Australia's health data push aims to shift care from 'break and fix' to 'predict and prevent'

Every time a patient walks into a hospital, a GP clinic, or any other part of the Australian health system, a record is created. Most of it sits in storage, opened only by the few clinicians involved in that person's care. That is how it should work, and it is the foundation of the system. But an opinion piece published by InSight+ argues the same data has a second job that is going largely undone: helping the system learn what works and get better.

The author's term for it is real-world data, or RWD. In many countries, deidentified RWD is already used by researchers to look for medical explanations, spot health trends, identify more effective treatments and improve recovery odds. The author also suggests it could help prevent some diseases before they appear. It differs from clinical trial data because it is gathered every day from millions of people of all backgrounds, not from a selected study group.

The piece points to two Australian examples. RWD helped show how COVID-19 affected ambulance ramping times at Queensland hospitals. It has also revealed trends in emergency department visits across the country.

The infrastructure to do more is now being built. Australia has committed $9.5 million over three years to set up the Australian Health Data Evidence Network, or AHDEN. Adelaide University leads it, in partnership with the Australian Research Data Commons, and the University of Queensland is an important participant. The goal is to harmonise RWD from hospital electronic medical records nationwide while keeping patient information private and secure. AHDEN is a co-investment with the ARDC through the People Research Data Commons, and the ARDC is enabled by the Australian Government's National Collaborative Research Infrastructure Strategy. The author says the team is currently developing the infrastructure and the means by which researchers will be able to access and learn from the information ethically.

Looking ahead, the author describes three horizons. The first is a single place where each person's health data lives, so a GP can see the full history, every scan and every blood test, without chasing specialists or hospital records. The second is analysis at the population level: how wait times are tracking, and how patient outcomes look at a particular clinic. The third is new models of care built on digital data. With information from thousands of patients and AI, the author says, it may be possible to predict when someone in hospital is likely to have a low blood sugar reading and prevent it, or to predict which chemotherapy would most accurately target a person's cancer. The author says work is currently focused on the first horizon, and that the new network could help move toward all three.

AI is central to the argument. The author calls it unavoidable and says healthcare should harness it. Current care models, in this telling, still reflect a 1950s mindset built around conditions like pneumonia and broken legs, with people seeing a doctor when something is wrong. Today many patients live with four or five chronic illnesses at once, and the author says older models are plainly not working. With shortages of doctors, allied health workers and nurses, the author considers some adoption of AI to manage the load inevitable, including managing the data burden attached to each patient.

The conclusion is that better data access and AI could, over the longer term, move Australia from a break-and-fix model to one of predicting and preventing illness. This is an opinion piece, and those predictions are aspirations, not results. What to watch is whether AHDEN's ethical access arrangements are actually built, and whether predictive care moves beyond the proposal stage.