The new OHDSI Data Quality Dashboard runs checks built directly from the specification of the OMOP Common Data Model. Running it against an OMOP instance built from MIMIC III will give a comprehensive assessment of where results of this ETL deviates from the OMOP spec. Helpfully, it will generate a JSON object that is viewable in an R Shiny app. This app gives summary results and also allows you to drill down to the code implementing all data checks that fail. This includes a human readable description of the check. Checks are at the Table, Variable, and Concept level and are classified into the Kahn harmonized DQ framework categories. Running the SQL code implementing a failed data check will generate the rows in the OMOP instance that fail the check.
This tool is already functional, but is being further developed, e.g. to automatically generate the rows that fail a check. The code is available here: https://github.com/OHDSI/DataQualityDashboard
vojtechhuser
commented
4 years ago
The new OHDSI Data Quality Dashboard runs checks built directly from the specification of the OMOP Common Data Model. Running it against an OMOP instance built from MIMIC III will give a comprehensive assessment of where results of this ETL deviates from the OMOP spec. Helpfully, it will generate a JSON object that is viewable in an R Shiny app. This app gives summary results and also allows you to drill down to the code implementing all data checks that fail. This includes a human readable description of the check. Checks are at the Table, Variable, and Concept level and are classified into the Kahn harmonized DQ framework categories. Running the SQL code implementing a failed data check will generate the rows in the OMOP instance that fail the check. This tool is already functional, but is being further developed, e.g. to automatically generate the rows that fail a check. The code is available here: https://github.com/OHDSI/DataQualityDashboard