The purpose of this study was to examine the relationship of social support and religion to mortality after elective open heart surgery in older patients.Of the 232 patients included in the study, 21 died within 6 months of surgery. Three biomedical variables were significant predictors of mortality and selected as adjustment variables for a multivariate analysis: history of previous cardiac surgery; greater impairment in presurgery basic activities of daily living; and older age. Among the social support and religion variables, two were consistent predictors of mortality in the multivariate analyses: lack of participation in social or community groups and absence of strength and comfort from religion. These results suggest that in older persons lack of participation in groups and absence of strength and comfort in religion are independently related to risk for death during the 6-month period after cardiac surgery.