Dose-response associations between physical activity and sedentary time with functional disability in older adults with or without frailty: a prospective cohort study.

PURPOSE: Evidence regarding the dose-response curve shapes of physical activity (PA) and sedentary time (ST) in older adults with functional disability (FD) is extremely limited. Moreover, these associations may differ depending on with/without frailty. We examined the dose-response associations between moderate-to-vigorous PA (MVPA) and ST with FD among older adults with/without frailty.

METHODS: We included 7,480 initially nondisabled adults (3,795 men and 3,685 women) aged 65-84 years in Ota City, Tokyo, Japan. MVPA and ST were evaluated using the International Physical Activity Questionnaire-Short Form. FD was prospectively identified using a nationally unified database of the long-term care insurance system. Frailty was determined using Check-List 15, validated against Fried's frailty criteria. Multivariate-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of MVPA and ST for FD were calculated, and dose-response curves were examined using restricted cubic splines.

RESULTS: During 3.6 years of follow-up, 1,001 (13.4%) participants had FD. Among all participants, compared with no MVPA, the HRs for FD reduced linearly up to approximately 2000 metabolic equivalents (METs)■min/week of MVPA, and the lowest HR (HR: 0.61, 95% CI: 0.51-0.74) was reached at around 3,000-4,000 METs■min/week. Although the shape of this association was consistent regardless of with/without frailty, the magnitude of the association tended to be stronger in frail older adults than in non-frail older adults. Compared with those for the median (300 min/day) of ST, the HRs for FD increased linearly as ST reached approximately 600 min/day or more, independent of MVPA, with a maximum HR of 1.31 (95% CI: 1.01-1.71) for 1,080 min/day among all participants. This association was more pronounced among non-frail older adults but not statistically significant among frail older adults.

CONCLUSION: Higher MVPA levels consistently reduced the incidence of FD regardless of frailty in a significant inverse nonlinear dose-response manner. A significant positive nonlinear dose-response association between ST and FD risk was identified among non-frail older adults but not among frail older adults. Increasing MVPA and reducing prolonged ST are important for preventing FD among non-frail older adults. However, reducing ST alone may be insufficient; increasing MVPA, even if by only small increments, is highly recommended for frail older adults.