View the Tech Talk on YouTube.
Angus Taylor:
Hi, my name is Angus Taylor and welcome to today’s Tech Talk. I’m absolutely delighted to be here today with Professor Fiona Stanley to talk about how we can better use data to solve some of our biggest policy issues. Now Fiona, you’ve been involved in a fascinating project in Western Australia called the Development Pathways Project. Can you tell us a bit about the project and the sort of benefits it is delivering and it can deliver into the future as well?
Fiona Stanley:
Yes, I’m delighted to. This is a whole of government collaboration between groups of researchers who are passionate about finding out about how children develop overtime and the government agencies who’s data we have. So it’s not just health. It includes health, education, disability, child protection, public housing, police, justice and corrective services. We can actually map how children from before birth are going to end up in corrective services. We can show how important it is to prevent things very early and it shows how joined up many of these big problems are as children move through their lives. The value of this is it is total population data. Now when you do surveys and studies, what happens is you get a biased response rate. This is everybody. So you can say what the cost effectiveness is of interventions and services, who gets it, who doesn’t, what are the risks and characteristics of the people who do well under such services and who doesn’t do well.
Angus Taylor:
So in practice that means we can test whether things governments are doing actually work.
Fiona Stanley:
Absolutely, and you can cost that out. And people like the police have been very keen for us to map hotspots of potential problems in crime and substance use so that they can map their services against them. So it’s got a predictive value as well. We’re interested as researchers because we’re interested in why children don’t develop well and what are their risk and preventative characteristics. There’s a wonderful example of the best outcome for education that’s not a lot to do with schools. It’s to do with where you live, how old your mum was, whether she was married, what her social class was, and what capacity she had to nurture you in the early years.
Angus Taylor:
And we can see then the impact of that beyond just their school life? Have we seen the results?
Fiona Stanley:
Absolutely. Well these children in our Developmental Pathways Project, it’s a total birth cohort, are now thirty five. We are even able to look at their own parenting now who have intergenerational data. So, this has been a research project but the wonderful thing that has happened is, this is now 11 years old, initially the bureaucrats and policy makers were quite scared about data. They didn’t have the analytical capacity; they were very excited about the possibility of joining up these datasets. What’s happened is that they realized the value of the data, they’ve actually started to get more competent and to think about why joined up data is so valuable. The thing that’s driven it is both the passionate users and the consumers. They are extremely keen that we use this information to improve the service delivery, not just in health, but in other services. And that’s really turned it around in Western Australia.
Angus Taylor:
So if you look more generally across the health research community, how important is getting access to data for that research community?
Fiona Stanley:
It’s crucial. If you do any surveys or studies where you can’t get total population data, you’ll get the wrong results. One day coffee is good for you, the next day coffee is bad for you. Unless you get the whole population, the people who aren’t in randomized control trials are people who, often, are the ones you really want to study. The smokers, the substance abusers, the comorbidities and so on. And so to get that total population data is absolutely key to get accuracy in what you’re doing in that system.
Angus Taylor:
And so what sort of benefits can we see, have we seen or can expect to see in the future from getting access to this data for researchers?
Fiona Stanley:
There’s been a huge number and I think we were able to confirm airplane travel and vascular diseases in the legs that led to implementation of the prevention project.
Angus Taylor:
So that came out of work that has been done on population data. Wow. The DVT issues.
Fiona Stanley:
Absolutely. And so we’ve had a large number of projects looking at things like evaluating public housing for people with mental illness and whether that’s effective or not. We’ve been able to look at the contribution of ADHD kids to incarceration and looked at the ways to improve that. We’ve been able to look at the fetal alcohol syndrome and how that’s contributing, not just in aboriginal children, to children in detention and so on. So there’s a huge number of ways we can use these data. But the most exciting one that we haven’t been able to do is the linkage of the pharmaceutical data to all of the health data.
Angus Taylor:
And so what would that do for us if we could get that linkage?
Fiona Stanley:
We would be able to say like that [finger click] whether a drug had an adverse effect, we’d be able to look at whether doctors were prescribing accurately, we’d be able to look at harmful effects of all interventions, not just drugs, and that would mean a huge cost saving for the nation.
Angus Taylor:
And also better outcomes presumably.
Fiona Stanley:
Better outcomes for people. You wouldn’t be killing them. We wouldn’t be making them sick. We’d actually have a much greater ability to look at cost effectiveness of that whole system and it’s really anguishing to me that we don’t have that system in place.
Angus Taylor:
So what’s holding us back? What do we need to do to get more access to this data to solve the sorts of problems you’ve described?
Fiona Stanley:
Well this is what is so exciting about what your government has done in terms of opening these datasets. We were unable to get commonwealth data. We could get the state data and we were hungry to link not just the pharmaceutical data but the Medicare data. To have that for the whole nation means that you would have everybody monitored overtime; trends in outcomes; whether services were effective; who was not responding to therapy; who was getting in appropriate investigations, the cost effectiveness would be enormous. If you could scale that up to the nation, it would be a most innovative use of the data you’ve been collecting for a long time and not used appropriately.
Angus Taylor:
So with access to that data, how would you expect the health sector and health services to change over the next five or ten years?
Fiona Stanley:
There’s some radical changes that would happen. What would happen is that there would be a real evidence base on which you would make your decisions. And you could almost imagine there’d be some ways in which you could control that to ensure that people were doing evidence-based decision making. Everyone’s talking about evidence-based medicine – but very few people are actually doing it. So that’s one thing that could happen.
Once we get electronic health records, not only will patients care improve, but we as epidemiologist and public health researchers will be able to give you the very best evidence on what you can do to prevent these problems because that’s where you’re going to get the biggest bang for your buck. And at the moment we’re stopped from doing that.
Angus Taylor:
Better targeted spending and that spending getting better outcomes.
Fiona Stanley:
Absolutely, and getting people in the health sector, and in the other areas of child protection, education and so on, to also have an evidence base to know what they’re doing is not harming people and to know what they’re doing is actually being good for the community. And I think that’s the most exciting way that you could use these data for the future.
Angus Taylor:
Well that’s fantastic. Thank you very much for your time today Professor Stanley.
Fiona Stanley:
Thank you Angus.