OBJECTIVES: To evaluate cost-effectiveness of biphasic insulin aspart 50 (BIAsp 50) versus biphasic human insulin 50(BHI 50) in people with type 2 diabetes mellitus (T2DM) in China. METHODS: The IMS Core Diabetes Model was used to project life expectancy, quality-adjusted life years (QALYs) and total direct medical costs over a 30-year time horizon. Baseline characteristics and treatment effects were based on a published 12-week head-to-head clinical trial for BIAsp 50 versus BHI 50 among people with T2DM in China. Insulin costs were calculated based on average bidding price and dose used in the trial. Diabetes management and complication unit costs were obtained from Chinese published data and adjusted to 2014. Costs and health outcomes were discounted at 3% annually. One-way sensitivity analyses were performed. RESULTS: In the base-case analysis, people with BIAsp 50 have more time alive and free of complications. BIAsp 50 was associated with longer life expectancy (12.246 versus 11.749 years), more QALYs (8.364versus 7.625QALYs), and higher total costs (CNY 196,519 versus 194,202), leading to an incremental cost-effectiveness ratio of CNY 3,140/QALY gained. Sensitivity analyses including discounting rate and treatment effectiveness parameters showed robustness of the results. BIAsp 50 was the dominant treatment choice over shorter time horizons (3.033 versus 2.872 QALYs and CNY 50,798 versus 51,290 for 5 years, 5.268 versus 4.942QALYs and CNY 97,854 versus 98,959 for 10 years, 7.726 versus 7.102 QALYs and CNY 168,453 versus 168,895 for 20-year time horizon). The higher costs over longer time horizon may be due to a longer life expectancy for BIAsp 50 versus BHI 50.    CONCLUSIONS: Treatment with BIAsp 50 was highly cost-effective (under GDP per capita of CNY 46,629) at a 30-year time horizon and dominant at a shorter year time horizon for people with T2DM in China compared to BHI50.