AIM: This observational study evaluated the clinical effectiveness of exenatide BID (exenatide) vs. insulin glargine (glargine) in patients with type 2 diabetes mellitus in ambulatory clinical practice. METHODS: Retrospective analyses were conducted using an electronic medical record (EMR) database among adult patients with type 2 diabetes mellitus initiating exenatide or glargine between 1 November 2006 and 30 April 2009. The cohorts were propensity-score matched to control baseline demographics, clinical measures, health status and medication use. The changes from baseline to a 12-month follow-up period for A1C (primary outcome), weight, body mass index (BMI), blood pressure and lipid levels were compared between the matched cohorts using paired tests. RESULTS: Propensity-score matching between the exenatide (n = 4494) and glargine (n = 5424) cohorts led to 2683 matched pairs with comparable characteristics, including age, gender and baseline clinical values. The exenatide cohort achieved a greater mean reduction in A1C (-0.6% vs. -0.4%, p < 0.01), weight (-2.6 kg vs. -0.2 kg, p < 0.01), BMI (-0.8 kg/m(2) vs. -0.04 kg/m(2) , p < 0.01) and systolic blood pressure (SBP) (-1.8 mmHg vs. -0.1 mmHg, p < 0.01) in the follow-up period. The changes in diastolic blood pressure and lipid levels were not significantly different between cohorts. CONCLUSIONS: Compared to glargine, exenatide-treated patients experienced significant reductions in A1C, weight, BMI and SBP. Acknowledging the limitations of observational research, exenatide showed greater clinical effectiveness than glargine from a large EMR database in the ambulatory care setting.