Validation of a nutrition screening tool: testing the reliability and validity

The aim of this study was to validate a nutrition screening tool for use in South Manchester University Hospitals Trust.

A sample of 100 patients was selected from medical, surgical and elderly care wards. To test the reliability of the screening tool, nurses and dietitians completed the screening tool on the same patient. These results were compared for interobserver error to determine whether the screening tool was reproducible with different observers. To ascertain if the screening tool identified malnutrition at ward level, four markers commonly used to assess nutritional status were collected. These included body mass index (BMI), mid upper arm circumference MUAC, percentage weight loss, and energy intake calculated from the patient's first full day in hospital and expressed as a percentage of their estimated average requirements (EAR).

There was a 95% level of agreement between nurses and dietitians within +/-3. The screening tool had a sensitivity level of 78% and a specificity of 52% when compared to all patients who had one or more markers indicating malnutrition. This association was found to be statistically significant (P < 0.005).

The screening tool is reliable when completed by different observers and is valid for wide scale nutritional assessment. The screening tool identifies an acceptable number of patients who are malnourished but overestimates patients at moderate risk.