The authors prospectively studied the overall effect of hormone replacement therapy (HRT) on hip fracture risk and the effect modification by behavioral habits and body mass index. A total of 6,159 postmenopausal women from the Copenhagen Center for Prospective Population Studies, Copenhagen, Denmark, with initial examination in 1976-1978 were followed until 1993. During follow-up 363 hip fractures were identified. Women who reported current use of HRT had a lower risk of hip fracture as compared with women who were nonusers (relative risk (RR) = 0.71; 95 percent confidence interval (CI): 0.50, 1.01). Use of HRT was associated with a lower risk of hip fracture in former (RR = 0.55; 95 percent CI: 0.22, 1.37) and current (RR = 0.61; 95 percent CI: 0.38, 0.99) smokers but not in never smokers (RR = 1.10; 95 percent CI: 0.60, 2.03). HRT was also associated with lower risk of hip fracture among alcohol drinkers (RR = 0.36; 95 percent CI: 0.14, 0.90) and among sedentary women (RR = 0.42; 95 percent CI: 0.18, 0.98) but not among nondrinkers (RR = 0.99; 95 percent CI: 0.61, 1.61) and physically active women (RR = 0.92; 95 percent CI: 0.42, 2.04). There was no evidence of interaction between use of HRT and body mass index. In conclusion, the protective effect of HRT on hip fracture appears to be strongest in women who ever smoked, in women who drink alcohol, and in women who are sedentary. The results suggest that history of behavioral habits offers important information concerning the probable degree of protection against hip fracture afforded by HRT.