Feasibility of using linked administrative data to support surveillance of bacterial nosocomial infections
Nosocomial infections are common, but the burden can be difficult to estimate and compare between settings.We provide an example of using linked administrative inpatient and microbiological data to improve the epidemiological understanding of nosocomial infections in Australia's Northern Territory (2012-2017).We developed criteria based on common definitions to identify bacterial nosocomial urinary tract infections (UTI), skin & soft tissue infections (SSTI), respiratory infections (RTI) and bloodstream infections (BSI).In the base analysis, an infection was defined as the presence of an infection-related International Classification of Diseases (ICD-10-AM) code and a bacterium isolated from an anatomically relevant specimen >2 days after admission.There were 72,583 patient episodes included in our base analysis, of which 1,544 (2.1%) had at least one nosocomial infection.The incidence rate of nosocomial UTIs, SSTIs, RTIs and BSIs was 7.4 [6.9-7.9],13.7 [13.0-14.4],5.8 [5.3-6.2] and 0.5 [0.4-0.6] per 10,000 patient-days.While sensitivity analyses highlighted the need to refine the assumptions used in defining infections as well as the need for formal validation and integration of additional clinical information, linking https://doi.
Authors
- Jodie McVernon (ORCID: https://orcid.org/0000-0001-9774-1961)
- Hannah Elizabeth Carter (ORCID: https://orcid.org/0000-0002-0046-4126)
- Will Cuningham (ORCID: https://orcid.org/0000-0001-5404-6088)
- Teresa Maria Wozniak (ORCID: https://orcid.org/0000-0003-3182-8348)
- Steven Y. C. Tong
Institutions
- Royal College of Surgeons of England (GB)
- Queensland University of Technology (AU)
- The University of Melbourne (AU)
- Charles Darwin University (AU)
- Menzies School of Health Research (AU)
- Australian e-Health Research Centre (AU)
Publication Details
- Journal
- Epidemiology and Infection
- Published
- 2026-11-19
- DOI
- https://doi.org/10.1017/s0950268826102428
- Primary Topic
- Infection Control in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00