I had an interesting exchange today through Twitter messages. I can now see why public figures disable this feature. A person from a major sports club told me that it was their club policy to have players, coaches, physios, anyone who might travel to & from other clubs, tested...
...by swab / PCR twice a week. Funny thing is, they said, it was commonplace for a couple or three positives to come back almost time, and on no occasion has anyone been symptomatic. Because the club is adequately funded & they want to know the real Covid19 status of their...
...person, they routinely order re-tests on anyone testing positive on the regular cycle. So far, every time, the re-tests have come back negative. I replied that’s very interesting. Would they kindly go back through their recent month or so, while testing in the country has...
...been in strong acceleration. A little while later, I got my reply. “5% pretty much” was the answer. Of course, I nearly fell over. Followers will confirm that recently, earlier than this exchange, I’d estimated a likely false positive rate of possibly as much as 5%.
Furthermore, I’d pointed out that, if the FPR really was 5%, then essentially all the positives obtained in mass screening, so called Pillar 2, are false.
Now, I am not saying that a few hundred tests performed for a sports club is any kind of formal sample for determining the...
...FPR in Pillar 2 at present. But it did give me a funny feeling in my gut.
Later the same day, this afternoon, I was on the phone to a professor of immunology at a U.K. university (critics: you don’t think I do all this on my own, do you? I’m still plugged into the research...
...network I’ve been part of all my adult life. People who were doing their PhDs when I did mine are now heads of faculty, textured professors, biotech CEOs and even U.K. Chief Scientific Advisors). I was checking some maths relating to herd immunity thresholds. He thinks I’m...
...correct in my assessment of the pandemic in U.K., by the way, as does a friend who is a professional mathematician & another who’s a pathologist. I wanted to pick his Braun’s about estimating possible scenarios for false positive rate in mass screening with PCR. As always...
...he reminded me that we were young scientists when the technique was invented, but the thorny problem of avoiding errors when you’re manually moving small but precise volumes of liquids from vessel to vessel, perhaps involving quite a few discrete steps, over & over again...
...is notoriously difficult. In short, you screw up periodically. Even if you’re highly motivated, very experienced & it’s your own research work, you still occasionally make mistakes. You pause, and realise you’ve “dropped a stitch”, and you’re trying to recall whether or not...
...you did at that ten microlitres of reagent. You can’t see it if you’ve added it even to a tiny tube if there’s already some liquid in it. You might be reading this, thinking how hard can it be to avoid errors? It’s not only the risk of momentary loss of concentration. It’s...
...also the case that there’s a great deal more skill involved in doing it well all day vs spilling tubes, losing the tip off your Gilson pipette (which of course you’re never dropped in the floor...have you? They can lose their calibration if you do) & even drawing up and then..
...transferring liquids without splashes, creating contaminating aerosols etc. Consider the humble screwdriver as quite a good analogy. I’m handy with engineering tools from screwdrivers to precision bore micrometers. I wasn’t always, of course. When you watch someone who’s...
...put in thousands of screws, there’s an admirable adeptness about how they engage the tip in the head of the fastener, how they apply just the right amount of torque etc. What they don’t do is dig the screwdriver into the paintwork or stab themselves in the hand. Similarly,
...watching a skilled lab technician moving about and getting it done well has that same professional economy of movement, assuredness in what they’re doing. But if you’re hiring new staff all the time, as we know they are in Lighthouse Labs, you’ve got the lab equivalent of...
...the hand-stabbing screwdriver wielder. Last week, a senior scientist resigned from one of those labs and was interviewed for BBC News at Ten. He made the same points I’ve just made. They are now recruiting staff who’ve never worked professionally in a lab. Previously, they...
...were getting seconded staff from universities but we’re way beyond that now. Not enough trained technical folk to do this as well as we used to. The prof and I agreed that a doubling of error rate compared with June/July was the absolute minimum, and it could easily be a...
...multiple of that. It’s important people realise that the errors giving rise to false positives especially mostly arise in the sample preparation stage, which is what I’ve attempted to describe. It’s not “in the kit”, nor is it fixed & invariant. That is why I’m appalled as...
...a lifer scientist to learn that PHE do not determine the operational false positive rate. Or if they do, it’s a state secret. It’s never discussed until people like me & others forced it out of Govt. It matters. While you cannot in a formal sense ‘correct for’ false...
....positives by subtracting the rate from your gross number of positives, knowing what the rate is, is vital when you’re trying to work out where we are with this virus. Anyway, the Prof & agreed that 2% mechanical error was fair as a base assumption. He went on to explain...
...that he’d been through the TDS (technical data sheets) of all the PCR test kits available in U.K. About two thirds of them use primers which cross react with pieces of common cold coronaviruses, of which 4 are endemic, and 3% of us are carrying one of them at any time now.
...That’s why I honestly think an operational FPR of 3-5% is likely. And IF I AM RIGHT then most or all of the positives or “cases” are false. I cannot do any more. Someone needs to force it out of Mr Hancock. If I’m right and NO ONE does anything about it, this nightmare will...
...never end. Even if the virus stopped circulating, you might still see thousands & thousands of “cases” every day, depending on how many tests are done. No, it’s not a fixed value, so don’t start pointing out there aren’t enough cases last Tuesday week for the FPR to be that...
...large. That’s not how it works.
So who’s going to ask the minister, this time?
“Mr Hancock, please tell us the operational FPR for population mass screening, from the last four formal internal QA runs, complete with the supporting report”.
You can follow @MichaelYeadon3.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: