Wisdom of the Crowds

Clinical (archive)
1

Hi All,

[This posting is related to Tony Shannon's great email regarding the
future of openEHR in 2009. I think some of it relates to the technical
list members as well so I have CC'd them.]

Not wanting to hi-jack Tony's thread I started this one where I hope I
have a few positive and possibly helpful suggestions. Some of them may
need funding to help protect some people's time and others are virtually
cost free.

I do have one negative comment and I'll get that out of the way first.
As I have indicated, I believe that I will have more success working on
some of the fringe areas of healthcare with openEHR. Therefore I tend
to talk to providers that are not in the main; primary care/family
medicine/general practice areas. I have been told by more than one of
these folks that they didn't feel very welcome to participate on this
(Clinical) list on issues that concerned their areas. Whether it was
lack of feedback on questions or actual comments about currently
focusing on archetypes for more general medicine. A bit more
consciousness about welcoming new people might be in order. :slight_smile:

Now for a few positive suggestions for the group and for individuals.
Some of you may already be doing some of these things but in my
experience in building open source teams they have helped me.

1. There are more than 450 members registered on this (Clinical) list.
With a few more than that on the technical list. Many are duplicates
(like myself). I would guess that at least 75% of the Clinical list
members have downloaded and tinkered with one of the archetype editors.
Probably created a few and then said; Now what?

No software to use them on (hopefully OSHIP will soon help with that) no
place for peer review and feedback. I suggest a section on the SVN
server labeled 'community' with the correct folder structure underneath
like the other areas. A group of the experts should receive an email
each time a commit is made to this section. One of the experts then
provides some kind of feed back on that commit. Maybe some of them are
good enough to be moved into the CKM for consideration? Maybe the
experts can provide enough feedback that these early community
committers get better. It is clearly true that Sam, Heather, the NHS
group and a handful of others cannot possibly build all the archetypes
needed. Sure, you'll get a lot of junk archetypes to sort through in
the beginning. You'll also need to spend more time in education but
there are a lot of resources on the website and wiki that you can point
to. But people like to participate in something meaningful. If they
enjoy it, they'll tell a friend. It shouldn't be too difficult to setup
a web page to show people when to get an SVN client along with a name
and email registration space where they can be sent a SVN password
automatically. Open this are up to the world. If it gets completely
out of control then change the rules or shut it down. Right now there
is no way to encourage "the crowd" to participate and share their
wisdom.

2. When you go to meetings and conferences. Do not hang out with
openEHR people. Meet new professionals and have a 15-30 sec comment
about how we are turning over the data design of healthcare applications
to the healthcare providers. Give them the URLs to get an editor and to
the community SVN website along with the mailing list info. Do not try
to explain openEHR or even archetypes to them at that point. Even if
they ask; give them a little more info and encourage them to join the
community. Leave them wanting to learn more.

3. Post comments on blog articles and healthcare related sites/online
magazines. Try one of these: http://www.hitsphere.com/

4. Prepare a guest blog entry. In fact two of those on the above site
have asked me and are waiting for me to prepare guest postings on
openEHR for their sites. Most of these guys WELCOME contributed content
that is of interest to their readers.

5. When you see stuff that is blatantly bull$$%$%^ on blogs and online
magazines, do not hesitate to say so. If you really believe in what you
are saying and doing then let people know. Certainly people like David
Kibbe have no problem with saying that CCR is the greatest thing to
happen to healthcare while at the same time thinking that openEHR is an
open source EMR project. Don't be afraid to put your ideas and
convictions out there. It usually only takes a few minutes. If we
spend all of our time discussing openEHR related matters on these lists
then we are only "preaching to the choir" and not recruiting new people
with new ideas.

Well, that's my top five. I hope they help promote and expand the
community.

Cheers,
Tim

(attachments)

oetrick.png

2

Good one - I am gonna be at HIMSS Asia Pac at the end of the month and will take your advice (although I am speaking on a ‘clinical perspective on health care standards’ and manage to spend a lot of my time talking about openEHR.) We have certainly noticed an increase in discussion at many conferences about openEHR…

3

Hi Tim,

Thanks for your posts and never-ending enthusiasm.

I have responded inline re your comments on archetype collaboration wrt CKM.

Regards

Heather

Tim Cook wrote:

Hi All,

[This posting is related to Tony Shannon's great email regarding the
future of openEHR in 2009. I think some of it relates to the technical
list members as well so I have CC'd them.]

