Examining factors associated with nonadherence and identifying providers caring for nonadherent subgroups

Objectives To examine racial/ethnic and regional differences in medication adherence in patients with diabetes taking oral antidiabetic, antihypertensive and cholesterol‐lowering medications and to identify the pharmacies and prescribers who serve these communities. Methods Administrative claims data were analysed for members enrolled in a large health plan in Hawaii (2008–2010) with diabetes mellitus who were taking three types of medications: (1) oral antidiabetic medications; (2) antihypertensive medications; (3) cholesterol‐lowering medications ( n = 5136). The primary outcome was medication adherence based on medication possession ratios. Multivariable logistic regression models were estimated to examine the association between race/ethnicity and region to adherence to each drug class separately, followed by nonadherence to all three. Covariates included age, gender, education level, chronic conditions, copayment level and number of prescribers and pharmacies from which the patients received their medications. Key findings After adjustment for other factors, Filipinos ( OR = 0.58, 95% CI : (0.45, 0.74)), Native Hawaiians ( OR = 0.74, 95% CI : (0.56, 0.98)) and people of other race ( OR = 0.67, 95% CI : (0.55, 0.82)) were significantly less adherent to antidiabetic and antihypertensive medications than Japanese. For cholesterol‐lowering medications, all racial and ethnic groups were significantly less adherent than Japanese, except mixed race. We also found that different racial/ethnic groups tended to use different pharmacies and prescribers, particularly in rural areas. Conclusion Adherence differed by race/ethnicity as well as age and region. Qualitative research involving subgroups (e.g. Filipinos, Native Hawaiians, people under age 50) is needed to identify how to adapt and enhance the effects of interventions shown to be efficacious in prior studies.