OBJECTIVES: Due to the progressive nature of type 2 diabetes mellitus (T2DM), therapy escalation or intensification is often required to maintain acceptable levels of glycaemic control. The objective of this study was to assess how differential dual therapy failure rates influence the cost effectiveness (CE) results in T2DM. METHODS: This study used the IMS Core Diabetes Model (CDM), a validated and established diabetes model, to evaluate the CE of metformin+ sulphonylurea (M+S) compared to metformin + DPP-4 (M+D).  Efficacy data for dual therapy was sourced from a published systematic review;  HbA1c and BMI change of -0.8% and 0.199kg/m2 (M+D) and -0.79% and 0.707kg/m2 (M+S) respectively were applied.  Rates of severe hypoglycaemia were 0.1612 and 1.538 per 100 patient years and 4.596 and 68.769 per 100 patient years for non-severe events in M+D and M+S respectively. Insulin rescue therapy was initiated at an HbA1c threshold of 7.5%. Base case analysis assumed M+D and M+S had the same HbA1c durability; lifetime CE was assessed assuming improvement in durability favouring M+D applied in 10% increments with costs (US$) and benefits discounted at 3.5%. RESULTS: In the base case analysis, annual HbA1c increase was 0.26% with mean time to therapy escalation of 5 years; and a predicted cost per quality adjusted life year (QALY) of $211,948.  Mean annual increments in HbA1c for M+D of 0.182%, 0.13% and 0.1% were necessary to achieve costs effectiveness at willingness to pay (WTP) thresholds of $100,000, $70,000 and $50,000 respectively. Published HbA1c durability for M+D of 0.052% per year was associated with a cost per QALY of $33,427. CONCLUSIONS: This analysis demonstrates that the annual rate of increase in HbA1c exerts considerable influence over predicted CE and is therefore an important variable to study when assessing the CE of new interventions for the management of T2DM.