Not wanting to hi-jack Tony's thread I started this one where I hope I
have a few positive and possibly helpful suggestions. Some of them may
need funding to help protect some people's time and others are virtually
cost free.

I do have one negative comment and I'll get that out of the way first.
As I have indicated, I believe that I will have more success working on
some of the fringe areas of healthcare with openEHR. Therefore I tend
to talk to providers that are not in the main; primary care/family
medicine/general practice areas. I have been told by more than one of
these folks that they didn't feel very welcome to participate on this
(Clinical) list on issues that concerned their areas. Whether it was
lack of feedback on questions or actual comments about currently
focusing on archetypes for more general medicine. A bit more
consciousness about welcoming new people might be in order. :slight_smile:

Now for a few positive suggestions for the group and for individuals.
Some of you may already be doing some of these things but in my
experience in building open source teams they have helped me.

1. There are more than 450 members registered on this (Clinical) list.
With a few more than that on the technical list. Many are duplicates
(like myself). I would guess that at least 75% of the Clinical list
members have downloaded and tinkered with one of the archetype editors.
Probably created a few and then said; Now what?
  

This is absolutely a major issue - and one of the key motivators for
CKM. For those not up to speed on the acronyms - Clinical Knowledge
Manager - found at www.openehr.org/knowledge. Further description can
be found at
http://www.openehr.org/wiki/display/healthmod/Clinical+Knowledge+Manager

In the first instance this is an archetype repository - with a
significant number of 'reasonably sound' archetypes uploaded in draft
status. (Reasonably sound as we know that people are starting to use
these archetypes in their draft form, pre-publication). The scope of
CKM will expand further to embrace other knowledge artefacts.

The CKM development to date has focused on 3 main functions:

    * asset management so that we know exactly what archetype is what,
      and can track all changes - revisions and versioning etc.

    * supporting the archetype publication lifecycle. It is true that
      we have just one archetype that has gone through the formal
      team/peer review process to be formally published (a temperature
      observation) and there are 4 well into the review process and
      another about to kick off. Publication occurs at the completion
      of content agreement.
      Archetype publication hasn't ended up being as rapid a process as
      I would have liked - there have been issues with limited numbers
      of reviewers, software refinement, etc. We have made significant
      process in refining and streamlining the usability and team review
      collaboration and we have had a group of about 35 clinicians from
      around the world participating actively to date. We actually
      can't review much faster or I fear that we will 'wear out the
      goodwill' of those currently involved - so while still in this
      beta phase, recruitment has largely been word of mouth, and now a
      few more are engaging through these lists this dynamic is likely
      to change.

4

Thanks Tim and Heather,

For encouraging me out of lurking mode.

I am a clinician (internist physician) and would like to contribute positively to your efforts.

On my last few months of subscribing to your email list I haven’t been able to figure out how
but with today’s email I realize their may be a way forward.

Hope to hear more on this positive note.

rakesh
http://peoplesgroup.academia.edu/RakeshBiswas

5

Dear Tim,
Following on Heathers email, I only wanted to stress the importance of bullet number 1: “after creating archetypes, now what?”
But I fear that my “now what?” is rather different.
I may not be completely up-to-speed, but I would say that the software released in openEHR, to date, does not allow to manage actual clinical data which adheres to these hundreds of archetypes available in the repository. I mean making persistent clinical data, which adheres to those archetypes, through the openEHR implementation. The persistent layer does not exist yet, or am I mistaken?
A few months ago I was working on the java implementation of openEHR, and I exchanged a few emails with Rong.
For people to be able to test and be interested in using openEHR, or another “two-level modeling” paradigm implementation for that matter, they need to be able to see it, and without the persistence layer, they can not see something actually somewhat usable (I’m sure it´s very useful, it´s just not usable right now).
A very simple “hello world” example, showing the whole life cycle of a very, very simple EHR is essential, I believe. If it has been created over that last few months and I missed, please correct me.
Best regards,

Jesús Bisbal

(attachments)

cilab.jpg
ciber.jpg

6

Hi everyone

I'd like to encourage Rakesh and other clinicians to self-register in
CKM and volunteer to review archetypes!!!! - Instructions here:
http://www.openehr.org/wiki/display/healthmod/Registration+in+CKM

No openEHR experience is necessary as primarily we need a broad range of
clinical input to make sure that the clinical content is correct.
Technically oriented openEHR people are also on the review teams as well
to provide guidance on design and implementation issues, so there are no
unrealistic expectations of you. Contributions of clinical and technical
nature are equally and gratefully received.

