A further study recruited undergraduate psychology students, suicide hotline volunteers, and graduate students in clinical and counseling psychology programs, which served as nonprofessional, paraprofessional, and professional groups, respectively. The authors of this study did not report distress or impairment data for participants in these separate groups.

The characteristics of each included study are reported in Table 1. The following is an overview of the study characteristics and findings.

Exclusion Criteria One study excluded telephone crisis support workers who had completed fewer than six shifts. The remaining six studies did not report exclusion criteria.

Study Characteristics

Findings

Year Published Included studies were published between 1973 and 2013.

Most studies were conducted between 2000 and 2013 (n = 4). Two studies were conducted in the 1990s and one in the 1970s.

Do Telephone Crisis Support Workers Experience ­Elevated Symptoms of Psychological Distress?

Six studies examined telephone crisis support workers’ symptoms of psychological distress. The most frequently examined type of psychological distress was vicarious trauma, explored in two studies. Both these studies reported sample mean disruptions in beliefs scores in the average range, and low sample mean hyperarousal, avoidance and intrusion scores. Dunkley and Whelan (2006) reported that participants with a higher trauma caseload reported significantly higher symptoms of hyperarousal, avoidance, and intrusion symptoms than did participants with a lower trauma caseload. Furlonger and Taylor (2013) found that participants with a personal trauma history reported significantly higher hyperarousal, avoidance, and intrusion symptoms. Those who engaged in avoidance as a coping strategy reported significantly higher disruptions in beliefs than those who used problem-solving or social support to cope. Participants who reported a weaker working alliance

Location Studies were conducted in the United States (n = 3), Australia (n = 2), Canada (n = 1), and the United Kingdom (n = 1).

Methodologies Employed All seven included studies were conducted using quantitative methods. Most studies (n = 6) used a self-report survey to collect data. Diagnostic interviews were conducted in one study.

Sample Characteristics The number of participants in included studies ranged from n = 54 to n = 131. Most studies (n = 5) had between Crisis (2018), 39(1), 13–26

© 2017 Hogrefe Publishing

Location

US

Australia

Australia

Author(s) and date

Cyr & Dowrick (1991)

© 2017 Hogrefe Publishing

Dunkley & Whelan (2006)

Furlonger & Taylor (2013)

54 M = 36.7 years, Range = 23.8–60.8 years 65.8% female Kids Help Line (n = 38) and Parentline (n = 16) telephone/online counsellors

62 M = 45.5 years, Range = 24.7–75.2 years 88.7% female Telephone crisis line workers from nine trauma-related services (no sig. diff. in TABS or IES-R scores, subsequently treated as single group) Excluded if worked fewer than six shifts

Survey Self-selecting

Survey Self-selecting

Survey random

Data collection method Participant selection method

N Age % female Population Exclusion criteria 130 Age not reported 69% female Crisis line volunteers at a sexual assault agency (n = 62), and crisis line volunteers at a suicide prevention agency (n = 68)

Study design

Sample characteristics

Table 1. Characteristics of included studies, in alphabetical order by author name (n = 7)

Disruptions in beliefs as a result of directly and indirectly experienced trauma (TABS; Pearlman, 2003) Hyperarousal, avoidance and intrusion (IES-R; Weiss & Marmar, 1997) Coping styles (Coping Strategy Indicator; Amirkhan, 1990)

Disruptions in beliefs as a result of directly and indirectly experienced trauma (Trauma Attachment and Belief Scale [TABS]; Pearlman, 2003 Hyperarousal, avoidance and intrusion (Impact of Event Scale-Revised [IES-R]; Weiss & Marmar, 1997) Coping styles (Coping Scale for Adults; Frydenberg & Lewis, 1997)

Burnout (novel measure, including: self-report direct question about burnout, “Have you ever felt burned out from working on the crisis line?”; a checklist of burnout stages and satisfaction factors)

Variable (measure)

Outcome variable (distress/ impairment)

Demographics (categorical items to assess age, gender, years of counselling experience, trauma caseload numbers, weekly contact with trauma clients) Perceived working alliance (Supervisory Working Alliance Inventory [SWAI]; Efstation et al., 1990)

Sample mean TABS score in average range.

Sample mean IES-R score in low range.

Participation in supervision associated with significantly higher than average use of problem-solving and seeking social support, and below average use of avoidance.

Sample mean disruptions in beliefs in average range.

Sample mean hyperarousal, avoidance and intrusion in low range.

Dealing with the problem most frequently used coping style, nonproductive coping least commonly used.

Nonproductive coping associated with significantly higher TABS scores.

Personal trauma history associated with significantly higher IES-R scores.

Stronger working alliance with supervisor associated with significantly lower TABS scores.

13

13

10

54% Participants identified as having felt burned out on the crisis line Of these 54% participants, 75% indicated that they noticed burnout within 1 year of beginning work on the crisis line, and 95% indicated that they continued to work on the crisis line although they felt burned out 97% of all participants had experienced at least one stage of burnout

Demographics (categorical items to assess age, number of volunteer hours per month, stress and support information)

Perceived working alliance (Supervisory Working Alliance Inventory; Efstation et al., 1990

Total points out of 15

Quality score

Results relevant to symptoms of psychological distress, ­impaired responses to callers due to ­symptoms of psychological ­distress, mechanisms of distress or impairment

Results

Variable (measure)

Study variables

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment 17

Crisis (2018), 39(1), 13–26

Location

US

Canada

Author(s) and date

McClure et al. (1973)

Mishara & Giroux (1993)

Table 1. continued

Crisis (2018), 39(1), 13–26 80 M = 28.6 years, Range = 19–64 years 49% female Active telephone crisis line volunteers from Suicide-Action Montreal

Survey Self-selecting

Perceived stress (visual analogue scale) Coping strategies (Ways of Coping Checklist; Lazarus & Folkman, 1984)

Psychiatric diagnosis by interviewer and two psychiatrists (diagnostic criteria unspecified)

Interviews conducted by researcher and two psychiatrists Self-selecting

125 Age not reported % female not reported Crisis line volunteers from Call for Help (CFH) suicide prevention line (n = 74), Youth Life Line (YLL) teen support hotline (n = 51)

Variable (measure)

Data collection method Participant selection method

N Age % female Population Exclusion criteria

Outcome variable (distress/ impairment)

Study design

Sample characteristics Results relevant to symptoms of psychological distress, ­impaired responses to callers due to ­symptoms of psychological ­distress, mechanisms of distress or impairment 22% of CFH volunteers versus 4% of YLL volunteers were currently ill (p < .05) Lifetime incidence of emotional problems was 54% in CFH volunteers versus 32% in YLL volunteers (p < .01) 35% of CFH volunteers reported history of suicidal thoughts versus 12% of YLL volunteers (p < .05) Sample mean light anticipatory stress before commencing shift, moderate stress during shift, and light to moderate retrospective stress after shift Sig. greater stress before shift associated with less experience Sig. greater stress during shift associated with taking more urgent calls, longer total length of calls, greater use of coping strategies magical thinking, detachment, and feeling personally responsible Sig. greater stress after shift associated with greater total length of calls during shift, more unreasonable expectations about interventions, greater use of coping strategy magical thinking and less use of coping strategy positive thinking No sig. relationship between personal experience with suicide and stress levels before, during, or after the shift

Demographics (categorical items to assess age, sex, marital status, occupation, reported history of psychiatric illness)

Demographics (categorical items to assess age, sex, education, occupation, prior experience in helping relationships, experience at Suicide-Action Montreal) Beliefs about interventions (open-ended questions which were content-analyzed)

Results

Variable (measure)

Study variables

9

6

Total points out of 15

Quality score

18 T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

© 2017 Hogrefe Publishing

Location

US

United Kingdom

Author(s) and date

Neimeyer et al. (2001)

Paterson et al. (2009)

Table 1. continued

© 2017 Hogrefe Publishing 106 M = 19.9 years, Range = 18–24 years Volunteers from a helpline at the University of Manchester (n = 54, 70% female) Control group – students from the same university who had never worked on a hotline (n = 52, 44% female) Survey Self-selecting

Mental health experiences (screening questions from the Mini International Neuropsychiatric Interview to assess lifetime prevalence of symptoms; Sheehan et al., 1998) “Have you ever been diagnosed with or treated for mental health difficulties?”

(Yes/No)

Suicidal distress (Suicidal Behaviours Questionnaire [SBQ]; Cotton, Peters, & Range, 1995) Competence in choosing appropriate therapeutic responses to suicidal individuals (Suicidal Intervention Response Inventory [SIRI]; Neimeyer & MacInnes, 1981)

Survey Self-selecting

131 M = 32.5 years, Range = 19–76 years 67.2% female Suicide hotline volunteers from the Memphis Suicide and Crisis Intervention Service, undergraduate psychology students at the University of Memphis, and graduate psychology students at the same university

Variable (measure)

Data collection method Participant selection method

N Age % female Population Exclusion criteria

Outcome variable (distress/ impairment)

Study design

Sample characteristics

No significant differences between groups on mental health symptoms or history of diagnosis of/treatment for mental health difficulties

Previous experience with suicide associated with significantly lower SIRI scores Higher SBQ scores associated with significantly lower SIRI scores

Demographics (categorical items to assess gender, age, marital status, racial identification, level of education, counseling experience and training, personal and professional experience with suicide, history of personal psychological counselling received) Death attitudes (Death Attitude Profile – Revised; Wong et al., 1994) Attitudes toward suicide (Suicide Opinion Questionnaire; Domino et al., 1982) Demographics (categorical items to assess gender, age, experience working for a helpline) Empathy (Interpersonal Reactivity Index; Davis, 1983) Personality traits (Big Five Inventory; Goldberg, 1993)

Results relevant to symptoms of psychological distress, ­impaired responses to callers due to ­symptoms of psychological ­distress, mechanisms of distress or impairment

Results

Variable (measure)

Study variables

10

9

Total points out of 15

Quality score

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment 19

Crisis (2018), 39(1), 13–26

20

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

with their supervisor also experienced significantly higher disruptions in beliefs than those who reported a stronger working alliance.

Mishara and Giroux (1993) examined telephone crisis support workers’ perceived stress before, during, and after a shift on the crisis line. On average, participants reported light perceived stress directly before commencing a telephone crisis support shift, moderate stress during the shift, and light to moderate stress in the week following a shift.

Less experienced telephone crisis support workers reported significantly greater perceived stress before the shift than those with more experience. More urgent calls and longer amounts of time on calls were associated with significantly greater perceived stress during the shift. Longer amounts of time on calls during the shift and unreasonable expectations about interventions with callers were associated with significantly greater stress during the week following the shift. While a significant proportion of participants reported personal experience with suicide (lifetime thoughts of suicide, suicide attempts, suicidal actions by others), this was not significantly related to perceived stress before, during, or after the shift.

Cyr and Dowrick (1991) examined telephone crisis support workers’ symptoms of burnout. More than half (54%) of their sample reported experiencing burnout as a result of working on the crisis line. Three quarters (75%) of these participants reported developing symptoms of burnout within their first year working on the crisis line, and 95% reported that they continued to work on the crisis line although they felt burned out.

McClure, Wetzel, Flanagan, McCabe, and Murphy (1973) compared telephone crisis support workers with workers from a non-crisis-focused teen support hotline on the basis of current and lifetime psychological disorders.

The majority (78%) of telephone crisis support workers did not currently meet criteria for any disorder. However, workers from the telephone crisis support line were significantly more likely to meet criteria for a lifetime disorder than workers from the non-crisis-focused teen support hotline. By contrast, Paterson, Reniers, and Vollm (2009) compared telephone crisis support workers with a general population control group, finding no significant difference between groups in current symptoms or in previous diagnosis of or treatment for mental health disorders.

Are Telephone Crisis Support Workers Impaired by Elevated Symptoms of Psychological Distress in Their Responses to Callers?

One study examined telephone crisis support workers’ impaired service delivery related to symptoms of psychological distress. Neimeyer, Fortner, and Melby (2001) reported that higher suicidal distress was associated with significantly lower competency in choosing appropriate Crisis (2018), 39(1), 13–26

therapeutic responses to suicidal individuals. This result suggested that workers who have experienced or are currently dealing with their own suicidal distress may not respond optimally to others who are suicidal. However, participants in this study included undergraduate and postgraduate psychology students as well as suicide hotline volunteers. No data were reported for suicide hotline volunteers alone.

Discussion Interpreting the Findings This systematic review sought to identify and describe what is currently known about telephone crisis support workers’ symptoms of psychological distress and impairment. A comprehensive search strategy was employed and, as no restrictions were placed on study design, we were able to include and describe a variety of studies that have been undertaken in this area.

The main finding of this systematic review is that conclusions about telephone crisis support workers’ symptoms of psychological distress and impairment are unable to be drawn due to the paucity, disparity, and methodological limitations of current data. Only seven studies met the selection criteria and were included in the review. Of these studies, none was a randomized controlled trial, only one included a control group, and participants included telephone crisis support workers from a variety of fields.

Six included studies examined telephone crisis support workers’ symptoms of psychological distress. These studies suggest that telephone crisis support workers may experience stress, symptoms of vicarious trauma, burnout, and psychological disorders, and that workers who engage in maladaptive coping strategies and have less experience and greater exposure to distressed callers are at increased risk. However, studies reported contradictory findings, none recruited a random sample, and the majority did not report participant inclusion criteria or the reliability of study measures.

One study examined telephone crisis support workers’ impairment related to elevated personal symptoms of psychological distress. Results indicated that participants who had experienced or were experiencing suicidal distress may not respond optimally to other suicidal individuals.

However, specific conclusions about telephone crisis support workers’ impairment were precluded by the treatment of participants, including undergraduate and postgraduate psychology students, as a single group.

© 2017 Hogrefe Publishing

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

Limitations Several limitations of the current review must be acknowledged. First, while the most comprehensive search strategy possible was adopted, resource constraints meant that conference abstracts were not searched and authors were not contacted for additional unpublished information. Consequently, some relevant papers may have been missed. Second, publication biases may mean that negative information about telephone crisis support workers’ symptoms of and responses to distress were more likely to have been published than neutral or positive information.

Future research is needed to investigate positive impacts of empathetic engagement with crisis callers on the telephone crisis support workers’ psychological well-being.

Conclusion and Future Directions This review found no published studies that have specifically examined whether telephone crisis support workers experience impaired functioning or impaired support skills related to known occupational hazards for helpers (e.g., elevated symptoms of general psychological distress, depression, anxiety, or compassion fatigue). Therefore, conclusions about telephone crisis support workers’ symptoms of psychological distress and impairment are unable to be drawn due to the paucity, disparity, and methodological limitations of current data. An urgent need for ongoing service development is to identify (a) the impact of the telephone crisis support workers’ role on their well-being, (b) determinants of worker well-being in the telephone crisis support context, and (c) the extent to which the well-being of telephone crisis support workers impacts their performance and caller outcomes. Future research should be driven by an integrated model of staff education and service support that examines relationships between: (a) recruitment, (b) training, (c) skill assessment, (d) personal development and individual processes to maintain well-being (e.g., self-care), (e) supervisor training and staff support strategies (e.g., learning, teaching, and facilitating a reflective practice model for supervision and staff professional development), and (f) service support strategies (e.g., organization of personal and professional support strategies that complement staff supervision; Kitchingman, Wilson, Woodward, Caputi, & Wilson, 2015).

Ideally, future studies should collect longitudinal data to examine changes in telephone crisis support workers’ symptoms of psychological distress and any consequent impairment over time. In the meantime, exploratory, cross-sectional studies are warranted to identify factors involved in the development of psychological distress and © 2017 Hogrefe Publishing

21

impairment. The use of random sampling and standardized measures would increase the validity and reliability of study findings.

References American Psychological Association. (2010). Survey findings emphasize the importance of self-care for psychologists. Retrieved from http://www.apapracticecentral.org/update/2010/08–31/ survey.aspx Amirkhan, J. H. (1990). A factor analytically derived measure of coping: The Coping Strategy Indicator. Personality Processes and Individual Differences, 59, 1066–1074.

Baird, S., & Jenkins, S. R. (2003). Vicarious traumatization, secondary traumatic stress, and burnout in sexual assault and domestic violence agency staff. Violence and Victims, 18, 71–86. doi:10.1891/vivi.2003.18.1.71 Burgess, N., Christensen, H., Leach, L. S., Farrer, L., & Griffiths, K. (2008). Mental health profile of callers to a telephone counselling service. Journal of Telemedicine and Telecare, 14, 42–47. doi:10.1258/jtt.2007.070610 Coman, G. J., Burrows, G. D., & Evans, B. J. (2001). Telephone counselling in Australia: Application and considerations for use. British Journal of Guidance and Counselling, 29, 247–258. doi:10.1080/03069880124904 Cotton, C. R., Peters, D. K., & Range, L. (1995). Psychometric properties of the suicidal behaviors questionnaire. Death Studies, 19, 391–397.

Cyr, C., & Dowrick, P. W. (1991). Burnout in crisisline volunteers. Administration and Policy in Mental Health, 18, 343–354.

Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44, 113–126.

Domino, G., Moore, D., Westlake, L., & Gibson, L. (1982). Attitudes toward suicide: A factor analytic approach. Journal of Clinical Psychology, 38, 257–262.

Dunkley, J., & Whelan, T. A. (2006). Vicarious traumatization in telephone counsellors: internal and external influences.

British Journal of Guidance and Counselling, 34, 451–469. doi:10.1080/03069880600942574 Efstation, J. F., Patton, M. J., & Kardash, C. M. (1990). Measuring the working alliance in counselor supervision. Journal of Counseling Psychology, 37, 322–329.

Frydenberg, E., & Lewis, R. (1997). Coping Scale for Adults: Administrator’s manual. Melbourne, Australia: ACER. Furlonger, B., & Taylor, W. (2013). Supervision and the management of vicarious traumatisation among Australian telephone and online counsellors. Australian Journal of Guidance and Counselling, 23, 82–94. doi:10.1017/jgc.2013.3 Gilroy, P. J., Carroll, L., & Murra, J. (2002). A preliminary survey of counseling psychologists’ personal experiences with depression and treatment. Professional Psychology Research and Practice, 33, 402–407. doi:10.1037/0735-7028.33.4.402 Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 26–34.

Gould, M. S., Kalafat, J., Munfahk, J. L. H., & Kleinman, M. (2007).

An evaluation of crisis hotline outcomes part 2: Suicidal callers. Suicide and Life-Threatening Behaviour, 37, 338–352. doi:10.1521/suli.2007.37.3.338 Guy, J. D., Poelstra, P. L., & Stark, M. J. (1989). Personal distress and therapeutic effectiveness: National survey of psychologists practicing psychotherapy. Professional Psychology Research and Practice, 20, 48–50. doi:10.1037/0735-7028.20.1.48 Crisis (2018), 39(1), 13–26

22

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

Hall, B., & Schlosar, H. (1995). Repeat callers and the Samaritan telephone crisis line: A Canadian experience. Crisis, 16, 66–71. doi:10.1027/0227-5910.16.2.66 Kalafat, J., Gould, M. S., Munfakh, J. L. H., & Kleinman, M. (2007). An evaluation of crisis hotline outcomes – part 1: Nonsuicidal crisis callers. Suicide and Life-Threatening Behavior, 37, 322–337. doi:10.1521/suli.2007.37.3.322 King, R., Nurcombe, B., Bickman, L., Hides, L., & Reid, W. (2003).

Telephone counselling for adolescent suicide prevention: Changes in suicidality and mental state from beginning to end of a counselling session. Suicide and Life-Threatening Behavior, 33(4), 400–411. doi:10.1521/suli.33.4.400.25235 Kinzel, A., & Nanson, J. (2000). Education and debriefing: strategies for preventing crises in crisis-line volunteers. Crisis, 21, 126–134. doi:10.1027//0227-5910.21.3.126 Kitchingman, T. A., Wilson, C. J., Woodward, A., Caputi, P., & Wilson, I. (2015). Preventing suicide requires more attention on technology-based crisis support services. Australian New Zealand Journal of Psychiatry, 50(2), 181–186. doi:10.1177/0004867415605643 Lazarus, R. S., & Folkman, S. (1984). Stress appraisal and coping.

New York, NY: Springer.

Lifeline Australia Research Foundation. (2013). Summary of research and evaluation of crisis helplines. Retrieved from https:// www.lifeline.org.au/about-lifeline/publications/re​search-andreports Mahoney, M. J. (1997). Psychotherapists' personal problems and self-care patterns. Professional Psychology: Research and Practice, 28(1), 14–16.

McCann, L., & Pearlman, L. A. (1990). Vicarious traumatization: A frame-work for understanding the psychological effects of working with victims. Journal of Traumatic Stress, 3, 131–149. doi:10.1007/bf00975140 McClure, J. N., Wetzel, R. D., Flanagan, T. A., McCabe, M., & Murphy, G. E. (1973). Volunteers in a suicide prevention service. Journal of Community Psychology, 1, 397–398. doi:10.1002/15206629(197310)1:4<397::AID-JCOP2290010413>3.0.CO;2-4 Mishara, B. L., Chagnon, F., Daigle, M., Balan, B., Raymond, S., Marcoux, I., … Berman, A. (2007). Comparing models of helper behavior to actual practice in telephone crisis intervention: A silent monitoring study of calls to the U.S. 1-800-SUICIDE Network. Suicide and Life-Threatening Behavior, 37, 291–307. doi:10.1521/suli.2007.37.3.291 Mishara, B. L., & Giroux, G. (1993). The relationship between coping strategies and perceived stress in telephone intervention volunteers at a suicide prevention center. Suicide and Life-Threatening Behavior, 23, 221–229.

Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Medicine, 6, 1–6. doi:10.1371/ journal.pmed.1000097 National Coalition for Suicide Prevention. (2014). One world connected: An assessment of Australia’s progress in suicide prevention. Retrieved from http://suicidepreventionaust.org/ statement/national-coalition-for-suicide-prevention-australian-response-to-who-world-suicide-report/ Neimeyer, R. A., Fortner, B., & Melby, D. (2001). Personal and professional factors and suicide intervention skills. Suicide and Life-Threatening Behavior, 31, 71–82.

Neimeyer, R. A., & MacInnes, W. D. (1981). Assessing paraprofessional competence with the Suicide Intervention Response Inventory. Journal of Counseling Psychology, 28, 206–209.

Paterson, H., Reniers, R., & Vollm, B. (2009). Personality types and mental health experiences of those who volunteer for helplines. British Journal of Guidance and Counselling, 37, 459–471. doi:10.1080/03069880903161419 Crisis (2018), 39(1), 13–26

Pearlman, L. A. (2003). Trauma Attachment and Belief Scale. South Windsor, CT: Western Psychological Services.

Pearlman, L. A., & MacIan, P. S. (1995). Vicarious traumatization: An empirical study of the effects of trauma work on trauma therapists. Professional Psychology Research and Practice, 26, 558–565. doi:10.1037/0735-7028.26.6.558 Pope, K. S., & Tabachnick, B. G. (1994). Therapists as patients: A national survey of psychologists’ experiences, problems, and beliefs. Professional Psychology Research and Practice, 25, 247–258. doi:10.1037/0735-7028.25.3.247 Pope, K. S., Tabachnick, B. G., & Keith-Spiegel, P. (1987). Ethics of practice. The beliefs and behaviors of psychologists as therapists. American Psychologist, 42, 993–1006. doi:10.1037/0003066X.42.11.993 Rupert, P. A., & Kent, J. S. (2007). Gender and work setting differences in career-sustaining behaviors and burnout among professional psychologists. Professional Psychology Research and Practice, 38, 88–96. doi:10.1037/0735-7028.38.1.88 Rupert, P. A., & Morgan, D. J. (2005). Work setting and burnout among professional psychologists. Professional Psychology Research and Practice, 36, 544–550. doi:10.1037/0735-7028.36.5.544 Shapiro, S. L., Brown, K. W., & Biegel, G. M. (2007). Teaching self-care to caregivers: Effects of mindfulness-based stress reduction on the mental health of therapists in training. Training and Education in Professional Psychology, 1, 105–115. doi:10.1037/19313918.1.2.105 Sherman, M. D., & Thelen, M. H. (1998). Distress and professional impairment among psychologists in clinical practice.

Professional Psychology Research and Practice, 29, 79–85. doi:10.1037/0735-7028.29.1.79 Thoreson, R. W., Miller, M., & Krauskopf, C. J. (1989). The distressed psychologist: Prevalence and treatment considerations. Professional Psychology: Research and Practice, 20(3), 153–158.

Van Uffelen, J. G. Z., Wong, J., Chau, J. Y., van der Ploeg, H. P., Riphagen, I., Gilson, N. D., ... Brown, W. J. (2010). Occupational sitting and health risks: A systematic review. American Journal of Preventative Medicine, 39, 379–388. doi:10.1016/j.amepre.2010.05.024 Weiss, D. S., & Marmar, C. R. (1997). The Impact of Event Scale – revised. In J. Wilson & T. M. Keane (Eds.), Assessing psychological trauma and PTSD (pp. 399–411). New York, NY: Guilford.

West, C. P., & Shanafelt, T. D. (2007). Physician well-being and professionalism. Minnesota Medicine, 90, 44–46.

Wise, E. H., Hersh, M. A., & Gibson, C. M. (2012). Ethics, self-care and well-being for psychologists: Reenvisioning the stress-distress continuum. Professional Psychology Research and Practice, 43, 487–494. doi:10.1037/a0029446 Wong, P. T., Reker, G. T., & Gesser, G. (1994). Death attitude profile – revised: A multidimensional measure of attitudes toward death.

In R. A. Neimeyer (Ed.), Death anxiety handbook: Research, instrumentation, and application (pp. 121–148). Washington, DC: Taylor & Francis.

Received December 10, 2015 Revision received November 6, 2016 Accepted November 26, 2016 Published online March 24, 2017 About the authors Taneile Kitchingman is a PhD (Clinical Psychology) candidate at the University of Wollongong, Australia. She holds a Bachelor of Psychology (Hons I) degree (2012).

Coralie Wilson is the academic leader of Personal and Professional Development, School of Medicine, University of Wollongong, Australia, a © 2017 Hogrefe Publishing

23

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

researcher with the Illawarra Health and Medical Research Institute, and a member of several national non-government suicide prevention expert advisory groups.

Australia. He is also a board director with Suicide Prevention Australia.

He holds a Bachelor of Business and a Master’s in Policy Studies (Social Science and Policy). Alan is a PhD candidate at the University of Melbourne.

Peter Caputi is Head of the School of Psychology at the University of Wollongong, Australia.

Ian Wilson is Associate Dean of Learning and Teaching, School of Medicine, University of Wollongong, Australia, and a general/family practitioner.

Alan Woodward is Executive Director of the Lifeline Research Foundation, with responsibility for the national research program for Lifeline

Taneile A. Kitchingman School of Psychology University of Wollongong Northfields Avenue Wollongong NSW 2522 Australia tak901@uowmail.edu.au

Appendix A Table A1. PRISMA checklist Section/topic

#

Checklist item

Reported on page #

1

Identify the report as a systematic review, meta-analysis, or both.

1

2

Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.

1

Rationale

3

Describe the rationale for the review in the context of what is already known.

1–3

Objectives

4

Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS).

3

Protocol and registration

5

Indicate if a review protocol exists, if and where it can be accessed (e.g., web address), and, if available, provide registration information including registration number.

N/A

Eligibility criteria

6

Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale.

3

Information sources

7

Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.

3

Search

8

Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated.

Appendix B

Study selection

9

State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).

3

Data collection process

10

Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.

4

Data items

11

List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.

4

Risk of bias in individual studies

12

Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.

4

Title Title Abstract Structured summary

Introduction

Methods

© 2017 Hogrefe Publishing

Crisis (2018), 39(1), 13–26

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

24

Table A1. continued Section/topic

#

Checklist item

Reported on page #

Summary measures

13

State the principal summary measures (e.g., risk ratio, difference in means).

N/A

Synthesis of results

14

Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I2) for each meta-analysis.

N/A

Risk of bias across studies

15

Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies).

8

Additional analyses

16

Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified.

N/A

Study selection

17

Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.

3

Study characteristics

18

For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations.

14–16

Risk of bias within studies

19

Present data on risk of bias of each study and, if available, any outcome level assessment (see Item 12).

4, Appendices C and D

Results of individual studies

20

For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.

N/A

Synthesis of results

21

Present the main results of the review. If meta-analyses are done, include for each, confidence intervals and measures of consistency.

4–7

Risk of bias across studies

22

Present results of any assessment of risk of bias across studies (see Item 15).

Appendices C and D

Additional analysis

23

Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]).

N/A

Summary of evidence

24

Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., health-care providers, users, and policy makers).

7–8

Limitations

25

Discuss limitations at study and outcome level (e.g., risk of bias), and at review level (e.g., incomplete retrieval of identified research, reporting bias).

8–9

Conclusions

26

Provide a general interpretation of the results in the context of other evidence, and implications for future research.

9

27

Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review.

