There’s a big flaw in his understanding of what should be his no. 1 health reform priority, but we have just the explainer he needs.
Asked this week about the well-overdue My Health Record app and the various problems with My Health Record overall, Mark Butler said we “need to do better on digital health”.
He then said that even though MHR was clunky and didn’t work very well, it was “really the mothership of our digital health system”.
Sorry? It still doesn’t work after $2 billion in spending, but he’s still all in on the idea that it is the “mothership” of our digital health strategy?
Over at our sister digital health publication, Wild Health, I wrote a piece about all this yesterday: Australia’s digital health problem as explained to a six-year-old.
I wanted to publish it here but my editor flat-out refused, mainly because it is nearly 8000 words, so I’ll need to give you the headline ideas here (but go and read it anyway when you can).
The key theme of the article is that digital health and our installed healthcare technology base in Australia might be the number one health reform issue in this country if we are serious about dealing properly with chronic care and, vitally, saving general practice by properly connecting it to the greater healthcare system.
As a part of understanding how far we are from being able to change the system, we all need to better understand the concept and technology behind the My Health Record and how it became redundant years ago as far as efficiently sharing data between providers and patients in an advanced and complex healthcare system.
To illustrate this point in my article, I explained how I had walked three football fields of technology exhibitors (over 1100 of them) at the world’s largest digital health conference in the US last year and could not find one which did not have some capability to share their data via a modern cloud-enabled application as a part of their technology.
I then explained that if I were walking the floor of equivalent exhibitions in Australia the reverse was true and you would find that well over 90% of the exhibitors did not have modern web-sharing capability, and many would be exhibiting on-premise server-bound technology that had its peak time somewhere in the 1990s.
Why such a vast difference in installed technology bases between our two countries (and a few others like Denmark) and why is no one worried about the difference?
How these countries work now is that they have made their software vendors and providers align over time to some sort of standard in modern web-based data-sharing technology so that they can at any time share any meaningful data they generate with any other provider and any patient (via their mobile).
The data generated at various points of a healthcare system where a patient is engaging is always going to be of the highest fidelity at that source, and with this technology you can largely leave it where it is and if you need it elsewhere in the system you can get it when you need it in real time.
Simple: no need for a “mothership”, no need to waste time collating and sending data to the middle or trying to get it out of the middle and interpret it.
This is what is known as a web-based distributed data model.
Leave most (not all) of the data where it is generated (so within hospitals, general practices, at pathology labs and so on) and when you need it for a patient encounter, in this system, you can get it, knowing that it is up to date and high fidelity because it’s coming from the source.
This is not new technology. It’s been used in other industry sectors like banking, travel and media for many years now.
Let’s compare this technology to our MHR mothership model.
A centralised data mothership for healthcare has these fairly obvious issues:
- An all-encompassing database of everyone in health with everything on it will never have everyone and everything on it, nor will it have everything in an up-to-date high-fidelity manner because someone has to always go to the trouble after the fact of sending their data (from a GP practice or pathology lab or a pharmacy) to the middle.
- It is not a matter of whether such a database will be hacked, it is only a matter of when, and what we will do when it is.
- Such a central database is not only a large and very expensive security problem (security necessarily keeps the whole system very slow and clunky in getting stuff in and out) it is also highly inefficient for data transfer and encourages low-fidelity data generation.
- The number one burnout issue for healthcare workers is lack of time. But we are asking every point of the system to stop and collate data and send it into the central health record, and we are expecting them to care about the process. They largely don’t (you will need to read my other article to understand why) and that is a very bad dynamic for the quality of data you are generating.
If the problems are so obvious, and with so many other countries now marching forward rapidly on an entirely different healthcare data sharing model, why do we have a federal Health Minister still seemingly supporting the “mothership” model, while notably damning it at the same time?
In my Wild Health article I postulate the following:
- Modern secure web-based distributed data sharing technology isn’t that easy to get your head around, and the old technology we subsist on in large parts of Australia (most particularly our general practice sector) still does talk to the web so it can at times become confusing to understand the difference.
- Without a burning need to understand the importance of the difference in the two technology bases (we still have a great healthcare system compared to most of the rest of the world) very few people in power in healthcare reform have got their heads around this issue.
- The My Health Record and its predecessor the PCEHR were OK ideas in their time, but now our governments have found that after spending more than $2bn it’s embarrassing to admit you might be wrong and stop. Dare I say it, but the continuing spruiking of it has overtones of the dynamics of political and public service dysfunction we are now seeing come out of the Robodebt inquiry.
- Like any mothership project a lot of egos and careers are tied to it being a success, no matter what the cost.
For years it has been politically expedient to tell Australians we are ahead of the game in digital health when in fact for years now we’ve been drifting to the back of the pack.
Transitioning the system to a chronic care one hangs on digital strategy, which hangs on data sharing – so it’s frustrating that our health reform experts can’t see how far we’re falling behind other countries.
