Background Falls are a major threat to older adults worldwide. Although various effective interventions have been developed, their comparative effectiveness remains unreported.

Method and findings Nine databases were searched up to December 31, 2016 to identify randomized controlled trials comparing fall interventions. A systematic review and network meta-analysis was conducted to determine the most effective interventions to prevent falls in community-dwelling adults aged 60 and over. A total of 49 trials involving 27 740 participants and 9271 fallers were included. Compared to usual care, multifactorial interventions (MFI) demonstrated the greatest efficacy (0.64, 95% CrI: 0.53 to 0.77) follow by interventions combining education and exercise (EDU + EXC) (0.65, 95% CrI: 0.38 to 1.00) and interventions combining exercise and hazard assessment and modification (EXC + HAM) (0.66, 95% CrI: 0.40 to 1.04). The effect of medical care performed the worst (1.02, 95% CrI: 0.78 to 1.34). Model fit was good, inconsistency was low, and publication bias was considered absent. The overall quality of included trials was high. The pooled odds ratios and ranking probabilities remained relatively stable across all sensitivity analyses.

Conclusions MFI and exercise appear to be effective to reduce falls among older adults, with MFI showing the highest efficacy. Those approaches should be considered first as service delivery options. Further investigation is necessary to verify effectiveness and suitableness of the strategies to at-risk populations.