Hospital malnutrition is becoming a clinical concern. Our aim was to determine the prevalence of hospital malnutrition through Nutritional Risk Screening 2002 (NRS) and to evaluate nutritional risk through a prospective study. Nutritional status was assessed collecting anthropometric parameters together with the data relating to the diseases in the medical records of patients admitted to the Department of Emergency Medicine of the "Sant'Eugenio" Hospital. One hundred and sixty patients were retrospectively enrolled during a 3-month observational period. The risk of malnutrition was detected in 52% of patients (of whom 38% at risk and 62% at serious risk). The NRS score was positively correlated with patient age, days between hospital admission and nutritional assessment, disease severity, length of hospital stay and catabolism (p less than 0.05); Basal Energy Expenditure (BEE) and mean arm circumference (MUAC) were negatively correlated with positive outcome (p less than 0.05). No correlations were found in the NRS score, gender, height, weight, Body Mass Index (BMI) and Total Energetic Expenditure (TEE) (p=n.s). A high prevalence of the risk of malnutrition may be detected in the emergency medicine setting, particularly in the geriatric population. The NRS score is not strictly related to BMI, but rather is an excellent tool for disease prognosis, as well as nutritional screening.