The DMs may ask you to delete records from the database so they can load an improved version of the file.
In brief the steps are -
SELECT DATE_LOADED FROM FILES_LOADED WHERE FILE_LABCODE = 'XXXXX' AND FILENAME = 'XXXXX_20160701_20160731_ALERTS.csv'
Copy the value and then do -
DELETE FROM ALERT_RAW WHERE DATE_LOADED = CAST('2020-09-15 08:11:29.000' AS DATETIME2)
The reason for doing it this way, rather than using DATE_OF_ALERT is that sometimes alerts are submitted outside the month. This ensures you've deleted everything that was loaded when that file was loaded.
And finally -
DELETE FROM FILES_LOADED WHERE DATE_LOADED = CAST('2020-09-15 08:11:29.000' AS DATETIME2)
If you don't delete the record from FILES_LOADED then the Validation will stop the replacement file from being loaded.
There is a Jira ticket about adding this as a function in the Validation to save doing it manually. https://jira.renalregistry.nhs.uk/projects/AKI/issues/AKI-125