Background--Aging is associated with impaired physiological functions such as reduced GI functions, decreased metabolic rate and diminished physical activity. These changes can lead to malnutrition and increase risk of disease and mortality. Proper nutrition is important in delaying aging and reducing these risks. Nutritional screening/assessment is the key to early detection of malnutrition. Thus, a fast, effective and low-cost tool is essential for routine screening of the risk of malnutrition.
Objective--The objective of this study was to compare the malnutrition-screening ability of the modified Taiwan versions of Mini Nutritional Assessment short-form (MNA-SF) with Malnutrition Universal Screening Test (MUST), and to determine the predictors of malnutrition in elderly Taiwanese.
Methods--The study employed a purposive design and recruited 200 (88 men and 112 women) ?65y elderly who attended regional hospital for annual health examination. After signing an informed consent, each subject was interviewed with a structured questionnaire for eliciting personal data, lifestyle and health statuses and answers to questions in the MNA. Biochemical data were taken from subjects' medical records. Each subject was rated for the risk of malnutrition with the MNA and MUST, each in three
versions, the original, Taiwan version-1 which adopted population-specific anthropometric cut-points, and version-2 which replaced CC for BMI in the scale. Statistical analyses were carried out with SPSS/Window 10.0 Statistical Software Package (Chicago, IL). Friedman Test and Wilcoxon Signed-Rand Test were used to determine the significance of differences among the versions. Spearman's correlation analysis was used to determine the significance of relationship between the MNA or MUST scores with each of the health- related variables. Multivariate linear regression analysis was conducted to determine the factors that may impact the nutritional risk status of the elderly.
Results--The original, T1 and T2 versions of short-form MNA rated 18.0, 13.0, and 11.0% of participants at risk of malnutrition. The original MUST version rated 5.5% at high risk and 7.5% at moderate risk; MUST-T1 rated 2.5% and 6.0%; and MUST-T2 rated 4.0% and 4.5%, respectively. The modified versions of the MNA- SF and MUST have improved Kappa values and the specificity over the original versions. But the MNA-SF versions have better malnutrition-predictive ability than MUST versions. Regression analyses indicate that age, serum albumin and appetite status are the major factors impacting one's nutritional risk status.
Conclusion--Results of this study suggest that MNA-SF versions have superior nutritional-risk predictive ability over MUST versions. But MUST versions appear to be better suited for predicting the nutritional risk of clinical patients. The replacement of CC for BMI in T2 versions maintains the predictive ability of the tool. Since measurement of body weight and height can be difficult in frail elderly, the T2 versions represent an improvement in nutrition assessment. Without requiring the measurement of weight and height, the tool can be used more easily and can help to make routine nutritional screening a realistic goal.
This study also finds age, serum albumin and appetite status the major predictors of malnutrition in Taiwanese elderly.
