[2022 US Symposium] Conical Set of CONCEPT_IDs for Vital Signs

[ericaVoss]

TYPE	NOTES
ITEM	Could we implement a canonical set of CONCEPT_IDs for highly used vital signs (height, weigh, BMI, BP, heart rate, respiratory rate, SPO2, temperature, etc.)?
FORUM POST	https://forums.ohdsi.org/t/2022-us-symposium-67-conical-set-of-concept-ids-for-vital-signs/17502
SOLUTION	TBD
NEXT STEPS	TBD

[sebastiaan101]

For this canonical set of concept_ids for vital signs we may look at codes that are often used in EHR systems - or seek to align with (future) manadatory codes in different countries. See for instance the FHIR vital signs Implementation Guide: https://build.fhir.org/ig/HL7/cimi-vital-signs/ Or the NHS discussion/choice: https://digital.nhs.uk/developer/guides-and-documentation/legacy-fhir-profiles/vital-signs-profiles Those two takes on the Snomed versus LOINC discussion. Another choice being made is to use a Snomed concepts to represent/codify the measuring procedure (e.g., 46973005 for measuring the BP) of a vital sign and LOINC concepts (like 85354-9 for Blood pressure) to represent the resulting measurement.

Apart from the choice of preferred concept_ids for specific vital signs, we also need to adress questions like (non-exhaustive, initial draft):

• which vital signs are to be included in the canonical set? Are the ones in the FHIR IG sufficient or do we need more/lees in OMOP-CDM?
• what are the conventions for representing more detailed measurements (specific blood panels, measurement with specific devices/procedures/...) in OMOP-CDM?
• can we or should we say something about conventions for representing less commonly used vital signs?
• what are conventions or potential issues for analists - e.g., if we would define standardized "common vital sign cohorts", what are the caveats in using these?

[MelaniePhilofsky]

Step 1: Themis provides a list of "preferred" concept_ids for the vital signs found in original post Step 2: Destandardize and map to the chosen concept_ids