-----
Re-platforming the MyHR
Posted on by Grahame Grieve
A couple of months ago I made a post proposing that the MyHR could be reworked to use the Argonaut interface. This change could open up the MyHR infrastructure so that providers and vendors could collaborate to build innovative local solutions without being bound to national political concerns about what is possible or not.
That caused quite a bit of discussion and media interest (e.g. in Pulse IT). This week I met with Tim Kelsey (CEO, Australian Digital Health Agency) as part of the follow up, and he pointed out an as-yet under-appreciated aspect of the Agency’s work plan for next year.
Quoting from the Budget Measure:
'This measure allows the Australian Digital Health Agency to implement plans that will increase efficiency and sustainability and reduce the ongoing operational costs of the system in the longer term through the transition of Department of Human Services’ support functions to the Agency, the re-tender of the national infrastructure operator and delivery of a new, more flexible platform."
I’ve heard lots of discussion about efficiency and sustainability, but it seems that the commentariat haven’t picked up on the significance of ‘a new, more flexible platform’.
The Agency will be conducting community consultation around this subject in the first half of next year. In advance of that, the community of interest in digital health in Australia – which is a pretty wide set of people and institutions – should start thinking about these questions:
- How should the Agency conduct the consultation to ensure that it gets good, honest open input from the community, without political noise overwhelming the process?
- What kinds of improvements could be made to the MyHR to make it a more flexible platform that are both politically and financially realistic? (and that align with the digital health strategy, since that’s that the official national strategy)
- How should the technical specifications that underpin a more flexible system be managed (e.g. what’s the standards process for digital health in Australia)
Tim will be commenting in public about this subject too – I’ll link to his comment from here once he makes it (maybe later in January).
Note: I have a contract with the Agency, to provide advice around strategic direction with regard to the Secure Messaging project, which overlaps a little with regard to this subject. But in this case, I’m not speaking on behalf of the agency in any official sense.
Here is the link:
http://www.healthintersections.com.au/?p=2765
-----
Now, most who read here would know I favour abandonment of the myHR and the use of other more modern and privacy respectful approaches for health information use and sharing but if we are to be forced to have a national application with most of the attributes of the myHR then shaping its outline and functions to be as worthwhile as it can be is vital.
Please use the comment section to share your views on the optimal future for the myHR and the questions posed by Grahame in his post.
David
Here is the link:
http://www.healthintersections.com.au/?p=2765
-----
Now, most who read here would know I favour abandonment of the myHR and the use of other more modern and privacy respectful approaches for health information use and sharing but if we are to be forced to have a national application with most of the attributes of the myHR then shaping its outline and functions to be as worthwhile as it can be is vital.
Please use the comment section to share your views on the optimal future for the myHR and the questions posed by Grahame in his post.
David
43 comments:
There's an important shift of language here - from "MyHR is a solution" to "MyHR has a flexible platform" - I think that your "use of other more modern and privacy respectful approaches for health information use and sharing" is not inconsistent with that.
Also, I didn't nominate where input should be submitted; I just noted that there will be a place for input, and people should make that part of their thinking. Of course, There's a circular redundancy about seeking comments about how to gain input as part of the input; people are always welcome to comment here or on my blog or to me directly (or to the agency directly, of course) about how to gain input. But I've always found that it's most productive for building community if people can make input in public, and comment on each other's input. It's too easy for the discussion to break down into a whole set of parallel disconnected discussions that divide the community rather than build it.
"a new, more flexible platform" by any stretch of the imagination this means - a complete architecture and system design.
Let's not be deluded - MyHR is a failed project.
IMHO, the biggest problem with MyHR is that the government gets to acquire, keep and use the medical data of most Australians without their consent. The MyHR isn't a platform, it's a document management system (not a very good one) built upon some very shaky infrastructure (e.g. the infrastructure doen't identify individuals, only institutions).
