Postmenopausal estrogen use is associated with a reduced risk of heart disease and hip fracture; in observational studies, different behaviors among hormone users and nonusers may partially explain these results. We examined risk of cardiovascular disease and hip fracture with medium-potency compared with low-potency or short-term estrogen use, and the effect of added progestin, among 9,236 women in Uppsala, Sweden, who responded to a mailed questionnaire in 1987-1988. Using population registries, we identified 213 cases of myocardial infarction, 289 strokes, and 114 hip fractures from 1987-1995. We found a reduced risk of myocardial infarction for medium-potency compared with low-potency or short-term estrogen use (relative risk = 0.75, 95% confidence interval (CI) = 0.56-0.99), with a similar decrease in the subgroup that took estrogens with progestin (RR = 0.69, 95% CI = 0.45-0.90). There was no relation of medium-potency estrogen to stroke (RR = 0.91, 95% CI = 0.71-1.17, and RR = 0.81, 95% CI = 0.61-1.10 for the subgroup taking progestin), and no effect of duration on either heart disease or stroke. We observed a reduction in hip fractures for medium-potency use (RR = 0.65, 95% CI = 0.45-0.95), and for use of combined estrogen-progestin therapy (RR = 0.64, 95% CI = 0.41-1.00). These data support a decreased risk of heart disease and hip fracture for medium-potency estrogen use alone or with progestin; self-selection to hormone use cannot explain these reductions.