Integration of an electronic hand hygiene auditing system with electronic health records using machine learning to predict hospital-acquired infection in a healthcare setting.

Cotia ALF, Scorsato AP, Victor EDS, Prado M, Gagliardi G, Vieira JE, Generoso JR, de Menezes FG, Hsieh MK, Lopes GOV, Edmond MB, Perencevich EN, Goto M, Wey SB, Marra AR. Integration of an electronic hand hygiene auditing system with electronic health records using machine learning to predict hospital-acquired infection in a healthcare setting. American journal of infection control. 2024 Sep 21.

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

If you have VA-Intranet access, click here for more information vaww.hsrd.research.va.gov/dimensions/

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.    Search Dimensions for VA for this citation
* Don't have VA-internal network access or a VA email address? Try searching the free-to-the-public version of Dimensions

Search for Abstract from PubMed

---

Abstract:

BACKGROUND: Hospital-acquired infections (HAIs) increase morbidity, mortality, and healthcare costs. Effective hand hygiene (HH) is crucial for prevention, but achieving high compliance remains challenge. This study explores using machine learning to integrate an electronic HH auditing system with electronic health records to predict HAIs. METHODS: A retrospective cohort study was conducted at a Brazilian hospital during 2017-2020. HH compliance was recorded electronically, and patient data were collected from electronic health records. The primary outcomes were HAIs per CDC/NHSN surveillance definitions. Machine learning algorithms, balanced with Random Over Sampling Examples (ROSE), were utilized for predictive modeling, including generalized linear models (GLM); generalized additive models for location, scale, and shape (GAMLSS); random forest; support vector machine; and extreme gradient boosting (XGboost). RESULTS: 125 of 6,253 patients (2%) developed HAIs and 920,489 HH opportunities (49.3% compliance) were analyzed. A direct correlation between HH compliance and HAIs was observed. The GLM algorithm with ROSE demonstrated superior performance, with 84.2% sensitivity, 82.9% specificity, and a 93% AUC. CONCLUSIONS: Integrating electronic HH auditing systems with electronic health records and using machine learning models can enhance infection control surveillance and predict patient outcomes. Further research is needed to validate these findings and integrate them into clinical practice.