Effect of body mass on the association between estrogen replacement therapy and mortality among elderly US women

In observational studies, estrogen replacement therapy is associated with d ecreased cardiovascular disease rates and increased breast cancer rates. Re cent evidence suggests that the impact of estrogen use on disease outcomes may vary by body mass. In a prospective study of 290,827 postmenopausal US women with no history of cancer or cardiovascular disease at enrollment in 1982, the authors examined the association between postmenopausal estrogen use and all-cause, coronary heart disease, stroke, all-cancer, and breast c ancer death rates and whether these associations differed by body mass. Aft er 12 years of follow-up, results from Cox proportional hazards models show ed that all-cause death rates were lower among baseline estrogen users than never users (rate ratio (RR) = 0.82, 95% confidence interval (CI): 0.78, 0 .87). The lowest relative risk was found for coronary heart disease (RR = 0 .66, 95% Cl: 0.58, 0.77). The inverse association between estrogen use and coronary heart disease mortality was strongest for thin women (body mass in dex <22 kg/m(2)) (RR = 0.49, p for interaction = 0.02). Breast cancer morta lity did not increase with estrogen use overall, and no increased risk was observed for thin or heavy women. in this population, the reduction in coro nary heart disease mortality among estrogen users was greatest for thinner women. Additional studies are needed to confirm or refute these results.