A national perioperative outcomes registry will facilitate quality assurance and research in Australia

Countries with a national approach to ethics and governance and to data collection have a distinct advantage in conducting large surgical outcomes studies. Knowing the Risk was a surgical outcomes study conducted as a quality improvement activity in the United Kingdom in 2011. ¹ The study provided population-level data on high-risk surgical patients and critical care utilisation, and the data were used to develop the Surgical Outcome Risk Tool (SORT), a generalisable perioperative mortality risk calculator for adults undergoing inpatient noncardiac surgery. ²

SORT was locally adapted for the New Zealand population using a large dataset from the New Zealand National Minimum Data Set. ³ SORT was found to underpredict 30-day mortality in New Zealand up to five-fold, and the model was locally updated and published as NZRISK. ³ The New Zealand experience confirms that risk-prediction models cannot be ‘exported’ to other countries without being adjusted for differences in local healthcare systems and demographics. ³

In 2019 we investigated the feasibility of conducting a national surgical outcomes study in Australia. Ideally, such a study would link identifiable hospital-level data to the national death index, to access the most accurate mortality data. In Australia, ethics and governance principles and privacy laws require studies using identifiable, sensitive healthcare information as well as those using information for a secondary purpose to be approved and overseen by a Human Research Ethics Committee (HREC). The process for obtaining national HREC approval in Australia is complex^4–6 and is further complicated by federal and state privacy legislation governing the collection, use and transmission of identifiable data.^4–6

Our feasibility study was coordinated by the Australian and New Zealand College of Anaesthetists (ANZCA) Clinical Trials Network at Monash University and was approved by the Alfred Health HREC (reference 47931) under the national mutual acceptance scheme, with a waiver of consent for the collection, storage and transmission of identifiable personal and health data. Separate ethics approval was obtained in the Northern Territory and Tasmania and at the St John of God Health Care private hospital network. Of the 22 hospitals recruited (19 public and three private), representing each state and territory of Australia, 18 gained site-specific approval under the national mutual acceptance scheme and three obtained site-specific approval under the separate ethics approvals. One private hospital was unable to gain site-specific approval. Site-specific approval required protocol variations at five of 21 hospitals, including additional privacy requirements at four hospitals and a requirement for full informed consent at one hospital. All patients (20/20) invited to participate at that hospital agreed to do so. Site-specific approval by local Research Governance Offices (RGOs) was the rate-limiting step, with half of all RGOs taking over 56 days to approve an observational pilot study gathering routinely collected data on 20 patients that already had HREC approval, the slowest taking 196 days and imposing no additional requirements.

Barriers to conducting a national surgical outcomes study in Australia include ethics and governance delays; privacy regulations relating to identifiable data; differing approaches to consent between the public and private sectors; ensuring that regional and rural hospitals are adequately represented; and ensuring the balance of public and private hospitals reflects the proportion of surgery performed in each sector.

An alternative source of national surgical outcomes data would be a national perioperative outcomes registry, modelled on international examples such as the Swedish Perioperative Outcomes Registry, the Multicentre Perioperative Outcomes Group or the National Surgery Quality Improvement Program.^7–9 The Australian Commission for Safety and Quality in Health Care has published a framework for the development of clinical quality registries in Australia, whereby initial ethics and governance approval allows ongoing collection of data to monitor the quality and safety of care provided by hospitals. ¹⁰ A national perioperative outcomes registry would facilitate risk prediction model development and updating, as well as population-based cohort studies, national audits, benchmarking of hospital performance, registry randomised controlled clinical trials and real-world monitoring of the translation of research into practice. These benefits should be attractive to regional, rural, remote and private hospitals, and their participation would ensure registry data accurately represent the Australian population.

We have subsequently established a pilot national perioperative outcomes registry, ANZCA Perioperative Clinical Outcomes Registry (PCORE): Pilot, with ethics approval under the national mutual acceptance scheme for three years (including a waiver of consent for the collection of identifiable data) and seed funding from an ANZCA Pilot Grant. We invite interested hospitals to contact the ANZCA Clinical Trials Network office to join the pilot registry.