BACKGROUND AND AIMS: Falls are particularly common among older people living in residential care facilities. The aim of this randomized controlled trial was to evaluate the effectiveness of a high-intensity functional exercise program in reducing falls in residential care facilities. METHODS: Participants comprised 191 older people, 139 women and 52 men, who were dependent in activities of daily living. Their mean±SD score on the Mini-Mental State Examination was 17.8±5.1 (range 10-30). Participants were randomized to a high-intensity functional exercise program or a control activity, consisting of 29 sessions over 3 months. The fall rate and proportion of participants sustaining a fall were the outcome measures, subsequently analysed using negative binominal analysis and logistic regression analysis, respectively. RESULTS: During the 6- month follow-up period, when all participants were compared, no statistically significant differences between groups were found for fall rate (exercise group 3.6 falls per person years [PY], control group 4.6 falls per PY), incidence rate ratio (95% CI) 0.82 (0.49-1.39), p=0.46, or the proportion of participants sustaining a fall (exercise 53%, control 51%), odds ratio (95% CI) 0.95 (0.52-1.74), p=0.86. A subgroup interaction analysis revealed that, among participants who improved their balance during the intervention period, the exercise group had a lower fall rate than the control group (exercise 2.7 falls per PY, control 5.9 falls per PY), incidence rate ratio (95% CI) 0.44 (0.21-0.91), p=0.03. CONCLUSIONS: In older people living in residential care facilities, a high-intensity functional exercise program may prevent falls among those who improve their balance.