FHIR Chat · Addiction · implementers

Stream: implementers

Topic: Addiction


view this post on Zulip Jose Costa Teixeira (Mar 24 2020 at 18:09):

Addiction is a Condition, more than than Observation, right?

view this post on Zulip Robin Bosman (Mar 26 2020 at 17:38):

"Social history like tobacco use" is explicitly mentioned in scope of Observation. Could one say e.g addiction to alcohol is an observation but liver chirrhosis is a Condition? Or even: addiction to a certain product is an observation but dependence is a condition?

view this post on Zulip Jose Costa Teixeira (Mar 26 2020 at 17:46):

@Eric Haas @Michelle (Moseman) Miller ?

view this post on Zulip Jose Costa Teixeira (Mar 26 2020 at 17:47):

perhaps "known history of abuse" is observation, "addiction" is a condition?

view this post on Zulip Jose Costa Teixeira (Mar 26 2020 at 17:48):

I prefer to err on the side of not overloading Condition.

view this post on Zulip Lloyd McKenzie (Mar 26 2020 at 17:52):

Observation is a point-in-time statement, Condition is a record with ongoing evolution

view this post on Zulip Robin Bosman (Mar 26 2020 at 17:58):

Not sure if this clarifies but as an example, in SNOMED-CT: nicotine abuse is a 'disorder' but smoker/heavy smoker/ cigarette smoker etc. are all 'finding'

view this post on Zulip Eric Haas (Mar 26 2020 at 23:15):

An addiction is a condition. Information about an addicts usage are Observations.

view this post on Zulip Jose Costa Teixeira (Apr 24 2020 at 13:02):

How would we quantify the addiction?
Suppose we need to say "2 packs of cigarettes per day" or " 5-7 beers"..

view this post on Zulip René Spronk (Apr 24 2020 at 13:45):

As pointed out by Eric: observations (on the condition).

view this post on Zulip Jose Costa Teixeira (May 05 2020 at 14:59):

how do we say "We know this patient's addiction is being treated"?
An extension?

view this post on Zulip Nathan Hall (May 05 2020 at 15:01):

CarePlan.addresses([Condition]) maybe?

view this post on Zulip Jose Costa Teixeira (May 05 2020 at 15:09):

I need 2 things - boolean and the treatment itself

view this post on Zulip Jose Costa Teixeira (May 05 2020 at 15:10):

and I think I need to start from the condition, not the careplan - in other words, i'd need a "Condition.addressedBy" construct... Or perhaps I will have to bundle things together...

view this post on Zulip Lloyd McKenzie (May 05 2020 at 15:27):

Based on FHIR principles, the addressing resource should point to the thing addressed and we try hard to avoid bi-directional links. Is there a reason you couldn't query using _has or _revinclude?

view this post on Zulip Jose Costa Teixeira (May 05 2020 at 16:27):

I can. I am in the design phase, so this means that My logical model will be mapped to 2 resources. I just wanted to be really sure.

view this post on Zulip Jose Costa Teixeira (May 05 2020 at 16:27):

and I still need the extension for the boolean

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 14:14):

1 How do I profile a contained Condition resource? Examples?

  1. How do I profile the "underTreatment" thing? I think a complex extension, with a boolean and a reference to CarePlan - but the CarePlan may or not exist

view this post on Zulip Lloyd McKenzie (Jul 31 2020 at 14:24):

  1. You profile it with a separate profile and indicate that's your target profile of whatever's referencing it and declare the reference to be 'contained'. If you have a contained reference that points to the container, I don't think there's a way
  2. Wouldn't the fact you have a CarePlan that 'addressses' the Condition be sufficient?

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 14:32):

  1. any way I'd see see it in the condition profile?

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 14:34):

I could rely on _has for getting the CarePlan when the plan predates the Addiction report, but when the CarePlan does not exist when reporting the addiction, I need to have something.

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 14:35):

do i make a "AddictionCarePlan" profile?

view this post on Zulip Lloyd McKenzie (Jul 31 2020 at 14:52):

Generally, references point from the thing created second - and management gets created after thing being managed.

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 14:52):

sometimes this is an undocumented management plan

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 14:53):

patient is addicted and say "but i'm in recovery" - and perhaps we do not have a careplan (not documented, not accessible)

view this post on Zulip Lloyd McKenzie (Jul 31 2020 at 14:55):

You could use Condition.category for "managed" vs "unmanaged". @Michelle (Moseman) Miller - other thoughts?

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 16:29):

using condition.category - that would definitely work instead of the boolean.

view this post on Zulip Jose Costa Teixeira (Jul 31 2020 at 16:32):

so the remaining thing is :
can I show in my Addiction profile it will be linked to by a CarePlan? (the care plan may exist or not, and may be remote or contained)

view this post on Zulip Lloyd McKenzie (Jul 31 2020 at 16:39):

Your Condition shouldn't point to your CarePlan. The CarePlan will point to the Condition. You could define an extension with the reverse reference, but it makes REST updating miserable.

view this post on Zulip Michelle (Moseman) Miller (Sep 17 2020 at 14:50):

@Jose Costa Teixeira For what it's worth, Patient Care had a brief discussion last week:

  • Is the condition is managed? While Condition.category is possible, it might be better to log a change request to consider a new element.
  • How the condition is managed? Use CarePlan

view this post on Zulip Jose Costa Teixeira (Sep 17 2020 at 14:56):

Thank you. So "how it is managed" can be a reference (logical or literal). That could work. I'll give it a try and provide an example

view this post on Zulip Brian Postlethwaite (Dec 16 2021 at 04:54):

Was an explicit code for the cigarettes per day identified?

view this post on Zulip Jose Costa Teixeira (Dec 16 2021 at 07:13):

we're considering an addicion quantifier extension (ratio)

view this post on Zulip Jose Costa Teixeira (Dec 16 2021 at 07:13):

this allows cigarettes per day or bottles per week


Last updated: Apr 12 2022 at 19:14 UTC