The strongest correlations between coffee consumption and serum cholesterol levels have been found in countries where people drink coffee brewed by mixing coffee grounds directly in boiling water (boiled coffee). In the present study of a population-based sample of 1625 middle-aged subjects (the Northern Sweden MONICA Study), approximately 50% of the participants were drinking boiled coffee, and 50% were drinking filtered coffee. Consumers of boiled coffee had significantly higher serum cholesterol levels than consumers of filtered coffee. Subjects who drank boiled coffee reported a higher intake of fat. A linear multiple regression analysis with serum cholesterol as the dependent variable confirmed that boiled coffee was an important independent determinant of cholesterol levels. We conclude that subjects who drink boiled coffee have higher serum cholesterol levels than those who drink filtered coffee, and that the most likely explanation for this finding lies in the type of brewing method.