There are many factors which affect the fracture risk in elderly women. Among them we stress low bone mass, increased occurrence of falls, and anthropometric characteristics and physical activity. Low bone mass in postmenopausal women may come from genetic and environmental factors. The non-modifiable factors are gender and age. The modifiable ones are related to change of habits, including a sedentary lifestyle. Low bone mass is prevalent in women and the problem worsens with aging. Fractures represent a high cost in hospitalization, rehabilitation, morbidity and mortality. PURPOSE: The aim was to analyze the correlation among the history of fractures over the age of 50, physical activity and body mass index (BMI) in elderly women. METHODS: The study was approved by the Research Ethics Committee of PUCRS, Letter No. 05029/35, and the sample consisted of 284 elderly from 60 to 94 years with an average of 71.2 ± 7.8 years, who had or had not done physical activity for at least 3 months on a regular and systematic basis, with a BMI between 14.3 and 44.8. Data were analyzed using descriptive and inferential statistics (chi-square test). RESULTS: Among the 284 elderly women, 76 had a history of fractures over the age of 50, 28 of these were physically active e the other 48 were not. Also, among the 97 physically active ones, 28 women (28.9%) had fractures, 11.1% with BMI ≤ 22, 28.2% with BMI 22-27, and 32.7% with BMI ≥ 27. Among the 187 non-physically active ones, 48 women (25.7%) had a history of fractures, 22.7% of those with BMI ≤ 22, 28.3% with BMI between 22-27 and 24.8% with BMI ≥ 27. CONCLUSION: Physical activity plays a fundamental role in increasing bone mass during childhood and adolescence, and it's a necessary stimulus for one to reach his/her peak bone mass in adulthood. Some studies in postmenopausal women are inconclusive on the association between physical activity and bone mass maintenance, and others, point out that only exercise or calcium supplementation alone can not compensate for the loss of bone mass and suggest that the benefits of exercise on low bone mass works if there is a high calcium and D vitamin intake, suggesting that physical activity, D vitamin and calcium would act synergistically. This leads to the need of correlating variables related to physical activity.