8 August 2018

Five things I like about MHR

ADHA Government Insights MHR

  1. It’s a brilliant idea that will, in the end, transcend the  ability of government to screw things up in ways we couldn’t predict.

The MHR is not:

  • A centralised database, largely of hard-to-read documents where useful atomised data can’t be accessed and organised easily.
  • Built on a platform that is already becoming outdated and being superseded by smart cloud and mobile solutions.
  • Locking up lots of important information in one place where it can get hacked all in one go.
  • A public relations disaster because nobody asked anyone – especially doctors – if it would be OK to publish all their data and make it accessible to health professionals without asking permission, thereby angering huge swathes of the smart and capable special interest groups ( for good reason) and causing mayhem
  • Endangering the whole project through trust and privacy concerns.
  • Making the majority of doctors work tune out, and likely to opt out.
  • Inhibiting short-term innovation in healthcare by making access to the system near mandatory for real innovators as a part of anything they do, which is expensive, time consuming and largely not necessary.
  • Diverting money away from the real innovators where that money would get a seriously better return on investment.
  • Targeting 25 million people, when its best return on investment would be in targeting the population that really needs it, which is only something like 200,000 to 300,000.

MHR is an idea. A bloody good idea.

To the ADHA’s significant credit, it never really let the “idea” get hijacked or tainted. It has stuck with it, partly because selling the idea gets it over the line on a less than perfect implementation of that idea. But also because the ADHA believe in it. That the ADHA believes so wholly in the idea and keeps it healthy and thriving is vital and not to be undervalued.

What about that the agency constantly trying to pull the wool over our eyes on how poorly the project has been rolled  out, and failing to get agile and adapt to the world around it in real time to meet the objectives of the “idea” in a much more efficient and productive way? Well, that isn’t great, but it could be much worse. We could not have set out on the journey in the first place and not have anyone even trying.

And frankly, that the ADHA has not been able to realise the idea in the format that it has been sold to us all is not surprising.

It’s probably the most difficult government project ever undertaken when you consider the component parts: health, digital transformation, doctors, hospitals, governments of state and federal persuasion trying to agree, death, risk, regulation, et cetera.

So a fair bit of kudos should still accrue  to all those who have put their hearts and minds into it so far.

The MHR concept was never going to work the first few times around. Even in normal IT projects things don’t work the first few times. There’s a very well known adage for IT platform projects that goes like this:

  • The first time you don’t know what you don’t know.
  • The second time you know enough to know why you failed the first time.
  • By the third time, you are only just starting to understand the real issues and plan for them adequately.

So the MHR should by rights be coming up to its third go. And it’s not dead in the water by any means. But it does need another round of change to adapt to what many think,  is still wrong.

In relative terms the ADHA (and NEHTA for that matter), has done a much better  job  than it is getting credit for, to get as far as it has so far.

  1. It isn’t all bad, by a long shot. In fact, there is a lot of practical and lasting good in this iteration of the MHR.

There is a huge amount of progress in some of the things the ADHA has achieved in its short tenure  This has been overshadowed by some key failures, probably the biggest being that it misjudged just how upset doctors and the general population would be with not being asked for consent on the opt out.

Here’s a short list of some very tangible and useful things they’ve achieved in this iteration of the MHR, by no means exhaustive:

  • Brought the big private pathology and imaging companies to the table to help the common good by making them open their data up for use and agree to stop the madness of locking up data in their bespoke, secure messaging systems that are hard to use and have remained so, largely for reasons of commercial advantage.
  • Brought state and federal health departments to the table and opened up the first decent access to public hospitals for digital exchange of information to the primary sector.
  • Some data collection that is vital that likely no one else will do, at least in the short term, because some data is just hard to get, and for many parties too expensive to access currently.
  • A lot of tangential work on interoperability that is important to the whole ecosystem of sharing patient data.
  • Legislation that is unique and provides a lot of important options and flexibility for the project.
  1. The future of the MHR is exciting and it doesn’t look too far away

If the MHR is a great idea and today’s version isn’t living up to that idea, then it doesn’t feel like we have that long to wait until we start seeing it in a much better format.

This may or may not have be an obvious takeout of a very recent exchange between the head of the ADHA, Tim Kelsey, and the founder of a revolutionary new standard for sharing health data on the web called FHIR (pronounced ‘fire’ for effect), Grahame Grieve. At last month’s Health Informatics conference in Sydney a very surprising exchange took place. Surprising and hopeful.

