OHDSI Home | Forums | Wiki | Github

What information is stored in the procedure table?


(Soyoung Yi) #1

I am confused about the procedure and the observation table.
Are CT, MRI, and x-ray tests stored in procedure tables?
Which table does the pathology test (ex, biopsy), EKG, EGD, etc. are stored in?


(Ron Ribitzky, MD) #2

Beyond the need to solve this problem here and now (sorry I can’t help with this one), the question POTENTIALLY illustrates a much larger, and foundational problem of Semantic Services Framework (SSF).

For example - - is CT (or MRI, x-ray) a ‘Procedure’ (assuming that a ‘Procedure’ may be defined as a concept like the ones expressed in CPT? Or would we be better off modeling these as devices (or methods?..)

Am curious if it has been addressed? If so – would it be worthwhile to start a separate thread?


(Patrick Ryan) #3

From the CDM specs: The PROCEDURE_OCCURRENCE table contains records of
activities or processes ordered by, or carried out by, a healthcare
provider on the patient to have a diagnostic or therapeutic purpose (
http://www.ohdsi.org/web/wiki/doku.php?id=documentation:cdm:procedure_occurrence).
The OBSERVATION table captures clinical facts about a Person obtained in
the context of examination, questioning or a procedure (
http://www.ohdsi.org/web/wiki/doku.php?id=documentation:cdm:observation)

The simple rule to always follow in OMOP CDM v5 and beyond: in your ETL,
you are going to map your source values/codes into standard concepts.
Each standard concept is assigned a domain. The domain tells you the
specific table that your data record should be stored in. So, you don’t
have to (and shouldn’t attempt to) decide whether a code belongs to the
procedure or observation table, you should let the vocabulary guide your
decision. This ensures that everyone across the community employs a
consistent convention of where in the data model each type of data is
placed, and avoid the quite reasonable possibility that two researchers
could subjectively interpret the same data differently. In cases like
this, we’d much rather all be consistently wrong, and inconsistently right,
because we can ensure there is only one unambiguous way to represent each
medical entity, and all researchers can rely on this rule to develop
standardized analytics that can be applied across the community.

So, to some of your examples, you can see MRI is a procedure:
http://www.ohdsi.org/web/atlas/#/concept/4013636

You can see that a ‘plan chest x-ray’ is a procedure (
http://www.ohdsi.org/web/atlas/#/concept/4163872) but having a chest
x-ray documented and reviewed is an observation (
http://www.ohdsi.org/web/atlas/#/concept/2106136)

The act of having a biopsy is a procedure (
http://www.ohdsi.org/web/atlas/#/concept/4311405) but having a bone biopsy
sample is a specimen (http://www.ohdsi.org/web/atlas/#/concept/4205816) and
noting that there was a biopsy interpretation is an observation (
http://www.ohdsi.org/web/atlas/#/concept/4155068).

Conducting a EKG with an exercise test is a procedure (
http://www.ohdsi.org/web/atlas/#/concept/4261794) , but the observing an
abnormal elecrocardiogram is a measurement (
http://www.ohdsi.org/web/atlas/#/concept/320536).


(Ron Ribitzky, MD) #4

Great! Thanks Ryan


(Nitish Kumar Jha) #5

Hi Ryan,

We are doing ETL for an oncology observational study.

In some raw data for example Condition data the start date of condition has been captured as a partial date e.g. Jan/2015 (day is missing) , 2016 (day and month both are missing)

So my first question is should we include this record and If Yes then should we impute the date to make it complete ? is there any date imputation method followed in CDM.

I hope I have explained it well!

Looking forward for your response


(qiongwang) #6

Hi Ryan,
In your exsamples, if a patient had a bone biopsy, there is a record in PROCEDURE_OCCURRENCE table representing the action of biopsy, a record in SPECIMEN table representing the specimen info, and a record in OBSERVATION table representing the biopsy analysis and results, and sometime it may be some records in MEASUREMENT table storing quantitative results.
Do I understand this correctly?


t