The risk of osteoporotic fracture is related to peak bone mass achieved at skeletal maturity and subsequent bone loss. Although premature menopause is a risk factor for osteoporosis, the effect of exposure to endogenous estrogen during a woman's reproductive years is poorly characterized. We analyzed the relationship between reproductive factors and estrogen exposure on bone mineral density (BMD) and incidence of atraumatic fracture in data from 1091 women (age: 70 +/- 7.2 yr; mean +/- SD) participating in the Dubbo Osteoporosis Epidemiology Study. Age- and weight-adjusted BMD among women who had used estrogen replacement therapy (ERT) for more than 5 yr was higher at the lumbar spine and femoral neck by 13.7% and 10.2% (P < 0.001), respectively, compared with women who had used ERT for less than 5 yr or nonusers. Duration of exposure to estrogen (years of menstruation plus postmenopausal ERT use) was associated with higher BMD, such that BMD increased by 2-3% for every 10-yr increase in years of estrogen exposure; thus women who menstruated for more than 40 yr had a 6-8% higher BMD than did women who menstruated for less than 30 yr. Higher BMD was also significantly associated with high parity, such that nulliparous women had 5-6% lower BMD than did their peers of the same age and weight. The incidence of atraumatic fractures among non-ERT users was higher than that of ERT-users [odds ratio (OR): 1.06; 95% confidence interval (CI): 0.94-1.16] and was significantly lower among parous women than among nulliparous women (OR 0.94; 95% CI: 0.84-0.98) in univariate analysis. Longer duration of menstruation was associated with lower fracture incidence (OR for 1 SD = 6.6 yr: 0.93; 95% CI: 0.86-1.02). Moreover, when all of these factors were considered simultaneously, parity remained a significant determinant of fracture as well as femoral neck BMD. We conclude that high parity and longer duration of exposure to estrogen, either through natural menstruation or postmenopausal ERT, have protective effects on BMD and are associated with a reduced incidence of atraumatic fracture in a population-based study.