Predicting Pressure Injury in Critical Care Patients: A Machine-Learning Model.

BACKGROUND: Hospital-acquired pressure injuries are a serious problem among critical care patients. Some can be prevented by using measures such as specialty beds, which are not feasible for every patient because of costs. However, decisions about which patient would benefit most from a specialty bed are difficult because results of existing tools to determine risk for pressure injury indicate that most critical care patients are at high risk. OBJECTIVE: To develop a model for predicting development of pressure injuries among surgical critical care patients. METHODS: Data from electronic health records were divided into training (67%) and testing (33%) data sets, and a model was developed by using a random forest algorithm via the R package "randomforest." RESULTS: Among a sample of 6376 patients, hospital-acquired pressure injuries of stage 1 or greater (outcome variable 1) developed in 516 patients (8.1%) and injuries of stage 2 or greater (outcome variable 2) developed in 257 (4.0%). Random forest models were developed to predict stage 1 and greater and stage 2 and greater injuries by using the testing set to evaluate classifier performance. The area under the receiver operating characteristic curve for both models was 0.79. CONCLUSION: This machine-learning approach differs from other available models because it does not require clinicians to input information into a tool (eg, the Braden Scale). Rather, it uses information readily available in electronic health records. Next steps include testing in an independent sample and then calibration to optimize specificity.

Midrange Braden Subscale Scores Are Associated With Increased Risk for Pressure Injury Development Among Critical Care Patients.

PURPOSE: The purpose of the current study was to examine the relationship between pressure injury development and the Braden Scale for Pressure Sore Risk subscale scores in a surgical intensive care unit (ICU) population and to ascertain whether the risk represented by the subscale scores is different between older and younger patients. DESIGN: Retrospective review of electronic medical records. SUBJECTS AND SETTING: The sample comprised patients admitted to the ICU at an academic medical center in the Western United States (Utah) and Level 1 trauma center between January 1, 2008 and May 1, 2013. Analysis is based on data from 6377 patients. METHODS: Retrospective chart review was used to determine Braden Scale total and subscale scores, age, and incidence of pressure injury development. We used survival analysis to determine the hazards of developing a pressure injury associated with each subscale of the Braden Scale, with the lowest-risk category as a reference. In addition, we used time-dependent Cox regression with natural cubic splines to model the interaction between age and Braden Scale scores and subscale scores in pressure injury risk. RESULTS: Of the 6377 ICU patients, 214 (4%) developed a pressure injury (stages 2-4, deep tissue injury, or unstageable) and 516 (8%) developed a hospital-acquired pressure injury of any stage. With the exception of the friction and shear subscales, regardless of age, individuals with scores in the intermediate-risk levels had the highest likelihood of developing pressure injury. CONCLUSION: The relationship between age, Braden Scale subscale scores, and pressure injury development varied among subscales. Maximal preventive efforts should be extended to include individuals with intermediate Braden Scale subscale scores, and age should be considered along with the subscale scores as a factor in care planning.