Background and aims: Falls efficacy has been shown to predict functional decline, but whether it is independently associated with frailty is still unclear. This study investigated the cross-sectional association between falls efficacy and the frailty phenotype in high-functioning older persons. Methods: Subjects (n=861) were a sub-sample of community-dwelling persons aged 65 to 70 years enrolled in the "Lc65+" cohort, who had gait assessment. Data included demographics, functional, cognitive, affective and health status, as well as measures of physical performance. Falls efficacy was measured with the Falls Efficacy Scale-International (FES-I) and frailty with Fried's criteria. Participants were categorized into robust (no frailty criterion) and vulnerable (1 or more criteria). Low falls efficacy was defined as a FES-I score in the lowest quartile. Results: Overall, 23.9% of participants were vulnerable. Compared with robust participants, they were more likely to report low falls efficacy (43.3% vs 19.1%, p<0.001) and had poorer health and functional and mental status. They had slower gait speed (1.07+/-0.18 vs 1.15+/-0.15 m/s, p<0.001) and increased gait speed variability (coefficient of variation 4.10+/-4.03 vs 3.33+/-1.45%, p<0.001), although only 6 participants (0.7%) fulfilled Fried's slow walking criterion. In multivariate analysis, low falls efficacy remained associated with being vulnerable (adjusted OR 1.80, 95% CI 1.19-2.74, p=0.006), independent of comorbidity, functional status, falls history and gait performance. Conclusion: In high-functioning older persons, low falls efficacy was associated with vulnerability, even after controlling for gait performance and falls history. Whether low falls efficacy is a potential target on the pathway leading to frailty should be further examined prospectively. (Aging Clin Exp Res 2010; 22: 212-218) (C) 2010, Editrice Kurtis