OBJECTIVE: To help health care decision-makers assess the cost-effectiveness of HDL-c raising therapy, a model was developed to project lifetime clinical and economic benefits of adding an HDL-c raising drug in statin-treated patients with persistently low HDL-c. METHODS: We developed a simulation model made of two analytic decision sub-models. The first sub-model (Monte-Carlo) created a cohort of patients based on lipid levels after drug treatment. The second sub-model (Markov) estimated long-term clinical and cost outcomes (Framingham risk equations). Validation analyses were performed. RESULTS: Internal validation: in patients with no history of CVD, the model predicted a 2 years risk of CHD events of 1.12% and 2.02% risk for males and females respectively versus 1% and 2% in the Framingham study. In patients with a history of CVD, the model predicted 2-year risk of CHD events of 9.15% and 3.20% for males and females respectively versus 10 and 3.5% respectively in the published values. External validation: 1) in patients with low HDL-c, the model predicted a 2.32% risk for CHD events associated with each 1mg/dL increment in HDL-c versus 2-3% in a published meta-analysis 2) by recreating the 4S cohort characteristics and simvastatin treatment effects, the model predicted cumulative incidences of 14.2% for non-fatal and 5.5% for fatal CHD events versus 15.9% and 5.0% in the 4S. CONCLUSIONS: Based on our validations, this computer simulation model appears to be a valuable and accurate tool to help evaluate HDL-c raising therapy for patients with persistently low HDL-c despite statin treatment.