Background: Consuming fruit and vegetables (FVs) may protect against frailty, but to our knowledge no study has yet assessed their prospective dose-response relation. Objective: We sought to examine the dose-response association between FV consumption and the risk of frailty in older adults. Design: Data were taken from 3 independent cohorts of communitydwelling older adults: the Seniors-ENRICA (Study on Nutrition and Cardiovascular Risk Factors in Spain) cohort (n = 1872), Three-City (3C) Bordeaux cohort (n = 581), and integrated multidisciplinary approach cohort (n = 473). Baseline food consumption was assessed with a validated computerized diet history (Seniors-ENRICA) or with a food-frequency questionnaire (3C Bordeaux and AMI). In all cohorts, incident frailty was assessed with the use of the Fried criteria. Results across cohorts were pooled with the use of a randomeffects model. Results: During a mean 2.5-y follow-up, 300 incident frailty cases occurred. Fully adjusted models showed that the pooled ORs (95% CIs) of incident frailty comparing participants who consumed 1, 2, or $3 portions of fruit/d to those with no consumption were, respectively, 0.59 (0.27, 0.90), 0.58 (0.29, 0.86), and 0.48 (0.20, 0.75), with a P-trend of 0.04. The corresponding values for vegetables were 0.69 (0.42, 0.97), 0.56 (0.35, 0.77), and 0.52 (0.13, 0.92), with a P-trend , 0.01. When FVs were analyzed together, the pooled ORs (95% CIs) of incident frailty were 0.41 (0.21, 0.60), 0.47 (0.25, 0.68), 0.36 (0.18, 0.53), and 0.31 (0.13, 0.48), with a P-trend , 0.01 for participants who consumed 2, 3, 4, or $5 portions/d, respectively, compared with those who consumed #1 portion/d. An inverse dose-response relation was also found between the baseline consumption of fruit and risk of exhaustion, low physical activity, and slow walking speed, whereas the consumption of vegetables was associated with a decreased risk of exhaustion and unintentional weight loss. Conclusions: Among community-dwelling older adults, FV consumption was associated with a lower short-term risk of frailty in a doseresponse manner, and the strongest association was obtained with 3 portions of fruit/d and 2 portions of vegetables/d.Am J Clin Nutr 2016;104:132-42.