10

Results

Discussion

Funding Funding

Crisis (2018), 39(1), 13–26

© 2017 Hogrefe Publishing

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

25

Appendix B Search strategy Search 1: CINAHL 1. telephone intervention OR crisis line OR hot line OR help line (Abstract) AND 2. worker* OR volunteer* OR staff* (Abstract) AND 3. stress OR distress OR anxi* OR depress* OR suicid* OR burnout OR secondary traumatic stress OR compassion fatigue OR help-seeking OR treatment OR supervis* OR service provision OR skill* OR performance (Abstract) Limits: None Search 2: Medline 1. telephone intervention OR crisis line OR hot line OR help line (Abstract) AND 2. worker* OR volunteer* OR staff* (Abstract) AND 3. stress OR distress OR anxi* OR depress* OR suicid* OR burnout OR secondary traumatic stress OR compassion fatigue OR help-seeking OR treatment OR supervis* OR service provision OR skill* OR performance (Abstract) Limits: None Search 3: PsycINFO 1. telephone intervention OR crisis line OR hot line OR help line (Abstract) AND 2. worker* OR volunteer* OR staff* (Abstract) AND 3. stress OR distress OR anxi* OR depress* OR suicid* OR burnout OR secondary traumatic stress OR compassion fatigue OR help-seeking OR treatment OR supervis* OR service provision OR skill* OR performance (Abstract) Limits: None Search 4: Scopus 1. telephone intervention OR crisis line OR hot line OR help line (Abstract) AND 2. worker* OR volunteer* OR staff* (Abstract) AND 3. stress OR distress OR anxi* OR depress* OR suicid* OR burnout OR secondary traumatic stress OR compassion fatigue OR help-seeking OR treatment OR supervis* OR service provision OR skill* OR performance OR impairment (Abstract) Limits: None

© 2017 Hogrefe Publishing

Crisis (2018), 39(1), 13–26

T. A. Kitchingman et al., Telephone Crisis Support Workers’ Distress and Impairment

26

Appendix C Table C1. Quality assessment for all included studies, per criterion, in alphabetical order by author name Criterion

Objective

Design

Target population and sample

Variables

Data sources/ collection

Measurement

Statistics

Score out of 15

1

2

3a

3b

3c

3d

3e

3f

3g

4

5a

5b

5c

6a

6b

Cyr & Dowrick (1991)

Y

Y

Y

N

Y

Y

N

Y

N

Y

Y

N

Y

Y

N

10

Dunkley & Whelan (2006)

Y

N

Y

N

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

13

Furlonger et al. (2013)

Y

Y

Y

N

Y

Y

N

Y

Y

Y

Y

Y

Y

Y

Y

13

McClure et al.

(1973)

Y

N

N

N

N

N

N

N

Y

Y

Y

N

N

Y

Y

6

Mishara & Giroux (1993)

Y

Y

Y

N

N

Y

Y

Y

N

Y

Y

N

N

Y

N

9

Niemeyer et al. (2001)

Y

N

Y

N

N

Y

N

Y

N

Y

Y

Y

Y

Y

N

9

Paterson et al.

(2009)

Y

Y

Y

N

Y

Y

N

N

Y

Y

Y

N

N

Y

Y

10

Note. Y = yes. N = no.

Appendix D Table D1. Number (%) of studies scoring a point for quality assessment criterion Item

Criterion

Description

n (%)

1

Objectives

Are the objectives or hypothesis of the research described in the paper stated?

7 (100)

2

Study design

Is the study design presented?

4 (57)

3a

Target population

Do the authors describe the target population they wanted to research?

6 (86)

3b

Sample

Was a random sample of the target population taken? AND was the response rate 60% or more?

0

3c

Sample

Is participant selection described?

4 (57)

3d

Sample

Is participant recruitment described or referred to?

6 (86)

3e

Sample

Are the inclusion and/or exclusion criteria stated?

2 (29)

3f

Sample

Is the study sample described? (minimum description = sample size, gender, age and an indicator of SES)

5 (71)

3g

Sample

Are the numbers of participants at each stage of the study reported? (Authors should report at least numbers eligible, numbers recruited, numbers with data at baseline, and numbers lost to follow-up)

4 (57)

4

Variables

Are the measures described?

7 (100)

5a

Data sources and collection

Do authors describe the source of their data AND did authors describe how the data were collected?

7 (100)

5b

Measurement

Was the reliability of the measure(s) mentioned or referred to?

3 (43)

5c

Measurement

Was the validity of the measure(s) mentioned or referred to?

4 (57)

6a

Statistical methods

Were appropriate statistical methods used and described, including those for addressing confounders?

7 (100)

6b

Statistical methods

Were the numbers/percentages of participants with missing data indicated AND If more than 20% of data in the primary analyses were missing, were methods used to address missing data?

3 (43)

Crisis (2018), 39(1), 13–26

(0)

© 2017 Hogrefe Publishing

Research Trends

Age at Exposure to Parental Suicide and the Subsequent Risk of Suicide in Young People Kuan-Ying Lee1,2, Chung-Yi Li2,3, Kun-Chia Chang1,2 , Tsung-Hsueh Lu2, and Ying-Yeh Chen4,5 Department of General Psychiatry, Jianan Psychiatric Center, Ministry of Health and Welfare, Tainan, Taiwan Department of Public Health, College of Medicine, National Cheng-Kung University, Tainan, Taiwan 3 Department of Public Health, College of Public Health, China Medical University, Taichung, Taiwan 1 2

Taipei City Psychiatric Center, Taipei City Hospital, Taipei, Taiwan Institute of Public Health and Department of Public Health, National Yang-Ming University, Taipei, Taiwan

4 5

Abstract. Background: We investigated the age at exposure to parental suicide and the risk of subsequent suicide completion in young people.

The impact of parental and offspring sex was also examined. Method: Using a cohort study design, we linked Taiwan’s Birth Registry (1978–1997) with Taiwan’s Death Registry (1985–2009) and identified 40,249 children who had experienced maternal suicide (n = 14,431), paternal suicide (n = 26,887), or the suicide of both parents (n = 281). Each exposed child was matched to 10 children of the same sex and birth year whose parents were still alive. This yielded a total of 398,081 children for our non-exposed cohort. A Cox proportional hazards model was used to compare the suicide risk of the exposed and non-exposed groups. Results: Compared with the non-exposed group, offspring who were exposed to parental suicide were 3.91 times (95% confidence interval [CI] = 3.10–4.92 more likely to die by suicide after adjusting for baseline characteristics. The risk of suicide seemed to be lower in older male offspring (HR = 3.94, 95% CI = 2.57–6.06), but higher in older female offspring (HR = 5.30, 95% CI = 3.05–9.22). Stratified analyses based on parental sex revealed similar patterns as the combined analysis. Limitations: As only register-­based data were used, we were not able to explore the impact of variables not contained in the data set, such as the role of mental illness. Conclusion: Our findings suggest a prominent elevation in the risk of suicide among offspring who lost their parents to suicide. The risk elevation differed according to the sex of the afflicted offspring as well as to their age at exposure.

Keywords: suicide, developmental stage, offspring, parental suicide, age, sex/gender

Familial clustering of suicidal behavior has been well documented (Brent & Mann, 2005; Burke et al., 2010; Goodwin, Beautrais, & Fergusson, 2004; Kim et al., 2005; Lieb, Bronisch, Hofler, Schreier, & Wittchen, 2005; Mittendorfer-Rutz, Rasmussen, & Wasserman, 2008). Mounting evidence from twin studies, adoption studies, and largescale data linkage research has helped to substantiate the link between parental suicide and an elevated risk of suicide in offspring. A recent meta-analysis exploring the intergenerational transmission of suicidal behaviors has concluded that parental suicidal behavior conveys an approximately two- to threefold increased risk of suicidal behavior to offspring (Geulayov, Gunnell, Holmen, & Metcalfe, 2012). The mechanisms underpinning this increase in risk have been suggested to be related to genetic vulnerability (Brent & Melhem, 2008), socioenvironmental factors (Brent et al., 2004; Brodsky et al., 2008; Chaffin, Kelleher, & Hollenberg, 1996; De Bellis & Thomas, 2003; Roberts & Hawton, 1980), and gene–environment interactions (Caspi et al., 2002; Kendler, Sheth, Gardner, & Prescott, 2002). © 2017 Hogrefe Publishing

Few studies, however, have investigated the effect that the age at exposure to parental suicide has on the subsequent suicide risk in offspring. The life-course approach suggests that there may be critical stages in an individual’s development that predispose them to an increased sensitivity to exposures (Ben-Shlomo & Kuh, 2002; Kuh, Ben-Shlomo, Lynch, Hallqvist, & Power, 2003). Thus, the impact of biological and social exposures may vary according to the developmental stage during which an individual is exposed. To our knowledge, only two studies have investigated the relationship between the age at exposure to parental suicidal behavior and an offspring’s subsequent suicide risk (Niederkrotenthaler, Floderus, Alexanderson, Rasmussen, & Mittendorfer-Rutz, 2012; Wilcox et al., 2010).

In the first study, Wilcox et al. analyzed Swedish registers to estimate the risk of suicide among offspring exposed to parental suicide during three developmental stages: childhood, adolescence, and young adulthood. While they found that exposure to parental suicide as children and adolescents (< 18 years old) increased the subsequent Crisis (2018), 39(1), 27–36 https://doi.org/10.1027/0227-5910/a000468

28

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

risk of suicide by about threefold, young adults (≥ 18 years old) exposed to parental suicide were not found to be at an increased risk for suicide compared with the offspring of non-suicide decedents (Wilcox et al., 2010).

In the second study, Niederkrotenthaler et al. analyzed linked datasets from Sweden and found that exposure to parental suicide, whether before the age of 10 years (odds ratio [OR] = 2.93, 95% CI = 1.67–5.15) or after (OR = 2.36, 95% CI = 1.46–3.80), was associated with an elevated risk of subsequent suicide completion (Niederkrotenthaler et al., 2012). Taken together, these two studies may suggest that exposure to parental suicide during childhood conveys a greater risk of suicide than exposure later in life.

However, neither of these two studies stratified subjects by either age or parental and offspring sex, both of which have been suggested by previous studies to play a role in subsequent offspring suicide risk (Niederkrotenthaler et al., 2012; Wilcox et al., 2010). In addition, these two studies were conducted in Sweden, which could differ markedly from other cultures, as different societal representations of death and suicide could mediate the relationship of intergenerational transmission of suicide in different ways.

Therefore, replications of this investigation in other countries and regions are warranted.

Therefore, to further explore these putative associations and any interaction between them, we utilized a linked dataset from Taiwan to investigate the age at exposure to parental suicide and the subsequent risk of suicide in offspring while conducting stratified analyses according to parental and offspring sex.

In Taiwan, the suicide rates in the age group of 15–19-year-olds were generally less than 4 per 100,000 population and did not fluctuate markedly in the past decade (Lin & Lu, 2008; Taiwan Suicide Prevention Center, 2016). These figures are relatively low compared with a mean global suicide rate of 7.4 per 100,000 population in teenagers aged 15–19 assessed during 2003–2014 (McLoughlin, Gould, & Malone, 2015). In many developed countries, such as the United States, Australia, Scandinavia, South Korea, and Japan, teenage suicide rates (15–19 year old) have generally ranged between 6 and 15 per 100,000 population throughout the past decade (McLoughlin et al., 2015). Although the overall youth suicide rate has been low in Taiwan, the male-to-female suicide ratio in teenagers (around 1.5:1) has been comparatively lower than in many Western countries (generally greater than 3:1) (Chen, Wu, Yousuf, & Yip, 2012), thus indicating the importance of exploring specific age and gender patterns of suicide risk factors in different sociocultural contexts.

Crisis (2018), 39(1), 27–36

Method Sources of Data This study analyzed a dataset produced by linking the Taiwan Birth Registry (1978–1997) with the Taiwan Death Registry (1985–2009) using national identification numbers. In Taiwan, it is a legal requirement that all live births and deaths be registered within 10 days. Various birth characteristics including sex, birth weight, gestational age, single/multiple birth, birth order, parental ages, education, and mother’s marital status are available for each live birth in the Taiwan Birth Registry. Both the Taiwan Birth Registry and Taiwan Death Registry have been evaluated for data quality and are considered valid and complete (Li et al., 2010; Lu, Lee, & Chou, 2000). Access to the birth and death registries was approved by the Department of Health.

Retrospective Cohort Design A total of 6,996,930 live births registered in Taiwan between 1978 and 1997 were identified; these children were born to 3,420,178 fathers and 3,456,183 mothers. To identify a cohort of offspring whose parents died by suicide, parental identification numbers were linked to the death registry. This yielded 13,723 fathers and 7,462 mothers who died by suicide. The identification numbers of these parental suicide decedents were re-linked to the birth registry to get a complete list of offspring who were born to parents who died by suicide. In total, there were 26,887 individuals exposed to paternal suicide and 14,431 individuals exposed to maternal suicide; 281 children experienced from both maternal and paternal suicide. In total, our exposed group contained 40,249 children who had experienced either maternal suicide, paternal suicide, or the suicide of both parents. The remaining children were candidates for the non-exposed group of the study. Ten children whose parents were still alive were matched to each exposed child on the same sex and birth year, ranging from 1978 to 1997. This yielded a total of 398,081 children for the non-exposed cohort.

Outcome Variable The outcome variable was suicide. Deaths certified as undetermined intent were also included because previous research has indicated that many deaths of undetermined intent are likely to be missed suicides (Chang, Sterne, Lu, & Gunnell, 2010). Overall, we used the following codes to © 2017 Hogrefe Publishing

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

identify suicides: International Classification of Diseases, Ninth Revision (ICD-9) E950-959 (intentional self-harm), ICD-9 E980-989 (undetermined intent), ICD-10 X60– X84 (intentional self-harm), and ICD-10 Y10-Y34 (undetermined intent).

Predictor Variable and Confounders Initially, we categorized the age at exposure to parental suicide according to three developmental stages: childhood (0–12 years), adolescence (13–17 years), and young adulthood (18–31 years). A trend test was conducted to examine whether there was a significant trend in the risk of offspring suicide according to the age at exposure. The statistical test was not significant for either sex (male, p = .30; female, p = .11). In addition, the sample size was very small for female offspring, with only six subjects having been exposed to parental suicide between the ages of 13 and 17 years. We hence combined the 0–12- and 13–17-year categories in the analysis of offspring suicide risk.

Covariates that may be causally related to the outcome of interest (i.e., offspring suicide) and may be associated with parental suicide were considered as potential confounders in the analysis. They included maternal age, parental educational attainment, and urbanization index.

Maternal age was grouped into four categories: < 20, 20– 24, 25–29, and > 30 years. Educational attainment was categorized into elementary or lower (< 6 years of education), junior high (7–9 years), senior high (10–12 years), and college or higher (> 12 years). We used the highest educational attainment among the parents to index parental educational attainment. The urbanization index was categorized into four groups: metropolitan, urban, suburban, and rural (Liu et al., 2006).

Data Analysis We first described the distribution of risk factors in the exposed and comparison group. We then analyzed data using the Cox proportional hazards model to estimate hazard ratios (HRs) and their 95% confidence intervals (CIs).

Both crude and adjusted models are presented. Potential confounders including maternal age, parental educational attainment, and urbanization index were controlled in the adjusted model. Analyses were stratified by age (0–17 years and 18–31 years) and sex (parental sex and offspring sex). The proportional-hazards assumption was verified using plots of log(-log(survival function)) versus time.

To take family clustering into consideration, we did a separate analysis using SAS concatenation procedure. In addition, there were 281 individuals who lost two parents. © 2017 Hogrefe Publishing

29

As the loss of two parents may engender a different effect than the loss of only one parent, we performed a separate sensitivity analysis to explore whether the subsequent risk of suicide differed between offspring bereaved by one or both parents.

Results In the exposed group, 39.9% (n = 16,049) experienced parental suicide at an age younger than 12 years, 23.5% (n = 9,452) experienced parental death between 13 and 17 years of age, and 36.6% (n = 14,748) experienced parental death between 18 and 31 years of age. The maleto-female sex ratio in our cohort was 1:0.92. The distributions of sociodemographic variables in the exposed (n = 40,249) and comparison group (n = 398,081) are presented in Table 1. Compared with the non-exposed cohort, offspring exposed to parental suicide tended to have higher proportions of young mothers (< 20 years) among the suicide decedents, low birth weight, and lower parental educational attainment. In addition, the exposed group was less likely to live in metropolitan cities, and, conversely, was more likely to live in rural areas.

Compared with the non-exposed group, offspring exposed to parental suicide were 3.91 times more likely to die by suicide after adjusting for baseline characteristics (95% CI = 3.10–4.92). This elevated risk in our exposed group was observed across all ages of exposure (Table 2). When stratified by sex, there was some indication of interaction between offspring gender and age at exposure to parental suicide. The overall HR for male and female offspring was estimated at 4.02 and 3.68, respectively. For female offspring, the risk of suicide among those exposed to parental suicide before age 18 was 2.76 times that of the comparison group (95% CI = 1.53–4.97), whereas the risk was 5.30 times that of the comparison group (95% CI = 3.05–9.22) for those exposed after 18 years of age. The relationship between age at exposure and suicide risk of female offspring was not statistically significant (p = .10).

Among male offspring exposed to parental suicide during adulthood the adjusted HR was 3.94 (95% CI = 2.57– 6.06), while the HR was 4.47 (95% CI = 3.07–6.50) for those who experienced parental suicide in childhood/adolescence (see Table 2). The relationship between age at exposure and suicide risk of male offspring was not statistically significant (p = .51).

To assess whether there was same-sex identification between parental suicide decedents and their offspring, we analyzed the impact of paternal and maternal suicide separately while stratifying by both offspring sex and age at exposure (Tables 3 and 4). The stratified analyses acCrisis (2018), 39(1), 27–36

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

30

Table 1. Offspring characteristics in the exposed group and the comparison group

Risk factors

Exposed group (N = 40,249)

Comparison group (N = 398,081)

N (%)

N (%)

p .89

Matching variables Age at exposure to parental suicide (years) 0–12

16,049 (39.9)

158,352 (39.8)

13–17

9,452 (23.5)

93,876 (23.6)

18–31

14,748 (36.6)

145,853 (36.6)

Period of birth

.96

1978–1982

14,354 (35.7)

142,520 (35.8)

1983–1987

10,681 (26.5)

105,476 (26.5)

1988–1992

8,640 (21.5)

85,239 (21.4)

1993–1997

6,574 (16.3)

64,846 (16.3)

Sex

.76

Male

21,049 (52.3)

207,871 (52.2)

Female

19,200 (47.7)

190,210 (47.8) < .0001

Control variables Maternal age (years) 3,782 (9.4)

17,599 (4.4)

20–24

14,818 (36.8)

123,422 (31.0)

25–29

14,657 (36.4)

172,885 (43.4)

6,813 (16.9)

82,155 (20.6)

179 (0.4)

2,020 (0.5)

< 20

> 30 = Unknown Paternal educational attainment

< .0001

Elementary or lower

14,379 (35.7)

74,021 (18.6)

Junior high

13,295 (33.0)

101,158 (25.4)

Senior high

9,776 (24.3)

144,821 (36.4)

2758 (6.9)

77,721 (19.5)

360 (0.1)

41 (0.1)

Metropolitan

9,974 (24.8)

108,793 (27.3)

Urban

9,347 (23.2)

95,485 (24.0)

College or higher Unknown Urbanization index

Suburban Rural Unknown

< .0001

7,338 (18.2)

69,761 (17.5)

13,296 (33.0)

120,666 (30.3)

294 (0.7)

3,376 (0.8)

cording to maternal and paternal suicide did not seem to differ substantially from the findings of the combined analysis. Regardless of the parental suicide decedent’s sex, exposure during adulthood and childhood/adolescence seemed to carry similar risks for future suicide among male offspring, while among female offspring, the risk of suicide was higher when the exposure occurred during adulthood.

The results of the SAS concatenation procedure are presented in Appendix A. Point estimates and their 95% conCrisis (2018), 39(1), 27–36

fidence interval did not seem to change after taking family clustering into account.

In our exposure cohort, there were 281 youths who lost both parents. To consider the possibility that individuals exposed to the death of two parents may differ from those who only lost one parent, we performed a sensitivity analysis eliminating individuals who lost two parents (Appendix B). The results did not differ substantially from those who only experienced the loss of one parent.

© 2017 Hogrefe Publishing

31

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

Table 2. The effect of parental suicide on the risk of offspring suicide by sex and age Age at father/mother suicide (years)

Exposed group

Comparison group

n

Rateb

n

Rateb

Crude HR

p

Adjusted HRa

0–17

62

22.75

133

4.92

4.63

3.43

6.26

< .0001

3.86

2.81

5.28

< .0001

18–31

53

79.26

115

17.27

4.59

3.31

6.35

< .0001

4.43

3.17

6.20

< .0001

115

33.89

248

7.36

4.61

3.69

5.75

< .0001

3.91

3.10

4.92

< .0001

95% CI

95% CI

p

All offspring

All Male offspring 0–17

46

32.12

88

6.19

5.20

3.64

7.43

< .0001

4.47

3.07

6.50

< .0001

18–31

31

89.36

74

21.45

4.17

2.74

6.34

< .0001

3.94

2.57

6.06

< .0001

All

77

43.28

162

9.16

4.72

3.60

6.20

< .0001

4.02

3.03

5.33

< .0001

16

12.38

45

3.51

3.53

2.00

6.25

< .0001

2.76

1.53

4.97

< .001

Female offspring 0–17 18–31

22

68.38

41

12.78

5.35

3.19

8.98

< .0001

5.30

3.05

9.22

< .0001

All

38

23.54

86

5.37

4.39

3.00

6.43

< .0001

3.68

2.47

5.48

< .0001

Note. Reference group is the comparison group who had no exposure to parental suicide. HR = hazard ratio. a Adjusted HR for maternal age, parental education level, and urbanization index. bRate indicates incidence rate per 100,000 person-years.

Table 3. The effect of paternal suicide on the risk of offspring suicide, by sex and age Age at paternal suicide (years)

Exposed group n

Rateb

Comparison group n

Rateb

Crude HR

95% CI

p

Adjusted HRa

95% CI

p

All offspring 0–17

34

20.95

133

4.92

4.44

3.05

6.48

< .0001

3.81

2.59

5.61

< .0001

18–31

33

72.29

115

17.27

4.18

2.84

6.16

< .0001

4.04

2.72

6.00

< .0001

All

67

32.22

248

7.36

4.43

3.38

5.81

< .0001

3.83

2.90

5.05

< .0001

27

31.90

88

6.19

5.36

3.49

8.26

< .0001

4.72

3.02

7.36

< .0001

Male offspring 0–17 18–31

19

80.35

74

21.45

3.74

2.26

6.20

< .0001

3.62

2.17

6.03

< .0001

All

46

42.48

162

9.16

4.69

3.38

6.51

< .0001

4.11

2.93

5.75

< .0001

0–17

7

9.01

45

3.51

2.70

1.22

5.98

0.01

2.20

0.98

4.94

0.06

18–31

14

63.62

41

12.78

4.97

2.71

9.12

< .0001

4.73

2.53

8.86

< .0001

All

21

21.07

86

5.37

3.98

2.47

6.42

< .0001

3.33

2.04

5.43

< .0001

Female offspring

Note. Reference group is the comparison group who had no exposure to parental suicide. HR = hazard ratio. a Adjusted HR for maternal age, parental education level, and urbanization index. bRate indicates incidence rate per 100,000 person-years.

Discussion Main Findings We found an approximately fourfold increased risk of suicide in offspring who experienced parental suicide. The potential influence of age at exposure tends to be different between male and female offspring. The risk of suicide seemed to be lower in older male offspring (HR = 3.94, 95% CI = 2.57–6.06), but higher in older female offspring (HR = 5.30, 95% CI = 3.05–9.22). Adult females (18–31 © 2017 Hogrefe Publishing

years) exposed to parental suicide exhibited the most prominent risk elevation (HR = 5.3), with those exposed in childhood/adolescence (< 18 years) only bearing approximately half the risk (HR = 2.76). The risk for subsequent suicide in male offspring exposed to parental suicide in adulthood (HR = 3.94) was not statistically different from male offspring exposed in childhood/adolescence (HR = 4.47).

These patterns were evident in both male and female offspring regardless of the parental suicide decedent’s sex.

Crisis (2018), 39(1), 27–36

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

32

Table 4. The effect of maternal suicide on the risk of offspring suicide, by sex and age Age at maternal suicide (years)

Exposed group

Comparison group

n

Rateb

n

Rateb

Crude HR

p

Adjusted HRa

0–17

31

27.65

133

4.92

5.32

3.60

7.87

< .0001

4.43

2.95

6.66

< .0001

18–31

21

97.08

115

17.27

5.62

3.53

8.95

< .0001

5.29

3.28

8.53

< .0001

All

52

38.87

248

7.36

5.18

3.84

6.99

< .0001

4.27

3.13

5.82

< .0001

95% CI

95% CI

p

All offspring

Male offspring 0–17

21

35.18

88

6.19

5.39

3.35

8.68

< .0001

4.79

2.92

7.88

< .0001

18–31

12

106.30

74

21.45

4.96

2.69

9.12

< .0001

4.69

2.51

8.74

< .0001

All

33

46.49

162

9.16

4.98

3.42

7.24

< .0001

4.23

2.87

6.23

< .0001

10

19.07

45

3.51

5.12

2.58

10.16

< .0001

3.81

1.87

7.71

.0002

Female offspring 0–17 18–31 All

9

87.01

41

12.78

6.82

3.32

14.04

< .0001

6.44

3.03

13.65

< .0001

19

30.26

86

5.37

5.53

3.36

9.09

< .0001

4.28

2.55

7.18

< .0001

Note. Reference group is the comparison group who had no exposure to parental suicide. HR = hazard ratio. a Adjusted HR for maternal age, parental education level, and urbanization index. bRate indicates incidence rate per 100,000 person-years.

Strengths and Limitations The strength of this study lies in its cohort design, use of a large representative sample (n = 438,330), and long follow-up period (1978–2009). Although existing studies tend to support an elevated risk of offspring suicide after parental suicide, few studies have investigated how the age at exposure to parental suicide may influence subsequent suicide risk. In addition to providing further evidence regarding the association between the age at exposure to parental suicide and subsequent offspring suicide from an Asian population, this is also the first study to explore the effects of offspring and parental sex on this association.

However, our current findings need to be considered in light of several limitations. First, we analyzed a dataset constructed by linking the Taiwan Birth Registry (1978– 2009) and Taiwan Death Registry (1985–2009). Since complete identification numbers for the Taiwan Death Registry only became available in 1985, any parents who died by suicide between 1978 and 1984 could not be identified. As this may have resulted in some subjects exposed to suicide between 1978 and 1984 having been included in the non-exposed cohort, it may have contributed to an underestimation of the suicide risk in the exposed cohort.

Secondly, we eliminated subjects (n = 14) who experienced parental suicide before 10 years of age because previous studies have indicated that children only develop an understanding of death sometime after the age of 10 (Dervic, Brent, & Oquendo, 2008; Mitchell et al., 2006; Shaffer, 1988). However, if there were some rare cases of suicide occurring before age 10, our estimated hazard ratio may have been slightly attenuated.

Crisis (2018), 39(1), 27–36

Third, as only register-based data were available, we were not able to assess any possible underlying mechanisms of the association, such as the role of mental illness.

We were additionally unable to gather any information regarding how parental loss was experienced by each individual, which may have varied widely. For example, some offspring may have witnessed the suicides, whereas others may have grown up with the events being concealed.

Interpretations The suicide risk elevation estimated in our investigation is larger than those reported in the literature. A recent meta-analysis summarized studies conducted between 1976 and 2011, and estimated that the relative risk of suicide in those who lost a parent to suicide was 1.94 (95% CI = 2.54–2.70; Geulayov et al., 2012), which is about half the hazard ratio reported in this study (HR = 3.91).

In addition to the potentially inadequate adjustment for confounding variables, such as familial psychiatric disorders, the cultural milieu in which our study was conducted may also have contributed to the greater risk observed in our sample. As opposed to the studies in the literature that were conducted in Western populations, this investigation was performed in a Confucian society where the breakdown of the family-oriented social unit has been suggested to lead to an even greater degree of social isolation and exclusion than in the West (Hung, Huang, Yip, Fan, & Chen, 2013; Lee et al., 2011). Thus, it is possible that the offspring of suicide decedents in this study suffered a comparatively greater social trauma than their © 2017 Hogrefe Publishing

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

Western counterparts, which may have predisposed them to a greater suicide risk.

In addition to estimating a larger magnitude of risk than previous studies, the results of our investigation also depart from the literature in terms of the impact of a sensitive period. In line with life course epidemiology, which posits that social phenomena do not have uniform effects throughout an individual’s lifetime (Ben-Shlomo & Kuh, 2002), previous studies have indicated that there is a sensitive period for the impact of parental suicide on children’s suicidal behaviors (Niederkrotenthaler et al., 2012; Wilcox et al., 2010). Namely, the suicide of a parent is especially difficult and disruptive when it occurs early in a person’s life, whereas when the same event occurs in adulthood, the impact is presumably weaker. Various explanations for this phenomenon have been raised. For example, younger children have been identified to be particularly sensitive to the loss of a parent. As we do not expect a parent to die while he/she is relatively young, such a death generally involves a higher level of stress, stronger feelings of grief, and fewer available coping strategies to deal with the death (Rostila, 2015). Moreover, parental death may also have social and economic consequences (such as financial difficulties, and reduced surveillance over deviant behaviors), and hence young children may be particularly vulnerable to parental loss in general (Rostila, 2015; Rostila & Saarela, 2011). Overall, the existing findings on the impact of parental death due to suicide, although scarce, tend to support this sensitive period proposition (Niederkrotenthaler et al., 2012; Wilcox et al., 2010); however, the results of the present study do not substantiate the sensitive period hypothesis.

