OHDSI Home | Forums | Wiki | Github

Observation_type_concept_id would you recommend to

Vocabulary Users
#1

Which observation_type_concept_id would you recommend if the provenance of the data is

  1. app generated
    or
  2. manually entered from the paper test form?
    I could not find one that I can used from the exisiting type_concept_id list. Wondering if you can create a new one that I can use.
#2

Thank you for your question @stephanieshong . If the provenance of data is manually entered from the paper test form, you may want to consider using “Information gathering from manual entry” (Athena) if it aligns well with the characteristics of your data. In order to create a new observation_type_concept_id for ‘app generated’ origin of data, kindly submit this as a community contribution by creating a GitHub issue here, and by checking our Community Contribution Guidelines here. Thank you.

1 Like
#3

Thank you I can use “Information gathering from manual entry”. As for the other one, I will submit a request for the app generated data type concept id.

#4

@janice - One quick question, this concept is not type concept class.
Athena Any issue for using this concept id as the type concept ?

#5

@m-khitrun @krfeeney - could you share your thoughts on my question above?
we are trying to map montreal cognitive assessment - place the assessment data into measurement domain.
Montreal Cognitive assessment data can come to us in two different paths; the data could come to us as part of the app generated data or it can also come to us in a paper form where we would have to then manually enter the data.
So we want to specify the provenance of the data source, how the data was entered. I did not see the proper type concept id that I can use. However, I do see the following two concepts in Athena: Athena and Athena.
My question is can I use these to concepts as the type_concept_id for the measurement domain?

#6

@Christian_Reich ^^ - could you share your thoughts on my question above.

#7

Sorry, this thing totally slipped my attention and not sure it is still relevant. But as @janice pointed out, you should add those SNOMED ones, or your own ones, into the community contribution machinery. They could be made children of Patient self-report.

#8

@stephanieshong, @janice, @Christian_Reich and cc- @Andrew, @Polina_Talapova

This is a great question.

I wonder if we can start a more formal discussion on this one with the broader community (including OMOP Psychiatry working group and the Survey working group)?

Question re:

When you say “app generated” do you mean the data comes via a patients’ smart device, an iPad used in the waiting room, computer based app used in clinic, or a third party app, or other?

Some additional information and background:

Here is a link to information about some work the SNOMED Mental and Behavioral Health group has organized to put together a formal proposal for work that we hope will make assessment-related concept models in SNOMED-CT more fit-for-purpose for use in EHRs and CDMS like OMOP. The models include those for assessment scores (observable entity concepts), assessment procedures (procedure concepts), and the assessments themselves (staging and scales concepts). If we get these models right, the relationships we need will automatically be available for use in the OMOP vocabulary files.

Our last meeting included representatives from LOINC, SNOMED International (of course), leads from several of the SNOMED International Working Groups, and members of the National Release Centers (and in some cases the ministries of health) of 6 different countries. There is a clear need for better representation of these data.

I personally am very keen to see how we might use the OMOP vocabulary tables to implement those parts of the models needed by the OHDSI community to meet our needs until the content and relationships are available via the source vocabularies (SNOMED, LOINC).

Piper

t