We examined the relation between menopausal estrogen use and all-cause and cause-specific mortality in a cohort of over 49,000 women followed between 1979 and 1989 in the Breast Cancer Detection Demonstration Project (BCDDP) Follow-Up Study. We found a lower all-cause mortality rate among women who took estrogens [rate ratio (RR) = 0.7; 95% confidence interval (CI) = 0.7-0.8], particularly current users (RR = 0.3; 95% CI = 0.2-0.4), than among women who never took them. Additional analyses, however, revealed that women who had recently stopped taking estrogens had a higher all-cause mortality rate than women who had never taken them (RR = 1.4; 95% CI = 1.2-1.7). Women who had recently stopped taking estrogens also had higher mortality rates from circulatory disease (RR = 1.3; 95% CI = 1.0-1.8) and cancer (RR = 1.6; 95% CI = 1.2-2.2) than women who never took them. The most likely explanation for these results is that women stop taking estrogens when they develop symptoms of serious illness. As a consequence of this "healthy estrogen user survivor effect," nonexperimental studies are susceptible to overestimating the benefits of menopausal estrogen use, particularly current use, on mortality.