The relation between coronary heart disease and disability was examined in 2,576 community-dwelling women and men ages 55-88 years. These Framingham Study participants were originally recruited in 1948-51 for an examination of cardiovascular disease. Twenty-seven years later, remaining members of the cohort were interviewed to ascertain physical abilities, and a score on a disability scale was assigned. Multivariate logistic analyses examined disability in relation to uncomplicated angina pectoris (AP), complicated AP, and coronary heart disease other than AP, controlling for possible confounders. In younger and older women and men, uncomplicated and complicated AP were associated with disability. Coronary heart disease other than AP was associated with disability only in the younger men. Congestive heart failure predicted disability only in the women. These results suggest that onset of AP should be recognized as a critical point in the development of disability and that AP is a better predictor of disability than is myocardial infarction or coronary insufficiency.