OBJECTIVES: To develop an epidemiological and economic model of first-onset cardiovascular disease (CVD, comprising myocardial infarction and ischemic stroke) in Korea that can be applied to cost-effectiveness analyses of interventions. METHODS: KIMCHI is a Markov model with yearly cycles and the health states ‘Alive without CVD’, ‘Alive with CVD’, ‘Dead from CVD’ and ‘Dead from non-CVD causes’. It is populated with 5270 CVD-naïve subjects aged ≥18 years from the 2005 Korea National Health and Nutritional Examination Survey. Annual probabilities of CVD are estimated for each individual using the Asian-specific risk equation by Wu, the covariates for which are: sex, age, total cholesterol (TC), systolic blood pressure (SBP), smoking, diabetes and body mass index (BMI). Age-and-sex-specific annual probabilities of death are based on national health data. To illustrate the function of KIMCHI, follow-up was simulated of Koreans aged ≥55 years until death or age 99 and the cost-effectiveness of atorvastatin for the primary prevention of CVD assessed using decision analysis. The TC-reducing efficacy and cost of atorvastatin were drawn from a meta-analysis and current drug pricing schedules, respectively. CVD costs were provided by the Korean Health Insurance Review and Assessment Services. A 5% annual discount rate was applied. RESULTS: KIMCHI predicted that 30.4% and 18.2% of CVD-naïve Koreans currently aged ≥55 years will develop non-fatal and fatal CVD, respectively, by age 99. Atorvastatin was predicted to reduce these figures to 25.4% and 15.4%, corresponding to numbers needed to treat of 20 and 36 to prevent non-fatal and fatal CVD, respectively. The estimated ICERs were 21.8 million KW/YoLS and 17.4 million KW/QALY saved. CONCLUSIONS: KIMCHI is a contemporary epidemiological and economic model of CVD in Korea that can predict future patterns of disease and be applied to cost-effectiveness analyses of interventions that alter any of TC, SBP, smoking, diabetes and BMI.