Here are the results of the poll.
Do You Expect More Success With An Approach To Health Information Interoperability And Sharing Based On Health Information Networks (US) Or A Centralised My Health Record (AUS)?
The Australian Way 3% (3)
The US Way 43% (46)
Neither 24% (26)
A Mixed Approach 30% (32)
I Have No Idea 1% (1)
Total votes: 108
What a really interesting poll again. The US approach was most approved. I would be very interested to understand what people were thinking with the other responses as to what might work!
Any insights on the poll welcome as a comment, as usual.
A more than reasonable turnout of votes for a holiday shortened week!
It must have been a fairly easy question as only 1/108 readers were not sure what the appropriate answer was.
Again, many, many thanks to all those that voted!
David.
David,
ReplyDeleteI suggest we first recognise that the 'interoperability problem' is a technology-based problem that we have largely inherited from the US Health system through the importation and use of US-based Hospital Information Systems. The document entitled: Death by a Thousand Clicks: Where Electronic Health Records Went Wrong--dated March 18, 2019, provides an interesting evidence base in this regard.
The way forward IMHO is through acknowledgement that we don't wish to continue down the American path and we chose, to borrow a well-known phrase, to 'take the road less travelled'.
The road less travelled would place people, specifically doctors, nurses and allied health clinicians, at the centre of the system with a focus on the timely use of information. We seek a social solution not a technology solution.
It is through the timely flows of information that both inform and enable appropriate health care action that the necessity for shared meaning, e.g. semantics, becomes of paramount importance.
Further, you don't have an 'interoperability problem' if the technology specification is informed by shared meaning and the technology solution is held to account by norms of communications established by the health sector for the health sector.
@John,
ReplyDeleteHow about we put patients and their problems at the centre of the system?
Patients have health problems, not technology problems.
Bernard,
ReplyDeleteYes, patients/citizens should be at the centre of what our health systems should be about.
However, if we are to make real progress on improving the flow of information we need to make the distinction between what might be called the 'internal' information flows--that is, the communications between and among clinicians, and the 'external' communications which are between clinicians and citizens and other stakeholders.
Our first priority has to the internal communications because without these being the subject of a revolution in the quality of collaboration, we will all suffer.
@5:31 PM said the "first priority has to the internal communications". I agree, as getting this right is the essence of effective collaboration and information sharing "between and among clinicians".
ReplyDeleteTwo questions for you John.
1. Was that not the intent and purpose of the Personal Controlled Electronic Health Record and now the My Health Record?
2. Can the My Health Record achieve that goal?
@9:02 PM,
ReplyDeleteI can't comment on the PCEHR and the MyHR has already demonstrated its lack of fitness-for-purpose.
The internal communications of health care are of a completely different order of complexity from the 'external' communications of the health sector. The internal communications have to adhere to more complex norms AND they are evolving as we move care to the community and home. 'Crossing boundaries' represents myriad challenges; challenges often hidden by Electronic Health Records.
In regard to the MyHR, we need to re-think, re-design, and then start building the contribution of digital technology for citizens.
The re-think arguably would focus on helping people make a contribution to improving and maintaining their own health. This isn't about cost shifting or pushing the can down the road.
The re-design would equally help with the necessity of re-telling their story--at least in part because the internal communications of health care are being addressed.
There are substantial asymmetries of knowledge and understanding as well as communication skills to be addressed as part of the re-design.
Finally, we need a higher level of TRUST, higher than currently exists.
We are increasingly operating in a 'world of networks'.
People will need to be convinced that the privacy and security of their personal data can be guaranteed.
IMHO, new sectoral infrastructure will be required to enable the various social networks and services to communicate / exchange information in this world of networks. TRUST has to be a core principle designed for and designed in to the operation.
First and foremost, we need a social solution; not a technology solution.
The experts say our architecture is not scalable!
ReplyDeleteBah! The experts are biased. I want to hear what the non-experts say.
They say we should listen to the experts.
Okay, how about the people who are neither experts nor non-experts?
[Source: Possibly Dilbert]
Can I suggest that we challenge a base (but largely unstated) assumption.
ReplyDeleteThe assumption is that a single system (myhr) meets the needs and requirements of both patients and healthcare practitioners.
IMHO, it does not and cannot.
Patients interested in their health (and that is most certainly not everyone) need more than access to their medical records. There is already plenty of health advice out there, usually ignored by most people.
Healthcare service providers are interested in and need access to far more detailed data than any summary system can contain. They also need help in interpreting it and they certainly do not want to be distracted by having to input data into a system that does nothing for them.
Interoperability is hard enough as it is. Introducing the complexity of myhr is madness. Unfortunately the government sees myhr as the driver of interoperability.
