Dr Michael Lydeamore1,2, Dr Tjibbe Donker3, Professor Ben Cooper4, Ms Marion Easton5, Dr Nicolas Geard6, Dr Claire Gorrie7, Daneeta Hennessy5, Professor Benjamin Howden7, Professor Anton Peleg8, Ms Annabelle Turner7, Professor Andrew Wilson2, Associate Professor Andrew Stewardson8
1Department of Econometrics and Business Statistics, Monash University, Melbourne, Australia
2SaferCare Victoria, Government of Victoria, Melbourne, Australia
3Institute for Infection Prevention and Hospital Epidemiology, University of Freiburg, , Germany
4Centre for Tropical Medicine & Global Health, Nuffield Department of Medicine, University of Oxford, , United Kingdom
5Health Protection Branch, Department of Health, Government of Victoria, Melbourne, Australia
6School of Computing and Information Systems, The University of Melbourne, Parkville, Australia
7Microbiological Diagnostic Unit Public Health Laboratory, Department of Microbiology & Immunology at the Peter Doherty Institute for Infection & Immunity, University of Melbourne, Melbourne, Australia
8Department of Infectious Diseases, The Alfred Hospital and Monash University, Melbourne, Australia
Introduction: The Victorian Carbapenemase-producing Enterobacterales (CPE) guideline considers colonisation to be lifelong. In the absence of a statewide alert system, patients and their GPs need to inform hospitals of their CPE status so that transmission-based precautions can be applied. We describe the frequency with which CPE-colonised patients are readmitted to hospitals, evaluating the level of reliance on this manual alert method.
Methods: We used a linked patient-level dataset containing hospital admissions and CPE notifications in Victoria from 2011 until 2020. CPE-positive patients were matched 1:1 to 100 cohorts of CPE-negative patients on age, time and hospital. Time zero was defined as the separation date from the health service for the admission of CPE diagnosis, or matched admission for CPE-negative patients. Time to next admission to a different health service was investigated using survival analysis.
Results: We included 1084 CPE-positive patients; 480 (44%) were female, median age bracket was 65-70, and 23 (2.1%) were discharged to aged care. The median time until next admission was 215 days (95% CI 159-284) for CPE-positive and 445 days (335-567) for CPE-negative patients. After one year, 541 (50%) and 641 (47%) of CPE-positive and negative patients, respectively, had been admitted to another health service. After five years, 717 (66%) and 914 (67%) of CPE-positive and negative patients, respectively, had been admitted to another health service.
Conclusion: Among hospitalised patients, admission to another health service is more common for CPE-positive than CPE-negative patients, emphasising the importance of communication by patients and GPs to trigger transmission-based precautions.
Biography: Dr Lydeamore is an infectious diseases modeller and data analyst. His research focusses on understanding the dynamics of antimicrobial resistance and how changes in health systems can impact on the spread and control of hospital-acquired conditions.
He has previously worked on quantifying the burden of healthcare acquired infections, the dynamics of COVID-19 (for the Victorian state government), and the control of skin infections in northern Australia.