The nutrition impact symptoms (NIS) score detects malnutrition risk in patients admitted to nephrology wards

Abstract

Background: Nutritional screening tools recommended for the general hospitalised population do not always adequately detect malnutrition risk in patients with kidney disease. The present study assessed the validity and reliability of the Nutrition Impact Symptoms (NIS) score as a nutrition screening tool for hospitalised inpatients prefer in nephrology wards.

Methods: Nutritional status was classified using Subjective Global Assessment (SGA). NIS scores were calculated from the total score of responses to questions assessing symptoms impacting upon nutritional status from the patient‐generated SGA. Concurrent validity of NIS score was assessed using a receiver operating characteristic curve to predict malnutrition risk against SGA. Predictive validity was examined against length of hospital stay (LOS) and 30‐day re‐admission using Poisson and logistic regression, respectively. Inter‐rater reliability of NIS scoring between assessors was determined using intraclass correlation.

Results: In 143 patients [90 males; mean (SD) age 57.8 (15.8) years], malnutrition prevalence was 38% (54/143) using SGA (rating B/C). Predicting malnutrition risk with an NIS score of ≥3 had a sensitivity of 0.89 and a specificity of 0.65 (area under the curve = 0.81 [95% confidence interval (CI) = 0.74–0.88]). For each 1‐point increase in NIS score, the model predicted a 1.9% rise in the risk of an increased LOS (P = 0.002). Thirty‐day re‐admission was not associated with NIS score. Inter‐rater reliability was moderate (mean difference = 0.53; intraclass correlation coefficient = 0.74; 95% CI = 0.57–0.85).

Conclusions: Nutrition impact symptoms score is a valid stand‐alone nutrition screening tool for identifying malnutrition risk in nephrology inpatients and is associated with LOS.