GLIM in nursing homes; practical implications

Background: Only very few papers have described malnutrition prevalence rates according to the Global Leadership Initiative on Malnutrition (GLIM) criteria in nursing homes, likely due to practical reasons such as missing data on body composition, dietary intake, or acute disease/inflammation. Methods: Data was collected in 5 different nursing homes. Food intake measurements took place over 3 days of observations, and intakes below 90% of energy or protein requirements were regarded as insufficient. The GLIM diagnosis was based on body weight loss and/or low BMI in combination with insufficient food intake. Additionally, we also studied the sensitivity of GLIM with the question from the Mini Nutritional Assessment Short Form (MNA-SF) on insufficient food intake (GLIM_MNA) versus GLIM with measured food intake. Results: Out of 176 participants, 21.0% were categorized as malnourished according to GLIM. Observations revealed an insufficient food intake in 81.3% (N = 143) of residents; only 39% of those (N = 56) scored positive on the MNA-SF question regarding low food intake. GLIM_MNA diagnosed 17.0% of residents as malnourished. Sensitivity of GLIM_MNA for GLIM was 62.2%, and specificity 95.0% (kappa = 0.61). Conclusion: Twenty-one percent of nursing home residents were diagnosed malnourished based on a limited set of GLIM criteria. The MNA question on insufficient food intake missed ∼60% of residents with a truly low food intake. Herewith, malnutrition prevalence rates with GLIM_MNA decreased to 17%. We advise measuring food intake for studies, and to be aware of too low prevalence rates of GLIM when an estimate of reduced food intake is applied.