How to fix CKM biggest issue

Ian McNicoll ian at freshehr.com
Fri Mar 13 12:52:33 EDT 2015


Hi Marcus,

Glad to see you pitch in!

There isn't really a problem with getting people to participate in reviews,
to help try to build a 'universal use case', or to persuade system builders
and vendors that this is good for them and their customers. I am sure you
are correct that we can do more to foster the kind of community
contribution that Gustavo has advocated but if anything, that may work
against the kind of 'single source of truth' that you would prefer to see.

The reason that there are multiple repositories is not because of any
commercial pressures or Foundation policy - it simply reflects that
national health bodies, in particular, are not yet convinced that the
effort and complexity of working internationally outweighs any
interoperability benefit. This is also a very new way of working for many
people and there is a natural desire to keep some sense of control and to
have the liberty of working to your own policies, in your own timescales
and to align with other national standards.

The current UK-CKM medication models work is a great example of the limits
of real-world interop. These models were built against existing UK models
(non-openEHR) to ensure a close fit with current UK GP messages, primarily
GP2GP, other key work like dm+d and historical dose syntax research. To get
clinicians, informaticians and vendors onside, it was important for them to
work with known, proven and locally implemented concepts. We could
certainly have adapted the international openEHR medication archetypes but
this would have required the UK community to engage at that level and, of
course, forced compromise which I judge right now would have been
unacceptable.

However, many of the other archetypes in the UK-CKM, from which examples of
UK Discharge Summaries, End of Life Summaries templates have been built etc
are actually referenced from the international CKM. i.e there is quite a
bit of

Where the limits of practical working and politics allow, we do make use of
shared resources. I expect this grow over time and for other non-CKM
repositories to appear.

So .. great question but the answer is simply that, right now, for many
good (and some bad reasons) there is simply not enough demand for a single
set of international resources from national standards bodies.

This is not an openEHR phenomenon.. We have seen the same thing happen in
the past with CDA, 13606 etc and will also happen with FHIR.

I actually think (unsurprisingly!) that the openEHR technical stack and
methodology is best placed to support vendors and national bodies to work
in a more unified, common fashion, but they have to want to do so first,
and be prepared to support a key, central editorial resource.

I think we are getting there, but it takes time to get people into their
comfort zone. I will be interested to hear the view from the Norwegian CKM
team, who more than any seem to me to espouse this philosophy, but are
still to some extent doing their own thing, at least for now).

Ian



Dr Ian McNicoll
mobile +44 (0)775 209 7859
office +44 (0)1536 414994
skype: ianmcnicoll
email: ian at freshehr.com
twitter: @ianmcnicoll

Director, freshEHR Clinical Informatics
Director, openEHR Foundation
Director, HANDIHealth CIC
Hon. Senior Research Associate, CHIME, UCL

On 13 March 2015 at 15:22, Marcus Baw <marcusbaw at gmail.com> wrote:

> At the risk of upsetting people, I am going to stick my neck out here in
> support of Gustavo.
>
> It has long puzzled me why a technology like openEHR - which is intended
> to foster sharing of Archetypes for the "universal use case" and trying to
> conquer the massive problem of international interoperability - has so many
> different CKMs, all doing their own thing albeit with some sharing.
>
> I think if the community knew that what it had built was going to be
> shared with the entire world by default (like Wikipedia) then there would
> be more motivation to participate. I have been involved with a couple of
> archetype reviews but if someone asked me where those archetypes are now I
> wouldn't be sure. UK CKM? Scotland CKM? HSCIC CKM?
>
> Apologies if this email gives away how little I know about openEHR or
> about how archetypes are shared in the real world*, but I would second
> Gustavo's call for a single, openly accessible, point of origin for
> canonical archetypes.
>
> * [I am one of a pretty small number of openEHR-trained individuals in the
> UK - so if I still don't get it yet, what's the hope for real
> direct-clinician-generated archetypes covering all of medicine.]
>
> I am attaching a heavy steel flame-proof helmet as I press Send. ;-)
>
> Here to Learn
>
> M
>
>
> On 13 March 2015 at 14:46, Gustavo Bacelar <gbacelar at gmail.com> wrote:
>
>> Dear Heather and everyone,
>> I'm really sorry, but you completely misunderstood the point. I'm not
>> critisizing you or Ian, on the contrary. I've always appreciated your work
>> and I'm a big fan of you both (I'm proud to say it in public). I was not
>> discussing the persons, but the policies. I don't think Ocean is an enemy,
>> never mentioned it.
>>
>> Differently of Wikipedia, where it doesn't matter to have other similar
>> wiki competitors, openEHR must have a single knowledge repository to
>> support semantic interoperability. The knowledge repository of openEHR, be
>> it CKM or not, must take advantage of the community.
>>
>> I agree with you that community is not as active as it should be, but
>> that's just because the current model doesn't help them to. I know you and
>> Ian are overloaded, and I don't blame you, but that's exactly why we need
>> to change the policy. If we want a more active community, we must provide
>> the means to achieve it.
>>
>> You asked me what do I propose and what am I going to do about it. I'm
>> already doing something.
>>
>> I want openEHR to be much bigger. I propose a more liberal approach for
>> CKM governance. I propose openEHR doesn't focus only on National
>> governments and big industry players, but also on startups and small
>> companies.
>>
>> During the last couple of years, I can tell you I've promoted openEHR in
>> Brazil, in Portugal and even in USA. I've presented lots of keynotes and
>> courses free of charge, started an unfunded project for public care,
>> created a website in Portuguese (www.openehrbrasil.com.br), written
>> papers and white papers in Portuguese. More recently, I'm writing a book
>> (an introductory guide) to be distributed for free. All about openEHR with
>> zero resources (and the list is probably missing many things, like
>> ophthalmology archetypes).
>>
>> Kind regards,
>> Gustavo Bacelar
>>
>>
>> 2015-03-13 5:10 GMT+00:00 Heather Leslie <
>> heather.leslie at oceaninformatics.com>:
>>
>>>  Thanks Evelyn,
>>>
>>>
>>>
>>> Even I forget the real roots… We should document it so we don’t lose the
>>> provenance.
>>>
>>>
>>>
>>> Regards
>>>
>>>
>>>
>>> Heather
>>>
>>>
>>>
>>> *From:* openEHR-clinical [mailto:
>>> openehr-clinical-bounces at lists.openehr.org] *On Behalf Of *Evelyn
>>> Hovenga
>>> *Sent:* Friday, 13 March 2015 2:53 PM
>>>
>>> *To:* 'For openEHR clinical discussions'
>>> *Subject:* RE: How to fix CKM biggest issue
>>>
>>>
>>>
>>> Thanks you for this historical overview Heather.  I’d like to add that
>>> the original CKM was developed, maintained and funded by Central Queensland
>>> University.  It was taken over by Ocean Informatics when that University
>>> decided to shut down its entire HI Research Centre at the end of 2007.
>>>
>>>
>>>
>>> Evelyn
>>>
>>> [image: EHE logo tree]
>>>
>>>
>>>
>>> Dr Evelyn J.S.Hovenga, FACS
>>>
>>> CEO & Director
>>>
>>> *   eHealth Education Pty Ltd, RTO 32279*
>>>
>>>   www.ehe.edu.au
>>>    e.hovenga at ehe.edu.au
>>>  Mob. 0408309839
>>>
>>>
>>>
>>> *From:* openEHR-clinical [
>>> mailto:openehr-clinical-bounces at lists.openehr.org