The government is not obliged to tell people if, when and how their data is being utilised internally. All the talk of access controls and audit logs is meaningless when the government, as system owner, accesses the data. Nothing gets deleted or overwritten, it's all there for perpetuity.
As as has been noted in another comment, even the ePIP bribes are not working well. GPs are starting to realise that the data in MyHR could be used against them and without their knowledge. Based upon the recent Centrelink robo-debt in-competencies, it is not beyond the bounds of possibility that the government could misinpterpret the data in the system, especially as it may well be incomplete, potential inaccurate, input by some other person or institution and be inconsistent with what they have put in.
GPs are already becoming distracted by their computer systems. Adding another one on top will only compound the problem. GPs are safer ignoring the MyHR.
There is no justification for the government to have such detailed data on Australian's medical treatment and it is highly likely that this issue will become the central focus of opposition to the move to opt-out.
The ADHA would do well to focus on true infrastructure capabilities and steer well clear of clinical applications and databases of medical data.
More flexible platform, sounds like elasticity sounds like cloud, sounds like someone has been reading Qantas inflight magazines, or maybe just PDF everything to date and archive it and start again. The platform does not address the applications.
Good to see some consideration of alternatives even if it does result in another five years of entrenchment of the GovHR.
I do hope the ADHA also brings on board experiences people that can go toe to toe with you Grahame, just to ensure a balanced view and scientific analysis
You can't polish a turd.
Out of interest why is the conversation a technology one? Health to me seems one of the System models where the only true model is the system itself. We seem to apply all these fantastic skillset towards modelling a set of constraints to a model a system based on those outside the models day to day use. There is more to healthcare than simple transactions. It is and always has been about healing, this is a human thing, technology has a role to play but somehow has become the dominate player that all others must obey.
All,
While I realise the myHR needs to be considered in a range of dimensions beyond the technical, one gap in most discussions on the myHR is exactly just what the technolog(ies) underpinning the myHR are and how they are assembled.
A decent diagram of just which goes with what as far as the myHR is concerned would be of interest to many here in contemplating what possibly comes next.
I have a feeling if we knew we may be saying "I would not choose to start from here!" to move things forward.
Does anyone have a good link?
David.
What is the problem to be solved?
"The Agency will be conducting community consultation around this subject in the first half of next year." Ah yes. Community consultation - the panacea approach!
Not another round of butcher's paper, breakout groups comprising a miscellaneae of interested parties (euphemism for 'stakeholders') cobbled together and led by ADHA facilitators. Committees, gathering a truckload of ideas and wishes then knitting them together, plopping them in the hot and fervent cauldron of urgency and stirring vigorously in every direction, before serving the delicious broth full of enticing spices for everyone to embrace as a solution. The bureaucrats approach to technology development at its best.
Some will have noticed the app for Android called Haven by Ed Snowden. Seems strange to me that at the user level there's no end of protections that can/should/must be used to protect one's own data, but we happily hand off the protection of data to the "cloud". We have no control over those transactions, whether they be performed by people with the highest integrity, or by clowns who may be corrupt, incompetent, under duress or impaired by alcohol and other drugs. By "strange" I mean dissonant, in that we willingly step into a void where we accept bland assurances like "Govt will own the data" as if it's the pure truth. No need to ask, no need to test, no need to bother, it's all OK.
If a system cannot be built with robust and auditable access controls, and an elected representative wears the cost of failure for life, then it shouldn't be.
They are mere trifles, though, and easily dealt with in other domains. The biggest failure of MyHR is that it is not linked to a national identity management scheme - because there isn't one.
"The biggest failure of MyHR is that it is not linked to a national identity management scheme - because there isn't one."
That's because MyHR is the core of a national identity management scheme. Even if you opt-out, you give the government so much information that they know who you are and can link you with a number of ID constructs. The government's Trusted Digital Identity Framework will be integrated under the covers with MyHR, as sure as heck.
