I was alerted to this a little while ago.
Semantically Aligned Design Principles At Core of Australian Electronic Health Records Platform
By Jennifer Zaino on July 30, 2014 9:13 AM
At the upcoming Semantic Technology & Business Conference in San Jose, Dr. Terry Roach, principal of CAPSICUM Business Architects, and Dr. Dean Allemang, principal consultant at Working Ontologist, will host a session on A Semantic Model for an Electronic Health Record (EHR). It will focus on Australia’s electronic-Health-As-A-Service (eHaas) national platform for personal electronic health records, provided by the CAPSICUM semantic framework for strategically aligned business architectures.
Roach and Allemang participated in an email interview with The Semantic Web Blog to preview the topic:
The Semantic Web Blog: Can you put the work you are doing on the semantic EHR model in context: How does what Australia is doing with its semantic framework compare with how other countries are approaching EHRs and healthcare information exchange?
Roach and Allemang: The eHaaS project that we have been working on has been an initiative of Telstra, a large, traditional telecommunications provider in Australia. Its Telstra Health division, which is focused on health-related software investments, for the past two years has embarked on a set of strategic investments in the electronic health space. Since early 2013 it has acquired and/or established strategic partnerships with a number of local and international healthcare software providers ranging from hospital information systems [to] mobile health applications [to] remote patient monitoring systems to personal health records [to] integration platforms and health analytics suites.
At the core of these investments is a strategy to develop a platform that captures and maintains diverse health-related interactions in a consolidated lifetime health record for individuals. The eHaaS platform facilitates interoperability and integration of several health service components over a common secure authentication service, data model, infrastructure, and platform. Starting from a base of stand-alone, vertical applications that manage fragmented information across the health spectrum, the eHaaS platform will establish an integrated, continuously improving, shared healthcare data platform that will aggregate information from a number of vertical applications, as well as an external gateway for standards-based eHealth messages, to present a unified picture of an individual’s health care profile and history.
The Telstra Health eHaaS initiative represents one of the largest and most ambitious non-government investments ever, in developing a national electronic health record. Well aware of the history of failed attempts by government-sponsored national health record schemes, Telstra’s approach is to provide an integrated healthcare platform that will offer contextualized access, sharing, and analytics to individuals, health care providers, devices and industry decision support and research. By 2020 Telstra will invest US$1B in a mix of six core capabilities including provider applications, health gateways, telehealth, care coordination, enabling solutions and business analytics.
A key challenge for this approach is the conflict between the integrated view of data in the healthcare platform and existing data that is already being used by the service and care providers across the industry. To alleviate this tension, eHaaS leverages the flexibility of RDF and other semantic technologies to manage distributed and varying data schemas. This fundamental commitment to semantically aligned design principles is a unique characteristic of the eHaaS platform, setting it apart from earlier (failed) efforts. The core information, messaging and process models supporting the eHaaS have been designed and built from the ground up on a semantic business architecture framework (CAPSICUM) and is optimized for interoperability across the major international health information standards.
The Semantic Web Blog: Can you provide a little more detail on just what the CAPSICUM semantic meta model is, and how its features are critical to the business architecture ontology that underlies the eHaas platform?
Roach and Allemang: The centerpiece of the eHaaS platform is a Common Information Model (CIM) that will be deployed on a number of separate MongoDB NoSQL databases. An over-riding concern with information privacy has driven the database architecture which has a unique characteristic of maintaining separate repositories for clinical data, demographic information and analytical data-stores. The CAPSICUM framework is used to manage a single view of this CIM, and deploy it out to the various operational data systems on an as-needed basis.
The design of the eHaas CIM has been based on the CAPSICUM Framework for semantically aligned business architecture. CAPSICUM follows a model-driven engineering approach with three modelling layers, namely:
- a computationally independent, conceptual model (CAPSICUM Business View) that is modelled in RDF/OWL;
- a platform independent, logical model (CAPSICUM Technical View) based on an entity-relation modelling paradigm;
- a platform specific, physical modelling layer (CAPSICUM Platform View) which is a repository of XSD-based application schemas.
Models at the three modelling layers are aligned through an integrated meta-model which provides a single consistent semantic representation, which is serialized on-demand into any of the three modelling paradigms. The framework supports round-trip transformation of model semantics across all three modelling layers so that, for example, an XSD schema can be imported into the Platform View of the framework and rendered graphically as both an Entity-Relation-Diagram in the Technical View and an Ontological Graph in the Business View. Analysts can then make modifications to the models at any of the layers and have these applied to the each of the models at the other two layers. Additionally the models can be serialized in the ISO11179 metadata standard.
Lots more here:
The guts of what we read here seems to me:
1. Telstra is developing a national EHR.
2. Telstra plans to invest $1 billion in this over the next few years.
3. Telstra is saying they are developing a semantically interoperable Common Information Model (CIM).
4. The scope of the eHealth as a Service (eHaas) is said to be “By 2020 Telstra will invest US$1B in a mix of six core capabilities including provider applications, health gateways, telehealth, care coordination, enabling solutions and business analytics.”
