Assessment of food intake in hospitalised patients: A 10-year comparative study of a prospective hospital survey

Background & aims: A food quality control and improvement permanent process was initiated in 1999. To evaluate the food service evolution, protein energy needs coverage were compared in 1999 and 2008 with the same structure survey in all hospitalized patients receiving 3 meals/day. Methods: Nutritional values of food provided, consumed and wasted over 24 h including non-exclusive nutritional support were calculated individually. Nutritional needs were estimated as 110% of Harris-Benedict formula for energy and 1.2 or 1.0 g protein/kg/day for patients <65 or >= 65 years old, respectively. Multivariate analysis identified factors associated with low nutritional intake in both populations standardized to body mass index (BMI) of 1999's patients. Results: Out of 1677 patients, 1291 were included. Mean BMI was higher in 2008 than 1999 (P < 0.001). The proportion of underfed patients was unchanged (69 vs. 70%, NS). The consumption of >= 1 oral nutritional supplements (ONS) daily increased the protein needs coverage from 80% to 115%(P < 0.001). The year 1999, high BMI, 1st week of hospital stay, specific diet. ONS absence and low meal quality were associated with low nutritional intakes. Conclusion: The nutritional needs coverage could have improved in 2008 if BMI was similar to 1999's. ONS consumption is associated with a lower risk of underfeeding in hospitalized patients. (C) 2010 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.