To examine the temporal stability of the Type A behavior pattern (TABP) after a cardiac event, both a Videotaped Clinical Interview (VCI) (formerly known as the Videotaped Structured Interview) and the Jenkins Activity Survey (JAS) were administered to 81 male and 19 female patients three times during the first year after hospitalization for an initial myocardial infarction, coronary artery bypass graft, or both. There was no intervention other than the usual treatment provided by the medical care system. The test-retest reliability coefficients were moderately high for all measures of TABP. The JAS provided the most reliable scores in both sexes. The Hard Driving and Competitive component of the JAS decreased significantly throughout the year in both men (p less than 0.001) and women (p less than 0.01), but in men all three overall indices of TABP and their components (other than Hard Driving and Competitive) decreased spontaneously between 1 and 3 months, and then increased again by 1 year to nearly the original levels. In women, there were no consistent changes over time among the indices other than that for the Hard Driving and Competitive component of the JAS. Intercorrelations among the indices demonstrate the failure of the JAS to assess the Hostility component of the TABP and the possibility that the Speed and Impatience subscale of the JAS measures different things in men and women. The newest method of scoring the VCI appeared to provide the best index for measuring TABP in male coronary patients, while the original scoring was more reliable for women.