Prevalence and correlates of self-harm in a Southern California driving under the influence treatment setting

Abstract

Abstract Background Suicide and driving under the influence of alcohol and other drugs are US public health epidemics. Research exploring the associations between driving under the influence and suicide presents a research gap. This study describes the prevalence and associations between demographic, alcohol and drug-related characteristics, depression, and self-harm (suicidal thoughts, behavior, and non-suicidal self-injury) among participants in a Southern California DUI Program. Methods Participants receiving driving under the influence treatment services (n = 1310) reported substance use, mental health and demographic data upon program intake between July 2019 and March 2020. A descriptive analysis was used to determine the prevalence of lifetime self-harm thoughts or behaviors. Bivariate analyses of associations between lifetime self-harm thoughts or behaviors and sociodemographic, alcohol, drug, and depression severity measures were explored. Manual backward elimination was used to obtain the final logistic regression model with variables statistically significant at p < 0.05 or had at least one category that was significant at p < 0.05 for non-dichotomous variables. For those reporting self-harm, a chart review extracted clinical themes to identify whether the driving under the influence event was reported as a deliberate suicide attempt. Results Ten percent (n = 135) of participants endorsed a lifetime self-harm history. Younger age, female gender, prior alcohol treatment, depressed mood, and family suicide history were significantly associated with self-harm. One individual reported the driving under the influence event as non-suicidal self-injury), and three reported it as a suicide attempt. Conclusion First- and multiple-time driving under the influence offenders represent a clinical population at increased risk for suicidal thoughts and behavior in need of specialized treatment referrals.