Imma Perfetto
Cosmos science journalist
A leading digital health expert says that current models of healthcare – driven as they are by economic and environmental pressures – are no longer sustainable.
The Director of the Queensland Digital Health Centre (QDHeC), a leading digital health research unit based at the University of Queensland, Clair Sullivan is working to create what she calls: “…a digitally enabled learning health system”.
“At the moment, we deliver healthcare … But what I would love to see is that every time we deliver healthcare, we monitor what we do and the outcomes, aggregate that data and learn how to do better next time,” Sullivan told Cosmos.
“That happens in many other industries, but healthcare is a little behind.”
The Australian Department of Health and Aged Care recognises the way in which Australia’s health systems collect, share, and analyse data must be transformed.
Sullivan was speaking ahead of her panel at the annual Australian Academy of Health and Medical Sciences (AAHMS) Annual Meeting to be held in Adelaide next week.
The theme of the conference is “Shaping tomorrow’s health: embracing innovation and transformation.”
A learning healthcare system is defined as one that monitors, aggregates, and reviews its data to iteratively improve outcomes across various aspects of the system – from clinical outcomes to providers’ and consumers’ experiences, and costs.
The problem is that Australia’s health data is siloed across many agencies and jurisdictions, limiting its effective use.
Sullivan and researchers at QDHeC are working across 3 areas: digitising traditional workflows; creating real-time analytics; and developing new care models.
“At the moment, our data is very much geographically based, so it’s stored where the care is delivered,” says Sullivan.
“I am an advocate for a consumer centred record. So, wherever the consumer is, the data should follow her throughout her care journey, whether it’s GP, hospital or residential aged care facility.”
There is a large body of work around figuring out exactly how to take those traditional, paper-based workflows and create digital ones, and make that data useable.
“Of course, we need the infrastructure to be able to aggregate that data, create real-time analytics and learn from it, and then, just as importantly, change our behaviour.”
Large datasets can be used to train artificial intelligence (AI) models, which the Health x Digital Transformation Report 2024-2025 identifies as 2024’s most hyped technology trend, to identify patterns and apply them to new cases.
“An example might be that we could predict who in hospital is going to have a low blood sugar and go and see them and adjust the insulin before it happens,” says Sullivan.
But realising these new care models requires the right framework and infrastructure to be established around them first.
“Many people don’t really have, I don’t think, a good understanding of how hard all of those first steps are and how expensive they are,” says Sullivan.
“But it’s important to do it correctly and make sure that you have your ethics, and your governance, and your privacy correct all along the way.
“Being entrusted with medical information is on honour and a privilege, but also, I think that our consumers expect us to monitor the outcomes and improve.
“It’s just making sure that we abide by all the cybersecurity and privacy regulations. But I don’t see them as a barrier. I see them as an enabler, really.”
Sullivan will participate in a panel session on the critical role of secure and integrated medical data sharing and university-industry-government partnerships in driving research excellence and healthcare innovation.
Cosmos is a media partner of the AAHMS Annual Meeting. You can read another preview article on responsible AI integration here.
