Substance use and criminogenic thinking: Longitudinal latent class analysis of veterans with criminal histories

Abstract

Introduction

The objective of this study was to inform clinical practice by identifying distinct subgroups of US veterans with criminal histories in residential mental health treatment. The study characterized veteran patients on their alcohol and drug use and criminogenic thinking. We also examined predictors and outcomes of subgroup membership.

Methods

Participants were 341 veterans with a criminal history in residential mental health care. A parallel latent growth trajectory model characterized participants' alcohol and drug use and criminogenic thinking at treatment entry and at 6- and 12-month follow-ups.

Results

The study identified four distinct classes: 53 % Normative Improvement, 27 % High Criminogenic Thinking, 11 % High Recurrence (of substance use), and 9 % High Drug Use. Compared to the Normative Improvement class, prior to treatment entry, patients in the High Recurrence class were less likely to be on parole or probation, and patients in the High Criminogenic Thinking class were more likely to be chronically homeless. Compared to the Normative Improvement class, at follow-ups, patients in the High Drug Use and High Criminogenic Thinking classes were more likely to recidivate, and patients in the High Drug Use class were more likely to report unstable housing. Depression scores were higher (nearly double) in the High Drug Use, High Recurrence, and High Criminogenic Thinking classes at follow-ups compared to the Normative Improvement class.

Conclusions

That the Normative Improvement class entered mental health residential treatment with relatively low alcohol and drug use and criminogenic thinking, and sustained these low levels, suggests that treatment does not need to be broadened or intensified to improve these domains for these patients with criminal histories. In contrast, findings for the High Drug Use, High Recurrence, and High Criminogenic Thinking classes, which composed 47 % of the sample, suggest that more integrated and sustained treatment may be needed to reduce recidivism, depression, and homelessness among these patients.