The government has learned the lessons of the Australia Card and the Human Service Access card and is carefully avoiding all talk of a national Identity Number. However, internally they could create a virtual identity scheme which operates by linking every personal database they can get their hands on. They've already linked Centrelink data with ATO data. They made a mess of assessing/reconciling welfare and income, but there hasn't been much talk about incorrect identities. Not everyone is known to the welfare system or pays tax. Health care on the other hand is far wider in scope and subject matter. As it says on myhealthrecord.gov.au "A My Health Record for every Australian in 2018".
Doctors and other health care professionals will be feeding this system under the pretext of health care. IMHO, such people should ask themselves - what do I get out of this compared with what I put in. If there is a net benefit to the health care professional and the patient, then OK. If not, they why do it?
And I wonder if anyone has told ADHA of this larger initiative? There is no reason why they should be told, all they have to do is make sure that the system gathers as much personal data as possible. Health outcomes are irrelevant. So is the lack of health care benefits from the $2b spend.
Why do you suppose MyHR is being re-platformed? When it was originally built it wasn't optimised for integration into other government systems. Now it can be.
Look at the reason for the re-platforming "This measure allows the Australian Digital Health Agency to implement plans that will increase efficiency and sustainability and reduce the ongoing operational costs of the system in the longer term through the transition of Department of Human Services’ support functions to the Agency, the re-tender of the national infrastructure operator and delivery of a new, more flexible platform."
Not a mention of health outcomes, only efficiency of operation and flexibility.
Flexibility is the key word. So flexible that the data can be used for "other" purposes.
And I have no special knowledge or privileged information. What the government is doing is all published on government websites and publicly available. All you have to do is join the dots. Which is exactly what the government is doing, except the dots are us, the people.
And I don't actually object to what they are trying to achieve, it's quite sensible. It's the dishonest way they are doing it and the lack of safeguards that should be baked into the system and which might have happened if it had been done in a transparent manner. Not to mention the government's track record for stuffing up large scale IT projects.
I for one welcome this news, it is a much needed conversation and if done right we might be able to build on the many outcomes (positive and less so) of the past 10-20 years. To do this it will need to be very open and objective. It will not be a quick and easy journey.
The ADHA needs to clean up some internal issues which are a left over of a need when forming but sound more and more surplus to requirement.
Build a set of strong leaders to orchestrate this undertaking
Ensure that communities have access to communication tools and access to developing designs, policies and decisions.
Adopt the Standards principles, which the do in part so adding a few more should be simple enough.
Is the My Health Record the solution or the problem? If the former why and to what? If the latter how can that be?
What will My Health Record deliver that we all need?
Not wanting to rain on anyone’s parade but I think this is yet another smoke and mirror con job. Intended to hash and/or distract people all the while the Government human records system gets further entrenched.
I am confused. What is the smoke and mirror con job to which you refer?
You ask "How should the technical specifications that underpin a more flexible system be managed?"
Embedded in this question are two unsubstantiated assumptions:
1. that a more flexible system has been defined?
2. that the MyHR will form the basis of a more flexible system.
Surely you have the horse and the cart all back to front.
What is a more flexible system?
Why should the MyHR form the basis of your "more flexible system"?
Should not the technical specifications that underpin the MyHR have been sorted out a long time ago?
Changing the fundamentals of a system is far harder than anything else.
And are the same people who didn't perform way back then, trying to fix things now?
It takes more smarts to get out of a problem than getting into it in the first place.
You would be most unwise to base your decisions on the Digital Health Strategy if it is at all deficient in depth, clarity, substance, pragmatism and logic regardless of the political reality that it is "the official strategy". We have to assume that Grahame fully supports this "official strategy" !!!
I fully support having a consultation about the direction of the MyHR (though you couldn't call it official policy yet). What we have now is not serving us well. We'll have to see how it gets done before we know whether the consultation will be done well - can we build a community that actually creates value? I'll be collaborating with the agency as long as I think there's a path towards us having a healthy community. The discussion here might be a good way for people to let off steam, but it misses the point.