There are several reasons why we may not have detected a sensitive period of exposure to parental suicide during childhood for subsequent suicide risk. First, it is possible that children who lose a parent at a very young age are not mature enough to adequately comprehend the meaning of suicide stigma or the depth of their loss. Thus, it is possible that the sense of shame, isolation, and exclusion associated with suicide stigma is stronger in older children, as they are more capable of understanding their social circumstances, and strongest among females because of the greater emphasis they place on social relationships (Umberson, Chen, House, Hopkins, & Slaten, 1996). As mentioned, the stigma and social consequences of suicide are enhanced in Chinese society (Tzeng & Lipson, 2004) and thus the greater social burden associated with parental suicide among adult Taiwanese may have worked to offset the effect of the sensitive period to parental suicide observed during childhood by researchers in Western countries.

One further explanation is the higher prevalence (15%) of three-generational families in Taiwan (National Statistics R.O.C (Taiwan), 2016), almost three times the rate of the © 2017 Hogrefe Publishing

33

United States (Ku et al., 2013). Grandparents caring for grandchildren is culturally encouraged in Taiwan (Ku et al., 2013) and may have somehow alleviated the psychological impact of losing a parent to suicide at a young age.

The marked increase in risk of suicide in female offspring who experienced parental suicide in adulthood may also be best explained through Asian sociocultural factors, namely, that the social stigma attached to suicide in Chinese society tends to hit women more severely. In Confucian culture, marriage for a woman usually indicates the departure from her own family, and her entry into an unfamiliar household (Watson & Ebrey, 1991). Since parental suicide in Chinese culture is regarded as a family dishonor, through marriage it is believed that the event may bring misfortune and shame to the new household (Watson & Ebrey, 1991). This social stigma may have a stronger labeling effect on female offspring who lost a parent to suicide in adulthood than those who experienced parental suicide early in life, as a long past event is easier to conceal than a more recent one. However, it is also possible that this observation was a chance finding as the event rate in women was relatively lower. We also considered the impact of birth order, as it is possible that the first-born older female children exposed to parental suicide were under the stress of caring for their younger siblings. However, a separate analysis did not show a significant difference in the suicide risk of first-born child versus non-first-born children following parental suicide (data not shown).

Since suicide-related stigma has not been rigorously explored in Chinese societies, qualitative studies are needed to provide an in-depth understanding of how the suicide taboo affects male and female suicide survivors differently, and how suicide stigma is experienced among survivors exposed at various life stages. Issues generated from these qualitative studies can be used to design further quantitative studies to explore the impact of different aspects of stigma on the grief process of suicide survivors.

We did not observe a stronger maternal influence on offspring suicide as suggested by a previous study (Agerbo, Nordentoft, & Mortensen, 2002), but rather, similar to Mittendorfer-Rutz et al., observed equal impacts of maternal and paternal suicide on subsequent suicide risk in the suicide decedent’s offspring (Mittendorfer-Rutz et al., 2008). We furthermore did not find any evidence in support of same-sex identification in the transmission of suicide risk, as has been suggested by a previous study performed in Taiwan (Cheng et al., 2014). This inconsistency may be related to differences in the study design, the study period, and the age groups investigated between the previous study and ours. The by Cheng et al. utilized a case-control design, while a retrospective cohort design was used in the current investigation. Our study sample was followed up from 1985 to 2009 with an age range of 11–31 years, Crisis (2018), 39(1), 27–36

34

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

whereas the study by Cheng et al. was restricted to suicide cases aged 15–19 that occurred during 1997–2007 (Cheng et al., 2014). Existing studies are still too few to draw a definite conclusion on whether parental and offspring sex/gender moderates the familial transmission of suicide.

Conclusion This study conducted in an Asian population yielded larger hazard ratios for suicide risk than the risk estimates reported by investigations conducted in Western countries. We further detected interesting patterns in risk by age at exposure and sex. We were able to provide explanations for these trends by working through the lens of a Confucian value system, and thus the same trends may not be present in other populations. These results suggest the need to explore the age and sex/gender effect among offspring exposed to parental suicide in different sociocultural contexts.

Acknowledgments Professors Ying-Yeh Chen and Tsung-Hsueh Lu contributed equally to this study.

Y-Y.C. was supported by the Ministry of Science and Technology (Grant number: MOST 103-2314-B-532 -003-MY2) and the Department of Health, Taipei City Government (Grant number:10501-62-062).

The authors declare that there are no conflicts of interest related to the subject of this study.

References Agerbo, E., Nordentoft, M., & Mortensen, P. B. (2002). Familial, psychiatric, and socioeconomic risk factors for suicide in young people: Nested case-control study. BMJ, 325(7355), 74.

Ben-Shlomo, Y., & Kuh, D. (2002). A life course approach to chronic disease epidemiology: conceptual models, empirical challenges and interdisciplinary perspectives. International Journal of Epidemiology, 31(2), 285–293.

Brent, D. A., & Mann, J. J. (2005). Family genetic studies, suicide, and suicidal behavior. American Journal of Medical Genetics.

Part C, Seminars in Medical Genetics, 133C(1), 13–24.

Brent, D. A., & Melhem, N. (2008). Familial transmission of suicidal behavior. The Psychiatric Clinics of North America, 31(2), 157–177.

Brent, D. A., Oquendo, M., Birmaher, B., Greenhill, L., Kolko, D., Stanley, B., & Mann, J. J. (2004). Familial transmission of mood disorders: Convergence and divergence with transmission of suicidal behavior. Journal of the American Academy of Child and Adolescent Psychiatry, 43(10), 1259–1266.

Brodsky, B. S., Mann, J. J., Stanley, B., Tin, A., Oquendo, M., Birmaher, B., & Brent, D. (2008). Familial transmission of suicidal beCrisis (2018), 39(1), 27–36

havior: Factors mediating the relationship between childhood abuse and offspring suicide attempts. The Journal of Clinical Psychiatry, 69(4), 584–596.

Burke, A. K., Galfalvy, H., Everett, B., Currier, D., Zelazny, J., Oquendo, M. A., & Brent, D. A. (2010). Effect of exposure to suicidal behavior on suicide attempt in a high-risk sample of offspring of depressed parents. Journal of the American Academy of Child and Adolescent Psychiatry, 49(2), 114–121.

Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., & Poulton, R. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 297(5582), 851–854.

Chaffin, M., Kelleher, K., & Hollenberg, J. (1996). Onset of physical abuse and neglect: Psychiatric, substance abuse, and social risk factors from prospective community data. Child Abuse & Neglect, 20(3), 191–203.

Chang, S. S., Sterne, J. A., Lu, T. H., & Gunnell, D. (2010). ‘Hidden’ suicides amongst deaths certified as undetermined intent, accident by pesticide poisoning and accident by suffocation in Taiwan. Social Psychiatry and Psychiatric Epidemiology, 45(2), 143–152.

Chen, Y. Y., Wu, K. C. C., Yousuf, S., & Yip, P. S. F. (2012). Suicide in Asia: Opportunities and challenges. Epidemiologic Reviews, 34(1), 129–144.

Cheng, C. C., Yen, W. J., Chang, W. T., Wu, K. C., Ko, M. C., & Li, C. Y. (2014). Risk of adolescent offspring’s completed suicide increases with prior history of their same-sex parents’ death by suicide. Psychological Medicine, 44(9), 1845–1854.

De Bellis, M. D., & Thomas, L. A. (2003). Biologic findings of post-traumatic stress disorder and child maltreatment. Current Psychiatry Reports, 5(2), 108–117.

Dervic, K., Brent, D. A., & Oquendo, M. A. (2008). Completed suicide in childhood. Psychiatric Clinics of North America, 31(2), 271–291.

Geulayov, G., Gunnell, D., Holmen, T. L., & Metcalfe, C. (2012). The association of parental fatal and non-fatal suicidal behaviour with offspring suicidal behaviour and depression: A systematic review and meta-analysis. Psychological Medicine, 42(8), 1567–1580.

Goodwin, R. D., Beautrais, A. L., & Fergusson, D. M. (2004). Familial transmission of suicidal ideation and suicide attempts: Evidence from a general population sample. Psychiatry Research, 126(2), 159–165.

Hung, G. C. L., Huang, M. C., Yip, P. S. F., Fan, H. F., & Chen, Y. Y. (2013).

Association between childhood adversities and suicide attempts among alcoholic inpatients in Taiwan. Journal of Studies on Alcohol and Drugs, 74(4), 559–564.

Kendler, K. S., Sheth, K., Gardner, C. O., & Prescott, C. A. (2002).

Childhood parental loss and risk for first-onset of major depression and alcohol dependence: The time-decay of risk and sex differences. Psychological Medicine, 32(7), 1187–1194.

Kim, C. D., Seguin, M., Therrien, N., Riopel, G., Chawky, N., Lesage, A. D., & Turecki, G. (2005). Familial aggregation of suicidal behavior: A family study of male suicide completers from the general population. The American Journal of Psychiatry, 162(5), 1017–1019.

Ku, L. J., Stearns, S. C., Van Houtven, C. H., Lee, S. Y., Dilworth-Anderson, P., & Konrad, T. R. (2013). Impact of caring for grandchildren on the health of grandparents in Taiwan. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 68(6), 1009–1021.

Kuh, D., Ben-Shlomo, Y., Lynch, J., Hallqvist, J., & Power, C. (2003).

Life course epidemiology. Journal of Epidemiology and Community Health, 57(10), 778–783.

Lee, S., Guo, W. J., Tsang, A., He, Y. L., Huang, Y. Q., Zhang, M. Y., & Kessler, R. C. (2011). The prevalence of family childhood adversities and their association with first onset of DSM-IV disorders in metropolitan China. Psychological Medicine, 41(1), 85–96. © 2017 Hogrefe Publishing

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

Li, C. Y., Chen, H. F., Sung, F. C., Chen, C. C., Lu, T. H., Yang, C. H., & Ko, M. C. (2010). Offspring birth weight and parental cardiovascular mortality. International Journal of Epidemiology, 39(4), 1082–1090.

Lieb, R., Bronisch, T., Hofler, M., Schreier, A., & Wittchen, H. U. (2005).

Maternal suicidality and risk of suicidality in offspring: findings from a community study. The American Journal of Psychiatry, 162(9), 1665–1671.

Lin, J. J., & Lu, T. H. (2008). Suicide mortality trends by sex, age and method in Taiwan, 1971–2005. BMC Public Health, 8, 6.

Liu, C. Y., Hung, Y. T., Chuang, Y. L., Chen, Y. J., Weng, W. S., Liu, J. S., & Liang, K. Y. (2006). Incorporating development stratification of Taiwan townships into sampling design of large scale health interview survey. Journal of Health Management, 4(1), 1–22.

Lu, T. H., Lee, M. C., & Chou, M. C. (2000). Accuracy of cause-of-death coding in Taiwan: Types of miscoding and effects on mortality statistics. International Journal of Epidemiology, 29(2), 336–343.

McLoughlin, A. B., Gould, M. S., & Malone, K. M. (2015). Global trends in teenage suicide: 2003–2014. QJM, 108(10), 765–780.

Mitchell, A. M., Wesner, S., Brownson, L., Gale, D. D., Garand, L., & Havill, A. (2006). Effective communication with bereaved child survivors of suicide. Journal of Child and Adolescent Psychiatric Nursing, 19(3), 130–136.

Mittendorfer-Rutz, E., Rasmussen, F., & Wasserman, D. (2008).

Familial clustering of suicidal behaviour and psychopathology in young suicide attempters. A register-based nested case control study. Social Psychiatry and Psychiatric Epidemiology, 43(1), 28–36.

National Statistics R.O.C. (Taiwan). (2016). Population and housing census. Retrieved from http://www.stat.gov.tw/np.asp?ctNode=546 Niederkrotenthaler, T., Floderus, B., Alexanderson, K., Rasmussen, F., & Mittendorfer-Rutz, E. (2012). Exposure to parental mortality and markers of morbidity, and the risks of attempted and completed suicide in offspring: An analysis of sensitive life periods. Journal of Epidemiology and Community Health, 66(3), 233–239.

Roberts, J., & Hawton, K. (1980). Child abuse and attempted suicide. The British Journal of Psychiatry, 137, 319–323.

Rostila, M. (2015). Commentary: Childhood parental loss and adulthood health: Discussing the role of parental cause of death, child’s age at death and historical context. Social Science & Medicine, 131, 190–192.

Rostila, M., & Saarela, J. M. (2011). Time does not heal all wounds: Mortality following the death of a parent. Journal of Marriage and Family, 73(1), 236–249.

Shaffer, D. (1988). The epidemiology of teen suicide: An examination of risk factors. Journal of Clinical Psychiatry, 49 Suppl, 36–41.

Taiwan Suicide Prevention Center. (2016). Survey and research. Retrieved from http://tspc.tw/tspc/portal/know/index.jsp?type=2 Tzeng, W. C., & Lipson, J. G. (2004). The cultural context of suicide stigma in Taiwan. Qualitative Health Research, 14(3), 345–358.

Umberson, D., Chen, M. D., House, J. S., Hopkins, K., & Slaten, E. (1996). The effect of social relationships on psychological well-being: Are men and women really so different? American Sociological Review, 61(5), 837–857.

Watson, R. S., & Ebrey, P. B. (1991). Marriage and inequality in chinese society. San Francisco, CA: University of California Press.

Wilcox, H. C., Kuramoto, S. J., Lichtenstein, P., Langstrom, N., Brent, D. A., & Runeson, B. (2010). Psychiatric morbidity, violent crime,

© 2017 Hogrefe Publishing

35

and suicide among children and adolescents exposed to parental death. Journal of the American Academy of Child and Adolescent Psychiatry, 49(5), 514–523.

Received June 8, 2016 Revision received February 16, 2017 Accepted February 16, 2017 Published online July 27, 2017 Dr. Kuan-ying Lee serves as a child psychiatrist in the Department of General Psychiatry, Jianan Psychiatric Center, Ministry of Health and Welfare, Tainan, Taiwan. Her research focuses on adolescent mental health and suicide prevention.

Chung-Yi Li is a professor at the National Taipei College of Nursing and National Chen Kung University. He received his doctorate degree on epidemiology and biostatistics from McGill University. His research expertise includes health risk assessment, health-care services research, epidemiology, and biostatistics.

Dr. Kun-Chia Chang serves as both psychiatrist and director of the Department of General Psychiatry in Jianan Psychiatric Center, MOHW, Taiwan. He is also a candidate for a PhD in public health from the National Cheng-Kung University, Taiwan. His research focuses on health burden, mortality, and quality of life among patients with mental problems.

Dr. Tsung-Hsueh Lu is a professor at National Chen Kung University, Tainan. He is the chair of the Department of Public Health, College of Medicine, National Cheng-Kung University, Tainan, Taiwan. His research focuses on quality of cause-of-death assessment, mortality data, social inequalities in injury, mortality, and suicide methods.

Dr. Ying-Yeh Chen is a professor at National Yang-Ming University, Taiwan, and an attending psychiatrist at Taipei City Psychiatric Center, Taipei City Hospital. She also serves as chief of the research division at Taipei City Suicide Prevention Center. Her research focuses on socioenvironmental influences on suicidal behaviors. She conducts studies on suicide attempters, covering topics on media influences, outcome assessment, and rationale for method choice.

Ying-Yeh Chen Department of General Psychiatry Taipei City Psychiatric Center 309 Songde Road Taipei City, 110 Taiwan ychen@tpech.gov.tw Tsung-Hsueh Lu Department of Public Health College of Medicine National Cheng-Kung University Tainan Taiwan robertlu@mail.ncku.edu.tw

Crisis (2018), 39(1), 27–36

K.-Y. Lee et al., Age at Exposure to Parental Suicide and Subsequent Suicide Risk

36

Appendix A Table A1. The effect of paternal suicide on the risk of offspring suicide, having controlled for family clustering, by sex and age Age at paternal suicide (years)

Adjusted

95% CI

Table A2. The effect of maternal suicide on the risk of offspring suicide, having controlled for family clustering, by sex and age Age at maternal suicide (years)

p

All offspring

Adjusted

95% CI

p

All offspring

0–17

3.82

2.58

5.66

< .0001

0–17

4.47

2.94

6.80

< .0001

18–31

4.03

2.71

5.99

< .0001

18–31

5.29

3.25

8.59

< .0001

All

3.83

2.89

5.07

< .0001

All

4.28

3.11

5.90

< .0001

4.81

3.07

7.54

< .0001

4.84

2.90

8.08

< .0001

Male offspring 0–17

Male offspring 0–17

18–31

3.56

2.14

5.91

< .0001

18–31

4.68

2.55

8.59

< .0001

All

4.11

2.93

5.77

< .0001

All

4.24

2.87

6.28

< .0001

0–17

2.15

0.96

4.81

.0630

0–17

3.83

1.87

7.82

.0002

18–31

4.83

2.53

9.24

< .0001

18–31

6.43

2.83

14.59

< .0001

All

3.32

2.02

5.47

< .0001

All

4.31

2.48

7.48

< .0001

Female offspring

Female offspring

Appendix B Table B1. The effect of parental suicide on the risk of offspring suicide by sex and age Age at father/mother suicide (years)

Exposed group n

b

Rate

Comparison group n

Rateb

Crude HR

95% CI

p

Adjusted HRa

95% CI

p

All offspring 0–17

59

21.81

133

4.92

4.44

3.27

6.04

< .0001

3.69

2.68

5.08

< .0001

18–31

52

78.26

115

17.27

4.53

3.27

6.29

< .0001

4.38

3.13

6.15

< .0001

111

32.94

248

7.36

4.48

3.58

5.61

< .0001

3.79

3.01

4.79

< .0001

44

30.96

88

6.19

5.02

3.49

7.20

< .0001

4.29

2.94

6.28

< .0001

All Male offspring 0–17 18–31

31

90.00

74

21.45

4.20

2.76

6.38

< .0001

3.97

2.59

6.11

< .0001

All

75

42.47

162

9.16

4.64

3.53

6.10

< .0001

3.94

2.96

5.23

< .0001

Female offspring 0–17

15

11.68

45

3.51

3.33

1.86

5.98

< .0001

2.62

1.43

4.79

.0018

18–31

21

65.62

41

12.78

5.14

3.04

8.69

< .0001

5.33

3.07

9.26

< .0001

All

36

22.44

86

5.37

4.19

2.84

6.18

< .0001

3.52

2.34

5.28

< .0001

Note. Reference group is the comparison group who had no exposure to parental suicide. HR = hazard ratio. a Adjusted HR for maternal age, parental education level, and urbanization index. bRate indicates incidence rate per 100,000 person-years.

Crisis (2018), 39(1), 27–36

© 2017 Hogrefe Publishing

Research Trends

Patient-Identified Priorities Leading to Attempted Suicide Life Is Lived in Interpersonal Relationships Niklaus Stulz1,2, Urs Hepp3, Dominic G. Gosoniu4,5, Leticia Grize4,5, Flavio Muheim6, Mitchell G. Weiss4,5, and Anita Riecher-Rössler7 Psychiatric Services Aargau, Brugg, Switzerland University of Bern, Department of Psychology, Bern, Switzerland

1 2

Integrated Psychiatric Services of Winterthur and Zurich Unterland (ipw), Switzerland Swiss Tropical and Public Health Institute, Basel, Switzerland 5 University of Basel, Switzerland 6 University of Basel Psychiatric Clinics, Department of Forensic Psychiatry, Basel, Switzerland 7 University of Basel Psychiatric Hospital, Center for Gender Research and Early Detection, Basel, Switzerland 3 4

Abstract. Background: Attempted suicide is a major public health problem. Aim: The aim of this study was to identify patient-identified problems and triggers typically leading to attempted suicide. Method: A representative sample of 66 adult patients was recruited from all clinical sites and psychiatrists who treat patients after attempted suicide in the Canton of Basel-City (Switzerland). Patients were diagnosed using the Structured Clinical Interview for DSM-IV (SCID) and interviewed with a local adaptation of the Explanatory Model Interview Catalogue (EMIC) to study underlying problems and triggers of attempted suicide. Results: Of the patients, 92.4% had at least one DSM-IV disorder, with depressive disorders being the most prevalent disorder. Although half (50.0%) of the patients identified a health problem, 71.2% identified an interpersonal conflict as underlying problem leading to the suicide attempt. Furthermore, an interpersonal conflict was identified as the trigger of the suicide attempt by more than half of the patients (54.5%). Limitations: The study included German-speaking patients only. Conclusion: According to patients, interpersonal problems often amplify underlying psychiatric problems, leading to suicide attempts. Social and interpersonal stressors should be acknowledged with integrated clinical and social interventions to prevent suicidal behavior in patients and populations.

Keywords: suicide attempt, deliberate self-harm, patient perspective, underlying problems, triggers

Attempted suicide denotes self-inflicted, potentially injurious behavior with a nonfatal outcome for which there is evidence (either explicit or implicit) of intent to die (Silverman, Berman, Sanddal, O’Carroll, & Joiner Jr., 2007a, 2007b). Attempted suicide is a major public health problem (Welch, 2001). Every year worldwide more than 800,000 people die from suicide, and the number of attempted suicides is many times higher (World Health Organization, 2014b). Together with completed suicides, attempted suicides constituted 1.4% of the so-called global burden of disease for the year 2012 (World Health Organization, 2014a).

Suicidal behavior (with or without intention to die), however, is in fact not a disease or disorder in and of itself, even though it may result from mental illness. It is widely asserted that over 90% of suicides are associated with mental illness, with depression and alcohol use disorders being particularly prevalent in people who completed suicide (Bertolote & Fleischmann, 2002a, 2002b; Bertolote, © 2017 Hogrefe Publishing

Fleischmann, De Leo, & Wasserman, 2004; Mann, 2002).

Research on suicidal behavior accordingly has a strong focus on the relationship between psychopathology and suicide (Bertolote, Fleischmann, De Leo, & Wasserman, 2003; De Leo, 2004; Knox, Conwell, & Caine, 2004), and some prevention programs seem to rely solely on psychiatric risk factors as a guide to population-based suicide prevention: “Because most people who commit suicide have a mental disorder […], suicide rates indicate potential need for mental health care” (US Department of Health and Human Services, 2001). Even though mental health care was demonstrably effective in preventing suicide for at-risk populations, comprehensive prevention measures should also account for psychosocial factors that may affect suicidal behavior. Mental disorders undoubtedly are a major risk factor for suicide, and the focus on psychopathological markers may help to explain and prevent suicidal behavior, but the majority of people with mental disorders do not attempt or complete suicide. Most of the data emphasizing Crisis (2018), 39(1), 37–46 https://doi.org/10.1027/0227-5910/a000473

38

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

the strong link between mental illness and suicidal behavior originated from Euro-American studies (Bertolote et al., 2004) and many of these studies were performed with subjects who had been admitted to psychiatric hospitals following their suicide attempt (Bertolote & Fleischmann, 2002a). This may have resulted in an overestimation of the global significance of psychiatric risk factors for suicidal behavior in the general population. In fact, experiences from Asian countries (India, China) suggest that cultural values and situational stressors may likewise play an important role in suicidal behavior (Eddleston & Phillips, 2004; Parkar, Nagarsekar, & Weiss, 2012). In prior studies we found that migration may also play an important role (Brückner, Muheim, Berger, & Riecher-Rössler, 2011; Yilmaz & Riecher-Rössler, 2008, 2012), and it is quite well known that gender-associated factors massively influence suicidal behavior, with women having a much higher rate of attempted suicide and men having a much higher rate of completed suicide (Canetto & Sakinofsky, 1998; Weissman et al., 1999). Suicidal behavior therefore became increasingly recognized as a complex public health problem involving psychological, social, biological, cultural, and environmental factors (Bertolote, 2004; De Leo, 2004; Hammond, 2001; Knox et al., 2004; Phillips et al., 2002; Welch, 2001).

In this study, we aimed to clarify the relative role of psychopathology and of situational and contextual stressors (e.g., interpersonal, health, work or financial problems) for suicidal behavior, pursuant to reports of patients who attempted suicide. The identification of patient-identified priorities leading to a suicide attempt in a sample from a qualitative survey may have important implications for both theory building and very practical considerations about ways of approaching suicide prevention and mental health policy. Patients’ explanations of attempted suicide may affect clinical practice and interventions as well as all levels of prevention: universal, selective, and indicated (Bertolote, 2004).

source of recruitment. Further sources were the Psychiatric University Clinics as the main psychiatric treatment center in the area, the hospitals surrounding Basel-City, and all psychiatrists in private practice in Basel-City. All of these sites were already referring patients to the clinical epidemiological WHO/EURO Multicentre Study on Suicidal Behavior (Muheim et al., 2013); hence, a mechanism for recruitment was already established and functioning effectively at the beginning of the current study.

Upon admission at all clinical sites, suicide attempters were regularly seen and assessed by a psychiatrist. The psychiatrist obtained basic sociodemographic and clinical data of the patient as well as information about the suicide attempt (e.g., the method) using the brief epidemiological clinical interview protocol of the WHO/EURO Multicentre Study on Suicidal Behavior in the Canton Basel-City (Muheim et al., 2013). The results of this representative study, which covered the period between January 2003 and December 2006, have been published (Muheim et al., 2013).

For our study, the patients of the WHO/EURO Multicentre Study were invited to take part in additional interviews to examine the reasons for the suicide attempt in more detail.

If patients agreed, they were interviewed by a member of our study team as soon as possible after their suicide attempt that had led to medical care.

Inclusion criteria for the current study were minimum age of 18 years, proficiency in German to participate in the research interviews, and written informed consent. In accordance with the definition of parasuicide guiding recruitment in the WHO/EURO Multicentre study (Platt et al., 1992), patients whose current behavior was characterized by the responsible psychiatrist as an instance of a habitual pattern of self-harming behavior, rather than an acute suicide attempt, were excluded from the current study.

For patients who declined participation or who did not show up for the scheduled interview, at minimum sociodemographic and basic clinical data, which were recorded for the WHO/EURO Multicentre Study and which did not rely on the informed consent and participation of the patient (Muheim et al., 2013), were available and were used for comparison with our study group.

The current study was approved by the local ethics committee and conducted according to the declaration of Helsinki.

Method Procedures All patients living in the Canton of Basel-City (Switzerland) with suicide attempts between November 2005 and February 2007 were contacted and interviewed after giving their consent. Participants were recruited from all clinical sites that were treating patients after attempted suicide in the Canton of Basel-City. The emergency unit of the University Hospital Basel, where most patients in need of medical care after suicide attempts are seen, was the main Crisis (2018), 39(1), 37–46

Assessments The Structured Clinical Interviews for DSM-IV Disorders (SCID-I and SCID-II) are considered the gold standard test for determining DSM-IV Axis I (major mental) disorders and Axis II (personality) disorders (First, Gibbon, Spitzer, & Williams, 1997; First, Spitzer, Gibbon, & Wil© 2017 Hogrefe Publishing

39

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

liams, 1997). Following the assessments with the SCID in the first appointment, a locally adapted explanatory model interview based on the framework of the Explanatory Model Interview Catalogue (EMIC; Weiss, 1997) was administered by the same member of the study group in a second appointment to elicit patients’ explanations and the sociocultural contexts of their suicide attempts. This EMIC interview includes questions about circumstances leading to suicide attempts (underlying problems and triggers) and about patterns of personal distress (including symptoms), social distress (including stigma), prior help-seeking, and perceived causes of attempted suicide.

Each section of the interview includes coding items to categorize the answers of the respondent. The interviewer codes a patient’s response on an extensive list of problems and triggers, that is, on an extensive list of various stressors that include specific interpersonal or social problems.

Narrative elaboration is included in the data set to explain these problems. The local adaption of the EMIC interview used here was based on the structure of other EMIC interviews for study of attempted suicide (Chowdhury et al., 2001; Parkar, Dawani, & Weiss, 2006, 2008).

Data Analysis First, sociodemographic data and the method of the suicide attempt were compared between the study sample and the patients who did not participate in our study. If a participant had multiple suicide attempts during the study period, only the event when he or she was also interviewed for this study was considered. If nonparticipants had multiple suicide attempts during the study period, the data of the most recent event were drawn from the WHO/EURO Multicentre Study database for comparison with the respective data in our study sample. Statistical tests used for group comparisons were Fisher‘s exact test, the χ2 test, and the Mann–Whitney U test.

Subsequently, we performed descriptive statistical analyses of the SCID-I and SCID-II data and of the coded quantitative variables of the EMIC interview on underlying problems and triggers of the suicide attempts.

Finally, some additional exploratory analyses were performed using χ2 tests to explore the role of age, gender, and nationality with respect to the patient-identified priorities leading to suicidal behavior.

All analyses were performed using SPSS, Version 18 (SPSS Inc., 2009).

© 2017 Hogrefe Publishing

Results Sample During the study period, 215 patients who fulfilled the inclusion criteria for the current study presented at the study sites after one or multiple suicide attempts. Of these patients, 66 (30.7%) agreed to participate in our study. The EMIC assessment (second interview) was performed on average 14.5 days (SD = 40.1) after the suicide attempt (Mdn = 5, range = 1–276). Participants in our study did not differ from the 149 nonparticipants with regard to available sociodemographics and the method of the suicide attempt (Table 1).

Mental Disorders Mental disorders were highly prevalent in our study sample: 92.4% of the patients had at least one DSM-IV Axis I disorder, with depressive disorders (53.0%) being the most prevalent main diagnosis (see Table 2). In all, 30 (45.5%) of the patients had at least one further diagnosis on Axis I; substance use disorders were the most frequent secondary diagnosis. On Axis II, borderline personality disorders (22.7%) were most prevalent among the patients who were in need of medical care after attempted suicide (see Table 2).

