Sunday, April 22, 2018

The Way Health Information Is Handled In Australia Looks Like Changing. Badly Needed Indeed.

A kind commenter sent this link to the blog today:
Following it you get this:

Interoperability and Connected Healthcare in Australia Early Engagement Paper

27 February 2018 V1.0  - Approved or external use
Document ID: Interoperability Early Engagement Paper Version 1.0
What is not clear is just why it is hosted on the Aust. College of Emergency Medicine Website? This also looks to be the only copy out there.
The whole document is well worth a read (Written by EY) but there are some highlights to me:
First:
What Can We Learn From Other Industries And Countries?

There is much to learn from the experience of other industries and other countries’ health systems.
Examples of successful interoperability in other industries include the financial services sector, the aviation sector, and the smartphone ecosystem. An examination of these industries by Ernst and Young shows common themes:
·         The need for clear regulatory and legislative frameworks. For example, in the financial services sector, clear rules and protocols are set and governed by the Reserve Bank of Australia’s Payments System Board and by the Australian Prudential Regulatory Authority.
·         Reliance on standards bodies. The telecommunications industry has a single global mobile standard: the Global System for Mobile Communications (GSM). In the smartphone environment, the rules for developers to participate in the ecosystem are set by the operating system controllers.
·         Clear leadership. For example, in the aviation industry, all airlines rely on the International Civil Aviation Organization (ICAO), and the European  Organisation for Civil Aviation Electronics.
·         Collaboration and self‐regulation. For example, the Australian Payments Network (AusPayNet) is a membership‐based industry association that acts as a self‐regulatory body.
This points to the need to resolve the issue of ‘who sets and enforces the rules’ for Australian healthcare interoperability, and the balance between government and self‐regulation.
Other countries have taken various approaches to health interoperability. A summary of these countries that are most relevant to Australia’s healthcare system may be found at
There are recurring themes across different health systems that point to issues that need to be resolved in the Australian healthcare context to promote interoperability, and in doing this, connected healthcare:
·         Most countries have a government body with some responsibility for standards and interoperability, although this varies from ‘mandating’ to ‘facilitating’. Some countries complement this with a degree of self‐regulation.
·         There are varying approaches to how data is exchanged, some with centralised data repositories and others with distributed models where patient data is stored in the original collection point and extracted when required.
·         Most countries have a system of unique patient identifiers, with some extending this to both health and social system (with clear benefits of an integrated social approach to healthcare).
·         Most countries draw on international terminology standards, rather than creating their own.
·         Various ‘levers’ have been used to promote adoption, ranging from mandating standards, to stipulations in funding agreements, to providing resource materials (toolkits, implementation guides).
Second:

What makes interoperability?


Enabler
Relevance to interoperability

Identity management
Uniquely identifying health consumers, health and care providers and health organisations in the healthcare ecosystem.
‘I am who I say I am’.
Authentication and authorisation
Ensuring appropriate levels of access to a patient’s health information.
‘As a clinician, I can get access to the information I need to provide care’.
‘As a consumer, I have access to my information and have control over who can access it’.
Integrated electronic health record
Allows health consumers, health and care providers and other participants to access health information, with the consumer at the centre.
‘My health information is available to clinicians involved in my care’.
Unified health directory services
Provide the technology for health consumers and health and care provider to find health services quickly.
‘I can easily find the health and care providers I need to communicate with electronically’
Information exchange
Developing clinical informatics specifications to facilitate information exchange in alignment with relevant data quality and clinical safety requirements.
‘My clinical software communicates with other systems to bring me what I need’.
National standards
Promoting the definition and adoption of consistent standards to facilitate information exchange and provide conformance and compliance mechanisms.
‘Minimum standards are accessible and easy to understand and implement’.
Security
Common standards, understanding and adhering to required policies to protect the privacy and integrity of information.
‘I can rest assured that my information is secure’.

Third:

What shifts to current Australian paradigms do we need to consider?

Comparing international and other industry approaches to interoperability exposes a number of opportunities for Australia. While individual technologies or implementation approaches will need be co‐ designed at a later time, this analysis suggests a number of possible paradigm shifts that may be appropriate in the Australian context. These are described below.