The question we should be asking is 'a more flexible platform' for what?
"And are the same people who didn't perform way back then, trying to fix things now?" - as in, the Australian government and community?: no, we didn't perform, but we don't have another government or community, so we (I) have to invest in having another go with the same participants under different conditions with the intent that we can build on what was an important start to make something more useful.
Hindsight is a great thing and a full told you so bucket brings a warm feeling. I don’t believe it is a case of performing or not. Everyone with passion performed. Choices where made and some panned our and others have not reached the potential first envisioned. None of this is wasted, evolution moves in mysterious ways. The ADHA has to deal with many challenges and has made some mistakes, in some instances to other industries advantage.
I hope 2018 starts to demonstrate a different set of conversations are emerging that look now at the opportunities presented. For one the ADHA needs to become a government entity and support the slow but essential architectures, policies and standards needed and with the community shape what conformance and compliance means.
Keep supporting them on the journey.
Grahame I'm sorry to have to tell you that this is the most stupid and naive thing I have ever seen you utter "so (I) have to invest in having another go with the same participants under different conditions. .. "
First, to be confident of making progress, get the government and bureaucrats out of the way.
re: "First, to be confident of making progress, get the government and bureaucrats out of the way."
Hear hear.
What does this really mean?
".... transition of the Department of Human Services' support functions to the Agency, the re-tender of the national infrastructure operator, and delivery of a new more flexible platform."
What should we make of this?
".... transition of the Department of Human Services' support functions to the Agency, the re-tender of the national infrastructure operator and delivery of a new more flexible platform."
Just a consolidation of services and the Accenture contract was most likely a 5-7 year Agreement.
Will not be an easy task on either front.
Far from just a consolidation of services.
1. The Health Department has already distanced itself from the 'project' having come to appreciate the risks and complexities are way beyond the skills and competencies of the public service.
2. The next step is to distance Human Services and transfer remaining risk to ADHA.
3. Next is to begin the process of readying ADHA for transference to the private sector.
4. First secure the National Infrastructure Operator via a 'tender' and in the process lock-in some long-term T's&C's behind the scenes the will give the government 100% access to all the information collected in exchange for a Commercial -in-Confidence arrangement (to be negotiated) which will underpin the operator's requirements for project viability, ROI, and reasonable profit margin.
5. The operator will then commence the progressive dismantling of the MyHR replacing it with the operator's proprietary platform under the guise of a "new more flexible platform".
6. Finally under the T's &C's the operator will, if it so chooses, float the project including IP as an IPO enriching management in the process.
We are watching here the slow steady unravelling of responsibility and the transfer of IP and ownership to private owners for maximum exploitation.
So the system will be owned by a private sector organisation with the government having full access to the data?
Either scenario is bad, put the two together and it becomes political suicide. Most people won't differential myhr from the government.
Re "That's because MyHR is the core of a national identity management scheme." (Bernard R-D December 23, 2017 4:04 PM).
I see your logic, Bernard, but rebutt with 2 words - Peter Dutton.
If ADHA and/or DTA was/were under Home Affairs, your line-up is inevitable. But, the fierce rivalry between Ministers (their ambitions) and the bureaucracies pretty much ensures the status quo. From my untrained perspective, I lost interest when the APPs were legislated. That would have been an obvious point to hold inter-Ministry chats on how the Govt defined its citizens and went about assigning unique identifiers. The APP law does not refer once to "unique".
It looked to me that the whole exercise of harmonising the Privacy Principles was to safeguard the rights of organisations and agencies, not so much the rights of individual Australians.
Of course, I stand to be corrected on any of that.
So why is my photo (mug shot) not compulsorily included?
As I said, it's all publicly available:
PUBLIC SECTOR DATA MANAGEMENT PROJECT
https://www.pmc.gov.au/public-data/public-sector-data-management-project
"In early 2015, we led a study about how public sector data helps the government with better service delivery and efficiencies. The study also looked at how the private sector can use data to stimulate economic activity."
