Improving nutritional screening of hospitalized patients: the role of prealbumin

Abstract

BACKGROUND: Limited resources prevent hospitals from having all patients formally evaluated by a nutrition expert. Thus, hospitals rely on nutrition-screening tools to identify malnourished patients. The purpose of this study was to determine the effectiveness of a nutrition-screening protocol, prealbumin (PAB), retinol binding protein (RBP), and albumin (ALB) in identifying malnourished hospitalized patients. METHODS: A nutrition screening protocol was prospectively used in medical and surgical patients and consisted of a nurse administering a questionnaire to patients and requesting formal evaluation by a registered dietitian (RD) only if nutritional issues were identified. Patients also had ALB, PAB, and RBP drawn, which were used to both screen and identify the malnourished. PAB, RBP, and ALB were compared as predictors of RD classification of patient nutritional status. RESULTS: The nutrition-screening protocol classified 104 of 320 patients (33%) as malnourished. However, 43% of the patients were not deemed at nutritional risk according to this protocol and therefore did not receive RD assessment. PAB was a significant predictor of RD-determined nutritional status (p < .05), whereas RBP and ALB were not. PAB screening/assessment identified 50% (162/320) of the patients as being malnourished. Notably, 50% of the patients (71 of 142) who were not evaluated by an RD were identified as malnourished using PAB criteria. The nutrition-screening protocol took 1.2 days longer to determine malnourishment compared with PAB (p = .0021). CONCLUSIONS: Use of screening questionnaires may miss or delay identification of malnourished patients. PAB screening/assessment may improve identification of those patients requiring nutrition intervention and thus enhance the care of hospitalized individuals. Few studies have evaluated the relative effectiveness of nutrition screening tools. This study indicates that prealbumin (PAB) identifies more malnourished patients more quickly than a screening questionnaire and that PAB is a significant predictor of nutritional assessment by a registered dietitian, whereas retinol binding protein and albumin are not.