Skip to content

Latest commit

 

History

History
52 lines (34 loc) · 1.74 KB

File metadata and controls

52 lines (34 loc) · 1.74 KB

Link to CommonDataModel

Remark

The proper etl.sql file is actually in MEASUREMENT/etl.sql

Principle

A specimen is linked to a measurement by the fact_relationship table and vice-versa. (relationship_id "has specimen"). Since MIMIC does not provide that level of information, they are populated with most appropriate information in order to keep OMOP philosophy.

  • for chartevents provide two timestamp : charttime and storetime
    • charttime is the prelevement timestamp and populates specimen_datetime
    • storetime populates measurement_datetime
  • for labevents and microbiologyevents, there is only one timestamp with populates specimen and measurement timestamp

Source Tables (mimic)

  • laboratory chartevents data generates a specimen for each result
  • laboratory data generates a specimen for each result
  • a specimen is generated for each single specimen id

Mapping used

  • it maps d_labitems.fluid to specimen_concept_id
  • it maps microbiologyevents.spec_type_desc to specimen_concept_id

Examples

Distinct specimen_type_concept_id

SELECT distinct specimen_type_concept_id
FROM specimen;

Distinct specimen_concept_id

SELECT distinct specimen_concept_id
FROM specimen;