A comparison of the effects of three types of endurance training on balance and other fall risk factors in older adults

We hypothesized that short-term endurance training improves balance in older adults, if training involves movements that 'stress' balance. We tested the hypothesis by looking for a dose-response relationship between movement during exercise and balance improvement. The study was a single-blinded, randomized controlled trial. Subjects were sedentary adults (N=106) aged 68-85 with at least mild deficits in balance. Exercise groups were: stationary cycle (low movement), walking (medium movement), and aerobic movement (high movement). Subjects attended supervised exercise classes three times a week for three months, followed by self-directed exercise of any type for three months. The primary test of the hypothesis compared changes in balance after three months of supervised exercise. One balance measure (distance walked on a six-meter narrow balance beam) improved in the hypothesized dose-response manner (cycle, 3% improvement; walking, 7% improvement; aerobic movement, 18% improvement: p<0.02, test of trend). Other balance measures did not improve with exercise. Only walking exercise improved gait speed (by 5%, p<0.02) and SF-36 role-physical score (by 24%, p<0.05). VO₂max improved with walking (18%, p<0.004) and aerobic movement (10%, p<0.01), but improved less with cycling (8%, p>0.1). Leg strength improved significantly in all exercise groups. The study hypothesis was supported only for one balance measure. Only walking improved at least one measure of all major outcomes (endurance, strength, gait, balance, health status), suggesting that walking is most useful for fall prevention. Cycle exercise appeared least useful.