FHIR Chat · condition - expected to be risk · implementers

Stream: implementers

Topic: condition - expected to be risk


view this post on Zulip Krishna Moorthi (Sep 13 2021 at 15:23):

Hello all,

I am trying to figure out how to model condition which is expected to be risk for patient & it is not diagnosed yet nor active.
Could you please guide me which clinicalStatus fit for this usecase in STU3?

Thanks for your feedback on this

Regards

view this post on Zulip Lloyd McKenzie (Sep 13 2021 at 16:53):

Typically, you'd us a Condition.code which indicated 'risk of' - and it would be an active condition. I.e. The patient has been diagnosed as "at risk of X"

view this post on Zulip Krishna Moorthi (Sep 14 2021 at 08:26):

Hi @Lloyd McKenzie thanks for the answer. how do we handle the risk condition become an active condition ? Example: Patient A is diagnosed with CKD which is previously a risk candidate.

CKD risk patient upon receiving their CKD diagnosis, we would remove the risk candidate status and replace it with diagnosed (or something comparable) ??

view this post on Zulip Lloyd McKenzie (Sep 15 2021 at 04:21):

There are a few options. You could just update the "risk of" Condition to be the "has it" Condition. (It's completely legitimate to update the Condition.code as the condition evolves and/or the understanding of it evolves.) Alternatively, you could end one condition and create a new one. The best approach is really driven by organizational or individual practitioner preference.

view this post on Zulip Krishna Moorthi (Sep 15 2021 at 08:06):

Thanks for the suggestion. can we update condition clinicalstatus as inactive ? is this considerable approach?

view this post on Zulip Lloyd McKenzie (Sep 15 2021 at 13:59):

I think so.

view this post on Zulip Krishna Moorthi (Sep 16 2021 at 07:10):

thanks

view this post on Zulip Krishna Moorthi (Sep 30 2021 at 12:30):

Hi @Lloyd McKenzie sorry for re-opening the topic. can we also use Risk Assessment resource for risk condition of the patient?
Thanks

view this post on Zulip Lloyd McKenzie (Sep 30 2021 at 14:04):

RiskAssessment is a point-in-time assessment of the probability of occurrence of an event over particular periods of time. If you just want to say "Patient is at risk of X", you can just use Condition.

view this post on Zulip Krishna Moorthi (Sep 30 2021 at 14:12):

There are a several discussions with architects where they are against it to model as condition. Their argument is that standard STU3 Condition scope and background does not specify this usecase. where as Riskassesment was documented as below :

  • A prognosis statement for a particular condition
  • Risk of health outcome (heart attack, particular type of cancer) on the basis of lifestyle factors and/or family history
  • List of potential health risks based on a patient's genetic analysis
  • A prediction of outbreak infection rates within a geography based on immunization rates

view this post on Zulip Krishna Moorthi (Oct 04 2021 at 09:47):

Lloyd McKenzie said:

RiskAssessment is a point-in-time assessment of the probability of occurrence of an event over particular periods of time. If you just want to say "Patient is at risk of X", you can just use Condition.

Thanks @Lloyd McKenzie totally agree.

view this post on Zulip Lloyd McKenzie (Oct 04 2021 at 14:20):

If you'd like the specification to be clarified, you can submit a change request.

view this post on Zulip Krishna Moorthi (Oct 05 2021 at 08:01):

Hi @Lloyd McKenzie I will discuss and explain the context with them again. if not, i will submit change request if necessary.

view this post on Zulip Krishna Moorthi (Oct 05 2021 at 11:36):

Hi @Lloyd McKenzie
Agreed approach is that

  • RiskAssessment.condition element reference would be point to condition resource (ex: "At risk for heart failure").
  • The diagnosed condition would be standalone condition resource (ex.. Heart failure).
    Is there a way to identify history of previously at risk condition become diagnosed condition?

Thanks

view this post on Zulip Lloyd McKenzie (Oct 05 2021 at 13:55):

Condition can have history. You'd have to invoke the history operation to see how the Condition had changed - evolving severity, evolving condition code, evolving stage, etc.

view this post on Zulip Krishna Moorthi (Oct 05 2021 at 14:30):

sorry my mistake. I could have asked the question differently. i will try to provide the flow :
#1. HCP input the condition "At risk for heart failure" -> we store two resources 1. RiskAssessment (points to condition) 2. condition (at risk of HF)
#2. After 6 months, HCP input the condition "Heart failure" as diagnosed -> we store standalone condition resource (heart failure)
In the above#2, standalone entry of condition resource is somehow related to previously identified as risk condition in #1. Is there a way to relate both conditions (Risk vs diagnosed)
thanks

view this post on Zulip Lloyd McKenzie (Oct 05 2021 at 14:40):

There are several candidate extensions - dueTo, occurredFollowing, outcome, related. Which is appropriate depends on the direction of the assertion and how strong an assertion you want to make.

view this post on Zulip Krishna Moorthi (Oct 05 2021 at 15:31):

thanks


Last updated: Apr 12 2022 at 19:14 UTC