Objective: To assess suitability and comparability of the most common methods of treatment adherence and persistence assessment, as applied to the same pharmacy dataset.
Methods: Data on drugs prescribed for cardiovascular primary prevention to participants in the Aragon Workers' Health Study (AWHS) were collected from a regional electronic drug prescription database. Several different approaches were used to measure treatment adherence (with the medication possession ratio [MPR]) and proportion of days covered [PDC]) and persistence in new users by therapeutic subgroup. Defined daily dose (DDD) was used as a proxy of the number of days' supply, or substituted with surrogate daily dose values.
Results: Higher mean adherence values and proportions of adherent patients were obtained using MPR versus PDC, with additional differences depending on the approach used. The proportion of adherent patients was lowest for oral antidiabetics (14.4%-30.6%) and highest for antihypertensives (70.2%-82.1%). The use of surrogate daily dose values increased adherence for antidiabetics and statins and decreased adherence for antihypertensives. After a 1 year follow-up, treatment persistence was observed for 21.1%, 58.7%, and 29.5% of antidiabetic, antihypertensive and statin users, respectively.
Conclusions: Our findings indicate that use of multiple measures of treatment adherence and persistence provides a more complete overview of medication use patterns, and certain limitations associated with DDD for some drug groups can be overcome with replacement by surrogate doses. The PDC indicator seems to provide a more accurate reflection of patient behavior and treatment continuity than the MPR. Any comparison of adherence/persistence should always consider the method used, variables analyzed, and corresponding data collection process.
Keywords: Medication adherence; administrative data; antidiabetics; antihypertensives; measurement; persistence; pharmacoepidemiology; statins.