Appraisal of Suicidal Risk Among Adolescents and Young Adults Through the Rorschach Test

Abstract

The aims of this study are to investigate suicidal behaviors among adolescents and young adults and to test an index composed using Rorschach test responses related to an increased risk of suicide. Using a cross-sectional design, 4 groups were studied (according to criteria of the Columbia Classification Algorithm of Suicide Assessment [Posner, Oquendo, Gould, Stanley, & Davies, 2007]): A group with suicidal ideation (n = 30), a group with parasuicidal behavior (n = 30), a group with near-lethal suicide attempts (n = 26), and a control group (n = 30). Responses to the Rorschach test yielded 6 potential indicators of suicidal behavior (scored according to Exner's Comprehensive System and the Suicidal Index for Adolescents; Silberg & Armstrong, 1992). Rorschach scores including at least 4 of these 6 indicators selected 69% of the people who had committed serious suicide attempts. The Rorschach Suicidal Index reached an acceptable reliability and was related to other criteria of suicide risk, such as the Beck Depression Inventory (BDI; Beck, Ward, Mendelsohn, Mock, & Erbaugh, 1961) and Linehan Reasons for Living Inventory (RFL–I; Linehan, Goodstein, Nielsen, & Chiles, 1983). Moreover, the Rorschach Suicidal Index showed incremental validity over the BDI and the RFL–I to predict suicidal behavior. A path analysis additionally showed that low social support was an important mediator between the Rorschach Suicidal Index and the number of suicide attempts committed by participants.

Notes

From our initial sample of 125 participants, 9 of the inpatients hospitalized after a suicide attempt were not included in the study for two reasons. First, they showed self-injurious behavior with unknown or no suicidal intent (i.e., an accidental injury with psychiatric symptoms only). The self-injurious behaviors of those persons according to C–CASA criteria belonged to the categories of indeterminate or potentially suicidal events and nonsuicidal events (Posner et al., 2007). Second, these young inpatients had other prominent Axis I disorders and Axis II disorders; specifically, major depression (3), anxiety disorder (2), drug abuse (2), and personality disorder (2). It was rare for other psychiatric disorders to be specified on this toxicology ward, which suggested these other symptoms were more prominent than the self-injurious behaviors.

^a n = 30.

^b n = 26.

*p < .05. **p < .01. ***p < .001.

*p < .01. **p < .001.

*p = .012.