OBJECTIVES:  Medication non-adherence leads to increased morbidity, mortality and treatment failures. Current literature lacks evidence about the effect of access to pharmacy services on medication adherence. Pharmacy desert (areas with low access to pharmacy services) may affect patient’s medication adherence behavior. This research aims to compare rates of medication adherence among elderly people living in pharmacy desert and non-desert areas in Pennsylvania for oral medications for treatment of type 2 diabetes mellitus. METHODS:  We conducted a retrospective observational cohort study. The Pharmaceutical Assistance Contract for the Elderly (PACE) program claims dataset along with prescription drug event (PDE) data from Medicare Part D claims of the period January to June 2013 (N=15,886) was used to conduct this study. Medication adherence was measured as the Proportion of Days Covered (PDC) for subjects living in pharmacy desert and non-desert areas of Pennsylvania. PDC is defined as total day’s supply over total number of days evaluated multiplied by 100. Patients with PDC ≥80% were dichotomized as adherent. Sequential multiple regression analysis was carried out to examine the factors responsible for difference in medication adherence rates. RESULTS:  Chester, Montgomery, Bucks, Delaware, Philadelphia, Fayette and Cumberland counties were considered as pharmacy non-desert areas (n=2015, Female=1376). Forest, Mifflin and Sullivan counties were considered as pharmacy desert regions (n=116, Female=87). Pharmacy deserts found to have low access to pharmacy services. The mean of PDC for patients living in pharmacy desert and non-desert regions found to be 89% and 91% respectively. After adjusting for selected covariates, the income level, age, sex and marital status, study found no significant difference in medication adherence rates among elderly patients living in pharmacy desert and non-desert areas. CONCLUSIONS:  Medication Adherence rate among elderly with diabetes not found to vary with the place of residence when pharmacy deserts and non-deserts across ten counties of Pennsylvania were compared.