Moving away from an exclusive reliance on a document paradigm

Currently, My Health Record receives information in one of a range of document types that are available for different purposes[13]. There are some advantages to this model: It reflects the traditional practice of healthcare to communicate in documents and does not require a high degree of common terminology. Specialists in Australia, for instance, are familiar with reading referral letters from General Practitioners who have trained and practiced globally, and have differences in terminology.
While documents will continue to play an important role in healthcare communication, some data may be better captured in an atomic form. Atomic data captures information in discrete packets that represent something about a person or service. Frequently, this is numerical data, representing a blood test result, a height, weight, heart rate or blood pressure

Strengthening the use of identifiers for patients, providers and services.

Most Australians have an Individual Healthcare Identifier[14]. However, not all regular care users in Australia have an Individual Healthcare Identifier (IHI). There are also variations in how IHI’s are looked‐up, stored and used in different care provision settings.
It is just as important that we are able to identify providers of care in order to make information on patients available to them. Current provider identification systems, including NASH certificates, are cumbersome and there are ways in which these can be streamlined without compromising security.

Develop new collaborative models for the governance and approval of standards and specifications

The implementation of co‐ordinated improvements will require a multitude of decisions as to “how”. Without agreement on how information will be made available, requested and transmitted, widespread implementation is impossible.
The standards and specifications that define all of this must be developed collaboratively with industry and with a strong focus on suitability for use in the clinical setting.
The process itself for achieving this must be developed in partnership with stakeholders.

Extending the role of My Health Record to include acting as a trust broker to facilitate access to information or resources that are held on other systems.

Currently, My Health Records acts primarily as a repository for information. Clinical information is uploaded into My Health Record by some people and downloaded by others. This limits the type and amount of information that My Health Record is able to facilitate access to. In the age of wearable devices and widespread home monitoring, there may be value in enabling consumers to use My Health Record to link information from disparate sources. Similarly, My Health Record could provide capacity to link a radiology report to the associated images (stored elsewhere) enabler clinicians to have faster access to diagnostic images.
Fourth:

Data rights and responsibilities in Australia

Globally, the issues associated with access and custodianship of information are being redefined in ways that reflect the changing nature of a consumer’s relationship with information about themselves or those that they care for.
The European Union has introduced the General Data Protection Regulation[15], which comes into force on 25 May 2018. These regulations includes provisions for:
Right to Access
Part of the expanded rights of data subjects outlined by the GDPR is the right for data subjects to obtain from the data controller confirmation as to whether or not personal data concerning them is being processed, where and for what purpose. Further, the controller shall provide a copy of the personal data, free of charge, in an electronic format. This change is a dramatic shift to data transparency and empowerment of data subjects.
and Data Portability
GDPR introduces data portability the right for a data subject to receive the personal data concerning them, which they have previously provided in a 'commonly use and machine readable format' and have the right to transmit that data to another controller.
A similar approach is envisaged in the United States with the US Office of the National Coordinator for Health Information Technology defining interoperability as[16]:
the ability of a system to exchange electronic health information with and use electronic health information from other systems without special effort on the part of the user.
This reference to ‘without special effort’ draws upon the growing body of agreed, open‐source, Application Programming Interfaces that allow consumer facing applications (like Apple’s new Health Application) to draw information from a range of clinical systems using a standardised approach with the consumer’s consent. Essentially, Government and Industry have collaborated to define a set of interfaces, increasing over time, by which systems share information that becomes the de facto standard of what is reasonable as opposed to being ‘special effort’.
In Australia, the Privacy Act 1988 governs how personal health information is managed by government and the private health sector, with additional legislative requirements applying in states and territories for state based health providers. The Privacy Act and the Australian Privacy Principles it contains, give the consumer the right to access to their own health information, with entities who hold that information obliged to provide it in the manner requested by the individual unless one of a number of lawful exceptions applies. The principles provide that an entity may charge the consumer the reasonable costs of providing this information in the manner requested.
The Australian Government Productivity Commission examined the issues associated with data availability and use, reporting in 2017[17]. This report made a number of observations regarding the potential to improve outcomes and the effectiveness of the health system more generally. Importantly, the report makes the following recommendation:
Marginal changes to existing structures and legislation will not suffice. Recommended reforms are aimed at moving from a system based on risk aversion and avoidance, to one based on transparency and confidence in data processes, treating data as an asset and not a threat. Significant change is needed for Australia’s open government agenda and the rights of consumers to data to catch up with achievements in competing economies.


