To develop the app, researchers used data on reported inpatient falls between January 2011 and August 2015 at Houston Methodist Hospital. The research team also collected the patients’ demographic information, which included sex, age and race as well as admission diagnoses and bone density measurements reported within one year, to develop a predictive model.
The app achieved accuracy levels significantly above those of previous models in predicting fall risk. Researchers also integrated the technology with the EHR, so the app can automatically flag or alert clinicians for high-risk fall patients when they are enrolled in the hospital.
Researchers concluded that there needs to be more research done on the app. The technology has the potential to provide an automatic severity index that will predict whether a fall in a patient will be severe, which can help hospitals draw appropriate interventions and more attention to fall-risk patients to prevent such accidents from occurring.
More articles on artificial intelligence:
Cigna to expand AI-enabled medication management solution
Cincinnati Children’s AI boosts radiologists’ ADHD detection
Physicians express doubts about Babylon Health’s symptom-checking AI