Markers of mercury (Hg) exposure have shown both positive and negative associations with cardiovascular disease (CVD). We assessed the association between serum Hg (S-Hg) and risk of cardiovascular disease in a prospective population-based cohort, with attention to the roles of dental health and fish consumption. Total mortality, as well as morbidity and mortality from acute myocardial infarction (AMI) and stroke, was followed up for 32 years in 1,391 women (initially age 38-60), in relation to S-Hg at baseline, using Cox regression models. Potential confounders (age, socioeconomic status, serum lipids, alcohol consumption, dental health, smoking, hypertension, waist-hip ratio, and diabetes) and other covariates (e.g., fish consumption) were also considered. Hazard ratios (HR) adjusted only for age showed strong inverse associations between baseline S-Hg and total mortality [highest quartile: hazard ratio (HR) 0.76; 95% confidence interval (CI) 0.59-0.97], incident AMI (HR 0.56; CI 0.34-0.93), and fatal AMI (HR 0.31; CI 0.15-0.66). Adjustment for potential confounding factors, especially dental health, had a strong impact on the risk estimates, and after adjustment, only the reduced risk of fatal AMI remained statistically significant. There was a strong inverse association between Hg exposure and CVD. Likely, reasons are confounding with good dental health (also correlated with the number of amalgam fillings in these age groups) and/or fish consumption. The results suggest potential effects of dental health and/or fish consumption on CVD that deserve attention in preventive medicine.