To better understand which women use estrogen replacement therapy, we examined the prevalence and determinants of estrogen replacement therapy in 9704 nonblack women, age greater than or equal to 65 years, who participated in the multicenter prospective Study of Osteoporotic Fractures. Overall, 13.7% of women reported current use of oral estrogen; 10.9% took estrogen alone and 2.8% took estrogen opposed by progestin. Four percent currently used parenteral estrogen compounds. Current use declined sharply with age from 17% at age 65 to 4% at age greater than or equal to 85. The primary determinant of estrogen replacement therapy was the type of menopause; the odds of using estrogen replacement therapy in current users compared with never users were approximately five times higher in women with a surgical menopause. Estrogen use was more common among women who had higher levels of education and were less obese. Furthermore, estrogen replacement therapy users were more likely to drink alcohol and to participate in sports and recreation. A diagnosis of osteoporosis was the major determinant of continued estrogen use, but only 24% of women with a diagnosis of osteoporosis used estrogen replacement therapy. We conclude that only a small proportion of elderly women in the United States use estrogen replacement therapy. Selection factors for use of estrogen are evident and may introduce bias in studies of estrogen and disease. In consideration of the distribution of these selection factors, estrogen users will tend to be at lower risk of coronary disease and possibly breast cancer but at greater risk for hip fractures.