Development of a new tool to monitor and identify inadequate oral intake in hospital.

Background: There is a need for quick and easy methods to monitor nutritional intake in hospital and identify patients with poor intake. Food record charts are often used in clinical practice, with low levels of accuracy and completion. This study aims to describe the development and evaluate the performance of a new tool to estimate energy and protein intake and identify poor nutritional intake amongst adult hospital patients.

Methods: Ninety trays were sampled and assessed independently using the new tool 'Meal Intake Points' and a weighed (reference) method. The performance was tested by measuring association (Spearman's correlation), agreement (proportion of meals within specified limits of reference method), and sensitivity and specificity to identify poor energy and protein intake.

Results: This new tool achieved very strong association for energy estimates (r = .91) and strong association for protein estimates (r = .86). Estimates for energy and protein were within 450 kJ and 4.5 g of the reference method in 77.8% and 62.2% of meals, respectively. It also displayed excellent performance as a screening tool (sensitivity 100%; specificity 76%-80%). Minor revision of the original tool was needed to optimise performance.

Conclusions: Meal Intake Points accurately estimates energy and protein intake and identifies patients with poor nutritional intake, providing a clinically relevant tool for use in hospitals to monitor intake and identify patients for proactive nutrition support. Further validation studies are needed to determine its performance in clinical practice and whether it is useful in predicting hospital-acquired malnutrition.