Use of antidepressant medications has been associated with increased risk of fracture, but prior studies have been limited by incomplete control of confounders or a limited number of fractures. Use of antidepressant medications by 8,217 community-dwelling women aged 69 and older from a population-based prospective cohort study at four US clinical centers was assessed by interview at four examinations over a 10-year period, beginning in 1992-1994. Use was coded as a time-dependent variable. Incident fractures occurring after the initial medication assessment until July 2007 were confirmed by radiographic reports. Potential confounders were included in multivariable models and updated at each follow-up visit. Compared to nonusers of antidepressant medications, women using SSRIs experienced a higher risk of nonspine fracture in age-adjusted models (HR = 1.36, 95% CI 1.11-1.67) and in multivariable models controlling for potential confounders (HR = 1.30, 95% CI 1.04-1.62). SSRI use was not associated with an increased risk of first hip fracture (HR = 1.01, 95% CI 0.71-1.44) but was associated with an increased risk of wrist fracture (HR = 1.54, 95% CI 1.01-2.36). TCA use was associated with an increased risk of nonspine fracture in age-adjusted models, but in multivariable models this risk was attenuated. SSRI use was associated with a higher risk of any nonspine fracture, but not hip fracture, in this cohort of older women. TCA use was associated with a higher risk of nonspine fracture, but this association was in part explained by confounding factors.