コホートの定義から実際にデータを抜いてくるまでの手順について

OHDSI in Japan
1

みなさま
お世話になっております。
基本的なことかもしれなくて恐縮ですが、
Book of OHDSIの4.2.2 General Conventions of Schemas
のことで教えてください！

分析を行うときにはおそらくcohort_definitionテーブルで分析対象の定義を行い、処理を行うと対象のコホートがcohortにリストされる感じなのでしょうか。。？
subject_idとperson_idは関連があるのでしょうか。
10章も読んているのですがよくわからなくて。。

よくわかっていないので、変な質問だと思いますがよろしくお願いします。

2

Let me answer to this question in English.

First of all, there is no stupid question in OHDSI.

‘Cohort’ table is a kind of ‘result’ table from the analysis, which is not usually generated during ETL process.
Usually, OHDSI ecosystem generates a cohort table for each analysis (or research project).

The cohort definition of OHDSI is: ‘A cohort is a set of persons who satisfy one or more inclusion criteria (a phenotype) for a duration of time’.

Subject_id of cohort is usually identical to the person_id of PERSON table. Cohort_definition_id represents unique identifier for the cohort. And cohort_start/end_date represents a duration of time for the cohort.

I hope this answer is helpful for you. @somacells

1 Like
3

Dear @SCYou,

Thank you for your information!

I crear about Cohort table is generated from analytical process.

Let me ask an additional question.
I would like to separate the ID used in the analysis from the ID used in the practice.
In this case, when I derive the data from the EHR to the omop system, do I convert the ID of the clinical practice to the ID for analysis in person_id and store it?
Actually, I understood person_ID to be patient IDs used in hospital practice, so I thought the ID would change when deriving subject_id from person_id.

4

@somacells
Indeed, in most institutions, we assign an unique new ID to each patient for OMOP-CDM during conversion (or ETL), and not using EHR’s patient identifier to minimize security concern.

1 Like
5

@SCYou
Thank you very much for your reply.
I will try to assign an another new ID during conversion or ETL.