At the moment every health minister from Victoria to NSW to Butler himself rants regularly about the need for integrated care teams to meet the future demands of our system.
It’s all talk.
Australia does not have the installed technology base it needs to facilitate that.
To manage chronic care in community care teams you need GPs to be able to talk seamlessly to allied care and to hospitals, and you need to generate data to measure outcomes so you can fund such care and manage it without wasting a lot of money.
At least 90% of our installed GP technology base is that 1990s-style on-premise server-bound tech I’ve described above.
Yes, these old platforms can talk clunkily to the internet and so sometimes people get confused. But they are generations behind what GPs are starting to use the US, Denmark, Israel and other countries.
These countries understood the implications of using distributed web based data sharing a while back and organised or assisted their general practice sectors to rapidly upgrade their tech in order to run that model.
Our GP sector is landlocked technologically and if it stays that way we will have no ability to share data to the degree needed to transform our healthcare system.
By the way, in Denmark, when they were radically starting to alter their hospital system they realised immediately it was useless without upgrading the GP sector, so the government stepped up and with the equivalent of the RACGP in that country funded the entire upgrade for GPs so they could integrate with the whole system.
Our “mothership” piece of infrastructure has to pay GP practices up to $50,000 a year to upload healthcare summaries which are neither comprehensive nor contextually very relevant to what is actually going on with the patients they are generating data on.
GP practices owners would not upload anything if they weren’t paid. They don’t see any clinical return or financial return on them doing it.
In his answer this week to the My Health Record app question Butler displayed quite a bit of ignorance by saying that GPs were fantastic at uploading data to the MHR, while specialists and pathology were terrible.
GPs are terrible too (don’t get upset, no one blames you).
Practice managers often automate the uploading to minimum standards just to get paid.
Even at $50,000 a year ePIP only requires a certain proportion of records to be uploaded so My Health Record has both bad data and not enough data on patient summaries to be really useful.
Specialists don’t do it because they aren’t paid and like everyone else they don’t have time.
Yet Butler is foreshadowing that somehow we are going to engage all these healthcare professionals to upload data and keep doing it for eternity.
He doesn’t get it.
How do you imagine we’re going to manage chronic care patients if we don’t have specialists engaged properly into a system of data sharing?
In the US and Denmark, which use the distributed data sharing models I have described above, these problems largely don’t exist.
A patient’s most important data set generally sits with their GPs and then with hospitals and with a distributed system the data stays largely with each system where the fidelity is high because it’s the source.
The data is pulled out and sent where it’s needed when it’s needed.
No one has to take time to extract data and send it to a honey pot of centralised government run data and the government doesn’t have to spend a fortune trying to incentivise time poor doctors and administrators to do it.
This isn’t all bad news.
Australia is in position to start on a program of upgrading its installed technology base like the US and Denmark, and interestingly there have been plans within the Department of Health and Age Care to do this via introducing a technology standards regime for a while now.
But the My Health Record mothership idea needs to stop.
Does such rhetoric persist because Mark Butler needs the problem explained to him as though he’s six?
He certainly would not be the first one to take on spec what he is being told by bureaucrats and minders – “all good here mate, check out this mothership over here…it’s a bit clunky sure, but it’ll do the job, we just need to modernise it (gulp) …”
As a part of the process of moving towards the 21st Century Cures Act, the legislation that underpinned the radical transformation of the US technology base in healthcare, the US government (which to be fair has been far more in need of a solution to its healthcare crisis than us for years now, so has been motivated) pulled together a series of very high-level experts who were very knowledgeable about technology and healthcare.
Ironically, one of those people was our very own Graham Grieve, who pretty much started the revolutionary new web data sharing resource and standard FHIR.
Butler would do well to put together something similar in Australia and get some one-on-one views from actual experts, including one or two experts from the US and Denmark, to give him some perspective.
So far, it looks like the only technology advice he has taken is via the Strengthening Medicare Taskforce, and the only group representing technology strategy on that taskforce was the Australian Digital Health Agency, who you might easily argue is hugely conflicted because it is the group that conceived, built and spent more than $2bn on this project.
It wouldn’t be hard for Butler to do a bit of checking and get himself some much better perspective on what is probably the single most important near-term piece of health reform this country should be looking at.
If you’re interested in learning more about the ideas and issues discussed in this article, most of them will be raised and discussed at the upcoming Wild Health health leaders reform summit in Canberra on May 3 and 4. The summit has drawn an array of senior healthcare leaders and influencers, including people like Stephen Duckett (former federal Health Secretary and current board member of Healthdirect); Hans Erik Hendriksen (former CEO of Healthcare Denmark); Dr Nicole Higgins (president of the RACGP); Dr Danielle McMullen (vice-president of the AMA); Jay Rebbeck (international co-commissioning expert); Elizabeth Koff (CEO of Telstra Health and immediate-past health secretary of NSW); and many more.
Check out our speakers and agenda and get your tickets HERE.