>>> <openehr-clinical-bounces at lists.openehr.org>] *On Behalf Of *Heather
>>> Leslie
>>> *Sent:* Friday, 13 March 2015 1:12 PM
>>> *To:* For openEHR clinical discussions
>>> *Subject:* RE: How to fix CKM biggest issue
>>>
>>>
>>>
>>> Hi Gustavo and the openEHR community,
>>>
>>>
>>>
>>> I’m really sad and disappointed if Gustavo’s opinion is mirrored
>>> elsewhere in the openEHR community.
>>>
>>>
>>>
>>> I’m sure it reflects a frustration with the slow process over past
>>> years. But anyone who has bothered to ask me about how I feel about the
>>> progress will hear that I am much more frustrated than any of you.
>>>
>>>
>>>
>>> We, as the openEHR community really need to do a bit of soul searching.
>>> From my point of view we’ve all been very passive about this modelling
>>> work, all waiting for someone else to do it or take responsibility for it.
>>>
>>>
>>>
>>> The reality is that when Ocean first launched the openEHR CKM, the work
>>> fell to Ocean people. Either Ocean funded it OR Ian and I did the editorial
>>> work in our own time… no other option, and has been the way for years.
>>> Truth is, after a couple of years and getting a couple of hundred
>>> archetypes publicly available on CKM, I was really burned out and unwell.
>>> No-one seemed to notice the effort, to be honest. Certainly no-one seemed
>>> to appreciate it.  I stopped doing the work in my own time and reclaimed my
>>> evenings and weekends. I hoped that there would be a cry of outrage from
>>> the community – “Why has the CKM work stopped?” But no one noticed; no one
>>> said anything, for at least 18 months, possibly more.
>>>
>>>
>>>
>>> This passivity has astounded me.
>>>
>>>
>>>
>>> Over 2 years ago, there was a bit of an epiphany – a special strategic
>>> board meeting was held in London where others were invited, including
>>> myself. The attendees all agreed that one of the highest priorities was to
>>> get archetypes published. I was able to present calculations on how much it
>>> would cost to fund some editorial work to get this happening. Nothing
>>> happened.
>>>
>>>
>>>
>>> Finally, in the second half of last year, the Industry Group has been
>>> able to offer the first funded work to Ian and myself to try to fast track
>>> some archetypes through to publication. This is the first funding that has
>>> been raised in the openEHR community for this critical modelling work ever.
>>> The scope is clearly limited to publishing 69 archetypes. Unfortunately
>>> there was no extra allocated for the extra time required to train or mentor
>>> others to do the work.
>>>
>>>
>>>
>>> The Industry Sprint hasn’t been as fast or as focussed as either Ian or
>>> I would like as we both have ‘day jobs’ that require our attention as well.
>>> However you will have seen a flurry of activity in the past couple of weeks
>>> – 9 archetypes have been refined and sent out for review in the past 10
>>> days. I really appreciate that the Industry Group has collaborated and
>>> committed to this support. And of course it is really exciting that this is
>>> one of the first times we will see potential competitive vendors working
>>> together to get clinical content standardised – breaking down the siloes!
>>>
>>>
>>>
>>> So the situation IS changing…
>>>
>>>
>>>
>>> And in addition, we need to recognise what we do have – an amazing set
>>> of building blocks and an approach to clinician engagement that has not
>>> been emulated in any other domain or standards work. This current openEHR
>>> approach is world-leading and with fairly modest resources we can do a lot
>>> more that needs to be done.
>>>
>>>
>>>
>>> The community has a fantastic problem. As of today we have 1300 users
>>> from 85 countries registered on the openEHR CKM. What a spectacular
>>> resource we have at our finger tips; 381 people have specifically
>>> volunteered to review and 199 to translate archetypes – all through word of
>>> mouth, no advertising. We have a purpose-built tool has been developed and
>>> provided free of charge to the community for over 7 years in order to
>>> manage the library, collaboration and governance of information models use
>>> that. We have only two trained Editors and a handful of others with limited
>>> experience and zero resources committed to managing it. So far it has been
>>> run on the ‘smell of an oily rag’ – not sure how that will translate
>>> outside of Australia – and this needs to change to become sustainable.
>>>
>>>
>>>
>>> From a tooling point of view, CKM has been purpose-designed and
