OBJECTIVES: The purpose of this study was to measure the relative outcome of treatment following initiation of treatment with insulin glargine versus insulin detemir in General Practice in people with both type 1 (T1DM) and type 2 (T2DM) diabetes mellitus. METHODS: Data were extracted from a proprietary database of primary care records (THIN). Cases were selected if treated exclusively with either glargine or detemir. Glycaemic control (HbA1c) was extracted for up to nine months following basal analog initiation and averaged over three quarterly periods, given the recent launch of detemir. Records of GP contacts for hypoglycaemia were also examined. RESULTS: There were 4844 subjects available for analysis in the glargine group and 528 subjects in the detemir group; baseline cohort characteristics were equivalent. In T1DM the glargine and detemir groups compared respectively: 40.8 years vs. 39.2 years old, 12.5 years vs. 12.2 years diabetes duration, 47.9% female vs. 42.9% female and 26.4 kg/m2 BMI vs. 26.9 kg/m2. In T2DM the glargine and detemir groups respectively: 61.4 years vs. 58.6 years old, 10.3 years vs. 9.0 years diabetes duration, 46.0% female vs. 45.9% female, 30.2 kg/m2 BMI vs. 31.0 kg/m2. The HbA1c deterioration profile was very similar prior to switching to either treatment. Following switching, diabetes control was superior in T1DM when treated with glargine; the mean reduction in HbA1c using glargine compared to baseline was 0.19%, 0.36%, and 0.37% for Q1 to Q3, respectively (p<0.01). A similar pattern was evident in T2DM, where the improvement was 0.52%, 0.82%, and 0.89%, respectively (p<0.001). There were fewer reports of GP-treated hypoglycaemia episodes with glargine vs. detemir (4.80 per 100 patient years [PHPY] vs. 6.40 PHPY; p<0.05). CONCLUSIONS: When used in general practice, glargine resulted in superior glycaemic control and fewer hypoglycaemic episodes than detemir in both Type 1 and Type 2 diabetes.