[Editorial] Promises and limitations of a digitalized infection control program

Abstract

Digitalization of health-related data from sources such as electronic health records (EHRs) and administrative and laboratory databases make them increasingly available. This availability is fostered by data-sharing agreements and technologies developing secured clouds. The improved communication between these databases facilitates exploration of documentation data for both descriptive and analytical analysis. Continuous monitoring of digitalized data aims to improve surveillance by offering a complete picture of infection control practices and situations at-risk, to alert healthcare workers when appropriate. Sophisticated algorithms currently build data-driven models to predict, classify, or cluster patients’ outcomes. These algorithms also called machine learning (ML), mainly aim to predict events, or target populations at-risk for specific care. Such applications of digitalized data hold promise for being translated into actionable results and integrated into the daily lives of hospital epidemiologists and infection preventionists. This article aims to highlight the pros and cons of digitalization and ML through several examples from the field of hospital epidemiology and infection control nursing.