Stream: genomics
Topic: task recommendation follow up
Larry Babb (Nov 06 2019 at 02:10):
The new 0.4 version of the TaskRecFollowup profile within the GenomicsReport seems to have a non-extensible required 1..1 code which is bound to an answer list that has 3 options
Genetic counseling recommended | Confirmatory testing recommended | Additional testing recommended
I do think the enhancements from the previous 0.3 version are an improvement, but it seems to be too limiting now.
In my first example report I need to code the following two recommendations...
It is recommended that correlation of these findings with the clinical phenotype be performed. Genetic counseling for the patient and at-risk family members is recommended.
Should we be constraining the codes?
Or am I able to use the code.text option instead of the bound answer list to apply the first recommendation.
Ideally I would like to have it be something like this
task 1
. status . requested
. intent . proposed
. code.text . "It is recommended that correlation of these findings with the clinical phenotype be performed"
. description . "It is recommended that correlation of these findings with the clinical phenotype be performed.task2
. status . requested
. intent . proposed
. code.system http://loinc.org
. code.code . LA14020-4
. code.display . "Genetic counseling recommended"
. description . "Genetic counseling for the patient and at-risk family members is recommended."
Any help would be appreciated.
Kevin Power (Nov 06 2019 at 14:51):
With a 'required' binding as we have today, I don't think your first task instance is valid. I could see making the binding strength 'extensible' which would allow what you have above.
Or would the correlation task also be performed by a genetic counselor? If so, perhaps it would make sense to deliver this as a single task? I really do not know, but wondering.
Larry Babb (Nov 06 2019 at 16:04):
i do not know either. but i think it is safe to say that we would likely not be able to come up with a "required" list, regardless of how many experts we convened. making it extensible is much more reasonable, unless the author of this concept had some very specific plans which are not apparent.
the other option is to forego making the "code" required (change it to 0..1) so that when folks want/need to simply communicate the recommendation as a descriptions string or paragraph, they can do so.
Kevin Power (Nov 06 2019 at 18:06):
Sounds like making extensible is the way to go here - do you mind logging a tracker?
Larry Babb (Nov 06 2019 at 20:41):
just added #25187
Bret H (Nov 18 2019 at 17:40):
just a generic word of caution. In Hl7v2, much of the genetic report came as comments. If the recommendation is actionable maybe consider looking at the profiles from the clinical decision support space. of course, having the genetic variant observation discreetly is a huge step. use your judgement, I'm just suggesting caution in general.
Last updated: Apr 12 2022 at 19:14 UTC