Characteristics of the Suicide Attempts Intoxication by drug overdose was the most frequent method of attempted suicide, accounting for more than three fourths (77.2%) of the incidents in our sample (Table 1). The vast majority of the patients (68.2%) reported that they attempted suicide impulsively (“Did you have a plan? Or would you say it happened impulsively?”); only relatively few of the suicide attempts were planned (16.7%) or at least partially planned (15.2%), pursuant to patients’ reports.

Patient-Identified Problems and Triggers Leading to Attempted Suicide Table 3 shows the distribution of the patient-identified underlying problems leading to suicide attempts (“What in your life do you believe has led to this event? What are the problems which led to this event?”), and the distribution of the patient-identified triggers (“What do you believe was the trigger for this event?”). The patients’ narratives often Crisis (2018), 39(1), 37–46

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

40

Table 1. Sample characteristics Characteristic Age (in years); M(SD)a

Total sample (n = 215)

Nonparticipants (n = 149)

Participants (n = 66)

p

40.1 (17.1)

41.4 (17.8)

37.0 (14.5)

.135

64 (31.4)

48 (32.9)

16 (27.6)

.463

140 (68.6)

98 (67.1)

42 (72.4)

100 (50.3)

60 (44.8)

40 (61.5)

Sex; n (%)b Male Female Marital status; n (%)

c

Single Married

53 (26.6)

39 (29.1)

14 (21.5)

Divorced or separated

38 (19.1)

28 (20.9)

10 (15.4)

8 (4.0)

7 (5.2)

1 (1.5)

Yes

91 (54.5)

63 (54.8)

28 (53.8)

No

76 (45.5)

52 (45.2)

24 (46.2)

Swiss

137 (67.5)

94 (67.1)

43 (68.3)

Other

66 (32.5)

46 (32.9)

20 (31.7)

Widowed Cohabitation; n (%)

.158

d

.910

Citizenship; n (%)e

Employment status; n (%)f

.876

.337

Employed

50 (28.7)

31 (25.6)

19 (35.8)

Homemaker

12 (6.9)

9 (7.4)

3 (5.7)

Education

23 (13.2)

13 (10.7)

10 (18.9)

Unemployed

27 (15.5)

18 (14.9)

9 (17.0)

Disability benefits

41 (23.6)

33 (27.3)

8 (15.1)

Old age pension

10 (5.7)

8 (6.6)

2 (3.8)

Other

11 (6.3)

9 (7.4)

2 (3.8)

160 (77.7)

116 (77.9)

44 (77.2)

8 (3.9)

5 (3.4)

3 (5.3)

Method of attempted suicide; n (%)a Intoxication Hanging Drowning

3 (1.5)

3 (2.0)

0

Firearms

1 (0.5)

0

1 (1.8)

Burning

5 (2.4)

4 (2.7)

1 (1.8)

Cutting

24 (11.7)

17 (11.4)

7 (12.3)

2 (1.0)

2 (1.3)

0

Railway

2 (1.0)

2 (1.3)

0

Car accident

1 (0.5)

0

1 (1.8)

Jumping from heights

.528

Note. aNine patients with missing data; b11 patients with missing data; c16 patients with missing data; d48 patients with missing data; e12 patients with f missing data; 41 patients with missing data.

described life stories with close connections between various problems over a prolonged period of time. Thus, when asked about the priorities leading to attempted suicide, the patients’ answers often were an accumulation of various problems (25.8%), otherwise referred to as “everything came together.”

However, interpersonal conflicts were most often identified by the patients as one of the underlying problems that led to the suicide attempt (71.2%), and 28.8% of the Crisis (2018), 39(1), 37–46

patients identified an interpersonal conflict as the main underlying problem. Additionally, more than half of the patients (54.5%) identified an interpersonal conflict that, in their view, had triggered the suicide attempt. The narratives of these patients described experiences of abandonment, rejection, loneliness, lack of understanding, and dependencies as part of their life story. In particular, the patients frequently described problems of communication with their partner (see Table 3) that pointed to conflicts in being able © 2017 Hogrefe Publishing

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

Table 2. Diagnoses (DSM-IV; n = 66) Diagnosis

n

%

Axis I: Main diagnosis Depressive disorder

35

53.0

Bipolar disorder

5

7.6

Schizophrenia / schizoaffective disorder

5

7.6

Psychotic disorder

1

1.5

Substance use disorder

6

9.1

Anxiety disorder

4

6.1

Obsessive-compulsive disorder

1

1.5

Adjustment disorder

2

3.0

Posttraumatic stress disorder

1

1.5

Dementia

1

1.5

No disorder

5

7.6

Histrionic personality disorder

1

1.5

Dependent personality disorder

1

1.5

Narcissistic personality disorder

1

1.5

Borderline personality disorder

15

22.7

No personality disorder

48

72.7

Axis II

to negotiate their needs within a relationship. For instance, a 40-year-old woman who had been married four times resorted to suicide attempts each time the relationship failed.

In the current case that woman attempted suicide after her partner, despite promising to do so, failed to accompany her to the graveyard of her son on the son’s death day: The partner only is concerned about himself. They do what they want, and this is all great, but it is bad when I raise my wishes and do things I want to do. When I want to talk with him, it is not possible. But in a relationship it is the most important thing that you talk to each other, isn’t it? To bring up the problems, otherwise all that will not work …

Health problems, in particular mental disorders, were likewise frequently identified by the patients as underlying problems and triggers of suicide attempts, but (mental) health problems were mentioned less often than interpersonal stressors (Table 3). Moreover, even those patients who accepted the relevance of their mental illness often specified an interpersonal component that led to or triggered their suicide attempts. For instance, a 28-year-old female immigrant explained: The doctor first listed several symptoms, which would account for a depression. All applied to me and made sense. I could accept the diagnosis and wanted to see if I can overcome it. … But for me, it was really everything together, while the death of my mother was probably the main problem. I always say that if she would have lived, everything would be different. I always say, that I had such a great life and now it turned so badly. © 2017 Hogrefe Publishing

41

The family of this patient had collapsed after her father had killed her mother. The patient found solace in a relationship, yet a miscarriage that threatened the relationship triggered her suicide attempt. Note that this patient furthermore described her frustration with the clinical treatment that exclusively focused on the diagnosis of depression.

The Role of Age, Gender, and Nationality in ­Patient-Identified Problems and Triggers We finally performed additional exploratory analyses to explore the role of age, gender, and nationality in the patient-identified priorities leading to suicidal behavior.

When compared with men, female suicide attempters more often identified problems with their parents to be one of the existing problems in their life (Fisher’s exact test: p = .015). By contrast, men more often named a physical illness to be an issue in their life that led to the suicide attempt (Fisher’s exact test: p = .014).

Furthermore, when compared with their older counterparts, patients younger than 35 years (median-split) more frequently reported interpersonal problems (χ2 = 15.391; df = 1; p < .001), particularly problems with their partners (χ2 = 3.956; df = 1; p < .001) and with their parents (χ2 = 16.176; df = 1; p < .001), to be among the priorities leading to the suicide attempt. Younger patients also more often mentioned stress at school (Fisher’s exact test: p = .023) and at work (χ2 = 10.949; df = 1; p = .001), as well as problems in their work environment in general (χ2 = 3.957; df = 1; p = .047), than their older counterparts did, while the latter more frequently reported health issues as an underlying problem leading to the suicide attempt (χ2 = 3.882; df = 1; p = .049).

Finally, when compared with Swiss citizens, immigrants more frequently reported work-related stressors (Fisher’s exact test: p = .028) to be among the priorities leading to the suicide attempt.

Apart from these relationships between underlying problems and sociodemographics, there were no differences between male and female suicide attempters or between older and younger patients or Swiss citizens and immigrants, neither with respect to underlying problems nor with respect to the most important problem or to the trigger of the suicide attempt (detailed statistics are not reported here due to space limitations). Note that even the aforementioned relationships between sociodemographics and underlying problems should be considered with caution since the statistically significant results of these exploratory analyses would not withstand Bonferroni correction to control for alpha error inflation due to multiple testing in the sample.

Crisis (2018), 39(1), 37–46

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

42

Table 3. Patient-identified problems and triggers of deliberate self-harm events (n = 66) Underlying problems All identifieda

Triggersa

Main identified

EMIC variables: triggers and problems

n

%

n

%

n

%

Interpersonal conflict

47

71.2

19

28.8

36

54.5

Partner

36

53.0

16

24.2

30

45.5

Parents

18

27.3

3

4.5

4

6.1

Children

6

9.1

0

0.0

0

0.0

Other family

3

4.5

0

0.0

4

6.1

Other interpersonal

5

7.6

0

0.0

3

4.5

33

50.0

14

21.2

11

16.7

Physical and mental health Pain

4

6.1

1

1.5

1

1.5

Physical illness

7

10.6

2

3.0

1

1.5

Mental illness

24

36.4

9

13.6

7

10.6

Substance abuse (self)

7

10.6

2

3.0

4

6.1

Substance abuse (others)

3

4.5

0

0.0

0

0.0

20

30.3

6

9.1

2

3.0

15

22.7

6

9.1

1

1.5

Social distress Financial problems School Work problems Mobbing Work stress

5

7.6

0

0.0

1

1.5

23

34.8

4

6.1

4

6.1

0

0.0

0

0.0

1

1.5

12

18.2

3

4.5

3

4.5

Work insecurity

6

9.1

1

1.5

0

0.0

Unemployment

5

7.6

0

0.0

0

0.0

Other work problems

1

1.5

0

0.0

0

0.0

10

15.2

1

1.5

4

6.1

Sexual abuse

4

6.1

1

1.5

1

1.5

Physical abuse

4

6.1

0

0.0

1

1.5

Victimization

Verbal abuse

4

6.1

0

0.0

3

4.5

14

21.2

3

4.5

2

3.0

10

15.2

3

4.5

1

1.5

Suicide

2

3.0

0

0.0

0

0.0

Other event

2

3.0

0

0.0

1

1.5

17

25.8

12

18.2

0

0

0

0

0

0

4

6.1

Events before the suicide attempts Death

Accumulation of problems/triggers Nothing Don’t know

0

0

0

0

2

3.0

Other

6

9.1

0

0

6

9.1

Missing

0

0

7

10.6

0

0

Note. Multiple answers possible. a

Discussion Corroborating findings from previous studies, mental disorders – particularly depressive and substance use disorders – were found to be highly prevalent in patients who attempted suicide (Bertolote & Fleischmann, 2002a; BerCrisis (2018), 39(1), 37–46

tolote et al., 2004; Mann, 2002; Muheim et al., 2013). In all, 92.4% of our patients in need for medical care following attempted suicide were diagnosed with at least one DSMIV Axis I mental disorder. Nearly half of them (47.0%) identified a mental health issue including substance use as an existing problem in their life that had led to the suicide © 2017 Hogrefe Publishing

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

43

attempt. However, the patients in our sample identified a multitude of other problems, particularly interpersonal conflicts, which in their view led to the suicide attempts: 71.2% of the patients mentioned an interpersonal conflict as one of the underlying problems, 28.8% identified such conflict as the most important underlying problem, and more than half of the patients (54.5%) named an interpersonal conflict as the trigger of the suicide attempt. This trigger function of interpersonal conflicts seems to require particular attention given that in most of the cases the suicide attempt happened intuitively without long hours of premeditation (68.2%) or was at least only partially planed (15.2%; Simon et al., 2001).

While denoting the important role of psychopathology in the genesis of suicide attempts, our findings suggest that clinicians and other people concerned with suicide prevention should pay great attention to social, and particularly interpersonal, conflicts and stressors with regard to the prediction and prevention of suicidal behavior. The crucial role of social factors in the advent of suicidal behavior has already been reported in previous studies from Asian countries (Eddleston & Phillips, 2004; Parkar et al., 2012) and among Turkish immigrants in Switzerland (Brückner et al., 2011; Yilmaz & Riecher-Rössler, 2008, 2012). In line with this are the recently emerging views that suicidal behavior constitutes a complex public health problem, with psychological, social, biological, cultural, and environmental factors being involved. If recognized as a multicausal incident, then suicidal behavior calls for integrated clinical and population-based prevention approaches (Bertolote, 2004; De Leo, 2004; Hammond, 2001; Knox et al., 2004; Phillips et al., 2002; Welch, 2001).

On a public health level, the social and cultural dimensions of suicide attempts point to the need of a broader framework for prevention strategies (Bertolote et al., 2003; Knox et al., 2004). Instead of exclusively targeting psychopathology, prevention strategies might be improved by including assessments of larger social and cultural problems, such as childhood maltreatment, alcoholism and substance abuse, homosexuality, or unemployment and poverty, particularly when they are highly stigmatized and thus create barriers to relate to others.

Likewise caution is warranted on a clinical level not to overemphasize the role of psychopathology for suicidal behavior. Ignoring patient-identified social factors related to suicidal behavior may lead to a bias in risk prediction and treatment of suicidal behavior as well as in deficient therapeutic attitudes. Due to the crucial role of interpersonal conflicts in the development of suicidal behavior pursuant to the reports of our patients, the treatment of suicidal behavior should have a strong focus on interpersonal relationships and problems. A promising intervention might be provided by the Attempted Suicide Short

Intervention Program (ASSIP), which focuses on the patient’s view of the suicidal event and is currently under evaluation (Gysin-Maillart & Michel, 2013). Focusing on interpersonal conflicts might be particularly important in younger women (whereas subjective suffering from physical illness seems to be a pronounced risk factor in older men). The perpetuation of suffering as a result of relationships described in terms of lack of understanding, lack of time, and rejection that finally led to the suicidal behavior must be avoided in the therapeutic setting under all circumstances. Psychiatric treatment must not be perceived by the patients to have the same characteristics as the unsatisfactory or disappointing interpersonal relationships in everyday life. In this regard, our findings support views that the therapeutic alliance is more important than any other treatment approach in the treatment of self-injurious behavior (Trepal, 2010). The significance of interpersonal relationships for the formation of and recovery from mental illness has been documented for both therapeutic and everyday settings (Beach & Kaslow, 2006; Denton, 2007; Lewis, 2000). While the need for a better understanding of dyadic processes in research and clinical practice is often discussed (Borrell-Carrio, Suchman, & Epstein, 2004), some go even further and argue for the centrality of relational processes to understand and treat mental disorders (Lewis, 1998). Consequently, some authors argued for the inclusion of relational disorders into the DSM-5 in order to guide mental health professionals in ways to more efficiently treat mental disorders (Beach & Kaslow, 2006; Beach, Wamboldt, Kaslow, Heyman, & Reiss, 2006; First, 2006). In any case, when assessing the risk of self-harm and suicide in patients, clinicians might be well advised to systematically ask about interpersonal problems and conflicts. Physical illness and associated coping strategies seem to be further important issues that merit the attention of clinicians in order to prevent suicidal behavior, particularly in older men. In immigrants work-related stressors might be particularly important regarding suicidal behavior. If these findings and the results from a recent meta-analysis showing that approximately half of all people who complete suicide communicate their suicidal intentions prior to their suicide (Pompili et al., 2016) are taken into account, this may help to prevent suicidal behavior in clinical settings.

© 2017 Hogrefe Publishing

Limitations and Future Perspectives Several limitations of this study have to be addressed.

First, we recruited our participants from health facilities where they were seeking help after attempted suicide. The multidisciplinary emergency unit of the University HospiCrisis (2018), 39(1), 37–46

44

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

tal Basel was the main source of recruitment. However, the remaining participants were mainly recruited from mental health facilities. This might have led to an overestimation of the prevalence of mental disorders in suicide attempters and hence might have hindered testing the impact of psychosocial factors independently from mental disorders.

Second, we only included patients with sufficient proficiency in German to participate in research interviews.

We hence do not know whether our findings would apply to foreign-language suicide attempters. The significance of social factors for the emergence of suicide attempts, however, has also been demonstrated in Turkish immigrants in Switzerland (Brückner et al., 2011; Yilmaz & Riecher-Rössler, 2008, 2012) and in studies from other (non-Western) countries (Eddleston & Phillips, 2004; Parkar et al., 2012).

Third, of the 215 patients who fulfilled the inclusion criteria for the current study, only 66 (30.7%) agreed to participate. Although these participants did not differ from the 149 nonparticipants in the available sociodemographics and in the method of the suicide attempt, a selection bias is still possible.

Fourth, for 11 (16.7%) of the patients SCID data were lacking. For these patients, ICD-10 diagnoses given by the psychiatrists seeing the patients after their suicide attempts were coded. Thus, the diagnoses of these patients without structured clinical interview may be overestimated in the absence of systematic criteria-based assessment with the SCID. Fifth, the exploratory analyses on the role of age, gender, and nationality should be considered preliminary owing to the small sample size. Addressing these questions provided information about the available sample from a qualitative survey, rather than a rigorous and generalizable test of hypotheses concerning age, gender, and nationality.

This should be the aim of further research.

Finally, it would have been interesting to see whether particular patient-identified problems or triggers of attempted suicides were associated with certain mental disorders. We desisted from performing these analyses in the current study owing to the small diagnostic subgroups in our sample. However, answering this question will be an important task for future research. This also applies to the examination of further characteristics of suicide attempters (e.g., sociodemographics such as marital status or education) or to characteristics of the suicide attempt (e.g., the method of the suicide attempt) that might be related to patient-identified priorities leading to suicidal behavior. Knowledge on such associations would help to identify high-risk constellations (e.g., that young depressive girls are particularly prone to suicide attempts in the case of interpersonal conflicts with their parents) and hence to

optimally tailor suicide prevention measures to individuals and subpopulations.

Crisis (2018), 39(1), 37–46

Conclusion The high prevalence of mental disorders among patients in need for medical care after attempted suicides reflects the important role of psychopathology in the development of suicidal behavior. From the patients’ point of view, however, there are often interpersonal stressors and problems that magnify underlying psychiatric problems, leading to attempted suicide. Interactions of social, and particularly interpersonal, stressors with psychopathology thus seem to be crucial with regard to the prevention and treatment of suicidal behavior. Overall, our findings stress the importance of taking into account patient explanations of suicidal behavior to improve the assessment and prevention of suicidal behavior in both clinical and public health settings (Flanagan, Davidson, & Strauss, 2007).

Acknowledgments This study was supported by the Swiss National Science Foundation (SNSF) Grant #3200 BO-105913 awarded to Mitchell Weiss and Anita Riecher-Rössler. The SNSF had no involvement in the collection, analysis, and interpretation of the data, in the writing of the report, and in the decision to submit the article for publication. We also thank all patients who took part in these personal interviews. Last but not least we thank Claudia Sauerborn and Cornelia Kneser for their support in data assessment and management.

The authors have no conflict of interest to report.

References Beach, S. R., & Kaslow, N. J. (2006). Relational disorders and relational processes in diagnostic practice: Introduction to the special section. Journal of Family Psychology, 20(3), 353–355. doi:10.1037/0893-3200.20.3.353 Beach, S. R., Wamboldt, M. Z., Kaslow, N. J., Heyman, R. E., & Reiss, D. (2006). Describing relationship problems in DSM-V: Toward better guidance for research and clinical practice. Journal of Family Psychology, 20(3), 359–368. doi:10.1037/0893-3200.20.3.359 Bertolote, J. M. (2004). Suicide prevention: At what level does it work? World Psychiatry, 3(3), 147–151.

Bertolote, J. M., & Fleischmann, A. (2002a). Suicide and psychiatric diagnosis: A worldwide perspective. World Psychiatry, 1(3), 181–185.

Bertolote, J. M., & Fleischmann, A. (2002b). Suicide rates in China.

Lancet, 359(9325), 2274; author reply 2274–2275. doi:10.1016/ S0140-6736(02)09268-1 Bertolote, J. M., Fleischmann, A., De Leo, D., & Wasserman, D. (2003). Suicide and mental disorders: Do we know enough? British Journal of Psychiatry, 183, 382–383. © 2017 Hogrefe Publishing

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

Bertolote, J. M., Fleischmann, A., De Leo, D., & Wasserman, D. (2004). Psychiatric diagnoses and suicide: Revisiting the evidence. Crisis, 25(4), 147–155.

Borrell-Carrio, F., Suchman, A. L., & Epstein, R. M. (2004). The biopsychosocial model 25 years later: Principles, practice, and scientific inquiry. Annals of Family Medicine, 2(6), 576–582. doi:10.1370/afm.245 Brückner, P., Muheim, F., Berger, P., & Riecher-Rössler, A. (2011).

[Characteristics of suicide attempts in immigrants from Turkey in the Canton Basel-Stadt: Results from the WHO/EURO-Multicentre Study]. Nervenheilkunde, 7, 517–522.

Canetto, S. S., & Sakinofsky, I. (1998). The gender paradox in suicide. Suicide and Life-Threatening Behavior, 28(1), 1–23.

Chowdhury, A. N., Sanyal, D., Dutta, S. K., De, R., Banerjee, S., Bhattacharya, K., … Weiss, M. G. (2001). Interrater reliability of the EMIC in a pilot study in West Bengal. International Medical Journal, 8, 25–29.

De Leo, D. (2004). Suicide prevention is far more than a psychiatric business. World Psychiatry, 3(3), 155–156.

Denton, W. H. (2007). Issues for DSM-V: Relational diagnosis: An essential component of biopsychosocial assessment. American Journal of Psychiatry, 164(8), 1146–1147. doi:10.1176/appi. ajp.2007.07010181 Eddleston, M., & Phillips, M. R. (2004). Self poisoning with pesticides.

BMJ (Clinical Research Ed.), 328(7430), 42–44. doi:10.1136/ bmj.328.7430.42 First, M. B. (2006). Relational processes in the DSM-V revision process: Comment on the special section. Journal of Family Psychology, 20(3), 356–358. doi:10.1037/0893-3200.20.3.356 First, M. B., Gibbon, M., Spitzer, R. L., & Williams, J. B. W. (1997).

Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). Washington, DC: American Psychiatric Association.

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1997). Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I), Clinical Version. Washington, DC: American Psychiatric Association.

Flanagan, E. H., Davidson, L., & Strauss, J. S. (2007). Issues for DSM-V: Incorporating patients’ subjective experiences. American Journal of Psychiatry, 164(3), 391–392. doi:10.1176/ajp.­ 2007.164.3.391 Gysin-Maillart, A., & Michel, K. (2013). Brief therapy after attempted suicide. ASSIP – Attempted Suicide Short Intervention Program. Bern, Switzerland: Hans Huber.

Hammond, W. R. (2001). Suicide prevention: Broadening the field toward a public health approach. Suicide and Life-Threatening Behavior, 32(1 Suppl), 1–2.

Knox, K. L., Conwell, Y., & Caine, E. D. (2004). If suicide is a public health problem, what are we doing to prevent it? American Journal of Public Health, 94(1), 37–45.

Lewis, J. M. (1998). For better or worse: Interpersonal relationships and individual outcome. American Journal of Psychiatry, 155(5), 582–589.

Lewis, J. M. (2000). Repairing the bond in important relationships: A dynamic for personality maturation. American Journal of Psychiatry, 157(9), 1375–1378.

Mann, J. J. (2002). A current perspective of suicide and attempted suicide. Annals of Internal Medicine, 136(4), 302–311.

Muheim, F., Eichhorn, M., Berger, P., Czernin, S., Stoppe, G., Keck, M., & Riecher-Rössler, A. (2013). Suicide attempts in the county of Basel: results from the WHO/EURO Multicentre Study on Suicidal Behaviour. Swiss Medical Weekly, 143, w13759. doi:10.4414/smw.2013.13759 Parkar, S. R., Dawani, V., & Weiss, M. G. (2006). Clinical diagnostic and sociocultural dimensions of deliberate self-harm in Mumbai, India. Suicide and Life-Threatening Behavior, 36(2), 223–238. doi:10.1521/suli.2006.36.2.223 © 2017 Hogrefe Publishing

45

Parkar, S. R., Dawani, V., & Weiss, M. G. (2008). Gender, suicide, and the sociocultural context of deliberate self-harm in an urban general hospital in Mumbai, India. Culture, Medicine and Psychiatry, 32(4), 492–515. doi:10.1007/s11013-008-9109-z Parkar, S. R., Nagarsekar, B. B., & Weiss, M. G. (2012). Explaining suicide: Identifying common themes and diverse perspectives in an urban Mumbai slum. Social Science & Medicine, 75(11), 2037–2046. doi:10.1016/j.socscimed.2012.07.002 Phillips, M. R., Yang, G., Zhang, Y., Wang, L., Ji, H., & Zhou, M. (2002).

Risk factors for suicide in China: A national case-control psychological autopsy study. Lancet, 360(9347), 1728–1736. doi:10.1016/S0140-6736(02)11681-3 Platt, S., Bille-Brahe, U., Kerkhof, A., Schmidtke, A., Bjerke, T., Crepet, P., … Sampaio Faria, J. (1992). Parasuicide in Europe: The WHO/EURO multicentre study on parasuicide. I. Introduction and preliminary analysis for 1989. Acta Psychiatrica Scandinavica, 85(2), 97–104.

Pompili, M., Belvederi Murri, M., Patti, S., Innamorati, M., Lester, D., Girardi, P., & Amore, M. (2016). The communication of suicidal intentions: A meta-analysis. Psychological Medicine, 46, 2239– 2253.

Silverman, M. M., Berman, A. L., Sanddal, N. D., O’Carroll, P. W., & Joiner Jr., T. E. (2007a). Rebuilding the Tower of Babel: A revised nomenclature for the study of suicide and suicidal behaviors Part 1: Background, rationale, and methodology. Suicide and Life-Threatening Behavior, 37(3), 248–263.

Silverman, M. M., Berman, A. L., Sanddal, N. D., O’Carroll, P. W., & Joiner Jr., T. E. (2007b). Rebuilding the Tower of Babel: A revised nomenclature for the study of suicide and suicidal behaviors Part 2: Suicide-Related ideations, communications, and behaviors. Suicide and Life-Threatening Behavior, 37(3), 264–277.

Simon, O. R., Swann, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., & O’Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide and Life-Threatening Behavior, 32(1 Suppl), 49–59.

SPSS Inc. (2009). PASW Statistics for Windows (Version 18.0) [Computer software]. Chicago, IL: SPSS Inc. Trepal, H. C. (2010). Exploring self-injury through a relational cultural lens. Journal of Counseling & Development, 88(4), 492–499.

US Department of Health and Human Services. (2001). Mental health: culture, race, and ethnicity – a supplement to mental health: a report of the surgeon general. Rockville, MD: US Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services.

Weiss, M. G. (1997). Explanatory Model Interview Catalogue: Framework for comparative study of illness experience. Transcultural Psychiatry, 34(2), 235–263.

Weissman, M. M., Bland, R. C., Canino, G. J., Greenwald, S., Hwu, H. G., Joyce, P. R., … Yeh, E. K. (1999). Prevalence of suicide ideation and suicide attempts in nine countries. Psychological Medicine, 29(1), 9–17.

Welch, S. S. (2001). A review of the literature on the epidemiology of parasuicide in the general population. Psychiatric Services, 52(3), 368–375.

World Health Organization. (2014a). Disease burden 2014. Retrieved from http://www.who.int/healthinfo/global_burden_disease/ estimates/en/index2.html World Health Organization. (2014b). Preventing suicide: A global imperative, 2015. Retrieved from http://www.who.int/mental_ health/suicide-prevention/world_report_2014/en/ Yilmaz, T. A., & Riecher-Rössler, A. (2008). Suicide attempts among first and second generation immigrants. Neuropsychiatrie, 22(4), 261–267.

Yilmaz, T. A., & Riecher-Rössler, A. (2012). Attempted suicide in immigrants from Turkey: A comparison with Swiss suicide attempters. Psychopathology, 45, 366–373.

Crisis (2018), 39(1), 37–46

46

N. Stulz et al., Patient-Identified Priorities and Attempted Suicide

Received July 13, 2016 Revision received March 10, 2017 Accepted March 11, 2017 Published online August 10, 2017

Flavio Muheim is a psychological scientist working in the public health sector performing awareness campaigns in the prevention of stress, affective disorders, and suicide. He finished his PhD in the WHO/EURO Multicentre Study on suicide attempts and currently works as psychotherapist.

Niklaus Stulz is a psychologist working in epidemiological and clinical studies. He is head of the research unit at the Psychiatric Services Aargau, member of the Department of Psychology at the University of Bern, and lecturer at the University of Zurich, Switzerland.

Mitchell Weiss is a health social scientist and psychiatrist working at the interface of public health and medical anthropology and contributing to cultural epidemiology and methods for sociocultural autopsy. He is Professor Emeritus at the Swiss TPH and University of Basel, Switzerland, and a corresponding member of the faculty at Harvard Medical School, USA.

Urs Hepp is Medical Director of the Integrated Psychiatric Services of Winterthur and Zurich Unterland (ipw). He is Professor of Psychiatry and Psychotherapy at the University of Zurich, Switzerland. His research on suicide prevention has focused on restriction of means.

Dominic Gabriel Gosoniu was a statistician working mainly in epidemiological and preclinical studies. He teaches mathematics at different schools in Vaud Canton. His research focused on developing Bayesian geostatistical methods for analyzing large epidemiological data to produce high-resolution mortality risk maps.

Leticia Grize is a public health scientist working in epidemiological and clinical studies. She is part of the biostatistics unit at the Swiss TPH and provides internal and external services in data management and statistical analytics. She is a member of the International Statistical Institute.

