ST0010 As TL I would like to improve the work_started report columns so that more studies get an actual date rather than NULL so that the report is more useful.

[TWJW-SANGER]

As TL I would like to improve the work_started report columns so that more studies get an actual date rather than NULL so that the report is more useful.

Acceptance Criteria

• [ ] The majority of reported studies have non-null values in the workstarted* columns.

Background Currently the work_started columns are based on the 'Working Dilution - Plate Reader v1.0' qc_result - as a proxy for a "work started" event. Unfortunately this does not seem to be common or filled in for the majority of studies / pipelines.

We could investigate using options such as:

• Expanding the number of qc_result types we are looking at (hopefully simple)
• Looking for the earliest point from a set of events including the qc_result types and mlwh_event types
• Looking the earliest mlwh_event timestamp for a sample that is NOT manifest.created AND NOT manifest.updated
• Something else...

[khelwood]

Rename the work_started to indicate that it refers to "working dilution". Add column(s) checking for "order made" event. Check how that looks.