Optimal PAI-A Cutoffs to Determine Risk for Non-Suicidal Self-Injury (NSSI) and Suicide-Related Behavior (SRB) among At-Risk Adolescents

Abstract

Objective This study examines psychopathology and personality correlates of non-suicidal self-injury (NSSI) and suicide-related behavior (SRB) in an understudied sample of adolescents who have exhibited behaviors (e.g., delinquent acts, premature high school termination) that place them at-risk for poor psychosocial outcomes. Method Participants included a predominantly White male sample of 182 adolescents (Mage = 16.82 years). In addition to information about NSSI and SRB histories, participants self-reported various facets of personality and psychopathology on the Personality Assessment Inventory-Adolescent (PAI-A). Results Logistic regression analyses indicated that the Suicidal Ideation (SUI) scale on the PAI-A was the strongest predictor of both NSSI and SRB history, as it outperformed other relevant PAI-A scales and the Suicide Potential Index (SPI), an aggregate scale that was designed to assess for suicide risk using the PAI for adults. Receiver operating characteristic (ROC) curve analyses were also conducted to determine optimal cutoff scores for significant PAI-A predictors. Conclusions Findings from the current study can be used to identify NSSI and SRB risk and target these life-threatening behaviors when working with at-risk adolescents. Highlights PAI-A SUI outperformed other PAI-A variables in predicting NSSI and SRB risk. PAI SPI did not perform as well in adolescents compared to adult samples. Cutoff scores in the current sample were well below those in the PAI-A manual.