·         At the centre of recommended reforms is a new Data Sharing and Release Act, and a National Data Custodian to guide and monitor new access and use arrangements, including proactively managing risks and broader ethical considerations around data use.
·         A new Comprehensive Right for consumers would give individuals and small/medium businesses opportunities for active use of their own data and represent fundamental reform to Australia’s competition policy in a digital world. This right would create for consumers:
o   powers comparable to those in the Privacy Act to view, request edits or corrections, and be advised of the trade to third parties of consumer information held on them
o   a new right to have a machine‐readable copy of their consumer data provided either to them or directly to a nominated third party, such as a new service provider
These proposed reforms, and the emergence of consumer‐facing applications nationally and internationally, demonstrate a growing expectation in the community that a consumer’s data both can and will be made usefully available.
----- End Extracts:
There is lots of news here:
1. The myHR is obsolete and hardly relevant. Large document storehouses are a health hazard!
2. Standards matter! And you actually need to understand what they mean and require.
3. There are big changes coming to data control and ownership including health and a GPDR like mechanism(s) is coming over time.
4. The myHR is going to be radically changed as to how and what it can hold and what control patients have. And it has to be able to properly delete old, unwanted material.
5. Primary systems are what clinicians are using – what a surprise!
Have a hunt for other nuggets. It looks to me – on this basis of this – that the ADHA is being dragged in at least way towards some more sensible directions!
Comments please and sorry the formatting is a bit poor!
David.

Added about 5pm. The document totally misses out on pointing out just how complex and difficult real interoperability is I believe. The learning curve has not really begun I reckon!

D.
 

29 comments:

  1. I agree David there are some bold statements in the document. I do wonder if the ADHA understands what they are putting out there, have they published anything relating to the consequences of interoperability?

    To point out some of my lack of faith. They state a need for standards, yet they only mention a framework Nehta undertook in 2007, not the revised version in 2012 and more importantly the Australian Standard.

    I agree there have been some changes technology wish and the way technology and services are delivered.

    As Bernard points out, why is this not public ally avalaible via the ADHA? This and so many other examples just points to an organisation that is not equipped to operate at a national level Jin this complex sphere.

    ReplyDelete
  2. Whoever wrote that document (and/or approved it) doesn't understand security, data integrity and privacy:

    Under The following seven enablers are the foundations of interoperability.

    "Security:
    Common standards, understanding and adhering to required policies to protect the privacy and integrity of information.

    'I can rest assured that my information is secure'."

    Security, data integrity and privacy are very different concepts. To assume that by addressing security you also address data integrity and privacy suggests a naive at best, dangerous at worst, understanding of the issue.

    If ADHA doesn't have the skills and expertise to write that document themselves, it is highly likely they don't understand its weaknesses, the above only being one.

    IMHO.

    It is also worth noting that the notion of a government owned, government controlled national database of copies of health data held in other data bases is never questioned.

    Section 5: says:

    "What Can We Learn From Other Industries And Countries?
    There is much to learn from the experience of other industries and other countries’ health systems.
    ...
    There are varying approaches to how data is exchanged, some with centralised data repositories and others with distributed models where patient data is stored in the original collection point and extracted when required."

    And the lessons are?????? Silence. There are enormous cost, system reliability, system integrity, system performance and privacy differences between the two approaches. Are they relevant? Important? Questioned? Silence.

    There's not even a cute little green box with consultation questions in it.

    This strongly suggests that the government getting its hands on Australian's health data and linking it to other data (i.e. in the MADIP project) is a non-negotiable objective.

    When it comes to interoperability, one of the most (if not the single most) important question should be: Should myhr exist or are there better ways of achieving the objectives and health outcomes.

    Oh, silly me, there are no health outcomes.

    The only place health outcomes are mentioned is in the context of rural health

    "Rural Doctors Association of Australia noting:

    Digital health is increasingly being promoted as a cost‐effective way to redress the inequities of access to health care and significantly poorer health outcomes that exist."

    ReplyDelete
  3. why is this not publicly avalaible via the ADHA?

