Type A/B behavior has been studied extensively as a risk factor for coronary heart disease. This study tests the hypothesis that Type A/B behavior is associated with ill health in general. All-cause mortality was examined in the original cohort of 3,154 men aged 39-59 years at baseline in the Western Collaborative Group Study in 1960-1961. Morbidity was examined in a group of 2,365 men aged 61-81 years who were surviving members of the cohort in 1982-1983 and who agreed to fill out a self-administered health status questionnaire. On the basis of these data, a three-level general morbidity index was constructed that included coronary heart disease and a number of other chronic conditions. Type A/B behavior was not significantly associated with all-cause mortality over the 22-year follow-up. However, among men 61-71 years of age, behavior type assessed at the baseline examination 22 years earlier (1960-1961) was positively associated with the morbidity index (p less than 0.001). Furthermore, behavior type and the morbidity index showed a dose-response association; extreme Type A personalities had the highest level of morbidity, while those with the fewest Type A characteristics had the lowest level of morbidity. This general pattern of association was not due to the presence of coronary heart disease or to any other particular disease condition in the morbidity index and did not change when an ordinal logistic regression model was used to control for other baseline risk variables. For the older men (aged 72-81 years), the pattern of association was similar but did not reach statistical significance. Thus, in a group of men over 60, a positive association was found between general morbidity status and Type A/B behavior measured over two decades earlier.