First steps in establishing surveillance of bloodstream infections from electronic health record derived data, EU/EEA countries, March 2023 to March 2025
Alexis Sentís, Tine Dalby, Tommi Κärki, Francisco Orchard, Angelo D'Ambrosio, Miriam Van den Nest, Bernhard Benka, Nathalie Shodu, Karl Mertens, Emanuel Brađašević, Antea Jezidžić, Markella Marcou, Linos Hadjihannas, Jan Kubele, Sophie Gubbels, Louise Stougaard, Liidia Dotsenko, Piret Mitt, Imke Wieters, Alexandra Hoffmann, Antonis Maragkos, Fortunato D'Ancona, Agnese Comelli, Auguste Salnaite, Andre Brincat, Michael A Borg, Manon Brekelmans, Stephanie van Rooden, Anders Skyrud Danielsen, Torunn Alberg, Irena Klavs, Daša Kavka, Lucía García-San Miguel, Pilar Gallego-Berciano, Hanna Billström, Olov Aspevall, Diamantis Plachouras, Hanne-Dorthe Emborg, Carlos Carvalho, Anthony Nardone,BACKGROUND
Enhancing surveillance of bloodstream infections (BSIs) by extracting data from electronic health records (EHR) is often the first step in automating the surveillance of healthcare-associated infections (HAIs).
AIM
We assessed existing BSI surveillance systems and the progress in implementing EHR-BSI surveillance within the ECDC project Surveillance from Electronic Health Data (SUREHD).
METHODS
We summarised information on BSI surveillance systems through meetings with national representatives and review of national surveillance protocols and reports.
RESULTS
Of the 30 EU/EEA countries invited to join the SUREHD project, 17 actively participated from March 2023 to March 2025. All 17 countries conducted BSI surveillance within HAI or antimicrobial resistance programmes, and EHR-BSI surveillance was being implemented nationally (n = 9), regionally (n = 2) or at hospital level (n = 6). Reported challenges included data standardisation (n = 13), IT capacity (n = 10), data linkage (n = 7), system integration (n = 7) and data protection (n = 6). Few countries had access to structured EHR data on patient symptoms (n = 1), infection origin (n = 2) and catheter-related variables (n = 5). Most countries had not yet implemented data validation processes (n = 12) nor decided on standardised vocabularies for most variables. All countries aimed to automate EHR-BSI surveillance.
CONCLUSION
Automation of BSI surveillance is critically relevant for the development of surveillance systems and to support infection and control measures (IPC) for HAIs. Implementation of automated EHR-based BSI surveillance varied between countries; a generic protocol and a network of experts have established a way forward. Future efforts should focus on harmonisation, data quality, and leveraging EHR-BSI data to strengthen HAI surveillance and to improve IPC.