Reducing unnecessary pathology testing in the emergency department – a controlled interrupted time series analysis — The Association Specialists
  1. Schedule
  2. Oral Presentations - Climate and healthcare access and use
  3. Reducing unnecessary pathology testing in the emergency department – a controlled interrupted time series analysis

Reducing unnecessary pathology testing in the emergency department – a controlled interrupted time series analysis (23968)

Erin Mathieu 1 , Tim Schlub 1 , Katy Bell 1 , Scott McAlister 2 , Brian O'Connell 3 , Craig Scowen 4 , Luise Kazda 1
  1. The University of Sydney, Camperdown, NSW, Australia
  2. The Healthcare Carbon Lab, Department of Critical Care, The University of Melbourne, Melbourne, Victoria, Australia
  3. Department of Emergency Medicine and Sustainable Healthcare Unit, Coffs Harbour Health Campus, Mid North Coast Local Health District, NSW Health, Coffs Harbour, NSW, Australia
  4. NSW Pathology, North Sydney, NSW, Australia

Objective

To evaluate whether the Rational Investigations Program (RIP), a multi‑component quality‑improvement intervention, reduces unnecessary pathology test utilisation per emergency department (ED) presentation and to quantify associated financial and environmental impacts while monitoring patient safety.

Methods

We conducted a controlled interrupted time series across two regional EDs in New South Wales, Australia (intervention: Coffs Harbour; control: Lismore) from July 2022 to June 2024. During the intervention period (July 2023 to June 2024) RIP targeted 17 high‑volume tests using clinician engagement and education, embedded guidelines, EMR order‑set optimisation, audit and feedback. All ED presentations were included. Pathology orders from NSW Health Pathology were linked to de‑identified ED data. Environmental impacts were estimated using published carbon footprint values. The primary outcome was monthly tests per presentation; secondary outcomes were cost, carbon footprint and adverse events. Poisson models estimated immediate and trend effects.

Results

Across the 24‑month period, the intervention site ED and the control site ED recorded 92,358 and 80,297 ED visits, respectively, with stable patient volumes and comparable demographics. Nearly half of patients at both sites received at least one pathology test. There was no evidence of a pre-intervention trend. RIP was associated with an immediate reduction in testing at the intervention site (RR 0.94, 95% CI 0.93–0.95) and a substantial slowing of testing growth compared to the control site (interaction RR 0.75, 95% CI 0.73–0.77). Over 12 months, an estimated 20,289 tests were avoided at the intervention site, yielding $419,703 savings and a reduction of 3.11 tonnes CO2e. No increase in adverse events was observed.

Conclusions

The RIP reduced unnecessary ED pathology testing, delivering significant financial and environmental benefits without compromising patient safety. This clinician-led, structured approach offers a scalable strategy for reducing low-value care.