I don't envy those of you working to solve this problem.
Talking about Dilbert...
ReplyDeletehttps://dilbert.com/strip/2019-04-27
and
https://dilbert.com/strip/2019-04-28
@11:39 AM Bernard said "I don't envy those of you working to solve this problem."
ReplyDeleteNeither do I Bernard. In that context what has bothered me for a very long time is how best to make sense of the Gordian knot.
These are just a few of the questions comprising the knot:
- 'Who are the people working to solve this problem?'; 'How do they relate to each other?'; 'Who is leading them?';'What is(are) the problem(s) they are working on solving?'; 'What approach are they using to bring that about?'; 'How are they being funded?'; "What is the(ir) budget?'.
John,
ReplyDeleteI'd add another question:
"Is interoperability the best problem to be working on?"
Is it the best use of the limited money available?
Would it deliver the most value and best outcomes for the money spent?
Hi Bernard,
ReplyDeleteits critical to work on the lowest level of interoperability first, and that starts with character encoding, message structure, escaping of characters and then starts to include display correctness and ack handling.
Unless all of that is working reliably actually making it easy to send data that is badly formed, or displayed incorrectly, or causes system crashes is a huge cause of diagnostic error.
Before you add voltage to an electricity network its good to make sure that the devices plugged into that network can function at that voltage and won't blow up and cause fires. Currently the ADHA want to energize a network where none of the work on whats plugged into it have been done. That is a recipe for widespread explosions.
I think providers are patient centric, they want reliable, fast access to patient data so decisions can be made and problems identified early. Giving patient access to data that providers have not seen makes it difficult to explain results to patients... It may be good to give patient access to their data, but the person who ordered and interprets the result must have higher priority for access.
Or even where should interoperability (as appears to integration) be prioritised?, how then to establish the trajectories of each participant in any given interoperability model, and how then to facilitate and support those journeys, ensuring they remain resilient over time and manage inevitable change.
ReplyDeleteI know a fair bit about the technology problems of large scale interoperability, but next to nothing about the higher layers where content and business processes live.
ReplyDeleteIt seems to me that there are many questions still unanswered, yet the ADHA is charging ahead, developing strategies and solutions.
Has anybody looked at the system wide aspects of interoperability concepts such as:
need-to-know;
consent;
content identification;
access control at the datum level;
transport level security;
content level security
timeliness of the content;
trustworhiness of the data (i.e. it is an hypothesis; observation; test results);
context of the data (i.e. was it gathered in isolation or as part of a co-ordinated set of tests that have relevance as a group);
the health status of the patient at the time (i.e. was it a baseline when the patient was in good health or was the patient suffering from a trauma; illness;
taking medication etc.)
In other words, what is the problem being sloved and what are the requirements of the solution.
Bernard @7:34 AM,
ReplyDeleteBernard, you asked a few questions.
'Interoperability' is not the best problem to be working on. This challenge is a by-product of starting with a technology solution, namely, electronic health/medical records.
The proper place to start is with the people at the centre of the health system--the clinicians.
If you start with people, you can help them to better articulate what it is they are trying to do together and what flows of information they need to support their activities.
The key point is that this is where shared meaning across the clinical team is established; shared meaning means shared SEMANTICS. These are the semantics of information and the semantics of information flows. Think 'Order a Chest X-Ray' for the patient.
Interoperability becomes the problem because we started in the wrong place with the wrong ideas and understanding of how to go forward together.'Closing the barn door after the horse has bolted' sounds about right.
Starting with the people and working collaboratively and cooperatively with them, particularly to remove barriers to information flows, will delivery results the like of which we have not seen.
Further, in regard to the points raised by Andrew McIntyre @12:17 PM, these would be second or third order challenges --if we had started in the right place with the right approach.
These fall out as resolvable because they are needed to ensure that we have assurance of dependable results. Again, you decide that you want a 220 volt electricity system or a 110 volt system at the design stage.
In regard to Andrew's final comment regarding providers being patient-centric. I agree.
However, we need to recognise that health care is 'TEAM-effort', whether formally acknowledged or not.
There are a range of clinical issues, visible or not, for each provider in the patient's care team that have to be addressed in order for each provider to have ready access to patient data to inform and enable clinical decision making and subsequent action. For example, who has the authority to say that this paediatric patient is actually an ACUTE, UNSTABLE ASTHMATIC patient?
The Respiratory Specialist is responsible for the designation of ACUTE. What we are seeking is meaningful, accurate, and trustworthy information to inform and guide action when and where it is needed.
