Beyond Malnutrition Screening: Appropriate Methods to Guide Nutrition Care for Aged Care Residents

BACKGROUND: Malnutrition is common in older adults and early and appropriate nutrition intervention can lead to positive quality of life and health outcomes. OBJECTIVE: The purpose of our study was to determine the concurrent validity of several malnutrition screening tools and anthropometric parameters against validated nutrition assessment tools in the long-term-care setting. STUDY DESIGN: This work was a cross-sectional, observational study. PARTICIPANTS/SETTING: Older adults (aged >55 years) from two long-term-care facilities were screened. MAIN OUTCOMES: Nutrition screening tools used included the Malnutrition Screening Tool (MST), Malnutrition Universal Screening Tool (MUST), Mini Nutritional Assessment-Short Form (MNA-SF), and the Simplified Nutritional Assessment Questionnaire. Nutritional status was assessed by Subjective Global Assessment (SGA), Mini Nutritional Assessment (MNA), body mass index (BMI), corrected arm muscle area, and calf circumference. Residents were rated as either well nourished or malnourished according to each nutrition assessment tool. STATISTICAL ANALYSIS: A contingency table was used to determine the sensitivity and specificity of the nutrition screening tools and objective measures in detecting patients at risk of malnutrition compared with the SGA and MNA. RESULTS: One hundred twenty-seven residents (31.5% men; mean age 82.7±9 years, 57.5% high care) consented. According to SGA, 27.6% (n=31) of residents were malnourished and 13.4% were rated as malnourished by MNA. MST had the best sensitivity and specificity compared with the SGA (sensitivity 88.6%, specificity 93.5%, κ=0.806), followed by MNA-SF (85.7%, 62%, κ=0.377), MUST (68.6%, 96.7%, κ=0.703), and Simplified Nutritional Assessment Questionnaire (45.7%, 77.2%, κ=0.225). Compared with MNA, MNA-SF had the highest sensitivity of 100%, but specificity was 56.4% (κ=0.257). MST compared with MNA had a sensitivity of 94.1%, specificity 80.9% (κ=0.501). The anthropometric screens ranged from κ=0.193 to 0.468 when compared with SGA and MNA. CONCLUSIONS: MST, MUST, MNA-SF, and the anthropometric screens corrected arm muscle area and calf circumference have acceptable concurrent validity compared with validated nutrition assessment tools and can be used to triage nutrition care in the long-term-care setting.