Further evidence that suicide risk is categorical: A taxometric analysis of data from an inpatient sample

Abstract

Recent research by Witte and colleagues (2017) revealed a taxonic structure for classifying suicide risk using a sample of predominantly military outpatients. The authors sought to replicate the Witte et al. (2017) findings using data from a sample of 2,385 psychiatric inpatients who completed measures of suicidal ideation and behavior upon admission to the hospital. The comparison curve fit index values for means above minus below a cut (.80), maximum eigenvalue (.71), and latent mode (.52) showed a similar taxonic structure (i.e., dichotomous rather than continuous). Consistent with Witte et al. (2017), differences between the taxon and complement groups were larger for variables conceptually directly related to suicide risk than to broader constructs such as hopelessness or depression. Support for this categorical distinction among a sample of long-term psychiatric inpatients, who are uniformly high in symptom severity, emphasizes the need for additional research on this high-risk group and development of further assessment methods.