Event-level relationship between methamphetamine use significantly associated with non-adherence to pharmacologic trial medications in event-level analyses

Background: Methamphetamine use has been previously associated with poor medication adherence, but, to date, there have been no studies that have conducted event-level analyses on correlates of medication adherence in studies of pharmacologic agents for methamphetamine dependence. Methods: We pooled data from two previous, randomized controlled trials (using bupropion and mirtazapine, respectively) for methamphetamine dependence and used a mixed effects logistic model to examine correlates of daily opening of the medication event monitoring system (MEMS) cap as a repeated measure. We explored whether periods of observed methamphetamine use via urine testing were associated with study medication adherence based on MEMS cap openings. Results: We found a significant negative association between methamphetamine-urine positivity and event-level study medication adherence as measured by MEMS cap openings (AOR: 0.69; 95% CI: 0.49-0.98). In addition, age (AOR: 1.07; 95% CI: 1.02-1.11) and depressive symptoms (AOR: 0.78; 95% CI: 0.64-0.90) were significantly associated with adherence. Finally, participants were more likely to open their study medication bottles on days when they presented for in-person urine testing. Conclusions: Our event-level analysis shows that methamphetamine use can be associated with reduced medication adherence as measured by MEMS cap openings in pharmacologic trials, which corroborates prior research. These findings may suggest that medication adherence support in pharmacologic trials among methamphetamine users may be needed to improve study compliance and could be targeted towards periods of time when there are more likely to not open their study medication pill bottles. (C) 2014 Elsevier Ireland Ltd. All rights reserved.