This was a retrospective cohort study that evaluated the differences in glycated haemoglobin (HbA_1c) and body mass index (BMI) in veterans with type 2 diabetes mellitus (T2DM), prescribed exenatide twice daily (BID) versus long-acting insulin analog (LAIA) two years after initiation in the United States (US) veteran population.

Patients were included if they were≥18 years old with T2DM, and initiated exenatide BID or LAIA at the Veterans Health Administration between January 1, 2006 and December 31, 2010. Multivariate models were used to evaluate the changes in HbA_1c and BMI between groups, controlling for potential confounders. Logistic regression was used to evaluate the odds of achieving≥0.5% HbA_1c reduction based on baseline HbA_1c stratifications: low,<7%; moderate, 7% to<9%; and high,≥9%.

A total of 446 exenatide BID and 51,531 LAIA patients met inclusion/exclusion criteria. On average, exenatide BID patients were significantly older (64 versus 60 years) with a higher BMI (37.8 versus 32.9kg/m²). Baseline HbA_1c was 8.2% and 8.8% for exenatide BID and LAIA patients, respectively (P<0.001); otherwise, patients were similar for all other characteristics. Exenatide BID treatment was significantly associated with a 0.32% (95%CI: 0.18–0.47%) greater reduction in HbA_1c at two years compared with LAIA. Similar findings were observed for BMI reduction (0.68kg/m²; 95%CI: 0.42–0.95kg/m²). Exenatide BID patients with moderate baseline HbA_1c had significantly higher odds of achieving≥0.5% HbA_1c reduction compared with LAIA patients (OR=1.5; 95%CI: 1.2–2.0).

Veterans treated with exenatide BID had significantly greater reduction in HbA_1c and BMI compared with patients treated with LAIA patients two years after initiation.