    Because we probably don’t know what website to put it on or who to ask. It may sound bizarre but the place is a nightmare to find information or who does what. The answer - IT rollout yet another option to collaborate, create and hide stuff, overseen by someone On a self promoting crusade with no grasp of IT or the business. After a year I still struggle how this organisation could grasp even basic information management and I have little faith that without outsourcing the MHR operations they could manage personal health information.

    ReplyDelete
  4. The paper makes a point about how information is captured into the My Healthrecord system using a document paradigm:
    "While documents will continue to play an important role in healthcare communication, some data may be
    better captured in an atomic form."

    I think they are quite confused. The concept of a 'document' (e.g. a discharge summary or a pathology result report) and 'data in atomic form' are NOT mutually exclusive. Even in the MyHR system, many documents can be and are submitted in atomic form (i.e. for CDA Level 3 documents). The problem with the My Health Record system is that it hasn't encouraged or enabled improvements in data quality for those who are still submitting documents in non-atomic form (e.g. in PDF format). It has a mix of documents in atomic and non-atomic form.

    Bit of a worry if the authors of this paper on behalf of ADHA don't understand or have misrepresented this - how can they seriously address the challenge of interoperability?


    ReplyDelete
  5. There was a co-design workshop in Interoperability held last September in Melbourne. It was hosted by Ernst & Young on behalf of ADHA. Given the timing, I assume this paper is the output from that assessment. I certainly haven't seen any other update for the participants.

    I was there on the assumption that we would look at the technical difficulties and constraints. As it turned out, it was taken for granted that technical solutions exist and we spent more time on what sort of problems need to be addressed. e.g sharing data for admission and discharge, adoption rates especially given variable capabilities amongst user groups, restrictions in regional areas etc.

    The focus was very much on the specific issue of Interoperability so MyHR was not discussed in much detail. I left feeling that we had not really achieved much although at least we all seemed to agree on what the problems were...

    ReplyDelete

  6. "I was there on the assumption that we would look at the technical difficulties and constraints. As it turned out, it was taken for granted that technical solutions exist"

    Peter I do not believe that is anywhere near true - does anyone else?

    David.

    ReplyDelete
  7. This is from ADHA's privacy page

    "Information provided via the internet

    The security of any information provided to the Agency via the internet cannot be assured as the internet is not a secure environment. Individuals should protect their own personal information appropriately."

    https://www.digitalhealth.gov.au/policies/privacy

    Was this problem identified and discussed? Is there a technical solution?

    If "... any information provided to the Agency via the internet" applies to myhr, it's a bit of a worry.

    But, of course, you can't trust anything you read on www.digitalhealth.gov.au, the terms of service say so.

    ReplyDelete
  8. If they are looking at technical solutions to the interoperability problem, they are not solving the interoperability problem they are trying to solve the fax problem.

    After over 10 years Health/NEHTA/ADHA still haven't solved anything, they are just making things worse.

    ReplyDelete
  9. two comments:

    "the co-design workshop in Interoperability held last September in Melbourne.. hosted by Ernst & Young ... I assume this paper is the output from that assessment"

    no. As yet, that document has not been released. This is a different document though it might be informed by that consultation to some degree.

    "The paper makes a point about how information is captured into the My Healthrecord system using a document paradigm:"While documents will continue to play an important role in healthcare communication, some data may be better captured in an atomic form."

    I think they are quite confused. The concept of a 'document' (e.g. a discharge summary or a pathology result report) and 'data in atomic form' are NOT mutually exclusive."

    This is a terminology question. Yes, CDA documents can contain atomic data - that's extremely well understood, and not the point of this comment. Perhaps this is loose language from a general point of view (precisely the sort of stuff an early engagement paper might be intended to identify), but atomic data inside a document is not in a data-centric framework where the data can be used to support healthcare process.

    "The problem with the My Health Record system is that it hasn't encouraged or enabled improvements in data quality for those who are still submitting documents in non-atomic form (e.g. in PDF format). It has a mix of documents in atomic and non-atomic form."

    Actually the program has encouraged this quite enthusiastically. It hasn't succeeded in some cases, but this speaks to wider issues than the MyHR itself. However in many cases the documents that include atomic data are still of insufficient consistency or quality to support many of the kinds of use that were anticipated at the start of the program. Though usually the program takes the blame for this, the issues are (again) much wider than that - relating to economics, clinical governance, and the general chaos in the healthcare system.

