|Abstract||Background: Studies conducted in Qatar have reported a high prevalence of diabetic nephropathy, retinopathy, and neuropathy. These complications are often associated with poor medication adherence and uncontrolled diabetes. The objectives of this study were to determine the rate of medication adherence among patients with uncontrolled diabetes, and to compare the characteristics and identified barriers between patients with good and those with poor medication adherence. Method: A cross-sectional quantitative study was conducted among patients living in Qatar with uncontrolled diabetes attending primary healthcare clinics from October 2016 to January 2017. An interviewer-administered questionnaire comprising three sections was utilized in the study: patients- characteristics, Adherence to Refill and Medications Scale in Diabetes (ARMS-D), and barriers to medication adherence. ARMS-D is a validated tool that is used to identify adherence level among patients with diabetes. Descriptive and inferential statistics including regression analysis were used for data analysis. Results: A total of 260 patients were included in the analysis. Overall, 73% (n = 191) were nonadherent to their diabetes medications (ARMS-D score above 11). Nonadherent patients reported the majority of the pre-determined barriers to medication adherence with forgetfulness being the most commonly reported barrier. Multivariate linear regression analysis found age, ethnicity, education level, income level and HbA1c to be independent predictors of adherence. Conclusion: The findings of this study reaffirm the notion that non-adherence to medications among patients with uncontrolled diabetes within primary care setting is higher than the general diabetes population. This high prevalence is concerning and necessitates urgent interventions. Nonetheless, an in-depth understanding of barriers to medication adherence often requires qualitative research approach as these barriers are very complex and multifactorial in nature.