Effects of implementing universal dysphagia screening at admission in a university hospital

In this study, we evaluated a hospital-wide swallowing screening system implemented for all inpatients at a university hospital since April 2018. We compared the original two-step screening method-combining the Eating Assessment Tool-10 (EAT-10) and the water-swallowing test-with the one-step screening using EAT-10 alone, introduced in April 2020 following the coronavirus disease 2019 pandemic.

A retrospective analysis of medical records was conducted to assess screening implementation rates, diagnostic accuracy (sensitivity and specificity), and the number of choking incidents.

The two-step method showed 90.1% sensitivity and 79.6% specificity, with no choking events, whereas the one-step method showed 73.6% sensitivity and 91.1% specificity, with three choking events. Both strategies significantly reduced choking incidents compared with those during the pre-implementation period (annual mean: 4.2 events). Sensitivity was higher for the two-step screening, although no significant difference in the number of choking events was observed between the two approaches.

The combined two-step screening using EAT-10 and the water-swallowing test offers high sensitivity and is effective in identifying patients at risk for choking. Meanwhile, screening with EAT-10 alone also offers preventive benefits and represents a feasible option for facilities with limited manpower. Implementing hospital-wide screening at admission contributes significantly to safer inpatient care.