FHIR Chat · checklist for profilers · implementers

Stream: implementers

Topic: checklist for profilers


view this post on Zulip René Spronk (Jan 25 2021 at 11:42):

t has been suggested by one of the trainees who attended our FHIR training course, that we publish a 'checklist for PROFILERS (not: users of profiles, but authors of profiles and implementation guides)'.
Feel free to make suggestions (key issues, high level), we'll create a curated list, which will be made available to the community as a whole.

view this post on Zulip René Spronk (Jan 25 2021 at 11:43):

Choose a tool for FHIR profiling

view this post on Zulip René Spronk (Jan 25 2021 at 11:43):

decide: open or closed model approach

view this post on Zulip René Spronk (Jan 25 2021 at 11:43):

what are your requirements for the IGs? Choice of publication tool

view this post on Zulip Mareike Przysucha (Jan 25 2021 at 12:24):

I once wrote a "guide" when asked in the German stream. I'll try to translate as good as possible.

view this post on Zulip Mareike Przysucha (Jan 25 2021 at 12:47):

Questions to answer, when I want to create one or more profiles:

  • What is my use case?
  • What are my data and information and how are they connected to each other?
  • Which resources do I need?
  • Is there anything on national or international level which I can use?
  • How about terminology?
  • In how far do I have to adjust the resources so that they fit to my data and data structures?

There are certainly more questions to ask, these are probably the first ones.

What is my use case?
This question sometimes sound trivial, as you have your use case in mind. Nonetheless it might not hurt to write it down. Do I want to send a document? Do I want to define an interface and write down / communicate, what it (should) support? In which sector am I: ambulatory, stationary, intersectoral, interdisciplinary, physician, nursing, ...? Do I define local, regional, national, international?

What are my data and information and how are they connected to each other?
It is also helpful to visualize what you want to represent. In a discussion with others, you might find one thing or another which does not fit or is missing. This discussion might be open or closed to only a few members within the same organization. This depends on the use case (see above).

Which resources do I need?
In order to answer this question, you should have a look at the FHIR resources. It might also help to look into the FHIR-Chat Zulip (also in international streams), whether there was a similar question or discussion before.

Is there anything on national or international level which I can use?
In our efforts, interoperability on national or even international level should play an important role. Thus national (and international) profiles should play an important role. [Hint to the German base profiles, I skip the part here...] On international level there are standard extensions, which consider single additional data fields, but do not fit the 80%-Rule. For similar use cases these international extensions should be preferred. It might also happen, that a similar use case was dealt with at another place, so you can get inspiration from there. There is the FHIR registry or the FHIR Implementation Guide registry, but also a search on Simplifier might help.

How about terminology
Depending on the use case, terminology can play a mayor or minor role. Fact is: You can't ignore that topic. For Conditions or Observations, you need a code to specify them: What is the condition about? What was observed? For this you should use (inter)national standards. The terminology to use again depends on the use case and the data you want to represent.

In how far do I have to adjust the resources so that they fit to my data and data structures?
This is by far the most time-consuming question: The profiling. For starting there is the Profiling Academy. Here the most important points are explained. As a tool, Forge is used, the resulting profiles might be published on Simplifier. [Addition: But there are also additional tools]. Based on personal experience I can say: The better you first points (use case, data and relationships) are elaborated, the easier is the profiling, as you know exactly, what you want to do.

view this post on Zulip Mareike Przysucha (Jan 25 2021 at 12:48):

I know, along answer and not really a "checklist", but it may help starting.

view this post on Zulip René Spronk (Jan 25 2021 at 12:59):

Ah yes, i did read that checklist in the German stream - thanks for reposting in English.

view this post on Zulip Mareike Przysucha (Jan 25 2021 at 13:45):

No problem. I hope it's understandable and it fits. Anyone who's more fluent in English is welcome improve the text.

view this post on Zulip Lloyd McKenzie (Jan 25 2021 at 15:39):

The notion of 'closed' profiles should generally be avoided as it makes interoperability much harder over the long term for the community. The IG Guidance implementation guide was intended to be a repository of such instructions and recommendations, but it hasn't been touched in quite a while. Pull requests welcome...

view this post on Zulip Lin Zhang (Jan 26 2021 at 04:36):

Valueable:+1:

view this post on Zulip Ardon Toonstra (Jan 26 2021 at 11:12):

Nice work @Mareike Przysucha !

view this post on Zulip Grahame Grieve (Feb 02 2021 at 21:40):

I wrote a check list for profile/ig reviewers. Not sure what happened to it..

view this post on Zulip Grahame Grieve (Feb 02 2021 at 21:41):

ahh: FHIR-IG-Review-Checklist.docx

view this post on Zulip Grahame Grieve (Feb 02 2021 at 21:41):

it's something...

view this post on Zulip Gay Dolin (Feb 03 2021 at 17:11):

Grahame's word doc list could be integrated into http://build.fhir.org/ig/FHIR/ig-guidance/index.html

view this post on Zulip Lloyd McKenzie (Feb 03 2021 at 17:49):

And some content from @Mareike Przysucha's document too. Any volunteers? :)


Last updated: Apr 12 2022 at 19:14 UTC