ABSTRACT
BACKGROUND: Estimates of survival in the older can be of benefit in various facets, particularly in medical and individual decision-making. We aim to validate the value of a combination of nutrition status evaluation and comorbidity assessment in predicting long-term survival among community-dwelling older. METHODS: The Charlson Comorbidity Index (CCI) was applied for comprehensive evaluation of comorbidities. Participants were classified into CCI score ≤ 2 and ≥ 3 subgroups. Nutritional status was assessed by using Mini Nutritional Assessment-Short Form (MNA-SF) and Geriatric Nutritional Risk Index (GNRI) evaluations. Mortality rates and survival curves over a 5-year period were compared among subgroups classified by CCI and/or MNA-SF/GNRI evaluations. RESULTS: A total of 1033 elderly male participants were enrolled in this study, with an average age of 79.44 ± 8.61 years. 108 deceased participants (10.5%) were identified during a follow-up of 5 years. Cox proportional hazards regression analysis showed that age, CCI, MNA-SF and GNRI were independent predictors of 5-year all-cause death in this cohort. Compared to those with normal nutrition status and CCI ≤ 2, the subgroup at risk of malnutrition and CCI ≥ 3 had a significantly higher 5-year all-cause mortality rate (HR = 4.671; 95% CI:2.613-8.351 for MNA-SF and HR = 7.268; 95% CI:3.401-15.530 for GNRI; P < 0.001 for both). Receiver operating characteristic curve analysis demonstrated that a combination of either MNA-SF or GNRI with CCI had significantly better performance than CCI, MNA-SF or GNRI alone in predicting all-cause death. CONCLUSION: The combination of nutritional assessment (MNA-SF or GNRI) with CCI can significantly improve the predictive accuracy of long-term mortality outcomes among community-dwelling older males.
