RESUMO
PURPOSE: The rising proportion of elderly population in high-income societies has resulted in increasing number of subjects with chronic disabling diseases and nutritional deficiency. Elderly's nutritional status is usually assessed through the mini-nutritional assessment (MNA®). However, its effectiveness may be influenced by weight excess or obesity. We tested the performance of MNA® questionnaire in subjects aged ≥ 65 years from Northern Sardinia, Italy, according to overweight/obesity, and we tried to identify the factors associated with malnutrition. METHODS: A modified version of MNA® (mMNA) test, not including BMI, was compared with the conventional MNA® (cMNA) test, and the overall test performance was assessed by calculating sensitivity, specificity and accuracy. In addition, indexes of cognitive health, disability, comorbidity and polypharmacy were compared between patients with concordant and discordant MNA tests. RESULTS: cMNA® sensitivity, specificity and accuracy were 67%, 99% and 84% compared with the mMNA test, due to malnourished patients misclassified as normal because of excess weight. Predictors of malnutrition were: depression (p < 0.0001), disability (p < 0.0001) and polypharmacy (p < 0.0001). Interestingly, the average scores of the "global", "subjective" and "dietary" components of the cMNA®, were significantly lower compared with the corresponding scores of the mMNA. CONCLUSIONS: Excess of weight, a condition progressively rising in the elderly population, may reduce the performance of cMNA® test in detecting malnutrition. LEVEL OF EVIDENCE: Level III, caseâcontrol analytic study.
