Grahame Grieve posted CIMI at the Crossroads recently. I can't disagree with a lot of the content, but maybe I'm a bit more of an optimist as I draw some slightly different conclusions. Grahame is totally right about what it has achieved so far:
- a significant membership roll that has never been achieved before
- a significant agreement of an initial approach to clinical models - a primary formalism of ADL 1.5/AOM with a commitment to support transformation to isosemantic UML models in a spirit of inclusivity and harmonisation.
And as he points out, the notion that the modelling methodology was chosen independently of the Reference Model is somewhat disconcerting.
"...the decision to choose ADL/AOM as the methodology, while deferring the choice of reference model. While I understood the political reality of this decision, choosing an existing methodology (ADL/AOM) but not the openEHR reference model committed CIMI to building at least a new tooling chain, a new community, and possibly a new reference model.
The cost of this is high; so high that the opportunity created by the foundation of CIMI may likely founder if we see another attempt to reinvent the health IT wheel, yet again.
There are many opinions, and everyone at the CIMI table has their own bias, history, experience. Organisational and personal investment in each existing solutions are high. No one wants to throw away their efforts and 'start again'; everyone wants their work to be the successful and sustained.
The CIMI community do need to make an objective decision if it is to move forward. It may not be result which wins a popularity contest. It is very likely that some members will walks away and keep working as they always have; maybe intending to return when a more mature solution is on offer.
In his paragraph on the pros and cons of openEHR, Grahame very eloquently states:
This is the first choice: pick the least worst established clinical modelling paradigm.
"Least worst" - Thanks Grahame! You could turn that around: the 'best' available so far, where there is no perfect solution!
But it's not a bad principle - to take the least worst and make it better!
The chair of the openEHR board, Sam Heard proposed the following to the openEHR community back in October 2011:
“If the CIMI group chooses to use ADL as the formalism then the openEHR community is prepared to explore the Foundation governance arrangements with the CIMI group and align the two efforts using the structures that are mutually agreed.
Changes to ADL and the openEHR Reference Model may be part of the process to meet the collective needs, and alignment of the shared RM and a reviewed RM for ISO 13606 would also be a major goal. ADL 1.5 would be submitted to ISO as part of this alignment.”
Seems sensible to me - start with a robust candidate and modify/enhance it to meet the collective needs. The latest version of the openEHR RM is clearly one candidate. It has evolved significantly from the 2005 version which forms the basis of ISO 13606. Given that ISO 13606 (parts 1-5) is due for revision this year, perhaps we have a great opportunity for harmonisation. The openEHR community is already starting to develop a proposal for the revision, but a greater achievement would be to align all of these efforts into a new 13606/openEHR/CIMI specification.
This is a difficult task that we are trying to solve. We know that because it has not been solved before.
This is definitely not the first crossroad that CIMI has encountered - don't underestimate the effort that has brought the group to this point - and it will definitely not be the last. What will determine success is keeping the end goal front and centre in CIMI's decision-making; cutting ruthlessly through the political and personal agendas; putting pragmatism ahead of perfection; and a willingness to compromise in order to move forward.
It may not be possible. It could be a hell of a ride. I still think it has the potential make a hell of a difference.