Journey to Interoperability – Part I


We are on a journey, a transition from health records being recorded on paper to a new paradigm of electronic health records (EHRs) and data interoperability.

We’ve all grown up in an era of paper records – the inability to read doctor’s handwriting, lost records, flood damage and overloaded filing cabinets have been the norm for decades. This is our common experience all over the world.

While they have been in use for much longer, it is only in the last 20 years that electronic health records have slowly been encroaching in everyday clinical practice, accelerating markedly in the last 5 or so. There are some areas where the benefits of electronic records have been a no brainer – for example, ability to generate repeat prescriptions in primary care were a major enabler in the mid 1990’s here in Australia. Yet despite some wins, the transition to EHRs has generally been much slower than we anticipated, much harder than we imagined, and it is not hard to argue that interoperability of granular health data remains frustratingly elusive.

But why? Why is it so hard?

Let’s start to explore this question using this map – paper records represented by the ‘Land of Missing Files’ in the left bottom corner and the ‘Kingdom of Data’ on the remainder of the map.


The ‘Kingdom of Data’ can be further divided into two – the ‘State of Connectivity’ and the ‘State of Interoperability’ which is located on top of the ‘Semantic Cliffs’.

Universally, our eHealth journey commenced in the Land of Missing Files, crossing through the ‘Heights of Optimism’ and dividing into two major paths.


The first path heads north to the ‘Maze of Monolithic Systems’ – the massive clinical solutions designed explicitly to encompass clinical needs across a whole health organisation or region – think Epic or Cerner as examples.  These systems may indeed provide some degree of connected electronic data across many departments as they will all use a common proprietary data model but departments that have different data or functional requirements are effectively marooned and isolated outside the monolith and there is enormous difficulty, time and cost in connecting them. The other harsh reality is that is often extremely difficult to extract data or to share data to the community of care that exists outside the scope of the monoliths influence. Historically the monolith vendors have been notorious for saying ‘if you want interoperability, buy more of my system’. It is likely that this attitude is softening but, due to the sheer size of these systems, any change requires months to years to implement plus huge $$$ .


The second path heads east to the ‘Forest of Solo Silos’. Historically this has been the natural starting point for most clinical system development, resulting in a massive number of focused software applications that have been created to solve a specific clinical purpose by well-meaning vendors but each with its own unique proprietary data model. Each data model has traditionally been regarded as superior to others and thus a commercial advantage to the vendor – the truth is that none of them are likely to be better than another, all built from the sole perspective of the developer alone and usually with limited clinician input.

Historically, our first priority was to simply turn paper health records into electronic ones – capture, storage and display – and we have been successful. However the systems we built were rarely designed with a vision of how we might collectively use this health information for other more innovative purposes such as data exchange, aggregation, analysis and knowledge-based activities such as decision support and research. This is still well entrenched – modern systems these days are still being built as silos with a local, proprietary data models and yet we still wonder why we can’t accurately and safely interoperate with health data.


In order to break through the limitations and challenges imposed by the solo silo and monolith approaches we have collectively trekked onwards into the ‘Swamp of Incremental Innovation’. It is a natural human trait to try to improve on what we have already built by implementing a series of safe incremental steps to extend the status quo. We have become very adept at this – small innovations building on the successful ones that have come before. And the results have been proportional – small improvements that have been glacially slow in development and adoption – and one key factor has been because we have been held back by our historical preference for disparate, closed commercial data models.

The natural consequence of incremental innovation on our journey to interoperability is that we are constantly looking down, looking where we will place our immediate next step, rather than raising our heads to see the looming ‘Semantic Cliffs’. I think that the large majority of vendors are stuck in this swamp, taking one step at a time and without any vision of strategy for the journey ahead. Clinical system purchasers are perpetuating this approach – just look at the number of jurisdictions procuring the monolithic systems. Nearly every other business abandoned this approach decades ago… except health.

We run the risk of becoming permanently stuck in this swamp, moving in never-ending circles or hitting the bottom of the Semantic Cliffs with nowhere to go and drowning in the ‘Quicksand of Despair’.

Beware, my colleagues, here be dragons!

Onward to Part II…


“Smart data, smarter healthcare”

Last week Hugh Leslie & I spent time in Shanghai and Huangzhou at the invitation of Professor Xudong Lu and Professor Huilong Duan. It was a privilege to be invited and particpate in the very first openEHR meetings to be held in China.

It follows on from a surprise discovery of Professor Lu’s work presented at the Medinfo conference in Brazil last year, and Prof Lu attending our openEHR training in Kyoto in Janauary.

Hugh & I presented at two events:

  • a seminar of invited guests and VIPs for the first openEHR meeting to be held in China on April 18 in Shanghai – an introduction to openEHR (Heather) and an overview of openEHR activity in Australia (Hugh); followed by
  • an introduction to openEHR – at the China mHealth & Medical Software conference, Shanghai on April 19

Watch my introduction to openEHR, ‘Smart data, smarter healthcare’ presentation, available via slideshare:

Archetypes: health data bridges

What do we want for our health data – silos of information models for different purposes or ones that bridge multiple use cases?

From a series of emails shared on the HL7 Patient Care email list in the past few days…

Grahame Grieve (FHIR, HL7):

“Heather, you need to keep in mind the difference between FHIR and clinical models: it’s not our business to say not to exchange data that people do have because some user in an edge case might not understand it. We define an exchange standard, not a clinical UI standard…”


“Heather, do not lose sight of the difference between a clinical standard for what care/records should be, and FHIR, which is an IT standard for how care records are.”

My response:

“…I am concerned about developing another standard that you state clearly is only designed for exchange and not for what care records should be. If we are not designing to try to harmonise data requirements for health information exchange, how clinical care records are and how clinical care records should be, then we are building siloes of data structures again, that will require mappings and transforms ad infinitum. I’d hate to see us end up with a standard for exchange that can’t be implemented for persistence”…

If we end up with models for exchange, models representing current data in systems (whether or not they represent good clinical practice and models that are regarded as the roadmap for good data, then what have we got? Three sets of data models that perpetuate the nightmare of non-interoperability.

Our openEHR archetypes are attempting to bridge all of these. Use them in whatever context you choose – messages, document exchange, EHR persistence, CDS, secondary use, aggregation and analysis, querying etc. The ‘secret sauce’ is the use of a second layer of modelling – the template, that allows the correct expression of the archetype appropriate for the context of use.

Mappings and transformations are acceptable where we don’t have any choice, especially with legacy data, but they open us up to vulnerabilities from errors, misinterpretation and ambiguity, concerns re data integrity and possible overt data loss. Given the choice, lets work towards creating high quality data that can be re-used in multiple contexts safely.