Results The cumulative incidence of ADL disability was lower in the exercise groups (37.1%) than in the attention control group (52.5%) (P = .02). After adjustment for demographics and baseline physical function, the relative risk of incident ADL disability for assignment to exercise was 0.57 (95% confidence interval, 0.38-0.85; P = .006). Both exercise programs prevented ADL disability; the relative risks were 0.60 (95% confidence interval, 0.38-0.97; P = .04) for resistance exercise and 0.53 (95% confidence interval, 0.33-0.85; P = .009) for aerobic exercise. The lowest ADL disability risks were found for participants with the highest compliance to exercise.