OBJECTIVE: To determine how well the current Pharmaceutical Benefits Scheme (PBS) eligibility criteria for subsidy of lipid-lowering drugs compare with current national guidelines for determining the population at high risk of developing cardiovascular disease (CVD). DESIGN AND PARTICIPANTS: Analyses of the population-based, cross-sectional Australian Diabetes, Obesity and Lifestyle (AusDiab) study, conducted in 1999-2000. The 1991 Framingham risk prediction equation was used to compute 5-year risk of developing first-time CVD in 8286 participants aged 30-74 years with neither CVD nor diabetes. Based on the National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand guidelines, people with either 5-year CVD risk > or = 15% or with 5-year CVD risk of 10%-< 15% and the metabolic syndrome were defined as having estimated high absolute CVD risk. MAIN OUTCOME MEASURES: 5-year CVD risk; estimated population with high CVD risk. RESULTS: Among participants without prevalent CVD or diabetes, 7.9% of men and 1.5% of women had a 5-year CVD risk > or = 15%. Of the estimated residential Australian population in 2000 aged 30-74 years without CVD or diabetes, 717 000 people were considered to be at high absolute CVD risk. Among the high-risk AusDiab participants without CVD or diabetes, only 16.9% of men and 15.4% of women were being treated with lipid-lowering drugs. Of the 9.6% of participants free of CVD and diabetes who were untreated but eligible for subsidy under PBS criteria, only 27.4% had an estimated high absolute CVD risk. CONCLUSION: Strategies for CVD prevention using lipid-lowering medications can be improved by adoption of the absolute-risk approach.