OBJECTIVES: The Study TO Prevent Non-Insulin Dependent Diabetes Mellitus (STOP-NIDDM) demonstrated that acarbose therapy over 3.3 years delays or prevents the development of type-2 diabetes as well as cardiovascular (CV) events in persons with impaired glucose tolerance (IGT). The aim of the present study is to assess the economic advantage of these clinical results for the German health care system. METHODS: A cost-effectiveness-analysis from the German payers’ perspective was conducted. The outcome measure was cost per transition to diabetes prevented or delayed. Analyses were conducted for the total population (acarbose (N=682) and placebo (N=686)), and subgroups at high risk for CV disease, for diabetes or both. Cost data were derived from published sources and adjusted to the year 2003 where necessary. Future costs and benefits were discounted by 5%. Acarbose therapy costs and the costs for managing new cases of diabetes, hypertension, and CV events were taken into account. Only direct healthcare costs were considered. RESULTS: For the total population treated over 3.3 years, the incremental cost-effectiveness-ratio was 772€ per case of diabetes delayed. For all high-risk subgroups, acarbose treatment was the dominant treatment option: Cost savings per person in the subgroup at high risk for CV disease, for diabetes and for both CV disease and diabetes were 674€, 42€ and 408€ respectively over 3.3 years. These results were robust to sensitivity analyses. CONCLUSIONS: In addition to the significant clinical benefits observed with acarbose therapy in pre-diabetic persons, the cost-effectiveness model estimated potential cost savings to the German healthcare system in those persons at high risk for diabetes and / or cardiovascular events. For the total population, acarbose therapy was estimated to be cost-effective.