Reduced alcohol use in patients prescribed pioglitazone

Background and Objectives Alcohol use disorder (AUD) is common and causes significant morbidity and mortality. Currently approved medications are moderately effective. Novel medications are needed to address AUD. Preliminary data suggests pioglitazone may reduce alcohol use. Methods Veterans seen at the Minneapolis VA Health Care System, who were prescribed pioglitazone for diabetes between October 1, 2015 and September 30, 2016, were identified using a national VA database (N = 49). Further chart review was performed to identify all Alcohol Use Disorder Identification Test-Consumption (AUDIT-C) scores prior to starting pioglitazone. Hierarchical Linear models were used to compare all AUDIT-C scores on and off pioglitazone and compare the change in AUDIT-C scores over time before and during pioglitazone was prescribed. AUDIT-C scores were nested within subject with fixed effects for pioglitazone and random intercept and slope for time. Results Forty-nine patients were prescribed pioglitazone and had AUDIT-C scores of 3 or more. The estimated mean AUDIT-C score prior to receiving pioglitazone was 3.98 (95% confidence interval [CI]: 3.51-4.44) and this was reduced to 2.89 (95% CI: 2.46-3.32), reflecting a significant change F(1, 323) = 43.3, p < .001 in the score. The primary reduction occurred within the first year of the pioglitazone prescription. This effect remained significant after controlling for age. Conclusion and Scientific Significance This is the first study of pioglitazone used in a clinical sample focused on alcohol use outcome. The data show that pioglitazone may reduce alcohol use in patients with heavy drinking. Clinical trials of pioglitazone are warranted in patients with AUD.