Stream: implementers
Topic: observation or clinicalimpression for emergency triage
Jens Villadsen (Dec 13 2018 at 14:05):
I'm looking into profiling emergency triage (more specifically a triage conducted in the ambulance) and so far I've looked towards a profile on Observation. I had a quick dive into SOAP notes trying to understand the 'A' which should be reflected in ClinicalImpression. The question is if an emergency triage is best modelled using ClinicalImpression or Observation. Any clinical/technical advice is much appreciated
Michel Rutten (Dec 13 2018 at 15:01):
(deleted)
Lloyd McKenzie (Dec 13 2018 at 15:10):
SOAP is still a matter of some debate. @Michelle M Miller ?
Johannes Faber (Dec 13 2018 at 15:29):
What exactly are you going to model regarding the emergency triage? From my understanding the resulting triage level can best be modeled as an Observation. We are currently evaluating using Questionnaire for the triage assessment as a basis for generating the triage level.
Jens Villadsen (Dec 13 2018 at 15:30):
Yes ... the questionnaire was also in my mind
Jens Villadsen (Dec 13 2018 at 15:31):
in terms of questionnaire it would be the questionnaireresponse that would be in focus
Lloyd McKenzie (Dec 13 2018 at 15:36):
QuestionnaireResponse is a useful way of doing data capture, but for data analysis, you'll want to move the data into Observations, MedicationStatements, AllergyIntolerance, etc.
Johannes Faber (Dec 13 2018 at 16:00):
Our current idea is to use questionnaire to model the possible choices of a triage system, e.g., the manchester triage system.
Using questionnaireresponse we'll then store the choices of a triage nurse during an assessment, whereas the resulting triage level will be stored as observation.
I'm not sure though that questionnaire is actually suited to model the major aspects of a system like manchester triage (MTS), because MTS also defines certain clinical checks based on vital signs. For instance, such a check can be something like 'if the temperature of a patient is higher than 41 centigrade, then suggest the triage nurse to select Temperature Very High in the questionnaire'. This certainly can be added as an extension to the questionnaire, but I'm not sure if this is overstretching the questionnaire too much. I'd be very interested if anyone has better ideas how to realize something like that.
Jens Villadsen (Dec 13 2018 at 21:50):
Thats sort of the same for my case. We're up against modelling DEPT which is sort of like ADAPT, which in turn is sort of like CTAS and ATS (https://sjtrem.biomedcentral.com/articles/10.1186/s13049-016-0312-6) - but basically @Johannes Faber it sounds like we are sort of stuck with the same problem. Interesting though that you actually plan to use the QuestionnaireResponse for capturing the all the choices made by the clinician.
Last updated: Apr 12 2022 at 19:14 UTC