    Interoperability is a people problem. Standing back and throwing mis-cued criticisms about language and claiming they call everything into question does not demonstrate any understanding of interoperability, and it will make genuine justified criticism almost certain to be ignored.

    ReplyDelete
  10. Agree the interoperability is a people problem, as it is a policy, business and other layers, all requiring sets of agreement and a language so as to identify and understand divergences.

    Just to help is the discussion, I think the consequences should be as important as any wished for benefits

    What should the role of government(s) be in encouraging or mandating interoperability?

    How does/will interoperability or a lack thereof affects competition between providers of services, technologies and health care?

    How interoperability or a lack thereof affects a move towards clouds services?

    How can the age old problems with competition arising when inevitably we start to see dominant suppliers with substantial market power starts to control technology needed to support interoperability.

    How to balance the challenge of “time it takes to achieve interoperability”, as individual firms often innovate more quickly than standards organisations?

    How to best articulate the costs of interoperability (including increased time to market) as compared to the benefits of interoperability (such as an increase in the ease with which consumers can use a product)?

    How do (if at all) technologies for protecting copyrighted works, often called “digital rights management,” affect interoperability?

    What is the role of patented or copyrighted technologies and licensing in achieving interoperability and is there a need for such guidance?

    How best to position the contribution of interoperability to innovation and economic growth?

    Examples of problems consumers face if they are “locked in” to a system that is not interoperable, should this be included?

    How best to identify, articulate and address security and privacy problems related to interoperability.

    How does interoperability increase clinical hazards?

    ReplyDelete
  11. 1:02 PM - good questions. A lifetime of study for multiple people in there (though we have to go with our imperfect state now). Particularly the economic ones are terribly under-studied. I might take up this list on my blog at some stage.

    ReplyDelete
  12. and then there's the drug companies.

    What's in MyHR that makes it of interest to GSK:

    "ELIZA BORRELLO: David Herd is the external affairs director at GSK (GlaxoSmithKline), a drug company which makes medicines and vaccines and is a strong advocate of My Health Record:

    DAVID HERD: We are, as a company, excited by the opportunity that a physician, at any point in time of utilising any of those medicines or any of the vaccines, has the latest piece of information on that patient."

    Privacy concerns about My Health Record system
    http://www.abc.net.au/radio/programs/am/privacy-concerns-about-my-health-record-system/9049856

    GSK are probably so excited about myhr, interoperability and secondary use of myhr data (whatever happened to that consultancy?), they are wetting themselves in anticipation.

    All those questions asked by anon @1:02 PM are highly relevant, so are many others about myhr and other government data grab and link initiatives.

    Will they be answered before the opt-out period starts? Will the answers be available on a trusted website (obviously not ADHA's or myhealthrecord.gov.au), maybe the website you go to in order to opt-out, although I'm not holding my breath.

    ReplyDelete
  13. Grahame, re your comment:
    "This is a terminology question. Yes, CDA documents can contain atomic data - that's extremely well understood, and not the point of this comment. Perhaps this is loose language from a general point of view (precisely the sort of stuff an early engagement paper might be intended to identify), but atomic data inside a document is not in a data-centric framework where the data can be used to support healthcare process."

    Ahhhh now I understand, now that you have semantically translated the 'loose language' and 'ambiguous terminology' to what they really meant to say. Since I work in the health service industry, in Health IT, I too am keen to have some early engagement, and it would be helpful if they explained it a little better, as you have. I can see now that what they meant to say was that interoperability may be helped through use of more defined standardised bundles of data that align with healthcare processes, rather than just clinical documents. Like openEHR archetypes or FHIR resources? Talking about 'atomic data' was irrelevant and misleading.

    Seems like they are working up to a case for a re-design of the My Health Record? They just need to use semantically understood language, so we (the people that need it) can understand the point of their comments.

    ReplyDelete
  14. "Interoperability may be helped through use of more defined standardised bundles of data that align with healthcare processes, rather than just clinical documents. Like openEHR archetypes or FHIR resources? Talking about 'atomic data' was irrelevant and misleading."

    Well, standardised bundles of data doesn't quite capture the idea - need process as well. API would capture that side of it. Orchestration... such a lot is needed. Perhaps subsequent work will not use that language that has become internal short-hand for a whole lot of things.