>>> gradually enhanced to do all the things that Gustavo dreams of – projects
>>> and incubators (acting as sandpits for raw archetype development); multiple
>>> roles for reviewers, editors, CKAs have all been there for at least a year;
>>> archetypes can be proposed in the next release of CKM. Community
>>> participation is the focus, and the capability is not currently being
>>> leveraged as it could, and the healthy tension between ‘bottom-up’ and ‘top
>>> down’ can be managed. But the real problem is that there are not enough
>>> people trained as Editors, and no one resourced to manage/govern the work.
>>>
>>>
>>>
>>> Bringing on new Editors is absolutely welcome – both Ian and I are very
>>> keen to share the Editorial and Clinical Knowledge Administrator load more
>>> broadly, because otherwise the CKM work is not sustainable. All this talk
>>> of the community being unable to participate is not actually fair or
>>> reasonable – when I’ve put out a call for Editors we’ve had a few people
>>> volunteer, true. To be honest though, most of those that I have discussed
>>> it with in more detail have then declined when I’ve explained the amount of
>>> commitment or they’ve simply participated in an editorial meeting. For
>>> those remaining, they need training and then ongoing mentoring. But who is
>>> to do this? How is this to be resourced? It absolutely does need to be
>>> resourced appropriately.
>>>
>>>
>>>
>>> By contrast, I have been working under contract with the Norwegian CKM
>>> team recently – they have been resourced to develop archetypes and develop
>>> processes for governance and in many aspects after only one year of
>>> activity they are now more advanced than the openEHR community. We are
>>> working closely with the Norwegian CKM team to ensure that we can develop
>>> processes for collaboration between CKMs. Silje Bakke from the Norwegian
>>> CKM agreed last week to co-edit the Problem/Diagnosis archetype with me and
>>> that archetype was sent out for review last night. other archetypes have
>>> had guest editors involved as well, under Ian and my mentorship.
>>>
>>>
>>>
>>> Key learning: in order for the openEHR work to accelerate, there needs
>>> to be modest financial resources committed to the archetype standardisation
>>> work, beyond the very limited scope of the sprint, and the resources need
>>> to be dedicated, not fitting it in between other work committments.
>>>
>>>
>>>
>>> As an aside, personally,  I’m sick and tired of personally being
>>> considered a ‘blocker’. If only you can imagine how keen I am to upskill
>>> others and share this onerous responsibility with others; of course at the
>>> same time this will ensure that this approach will be sustainable into the
>>> future, and all my work, passion and vision will have been worth it. If I
>>> keep ‘control’, as some choose to view it, then I can be sure that all this
>>> effort will have been in vain.
>>>
>>>
>>>
>>> And I’m thoroughly sick of Ocean involvement being regarded as ‘the
>>> enemy’. I’m not going to address accusations of ‘conflict of interest’ in
>>> this forum – the assumption of huge commercial advantage never gets
>>> balanced by the huge cost of volunteering leadership. Perhaps one day one
>>> of us will write our memoirs… J
>>>
>>>
>>>
>>> Back to the main point again - the community should be rightly feeling
>>> indignant about a lot of things, but rather than complaining or ‘thinking
>>> about it’ we need to be actively doing something about it. We have a new
>>> openEHR Management Board – I hope they will do something about this? But,
>>> also, if you are one of the indignant what are YOU personally going to do
>>> about it?
>>>
>>>
>>>
>>> I’ve done what I can with essentially zero resources, now what do you
>>> propose?…
>>>
>>>
>>>
>>> Regards
>>>
>>>
>>>
>>> Heather
>>>
>>>
>>>
>>>
>>>
>>> *From:* openEHR-clinical [
>>> mailto:openehr-clinical-bounces at lists.openehr.org
>>> <openehr-clinical-bounces at lists.openehr.org>] *On Behalf Of *Gustavo
>>> Bacelar
>>> *Sent:* Friday, 13 March 2015 3:51 AM
>>> *To:* For openEHR clinical discussions
>>> *Subject:* Re: How to fix CKM biggest issue
>>>
>>>
>>>
>>> Hi Ian, Sebastian and everyone,
>>>
>>> on early 2009 Microsoft discontinued its encyclopedia, Encarta. MS
>>> Encarta had a limited selection of *professionally edited content*, but
>>> it was defeated by an initiative of *non-professional edited content*: Wikipedia.
>>> By that time, Wikipedia offered *2.7 million articles* in English,
