Impact of the Controlling Nutritional Status (CONUT) score as a prognostic factor for all-cause mortality in older patients without cancer receiving home medical care: hospital ward-based observational cohort study

Abstract

ObjectivesMalnutrition in cancer-free older patients receiving home medical care may affect prognosis, but indicators of long-term nutrition-related prognosis have not been developed. This study investigated the utility of the Controlling Nutritional Status (CONUT) score as a prognostic factor for older patients without cancer receiving home medical care.DesignThis was a single-centre, hospital ward-based observational cohort study.Setting and participantsIn total, 625 cancer-free older patients (median age, 81.0 years; 47.4% males) receiving continuous home medical care through clinics were enrolled on admission to a hospital ward from March 2011 to September 2018.Primary outcome measuresContinuous cumulative survival curves were obtained using the Kaplan-Meier method after dividing the CONUT score into four groups. The prognostic factors for overall mortality were evaluated using the Cox proportional hazards model. Comparisons with other predictive tools were performed.ResultsThe Kaplan-Meier curves of CONUT scores revealed a stepwise shortening of the median survival time with increasing scores. The HR of CONUT scores adjusted by age, sex and other confounding variables was 1.422 (95% CI 1.232 to 1.643, p<0.001). The areas under the receiver operating characteristic curve of the CONUT score for 1-year and 5-year survival were 0.684 and 0.707, respectively. The CONUT score displayed greater predictive utility than other nutrition-related predictive tools.ConclusionsThe CONUT score on hospital admission could be used to predict overall mortality in older patients without cancer receiving home medical care. It is expected to be a simpler and cheaper screening tool for assessing the nutritional status in the field of home medical care.