Objective: To examine the association between socioeconomic status, as indicated by neighborhood median income levels, and physician drug selection between older, less expensive generic drugs and newer, more expensive brand-name drugs for elderly patients initiating drug therapy in a universal healthcare system.
Methods: We conducted a population-based, retrospective, cross-sectional study. Using healthcare administrative databases, we assessed the medication profiles of 128 314 patients from more than 1.4 million residents of Ontario > or =65 years old initiating antipsychotic, hydroxymethylglutaryl-coenzyme A reductase inhibitor (statin), or ocular beta-blocker drug therapy from January 1, 1998, through December 31, 1999. We examined the selection of older generic drugs relative to newer brand-name agents for patients in each of 5 income quintiles.
Results: Overall, brand-name drug prescribing modestly increased with increasing income quintile after adjusting for patient age and gender (61.2% in the lowest income quintile vs. 64.1% in the highest income quintile; p value for trend < 0.001). Significant risk ratios comparing the highest with the lowest income-quintile patients were observed for selection of newer, brand-name antipsychotics (RR 1.14; 95% CI 1.06 to 1.23), older generic statins (RR 0.86; 95% CI 0.77 to 0.95), and newer, brand-name ocular beta-blockers (RR 1.13; 95% CI 1.02 to 1.25).
Conclusions: This study suggests that income-related differences in treatment selection by physicians may exist. The reasons for these differences and subsequent impact on health outcomes warrant further investigation.