The effect of age and sex on the routinely used measurements to assess the nutritional status of hospitalized patients

Abstract

Nutritional assessment measurements were performed on groups of 40 healthy young and elderly males and females. Their values were compared to measurements in 65 young and 44 elderly subjects with protein-calorie malnutrition. Creatinine height index was a good predictor of reductions in lean body mass in malnourished young males but was of less value in young females. In healthy elderly creatinine height index was significantly lower than in the young and wide overlaps between well and malnourished elderly groups were found. Relating creatinine excretion to total arm length improved the ability to predict malnutrition in young females and in elderly males, provided an appropriate lower limit of normal was used. In both sexes and all age groups percentage ideal body weight and arm muscle circumference were of limited value in detecting malnutrition and triceps skin fold thickness was only useful in elderly females. Anemia was usually found in young subjects with malnutrition but only distinguished well from malnourished elderly if a lower limit of normal for Hb was used (12 g/dl for males; 10 g/dl for females). The total iron-binding capacity was a good indicator of malnutrition in males of all ages but was of less value in females in whom iron stores, rather than nutritional status, exerted a major influence on the measurement. The best predictor of malnutrition in any age group was the serum albumin. In both young and old healthy subjects a value below 4 g/dl was unusual and no overlap occurred between the well and malnourished groups. These observations demonstrate the relative usefulness of nutritional assessment measurement in males and females of different ages. They also indicate the need for more suitable standards for the elderly.