Longitudinal effects of physical exercise on health-related outcomes based on frailty status in community-dwelling older adults.

Aim: To clarify the difference in the longitudinal effects of physical exercise on health-related outcomes according to the baseline frailty status (frail or non-frail) in community-dwelling older adults.

Methods: Participants included 177 adults aged ≥65 years who carried out multicomponent physical exercises (strength, aerobic, gait and balance) for 40 min, one to three times per week, for 1 year at a day-care center. Bodyweight, comfortable walking speed, 6-min walking distance and Mini-Mental State Examination were measured at baseline and every 3 months. For longitudinal trend, we analyzed the change in scores from baseline for each outcome using the linear mixed effects model. Fixed effects included "group" (frail or non-frail), "time" (4 time points every 3 months, from 3 to 12 months) and "interaction between group and time."

Results: The effect sizes from baseline showed almost all positive values for each outcome. The linear mixed effects model showed significant effects on "interaction between group and time" in changes in bodyweight (P = 0.033), "group" in changes in walking speed (P = 0.013) and "time" in changes in the Mini-Mental State Examination (P < 0.001). Bodyweight showed a decreasing trend in the non-frail group after 3 months, unlike in the frail group. For walking speed, moderate effect sizes (d = 0.67-0.74) were sustained over time in the frail group, as did lesser effect sizes (d = 0.26-0.40) in the non-frail group.

Conclusions: Exercise-based multicomponent interventions were effective for both groups. The longitudinal effects on walking speed and bodyweight were greater in the frail group. Geriatr Gerontol Int 2022; 22: 213-218.