Objective Examine the relation between IQ in early adulthood and later coronary heart disease (CHD) mortality, and assess the extent to which established risk factors measured in middle-age might explain this gradient. Design Cohort study of 4316 male former Vietnam-era US army personnel with IQ scores (mean age 20.4 years), risk factor data (mean age 38.3 years) and 15 years mortality surveillance. Results In age-adjusted analyses, lower IQ scores were associated with an increased rate of CHD mortality (hazard ratio (per SD decrease in IQ); 95% confidence interval: 1.34; 1.00, 1.79). Adjustment for later chronic disease (1.22; 0.91, 1.64), behavioural (1.29; 0.95, 1.74) and physiological risk factors (1.19; 0.88, 1.62) led to some attenuation of this gradient. This attenuation was particularly pronounced on adding socioeconomic indices to the multivariable model when the IQ-CHD relation was eliminated (1.05; 0.73,1.52). A similar pattern of association was apparent when cardiovascular disease was the outcome of interest. Conclusion High IQ may lead to educational success, well remunerated and higher prestige employment, and this pathway may confer cardio-protection.