Automatic strategy to identify the risk of malnutrition in primary care by means of Controlling Nutritional (CONUT) Score: a large population study.

Abstract Background Risk of malnutrition can be evaluated through Controlling Nutritional (CONUT) Score based on total peripheral lymphocytes, serum albumin, and total cholesterol. Our aim was to automatically calculate CONUT in primary care when involved markers were requested and compare its performance with albumin to identify patients at risk of malnutrition. CONUT was evaluated according to patient age, and calculated the potential expenses to calculate CONUT, when serum markers were no requested, and measured. Methods We calculated CONUT when the three laboratory markers were requested, compared CONUT and albumin, considering risk of malnutrition CONUT ≥ 2, and albumin < 35 g/L, and counted the number of albumin and/or cholesterol tests that would have been measured to calculate CONUT in patients with total lymphocytes and serum availability, and calculated hypothetical cost. Results of the 74743 requests from primary care, CONUT was calculated in 9353 patients. 2256 (24.1%) showed risk of malnutrition according CONUT; 630 (6.8%) through albumin. Albumin alone identified 17.5% of mild cases based on CONUT, but every severe case. The risk of malnutrition increased with patient age. In one year, through an expense of 5430 € in measuring the serum non requested test, we could have screened for risk of malnutrition the entire primary care population that underwent laboratory testing. Conclusion CONUT score calculated at no cost when requested involved laboratory markers, detected more patients at risk of malnutrition than albumin, increased with patient age, and could be calculated in all primary care patients by measuring unrequested tests, at a very affordable cost.