Electronic-Medical-Record-Driven Machine Learning Predictive Model for Hospital-Acquired Pressure Injuries: Development and External Validation

Abstract

Background: Hospital-acquired pressure injuries (HAPIs) affect approximately 2.5 million patients annually in the United States, leading to increased morbidity and healthcare costs. Current rule-based screening tools, such as the Braden Scale, lack sensitivity, highlighting the need for improved risk prediction methods. Methods: We developed and externally validated a machine learning model to predict HAPI risk using longitudinal electronic medical record (EMR) data. This study included adult inpatients (2018-2023) across five hospitals within a large health system. An automated pipeline was built for EMR data curation, labeling, and integration. The model employed XGBoost with recursive feature elimination to identify 35 optimal clinical variables and utilized time-series analysis for dynamic risk prediction. Results: Internal validation and multi-center external validation on 5510 hospitalizations demonstrated AUROC values of 0.83-0.85. The model outperformed the Braden Scale in sensitivity and F1-score and showed superior performance compared to previous predictive models. Conclusions: This is the first externally validated, cross-institutional HAPI prediction model using longitudinal EMR data and automated pipelines. The model demonstrates strong generalizability, scalability, and real-time applicability, offering a novel bioengineering approach to improve HAPI prevention, patient care, and clinical operations.

Keywords: automated EMR integration; clinical decision support; electronic medical records; external validation; hospital-acquired pressure injury; machine learning; multi-center validation; predictive model; pressure ulcer; wound care management.

Figure 1

Funnel graph showing the number of similar published studies [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16] by criteria of review.

Figure 2

Patient flow and inclusion/exclusion criteria in the development cohort.

Figure 3

Sampling strategy for observational variables.

Figure 4

Data flow for the HAPI labeling logic.

Figure 5

Variable importance of the final XGBoost model by descending information gain. Refer to Appendix A Table A2 for the definitions of the variables.

Figure 6

Receiver operating characteristic curves of the XGBoost model on the test set, the internal validation set, and all the external validation set, and their respective areas under the curve and 95% CIs.

Figure 7

Comparison graphs of the respective receiver operating characteristic curves of the XGBoost model (solid lines) and the Braden Scale (dashed lines) on the internal validation set (a) and on the external validation sets (Facility B (b), Facility C (c), Facility D (d), Facility E (e)), and their respective areas under the curve and 95% CIs.