Predictive inpatient falls risk model using Machine Learning

Abstract

Aim: To create a model that detects the population at risk of falls taking into account fall prevention variable and to know the effect on the model´s performance when not considering it. Background: Traditionally, instruments for detecting fall risk are based on risk factors, not mitigating factors. Machine learning (ML), which allows working with a wider range of variables, could improve patient risk identification. Methods: The sample was composed of adult patients admitted to the Internal Medicine service (total, n=22515; training, n=11134; validation, n=11381). A retrospective cohort design was used and we applied ML technics. Variables were extracted from electronic medical records (EMR). Results: The Two-Class Bayes Point Machine algorithm was selected. Model-A (with fall prevention variable) obtained better results than Model-B (without it) in sensitivity (0.74 vs 0.71), specificity (0.82 vs 0.74) and AUC (0.82 vs 0.78). Conclusions: Fall prevention was a key variable. The model that included it detected the risk of falls better than the model without it. Implications for Nursing Management: We created a decision-making support tool that helps nurses to identify patients at risk of falling. When it´s integrated in the EMR, it decreases nurses’ workloads by not having to collect information manually.