Objectives: This study sought to evaluate the impact of exercise on balance function in individuals with knee osteoarthritis (KOA) and determine the most effective exercise protocols for balance enhancement. Methods: A systematic literature search was performed across five major electronic databases (PubMed, Web of Science, EBSCO, Cochrane, Scopus) until 13 September 2024. Statistical synthesis was conducted using weighted mean differences (WMDs) with 95% confidence intervals under a random-effects model. Results: Analysis of 22 studies revealed significant improvements in balance function following exercise interventions. Outcomes measured by the Berg balance scale (BBS, WMD, 2.65, p < 0.00001) and timed up and go test (TUG, WMD, −0.59, p < 0.0001) demonstrated clinically relevant enhancements in KOA populations. Subgroup analyses revealed that multicomponent training (WMD, 6.25, p = 0.003), interventions lasting ≥ 8 weeks (WMD, 4.92, p = 0.002), sessions ≥ 60 min (WMD, 7.42, p = 0.002), frequency ≥ 3 times per week (WMD, 2.83, p = 0.0005), weekly time ≥ 180 min per week (WMD, 7.42, p = 0.002), and patients < 60 years (WMD, 6.71, p = 0.002) were associated with greater improvement in BBS. Conclusions: Exercise significantly improved balance function in KOA patients, with multicomponent training emerging as the most effective intervention. Based on the findings of this meta-analysis, clinicians should recommend that KOA patients engage in exercise at least three times per week, with each session lasting at least 60 min, to achieve a total weekly time of 180 min. These recommendations are particularly relevant for patients less than 60 years, who may experience greater benefits from exercise interventions.

Background: Although some studies have examined the association between exercise and falls, most have focused on specific exercises, and the results have been inconsistent. In addition, there is a lack of evidence on elderly Chinese women who have different living and exercise habits compared to those in other countries. Therefore, this study aimed to investigate whether physical exercise is associated with falls in elderly Chinese women. Methods: This cross-sectional study included 1429 elderly Chinese women with a mean age of 69.2 years. Information on physical exercise habits and fall experiences was collected using a self-report questionnaire. Logistic regression models were used to analyze the association between physical exercise habits and falls. Results: The results showed that 15% participants had a fall in the past year. After adjusting for confounding factors, the odd ratios (ORs) and 95% Confidence Intervals (CIs) for fall experiences across categories of exercise frequency were as follow: 1 (reference) for no exercise behavior, 0.50 (0.29, 0.85) for exercise 1 to 5 times a week, and 0.37 (0.25, 0.55) for exercise more than 6 times a week. Furthermore, the ORs (95% CIs) across categories of exercise insistence were 1 (reference) for less than 1 year, 0.78 (0.37, 1.65) for 1 to 3 years, and 0.38 (0.20, 0.74) for more than 3 years. In terms of exercise duration, the ORs (95% CIs) for < 1 h/day, 1–2 h/day, and > 2 h/day were 1 (reference), 0.85 (0.53, 1.36), and 2.80 (1.30, 6.05). Unlike other variables, longer exercise duration was associated unfavorably with falls. Conclusion: Physical exercise habits were associated with falls in elderly Chinese women. Keeping a proper exercise habit may contribute to lower risk of falling in elderly women.