Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Social Work


Social Work

First Advisor

Dr. Melanie D. Otis


Stress is associated with poor mental health, specifically anxiety and depression, and stress and mental health problems are predictors of substance dependence and relapse. Social characteristics, such as racial/ethnic minority status, female gender, and low socioeconomic status, are often associated with increased psychological distress and substance use disorders. Pearlin’s social stress theory postulates that this association is due to increased exposure to stress and subsequent experiences of distress related to social disadvantage and decreased access to resources for coping with stress. This project uses a social stress theoretical perspective to examine predictors of substance use after treatment entry and follow-up addiction severity in a large sample of Kentucky substance abuse treatment participants (N = 1123). A conceptual model is tested to determine if social characteristics along with psychological distress, perceived stress, and economic hardship are predictors of substance use and follow-up addiction severity. In addition, since recovery support, efficacy, and self-control have been previously identified as mediators in the stress and relapse processes these factors were included as mediators in the model tested. The conceptual model was tested with three outcome variables, substance use between baseline and 12-month follow-up, follow-up alcohol addiction severity, and follow-up drug addiction severity. Bivariate and multivariate analyses, including logistic regression and ordinary least squares regression, were used to test conceptual models with the full sample and also with a subsample with baseline substance use indicative of potential substance dependence. Findings indicated that significance of predictors varied depending on outcome variable, although recovery support, efficacy, and self-control were significant predictors of all three dependent variables. Findings for each outcome variable are discussed, as well as limitations of the present study, implications for social work practice, and implications for future research.