BackgroundExercise interventions are highly effective at preventing falls in older people living in the community. In residential aged care facilities (RACFs), however, the evidence for effectiveness is highly variable, warranting exploration of drivers of successful trials. This study aims to identify the conditions of randomised controlled trials (RCTs) that are associated with reducing falls in RACFs and test whether it can explain the variability.MethodsRCTs testing exercise interventions in RACFs compared to usual care, reporting rate or risk of falls from the 2018 Cochrane Collaboration review and a search update to December 2022 were included. Two authors independently extracted and coded trial conditions and outcomes according to a theory developed from prior Intervention Component Analysis. Trial outcomes were coded as successful or unsuccessful based on point estimates for the rate or risk ratio for falls, or p value. Qualitative Comparative Analysis (QCA), utilising Boolean minimisation theory, was conducted to determine the key conditions driving trial success. A subgroup meta-analysis and the GRADE approach was applied to the final theory.ResultsEighteen trials undertaken in 11 countries with 2,287 residents were included. Participants were predominately ambulant females aged 70 to 80 with cognitive impairment. Most interventions were fully supervised or supervised at the start of the intervention. QCA identified two configurations as drivers of successful exercise falls prevention programs: (i) group exercise that is moderate or low intensity, or (ii) for independent ambulatory residents, exercise for more than 1 h per week. The combination of configuration (i) and (ii) had consistency and total coverage scores of 1, indicating all trials were explained. This combination was associated with a reduction in falls (rate ratio 0.45, 95%CI 0.34 to 0.59; risk ratio 0.66, 95%CI 0.53 to 0.82; low certainty evidence).ConclusionTo successfully reduce falls in RACFs, exercise programs should provide continuous supervised moderate-intensity group exercise. For programs that mostly include independent ambulatory residents, exercise for at least 80 min per week should be provided. As many current residents are frail, tailored exercise is likely necessary and an individualised dose may be required. Future trials should test exercise interventions for less mobile residents.