....
"To overcome these barriers, the project established a road map to make better use of public data. It began with a period to build confidence and momentum across the APS. After that there was a longer term set of ideas and programs to make it easier for the APS to use, publish and share public data."
This is the road map.
https://www.pmc.gov.au/sites/default/files/publications/public_sector_data_mgt_project.pdf
This is a report on its implementation
https://www.pmc.gov.au/sites/default/files/publications/Implementation-Public-Sector-Data-Management-Report_0.pdf
A lot of the information in these documents refers to research data and data that can be publicly released, but if you look carefully and from the perspective of agencies sharing personal data internally (and where the APPs are less rigorously applied), then you can only concluded that, not only is it inevitable it is happening now. The Centrelink robo-debt initiative was built on data sharing across agencies.
And if you look carefully at the last slide in the road map slide pack - "Appendix 6 A trusted-access model", data held by the states is included in the scope of the project.
Not only that, but this was the project team:
Secretaries Michael Thawley, Jane Halton and Drew Clarke with guidance from an inter-agency Steering Committee.
Recognise any of those names?
It's all about better service delivery. IMHO, the government is more interested in the cost of the health care system than the health of individuals.
Ministerial ambitions will always be constrained and shaped by the man at the top - if they don't have their job, their ambitions will come to nothing.
Furthermore, students of Yes (Prime) Minister, will know where the real power lies - the secretary level, especially after John Howard politicised the APS.
Bernard, any anxiety I may have had re covert implementation of centralised identity apparatus is cooled by another two words - birth certificates.
IMHO it would take at least 5 years from the first blush, and probably more like ten, to achieve nationalisation of births, deaths & marriages. That's the duration of at least two Parliaments and who knows how many PMs and relevant Ministers. Since the legislation may require a referendum, no Secretary (even if named Michael, Jane or Drew) will hoist a project that could end in a failed referendum.
Hence, a program like that has to be embedded in the living will of the electorate, and accomplished within a time-span that can keep a "memory" alive.
So, while we all may "benefit" from that nationalisation, most are content with the status quo. Then, the point from which to tackle & sustain a good hard look at all the ins & outs would be to address those who are disadvantaged by the current system. I reckon we'd find that some indigenous people are, at least, on the margins of current benefits. Anything, though, like citizenship rights in that (ATSI) domain will be complicated by a broad and troublesome ongoing agenda.
Not to forget to mention that Human Rights isn't going anywhere, and that any proposals for defining when life (ie citizenship) begins will mire down hard in the quagmire of "religious freedoms".
Trevor,
The point I am making is that they are doing this linking of data as much as they can.
I am not suggesting that they will succeed, especially when it comes to collecting and utilising personal medical data. You have highlighted some problems/difficulties there are many more.
BTW, are you aware of the The Document Verification Service (DVS)
https://www.ag.gov.au/RightsAndProtections/IdentitySecurity/Pages/DocumentVerificationService.aspx
"one of the key initiatives of the Council of Australian Governments' National Identity Security Strategy, which is managed by the Attorney-General's Department on behalf of all jurisdictions."
The system permits, among other things, verification of birth, marriage and change-of-name certificates
Interestingly they say this:
"The DVS is not a database. It does not store any personal information."
Why not use the same system level architecture for medical information? Clinicians, when they need answers to medically relevant questions, could ask the question and get accurate, complete and up-to-date answers - if there are such answers.
You've got me poking around in the weeds, Bernard.
The National Identity Security Strategy has (is led by?) a National Identity Security Coordination Group, on which information is, er, sparse. Set up in 2007, the Govt Directory records "Annual Report Prepared and tabled? No."
Back to the DVS, that page does refer to "other identity verification methods". I wonder if banks & other financial institutions use DVS. I wonder if DVS is auditted and its performance has been spoken of in Senate Estimates. Too many queries.
I do not want to look at Face Matching Services, and start worrying what the new A-G will do with it. Although, it would be amusing if he suggests banks should be using facial identification to authorise deposits of large bundles of cash at ATMs.
