Abstract
OBJECTIVE: Frailty is a significant public health issue facing aging societies and can be reduced by physical activity (PA), but the dose-response relationship between PA and frailty is not clear. This systematic review and dose-response meta-analysis aimed to assess the effect of PA on frailty in adults by aggregating data from observational studies. METHODS: PubMed, Embase, Web of Science, Cochrane Library, Scopus, SAGE Reference Online, SinoMed, CINAHL and CNKI were retrieved for articles published before May 2024. After quality evaluation, data on PA and the risk of frailty were extracted. Stata/MP 17.0 was used for dose-response meta-analysis. RESULTS: A total of 15 articles were included, involving 34,754 participants, including 4250 subjects with frailty or pre-frailty. The consequence of the dose-response meta-analysis revealed that compared with those who were not active at all, a 22 % (95 % CI, 16 %-28 %) reduction in the risk of frailty in individuals with 11.25 MET h/week of cumulative activity and a 55 % (95 % CI, 44 %-63 %) reduction in the risk of frailty in those with 22.5 MET h/week of cumulative activity; for higher activity levels (36.75 MET h/week), the risk of frailty was reduced by 68 % (95 % CI, 58 %-76 %) and continued to be reduced as PA volum increased. CONCLUSIONS: There is a non-linear dose-response relationship between PA and frailty risk. Even small amounts of PA could reduce the risk of frailty. Meeting the minimum recommended PA target could reduce some risks, and doubling the recommended PA volumes could reduce most risks, which continue to increase as the volum of PA accumulates.