Thread by @chrischirp

SHORT THREAD: about Covid testing, system strain and potential fall out from Immensa scandal on getting PCR tests...

Am highlighting stuff from @ProfColinDavis , @fascinatorfun , @cfinnecy, @samueljlovett & @Gabriel_Pogrund

TLDR: it's not just Immensa, system is straining 1/11

When you do a PCR test, the most common result is either a positive or negative. But some tests return a "void" result, which can mean a bad sample or bad testing process.
Or an "unknown" result where the test just gets lost in the system 2/11

https://twitter.com/ProfColinDavis/status/1452285350651605000?s=20

@fascinatorfun has pointed out that voids are more common if testing capacity is under strain - rushed tests, cutting corners on equipment maintenance or procedures etc can lead to more voids.

Tests also more likely to be lost in a straining system 3/11

We know from @Gabriel_Pogrund that there are problems in the supply chain that means labs cannot process all the tests they are being sent - many samples will never get analysed (i.e. become "unknown"). 4/11

https://twitter.com/Gabriel_Pogrund/status/1305047753983225858?s=20

We know from @samueljlovett that poorly supported novice workers at Immensa were pressured to run tests quickly on poorly maintained equipment - likely to result in more voids. 5/11

independent.co.uk/news/health/co…

So if we plot the proportion of unknown/voided PCR tests over time, we could infer that higher unknown/void rates correspond to more system stress.

So that what's @ProfColinDavis did and I basically copied his method to do the same 6/11

The red lines represent the range in which 80% of the proportion of unknown/voided tests lie which were taken between Oct 2020 and end Aug 2021.

The strain in winter peak is obv - but since Sept voids have been climbing out of the "typical zone" & are now atypically v high 7/11

@ProfColinDavis and @OliasDave have shown that this is NOT just a SW problem.
And since voids remain high this latest week (after Immensa shut down), it's *not* just an Immensa issue either. Fitting in with @Gabriel_Pogrund's report of larger supply issues. 8/11

Finally - if you get a positive lateral flow test (LFD) you have to isolate - unless you get a negative PCR test. And govt guidance is to get a confirmatory PCR test.

So on the whole, people who get a positive LFD test *do* get a confirmatory PCR test.

BUT 9/11

This most recent week (to 20 Oct), @cfinnecy noticed that the % of positive LFD tests *not* matched to a confirmatory PCR test has

It's just an idea but this could be a consequence of the Immensa false neg scandal breaking on 15 Oct and people losing faith in the system? 10/11

Will this be just a blip? What does it mean for interpreting case numbers? Not sure yet...

But certainly I think there we urgently need far more clarity about the current testing situation in England and how quality & testing rigour is being assured by UKHSA. 11/11

PS orig tweet threads here:

https://twitter.com/ProfColinDavis/status/1452285348453752840?s=20

https://twitter.com/ProfColinDavis/status/1453785567628611600?t=48R3LlyzLVHvZGhin_zJTg&s=19

https://twitter.com/cfinnecy/status/1453757631382933510?s=20

https://twitter.com/OliasDave/status/1451900546755989513?s=20