My first reaction is just to wonder what is the business model Telstra has in mind to get a return on their $1B - given that Telstra is not a charity and has a lot of shareholders who will continue to demand dividends etc.
Possibilities might be that they take over from the Government in running the PCEHR and charging the Government a fee for operations and for providing a better products that are presently in place. Alternatively or additionally they might choose to charge practitioners to provide systems for them or State Governments to run / deliver hospital systems. They may also be thinking in the Asia / Pacific Region. Any other ideas?
My other reaction is what is said regarding the Common Information Model (CIM) sounds very ambitious and rather ahead of what I understand to be the ‘state of the art’. There is a long history of such initiatives taking a very long time to reach their objective - if ever!
I look forward to comments on what others think Telstra are up to and where it fits in the Australian e-Health echosystem!
David.
9 comments:
David, its seems like more of the same, half-baked, software/informatics-will-automatically-fix-healthcare thinking that saw the NHS run enthusiastically toward and then off the eHealth cliff 10 years ago. Indeed, on the basis of its audacity and grandiosity, I wouldn't be surprised if Granger was lurking.
Their public pronouncements have been vacuous at best, and farcical at worst. A recent PulseIT article from ITAC quoting their spokespeople as saying “Telstra plans to ditch the pilot and roll out the jumbo” seems to summarise their strategy of hoping a 747 will take off without a pilot on board.
Senior people from within suggest the idea was born of a thought bubble from someone high up in the organisation along the lines of "Surely with all that money sloshing around in healthcare, why aren’t we taking our piece?". The concept was sold to the board on the back of a flashy management consulting document that made sense on PowerPoint, but has since been discarded when tested against reality.
I'm told there is some expectation that their acquisitions will be expected to generate revenue in the near future, but they haven't been able to lure any serious Australian eHealth players into the fold (e.g. GP software developers) and none of their back-of-envelope-assessed hodgepodge acquisitions fulfil that end in a way that justifies the salaries of their very top-heavy executive made up almost exclusively of former bureaucrats.
If they were buying these businesses to acquire customers, it might have been sensible for them to have a product ready for market first.
I'd be disappointed if I were a TLS shareholder.
Lest we forget. By 2020 Telstra will invest $US1 Billion in a mix of six core eHealth capabilities.
The CAPSICUM Framework is a proprietary meta-model (US patent pending) for the design of an ontology for business architecture. It was designed and validated through a 5 year post graduate research programme at the University of New South Wales in Sydney, Australia.
Move over NEHTA The Government's eHealth exit strategy is revealed.
eHealth-as-a-Service (eHaaS) : towards universal stakeholder engagement
Black, Alofi Shane & Sahama, Tony R. (2014) eHealth-as-a-Service (eHaaS) : towards universal stakeholder engagement. e-Health Technical Committee Newsletter, 3(1), pp. 1-3.
http://eprints.qut.edu.au/72739/
Fast moving, slick and supremely confident - trademarking eHaaS - eHealth as a Service.
While I am sure Telstra would not be quite as brain dead as DOHA its still a minefield that can consume large players and I am not sure the currently stated directions and recent purchases fill me with confidence that they know what they are doing. Its clear that throwing money at the problem is not a solution. Mayne health's well funded venture into private hospitals springs to mind as an example of how the attempt might progress if they are not careful.
Shane Solomon is in charge of Telstra's Star Ship eHealth exploring the eHealth Galaxy, going where no man has gone before.
David,
re "Telstra is developing a national EHR."
That's not my reading of the situation. I suggest a more correct statement would be: "Telstra is developing an eHealth data transport infrastructure."
My conclusion is based on this statement "The eHaaS platform facilitates interoperability and integration of several health service components over a common secure authentication service, data model, infrastructure, and platform."
Nothing I have been able to find in any of the links has anything on eHealth data itself or the use of that data.
The Telstra initiative is exactly the same approach as the PCEHR - "collect as much health data as you can, throw it at health care professionals without any regard to its value, quality, accuracy, timeliness or usefulness and let them sort out what to do with it".
IMHO, unless you understand health data itself and the best way to manage, treat and use it, all initiatives are like throwing mud at a wall and hoping some will stick - a risky way to spend money, and more than likely to fail.
It all seems a bit academic, but I do see what they are proposing to do with the eHAAS. It is something that the PCEHR cannot do: i.e. provide semantically integrated views and reporting across the data that is submitted to it from health service providers. A ‘consolidated view’ was proposed in the PCEHR Concept of Operations; however when the PCEHR was designed and implemented, they realised that a safe consolidated view was not possible when each clinical document defines its data differently according to its own source and context. Thus we now have a PCEHR that is merely another collection of documents (some of which may not be there).
The concept of a health enterprise information model and framework that would allow for data from a variety of sources to be semantically combined, to allow for meaningful views and reporting has been around for a long time (at state and national levels), but there has been little progress. No doubt there are more sophisticated technologies and tools available now that could be used, and this seems to be the focus of the eHAAS.
If the PCEHR serves any good purpose then perhaps it will prove the need to do it differently next time.
Post a Comment