Background: As nutritional preventative measures are more effective in elderly people assessed as non protein-energy malnourished (non-PEM) than such interventions are in those who are malnourished, early detection of those at risk of malnutrition is important.
Objective: This study tests the hypothesis that health problems measured by the Nottingham Health Profile (NHP) can predict residents at risk of malnutrition.
Design: Nutritional status was assessed in 261 residents newly admitted to municipal care using a combination of anthropometry and serum protein measurements. From this sample, 20 non-PEM residents, simultaneously assessed as moderately malnourished according to a subjective method, were consecutively included. Using a paired matched design, 20 other non-PEM residents, who were simultaneously subjectively assessed as well nourished, completed the pairs.
Results: Univariate logistic regression analyses showed that the dimensions of emotional reactions, energy, pain, physical mobility and sleep had significant power to predict residents at risk of malnutrition. In the multiple logistic regression analysis, energy had the highest explanatory power.
Conclusion: Non-PEM residents, assessed as moderately malnourished according to the Subjective Global Assessment, perceived significantly greater health problems than non-PEM residents subjectively assessed as well nourished. Measurement of health problems adds important information to that used in early detection of residents at risk of malnutrition.