Author ORCID Identifier

Year of Publication


Degree Name

Master of Science (MS)

Document Type

Master's Thesis




Pharmaceutical Sciences

First Advisor

Dr. Linda P. Dwoskin

Second Advisor

Dr. Jeffery C. Talbert


Healthcare big data are a growing source of real-world data with which to identify and validate medications with repurposing potential. Previously, we developed a claims-based workflow to evaluate medications with potential to treat stimulant use disorders. In order to test the workflow, the framework was applied in the context of opioid use disorders (OUDs), for which there are medications with known efficacy. Using the Truven Marketscan Commercial Claims Database, a nested case-control analysis was conducted to determine the association between OUD medications (buprenorphine, naltrexone) and remission. Cases were defined as enrollees with a remission diagnosis and matched (1:4) to controls (individuals without remission) using incidence density sampling, with age group, sex, region, and index year as additional matching variables. After adjusting for behavioral health visits, polysubstance use disorders, and psychiatric disorders using conditional logistic regression, the odds of OUD medication exposure were 3.8 (99% confidence interval: 3.0 – 4.9) times higher in cases than controls. Evaluation of angiotensin converting enzyme inhibitors (e.g. lisinopril) as a negative control revealed no significant association between the medication and remission. This work demonstrates the feasibility of using administrative health claims data to evaluate the effectiveness of medications to treat substance use disorders.

Digital Object Identifier (DOI)

Funding Information

This research was supported by grants from the National Institutes of Health NIDA T32 DA016176, NIDA F32 DA045483 and grants from the National Center for Advancing Translational Sciences CTSA UL1TR001998, CTSA UL1TR000117.

Available for download on Thursday, May 13, 2021