Crisis (2018), 39(1), 37–46

Anita Riecher-Rössler is Professor of Psychiatry and Head of the Center for Gender Research and Early Detection at the University of Basel Psychiatric Hospital, Switzerland. Her research focuses on schizophrenic psychoses, women‘s mental health, and gender aspects of mental disorders in general.

Niklaus Stulz Psychiatric Services Aargau P.O. Box 432 CH-5201 Brugg Switzerland niklaus.stulz@pdag.ch

© 2017 Hogrefe Publishing

Research Trends

Effects of Awareness Material on Suicide-Related Knowledge and the Intention to Provide Adequate Help to Suicidal Individuals Florian Arendt1, Sebastian Scherr1, Thomas Niederkrotenthaler2, Sabrina Krallmann1, and Benedikt Till2 Department of Communication Science and Media Research, University of Munich (LMU), Germany Department of Social and Preventive Medicine, Center for Public Health, Medical University of Vienna, Austria

1 2

Abstract. Background: Little is known about the impact of educative media reports on the intention to provide help to suicidal individuals and on suicide-related knowledge. Aims: To test whether material debunking widely shared myths influences knowledge and the intention to provide adequate help to others, and if such information reduces reading enjoyment. Method: A randomized controlled trial was utilized. Participants allocated to the intervention group were exposed to awareness material explicitly addressing suicide myths. Results: Analyses show that exposure to printed awareness material increased knowledge, which in turn positively influenced intentions to provide help. The inclusion of information regarding suicide myths did not reduce reading enjoyment. Limitations: The awareness material used in this study only addressed two suicide myths that were considered to be especially important. Conclusion: Information debunking suicide myths in suicide-related media reports is therefore both feasible and potentially helpful.

Keywords: suicide prevention, myths, media, awareness material, RCT

Despite suicide being preventable, worldwide, over 800,000 people die from suicide each year (World Health Organization [WHO], 2014). Of special interest to the present study, suicide can have devastating and far-reaching effects on families, friends, and communities (WHO, 2014). From a public health standpoint, the mass media are considered a key factor for suicide prevention (Mann et al., 2005; Niederkrotenthaler, Reidenberg, Till, & Gould, 2014; WHO, 2008). Importantly, the role of the media in suicide prevention can be considered a double-edged sword (Arendt, Till, & Niederkrotenthaler, 2016; Scherr, 2016): On the one hand, suicide reporting can be a risk factor for imitative suicides. Particularly sensationalist news reporting on suicide is associated with subsequent increases in suicides in the population, also known as the Werther effect (Philipps, 1974; Stack, 2005). On the other hand, responsible reporting emphasizing alternatives to suicide and providing examples of individuals who successfully overcame their suicidal crises can increase awareness and educate the public about suicidality; this may potentially decrease suicidal behavior among audiences, which is known as the Papageno effect (Niederkrotenthaler et al., 2010). © 2017 Hogrefe Publishing

The media may not only have direct effects on suicidal individuals themselves, but also indirect effects on people in their social environment. Importantly, these people have the ability to provide help (Sonneck, Kapusta, Tomandl, & Voracek, 2012). Unfortunately, widely shared suicide myths might hinder helping.

People often feel unconfident when others overtly express suicidal thoughts and they do not know how to react adequately (Aldrich, 2015; Hunt & Eisenberg, 2010). Uncertainty may exist around whether it is beneficial to talk about suicidal thoughts or if talking about suicide may encourage others to act out their suicidality. Some may feel there is no need to address this topic because they think people who talk about suicide will not actually do it. These uncertainties are reflected in widely shared suicide myths, that is, beliefs that “individuals who talk about suicide will not actually do it” and “talking about suicide encourages suicidal behavior” (Sonneck et al., 2012, pp. 256–257).

In the present paper, we tested the effects of printed suicide-awareness material on suicide-related knowledge and on the behavioral intention to provide adequate help to others. We assumed that media reporting can positiveCrisis (2018), 39(1), 47–54 https://doi.org/10.1027/0227-5910/a000474

48

ly contribute to suicide prevention by debunking widely shared suicide myths. We assumed that awareness-intervention material that explicitly addresses suicide myths may increase suicide-related knowledge that, in turn, will have an effect on the behavioral intention to provide adequate help to others.

Suicide Myths Suicide myths can be defined as widely shared misconceptions about suicide that are based on false facts (Domino, 1990; Katz-Sheiban & Eshet, 2008). These myths include common misconceptions such as “talking about suicide encourages suicide” (Niederkrotenthaler et al., 2010; Schurtz, Cerel, & Rodgers, 2010; Sonneck et al., 2012) or “a person who threatens suicide will not carry out the threat” (Domino, 1990). Other examples of suicide myths are “suicide happens without warning” (Neuringer, 1988), “suicidal people clearly want to die” (Domino, 1990), and “once suicidal, suicidal forever” (Domino, 1990; Niederkrotenthaler et al., 2010).

The dissemination of suicide myths is considered a tremendous public health problem: Suicide myths result in distorted knowledge on suicide, which can contribute to a stigmatization of suicidal individuals (Joiner, 2010; Niederkrotenthaler et al., 2014). In addition, suicide myths can inhibit help-seeking behavior among suicidal individuals as well as adequate reactions from families, friends, and colleagues in terms of providing help (Schurtz et al., 2010).

Furthermore, one recent study showed that newspaper reports that disseminate suicide myths increase the risk of imitational suicides in the population, which was not the case for suicide reports debunking these myths (Niederkrotenthaler et al., 2010). Thus, educating the public about suicide is considered an essential component of suicide prevention (Niederkrotenthaler et al., 2014). Accordingly, debunking suicide myths is recommended in media guidelines on how to report about suicide (WHO, 2008).

Responsible media reports may serve as an effective tool in debunking suicide myths by educating the public on suicide and providing accurate facts and expert knowledge.

Although there is already research testing the effects of awareness materials (see Dumesnil & Verger, 2009, for a review), the effectiveness of educating nonsuicidal individuals about suicide myths via newspaper reports has never been tested in previous studies.

Myths Targeted in the Present Study The present study focused on two widely shared, suicide myths: Crisis (2018), 39(1), 47–54

F. Arendt et al., Effects of Suicide-Awareness Material

1. “Those who talk about suicide will not actually do it” Individuals who communicate suicidal thoughts to others should be taken seriously. In fact, suicide experts emphasize that a great majority of those who die by suicide have given some clue or warning (Goodwin, 2003; Kuo, Gallo, & Tien, 2001; Mandrusiak et al., 2006; Rudd et al., 2006; Schurtz et al., 2010; Sonneck et al., 2012). This myth is assumed to cause others to misunderstand or misinterpret the situation. They may not see a need for action even if the suicidal crisis is severe (Sonneck et al., 2012).

2. “Talking about suicide encourages suicidal behavior” Talking with a suicidal individual is one important way to contribute to suicide prevention as this can result in relief and may be an indispensable action toward treatment. Unfortunately, many people believe that they trigger suicide plans when they talk about suicidal thoughts with the suicidal individual (Domino, 1990).

Conversely, bringing up the subject of suicide and talking about it openly is a very helpful method “others” can employ (Gould et al., 2005; Sonneck et al., 2012).

Hypotheses of the Present Study We assumed that media reporting can effectively debunk widely shared suicide myths. Thus, explicitly mentioning the facts noted in the previous section in news coverage may elicit beneficial effects on readers’ knowledge and behavior. In fact, we predicted that exposure to awareness-intervention material explicitly addressing suicide myths will increase suicide-related knowledge (Hypothesis 1). This, in turn, increases the intention to provide adequate help to others (Hypothesis 2). The latter implicates a mediator model in which awareness-material exposure increases behavioral intentions through its influence on suicide-related knowledge.

Method This was a Web-based randomized controlled trial. Participants were randomly allocated to one of three groups.

Participants allocated to the intervention group were exposed to awareness material explicitly addressing suicide myths. After reading, we administered a questionnaire assessing suicide-related knowledge and the intention to provide help to suicidal individuals. In addition, we used two control groups: Whereas participants allocated to the first control group did not read an article at all, participants allocated to the second control group read a newspaper article about a protagonist who had coped with his suicidal © 2017 Hogrefe Publishing

49

F. Arendt et al., Effects of Suicide-Awareness Material

crisis by contacting a crisis-intervention center. This was the same article as was used in the intervention group, but it did not explicitly address suicide myths.

Participants Participants were enrolled through social networking sites using convenience sampling techniques. A total of 273 individuals participated and the majority were female (74%).

The participants ranged in age from 18 to 69 years (M = 26.93, SD = 10.08). About half of the sample had a high school diploma (52%) and approximately one third had a university degree (37%). Almost all participants were German citizens (97%).

Experimental Manipulation Participants allocated to the suicide-unrelated control group (n = 99) did not read awareness-intervention material or any other media content. They simply filled out the questionnaire. Participants allocated to the suicide-related control group (n = 87) read a newspaper article featuring a protagonist who had coped with his suicidal crisis by contacting a crisis-intervention center. We used this additional suicide-related control group because simply reading a suicide report might influence the outcomes of the present study. Using an experimental design with two control groups, we have a confident baseline with which we can compare the intervention group. The stimulus material presented to the suicide-related control group was adapted from a previous study (Arendt et al., 2016). The article described a protagonist who experienced a severe suicidal crisis. He prepared for his suicide, but only seconds prior to his planned suicide, he reconsidered his plans and decided to call the telephone emergency line. The article finished by emphasizing how the protagonist was glad about his decision to seek help and not to have died by suicide.

Participants allocated to the intervention group (n = 87) read the same article, but two paragraphs were added.

The first short paragraph, located in the middle of the article, emphasized how the protagonist was not an isolated case. Conversely, it stated that academic research clearly showed that “most people who attempt suicide tell others about their plan.” By doing so, “they give others the chance to help them.” The second paragraph, located at the end of the article, noted that “many people shy away from talking to suicidal persons” due to the fear that they could potentially do something wrong. However, “experts advise” that “if someone has the impression that a person might be thinking about suicide, then one should address this clearly and talk about the suicidal thoughts of this person.” This © 2017 Hogrefe Publishing

would be “helpful for most people,” because, when being overtly asked, “they have the ability to speak about their sorrows and fears with others.” The two paragraphs providing additional information thus explicitly emphasized that (a) most people told others about their suicide plans before they attempted suicide and that (b) others should directly address these suicidal thoughts (i.e., talk to them about their suicidal thoughts).

Measures Suicide-Related Knowledge Participants were presented with statements and were asked to indicate whether they believed that they were true or false. Ten statements targeted suicide-related knowledge with presumed helpful consequences (e.g., “People often tell others about their suicide plans before they attempt suicide,” “In most cases, suicide happens without warning signs,” “It is better to speak with individuals who talk about suicide about their suicidal thoughts,” and “When one talks to suicidal individuals about suicide, then one prompts them to do it”). We coded whether participants correctly (coded as 1) or incorrectly (coded as 0) rated each of the statements as true or false. The measure is expressed as the relative frequency of giving accurate answers (M = 0.80, SD = 0.16).

Intention to Provide Adequate Help to Suicidal Individuals We used a vignette to measure behavioral intentions.

Therefore, we presented a hypothetical situation for participants to evaluate. In a first step, participants were asked to imagine the following situation: You meet a distant acquaintance who you know from the past.

You talk a bit. You talk about the past and the weather. You make small talk. Suddenly, he starts to talk about his life. He tells you that he has serious problems. He is in a difficult financial situation and his mother died recently. He feels extremely lonely, helpless, and hopeless. He tells you that he thinks no one cares about him. Alcohol only helps for a short period of time; by the next day, everything else remains the same. Suddenly he uses the word suicide – he uses the word, but does not mention anything concrete.

Afterwards, we assessed how participants would react in this hypothetical situation. We presented a series of different statements describing a variety of behavioral options (e.g., “I quickly say goodbye and leave,” “I ask him whether he thinks about suicide,” “I encourage him to seek help at a telephone counseling service or with a psychologist,” “I ask him whether he wants to order another round, which Crisis (2018), 39(1), 47–54

F. Arendt et al., Effects of Suicide-Awareness Material

50

would enable me to spend additional time with him”). Participants were asked to estimate their likelihood of reacting in the way described for each of the behavioral options on a 7-point scale ranging from very unlikely (coded as 1) to very likely (coded as 7). Items describing unhelpful behavior were coded in reverse, with higher scores being indicative of more helpful intentions (M = 5.26, SD = 0.84, α = 0.81).

Enjoyment For the explorative analysis, we measured whether participants enjoyed reading the article. As only two groups read an article, this variable was only measured for the suicide-related control group and the intervention group. We used two items (“The text was interesting” and “I found the text to be exciting”) to assess enjoyment. Participants rated each item on a 7-point scale ranging from I totally disagree (coded as 1) to I totally agree (coded as 7). Participants somewhat enjoyed reading the article (M = 4.79, SD = 1.24, α = 0.72).

Perceived Knowledge Gain We also measured whether participants subjectively believed that they had learned something based on their reading. We used two items (“I have the feeling that I learned something through reading” and “The text has resulted in me having a better understanding of the issue”) to assess perceived knowledge gain. Participants rated each item on a 7-point scale ranging from I totally disagree (coded as 1) to I totally agree (coded as 7). Perceived knowledge gains were on a moderate level (M = 3.68, SD = 1.43, α = 0.77).

Data Analysis To test Hypothesis 1, we used analysis of variance (ANOVA) and simple bivariate t tests, and for Hypothesis 2, we relied on mediation analysis (Hayes, 2013). In a first step, we dummy-coded the experimental group variable. Each dummy represents the effect of an article group (Dummy 1 = effect of suicide article without explicitly addressing suicide myths; Dummy 2 = effect of awareness-intervention material explicitly highlighting suicide myths; reference = suicide-unrelated control group). We specified the following mediator model: First, we predicted suicide-related knowledge by both dummies. Second, we predicted behavioral intentions simultaneously by both dummies and suicide-related knowledge. We present unstandardized coefficients from ordinary least squares regressions.

Indirect effects were tested and 95% confidence intervals (CIs) based on 10,000 bootstrapped samples are provided.

Results

on

ti

In

te

rv

en

tr

on

C

C

on

tr

ol

ol

2

1

K n o w le d g e

S u ic id e - R e la t e d

Hypothesis 1 predicted that awareness-intervention material explicitly addressing suicide myths increases suicide-related knowledge. To test this hypothesis, we used a one-way ANOVA. We found that the experimental manipulation elicited a significant impact on suicide-related knowledge, F(2, 263) = 9.41, p < .001, η2 = .07. As shown in Figure 1, this main effect was largely driven by the intervention group. Accordingly, the intervention group (M = 0.86, SD = 0.16) showed a significantly higher level of suicide-related knowledge compared with the suicide-­ EFFECTS OF SUICIDE-AWARENESS MATERIAL 23unspecific control group (M = 0.76, SD = 0.17), t(182) = 3.99, p < .001, d = 0.61, and the suicide-specific control group (M = 0.78, SD = 0.15), t(166) = 3.44, p = .001, 1 .0 d = 0.52. There were no differences between the two control groups, t(178) = 0.57, p = .570, d = 0.12. This sup0 .9 ports Hypothesis 1: The reading of awareness material ex0 .8 plicitly addressing suicide myths elicited a beneficial effect on suicide-related knowledge.

0 .7 Hypothesis 2 predicted that reading awareness-intervention material debunking suicide myths also has effects 0 .6 on the behavioral intention to provide adequate help to 0 .5 suicidal individuals through its impact on suicide-related knowledge. We used mediation analysis to test this hypothesis (see Figure 2). The total effect of Dummy 2 (i.e., effect of the intervention group) on behavioral intentions E x p e r im e n t a l G r o u p was significant, Coeff = 0.25, SE = 0.12, p = .04, 95% Figure 1. Effects of reading awareness-intervention material on suCI = 0.01–0.50, indicating that reading awareness materiknowledge.

Error barsmaterial indicate confidence knowledge. intervals Error Figure 1.icide-related Effects of reading awareness-intervention on suicide-related al explicitly addressing suicide myths elicited a beneficial (95%). Control 1 = suicide-unrelated control group. Control 2 = suibars indicate confidencecontrol intervalsgroup.

(95%).Intervention Control 1 = suicide-unrelated control group. Control 2effect on the behavioral intention to provide adequate help cide-related = treatment group exposed to awareness material explicitly addressing suicide myths. to suicidal individuals. = suicide-related control group. Intervention = treatment group exposed to awareness material explicitly addressing suicide myths.

Crisis (2018), 39(1), 47–54

© 2017 Hogrefe Publishing

cri_a000474_Arendt_to_MC_2017-05-29.pdf 1 27.07.2017 06:24:44

51

F. Arendt et al., Effects of Suicide-Awareness Material EFFECTS OF SUICIDE-AWARENESS MATERIAL

24

Total Effects Model AwarenessIntervention Material (Dummy 2)

Behavioral Intentions

0.25 n.s.

Figure 2. Mediation analysis: Effects of reading awareness material highlighting suicide myths on behavioral intentions through its effect on suicide-related knowledge. The figure only shows the significant unstandardized regression coefficients. See text for details on the statistical analysis.

(Dummy 1)

Mediation Model Suicide-Related Knowledge 0.10

AwarenessIntervention Material (Dummy 2)

1.34

n.s. n.s.

Behavioral Intentions

n.s.

Dummy 1

We also found supporting evidence for a mediation terial explicitly addressing suicide myths (M = 4.70, SD = Figure 2. Mediation analysis: Effects of reading awareness material highlighting suicide myths process: Exposure to the awareness material explicitly ad1.15) did not lead to a significantly reduced level of enjoyon behavioral intentions through its effect on suicide-related knowledge. The figure only shows dressing suicide myths significantly increased suicide-rement compared with the suicide-related control condition lated knowledge, Coeff = regression 0.10, SE = 0.02,See p text < .001, 95% (M = 4.89, SD = 1.33), t(163) = 1.00, p = .32, d = 0.15. the significant unstandardized coefficients. for details on the statistical CI = 0.05–0.14, which in turn influenced behavioral inWe also tested whether including explicit awareness inanalysis. tentions, Coeff = 1.34, SE = 0.31, p < .01, 95% CI = 0.72– formation influenced perceived knowledge gain. In fact, 1.96]. As expected, there was a significant indirect effect awareness material explicitly addressing suicide myths of Dummy 2 on behavioral intentions through suicide-­ (M = 3.96, SD = 1.40) increased perceived knowledge related knowledge, Coeff = 0.13, Boot SE = 0.05, 95% CI = gain compared with the suicide-related control condition 0.05–0.25. Yet, the direct effect of Dummy 2 on behavio(M = 3.38, SD = 1.40), t(163) = 2.70, p = .008, d = 0.41. ral intentions was not significant, Coeff = 0.13, SE = 0.12, p = .31, 95% CI = −0.12–0.37, indicating that the effect of awareness-material exposure on behavioral intentions was substantially mediated by suicide-related knowledge.

Discussion Consistent with the analysis presented earlier, mediation analysis showed that Dummy 1 (i.e., effect of the conThis study experimentally tested whether reading a sutrol group’s suicide article without explicitly addressing icide report that explicitly addresses suicide myths and suicide myths) did not have an effect on suicide-related debunks them has beneficial effects on suicide-related knowledge, Coeff = 0.01, SE = 0.02, p = .57, 95% CI = knowledge and the intention to provide adequate help to −0.03–0.06. Furthermore, Dummy 1 did not elicit a sigothers. We found that reading such suicide-awareness manificant total effect on behavioral intentions, Coeff = 0.16, terial increased suicide-related knowledge, and that these SE = 0.13, p = .21, 95% CI = −0.09–0.40, indicating that increased knowledge levels, in turn, exerted a positive efreading suicide-awareness material that does not explicitfect on behavioral intentions to provide adequate help to ly debunk suicide myths did not influence behavioral insuicidal individuals. Suicide-awareness material aiming to tentions. educate the public about suicide may have elicited a potentially preventive effect on the audience.

Numerous suicide researchers have pointed out the detrimental effects on suicidal individuals of disseminating Additional Explorative Analysis suicide myths (e.g., Niederkrotenthaler et al., 2010, 2014; Schurtz et al., 2010). Thus, educating the public about suMoreover, we explored whether the inclusion of two icide and debunking common misconceptions about suimyth-debunking paragraphs in a suicide-related article cide are key factors for suicide prevention. An important decreases reading enjoyment. Importantly, awareness ma© 2017 Hogrefe Publishing

Crisis (2018), 39(1), 47–54

52

implication of the present study is that journalists can actively contribute to the reduction of stigma attached to suicide and the increase of knowledge about adequate help for suicidal individuals by explicitly addressing suicide myths.

Our study demonstrated that debunking suicide myths increased suicide-related knowledge among readers, which contributed to increased intentions to provide adequate help to individuals in suicidal crises. A suicide report about a person overcoming a crisis that did not explicitly address suicide myths did not elicit beneficial effects.

Importantly, including two awareness-information paragraphs did not decrease reading enjoyment. This is an important finding because journalists and media managers may refrain from including educational information because they expect a negative effect on their readers (which ultimately may reduce circulation and thus economic success). Journalists may believe that explicit awareness information will be uninteresting to their readers and that explicitly addressing suicide myths will be perceived as “finger-wagging” and thus may elicit negative effects (e.g., reactance, see Brehm & Brehm, 1981). Crucially, and countering this speculation, including explicit awareness information did not reduce reading enjoyment. Furthermore, readers experienced a greater gain in knowledge when the two paragraphs debunking suicide myths were included. Subjectively perceived knowledge-gain levels may be related to self-confidence, which in turn will help individuals dare to help suicidal individuals. Thus, increases in perceived knowledge gain can be deemed a positive outcome.

These findings support widely implemented media guidelines on suicide-prevention strategies (Bohanna & Wang, 2012; WHO, 2008). Many national and international guidelines emphasize that the media should contribute to debunking widely shared suicide myths. The present study shows that following this recommendation has a potentially protective effect, which may help to save lives.

For many years it was felt that not reporting on suicide was the only option to prevent imitative suicides (Hawton & Williams, 2002). The discovery of the Papageno effect demonstrated that reports about suicidal individuals adopting constructive coping strategies had the potential to decrease suicide rates (Niederkrotenthaler et al., 2010).

The present study shows that explicitly debunking suicide myths in a newspaper article increases the intention to adequately help suicidal individuals via an increase in suicide-related knowledge. Importantly, the educative components of such news reports appear to be relevant components in media-related suicide-awareness strategies.

Crisis (2018), 39(1), 47–54

F. Arendt et al., Effects of Suicide-Awareness Material

Limitations The present study has its limitations. First, convenience sampling techniques were used. Women, highly educated people, and ethnic minorities were overrepresented in our sample, limiting the generalizability of the results. Furthermore, the study was conducted in Germany. It is important to note that the effects of culture and context can be profound across the entire help-seeking pathway, from problem identification to choice of treatment providers as previous research has shown (Cauce et al., 2002; see also Bruffaerts et al., 2011). There is evidence that such differences can emerge within societies as well: For example, in one study Asians and Asian Americans report seeking support from their close others less often compared with European Americans (Mojaverian, Hashimoto, & Kim, 2013). Comparable evidence has been revealed in the suicide context (Chan & Thambu, 2016). Future research should replicate our findings in different cultural contexts and test whether media effects are weaker or stronger in specific cultural contexts (e.g., in the East vs. the West).

Second, the awareness material used in this study explicitly addressed two suicide myths that were considered to be especially important. Nevertheless, awareness material targeting other suicide myths may also have beneficial effects. Future research should address them. Third, we only measured behavioral intention instead of actual behavior. This is an important limitation as people may actually behave differently to their behavioral intentions.

According to the theory of planned behavior (Fishbein & Ajzen, 2010), self-efficacy (Bandura, 1977) plays a crucial role when behavioral intentions become real behaviors (Sniehotta, Scholz, & Schwarzer, 2005), thus those with lower levels of self-efficacy (e.g., those with severe depression) might not feel confident enough to actually intervene when confronted with a suicidal person. However, behavioral intention is an important predictor of actual behavior and is considered to be an immediate antecedent of it (Fishbein & Ajzen, 2010). As the measurement of actual helping behavior in the suicide context is very difficult, the measurement of intentions seemed to be a justifiable decision. Fourth, the term myth should be used with caution.

For example, we identified the belief that “those who talk about suicide will not actually do it” as a myth. This decision was based on the finding that a great majority of those who die by suicide have made some announcement or given a clue or warning (Sonneck et al., 2012). However, it has been estimated that for each adult who died of suicide – over 800,000 people die due to suicide every year – there may have been more than 20 others attempting suicide, and many more who experienced suicidal ideation (WHO, 2014). Thus, although the vast majority of individuals who talk about suicide do not die by suicide, most of those who © 2017 Hogrefe Publishing

F. Arendt et al., Effects of Suicide-Awareness Material

die by suicide provide some sort of warning or clue about that intent. Therefore, this public myth can be considered as part myth and part reality. From an educational perspective, it is of high relevance to focus on the true positives, that is, to encourage others to pay serious attention to any suicide announcements, gestures, and communication, because this group will include the majority of those who subsequently die by suicide.

Conclusion The media may not only have effects on suicidal individuals themselves, but also on people in their social environment. The present study shows that news reporting can positively contribute to suicide prevention by debunking common suicide myths with awareness information. Debunking the beliefs that “individuals who talk about suicide will not actually do it” and “talking about suicide may encourage suicidal behavior” elicited beneficial effects on suicide-related knowledge and the intention to provide adequate help to others.

Thus, it would appear to be beneficial for suicide-­ prevention experts to inform media professionals about the positive effects of the inclusion of awareness material in regular suicide-related news coverage. This may contribute to more responsible forms of suicide reporting, which ultimately may have preventive effects.

References Aldrich, R. S. (2015). Using the theory of planned behavior to predict college students’ intention to intervene with a suicidal individual. Crisis, 36, 332–337.

Arendt, F., Till, B., & Niederkrotenthaler, T. (2016). Effects of suicide awareness material on implicit suicide cognition: A laboratory experiment. Health Communication, 31, 718–726.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215.

Bohanna, I., & Wang, X. (2012). Media guidelines for the responsible reporting of suicide. Crisis, 33, 190–198.

Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance: A theory of freedom and control. New York, NY: Academic Press.

Bruffaerts, R., Demyttenaere, K., Hwang, I., Chiu, W., Sampson, N., Kessler, R., … Nock, M. (2011). Treatment of suicidal people around the world. The British Journal of Psychiatry, 199, 64–70.

Cauce, A., Domenech-Rodríguez, M., Paradise, M., Cochran, B., Shea, J., Srebnik, D., & Baydar, N. (2002). Cultural and contextual influences in mental health help seeking: A focus on ethnic minority youth. Journal of Consulting and Clinical Psychology, 70, 44–55.

Chan, L., & Thambu, M. (2016). Cultural factors in suicide prevention. In R. C. O’Connor & J. Pirkis (Eds.), The International Handbook of Suicide Prevention (2nd ed., pp. 541–555). Hoboken, NJ: Wiley-Blackwell © 2017 Hogrefe Publishing

53

Domino, G. (1990). Popular misconceptions about suicide: How popular are they? Omega: Journal of Death and Dying, 21, 167– 175.

Dumesnil, H., & Verger, P. (2009). Public awareness campaigns about depression and suicide: A review. Psychiatric Services, 60, 1203–2013.

Fishbein, M., & Ajzen, I. (2010). Predicting and changing behavior: The reasoned action approach. New York, NY: Psychology Press (Taylor & Francis).

Goodwin, F. K. (2003). Preventing inpatient suicide. Journal of Clinical Psychiatry, 64, 12–13.

Gould, M. S., Marrocco, F. A., Kleinman, M., Thomas, J. G., Mostkoff, K., Cote, J., & Davies, M. (2005). Evaluating iatrogenic risk of youth suicide screening programs: A randomized controlled trial. Journal of the American Medical Association, 293, 1635– 1643.

Hawton, K., & Williams, K. (2002). Influences of the media on suicide: Researches, policy makers, and media personnel need to collaborate on guidelines. British Medical Journal, 325, 1374– 1375.

Hayes, A. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: The Guilford Press.

Hunt, J., & Eisenberg, D. (2010). Mental health problems and help-seeking behavior among college students. Journal of Adolescent Health, 46(1), 3–10.

Joiner, T. (2010). Myths about suicide. Cambridge, MA: Harvard University Press.

Katz-Sheiban, B., & Eshet, Y. (2008). Facts and myths about suicide: A study of Jewish and Arab students in Israel. Omega: Journal of Death and Dying, 57, 279–298.

Kuo, W. H., Gallo, J. J., & Tien, A. Y. (2001). Incidence of suicide ideation and attempts in adults: The 13-year follow-up of a community sample in Baltimore, Maryland. Psychological Medicine, 31(7), 1181–1191.

Mandrusiak, M., Rudd, M. D., Joiner, T. E., Jr., Berman, A. L., Van Orden, K. A., & Witte, T. (2006). Warning signs for suicide on the Internet: A descriptive study. Suicide and Life-Threatening Behavior, 36(3), 263–271.

Mann, J. J., Apter, A., Bertolote, J., Beautrais, A., Currier, D., Hass, A., … Hendin, H. (2005). Suicide prevention strategies: A systematic review. Journal of the American Medical Association, 294, 2064–2074.

Mojaverian, T., Hashimoto, T., & Kim, H. (2013). Cultural differences in professional help seeking: A comparison of Japan and the U.S. Frontiers in Psychology, 3, 1–8. doi:10.3389/fpsyg.2012.00615 Neuringer, C. (1988). The meaning behind popular myths about suicide. Omega: Journal of Death and Dying, 18, 155–162.

