Measurement and prediction of medication compliance in problem drinkers

A variety of methods have been used to measure medication compliance. Although electronic monitoring has been considered to be the best method, it has some potential for error and its high cost may limit its feasibility. This study examined the concordance of data on medication compliance that was obtained by using an electronic monitoring system (Medication Event Monitoring System, or MEMS), daily diary reports, and tablet counts.

Subjects were 150 problem drinkers participating in a pharmacotherapy trial. Subjects were randomly assigned to receive naltrexone 50 mg (n = 75) or placebo (n = 75) and to receive either daily (n = 72) or targeted (n = 78) medication administration. Comparisons between methods were conducted at the aggregate and individual daily levels, and the association between compliance and three alcohol-related outcomes was evaluated.

Daily diaries showed good day-to-day concordance with MEMS and a higher correlation with MEMS than did tablet counts. However, high-compliance subjects did no better than low-compliance subjects on any of three alcohol-related outcome measures.

Compliance measurement using a daily diary method yielded results that are comparable to those obtained with electronic monitoring, which served as a criterion measure. In contrast, tablet counts provided data that were less concordant with the criterion measure. The unique nature of the study sample (i.e., the majority of subjects were not committed to a goal of abstinence), combined with the targeted approach to treatment, may explain the lack of association between compliance and alcohol-related outcomes. Further research should aim to ascertain the factors affecting the relationship between compliance and outcomes.