Move over, Braden? Mathematical models built on readily available electronic health record data predicted the development of pressure wounds better than the ubiquitous Braden scale tool in a large study in Korea.
Researchers pulled data for about 15,000 patients at a teaching hospital between January 2015 and May 2016 and used the information to develop three distinctive predictive models. The logistic regression, Cox proportional hazards regression and decision tree modeling were then compared against Braden scale findings to determine which most accurately predicted the development of 202 pressure wounds that occurred among the real patients.
The logistic regression model was most efficient, followed by the decision tree model, Cox proportional hazards regression model and the Braden Scale, according to study results published by the Korean Society of Nursing Science.
Decreased mobility was the most significant factor in the logistic regression and Cox proportional hazards models, and the endotracheal tube was the most important factor in the decision tree model.
“The models developed in this study can be used to develop a clinical decision support system that automatically assesses risk for pressure ulcers to aid nurses,” wrote a group of researchers from the College of Nursing at Seoul National University.