One hundred and seventy-nine women aged 60-85 years (mean age 71.6 years, SD 5.3 years) were randomly recruited from the community to participate in a 12-month randomized controlled trial to determine whether a program of twice-weekly structured exercise has beneficial effects on three factors associated with osteoporotic fractures: quadriceps strength, postural sway and bone density. At initial testing, there were no significant differences in the strength, sway and bone density measures (assessed at the hip and lumbar spine) between the exerciser and control groups. The exercise classes included strengthening, coordination and balance exercises, and approximately 35 min of each class comprised weight-bearing exercise. The mean number of classes attended for the 68 exercisers who completed the program was 59.8 of the 82 classes (72.9%). At the completion of the trial, the intervention group showed significant improvements in quadriceps strength and sway but not bone mineral density when compared with the control group. Indices of fracture risk, indicated by (i) the sum of standard score results and (ii) the sum of quartile grades of the femoral neck bone density, sway and strength measures, decreased significantly in the exercisers at the end of the trial compared with the controls. In conclusion, the program of general aerobic exercise may have reduced overall fracture risk, even though it did not significantly increase bone density. Further long-term studies are required that include acceptable weight-loaded exercises to determine optimal programs for reducing fracture risk factors by improving bone density as well as strength and balance.