Smart surveillance: The future of infection prevention in hospitals 

By Dr Lyn-li Lim, Infectious Disease Physician at VICNISS Coordinating Centre, Doherty Institute 

In Australia, around 1 in ten adult patients admitted to hospital develop healthcare-associated infections, often linked to procedures or healthcare devices such as intravenous (IV) cannulas. Many of these infections are preventable with optimal care and the right systems in place. By analysing patient care data within electronic medical record (EMR) systems, hospitals can detect these infections early and identify areas where clinical processes can be improved to prevent them in the future.

The VICNISS team at the Doherty Institute, in collaboration with the University of Melbourne’s Centre for Digital Transformation of Health, examined if and how Victorian hospitals harness EMR data to automate the identification of healthcare-associated Staphylococcus aureus (‘Golden Staph’) bloodstream infections. These infections can have serious long-term health impacts, including death.

In our study, published in BMJ Health & Care Informatics, we surveyed 12 hospitals across Victoria. We found that while many hospitals had EMR in place, none had automated systems to detect healthcare-associated ‘Golden Staph’ bloodstream infections using EMR.

Instead, most facilities relied on Infection Prevention and Control-trained nurses to manually review the patient’s EMR record to determine if infections were healthcare-associated.

Our review showed that EMR systems already contained most of the patient data required to automate the detection of healthcare-associated infections. However, we found great variation in the systems that hospitals had in place and how they implemented them, resulting in significant differences in how data was collected. In some instances, clinical information was also missing or documented in a different place to where it should have been in EMR. These factors limit the ability to standardise data extraction and reporting across hospitals.

EMR holds great promise for hospitals to automate the identification of ‘Golden Staph’ bloodstream infections and to determine whether they are associated with healthcare interventions and potentially preventable.

Several countries overseas have successfully harnessed EMR to monitor hospital safety and quality. This is also within reach in Australia.

Offering Australian hospitals an implementation guide that includes an externally validated and standardised surveillance approach with standardised EMR data elements can reduce the work required at a hospital level to introduce automated identification of healthcare-associated infections.

Our study represents an important first step, demonstrating the feasibility of automating healthcare-associated infection surveillance in Victorian hospitals. With further development, this approach could improve efficiency and reduce human resources required for infection surveillance.

Peer review: Lim LL, et al. Feasibility of semiautomated surveillance of healthcare-associated Staphylococcus aureus bloodstream infections using hospital electronic health records in Victoria, Australia. BMJ & Health Care Informatics (2025). DOI: http://www.doi.org/10.1136/bmjhci-2024-101427

Collaboration: This work, led by VICNISS at the Doherty Institute, is the result of a collaboration with the University of Melbourne’s Centre for Digital Transformation and The Royal Melbourne Hospital.

Funding: This work was supported by a grant from the Innovation Acceleration program from the University of Melbourne and The Royal Melbourne Hospital.