Niederkrotenthaler, T., Reidenberg, D., Till, B., & Gould, M. (2014).

Increasing help-seeking and referrals for individuals at risk for suicide by decreasing stigma: The role of mass media. American Journal of Preventive Medicine, 47, 235–243.

Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., … Sonneck, G. (2010). The role of media reports in completed and prevented suicide: Werther versus Papageno effects. British Journal of Psychiatry, 197, 234–243.

Phillips, D. P. (1974). The influence of suggestion on suicide: Substantive and theoretical implications of the Werther effect.

American Sociological Review, 39, 340–354.

Rudd, M. D., Berman, A. L., Joiner, T. E., Jr., Nock, M. K., Silverman, M. M., Mandrusiak, M., … Witte, T. (2006). Warning signs for suicide: Theory, research, and clinical applications. Suicide and Life-Threatening Behavior, 36(3), 255–262.

Scherr, S. (2016). Depression – Medien – Suizid: Zur empirischen Relevanz von Depressionen und Medien für die Suizidalität [Depression – media – suicide: On the empirical relevance of Crisis (2018), 39(1), 47–54

54

depression and media for suicidality]. Wiesbaden, Germany: Springer VS. Schurtz, D. R., Cerel, J., & Rodgers, P. (2010). Myths and facts about suicide from individuals involved in suicide prevention. Suicide and Life-Threatening Behavior, 40(4), 346–352.

Sniehotta, F. F., Scholz, U., & Schwarzer, R. (2005). Bridging the intention–behaviour gap: Planning, self-efficacy, and action control in the adoption and maintenance of physical exercise.

Psychology & Health, 20(2), 143–160.

Sonneck, G., Kapusta, N., Tomandl, G., & Voracek, M. (Eds.) (2012).

Krisenintervention und Suizidverhütung [Crisis intervention and suicide prevention]. Vienna, Austria: Facultas.

Stack, S. (2005). Suicide in the media: A quantitative review of studies based on nonfictional stories. Suicide & Life-Threatening Behavior, 35, 121–133.

World Health Organization. (2008). Preventing suicide: A resource for media professionals. Geneva, Switzerland: Author.

World Health Organization. (2014). Preventing suicide: A global imperative. Geneva, Switzerland: Author.

Received September 13, 2016 Revision received February 28, 2017 Accepted March 11, 2017 Published online August 10, 2017 Florian Arendt, PhD, is a postdoctoral researcher at the Department of Communication Science and Media Research, LMU Munich, Germany.

His research focuses on health communication, media stereotyping, and media effects.

Crisis (2018), 39(1), 47–54

F. Arendt et al., Effects of Suicide-Awareness Material

Sebastian Scherr, PhD, is a postdoctoral researcher at the Department of Communication Science and Media Research, LMU Munich, Germany. His research interests include suicide prevention, health communication, media effects, political communication, and empirical methods.

Sabrina Krallmann, BA, is a student in the Department of Communication Science and Media Research, LMU Munich, Germany.

Thomas Niederkrotenthaler, MD PhD MMs, is Associate Professor and Head of the Suicide Research Unit/Department of Social and Preventive Medicine, Center for Public Health, Medical University of Vienna, Austria. He is the co-chair of the Media & Suicide Special Interest Group of IASP, and chair of the Wiener Werkstaette for Suicide Research.

Benedikt Till, PD, DSc, is Head of the Suicide Research Unit, Department of Social and Preventive Medicine, Center for Public Health, Medical University of Vienna, Austria. He works in the field of media psychology and suicide research, and he is board member of the Wiener Werkstaette for Suicide Research.

Florian Arendt Department of Communication Science and Media Research University of Munich (LMU) Oettingenstraße 67 80538 Munich Germany florian.arendt@ifkw.lmu.de

© 2017 Hogrefe Publishing

Research Trends

Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices The Impact of Fear of Suicide-Related Outcomes and ­Comfort Working With Suicidal Individuals Jared F. Roush1, Sarah L. Brown1, Danielle R. Jahn2, Sean M. Mitchell1, Nathanael J. Taylor1, Paul Quinnett3, and Richard Ries3 Department of Psychological Sciences, Texas Tech University, Lubbock, TX, USA Primary Care Institute, Gainesville, FL, USA 3 The QPR Institute, Inc., Spokane, WA, USA 1 2

Abstract. Background: Approximately 20% of suicide decedents have had contact with a mental health professional within 1 month prior to their death, and the majority of mental health professionals have treated suicidal individuals. Despite limited evidence-based training, mental health professionals make important clinical decisions related to suicide risk assessment and management. Aims: The current study aimed to determine the frequency of suicide risk assessment and management practices and the association between fear of suicide-related outcomes or comfort working with suicidal individuals and adequacy of suicide risk management decisions among mental health professionals. Method: Mental health professionals completed self-report assessments of fear, comfort, and suicide risk assessment and management practices.

Results: Approximately one third of mental health professionals did not ask every patient about current or previous suicidal thoughts or behaviors. Further, comfort, but not fear, was positively associated with greater odds of conducting evidence-based suicide risk assessments at first appointments and adequacy of suicide risk management practices with patients reporting suicide ideation and a recent suicide attempt.

Limitations: The study utilized a cross-sectional design and self-report questionnaires. Conclusion: Although the majority of mental health professionals report using evidenced-based practices, there appears to be variability in utilization of evidence-based practices.

Keywords: suicidal patients, evidence-based practice, suicide attempts, suicide

Suicide remains the 10th leading cause of death in the United States (Centers for Disease Control and Prevention, 2016), and approximately 20% of individuals who die by suicide have had contact with a mental health professional within 1 month prior to their death (Ahmedani et al., 2014; Luoma, Martin, & Pearson, 2002). Mental health professionals are in a key position to appropriately assess suicide risk and make potentially life-saving decisions about suicide risk management and interventions.

Therefore, it is important to determine whether providers are utilizing evidence-based suicide risk assessment and management (EBAM) and if there are factors associated with the implementation of these practices.

Mental health professionals frequently make important clinical decisions related to suicide risk assessment and management despite a lack of training in EBAM (e.g., Feldman & Freedenthal, 2006; Jahn, Quinnett, & Ries, 2016). © 2017 Hogrefe Publishing

A comprehensive review of clinical practice guidelines indicates a consensus supporting the use of empirically supported assessments and interventions for suicide risk (e.g., restricting access to lethal means; Bernert, Hom, & Roberts, 2014). Other guidelines specifically advocate for a suicide risk assessment during an initial interview with a patient or upon admission to a facility (Risk Management Foundation of the Harvard Medical Institutions, 2002), obtaining consent to speak with significant others in the patient’s social support network about suicide risk (American Psychiatric Association, 2003), utilizing suicide safety planning (i.e., a collaborative research-informed tool designed to identify resources and coping strategies for use by a patient during a suicidal crisis; Stanley & Brown, 2012), and providing contact information for suicide crisis hotlines (Gould, Kalafat, Harris Munfakh, & Kleinman, 2007) in place of interventions with a lack of empirical Crisis (2018), 39(1), 55–64 https://doi.org/10.1027/0227-5910/a000478

56

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

support (e.g., no-suicide contracts; Bernert et al., 2014).

Given that these guidelines and recommendations are empirically supported, it is crucial that current clinical practices are consistent with these approaches to promote improved clinical outcomes for suicidal individuals.

Although formalized training in EBAM is available to mental health trainees and professionals (e.g., Question, Persuade, Refer, and Treat; Quinnett, 1995 as cited in Pisani, Cross, & Gould, 2011), it is not widely disseminated (e.g., Debski, Spadafore, Jacob, Poole, & Hixson, 2007). In fact, research suggests that fewer than 25% of social workers (Feldman & Freedenthal, 2006) and fewer than 50% of psychologists and psychologists-in-training (Debski et al., 2007; Dexter-Mazza & Freeman, 2003) receive training in suicide prevention. Additionally, nearly one third of mental health professionals believe they have received insufficient suicide-focused training, although only 9% reported not having worked with suicidal patients (Jahn et al., 2016). Further, training may consist of outdated or inadequate practices; for example, 77% of psychiatrists indicated that no-suicide contracts were helpful (Kroll, 2000), despite a complete lack of empirical support for the practice (Rudd, Mandrusiak, & Joiner, 2006). Thus, research is needed to determine whether EBAM is reliably being provided to suicidal individuals.

Further, there is a dearth of research on factors that influence the use of EBAM among mental health professionals. Mental health professionals’ fear and anxiety may influence a variety of clinical practices related to suicide risk (e.g., Hendin, Haas, Maltsberger, Koestner, & Szanto, 2006; Meyer, Farrell, Kemp, Blakey, & Deacon, 2014; Pope & Tabachnick, 1993). Given that patient death by suicide is the most common fear reported by mental health professionals (Pope & Tabachnick, 1993), fear and comfort related to working with suicidal patients may be associated with the use of EBAM (Hendin et al., 2006). Whether fear and comfort are related to a lack of training (Schmitz et al., 2012), self-perceived incompetence (Oordt, Jobes, Fonseca, & Schmidt, 2009), and/or previous negative experiences with suicidal patients (e.g., suicide attempts or death by suicide; Kleespies, Penk, & Forsyth, 1993; Rothes, Henriques, Leal, & Lemos, 2014), these experiences may play an important role in mental health professionals’ clinical decisions.

Fear of suicide-related outcomes or comfort related to working with suicidal individuals may impact various aspects of practice (e.g., consultation, hospitalization, access to lethal means, self-harm contracts; Hendin et al., 2006). In a recent study, Jahn et al. (2016) found that 88% of mental health professionals reported at least some fear of patient death by suicide and 15% reported not feeling comfortable working with suicidal patients. In another study, mental health professionals reported only “modCrisis (2018), 39(1), 55–64

erate comfort” with suicide risk management, on average (Berman, Stark, Cooperman, Wilhelm, & Cohen, 2015).

Despite fear of suicide-related outcomes and moderate levels of comfort being relatively common among mental health professionals, research suggests that fear of patient suicide attempt or death by suicide and comfort in working with suicidal individuals are only moderately negatively correlated (Jahn et al., 2016). Taken together, fear of suicide-related outcomes and comfort working with suicidal individuals may be associated with a decreased use of EBAM, but the association between comfort working with suicidal individuals and use of EBAM may be weaker among those who report greater fear of suicide-related outcomes.

The current study aimed to determine the frequency with which mental health professionals use various suicide risk assessment and management practices and to elucidate the relation between fear of suicide-related outcomes or comfort in working with suicidal individuals and the adequacy of EBAM among mental health professionals.

We hypothesized that fear would be negatively associated and comfort would be positively associated with the use of EBAM: during the first visit with patients, with patients reporting active suicide ideation, and with patients reporting a recent suicide attempt. Further, we expected that fear would moderate the association between comfort and the use of EBAM in each of these clinical contexts, such that mental health professionals lower in fear would evidence a stronger association between comfort and the use of EBAM.

Method Participants Mental health professionals (N = 289) participated in this study. The majority were from the United States (n = 263, 91.0%), followed by Australia (n = 6, 2.1%), the United Kingdom (n = 5, 1.7%), Canada (n = 2, 0.7%), Ukraine (n = 2, 0.7%), Botswana (n = 1; 0.3%), Brazil (n = 1, 0.3%), Denmark (n = 1, 0.3%), Ireland (n = 1, 0.3%), Romania (n = 1, 0.3%), Serbia (n = 1, 0.3%), and Spain (n = 1, 0.3%), and data were missing for four participants (1.4%). The majority were women (n = 219; 75.8%; men: n = 70; 24.2%). Most professionals had a master’s degree (n = 192; 66.4%), followed by doctoral degree (n = 61, 21.2%), 4-year degree (n = 18, 6.2%), and other training (n = 18, 6.2%). Most participants identified as social workers (n = 73; 25.3%), psychologists (n = 68; 23.5%), and mental health counselors (n = 67; 23.3%). We did not collect information specifically on prescriptive authority of © 2017 Hogrefe Publishing

57

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

mental health professionals. On average, participants were 44.3 years old (SD = 12.6). The average number of years of professional experience among participants was 14.5 (SD = 11.2). Few participants reported being a student (n = 5; 5.2%), a trainee (n = 8; 2.8%), or both a student and a trainee (n = 27; 9.3%). See Jahn et al. (2016) for additional details regarding the sample.

Measures Suicidal Patient Comfort Survey (SPCS) The SPCS, which has five separate domains, was created for this study (Jahn et al., 2016). The domains include: demographic information, self-reported knowledge and attitudes about suicide, personal reactions to suicide and suicide risk, self-efficacy in assessment and management of suicide risk, and current practices related to assessment and management of suicide risk. The domains of the SPCS have demonstrated convergent and discriminant validity (Mitchell et al., 2017). Specific items within the SPCS are also face valid, as they include wording that specifically identifies the target construct (Jahn et al., 2016).

Fear and comfort. Items assessing fear of suicide-related outcomes (i.e., “Patient attempting suicide while in your care” and “Patient killing him/herself while in my care”) were reported on a 5-point ordinal response scale ranging from 1 (not at all) to 5 (completely), in response to the prompt, “Of the possible situations that could occur in a clinical setting, I would fear…” A score for suicide-related fear was calculated by computing an average of these two items for each participant and was used as a predictor. A score for non-­suicide-related fear was calculated by computing the average of five items assessing non-suicide-related fear; this score was used as a covariate in the analyses below. In the current study, Cronbach’s α was 0.90 for fear of suicide-­ related outcomes and 0.79 for non-suicide-related fear.

Sixteen items assessing comfort were answered on a 5-point ordinal response scale ranging from 1 (very uncomfortable) to 5 (very comfortable) in response to the prompt, “Given your training and experience as a mental health professional, how comfortable do you feel working with patients who are experiencing…” The suicide-related item (i.e., “Suicidal thoughts, feelings, and impulses”) was included as a predictor in the analyses. Additionally, a non-suicide-related comfort score was determined by averaging across the other 15 items (e.g., mood disorders, factitious disorder, marital/family problems) and was used as a covariate in the subsequent analyses. In the current study, Cronbach’s α for non-suicide-related comfort was 0.87.

Suicide risk assessment and management practices.

Information about mental health professionals’ current EBAM was used as the outcome in the analyses. Partici© 2017 Hogrefe Publishing

pants were asked to respond to a question about their current practice regarding suicide risk assessment on all first visits with new patients, and prompted to select only one response that best fit their typical practice. Responses were coded as 0 for those that were not evidence-based (i.e., “Ask about current, recent or remote suicidal thoughts and behaviors only with my most distressed patients, e.g., estimated GAF of 50 or lower and suicide warning signs are present,” “Only ask about current, recent or remote suicidal thoughts and behaviors when the patient broaches the subject,” or “Routinely do not ask any patients about suicidal thoughts and behaviors”) or 1 for the response that was evidence-based (i.e., “Routinely ask every patient about current, recent or remote suicidal thoughts and behaviors”).

Participants also responded to items about their utilization of assessment and management practices in two clinical scenarios: when a patient is thinking about suicide and when a patient has made a recent suicide attempt. Participants could select all relevant responses out of 15 options.

Difference scores were computed to assess the degree of EBAM utilized by participants by subtracting the number of practices that are not evidence-based from the number of practices that are evidence-based (for categorization of practices provided in tables). The difference score reliability coefficient for EBAM for a new patient actively thinking about suicide is 0.70 and EBAM for a new patient has made a non-fatal suicide attempt was 0.65. Although the brief nature of the assessment limited our ability to thoroughly assess each construct, single-item assessments have been previously used to measure aspects of suicide-related training with evidence of strong validity (e.g., Feldman & Freedenthal, 2006; Oordt et al., 2009).

Procedures The appropriate Institutional Review Board deemed the study to be exempt; participants were not required to provide informed consent. Participants were recruited via e-mails sent to professional mental health-related listservs, and individuals who were registered for a suicide gatekeeper training program were also invited to participate. Snowball sampling (i.e., chain referral sampling) was also utilized. Participants who agreed to participate completed an anonymous online survey through the Survey Monkey software program.

Data Analytic Strategy and Preparation Analyses were conducted using IBM SPSS (Version 21).

We first examined descriptive statistics and correlations Crisis (2018), 39(1), 55–64

58

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

Table 1. Descriptive statistics and correlations 1 1. Age 2. Fear (suicide-related) 3. Comfort (suicide-related) 4. Fear (nonsuicide)

2

3

4

5

6

7

8

– −.11

–

.08

−.22**

−.12*

–

.58**

−.22**

–

5. Comfort (nonsuicide)

.09

−.12**

.52**

−.33

6. Evidence-based practices on first visit

.04

.02

.12*

.01

.01

–

7. Evidence-based practices for suicide ideation

.06

.06

.24**

.04**

.13**

.03**

8. Evidence-based practices for suicide attempts

.04

.05

.18**

.05

.16**

.09

M

44.33

3.37

4.08

2.97

3.37

0.83

3.15

2.84

SD

12.63

1.01

0.78

0.97

0.51

– – .81**

–

0.38

2.15

1.98

Minimum

21

1

1

1

1

0

−1

−3

Maximum

76

5

5

5

5

1

7

7

Note. *p < .05; ** p < .01.

Table 2. Descriptive statistics for the item, “Of the possible situations that could occur in a clinical setting, I would fear:” Response option

M (SD)

Patient killing him/herself while in my care

3.49 (1.15)

Patient attempting suicide while in your care

3.24 (1.06)

Being sued for malpractice

2.82 (1.15)

Patient filing a formal complaint against me

2.64 (1.18)

False information being spread about me

2.40 (1.25)

Patient physically assaulting me

2.30 (1.01)

Patient telling me he/she is sexually attracted to me

1.74 (0.81)

Note. Item response options include 1 = not at all, 2 = a little bit, 3 = somewhat, 4 = very much, 5 = completely. Observed range is 1–5 for all response options.

Table 3. Descriptive statistics for the item, “Given your training and experience as a mental health professional, how comfortable do you feel working with patients who are experiencing:” Response option

M (SD)

Factitious disorders

2.95 (0.90)

Dissociative disorders

3.01 (1.03)

Schizophrenia and other psychotic disorders

3.23 (1.16)

Psychotic disorders

3.25 (1.13)

Somatoform disorders

3.25 (0.97)

Eating disorders

3.26 (1.01)

Sexual and gender identity disorders

3.58 (1.01)

Personality disorders

3.65 (0.96)

Sleep disorders

3.69 (0.89)

Substance-related disorders

3.73 (0.99)

Co-occurring disorders (e.g., substance abuse and depressive disorder)

3.94 (0.90)

Suicidal thoughts, feelings, and impulses

4.08 (0.82)

Marital/family problems

4.15 (0.86)

Adjustment disorders

4.34 (0.83)

Anxiety disorders

4.36 (0.69)

Mood disorders

4.37 (0.72)

Note. Item response options include 1 = very uncomfortable, 2 = uncomfortable, 3 = neutral, 4 = comfortable, 5 = very comfortable; Observed range is 1–5 for all response options.

Crisis (2018), 39(1), 55–64

for variables of interest. Nonparametric bootstrapped moderation procedures were used to test hypotheses using the PROCESS macro for SPSS Version 21 (Model 1; Hayes, 2013). Two separate analyses were conducted, with the difference scores for EBAM versus non-EBAM in two clinical contexts (i.e., practices with patients reporting active suicide ideation, and practices with patients reporting a recent suicide attempt) as the outcome in separate analyses. Fear of suicide-related outcomes was entered as a moderator of the relation between comfort in working with suicidal individuals and the EBAM difference scores for each of the two clinical scenarios. For all analyses, non-suicide-related fear and non-suicide-related comfort served as covariates to control for individual variability in general fear and comfort related to clinical issues. Ordinal variables were treated as continuous given the robustness of nonparametric analyses with regard to ordinal data (Norman, 2010). We utilized 10,000 bootstrap samples to construct 95% bias-corrected confidence intervals. Missing data were handled using mean imputation and total scores were calculated using imputed values.

Eleven participants were excluded from the analysis related to working with patients reporting suicide ideation because these participants indicated never having encountered a patient reporting suicide ideation, resulting in a total of 278 participants for that analysis. A further 42 participants were excluded from the analysis related to working with a patient reporting a non-fatal suicide attempt within the past 30 days because these participants indicated never encountering this clinical situation, resulting in a total of 247 participants for that analysis. Thus, participants were only excluded from analyses that did not apply to them, as results would otherwise be skewed and thereby uninterpretable. Predictor variable scores were standardized (z-score transformed) and the analyses were © 2017 Hogrefe Publishing

59

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

Table 4. Frequencies: “On all first visits with new patients, I:” Response option

Yes

Routinely ask every patient about current, recent, or remote suicidal thoughts or behaviors ­(evidence-based practice)

No

199 (68.9)

90 (31.1)

Ask about current, recent, or remote suicidal thoughts and behaviors only with my most distressed patients, e.g., estimated GAF of 50 or lower and suicide warning signs are present

38 (13.1)

251 (86.9)

Only ask about current, recent, or remote suicidal thoughts and behaviors when the patient ­broaches the subject

12 (4.2)

277 (95.8)

Routinely do not ask any patients about suicidal thoughts and behaviors

11 (3.8)

278 (96.2)

Note. Percentage of sample appears in parentheses. The sum of percentages for endorsement of a response does not equal 100% because of missing data.

Table 5. Frequencies: “In actual practice, when I learn that a new patient is actively thinking about suicide, but has not yet made an attempt, I:” Response option

Yes

No

This has never happened to me

11 (3.8)

278 (96.2)

Immediately contact mental health authorities for further evaluationa

19 (6.6)

270 (93.4)

Depending on the quality of the therapeutic relationship established, continue to see the patient

128 (44.3)

161 (55.7)

Provide the patient with a crisis line number in case the crisis worsens

184 (63.7)

105 (36.3)

Ask for signed permission to speak with significant others

116 (40.1)

173 (59.9)

Conduct a suicide risk assessment and adjust my treatment plan accordingly (e.g., increase visits, speak to family)

223 (77.2)

66 (22.8)

Depending on the quality of the therapeutic relationship, develop a collaborative patient-safety/ crisis response plan and negotiate a treatment plan that matches the level of assessed risk

196 (67.8)

93 (32.2)

Conduct a means restriction counseling session, with family included if available

98 (33.9)

191 (66.1)

Refer the patient to a psychiatrist for evaluation and continue treatment as indicated

79 (27.3)

210 (72.7)

Immediately refer the patient to a hospital emergency room for an evaluation

22 (7.6)

267 (92.4)

Immediately refer the patient to an inpatient psychiatric unit

11 (3.8)

278 (96.2)

Arrange for the patient to see a trusted nonpsychiatrist colleague for a second opinion

10 (3.5)

279 (96.5)

Refer the patient to nonpsychiatrist physician for a medication evaluation

23 (8.0)

266 (92.0)

Make sure I get a signed no-suicide contract in the medical record

64 (22.1)

225 (77.9)

Immediately refer the patient to a state-licensed mental health agency

10 (3.5)

279 (96.5)

Evidence-based practices

Non-evidence-based practices

Note. Percentage of sample appears in parentheses. a This practice was not coded because there are cases in which it would be considered evidence-based and appropriate practice (e.g., in the case of a supervised mental health professional), whereas in other cases it would be inappropriate (e.g., in the case of a licensed independent mental health professional).

conducted again. This produced standardized effect sizes (i.e., standardized odds ratios), which can be more directly compared because they are based on a z-score scale.

Results Descriptive statistics and correlations are presented in Table 1. Average fear of various clinical situations is presented in Table 2, and comfort working with different clinical diagnoses and issues is presented in Table 3. Of note, patient death by suicide was the most highly rated fear, fol© 2017 Hogrefe Publishing

lowed by patient suicide attempt. However, respondents reported feeling more comfortable working with suicide risk as a clinical issue than many other diagnoses and issues. Frequencies for various suicide risk assessment and management practices are presented in Tables 4–6; Table 4 reflects practices conducted at the first appointment with new patients, Table 5 provides information about practices with patients endorsing suicide ideation (EBAM difference score M = 3.15, SD = 2.15), and Table 6 presents frequencies for practices with patients reporting a recent suicide attempt (EBAM difference score M = 2.84, SD = 1.98).

Only 68.9% of the sample reported routinely asking every patient about suicide ideation at a first appointment. AddiCrisis (2018), 39(1), 55–64

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

60

Table 6. Frequencies: “In actual practice, when I learn that a new patient has made a nonfatal suicide attempt within the past 30 days known only to me, I:” Response option

Yes

This has never happened to me

No

42 (14.5)

247 (85.5)

13 (4.5)

276 (95.5)

Depending on the quality of the therapeutic relationship established, continue to see the patient

123 (42.6)

166 (57.4)

Provide the patient with a crisis line number in case the crisis worsens

158 (54.7)

131 (45.3)

Ask for signed permission to speak with significant others

114 (39.4)

175 (60.6)

Conduct a suicide risk assessment and adjust my treatment plan accordingly (e.g., increase visits, speak to family)

197 (68.2)

92 (31.8)

Depending on the quality of the therapeutic relationship, develop a collaborative patient-safety/ crisis response plan and negotiate a treatment plan that matches the level of assessed risk

179 (61.9)

110 (38.1)

Conduct a means restriction counseling session, with family included if available

100 (34.6)

189 (65.4)

90 (31.1)

199 (68.9)

Immediately refer the patient to a hospital emergency room for an evaluation

8 (2.8)

281 (97.2)

Immediately refer the patient to an inpatient psychiatric unit

5 (1.7)

284 (98.3)

a

Immediately contact mental health authorities for further evaluation Evidence-based practices

Refer the patient to a psychiatrist for evaluation and continue treatment as indicated Non-evidence-based practices

9 (3.1)

280 (96.9)

Refer the patient to nonpsychiatrist physician for a medication evaluation

28 (9.7)

261 (90.3)

Make sure I get a signed no-suicide contract in the medical record

52 (18.0)

237 (82.0)

Immediately refer the patient to a state-licensed mental health agency

11 (3.8)

278 (96.2)

Arrange for the patient to see a trusted nonpsychiatrist colleague for a second opinion

Note. Percentage of sample appears in parentheses. a This practice was not coded because there are cases in which it would be considered evidence-based and appropriate practice (e.g., in the case of a supervised mental health professional), whereas in other cases it would be inappropriate (e.g., in the case of a licensed independent mental health professional).

tionally, 22.1% reported ensuring that they had a no-suicide contract in the medical record for individuals who reported suicide ideation, and 18.0% reported this practice for individuals who reported a recent suicide attempt.

In partial support of our hypothesis, comfort working with suicidal individuals was associated with greater odds of conducting evidence-based suicide risk assessment at first appointments with patients, standardized odds ratio (OR) = 1.49, p = .03 (95% CI = 1.04, 2.12; see Table 7).

However, fear of suicide-related outcomes was not associated with greater odds of conducting evidence-based suicide risk assessment at first appointments with patients, standardized OR = 1.12, p = .56 (95% CI = 0.76, 1.66).

The interaction between fear and comfort working with patients reporting suicide ideation was also not significant, standardized OR = 1.23, p = .14 (95% CI = 0.94, 1.62).

Similarly, comfort working with suicidal patients was positively associated with adequacy of suicide risk management practices with individuals reporting suicide ideation, β = 0.24, p = .001 (95% CI = 0.10, 0.38), and individuals reporting a recent suicide attempt, β = 0.16, p = .04 (95% CI = 0.01, 0.32). However, there was not a significant association between fear of suicide-related outcomes Crisis (2018), 39(1), 55–64

and adequacy of suicide risk management practices with individuals reporting suicide ideation, β = 0.11, p = .11 (95% CI = −0.03, 0.25), and individuals reporting a recent suicide attempt, β = 0.10, p = .20 (95% CI = −0.05, 0.26). Upon examination of potential moderating effects, the interaction between fear and comfort was also not a significant predictor of adequacy of suicide risk management practices with individuals reporting suicide ideation, β = −0.04, p = .47 (95% CI = −0.15, 0.07), or a recent suicide attempt, β = −0.03, p = .68 (95% CI = −0.14, 0.95).