Please 'adopt' the archetypes that you would like to be involved in -
http://www.openehr.org/wiki/display/healthmod/Adopt+an+archetype - that
way you will definitely be invited to participate in your archetype's
review. And at other times you may be invited to become involved in a
review where we consider that we need your skills to balance out the
current team of adopters.

For each archetype we are seeking a range of views - from a variety of
professions, and not just limited to doctors and nurses; from a variety
of geographical locations, to make sure we can capture diverse clinical
and cultural practice; from a variety of health domains, so that all
use-cases can be part of the archetype's maximal dataset.

While we strive for the maximal dataset within the archetype, we are
pragmatic and realistic and know that we won't get it 100% right on the
first go. However I would go as far as to suggest that a small group of
clinicians with complementary skills and expertise can create and
develop an archetype and get it to about 80-85% complete. Review by a
team of clinicians from a range of professions, countries, institutions,
research, and health domains will contribute and refine the archetype
further - maybe this still will only get it to 90% complete; but maybe
more. Over time it will be interesting to see how the models evolve - no
doubt a good research topic!

Having agreed archetypes in this manner, even if in retrospect we find
they are only 90% complete, is a major step forward and all the flow on
benefits that come from using a shared set of clinical specifications
for EHRs can potentially be great. The capability to further review and
refine the archetypes and managing this is also part of the governance
process - being documented and refined as we write.

So all clinicians are welcome to get involved in CKM - we will certainly
set you to work very quickly! We expect that by contributing domain
expertise and insights, clinicians will also benefit personally by
gradually developing openEHR understanding and expertise as part of the
experience.

And then of course, there is also the contribution to the good of
mankind... :wink:

Kind Regards

Heather

Rakesh Biswas wrote:

7

Jesus,
You have hit the nail on the head. What one needs is a solution. Something, as follows, is what most of us are looking for:

  1. Download the exe, zip or rar file
  2. unRAR or unzip and execute it
  3. App runs and opens a help file.
  4. Help file takes you thru the steps of set up users and permissions
  5. Set up a few users load some patient data and get productive

Much later…

  1. Take time out to read through the tutorials to tinker with the program to write clinical pathways, modify programming logic and the UI.

Now that is what I would love.

With warm regards,

Dr D Lavanian
MBBS,MD
Certified HL7 Specialist
Member- American Medical Informatics Association
Member- HIMSS
Senior Consultant and Domain Expert - Healthcare Informatics and TeleHealth
Former Vice President - Healthcare Products, Bilcare Ltd
Former Vice President - Software Division, AxSys Healthtech Ltd
Former Co-convener Sub committee on Standards , Governmental Task force for Telemedicine
Former Vice President - Telemedicine (Technical), Apollo Hospitals Group
Former Deputy Director Medical Services, Indian Air Force
Mobile: +91-9970921266

8

Great post Heather.

This is my idea #1 on steroids. GREAT job. Now we just needd to
promote it to the larger clinical community.

Cheers,
Tim

9

This is where you lobby your government to actually put some funds where
it would help :wink:

Dr Lavanian wrote:

10

Hi Thomas,
You are right there, government funds may be a way forward.
Of course, requesting funds to add a basic persistence layer to OpenEHR will not be very appealing to any funding agency nowadays. But anyway, we could try to sell the idea some other way, and hopefully get that layer as a by product… and we are on to it.

But that does not mean that an “open source community” like OpenEHR should not consider as a priority to release a simple version of a working prototype for people to experiment with. No matter how simple and low performance (and keep the fancy, fully-fledge, and high performance implementations for sale… Ocean, etc).
Having a “reference implementation” which is a lot of code that no one, besides the contributor, can actually test without a very, very significant amount of programming effort, I´m not sure it’s the best way to promote an “open source community”, or even the OpenEHR platform.

Lobby our governments is not a substitute for an open source community to have a simple (no matter how simple, I insist) working prototype as their priority. They are different things, really.
That´s my humble opinion, I could of course be completely mistaken.
All the best,

Jesús

Thomas Beale wrote, on 03/02/2009 13:25: