Twenty-eight RCTs were included (n=1,707, range 24 to 185). Study quality was reported to be good: 50% of the studies had a Jadad score of 3 and 18% of the studies had a Jadad score of 0 or 1; and 82% of the studies had a PEDro score of at least 5.
Fall and fracture risk (six RCTs): Four studies reported a reduction in falls, one also reported a reduction in fracture risk and one study showed no reduction in vertebral height with exercise compared to controls; results were not given for one study.
BMD and bone strength (11 RCTs): Results were inconsistent across 10 RCTs that compared exercise to an inactive control, although all trials reported benefits on at least one outcome. One trial reported benefits of power training over strength training.
Muscle strength (14 RCTs): Six of eight RCTs reported improved lower extremity muscle strength for exercise compared to inactive control. Three of five RCTs reported improved trunk muscle strength. In comparisons of different exercise programs, one RCT reported benefits of multiple-set over single-set muscle training and a second RCT reported benefits of power over strength training.
Balance (12 RCTs): Eight out of 10 RCTs that compared exercise to inactive control or a sham intervention reported a positive impact on balance.