Purpose: Assess the effect of a neuromuscular exercise program in the risk of falls in the elderly residing in nursing homes. Methodology: Comparative cross-sectional pre-experimental study. A total of 20 seniors, ages 68 to 80 residing in a nursing home, participated in a neuromuscular training program 3 times a week for 12 weeks. Participants were evaluated on their risk of falling using the Tinetti test prior to training, 6 and 12 weeks. Normal data distribution was verified with the Shapiro Wilk test, while instrument reliability was calculated using Cronbach's alpha and scores differences using repeated measures ANOVA. The value p < 0.05 was considered significant. Results: The analysis of data distribution showed no significant difference (p > 0.05); instrument reliability was 0.96; variance analyses identified significant differences between risk scores measured at the beginning v/s 6 weeks, 6 weeks v/s 12 weeks and at the beginning v/s 12 weeks exercise (p < 0.05). Conclusion: Neuromuscular exercises reduce the risk of falls in institutionalized elderly. Implication: In virtue of the results obtained, this type of exercise could be used as an alternative to physical activity for residents to continue to be functional, thereby reducing social, psychological and economic costs involved in a fall.