Display number of prior screeners in dual screening#432

When adjudicating dual screening, I sometimes see 3 results. My guess is that two staff members were screening at the same time and NK allowed both to become Screener #2. Screenshot attached.

Good observation - is your team screening with inspector, or using sequential screening (i.e. using the big “Screening” menu link)?

Inspector is not collaboration safe. Since you, the user, can filter to whatever records you want by design, we can’t guarantee that another user isn’t also modifying that record.

Sequential screening does guarantee that only one person is looking at a record at a time and that a record will only be screened 2 times (in dual screening, & only once if the nest is configured for single screening). This guarantee extends to the situation where 3 team members access sequential screening at the exact same moment - all 3 will recieve different studies. Once a study is screened twice, it is removed from the queue.

For this particular study, I see the 3 times of screening were 1-2 hours apart, suggesting that the last screener added their decision via inspector, or possibly a very stale browser tab.

It’s also possible we have a bug, although unlikely, as this is a high volume area of the site. We are approaching 100k reviewer-level dual screening decisions, and triplicates have only happened 90 times across all nests. Our development team has also been working in this part of the code recently and everything appears in order.

Perhaps we should add a UI tip to inspector, when a study already has 2 reviewer-level decisions? That may eliminate this confusion entirely.

(Side note - what extension are you using for dark mode? That looks suprisingly nice, and we’d love to share with other users e.g. #174)

Our team does sequential screening. These triplicates are pretty rare on our end and it seems like it’s rare across all of NK so I’d file this under low-priority!

The dark mode extension is https://darkreader.org/

It’s the first thing I install on a fresh browser :)

OK - I’m going to co-opt this issue to simply display the number (only the number) of reviewers that have already screened a record, when viewing a record in dual screening, sequential or inspector. That would solve this problem, if caused by using inspector for screening, or make it obvious if we have a bug in our sequential study queueing.