Author ORCID Identifier 0000-0003-3775-3711

Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Educational, School, and Counseling Psychology

First Advisor

Dr. Joseph Hammer

Second Advisor

Dr. Danelle Stevens-Watkins


The Gelberg-Anderson Behavioral Model for Vulnerable Populations was used to identify factors related to buprenorphine use in general and through diversion (versus formal treatment) among rural Appalachian women using substances in the six months prior to their incarceration. This study is a secondary analysis of data from interviews completed with rural Appalachian women (N=400) residing in rural Kentucky jails. Independent variables were analyzed using chi square and independent t-tests for buprenorphine use generally and diverted buprenorphine use. The significant predictors were then evaluated through hierarchical logistic regression to explore which factors account for the most variance in general and diverted buprenorphine use. For buprenorphine use generally, the best fitting model, X2(8) = 57.81, p < .001, was comprised of need variables with prescription opioid use (OR = 1.030), having ever injected prescription pain relievers (OR = 1.965), and endorsing withdrawal symptoms in the year prior to incarceration (OR = 2.883) accounting for the most variance. For diverted buprenorphine use, the model including enabling, predisposing, and need variables was the best fitting model, X2(4) = 16.26, p = .003, with dependence on the GAIN SPS (OR = 12.44) and not having lost custody of a child in the six months prior to incarceration (OR = 0.23) accounting for the most variance. While the results for buprenorphine use generally use is based on need, results for diverted buprenorphine use suggest that enabling factors may play a significant role in a person’s decision-making regarding what avenue buprenorphine is obtained through.

Digital Object Identifier (DOI)