Evidence on physical activity and falls prevention for people aged 65+ years: systematic review to inform the WHO guidelines on physical activity and sedentary behaviour

BACKGROUND: Exercise prevents falls in older adults. Regular updates of estimated effects of exercise on falls are warranted given the number of new trials, the increasing number of older people globally and the major consequences of falls and fall-related injuries. METHODS: This update of a 2019 Cochrane Review was undertaken to inform the World Health Organization guidelines on physical activity and sedentary behaviour. Searches were conducted in six databases. We included randomised controlled trials evaluating effects of any form of physical activity as a single intervention on falls in people aged 60+ years living in the community. Analyses explored dose-response relationships. The certainty of the evidence was assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE). RESULTS: This review included 116 studies, involving 25,160 participants; nine new studies since the 2019 Cochrane Review. Exercise reduces the rate of falls by 23% (pooled rate ratio (RaR) 0.77, 95% confidence interval (CI) 0.71 to 0.83, 64 studies, high certainty evidence). Subgroup analysis showed variation in effects of different types of exercise (p < 0.01). Rate of falls compared with control is reduced by 24% from balance and functional exercises (RaR 0.76, 95% CI 0.70 to 0.82, 39 studies, high certainty evidence), 28% from programs involving multiple types of exercise (commonly balance and functional exercises plus resistance exercises, RaR 0.72, 95% CI 0.56 to 0.93, 15 studies, moderate certainty evidence) and 23% from Tai Chi (RaR 0.77, 95% CI 0.61 to 0.97, 9 studies, moderate certainty evidence). The effects of programs that primarily involve resistance training, dance or walking remain uncertain. Interventions with a total weekly dose of 3+ h that included balance and functional exercises were particularly effective with a 42% reduction in rate of falls compared to control (Incidence Rate Ratio (IRR) 0.58, 95% CI 0.45 to 0.76). Subgroup analyses showed no evidence of a difference in the effect on falls on the basis of participant age over 75 years, risk of falls as a trial inclusion criterion, individual versus group exercise, or whether a health professional delivered the intervention. CONCLUSIONS: Given the strength of this evidence, effective exercise programs should now be implemented at scale. SUPPLEMENTARY INFORMATION: Supplementary information accompanies this paper at 10.1186/s12966-020-01041-3.