Discussion The current study sought to determine the frequency of various suicide risk assessment and management practices among mental health professionals, and to identify factors that may influence their implementation. The results suggest that while the majority of mental health professionals report routinely asking every patient about suicidal thoughts or behaviors, approximately 30% indicated they do not ask every patient and 4% routinely do © 2017 Hogrefe Publishing

61

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

Table 7. Moderation results: Suicide-related fear as a potential moderator of the relations between suicide-related comfort and use of ­evidence-based practices Predictor variable

Unstandardized coefficient

SE

t/Wald

CI (LL, UL)

p

Criterion variable: first session (Nagelkerke R2 = 0.05) Constant

0.42

1.32

0.31

−2.17, 3.00

.753

Fear (non-suicide)

0.02

0.20

0.11

−0.37, 0.41

.911

−0.40

0.36

−1.11

−1.10, 0.30

.267

Fear

0.11

0.19

0.60

−0.26, 0.49

.550

Comfort

0.54

0.23

2.36

0.09, 0.98

.018

Constant

3.00

1.38

2.17

0.29, 5.71

.030

Fear (non-suicide)

0.01

0.20

0.07

−0.38, 0.40

.946

−0.38

0.35

−1.06

−1.07, 0.31

.288

Fear

0.11

0.20

0.58

−0.27, 0.50

.563

Comfort

0.51

0.23

2.19

0.05, 0.96

.029

Fear × Comfort

0.26

0.18

1.48

−0.84, 0.61

.139

0.35

−2,47, 1.72

.727

Comfort (non-suicide)

Comfort (non-suicide)

Criterion variable: practices for suicide ideation (F[5,272] = 3.57, R2 = 0.06, p = .004) Constant

−0.37

1.06

Fear (non-suicide)

0.03

0.19

0.13

−0.36, 0.41

.898

Comfort (non-suicide)

0.09

0.27

0.31

−0.45, 0.62

.754

Fear

0.23

0.15

1.52

−0.07, 0.52

.130

Comfort

0.65

0.19

3.42

0.28, 1.02

.001

Constant

2.50

1.13

2.21

0.27, 4.73

.028

Fear (non-suicide)

0.08

0.21

0.37

−0.33, 0.48

.713

Comfort (non-suicide)

0.12

0.29

0.41

−0.45, 0.69

.679

Fear

0.23

0.16

1.42

−0.09, 0.54

.157

Comfort

0.66

0.20

3.22

0.26, 1.06

.001

0.16

−1.13

−0.50, 0.13

.261

−4.10, 1.12

.262

Fear × Comfort

−0.18 2

Criterion variable: practices for suicide attempts (F[5,241] = 2.59, R = 0.05, p = .03) Constant

−1.49

1.33

−1.13

Fear (non-suicide)

0.18

0.25

0.73

−0.31, 0.66

.468

Comfort (non-suicide)

0.32

0.35

0.94

−0.35, 1.01

.351

Fear

0.27

0.18

1.46

−0.93, 0.63

.145

Comfort

0.60

0.25

2.37

0.10, 1.10

.019

Constant

1.65

1.31

1.26

−0.93, 4.24

.208

Fear (non-suicide)

0.09

0.23

0.39

−0.36, 0.54

.682

Comfort (non-suicide)

0.43

0.34

1.26

−0.23, 1.09

.208

Fear

0.23

0.18

1.28

−0.12, 0.59

.202

Comfort

0.50

0.24

2.07

0.02, 0.97

.040

−0.08

0.19

–0.41

−0.45, 0.98

.696

Fear × Comfort

Note. SE = standard error. CI = 95% confidence interval. LL = lower limit. UL = upper limit. Fear = centered average of two fear scores (i.e., patient suicide attempt, patient death by suicide) on the Suicidal Patient Comfort Survey. Comfort = centered score for the comfort in working with suicidal thoughts, feelings, and impulses as a clinical issue item on the Suicidal Patient Comfort Survey. Fear (nonsuicide) = centered average of five fear items on the Suicidal Patient Comfort Survey not related to suicide. Comfort (nonsuicide) = centered average of 15 comfort items on the Suicidal Patient Comfort Survey not related to suicide. Fear × Comfort = multiplied effects of Fear and Comfort. Unstandardized coefficients for the first regression model (criterion variable: first session) are logits; unstandardized coefficients for second and third regression models are unstandardized beta values.

© 2017 Hogrefe Publishing

Crisis (2018), 39(1), 55–64

62

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

not conduct any suicide risk assessment. Fewer than half of participants reported EBAM, such as asking permission to speak with significant others and conducting a means safety session, both of which are important suicide prevention strategies (American Psychiatric Association, 2003; Bernert et al., 2014). Importantly, the majority of mental health professionals indicated that they would not implement non-EBAM. Although the majority of mental health professionals reported using EBAM, there appears to be variability in such utilization, especially with patients who recently attempted suicide. For instance, it is plausible that a suicidal patient receiving mental health services following a recent suicide attempt may receive suicide risk management that has no empirical support (e.g., a no-suicide contract; Hendin et al., 2006), and may not receive other important EBAM (e.g., means safety counseling; Bryan, Stone, & Rudd, 2011). The inconsistency of EBAM may be an important factor contributing to suicide rates despite recent contact with mental health professionals (Luoma et al., 2002).

Previous research suggests that mental health professionals’ fears and anxiety may negatively influence clinical decision-making and practice (e.g., Hendin et al., 2006); however, the current findings suggest that mental health professionals’ fear of patient suicide attempt or death by suicide does not influence their ability to implement EBAM. Consistent with previous research (Pope & Tabachnick, 1993), mental health professionals in the current study were most fearful of a patient killing himself or herself, followed by having a patient attempt suicide while in their care; however, neither fear nor the interaction between fear and comfort were significant predictors of the use of EBAM. Although perceived sufficiency of suicide-focused training is negatively associated with fear of patient death by suicide (Jahn et al., 2016), some fear related to a patient’s suicide attempt or death by suicide may be an expected and normative reaction that reflects appropriate concerns given the potential risks to patient safety. As such, fear may not be a direct reflection of mental health professionals’ training or ability to implement EBAM. The association between comfort related to working with suicidal patients and the implementation of EBAM was also examined. On average, mental health professionals reported feeling “comfortable” (i.e., a score of 4.08 on a scale in which 4 represented comfortable) working with patients experiencing suicidal thoughts. This is a positive finding for the mental health field, as comfort related to working with suicidal patients was associated with greater use of EBAM during the initial therapy session. Notably, mental health professionals reported greater comfort working with patients experiencing suicidal thoughts than several other clinical issues, which may be due in part to Crisis (2018), 39(1), 55–64

the high rate at which mental health professionals treat suicidal individuals (Dexter-Mazza & Freeman, 2003; Feldman & Freedenthal, 2006).

The results from this study suggest there may be gaps in the implementation of EBAM due to mental health professionals’ comfort working with suicidal patients. Based on the current findings, some EBAM (i.e., working with suicidal patients to incorporate significant others, reducing access to means, making appropriate referrals for psychiatric evaluations) are areas where there is marked variability in implementation, which needs to be addressed to improve patient safety. Further, there is a continued need to dispel the use of non-evidence-based no-suicide contracts (Rudd et al., 2006) and increase efforts to provide mental health professionals with evidence-based training (Bernert et al., 2014).

Limitations There are limitations to the current study. The findings are based on cross-sectional data; therefore, temporal relations between comfort and EBAM cannot be established.

Additionally, the current study utilized one item to assess mental health professionals’ comfort working with suicidal patients. Future research should utilize a more comprehensive assessment of comfort that would allow for the identification of specific aspects of working with suicidal patients that may impact mental health professionals’ comfort and subsequent implementation of EBAM. Further, the use of self-report and future forecasting of behaviors may not reflect actual behaviors and may be subject to bias. It should also be noted that although difference scores were used in the current study given its high face validity, research indicates difference scores may have low internal consistency reliability (e.g., Rogosa & Willett, 1983).

Future studies may benefit from using objective measures to address potential issues related to reporting biases. Future research should also assess for additional factors that may be associated with mental health professionals’ comfort working with suicidal patients, such as the amount of clinical training and supervision received, previous experiences with suicidal patients, and the type of clinical setting (e.g., inpatient vs. outpatient).

Conclusion Taken together, these findings suggest that some EBAM practices are being used by the majority of mental health professionals; however, a portion of mental health profes© 2017 Hogrefe Publishing

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

sionals may be providing patients with less-than-adequate care by inconsistently implementing EBAM. This study also suggests that mental health professionals who feel more comfortable working with suicidal patients, regardless of their level of fear of suicide-related outcomes, are more likely to implement EBAM. Research aimed at identifying specific factors that may be associated with mental health professionals’ comfort working with suicidal patients is needed to ensure consistent implementation of EBAM and adequate patient care.

References Ahmedani, B. K., Simon, G. E., Stewart, C., Beck, A., Waitzfelder, B. E., Rossom, R., … Operskalski, B. H. (2014). Health care contacts in the year before suicide death. Journal of General Internal Medicine, 29, 870–877. doi:10.1007/s11606-014-2767-3 American Psychiatric Association. (2003). Practice guideline for the assessment and treatment of patients with suicidal behaviors.

Retrieved from http://psychiatryonline.org/pb/assets/raw/sitewide/practice_guidelines/guidelines/suicide.pdf Berman, N. C., Stark, A., Cooperman, A., Wilhelm, S., & Cohen, I. G. (2015). Effect of patient and therapist factors on suicide risk assessment. Death Studies, 39, 433–441. doi:10.1080/074811 87.2014.958630 Bernert, R. A., Hom, M. A., & Roberts, L. W. (2014). A review of multidisciplinary clinical practice guidelines in suicide prevention: Toward an emerging standard in suicide risk assessment and management, training and practice. Academic Psychiatry, 38, 585–592. doi:10.1007/s40596-014-0180-1 Bryan, C. J., Stone, S. L., & Rudd, M. D. (2011). A practical, evidence-based approach for means-restriction counseling with suicidal patients. Professional Psychology: Research and Practice, 42, 339. doi:10.1037/a0025051 Centers for Disease Control and Prevention. (2016). WISQARS: Leading causes of death. Retrieved from http://www.cdc.gov/ injury/wisqars/leading_causes_death.html Debski, J., Spadafore, C. D., Jacob, S., Poole, D. A., & Hixson, M. D. (2007). Suicide intervention: Training, roles, and knowledge of school psychologists. Psychology in the Schools, 44, 157–170. doi:10.1002/pits.20213 Dexter-Mazza, E. T., & Freeman, K. A. (2003). Graduate training and the treatment of suicidal clients: The students’ perspective. Suicide and Life-Threatening Behavior, 33, 211–218. doi:10.1521/ suli.33.2.211.22769 Feldman, B. N., & Freedenthal, S. (2006). Social work education in suicide intervention and prevention: An unmet need? Suicide and Life-Threatening Behavior, 36, 467–480. doi:10.1521/ suli.2006.36.4.467 Gould, M. S., Kalafat, J., Harris Munfakh, J. L., & Kleinman, M. (2007). An evaluation of crisis hotline outcomes part 2: Suicidal callers. Suicide and Life-Threatening Behavior, 37, 338–352. doi:10.1521/suli.2007.37.3.338 Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford.

Hendin, H., Haas, A. P., Maltsberger, J. T., Koestner, B., & Szanto, K. (2006). Problems in psychotherapy with suicidal patients.

American Journal of Psychiatry, 163, 67–72. doi:10.1176/appi. ajp.163.1.67 © 2017 Hogrefe Publishing

63

Jahn, D. R., Quinnett, P., & Ries, R. (2016). The influence of training and experience on mental health practitioners’ comfort working with suicidal individuals. Professional Psychology: Research and Practice, 47, 130–138. doi:10.1037/pro0000070 Kleespies, P. M., Penk, W. E., & Forsyth, J. P. (1993). The stress of patient suicidal behavior during clinical training: Incidence, impact, and recovery. Professional Psychology: Research and Practice, 24, 293–303. doi:10.1037/0735-7028.24.3.293 Kroll, J. (2000). Use of no-suicide contracts by psychiatrists in Minnesota. American Journal of Psychiatry, 157, 1684–1686. doi:10.1176/appi.ajp.157.10.1684 Luoma, J. B., Martin, C. E., & Pearson, J. L. (2002). Contact with mental health and primary care providers before suicide: A review of the evidence. American Journal of Psychiatry, 159, 909– 916. doi:10.1176/appi.ajp.159.6.909 Meyer, J. M., Farrell, N. R., Kemp, J. J., Blakey, S. M., & Deacon, B. J. (2014). Why do clinicians exclude anxious clients from exposure therapy? Behaviour Research and Therapy, 54, 49–53. doi:10.1016/j.brat.2014.01.004 Mitchell, S. M., Taylor, N., Jahn, D. R., Roush, J. F., Brown, S. L., Quinnett, P., & Ries, R. (2017). Mental health professionals’ anxiety and attitudes toward suicidal clients: The roles of training and self-efficacy. Manuscript submitted for publication.

Norman, G. (2010). Likert scales, levels of measurement and the “laws” of statistics. Advances in Health Sciences Education, 15, 625–632. doi:10.1007/s10459-010-9222-y Oordt, M. S., Jobes, D. A., Fonseca, V. P., & Schmidt, S. M. (2009).

Training mental health professionals to assess and manage suicidal behavior: Can provider confidence and practice behaviors be altered? Suicide and Life-Threatening Behavior, 39, 21–32. doi:10.1521/suli.2009.39.1.21 Pisani, A. R., Cross, W. F., & Gould, M. S. (2011). The assessment and management of suicide risk: State of workshop education. Suicide and Life-Threatening Behavior, 41, 255–276. doi:10.1111/ j.1943-278X.2011.00026.x Pope, K. S., & Tabachnick, B. G. (1993). Therapists’ anger, hate, fear, and sexual feelings: National survey of therapist responses, client characteristics, critical events, formal complaints, and training. Professional Psychology: Research and Practice, 24, 142–152. doi:10.1037/0735-7028.24.2.142 Quinnett, P. (1995). QPR: Certified QPR gatekeeper instructors training manual. Spokane, WA: The QPR Institute.

Risk Management Foundation of the Harvard Medical Institutions.

(2002). Guidelines for the identification, assessment, and treatment planning for suicidality. Retrieved from https://www.rmf. harvard.edu/Clinician-Resources/Guidelines-Algorithms/2003/ Suicide-Identification-Assessment-and-Treatment-Planning Rogosa, D. R., & Willett, J. B. (1983). Demonstrating the reliability the difference score in the measurement of change. Journal of Educational Measurement, 20, 335–343. doi:10.1111/ j.1745-3984.1983.tb00211.x Rothes, I. A., Henriques, M. R., Leal, J. B., & Lemos, M. S. (2014). Facing a patient who seeks help after a suicide attempt. Crisis, 35, 110–122. doi:10.1027/0227-5910/a000242 Rudd, M. D., Mandrusiak, M., & Joiner, T. E., Jr. (2006). The case against no-suicide contracts: The commitment to treatment statement as a practice alternative. Journal of Clinical Psychology, 62, 243–251. doi:10.1002/jclp.20227 Schmitz, W. M., Allen, M. H., Feldman, B. N., Gutin, N. J., Jahn, D. R., Kleespies, P. M., … Simpson, S. (2012). Preventing suicide through improved training in suicide risk assessment and care: An American Association of Suicidology Task Force report addressing serious gaps in US mental health training. Suicide and Life-Threatening Behavior, 42, 292–304. doi:10.1111/j.1943-278X.2012.00090.x Stanley, B., & Brown, G. K. (2012). Safety planning intervention: A brief intervention to mitigate suicide risk. Cognitive and Behavioral Practice, 19, 256–264. doi:10.1016/j.cbpra.2011.01.00 Crisis (2018), 39(1), 55–64

64

J. F. Roush et al., Mental Health Professionals’ Suicide Risk Assessment and ­Management Practices

Received October 14, 2016 Revision received March 23, 2017 Accepted March 26, 2017 Published online September 15, 2017

Nathanael Taylor, MA, is a clinical psychology candidate at Texas Tech University, USA. He is primarily interested in research related to suicide risk modeling, assessment, and prevention in populations with severe mental illness.

Jared F. Roush, MA, is a clinical psychology doctoral candidate at Texas Tech University, USA. He received his master’s degree in psychology from Texas Tech University. His primary research interests include suicide risk in medical settings and among individuals with severe mental illness.

Paul Quinnett, PhD, is a clinical psychologist and founder of the QPR Institute. He also currently serves as Clinical Assistant Professor in the Department of Psychiatry and Behavioral Science at the University of Washington School of Medicine, Seattle, USA. His professional interests focus on suicide prevention and training of gatekeepers and providers.

Sarah L. Brown, MA, graduated with a master’s degree in psychology from Texas Tech University, USA, and is currently a clinical psychology doctoral candidate at Texas Tech University. Her research focuses on theoretical models of suicide risk, issues related to the assessment of suicide capability, and utilizing experimental designs.

Richard Ries, MD, is Professor of Psychiatry at the University of Washington Medical School in Seattle, USA. He is Director of Outpatient Psychiatry, Dual Disorder Programs, and the Chemical Dependency Project at Harborview Medical Center. He is involved in clinical research and training physicians about substance abuse and suicide.

Danielle R. Jahn, PhD, is a licensed psychologist affiliated with Primary Care Institute in Gainesville, Florida, USA. She received her PhD in clinical psychology from Texas Tech University. Her primary research interests are in suicidology, geropsychology, and primary care–mental health integration.

Sean M. Mitchell, MA, graduated with a bachelor’s degree in psychology and Spanish and a master’s degree in psychology from Texas Tech University, USA. He is currently a clinical psychology doctoral candidate at Texas Tech University. His primary research interests include interpersonal risk factors for suicide.

Crisis (2018), 39(1), 55–64

Paul Quinnett The QPR Institute, Inc. P.O. Box 2867 Spokane, WA 99220 USA pquinnett@mindspring.com

© 2017 Hogrefe Publishing

Short Report

Investigating the Dynamics of ­Suicidal Ideation Preliminary Findings From a Study Using Ecological ­Momentary Assessments in Psychiatric Inpatients Nina Hallensleben1, Lena Spangenberg1, Thomas Forkmann2, Dajana Rath2, Ulrich Hegerl3, Anette Kersting4, Thomas W. Kallert5, and Heide Glaesmer1 Department of Medical Psychology and Medical Sociology, University of Leipzig, Germany Institute of Medical Psychology and Medical Sociology, University Hospital of RWTH Aachen University, Germany 3 Department of Psychiatry and Psychotherapy, University of Leipzig, Germany 4 Clinic of Psychosomatic Medicine and Psychotherapy, University of Leipzig, Germany 5 Psychiatric Health Care Facilities of Upper Franconia, Bayreuth, Germany 1 2

Abstract. Background: Although the fluctuating nature of suicidal ideation (SI) has been described previously, longitudinal studies investigating the dynamics of SI are scarce. Aim: To demonstrate the fluctuation of SI across 6 days and up to 60 measurement points using smartphone-based ecological momentary assessments (EMA). Method: Twenty inpatients with unipolar depression and current and/or lifetime suicidal ideation rated their momentary SI 10 times per day over a 6-day period. Mean squared successive difference (MSSD) was calculated as a measure of variability. Correlations of MSSD with severity of depression, number of previous depressive episodes, and history of suicidal behavior were examined. Results: Individual trajectories of SI are shown to illustrate fluctuation. MSSD values ranged from 0.2 to 21.7. No significant correlations of MSSD with several clinical parameters were found, but there are hints of associations between fluctuation of SI and severity of depression and suicidality. Limitations: Main limitation of this study is the small sample size leading to low power and probably missing potential effects. Further research with larger samples is necessary to shed light on the dynamics of SI. Conclusion: The results illustrate the dynamic nature and the diversity of trajectories of SI across 6 days in psychiatric inpatients with unipolar depression. Prediction of the fluctuation of SI might be of high clinical relevance. Further research using EMA and sophisticated analyses with larger samples is necessary to shed light on the dynamics of SI. Keywords: suicidal ideation, variability, depression, ecological momentary assessment

The onset, fluctuation, and remission of suicidal ideation (SI) and its transition to suicidal behavior are caused by a complex interplay of interpersonal and contextual factors (Hawton & van Heeringen, 2009; Joiner, 2005; O’Connor & Nock, 2014). Despite recent advances in the identification of distal and proximal risk factors of such alterations, evidence concerning the prediction of suicidal behavior is still limited (de Beurs, Kirtley, Kerkhof, Portzky, & O’Connor, 2015). It seems worthwhile that SI is an important predictor of suicidal behavior, because intensity of SI has been associated with previous suicide attempts (Joiner & Rudd, 2000), future suicide attempts (Lewinsohn, Rohde, & Seeley, 1996), and future suicide completion (Beck, Brown, Steer, Dahlsgaard, & Grisham, 1999). Therefore, detecting and monitoring SI seems highly relevant, especially in clinical populations with an elevated risk of suicidal behavior, and investigations of the trajectories of SI and its real-time predictors are required. © 2017 Hogrefe Publishing

Although the fluctuating nature of SI has been described (Joiner & Rudd, 2000; Witte, Fitzpatrick, Joiner, & Schmidt, 2005), most of the available data about SI have been collected from studies with cross-sectional or retrospective designs with a rather vague assessment of SI. Assuming that there are subgroups of patients with more or less rapidly cycling SI, it seems necessary to investigate the trajectories of SI with an intensive sampling.

In recent years, ecological momentary assessments (EMA) were established in the investigation of current experiences, behaviors, and emotions in the natural environment (Trull & Ebner-Priemer, 2009, 2013). EMA involves repeated sampling of subjects’ current behaviors and experiences in real time and in subjects’ natural environments (see Shiffman, Stone, & Hufford, 2008, for an overview), for example, via smartphones. Although EMA is a promising methodology, it is fairly neglected in suicidology (Davidson, Anestis, & Gutierrez, 2016; de Beurs et Crisis (2018), 39(1), 65–69 https://doi.org/10.1027/0227-5910/a000464

66

al., 2015). There are very few studies using EMA in suicidology with rather small and/or selective samples, few assessment points per day, and mostly with paper–pencil or PDA documentation (Spangenberg, Forkmann, & Glaesmer, 2015). To fill this gap, our study investigates SI using smartphone-based intensive real-time assessments.

We aim at showing preliminary results on the fluctuation of SI and its associations with several clinical indicators in psychiatric inpatients with a unipolar affective disorder.

Method Subjects and Procedures The study conducted between September 2015 and March 2016 included 20 psychiatric inpatients (from an ongoing study1), with a primary diagnosis of a unipolar affective disorder (major depression, dysthymia) and current and/or lifetime SI. Participants were included if they had a score above the cut-off value in the Suicide Behaviors Questionnaire Revised (SBQ-R) and/or reported SI in the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I). Of the participants, 80% (n = 16) were female, and the participants’ age was between 23 and 58 years (M = 35.9, SD = 9.3). After baseline assessment, participants underwent a 6-day EMA assessment with 10 randomly distributed assessment points per day using an experience-sampling software on smartphones (movisens GmbH) resulting in a maximum of 60 assessments per participant. All participants provided written informed consent, and the study was approved by the ethics committee of the Medical Faculty of the University of Leipzig, Germany.

Measures Participants underwent a baseline assessment including: the SCID-I (German version; Wittchen, Zaudig, & Fydrich, 1997) to confirm the diagnosis of unipolar affective disorders and to assess SI; the Rasch-based Depression Screening (DESC; Forkmann et al., 2009, 2010) – a 10-item self-report instrument to assess severity of depression on a 5-point Likert scale from 0 (never) to 4 (always), with Cronbach’s α =.92, sum scores ≥12 are indicative of a current depressive episode; and SBQ-R (Osman et al., 2001) – a 4-item measure assessing different aspects of suicidality, with Cronbach’s α = .87.

N. Hallensleben et al., Investigating the Dynamics of Suicidal Ideation

During the EMA phase, SI was assessed with four items; two covered more passive SI (e.g., “At the moment I feel that life is not worth living”) and two covered active SI (e.g., “At the moment I’m thinking about killing myself ”).

Each item was rated on a 5-point scale from 1 (not at all) to 5 (very much). To analyze SI, a total score of these four items was computed (range = 4–20). The EMA signals occurred randomly between 8:00 a.m. and 8:00 p.m. with a minimum interval of 30 min between alarms.

Data Analysis To illustrate the dynamic nature of SI, we plotted the fluctuation of SI across 60 assessment points for each participant. The plots were prepared using R (R Core Team, 2015), ggplot2 package (Wickham, 2009). To describe the variability of SI across time, we calculated the mean squared successive difference (MSSD). The MSSD represents point-to-point variability in time series (Woyshville, Lackamp, Eisengart, & Gilliland, 1999). It is a sort of sum score with arbitrary units and with higher values indicating higher fluctuation. Of the 20 participants, 19 missed at least one alarm with a maximum of 12 missing assessment points corresponding to a maximum of 20% missing data per participant. Missing data were handled with a procedure reported and validated by Woyshville et al. (1999); that is, erasing missing data points, shortening the time series by the missing points. Nonparametric correlation coefficients (Spearman’s ρ) were calculated for the association of MSSD with parameters of depression (DESC score and number of depressive episodes in SCID-I) and history of suicidal behavior (SBQ-R score).

Results All participants had a DESC score (depressiveness) above the proposed cut-off of ≥ 12 (range = 15–35, M = 24.5, SD = 6.1). Half of them (n = 10) exhibited an SBQ-R score (history of suicidal behavior) above the cut-off of ≥ 8 (range = 5–16, M = 9.2, SD = 3.3).

The MSSD values ranged from 0.2 to 21.7 (M = 5.1, SD = 6.1). Figure 1 shows the individual trajectories of SI (total score of four items) over 60 assessment points for the 20 participants, sorted by level of MSSD, including the individual MSSD values.

The nonparametric correlations between MSSD and severity of depression (DESC total score) was ρ = .362

1 Investigating dynamics and acute risk factors of suicidal ideation: A real-time assessment using experience sampling method (ESM) in inpatients with major depression (see http://medpsy.uniklinikum-leipzig.de/medpsych.site,postext,projekte-versorgungsforschung,a_id,5450.html) Crisis (2018), 39(1), 65–69

© 2017 Hogrefe Publishing

N. Hallensleben et al., Investigating the Dynamics of Suicidal Ideation

67

Figure 1. Trajectories of SI for each of the 20 participants, sorted by level of MSSD. Assessment points range from 1 to 60 (over a period of 6 days) and for SI from 4 to 20 (total score of 4 SI items). Individual MSSD values as indicator of variability are displayed in italics. SI = suicidal ideation.

MSSD = mean squared successive difference.

(p = .117), with the number of depressive episodes (SCID-I) at ρ = .069 (p = .780) and with different aspects of suicidality (SBQ-R total score) at ρ = .345 (p = .161).

Discussion Our preliminary study results illustrate that it is feasible to collect data on suicidality using EMA. Furthermore, it demonstrates the dynamic nature of SI in psychiatric inpatients with unipolar depression. Additionally, it clearly underlines the diversity of trajectories of SI. The main limitation of this study is the small sample size leading to lower power and, hence, probably missing potential effects. Therefore, it is meant to be an initial exploratory approach to the complex and largely unexplored issue of the dynamics of SI in psychiatric inpatients.

Although we failed to find significant associations between the MMSD score and several clinical parameters, ρ values of >.30 may hint at an association between fluctuation of SI with severity of depression and suicidality, © 2017 Hogrefe Publishing

which may fail to reach significance because of the small sample size. This would support the finding by Witte et al.

(2005) and Witte, Fitzpatrick, Warren, Schatschneider, and Schmidt (2006) of a relationship between variability in SI and previous history of suicide attempts. With reference to affect regulation literature, Witte and colleagues (2006) state that a variable pattern of SI might be experienced to be far more distressing than a stable pattern as one possible explanation for this association. If so, these mechanisms might strengthen each other in terms of a vicious circle. This highlights the potential relevance of the dynamics of SI for suicidality – which may be even higher than the overall level of risk factors (Witte et al., 2006).

Our study provides evidence that EMAs are able to shed light on the complex dynamics of SI and raises important questions as well as directions for further research: With regard to the dynamics of SI, further investigations should focus on the relationships between fluctuation of SI, depressiveness, and suicidality in order to determine the clinical relevance of these dynamics. Regarding the diversity of SI trajectories – is it possible to differentiate types of Crisis (2018), 39(1), 65–69

68

trajectories of SI? If so, are there predictors of such trajectories and what is their clinical impact? What therapeutic implications can be derived? Prospectively, larger samples and more sophisticated statistical approaches are needed for a deeper understanding of the dynamics of suicidality.

The data reported here are from an ongoing study. We are planning to include up to 75 patients, which will give us the opportunity to run complex analyses on different aspects of the fluctuation of SI in psychiatric inpatients with unipolar depression.

References Beck, A. T., Brown, G. K., Steer, R. A., Dahlsgaard, K. K., & Grisham, J. R. (1999). Suicide ideation at its worst point: A predictor of eventual suicide in psychiatric outpatients. Suicide and Life-Threatening Behavior, 29(1), 1–9.

Davidson, C. L., Anestis, M. D., & Gutierrez, P. M. (2016). Ecological momentary assessment is a neglected methodology in suicidology. Archives of Suicide Research, 21, 1–11. de Beurs, D., Kirtley, O., Kerkhof, A., Portzky, G., & O’Connor, R. C. (2015). The role of mobile phone technology in understanding and preventing suicidal behavior. Crisis, 36(2), 79–82.

Forkmann, T., Boecker, M., Wirtz, M., Eberle, N., Westhofen, M., Schauerte, P., … Norra, C. (2009). Development and validation of the Rasch-based Depression Screening (DESC) using Rasch Analysis and structural equation modelling. Journal of Behavior Therapy and Experimental Psychiatry, 40, 468–478.

Forkmann, T., Boecker, M., Wirtz, M., Glaesmer, H., Brähler, E., Norra, C., & Gauggel, S. (2010). Validation of the Rasch-based Depression Screening in a large scale German general population sample. Health and Quality of Life Outcomes, 8:105.

Hawton, K., & van Heeringen, K. (2009). Suicide. Lancet, 373(9672), 1372–1381.

Joiner, T. E. (2005). Why people die by suicide. Cambridge, MA: Harvard University Press.

Joiner, T. E., & Rudd, M. D. (2000). Intensity and duration of suicidal crises vary as a function of previous suicide attempts and negative life events. Journal of Consulting and Clinical Psychology, 68(5), 909–916.

