Psychopathy and Suicide: A Multisample Investigation Kevin S. Douglas Simon Fraser University Sylvia Herbozo and Norman G. Poythress University of South Florida Henrik Belfrage Mid-Sweden University and Sundsvall Forensic Psychiatric Hospital John F. Edens Southern Methodist University Evidence suggests that behavioral aspects of psychopathy are associated with suicidal behavior, whereas the affective and interpersonal aspects are not. The authors tested the robustness of this bifurcated association across 1,711 persons and 12 samples of adult and juvenile criminal offenders, forensic psychiatric patients, and civil psychiatric patients. The authors observed a small but significant partial correlation (.13) between the behavioral/impulsive lifestyle features of psychopathy and suicidality, but no effect for affective/interpersonal features. Several method and sample features (mental dis- order; psychopathy and suicidality measurement format) significantly strengthened or weakened this association. The authors conclude that it is not possible to speak of “the” association between psychopathy and suicide, but that this relationship appears to be partially dependent on methodological (i.e., self-report vs. clinician-administered psy- chopathy measures) and sample composition (i.e., age; mental illness) factors. Recom- mendations for practice are provided, including that clinicians should not consider psychopathy a buffer against suicidal behavior. Keywords: suicide, psychopathy, antisocial personality disorder, criminal offenders, forensic patients Risk for suicide is a salient concern in public sector penal and mental health institutions. Re- search indicates that suicide attempts and comple- tions are serious problems in adult and juvenile offender (Alessi, McManus, Brickman, & Grap- entine, 1984; Jenkins et al., 2005; Kempton & Forehand, 1992) and psychiatric samples (Baxter & Appleby, 1999; Mortensen, Agerbo, Erikson, Qin, & Westergaard-Nielsen, 2000) and that sui- cide is one of the leading causes of death in such institutions (Burtch & Ericson, 1979; Hayes, 1995). For this reason, psychologists and other mental health professionals who work in such settings must be knowledgeable about suicide risk factors, so as to be able to assess a given individ- ual’s risk for suicidal behavior and to intervene when necessary. A number of suicide risk factors exist in the research corpus, with a particular em- phasis on markers of depression. Overwhelming evidence links depressive symptomatology (e.g., hopelessness, guilt, anhedonia, insomnia, dimin- ished concentration) and suicide, leading research- ers to describe severe depression as a risk factor for suicidal behavior (Fawcett et al., 1990; Maris, Berman, & Silverman, 2000; Tanney, 1992). A wealth of research also identifies substance use problems as a strong risk factor for suicide (Maris et al., 2000). Fewer studies have examined suicide in relation to externalizing behaviors, which are dispropor- Kevin S. Douglas, Department of Psychology, Simon Fraser University, Canada; Sylvia Herbozo, Department of Psychology, University of South Florida; Norman G. Poy- thress, Department of Mental Health Law & Policy, Uni- versity of South Florida; Henrik Belfrage, Mid-Sweden University and Sundsvall Forensic Psychiatric Hospital; John F. Edens, Department of Psychology, Southern Meth- odist University. Kevin S. Douglas is also affiliated with Mid-Sweden University as a Guest Professor of Applied Criminology. Correspondence concerning this article should be ad- dressed to Kevin S. Douglas, Department of Psychology, Simon Fraser University, 8888 University Drive, Burnaby, BC, Canada, V5A 1S6. E-mail: douglask@sfu.ca Psychological Services Copyright 2006 by the American Psychological Association 2006, Vol. 3, No. 2, 97–116 1541-1559/06/$12.00 DOI: 10.1037/1541-1559.3.2.97 97 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.