    "Seems like they are working up to a case for a re-design of the My Health Record?" - well, yes. See http://www.healthintersections.com.au/?p=2765 - we should expect them to be doing preparative work for that discussion. Hopefully it will open soon.

    ReplyDelete
  15. http://www.healthintersections.com.au/?p=2765

    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)

    Are they going to redesign this thing before or after opt-out? If they do it after, then that's just plain fraudulent.

    ReplyDelete
  16. > What kinds of improvements could be made to the MyHR to make it a
    > more flexible platform that are both politically and financially realistic?

    there's plenty. I've got my own ideas, and so have other people. I'm not going to say more than I already have for now (see, say, http://www.healthintersections.com.au/?p=2720). The point is to have some open discussion

    > Are they going to redesign this thing before or after opt-out? If they do
    > it after, then that's just plain fraudulent.

    After. But that's *politics*.

    ReplyDelete
  17. @Grahame, happy for the list to reside in a caring home. I hope these discussions sprinkle a little shared knowledge about the place. I would suggest that they start small and focused. Government has to measure measure measure. They might start with where ‘they’ might want to see interoperability and then work out how to measure it.

    A CIO of a group of hospitals will most likely have a different success criteria than say a jurisdiction CIO or a health minister, certainly different from a care plan team

    ReplyDelete
  18. Colleagues,

    I agree with Grahame's statement at 10:24AM that: "Interoperability is a people problem'.

    Grahame further stated at 3:07PM: "... standardised bundles of data doesn't quite capture the idea - need process as well. API would capture that side of it. Orchestration... such a lot is needed. Perhaps subsequent work will not use that language that has become internal short-hand for a whole lot of things."

    This succinctly captures three critical challenges not presently in the solution framework that is the My Health Record. In saying this I am very mindful that the conversations that we have been allowed into have impaired / hindered the development of a more mature and strategically relevant solution framework bearing on the health care needs of Australians.

    Here are my three points:
    1. The '...standardized bundles of data' doesn't capture the idea because we are talking about flows of information--information that is needed to inform and enable some health care decision or action. There are 'standardized bundles of data' required but the discussion is obstructed because the human context of the flows of these 'standardized bundles of data' is missing in action.

    2. We need process most definitely because health care is about cooperation and collaboration which is organized and purposeful. Think about a Pathology request with its standardized bundle of requested tests. These processes and their enabling information flows have to be worked out together and the working out has huge implications for software vendors--not the least those vendors supporting pathology service providers.

    In regard to issues of process, I am firmly of the view that we need NEW METAPHORS and agree with Grahame that 'Orchestration' is one good metaphor. We need these to enable the conversation at the human level. I equally like the metaphor of 'Choreography'.

    I prefer to use 'Orchestration' when we are dealing with specific information flows whereas Choreography, for me, speaks to the understanding of process--namely, who is doing what and under what conditions, etc.

    The intended choreography establishes the desirability / necessity of particular information flows--and by extension establishes the health care perspective for and assurances sought in regard to 'interoperability'.

    3. Finally, I have trouble with the assumption that the human considerations (clinical, economic, social, legal, governance, etc.) can be reduced to an API and effectively hidden behind the notion that Vendors will use these to assist clinicians do their work.

    This speaks critically to the absence of a mature solution framework.

    APIs have a critical and valuable contribution to make, BUT, restricting our vision to the API level does not address the very real issues Grahame has raised about new language and a lot of work to do at the human level.

    ReplyDelete
  19. Grahame, I was asking about the timing - the question was from your website.
    You're correct about politics, the whole thing is about politics, not health care.

    ReplyDelete
  20. Talking about terminology - in an IT sense "orchestration" and "choreography" have different meanings and uses. In orchestration, there is a conductor who directs the process and mediates the flow of control between participants. By contrast, a choreography involves units working mostly independently and passing tasks back and forth as necessary, albeit usually as part of a pre-defined plan.

    This is relevant because I think it illustrates one of the key issues with ADHA's approach (to date). The Health industry is, by its very nature, a set of very independent people and groups working together. Even where central control exists, its authority is limited (how much can hospital admin dictate to medicos?). This is not how government works, nor most big business.
    Hence, a choreography paradigm is going to be much more appropriate than an orchestration. Federation rather than consolidation.