I'd better call it a day. Thanks.
@12:39PM point 2 transferring operational risks from Department of Human Services to the ADHA makes sense.
DHS is a payments Agency - pensions, PBS, MBS, etc. MyHR does not involve payments, it is an information data dumping ground, riddled with huge complexity and risk. DHS needs MyHR like a hole in the head. It is a distraction from its core business.
I cannot help wondering how this conversation could lead to anything but further investment in the MyHR. The title itself promotes this constraint. I would be far more comfortable if it either left the brand out or asked is the current approach a solution of a problem in the current technology and healthcare landscape.
As others have alluded to we need to revisit the question of the problems to be solved, this should also involve processes, clinical workflows, policies and politics, funding and other resourcing needs. The clinical workflows and processes really do need looking into as I am not convinced they were developed with interoperability in mind. Not that they all need changing, it maybe as simple as new ones are required
@ 9:47 AM. Agree, it does come across as ‘lets All agree on the MyHR but slap FHIR logo in it to pretend it is something new and bold’
So the conversation should be a combination of political, economic, socio-cultural and technological viewpoints. That sort of talk is a PEST and would require far to much honesty.
10:55 AM FHIR is still in its infancy. It has a long way to go to reach maturity along the R&D pathway towards commercialisation. It seems promising if the claims of its developers are to be believed. Cerner has secured an early position which should ensure Cerner is well placed to exert influence over its IP should a commercial opportunity to do so arise.
> ‘lets All agree on the MyHR but slap FHIR logo in it to pretend it is something new and bold’
That bit has already been done but as we all knew, it wasn't something new or bold. So now we get to talk about what that might be.
> FHIR is still in its infancy
I think we're out of that now, but there is still a long way to go
> Cerner has secured an early position which should ensure Cerner is well placed to exert
> influence over its IP should a commercial opportunity to do so arise.
I don't know what this means, but want to underscore, since rumours of claims to the contrary reach me: FHIR is locked in to open source by it's licensing model; there is no opportunity to influence it's IP. Neither me nor anyone else will ever be in a position to extract payment from anyone for using the standard. (Not that anyone can ask anyone to pay for their own implementation - perhaps that's what is meant here, but it's not clear)
@1:36 PM "FHIR is still in its infancy." " I think we're out of that now, but there is still a long way to go."
OK so we all agree MyHR is the wrong solution to a problem which has not yet been defined!
Help!
And FHIR is an open source 'technology' which has a long way to go.
Double help!!
This all means to me - start again, go back to the drawing board, analyse and define the problem and let FHIR progress to where it wants to go. By all means employ it iteratively in small steps to solve some small problems to get a better understanding of what it can and cannot do, without exposing a huge project of many hundreds of millions of dollars to FHIR when the project itself (herewith MyHR) doesn't even understand what problem it is attempting to solve!
Triple help !!!
24 Million is not that big, that is an average city in parts of the world. I do agree though we should be careful to seperate the standards from an implementation. The MyHR was designed for a purpose, repurposing or replatforming technology is more complex than a power point slide or an article in a newspaper.
I still don’t understand if we are investing in exchanges to support interoperability why a document store is needed, surely the sharing of information is a far different model than a push and pull one revolving around a single point.
FHIR is at least and open and authentic movement, one that should aspire better ways of working together.
@3:45 PM "The MyHR was designed for a purpose".
Purpose? Mmm. What exactly was that 'purpose'? Was it solving a problem which had not been clearly defined?
http://www.health.gov.au/internet/budget/publishing.nsf/Content/budget2010-hmedia09.htm
Interestingly it states this
Australians will be able to check their medical history online through the introduction of personally controlled electronic health records, which will boost patient safety, improve health care delivery, and cut waste and duplication.
To be picky, that does not mean a single government run document library. It simply means health records that can be accessed online and some form of personal control.
Post a Comment