OHDSI Home | Forums | Wiki | Github

Concept for ECG reading of QTc interval


Which is the most appropriate concept to use to represent an electrocardiogram reading of the corrected QT interval. I have come across the following:

4116637 QT interval duration Observation SNOMED Observable Entity S 301117000 1970-01-01 2099-12-31
4216826 QT interval feature Observation SNOMED Observable Entity S 81435004 1970-01-01 2099-12-31
4273023 QT interval - finding Measurement SNOMED Clinical Finding S 365417009 1970-01-01 2099-12-31
40298284 QT interval Observation SNOMED Observable Entity 142127008 1970-01-01 2013-07-31 U


4273023, because it’s a Measurement. But I would use the LOINC Concepts.

Hello Christian,

Happy 2018! Thanks for the reply. How about the concept to represent ECG reading of PQ interval?



You mean the P-R interval? Not sure P-Q is measured.

You are right, PQ and PR intervals are the same.
So I would use 3007794 P-R Interval

Hi, I am setting up OMOP mapping to ECG concepts for both measurements and diagnosis statements. This is part of incorporating device data like ECG into our OMOP research warehouse. I’ve done mapping for GE12SL, GLASGOW, and Philips 0B/0C. ECG measurements are largely LOINC codes as others on this thread indicated. Other measurements and diagnosis statements not covered by standard OMOP concepts are going to be compiled and submitted as new ECG vocabulary.

I would appreciate any thoughts or suggestions on this topic and/or understand what use cases are out there.

1 Like

Great, @mkwong. Bring it in. Let us know if you need help.

Use cases: There is a strong need for EKG information for QT prolongations and Torsades de pointes as drug side effects. That’s a big deal for the pharma industry and FDA.

Hi Christian - that is my plan is to get a draft vocabulary put together covering the top 3 ECG device manufacturers in the US - part standard OMOP mapping, part new concepts (common across all vendors), and the remainder as custom device/vendor specific vocabularies. Then have your vocabulary team review and suggest revisions.


I meant P-Q interval. Thanks for the confirmation Anna, as most of the literature I came across stated that the P-Q and P-R intervals are generally the same. Its just that the 2 readings are available in the source data and are in different ranges.


As @Christian_Reich and @mkwong discussed in this thread, we need convention list of concept_ids for measurement. (For ECG and echocardiography and so on). I also plan to convert ECG data into CDM, and I don’t want to use different concept_ids from @vjacob’s.

@Christian_Reich, do you know where we can store these information?
I cannot enter the THEMIS WG I google doc currently (I don’t know why. But possible reason is that I’m in Singapore now.).


That would be THEMIS Focus Group 1. Ping @mvanzandt.