Criterion validity of the Global Leadership Initiative on Malnutrition criteria for malnutrition diagnosis compared with the Subjective Global Assessment: Results from a large observational study.

Abstract

The aim of this study was to assess the criterion validity of the Global Leadership Initiative on Malnutrition criteria compared to the subjective global assessment in a diverse inpatient population. This cross-sectional study was a retrospective analysis of point prevalence audit data. The prevalence of malnutrition determined by the Global Leadership Initiative on Malnutrition criteria was compared to the Subjective Global Assessment. Validity statistics were determined using all of the Global Leadership Initiative on Malnutrition criteria concurrently as well as each pair that could be used to diagnose malnutrition. Subgroup analysis was undertaken based on severe malnutrition, treatment group, age and body mass index. Nine hundred and eighty-one patients were included (65.1 ± 18.6 years, 54.8% male). The prevalence of malnutrition was 36.7% using the Subjective Global Assessment and 36.1% using the Global Leadership Initiative on Malnutrition criteria. More patients were classified as severely malnourished using the Global Leadership Initiative on Malnutrition criteria (9.8% vs. 6.0%), whilst more rehabilitation patients were classified as malnourished using the Subjective Global Assessment (42.2% vs. 33.6%). The criterion validity of the Global Leadership Initiative on Malnutrition criteria was good, with a sensitivity of 92.5% (95% CI 90.9-94.2) and specificity of 96.6% (95% CI 95.5-97.8). There was a downward trend in sensitivity with increasing body mass index and a lower sensitivity in the rehabilitation population. The criterion validity was fair at best when each pair of the Global Leadership Initiative on Malnutrition criteria was considered independently of other criteria. When all criteria are considered concurrently, the Global Leadership Initiative on Malnutrition criteria present good criterion validity and can be applied in clinical practice to diagnose malnutrition.