Objective: The British nutrition screening tool is a questionnaire designed to assess the nutrition status of hospitalized patients by nurses. The aim of this study was to evaluate the validity and reliability of this questionnaire in patients on admission to the hospital in Tehran.
Methods: For 6 mo, 446 patients aged over 18 y admitted to different wards of a general hospital were studied within the first 24 h of admission. A nutritionist undertook nutritional assessment in all patients to determine their nutritional status as an objective standard. Then a nurse completed the screening tool for patients. Results obtained using the screening tool were compared with those of the nutritional assessment to calculate the sensitivity, specificity, and predictive values. The interrater reliability of the tool was assessed by two nurses who completed the screening tool separately during the first 24 h following admission of each patient. It was also completed by a nurse within 2-d period of admission to test the intrarater reliability.
Results: Study participants included 229 women and 217 men with mean age of 49.5 ± 16.0 y. Sensitivity and specificity of the questionnaire were 86.7% (95% CI: 83.9-90.3%) and 61.7% (95% CI: 57.5-66.5%), respectively. Positive and negative predictive values were 79.1% (95% CI: 68.9-77.1%) and 73.1% (95% CI: 75.1-82.9), respectively. The sensitivity of the tool was over 80% for both genders, for all body mass index grades, and for data obtained from gastroenterology, transplant, oncology, and hematology wards. The interrater reliability of screening tool was interpreted as substantial, being k = 0.68 and k = 0.74 on both the first and second days, respectively. The intrarater reliability of the screening tool was also interpreted as substantial, being k = 0.77.
Conclusion: The nutrition screening tool is a simple, valid, and reliable tool that can be used by nurses to facilitate identification of patients requiring nutritional interventions.
Copyright © 2011 Elsevier Inc. All rights reserved.