Regional variation in the use of medications by older Canadians--a persistent and incompletely understood phenomena.

BACKGROUND: We have previously reported on regional variability in medication consumption by older Canadians. In this study, we used longitudinal data to determine whether regional differences in commonly consumed medications persisted and to explore potential explanatory factors for observed differences.
METHODS: We utilized data from the second phase of the Canadian Study of Health and Aging to assess the number, types, and variability of medications used between regions. Linear and logistic regressions (LRs) were used to predict the number of medications and the use of specific agents where significant regional variability was found to exist.
RESULTS: There were significant regional differences in the number of medications consumed and in the prevalence of use of acetaminophen (p < 0.002), benzodiazepines (p < 0.020), nitrates (p = 0.040), and complementary and alternative medicines (CAMs; p < 0.020). The proportion of subjects using acetaminophen was highest in British Columbia (44.6%) and lowest in Quebec (27.3%). Benzodiazepine and nitrate consumption was highest in Quebec (35.9 and 19%, respectively) and lowest in the Praires (18.2%) and Atlantic Canada (6.6%). CAM use was highest in British Columbia (47.1%) and lowest in the Atlantic region (26.8%). Similar inter-regional differences had been found 5 years previously. There were no significant regional differences in the prevalence of hypertension, myocardial infarction, diabetes, arthritis/rheumatism, or depression. Region remained a significant explanatory variable for the number of medications and nitrate, benzodiazepine, and CAM use in our multivariate models.
CONCLUSIONS: Regional differences in medication use persisted over the course of this longitudinal study. Much of the variability remains unexplained. The reasons for regional differences in consumption of drugs and their clinical significance should be addressed.