OBJECTIVES:  This post hoc analysis analyzed the number of patients needed to be treated (NNT) with the GLP-1 receptor agonist exenatide once weekly (ExQW) vs titrated insulin glargine (IG) over 26 weeks to allow one additional patient to achieve single or combined recommended treatment goals. METHODS:  Data from the DURATION-3 trial was analyzed retrospectively. Treatment targets included: 1) glycaemia (HbA1c ≤6.5% or fasting plasma glucose (FPG) <7 mmol/L); 2) systolic blood pressure (SBP <130 mmHg); 3) low-density lipoprotein cholesterol (LDL <2.59 mmol/L); and 4) weight loss or maintenance. Hypoglycemic events were also assessed. NNT was calculated for the entire intent-to-treat (ITT) population (ExQW n=233, IG n=223) and for subpopulations of patients on different background therapies (metformin ± sulfonylurea). NNT was calculated using 1/Absolute Risk Reduction (percent of patients reaching goal in the ExQW treatment arm ‑ percent of patients reaching goal in the IG treatment arm). RESULTS:  Baseline mean characteristics were similar for both treatment groups: 45-48% women, age 58 years, HbA1c 8.3%, and body mass index 32 kg/m2. Regardless of background therapy, 8 patients would need to be treated with ExQW in place of IG for 26 weeks to allow one additional patient to attain the HbA1C goal. Five patients (ITT) would need to be treated with ExQW vs IG to allow one additional patient to attain the HbA1c goal with weight control and the absence of hypoglycaemia. Furthermore, 14 patients (ITT) would need to be treated with ExQW versus IG to allow one additional patient to reach combined HbA1c, SBP, and LDL goals. Only the FPG goal favored insulin use with an NNT of -8 (ITT). Minor differences in the NNT values were observed between subpopulations for most goals. CONCLUSIONS:  Both ExQW and IG treated patients achieved therapeutic target goals. In this post hoc analysis the NNT results favored ExQW for most goals.