    ReplyDelete
  21. @AnonymousApril 23, 2018 1:02 PM. These are great discussion points. @Grahame it is true the current state is not perfect. However it might be unwise to ignore various attractors and detractors from the change model. Not including these and other influences will risk any proposed long term investments failing, simply because not all market forces where applied in the modelling but appear in application.

    I would also suggest interoperability is a system issue (human and technological) the people aspect perhaps is the driving need we have to communicate, this has formed many systems over the ages, each human and technology system augmented to expand our need to share and other less savoury aspects of our species

    ReplyDelete
  22. There is a people problem wrt to inter-operability and its that the people charged with trying to make it happen do not have a clue what they are doing and have prevented progress for what is approaching several decades.

    While no technical solution is perfect the level of functionality that would be possible if 20yr old standards were well implemented would be amazing. There has been no effort to actually require compliance with the standards which is the first step in enabling interoperability and a building block for more semantic messaging. Instead we have wasted $2B giving people without any real domain knowledge, either in healthcare or IT to try and invent a "new" solution to problems they barely understand. Surprisingly HealthConnect, NEHTA 1&2 and ADHA have failed in this quest and we have MyEHR which is a good solution for nothing and a poor solution for everything. In the standards world we had a 15+ yr effort to develop HL7 V3, which ultimately failed and now a 5yr history of FHIR which could eventually succeed, but is not a current solution. Meanwhile HL7V2 provides the backbone of everything that works, warts and all.

    Meanwhile highly flawed solutions like tcp/ip, http, html and javascript have improved in their reliability and work cross platform and cross browser without being replaced, but being standardized in an evolutionary way. There were attempts to replace internet technology with "better" things like "MSN" that failed. Evolution remains a powerful force that intelligent design cannot match.

    What would healthcare IT be like if we had people in charge encourage standardization, compliance and clinical modelling using the working format of HL7V2? Today its still the best way to transfer clinical information but the quality of implementations has only progressed in a minor way, but it works.

    In some ways it is a people problem, but that people problem is the hubris that non medical, non technical people can solve a wicked problem if you just throw lots of $$ at it, and without understanding whats working, you replace it with the latest buzzword technology which is unproven in practice.

    ReplyDelete
  23. Dr Ian ColcloughApril 25, 2018 11:08 AM

    A masterly synopsis of the issues Andrew, thank you.

    ReplyDelete
  24. Progress towards interoperability has seen slow progress, dependant on one view of ‘progress’ this can in part be due to some stakeholders unfamiliarity of its inherent complexity, this may also be compounded through a need to reflect it against other industries who on the outside appear to be more mature, however I believe these same people have no formal understanding on just what took place over decades to achieve communication, banking and military interoperability outcomes.
    These same people may happily default to previous efforts having been hampered by overly complex technical standards that were not always fit for purpose, clinical cultures where information sharing was not the default position, vendor commercial models that did not support interoperability blah blah etc, etc…
    It is also clear from the current leadership that historically, those working in healthcare information interoperability have not done a good job of articulating the importance and ultimate outcomes of this vital work, rather than perhaps exposing their own short-comings.
    To many supposedly trying to lead eHealth at a National level, interoperability is seen as a largely technical exercise, focused on data types, data structures and complex standards developed by health informaticians divorced from the realities of front-line healthcare. I guess in much than same way astrophysicists and mathematicians are divorced from rocket science.

    ReplyDelete
  25. Maybe defaulting to interoperability being largely a technical issue is a form of defence, easier to deflect attention than admit interoperability in business, clinical, policy or legal domains is hard

    ReplyDelete
  26. Dr Ian ColcloughApril 26, 2018 9:07 AM

    It bears repeating time and again - interoperability is a people problem and among other things that includes the need for enlightened, informed and competent leadership.

    ReplyDelete
  27. @9:07 AM. .... and it should also be said that the people called politicians and bureaucrats are some of the biggest contributors to the problem.

    ReplyDelete
  28. How is interoperability to work when competing organisations are trying to increase or protect their own marketshare?

    ReplyDelete
  29. @3:32 that is simple, they are all in Washington for the Global Digital Wealth Thing cementing dominate players. One system is the idea of interoperability

    ReplyDelete