To understand you need the short version what FHIR might mean.

FHIR is a standard that is designed to translate hugely complex healthcare data which is shared throughout complex off-the-cloud healthcare databases via other more complex standards – the main one is called HL-7 –  and get it easily, but very securely, moving across the web interface so everyone can start sharing data like they do in normal digital markets, such as finance, travel and retailing.

In theory, it will allow consumer applications to talk to distributed databases in health, via the web, with much more security than the MHR, because it is only dealing in data that a consumer wants to share or needs to obtain through their app. And it is on their mobile, which is protected by Apple or Google, and only talking to relevant databases, not one giant central one.

The bottom line of the emerging FHIR revolution is that very soon, all things being equal, most the data being centralised in the MHR will simply not be needed to be stored that way. It will be accessed by a patient’s mobile via their doctor’s patient management system, or directly via the hospital, the pharmacist, or the pathology lab they just left. The concept is truly revolutionary.

When Mr Grieve was asked what FHIR means for the MHR, not surprisingly, he said the MHR was an idea and it was evolving and the natural evolution was a distributed, modern version based on FHIR where key patient data was served from a whole lot of FHIR-enabled distributed databases, not just one giant central one.

To everyone’s surprise, when Mr Kelsey was asked the same question he said: “What Grahame said”. In other words, the ADHA does acknowledge a good degree of its current shortfall, and is at least looking hard at the technology that can make the MHR idea really fly the way it is meant to.

FHIR has a long way to go and part of that journey requires the government stops pussy-footing around and helps fund the faster development of this revolutionary technology. The US and UK governments have embraced FHIR. Australian hasn’t yet, probably because it signals a future that means the giant, centralised current iteration of MHR is already outdated.

But at least the ADHA admits it. That’s a great start.

  1. It’s been a lightning rod of controversy around which big changes have begun to catalyse in digital health, and not just on electronic health records.

Being controversial really isn’t such a bad thing really if you want people to focus on something that is hard to understand, vitally important, but has been ignored for many years. Many people would describe the idea of secure and distributed data sharing between patients and our healthcare system as a “wicked” problem – one that is near impossible to solve without massive collaboration and thinking.

The MHR has focussed the nation on something that is really important to all our futures. The ADHA has made a few mistakes. And upset a lot of people, doctors included. But there is a silver lining in that in that digital health is getting the attention it deserves. And some of the debate is surely leading us all to understand just a bit better how hard the job of the ADHA has been and that maybe  it needs help and love as well as the constructive criticism.

Let’s look at opt out as a clever way to engage Australians in a subject they were never really going to get their heads into unless we really upset them with some decisions we made without them. It worked. Now let’s take all that feedback and get on with things. Rome wasn’t built in a day.

  1. Collaboration is not a dirty word in healthcare anymore

One of the amazing things that is carrying the FHIR web sharing standard into  places it should never had been this quickly is what Mr Grieve, describes as “the community”. The community, at its core, is most of the developers who understand things such as HL-7, and the difficulty of sharing data in health, as opposed to other, far more advanced,  digital markets such as finance, travel and retail.

Health hasn’t been disrupted like those markets because it is much more complex and harder to do. It involves governments, life and death and risk, lots of regulation, and, doctors, whose minds are not easily changed.

But “the community” is rallying around digital health, and around the FHIR standard. At its beating heart, this community is all about collaboration these days.

FHIR had its origins in collaboration. Its founder just wanted to release the power of the web to health, so developed it initially as means of more easily getting complex offline data onto the web to be shared more seamlessly.

He then gave the standard to the HL-7 global organisation for free, as an open standard. Like other famous open standard software projects, such as Linux, Apache, Git, Java and Android even, Mr Grieve worked out that things would progress if he shared his work with the community and encouraged that community to keep sharing.

Healthcare software’s history has been pretty much the opposite to collaboration. But FHIR, and the MHR have forced so many parties together for the greater good.

That “collaboration”, both among developers, and now increasingly among competing companies (Cerner vs Epic for example) and governments (state vs federal), is the new black.  MHR has helped that process.

Every other market that has been digitally transformed woke up to collaboration. Now health has as well, thanks, in part, to the MHR.