Predictive Effect of Malnutrition on Long-Term Clinical Outcomes among Older Men: A Prospectively Observational Cohort Study.

Abstract

To determine whether nutritional status can predict 3-year cognitive and functional decline, as well as 4-year all-cause mortality in older adults. Prospectively longitudinal cohort study. The study recruited 354 men aged 65 years and older in the veteran's retirement community. Baseline nutritional status was evaluated using the Mini-Nutritional Assessment-Short Form (MNA-SF). Cognitive function and Activities of Daily Living (ADL) function were determined by the Mini-Mental State Examination (MMSE) and the Barthel Index, respectively. Three-year cognitive and functional decline were respectively defined as a >3 point decrease in the MMSE scores and lower ADL scores than at baseline. Univariate and multivariable logistic regression analyses were conducted to identify nutritional status as a risk factor in poor outcome. The Kaplan-Meier method and Cox proportional regression models were used to estimate the effect of malnutrition risk on the mortality. According to MNS-SF, the prevalence of risk of malnutrition was 53.1% (188/354). Multivariate logistic regression found risk of malnutrition significantly associated with 3-year cognitive decline (Adjusted odds ratio [OR] 2.07, 95% Confidence Interval [CI] 1.05-4.08, P =0.036) and functional decline (Adjusted OR 1.83, 95% CI 1.01-3.34, P =0.047) compared with normal nutritional status. The hazard ratio (HR) for all-cause mortality was 1.8 times higher in residents at risk of malnutrition (Adjusted HR 1.82, 95% CI 1.19-2.79, P =0.006). Our results provide strong evidence that risk of malnutrition can predict not only cognitive and functional decline but also risk of all-cause mortality in older men living in a veteran retirement's community. Further longitudinal studies are needed to explore the causal relationship among nutrition, clinical outcomes, and the effect of an intervention for malnutrition.