Returning to Bernard's questions @2:22 PM, the issues identified are better addressed at the human (social) level because they are intimately related to TRUST and resolution and assurance of Norms for the interchange of personally identifiable information. This includes information flows related to vulnerability, threat and risk.
You need to start with the people and have an architecture that enables these issues to both have a home and a method/mechanism of resolution.
@John Scott: "'Interoperability' is not the best problem to be working on" - I very much disagree, and I think it's important.
ReplyDeleteOur mantra in the FHIR team is that "interoperability is a people problem" - so while I agree with you about where to start, this is still about interoperability - just interoperability between people. It's a misnomer that is making things harder when people dismiss interoperability as some technical challenge - our problem is getting people to proiritise working together across the board, whether they are clinicians, policy makers, vendors, payers, etc.
You might regard this is a snit about which word to use, but if we can't get solid semantic definitions about the problem, how are we going to get a solution?
There certainly is a people problem, or to be more accurate a "bureaucracy" problem thanks to the "help" we have received from government.
ReplyDeleteA small group of users and technical people did create solutions in the 1990s and that is what HL7 V2 is all about. Despite a lot of neglect since then it is still capable of solving the clinical problems I see and in fact is solving many of them every day. My issue is moving from the base level to good decision support, the problem is not that it can't be done, just that the quality and compliance of the data is way below what V2 could support.
FHIR is probably appropriate for new things, but if we want to see real progress in my working life then HL7V2 is the path that covers 95% of whats needed to have good eHealth with decision support. Whats really needed beyond some actual technical correctness is some decent terminology and clinical models, but its pointless to even try with this without message compliance.
Starting again, is the mistake made by HealthConnect, Nehta I and II and well the ADHA just don't have a clue. Fixing whats in actual use and having a culture of high quality correct implementations is a good starting block for any system, and we do not have that culture. Insisting on it in HL7V2, which is in widespread use and very useful is the cultural change that is needed now, well 20 years ago in reality, but its urgent now.
@Graham Grieve 7:51 PM
ReplyDeleteGraham, we are in agreement that the current 'interoperability' problem is important. However, I was asked was it the most important problem to be working on. My answer was NO.
In regard to language, I view 'interoperability' as technology language; it is not the language of clinicians. For them, information flows to inform and enable work flow resonates.
In regard to getting people to prioritize working together across the board I agree that it is a major challenge. We have not put the 'quality of collaboration' as our metric of performance on the design table. This has to change.
This would go some way to addressing the issues raised by Andrew @10:02 PM.
In the process we could also begin to address the issues of TRUST, COMPETENCE and CAPABILITY.
In regard to the role and contribution of government, it needs help to determine where it can lead, to understand where it should collaboration through new structures, and to act in a responsible and responsive fashion.
IMHO the crises we face in health care are not being solved by the present myopic focus on technology. They are equally not helped by proxy 'collaboration' entities that do not and cannot REFLECT the diverse interests within our modern health care system. It is time for a new idea and a new approach. All the rest is simply marking time.
and well the ADHA just don't have a clue.
ReplyDeleteThat is a most concerning observation and not one I can find an argument against. In three-years we still have no strategic framework in guiding how finders, vendors, operators etc.. can build infrastructures platforms and applications that enable various standards (V2, CDA, FHIR and the various clinical medical and administrative terminologies) to co-exist and then how to build trajectories towards some other chosen future state.
If ADHA cannot produce such guidance then how on earth can the sector move forward. Yes it is a people problem, but let us recognise that we as a collective rely heavily on information communication technologies to communicate and share information, interoperability efforts should not seperate and isolate the concerns but rather recognise the sets of layered agreements and establish a language where we can identify divergences, clearly and without ambiguity.
Looking back at early Australian interoperability efforts the Interoperability Framework has a very reasonable definition -
ReplyDelete‘Interoperability is the continued ability of an organisation (or a system) to use or offer business (or technical) services from or to another organisation (or system) and accordingly, exchange information (or data) with other organisations (or systems) to achieve a specified purpose in a given context.’
The rest of the work is pretty good but I admit some areas are intended for wiser persons than myself. What worries me is given all this collective knowledge provided to ADHA why they seem to be at a loss and if nothing else adding to the confusion around what is a well understood problem state which for me is that Healthcare is complex and fragmented. A long standing challenge for healthcare (digital or otherwise) is to recognise and enable opportunities for managed data flow while maintaining the integrity and context of use for that data thus ensuring safe and effective exchange of information semantics.
Yes it is a people problem, the problem being the people best suited to provide the rest of us a steady hand and work through these problems seem to be kept apart, banished in some cases but certainly not being utilised to the nations advantage.
Maybe the ADHA does not meet these people’s basic standards. ADHA leadership fails to listen to understand.
ReplyDelete