Data Quality-Handling Missing Birthdate Components in plausibleAfterBirth Check

General
1

Hi everyone,

I am using Data Quality Dashboard R package for one CDM data. to assess one of our CDM datasets.

After running the script, I encountered the following error related to the 'plausibleAfterBirth' check: ‘Error:net.snowflake.client.jdbc.SnowflakeSQLException: Can’t parse ‘00101’ as date with format ‘YYYYMMDD’’

I think the issue is our data only have information in YEAR_OF_BIRTH but MONTH_OF_BIRTH, DAY_OF_BIRTH and BIRTH_DATETIME are null.

Would it be possible to modify the code associated with the 'plausibleAfterBirth' check to handle this case?

Thanks

2

Hi there,

Could you please file your bug report in GitHub here? GitHub · Where software is built

In the issue, please include the full rendered SQL query that threw this error.

Thanks so much.

3

Thank you for the comment. Looks like the issue is due to a null birth year in the source data, so we put year_of_birth value = 0 for CDM conversion. What would be an appropriate default value to use when the birth year is missing?

4

Birth year is so crucial that often lack of it is a reason to exclude the person from the dataset.

5

Please see the THEMIS convention for guidance on this topic: How to populate year_of_birth when it is missing from the source | THEMIS Convention Repository for the OMOP CDM

1 Like