Clarifying Psychological Risk Factors for Self-Injury and Suicidal Behaviors: Clinical Applications of Behavioral Measures

Self-injurious thoughts and behaviors are life-threatening, prevalent, and challenging clinical outcomes to predict. This dissertation explores the use of behavioral measures to improve prediction of nonsuicidal self-injury (NSSI) and suicidal ideation. This builds on the growing body of literature supporting the clinical application of behavioral measures such as the Implicit Association Test (IAT) and emotional Stroop Task with self-injurious and suicidal individuals. I tackle three questions to inform continued research and application. First, can the Self-Injury IAT be used in acute care settings to predict NSSI? Study 1 shows that the Self-Injury IAT can be used for short-term prediction: it predicted NSSI occurring during hospital stays above and beyond other risk factors, but did not predict NSSI occurring after hospital discharge. Surprisingly, patients’ explicit self-report was a more robust predictor of NSSI than the Self-Injury IAT. Admission-to-discharge change in patients' explicit self-report, but not IAT performance, predicted whether they engaged in NSSI after hospital discharge. Second, how does transient mood affect the predictive validity of the Suicide IAT and Stroop task? In Study 2, suicide ideators demonstrated significantly stronger implicit identification with death after (vs. before) the mood induction, as indicated by post-induction IAT performance. Controlling for history of suicidal ideation, post-induction performance was most predictive of suicidal ideation when assessed categorically (i.e., identification with Death vs. Life). Suicide Stroop performance remained unrelated to suicidal ideation on its own, but enhanced prediction of suicidal ideation when combined with Suicide IAT performance. All baseline suicide ideators who achieved one particular type of IAT/Stroop scoring profile experienced suicidal thoughts six months later. Third, is it safe to administer behavioral measures related to self-injurious thoughts and behaviors? Study 3 reveals that there is minimal change in self-injurious or suicidal urges from before to after completing Suicide and Self-Injury IATs. This was found across three distinct samples. A small to moderate mood decline was consistently detected, which was isolated to female respondents and one type of IAT that presented NSSI-related images. Female participants’ negative mood after viewing NSSI-related images appeared to be transient in nature--possibly be alleviated by viewing positive images. This collection of studies balances clinical application and psychological science, and presents a number of important considerations for future research and practice.Psycholog