Retrospective Observational Assessment of Statin Adherence Among Subjects Patronizing Different Types of Community Pharmacies in Canada

BACKGROUND: Community pharmacies vary widely in terms of ownership structures, location, and dispensing policies. It is unknown if an association exists between the type of community pharmacy and the degree of medication adherence exhibited by patrons-patients. OBJECTIVES: To describe adherence to statin therapy among subjects patronizing different types of community pharmacy categories (department-mass merchandise, chain-franchise, and independent-banner) in Saskatchewan, Canada, between 2000 and 2005. METHODS: Study data were obtained from the Saskatchewan Drug Plan and Extended Benefits database, which is maintained by the government of Saskatchewan, Canada. The study included all subjects who (a) filled a statin prescription within selected community pharmacies between January 1, 2000, and December 31, 2005; (b) had no record of statin prescriptions during the year prior to the first statin prescription, according to the records of the Saskatchewan Drug Plan and Extended Benefits; and (c) demonstrated active utilization in the drug plan database for at least 1 year after the first statin prescription. The proxy criterion for activity was any dispensing record for statin or nonstatin medications at least 1 year following the index claim. Statin adherence level was estimated as tablets per day, defined as the total number of tablets dispensed divided by the total number of days of observation. Each subject's observation period began on the index date and ended on the earlier of (a) 30 days after the last recorded fill for any type of prescription medication (statin or nonstatin), or (b) December 31, 2005. The primary end point was the proportion of subjects within each pharmacy category who maintained an adherence level of 80% or greater during their individual observation period. Additional adherence calculations were performed for each of 3 time periods, beginning on the index date and ending on days 365, 729, and 1,094 (i.e., 1, 2, and 3 years). Patients were included in the analysis for each time period if they met a proxy criterion for availability for observation, defined as the dispensing of any drug at least 1 day after the end date of each period. Pearson chisquare tests were used to assess the significance of differences in baseline characteristics and adherence proportions, comparing pharmacy categories. Logistic regression analysis estimated the odds of an adherence level of at least 80% during the individual observation period, adjusting for pharmacy category, sex, age 65 years or older, known low-income drug coverage, number of distinct drug classes filled concurrently during the first year of observation, loyalty to index pharmacy, and length of observation. Using similar methods, we also estimated pharmacy loyalty by calculating the proportion of subjects who refilled 75% or more of their statin prescriptions at the pharmacy that dispensed their first statin prescription. RESULTS: From an initial sample of 12,818 subjects who had at least 1 pharmacy claim for a statin in the period from January 1, 2000, through December 31, 2005, 8,699 subjects met the inclusion criteria. Subjects were observed for a mean (SD, range) of 3.7 (1.7, 1.0-7.0) years after the index statin prescription. During the first year following the index claim, statin adherence rates were at least 80% for 1,799 of 3,761 (47.8%) patrons of department-mass merchandise, 1,778 of 3,235 (55.0%) patrons of chain-franchise, and 921 of 1,703 (54.1%) patrons of independent-banner stores (P less than 0.001). Measured from the index date through day 1,094, 869 of 2,292 (37.9%), 874 of 1,887 (46.3%), and 457 of 975 (46.9%) subjects in the department-mass merchandise, chain-franchise, and independent banner categories, respectively, had a statin adherence level of at least 80% (P less than 0.001). In logistic regression analysis, pharmacy category type was significantly associated with statin adherence.