OBJECTIVES: Exenatide and liraglutide are the two therapeutic options in the GLP-1 anti-diabetic medication class, to improve glycemic control in adults with type 2 diabetes (T2D). This study evaluated patient and prescriber characteristics, treatment patterns, average daily dose (ADD), and glycemic control of patients initiating GLP-1 medications in Germany. METHODS: The LifeLink™ EMR-EU database contains records for over 15 million German patients and 3,000 physicians. The cohort included patients who initiated exenatide or liraglutide during the index period (01/01/2009 - 04/30/2010). Patients also had ≥180 days history, pre-index; 90-540 days follow-up, post-index; and a T2D diagnosis (ICD-10 E10-E14), pre-index. Univariate tests were conducted at α=.05. RESULTS: The cohort included 692 patients (exenatide 292, liraglutide 400): mean (SD) age 59 (±10) years, 59% male.  Diabetologists prescribed liraglutide more frequently than exenatide (65% vs. 35%) compared to non-diabetologists (51% vs. 49 %). Choice of GLP-1 was not associated with patient age, sex, number of pre-index antidiabetic medications (1.9±0.9), pre-index HbA1c (8.2±1.5%), or Charlson Comorbidity Index (0.45±0.78, all p>.05). Mean (SD) ADD was 16.7 mcg (±9.22; label range 10-20 mcg) for exenatide patients and 1.43 mg (±0.69, label range 0.6-1.8 mg) for liraglutide patients. Among patients with post-index HbA1c tests, mean values did not differ at the first (7.9), second (7.8), or third (7.8, all p>.05) tests. Exenatide patients were more likely than liraglutide patients to continue pre-index anti-diabetic medications (67.1% vs. 60.3%, p=.027) or to start concomitant anti-diabetic medications at index (32.2% vs. 25.0%, p=.013); however, exenatide patients were less likely to augment treatment post-index (15.8% vs. 22.5%, p=.027). Post-index, 9.3% exenatide and 10% liraglutide patients discontinued GLP-1 therapy (p>.05). CONCLUSIONS: Results suggest that some differences exist between German patients initiating exenatide or liraglutide, with respect to prescribing physician specialty, pre-and post-index treatment patterns, and ADD.  Both GLP-1s show comparable post-index HbA1c.