>>> Encarta had *42,000 entries*.
>>>
>>>
>>>
>>> Encarta did try to adapt, inviting users to submit suggestions for
>>> changes to articles, but those suggestions *first had to be checked by
>>> a member of the Encarta staff*. And Encarta *did not allow users to
>>> submit new entries*.
>>>
>>>
>>>
>>> My point is: openEHR has a *huge potential*, but it is still
>>> too bureaucratic. It must be free to follow its path.
>>>
>>>
>>>
>>> Someone can say: "but the quality of wikipedia is questionable, the
>>> Editors are not professionals!". In 2005, Nature famously reported
>>> <http://www.jimgiles.net/pdfs/InternetEncyclopaedias.pdf> that
>>> Wikipedia articles on scientific topics contained just four errors per
>>> article on average, compared to three errors per article in the online
>>> edition of Encyclopaedia Britannica.
>>>
>>>
>>>
>>> ​I've been spreading the word about openEHR through courses (
>>> http://goo.gl/KvNCvb) and consulting and I can see more and more people
>>> aware of it, but the barriers are not moving. I've tested the beta version
>>> of CKM, thanks Sebastian! It seems to be a very important upgrade,
>>> including the CKA role.
>>>
>>>
>>>
>>> I understand that the focus of the Editorial group is to get green
>>> ticks, but if there were more Editors would be more green ticks as well. By
>>> mid-2012 there was a Call for CKM Editors. There were at least three people
>>> interested: Domingo Liotta, ​Pablo Corradini and I. Nothing happened since
>>> then.
>>>
>>>
>>>
>>> When it comes to using CKM to local projects, I really think it would be
>>> much better for the community. It a local project would like to develop new
>>> archetypes, it would be better to do it within an international context
>>> instead of developing them locally.
>>>
>>>
>>>
>>> It is important to separate the interests of openEHR Foundation from
>>> Ocean's, at his time there in conflict. Ocean wants to sell their products,
>>> I don't blame it, but the international CKM needs to be a central hub for
>>> archetype development. It doesn't matter if its for commercial projects or
>>> not as long as the content:
>>>
>>>    - Is of interest and not repeated (e.g. a local version of an
>>>    existing archetype)
>>>    - Is not a specific admin data for particular use
>>>    - Is available in CKM for community.
>>>
>>>  Many of the existing archetypes in CKM were created to fulfill
>>> commercial use, so it should be used as a source of resources. I will use
>>> the words on openEHR website and openEHR Wiki:
>>>
>>>
>>>
>>> "The openEHR CKM has gathered an active Web 2.0 community (...) for
>>> *sharing* health information between individuals, clinicians and
>>> *organisations*; between applications, and across *regional* and
>>> *national* *borders*."
>>>
>>>
>>>
>>> ​A moral liberal approach would put more load on the Editors, but only
>>> if we don't increase the Editorial team. It is better to have many useful
>>> incubated archetypes ​than not having then in CKM. If these archetypes are
>>> so important, we will be able to see and improve them as soon as possible.
>>> Let's think about Encarta.
>>>
>>>
>>>
>>> Best regards!
>>>
>>> --
>>>
>>> Gustavo Bacelar
>>>
>>> MD + MBA + MSc Med Informatics
>>>
>>> Skype: gustavobacela
>>>
>>> ​r
>>>
>>> LinkedIn: pt.linkedin.com/in/gbacelar
>>>
>>> _______________________________________________
>>> openEHR-clinical mailing list
>>> openEHR-clinical at lists.openehr.org
>>>
>>> http://lists.openehr.org/mailman/listinfo/openehr-clinical_lists.openehr.org
>>>
>>
>>
>>
>> --
>> Gustavo Bacelar
>> MD + MBA + MSc Med Informatics
>> Skype: gustavobacela
>> ​r
>> LinkedIn: pt.linkedin.com/in/gbacelar
>>
>> _______________________________________________
>> openEHR-clinical mailing list
>> openEHR-clinical at lists.openehr.org
>>
>> http://lists.openehr.org/mailman/listinfo/openehr-clinical_lists.openehr.org
>>
>
>
> _______________________________________________
> openEHR-clinical mailing list
> openEHR-clinical at lists.openehr.org
>
> http://lists.openehr.org/mailman/listinfo/openehr-clinical_lists.openehr.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openehr.org/pipermail/openehr-clinical_lists.openehr.org/attachments/20150313/e0fee4e2/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 2251 bytes
Desc: not available
URL: <http://lists.openehr.org/pipermail/openehr-clinical_lists.openehr.org/attachments/20150313/e0fee4e2/attachment.jpg>


More information about the openEHR-clinical mailing list