Differential impact of longitudinal medication non-adherence on mortality by race/ethnicity among veterans with diabetes

OBJECTIVE: To examine the differential effect of medication non-adherence over time on all-cause mortality by race/ethnicity. RESEARCH DESIGN AND METHODS: Data on a longitudinal cohort of veterans with type 2 diabetes was examined. The main outcome was time to death. Primary independent variables were race/ethnicity and mean medication possession ratio (MPR) categorized into quintiles over the study period. Cox regression was used to model the association between time to death and MPR quintiles and race/ethnicity, adjusting for relevant covariates. RESULTS: The cohort of 629,563 veterans was followed for 5 years. After adjusting for all covariates, the hazard ratios (HR) for subjects in the lowest versus highest MPR quintile was 12.21 (95 % CI 11.89, 12.55) for non-Hispanic white (NHW), 10.01 (95 % CI 9.18, 10.91) for non-Hispanic black (NHB), 12.65 (95 % CI 11.10, 14.43) for Hispanic and 10.41 (95 % CI 9.06, 11.96) for Other race veterans. Furthermore, type of diabetes therapy (oral versus insulin) maintained a significant relationship with mortality that varied by racial/ethnic group. CONCLUSIONS: This study demonstrates the differential impact of medication non-adherence on mortality by race. It also demonstrates that type of diabetes therapy (insulin with or without oral agents) is associated with mortality and varies by racial/ethnic group.