Lewinsohn, P. M., Rohde, P., & Seeley, J. R. (1996). Adolescent suicidal ideation and attempts: Prevalence, risk factors, and clinical implications. Clinical Psychology: Science and Practice, 3(1), 25–46. movisensXS (Version 0.7.4162) [Computer software]. Karlsruhe, Germany: movisens GmbH. O’Connor, R. C., & Nock, M. K. (2014). The psychology of suicidal behaviour. Lancet Psychiatry, 1(1), 73–85.

Osman, A., Bagge, C. L., Gutierrez, P. M., Konick, L. C., Kopper, B. A., & Barrios, F. X. (2001). The Suicidal Behaviors Questionnaire-Revised (SBQ-R): Validation with clinical and nonclinical samples.

Assessment, 8(4), 443–454.

R Core Team. (2015). R: A language and environment for statistical computing. Vienna, Austria: R Foundation of Statistical Computing. Retrieved from http://www.R-project.org/ Shiffman, S., Stone, A. A., & Hufford, M. R. (2008). Ecological momentary assessment. Annual Review of Clinical Psychology, 4, 1–32.

Spangenberg, L., Forkmann, T., & Glaesmer, H. (2015). Investigating dynamics and predictors of suicidal behaviors using ambulaCrisis (2018), 39(1), 65–69

N. Hallensleben et al., Investigating the Dynamics of Suicidal Ideation

tory assessment. Neuropsychiatrie: Klinik, Diagnostik, Therapie und Rehabilitation, 29(3), 139–143.

Trull, T. J., & Ebner-Priemer, U. W. (2009). Using experience sampling methods/ecological momentary assessment (ESM/EMA) in clinical assessment and clinical research: Introduction to the special section. Psychological Assessment, 21(4), 457–462.

Trull, T. J., & Ebner-Priemer, U. W. (2013). Ambulatory Assessment.

Annual Review of Clinical Psychology, 9(9), 151–176.

Wickham, H. (2009). ggplot2: Elegant graphics for data analysis.

New York, NY: Springer.

Wittchen, H.-U., Zaudig, M., & Fydrich, T. (1997). Strukturiertes Klinisches Interview für DSM-IV [Structured clinical interview for DSM-IV]. Göttingen, Germany: Hogrefe.

Witte, T. K., Fitzpatrick, K. K., Joiner, T. E., & Schmidt, N. B. (2005).

Variability in suicidal ideation: A better predictor of suicide attempts than intensity or duration of ideation? Journal of Affective Disorders, 88(2), 131–136.

Witte, T. K., Fitzpatrick, K. K., Warren, K. L., Schatschneider, C., & Schmidt, N. B. (2006). Naturalistic evaluation of suicidal ideation: Variability and relation to attempt status. Behaviour Research and Therapy, 44(7), 1029–1040.

Woyshville, M. J., Lackamp, J. M., Eisengart, J. A., & Gilliland, J. A. (1999). On the meaning and measurement of affective instability: Clues from chaos theory. Biological Psychiatry, 45(3), 261–269.

Received May 19, 2016 Revision received January 18, 2017 Accepted February 8, 2017 Published online May 3, 2017

About the authors Nina Hallensleben is a PhD student at the Department of Medical Psychology and Medical Sociology, University of Leipzig, Germany, and soon-to-be cognitive behavioral therapist. She is currently investigating the dynamics of suicidal ideation and acute risk factors for suicidality by means of ecological momentary assessment (EMA).

Dr. Spangenberg is soon-to-be cognitive behavioral therapist as well as a post-doctoral researcher at the Department of Medical Psychology and Medical Sociology, University of Leipzig, Germany. Besides psychometric investigations, she focuses on suicidality research, for example, with population-based studies and longitudinal studies in clinical populations, on the short-term prediction of suicidality.

Dr. Forkmann is a psychotherapist, postdoctoral researcher, and vice-director of the Institute of Medical Psychology and Medical Sociology, University Hospital of RWTH Aachen University, Germany. He has published research on the epidemiology, diagnostics, and treatment of suicidality and depression.

Dajana Rath is a PhD student working as a soon-to-be cognitive behavioral therapist as well as a researcher at the Institute of Medical Psychology and Medical Sociology, University Hospital of RWTH Aachen University, Germany. She is currently investigating (acute) risk factors of suicidality (e.g., by means of ecological momentary assessment).

Dr. Hegerl is Professor of Psychiatry and Director of the Department of Psychiatry and Psychotherapy, University of Leipzig, Germany. He is also Head of the German Depression Foundation and the European Alliance against Depression. Amongst others, his research focus is on depression and suicide prevention. © 2017 Hogrefe Publishing

69

N. Hallensleben et al., Investigating the Dynamics of Suicidal Ideation

Dr. Kersting is Professor for Psychosomatic Medicine and Director of the Clinic of Psychosomatic Medicine and Psychotherapy, University of Leipzig, Germany. Her research field focuses on Internet psychotherapy, eating disorders, pathological grief, and posttraumatic stress disorders.

Dr. Kallert is Medical Director of all psychiatric health-care facilities in the district of Upper Franconia, Germany. He is Professor of Psychiatry at the Dresden University of Technology, Germany. He has received 20 research grants focused on the evaluation of mental health services and complex interventions. Among his awards is the Honorary Membership of the World Psychiatric Association.

Dr. Glaesmer is a trained psychologist and psychotherapist (CBT). She is the vice head of the Department of Medical Psychology and Medical

Š 2017 Hogrefe Publishing

Sociology of the Medical School of the University of Leipzig, Germany.

Beside her research on psychotraumatology and migration, she conducts several research projects in suicidology, including psychometric studies, population-based studies, and longitudinal studies in clinical populations about the prediction of suicidality.

Nina Hallensleben Department of Medical Psychology and Medical Sociology University of Leipzig Philipp-Rosenthal-Str. 55 04103 Leipzig Germany nina.hallensleben@medizin.uni-leipzig.de

Crisis (2018), 39(1), 65â&#x20AC;&#x201C;69

Short Report

Framing Suicide – Investigating the News Media and Public’s Use of the Problematic Suicide Referents Freitod and Selbstmord in GermanSpeaking Countries Florian Arendt Department of Communication Science and Media Research, University of Munich (LMU), Germany Abstract. Background: In German-speaking countries, suicide experts recommend not using the suicide referents Freitod and Selbstmord, as their associative meanings relate to problematic concepts such as free will and crime. Aims: To investigate which terms – the neutral and recommended Suizid or Freitod and Selbstmord – have dominated news coverage and to reveal what terms the public actually used. Method: A retrospective database study was undertaken on data from the period 2004–2016. First, we investigated how frequently the terms were used in news coverage via an automated content analysis. Second, we investigated how often individuals used the terms for information-seeking via Google’s search engine, since it can be used as an indicator of the popularity of a given term within a given period. Results: Analyses revealed that Selbstmord was the most frequently used term in the news and by the public. Importantly, the use of Suizid increased in both datasets, nearly approaching the Selbstmord level in the later years. Although on a low level, the highly problematic term Freitod has also been in regular use. Conclusion: Media interventions should continue trying to increase journalists’ awareness so that they use appropriate terms when reporting on suicide.

Keywords: suicide prevention, framing, media, news, Google Trends

Over 800,000 people die from suicide each year worldwide (World Health Organization, 2014). The news media are considered a key factor for suicide prevention (Mann et al., 2005), as responsible reporting may potentially decrease suicidal behavior among audiences (Niederkrotenthaler et al., 2010). Thus, media guidelines have been developed to increase responsible reporting (Bohanna & Wang, 2012; Pirkis, Blood, Beautrais, Burgess, & Skehan, 2006). Among other elements of journalistic reporting, these guidelines emphasize the language used when referring to suicide. In the English language, for example, the recommendation is not to refer to suicide as successful or as a failed attempt because these terms may elicit (undesired) associative meanings. In the German language, the recommendation is not to use the Freitod and Selbstmord referents because of their ability to elicit (undesired) associative meanings related to concepts such as free will and crime (Tomandl, Sonneck, & Stein, 2008).

We present the results of a retrospective database study in which we investigated how frequently the Suizid (recCrisis (2018), 39(1), 70–73 https://doi.org/10.1027/0227-5910/a000467

ommended) as well as Freitod and Selbstmord (not recommended) referents appeared in news coverage. As a supplement, we investigated how the public had been using these terms in the same period. Since previous research has theorized on the possible effects of words with different associative meanings by using the framing concept, we first provide a succinct review on framing.

Framing Framing occurs when some aspects of a phenomenon are selected and made more salient than others so that a specific problem definition, causal interpretation, moral evaluation, or treatment recommendation will be promoted (Entman, 1993). Consequently, the news media may select, accentuate, or exclude some aspects of the suicide phenomenon that can evoke a specific, sometimes substantially constricted perspective on suicide. Yet, such framing effects are not knowledge effects (Scheufele & © 2017 Hogrefe Publishing

71

F. Arendt, Framing Suicide

Tewksbury, 2009): Knowledge gain results from a learning process where news consumers acquire something new on an issue. Conversely, framing effects occur when words, phrases, or images suggest a particular interpretation.

Even single words can elicit framing effects: In a randomized controlled trial (Simon & Jerit, 2007), participants read an article describing the so-called partial-birth abortion debate on a law prohibiting a form of late-term abortion. In the first condition, the term fetus was used throughout the article, while baby was used in the second condition. After reading the newspaper article, participants were asked to summarize it in their own words and to fill out a survey. Participants reading the fetus-frame used the word fetus more often and analyses revealed that reading this frame reduced support for the partial-birth abortion ban.

In the German language, the recommendation is to use the neutral term Suizid when referring to suicide. Importantly, two other words are generally considered problematic (Tomandl et al., 2008): Freitod (English: “free death”) refers to the concept of free will. The Freitod frame implicitly conveys how individuals are free, as they can freely, and thus probably rationally, choose from several behavioral options. Yet, research has shown that individuals in a suicidal crisis exhibit an emotional narrowing that does not support the free will and rational choice assumptions (Sonneck, Kapusta, Tomandl, & Voracek, 2007). Further, this framing may also lead to a heroic and/or romanticizing interpretation of the suicidal act, a further undesired outcome of suicide prevention (see Bohanna & Wang, 2012; Pirkis et al., 2006).

The suggestion is also that the Selbstmord (English: “self-murder”) referent should not be used as it refers to crime based on a diffuse, at least partly religion-laden understanding of the suicidal act (Tomandl et al., 2008); thus it is viewed as problematic because suicide is not a murder in the criminological sense. A crime-loaded term may make a neutral perspective on suicide less likely, ultimately contributing to a biased view on suicide (Sonneck et al., 2007).

Method This study is a retrospective database study. We investigated which term (Suizid, Freitod, Selbstmord) dominated the German-speaking news coverage over the past decade by using an automated content analysis. Furthermore, we tested how the public had been using these terms by using online information-seeking as an indicator.

© 2017 Hogrefe Publishing

News Coverage: Automated Content ­Analysis We used an archive of German-speaking newspapers (Austrian Press Agency, 2017) including 61 German-speaking newspapers. The archive includes important newspapers from Austria, Germany, and Switzerland. We searched the archive for articles including the terms Suizid, Freitod, and Selbstmord. For each year of the observation period (2004– 2016), we counted how many articles the terms appeared in.

The Public’s Use: Online InformationSeeking as a Proxy We investigated how often individuals used the three terms for information-seeking via the search engine Google. This is an indicator of which term people also use in everyday life. The data are based on the search terms Suizid, Freitod, and Selbstmord that users entered into the Google search engine. We downloaded the raw data from Google Trends (i.e., Google’s tool to access search-volume data). The observation period was 2004–2016 as search-volume data are not available prior to 2004.

Google Trends data reflect the frequency of the three search terms entered into Google relative to the total search volume. Raw data are normalized scores. A value of 100 is defined as the peak search volume for the specified time period and search terms. For example, a query share value of 30 represents 30% of the highest observed search proportion during the observation period for the given search term. For reasons of comparability to the news coverage analysis, we used query share per year as the target measure.

Although Google Trends data have limitations (Lazer, Kennedy, King, & Vespignani, 2014), research has shown that query share data are reliable predictors of information-seeking and can be used as an indicator of how popular a given term is within a given period (Nuti et al., 2014).

Results News Coverage: Journalists’ Use As can be seen in Figure 1, Selbstmord was the most frequently used search term. Importantly, there was a strong increase in the use of Suizid between 2004 and 2016, with Suizid nearly approaching the Selbstmord level in the later years. Although at a low level, Freitod was also used regCrisis (2018), 39(1), 70–73

FRAMING SUICIDE

11

72

N u m b e r o f A r tic le s

8000

S e lb s tm o rd F re ito d

6000

S u iz id

F. Arendt, Framing Suicide

Figure 1. Number of newspaper articles containing the terms Suizid, Freitod, and Selbstmord per year. Journalists increasingly use the term Suizid, consistent with recommendations in the media guidelines.

4000

2000

6

5

1

4

0

12

2

2

0

1

3 2

0

1

2 2

0

1

1 2

0

1

0 2

0

1

9 2

0

1

8 2

0

0

7 2

0

0

6 2

0

0

5

0

2

0

2

2

0

FRAMING SUICIDE

0

0

0

4

0

Y ear

S e a r c h Q u e r y V o lu m e

100 Figure Figure 1. Number of newspaper articles containing the terms Suizid, Freitod, and Selbstmord per 2.

S e lb s tm o rd

Search query volume of the terms Suizid, Freitod, and Selbstmord as a function of time.

F re ito d year. Journalists8 0increasingly use the term Suizid, consistent with recommendations in the media

guidelines.

S u iz id

60 40 20

6

5

1 0

2

2

0

1

4

3

1 0

2

2

0

1

2

1

1 0

2

2

0

1

0

2

0

1

9

2

0

0

8

2

0

0

7

2

0

0

6

2

0

0

5 0

0 2

2

0

0

4

0

Y ear

German-speaking 2 yearsand (2011 Figure 2. ularly Searchinquery volume of the newspapers. terms Suizid,In Freitod, Selbstmord as a function of Discussion

time.

and 2014), Freitod even passed the 1,000-article mark.

The Public’s Use: Online InformationSeeking Figure 2 represents the search query volume for each of the three search terms as a function of time. As can be seen in the figure, the total search volume decreased between 2004 and 2016. This general trend indicates that web users tended to search for suicide less often in more recent times. More importantly, the relative difference between the three search terms per year changed over time, similar to the pattern revealed for news coverage. Although Selbstmord still led the query share in 2016, Suizid nearly approached the Selbstmord level. This convergence was mostly due to the strong decrease in Selbstmord, with Freitod having the smallest query share.

Crisis (2018), 39(1), 70–73

Suicide experts recommend not using the Freitod and Selbstmord referents but instead recommend using Suizid (Tomandl et al., 2008). The news media are pervasive and may thus influence the dominant associative meanings of a society’s perspective on suicide. Analyses revealed that the use of Suizid steadily increased, nearly approaching the Selbstmord level in the later years. Although at a low level, the problematic term Freitod was also used regularly.

As a supplement, we found a similar trend in the public’s use of these terms by using online information-seeking via the search engine Google as an indicator.

Limitations The study has its limitations. First, newspaper archives do not provide the same newspapers every year, as, for example, new newspapers are regularly added to the archives.

This may influence the total number of suicide articles per year. However, we focused on the relative difference be© 2017 Hogrefe Publishing

73

F. Arendt, Framing Suicide

tween the number of articles using one of the three terms within a year. Thus, even if there is some uncertainty in the interpretation of year-by-year variations using absolute frequency scores, the interpretation of the relative difference is unambiguous. Second, Google Trends data have well-known limitations (Lazer et al., 2014) such as there being a difference between the social structure of web users and the real (offline) population. This possibly influenced the generalizability of our findings.

Conclusion The increase in the news media and public’s use of the term Suizid can be deemed as a beneficial change on a societal level. Unfortunately, Selbstmord and (the possibly even more detrimental) Freitod are still in regular use. Therefore, media interventions should continue trying to reduce the use of these terms by increasing journalists’ awareness.

References Austrian Press Agency. (2017). APA DeFacto. Retrieved from http:// www.apa-defacto.at Bohanna, I., & Wang, X. (2012). Media guidelines for the responsible reporting of suicide. Crisis, 33, 190–198.

Entman, R.M. (1993). Framing: Towards clarification of a fractured paradigm. Journal of Communication, 43(4), 51–58.

Lazer, D., Kennedy, R., King, G., & Vespignani, A. (2014). The parable of Google flu: Traps in big data analysis. Science, 343, 1203– 1205. doi:10.1126/science.1248506 Mann, J. J., Apter, A., Bertolote, J., Beautrais, A., Currier, D., Hass, A., … Hendin, H. (2005). Suicide prevention strategies: A systematic review. Journal of the American Medical Association, 294, 2064–2074.

Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., … Sonneck, G. (2010). The role of media re-

© 2017 Hogrefe Publishing

ports in completed and prevented suicide: Werther versus Papageno effects. British Journal of Psychiatry, 197, 234–243.

Nuti, S., Wayda, B., Ranasinghe, I., Wang, S., Dreyer, R., Chen, S., & Murugiah, K. (2014). The use of Google Trends in health care research: A systematic review. PLoS ONE, 9, e109583. doi:10.1371/ journal.pone.0109583 Pirkis, J. E., Blood, R. W., Beautrais, A., Burgess, P. M., & Skehan, J. (2006). Media guidelines on the reporting of suicide. Crisis, 27, 82–87.

Scheufele, D., & Tewksbury, D. (2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of Communication, 57, 9–20.

Simon, A., & Jeirt, J. (2007). Toward a theory relating political discourse, media, and public opinion. Journal of Communication, 57, 254–271.

Sonneck, G., Kapusta, N., Tomandl, G., & Voracek, M. (2007). Krisenintervention und Suizidverhütung [Crisis intervention and suicide prevention]. Vienna, Austria: UTB. Tomandl, G., Sonneck, G., & Stein, C. (2008). Leitfaden zur Berichterstattung über Suizid [Guidelines for news coverage about suicides]. Vienna, Austria: Kriseninterventionszentrum. Retrieved from http://www.suizidpraevention.at/downloads.html World Health Organization. (2014). Preventing suicide: A global imperative. Geneva, Switzerland: Author.

Received February 1, 2017 Accepted February 16, 2017 Published online July 27, 2017 Florian Arendt, PhD, is a postdoctoral researcher at the Department of Communication Science and Media Research (LMU Munich), Germany.

His primary research interests lie in the field of media processes and effects, in particular media priming, implicit cognition, stereotyping, and health communication.

Florian Arendt Department of Communication Science and Media Research University of Munich (LMU) Oettingenstraße 67 80538 Munich Germany florian.arendt@ifkw.lmu.de

Crisis (2018), 39(1), 70–73

News, Announcements, and IASP Announcements The 51st Annual Conference of the American Association of Suicidology (AAS), entitled “Integrating Science, Experience and Political Will: Informed Action to Prevent Suicide,” is taking place April 18–21, 2018, in Washington, DC, USA. For more information go to http://www. suicidology.org/Annual-Conference/51st-Annual-Con​ ference The 30th Annual Healing After Suicide Conference of the American Association of Suicidology (AAS) is taking place April 21, 2018, in Washington, DC, USA. For more information go to http://www.suicidology.org/ Annual-­Conference/30th-Annual-Healing-Conference The IASP 2018 Asia Pacific Conference is taking place May 2–5, 2018, in the Bay of Islands, New Zealand. For more information go to https://iasp.info/newzealand/

Crisis (2018), 39(1), 74–75 https://doi.org/10.1027/0227-5910/a000537

The 2018 Conference of the International Society for the Study of Self Injury (ISSS) is taking place June 22–23, 2018 in The Hague, Netherlands. For more information go to http://itriples.org/2018-meeting/ The Bi-Annual Australian Grief and Bereavement Conference of the Australian Centre for Grief and Bereavement is taking place June 2018 in Sydney, NSW, Australia.

For more information go to https://www.grief.org.au The 17th European Symposium on Suicide and Suicidal Behaviour (ESSSB) is taking place September 5–8, 2018, in Ghent, Belgium. For more information go to http://www.esssb17.org

© 2018 Hogrefe Publishing

75

News, Announcements, Reviewers, and IASP

International Association for Suicide Prevention Would you like to join IASP?

−− to carry out programs of research, especially those which can be pursued through international joint cooperation.

IASP is a nonprofit organization for those interested and working in suicide prevention. The membership consists of individuals (clinicians, scientists, crisis workers, volunteers and persons who have lost a family member by suicide) and community, national and international organizations. At this moment the membership extends over 50 countries and is affiliated with the World Health Organization as the key organization concerned with suicide prevention.

The activities of the association focus on the importance of sharing information, research and knowledge in order to address the issues of suicide and its prevention. The biennial congress, the bimonthly journal Crisis, biannual newsletter and the website serves as a medium for international exchange and a directory of members involved in suicidology is made available.

Candidates for IASP membership may apply directly to the Central Administration Office membership@iasp.info or join IASP online at http://www.iasp.info Membership consists of: −− Regular membership for individuals regularly engaged in suicide studies, clinical management of suicidal patients, and/or suicide prevention activities, and who are interested in furthering the work of the Association. −− Organizational membership is for national or international voluntary or other nonprofit organizations (incorporated or unincorporated) working in the field of suicide studies, clinical management of suicidal patients and/or suicide prevention activities under the terms of their governing document. −− Associate membership is for individuals and organizations who are interested in working toward the goals of the Association, but who are not qualified for regular membership.

Please contact the IASP Central Administration Office at membership@iasp.info.; on the web at http://www.iasp.info

The main objectives of IASP are: −− to provide a common platform for all representatives of the many professions who are engaged in the field of suicide prevention and crisis intervention; −− to allow interchange of acquired experience in this area in various countries, especially through the exchange of literature; −− to promote the establishment of national organizations for suicide prevention; −− to facilitate the wider dissemination of the fundamentals of effective suicide prevention to professional groups and to the general public; −− to arrange for specialized training of selected persons in the area of suicide prevention in selected training centers;

IASP membership fees (include a subscription to Crisis) Zone

Crisis

Individual Membership

Organizational Membership

1

Hard copy

US $190.00 (early bird US $180.50, 3 years US $515.00)

US $235.00 (early bird US $225.00, 3 years US $635.00)

Online only

US $162.00 (early bird US $153.00, 3 years US $438.00)

US $200.00 (early bird US $190.00, 3 years US $543.00)

Hard copy

US $160.00 (early bird US $152.00, 3 years US $430.00)

US $180.00 (early bird US $170.00, 3 years US $485.00)

Online only

US $136.00 (early bird US $129.00, 3 years US $368.00)

US $153.00 (early bird US $145.00, 3 years US $414.00)

Hard copy

US $135.00 (early bird US $128.00, 3 years US $365.00)

US $160.00 (early bird US $150.00, 3 years US $430.00)

Online only

US $115.00 (early bird US $109.00, 3 years US $310.00)

US $136.00 (early bird US $129.00, 3 years US $368.00)

Hard copy

US $115.00 (early bird US $109.00, 3 years US $310.00)

US $125.00 (early bird US $120.00, 3 years US $340.00)

Online only

US $98.00 (early bird US $93.00, 3 years US $239.00)

US $106.00 (early bird US $101.00, 3 years US $288.00)

Online only

US $115.00 (early bird US $109.00, 3 years US $310.00)

For Advanced Organization Membership fees go to http://www.iasp.info/application.php

2 3 4 All zones: students, volunteers, & retirees

IASP Executive Committee 2017–2019 President: Vice President 1: Vice President 2: Vice President 3:

Dr. Murad Khan Prof. Jane Pirkis Prof. Rory O’Connor Prof. Ping Qin

General Secretary: Dr. Daniel Reidenberg Treasurer: Dr. Thomas Niederkrotenthaler Representative of the Council of Presidents: Prof. Ella Arensman

IASP Council of National Representatives 2017–2019 Co-Chairs: Prof. Madelyn Gould and Prof. Jie Zhang For a current listing of the IASP Council of National Representatives please refer to https://www.iasp.info/council_of_national_represent​ atives.php © 2018 Hogrefe Publishing

Crisis (2018), 39(1), 74–75

Instructions to Authors – Crisis: The Journal of Crisis Intervention and Suicide Prevention Crisis – The Journal of Crisis Intervention and Suicide Prevention is an international periodical that publishes original articles on suicidology and crisis intervention. Papers presenting basic research as well as practical experience in the field are welcome.

Crisis also publishes potentially life-saving information for all those involved in crisis intervention and suicide prevention, making it important reading for clinicians, counselors, hotlines, and crisis intervention centers.

Crisis: The Journal of Crisis Intervention and Suicide Prevention publishes the following types of articles: Research Trends, Short Reports, and Clinical Insights.

Manuscript Submission: All manuscripts should be submitted electronically at http://www.editorialmanager.com/cri Detailed instructions to authors are provided at http://www. hogrefe.com/j/cri Copyright Agreement: By submitting an article, the author confirms and guarantees on behalf of him-/herself and any coauthors that he or she holds all copyright in and titles to the submitted contribution, including any figures, photographs, line drawings, plans, maps, sketches and tables, and that the article and its contents do not infringe in any way on the rights of third parties. The author indemnifies and holds harmless the publisher from any third-party claims. The author agrees, upon acceptance of the article for publication, to transfer to the publisher on behalf of him-/herself and any coauthors the exclusive right to reproduce and distribute the article and its contents, both physically and in nonphysical, electronic, and other form, in the journal to which it has been submitted and in other independent publications, with no limits on the number of copies or on the form or the extent of the distribution. These rights are transferred for the duration of copyright as defined by international law. Further-

Crisis (2018), 39(1)

more, the author transfers to the publisher the following exclusive rights to the article and its contents: 1. The rights to produce advance copies, reprints, or offprints of the article, in full or in part, to undertake or allow translations into other languages, to distribute other forms or modified versions of the article, and to produce and distribute summaries or abstracts.

2. The rights to microfilm and microfiche editions or similar, to the use of the article and its contents in videotext, teletext, and similar systems, to recordings or reproduction using other media, digital or analog, including electronic, magnetic, and optical media, and in multimedia form, as well as for public broadcasting in radio, television, or other forms of broadcast.

3. The rights to store the article and its content in machine-readable or electronic form on all media (such as computer disks, compact disks, magnetic tape), to store the article and its contents in online databases belonging to the publisher or third parties for viewing or downloading by third parties, and to present or reproduce the article or its contents on visual display screens, monitors, and similar devices, either directly or via data transmission.

4. The rights to reproduce and distribute the article and its contents by all other means, including photomechanical and similar processes (such as photocopying or facsimile), and as part of so-called document delivery services.

5. The right to transfer any or all rights mentioned in this agreement, as well as rights retained by the relevant copyright clearing centers, including royalty rights to third parties.

Online Rights for Journal Articles: Guidelines on authors’ rights to archive electronic versions of their manuscripts online are given in the document “Guidelines on sharing and use of articles in Hogrefe journals” on the journal’s web page at www.hogrefe. com/j/cri.

February 2016

© 2018 Hogrefe Publishing

An innovative and highly effective brief therapy for suicidal patients “ASSIP is perhaps the most significant innovation we have seen in the assessment and treatment of suicidal risk...”

David A. Jobes, PhD, Professor of Psychology, The Catholic University of America, Washington, DC, USA Past President, American Association of Suicidology

Konrad Michel / Anja Gysin-Maillart

ASSIP – Attempted Suicide Short Intervention Program A Manual for Clinicians

2015, x + 114 pp.

US $59.00 / € 41.95 ISBN 978-0-88937-476-8 Also available as eBook Attempted suicide is the main risk factor for suicide. The Attempted Suicide Short Intervention Program (ASSIP) described in this manual is an innovative brief therapy that has proven in published clinical trials to be highly effective in reducing the risk of further attempts. ASSIP is the result of the authors’ extensive practical experience in the treatment of suicidal individuals. The emphasis is on the therapeutic alliance with the suicidal patient, based on an initial patientoriented narrative interview.

www.hogrefe.com

The four therapy sessions are followed by continuing contact with patients by means of regular letters.

This clearly structured manual starts with an overview of suicide and suicide prevention, followed by a practical, step-by-step description of this highly structured treatment. It includes numerous checklists, handouts, and standardized letters for use by health professionals in various clinical settings.

A unique and comprehensive handbook presenting the state-of-the-art in suicide bereavement support

Karl Andriessen / Karolina Krysinska / Onja T. Grad (Editors)

Postvention in Action

The International Handbook of Suicide Bereavement Support 2017, xviii + 424 pp.

US $94.00 / â&#x201A;Ź 74.95 ISBN 978-0-88937-493-5 Suicide is not merely the act of an individual; it always has an effect on others and can even increase the risk of suicide in the bereaved. The International Association for Suicide Prevention, the World Health Organisation, and others have recognized postvention as an important strategy for suicide prevention. This unique and comprehensive handbook, authored by nearly 100 international experts, including researchers, clinicians, support group facilitators, and survivors, presents the state-of-theart in suicide bereavement support.

The first part examines the key concepts and the processes that the bereaved experience and illustrates them with illuminating clinical vignettes. The second and third parts

www.hogrefe.com

look in detail at suicide bereavement support in all the most relevant settings (including general practices, the workplace, online and many others) as well as in specific groups (such as health care workers).

In the concluding section, the support provided for those bereaved by suicide in no less than 23 countries is explored in detail, showing that postvention is becoming worldwide strategy for suicide prevention. These chapters provide useful lessons and inspiration for extending and improving postvention in new and existing areas. This unique handbook is thus essential reading for anyone involved in suicide prevention or postvention research and practice.