Oropharyngeal dysphagia is underdiagnosed and current screening is costly. We aimed: (a) to develop an expert system (ES) based on machine learning that calculated the risk of OD from the electronic health records of all hospitalized older patients during admission, and (b) to implement the ES in a general hospital. In an observational, retrospective study, electronic health records (EHR) and swallowing assessment using the volume-viscosity swallow test for oropharyngeal dysphagia were captured over 24 months in patients > 70yr admitted to Mataró Hospital. We studied the predictive power for dysphagia of 25,000 variables.
ES was obtained using feature selection, the final prediction model was built with non-linear methods (Random Forest). The psychometrics of the expert system built with a non-linear model were: Area under the ROC Curve of 0.840; sensitivity 0.940; specificity, 0.416. The expert system screens all patients admitted to a 419-bed hospital in 6 seconds, identifies patients at greater risk of OD, and shows the risk for OD in the clinician’s workstation.
AIMS-OD provides accurate, systematic and universal screening for oropharyngeal dysphagia in real-time during hospital admission of older patients, allowing the most appropriate diagnostic and therapeutic strategies to be selected for each patient.
