OBJECTIVES: This study, conducted in Type 2 diabetes mellitus (T2DM), evaluated the cost utility of glargine versus NPH in previously insulin naïve (IN) and non insulin naïve (NIN) patients switching from NPH to insulin glargine in the UK using observational data. The study assessed the combined effect of HbA1c and hypoglycaemia reduction. METHODS: A discrete event life time simulation based on UKPDS 68 was adapted to include the effects of HbA1c and hypoglycaemia reduction using published meta-regression results from 11 randomised clinical trials. Direct costs and health utility (EQ5D) were derived from published sources and the HODaR database respectively; costs and benefits were discounted at 3.5%. This model used the demographic and efficacy profiles of T2DM patients who were IN or NIN who switched from NPH to glargine identified via the THIN database. Analysis was conducted on 1,496 and 174 IN and NIN patients respectively; the primary outcome measure was Hba1c change. As hypoglycaemia was not directly collected from the THIN database, sensitivity analysis was performed taking into account HbA1c benefit only. RESULTS: The mean age and duration of diabetes at switch was 63 years and 7.5 years (IN) and 70 years and 10.2 years (NIN) respectively. After adjustment for baseline profiles IN patients starting glargine showed a significant reduction in HbA1c of 0.21%( p=0.029) 12 months post initiation versus NPH. For NIN patients switching from NPH to glargine the adjusted HbA1c reduction was 0.46% (p=0.0093). The cost per QALY for a simulated cohort of 10,000 patients was £5,806 and £3,415 for IN and NIN patients. In sensitivity analysis considering an HbA1c reduction only the cost per QALY was £18,179 and £7,973 for IN and NIN patients respectively. CONCLUSION: Based on real life observational data, in both IN and NIN patients T2DM patients, glargine is cost-effective compared to NPH.