Saturday 9 May 2020

Day 54 of self-isolation - the UK media

I don't understand it. The UK media are concentrating on whether the government has reached the "target" of 100,000 tests per day or only managed 80,000. Is that the most important issue of the day? Is it even one of the important issues?

No-one is asking where this number came from. Why 100,000? There's 63 million of us. If we had a million tests per day, then we could be tested once every nine weeks, which sounds rather inadequate to me. At 100,000 tests per day, we could be tested once every 22 months. It's pathetic.

No-one is asking how good the tests are. What percentage of tests give false positives (say you're infected when you aren't)? What percentage of tests give false negatives (say you're clean when you aren't)?

How long before the test results are known? Is it 15 minutes? A day? A week?

What are the costs of these test?

Why are the UK media so incurious? I'm not a conspiracy theorist; I usually ascribe bad outcomes to incompetence rather than malevolence, but the UK media must be spectacularly incompetent. And our government must be similarly incompetent for not publishing these data.

In the next phase of the fight against the virus, testing and tracing are going to be uppermost. When the Isle of Wight trials are in place, will the results be published? Even before they are in place, it should be pretty simple to know how many people have installed the tracing app, and are sending their data. I'd like to see that day by day - it wouldn't violate anyone's privacy, and statistics on app installation are available for Candy Crush, so why not NHSX?

Now that the UK has officially discovered that black people (I don't use the euphemism "people of colour" because there's a significant difference between black and brown) are MUCH more likely to have a bad outcome (i.e., fatal) if infected than white, why aren't any of the media demanding an investigation into whether the simple and cheap application of vitamin D to darker-skinned people makes a difference? To ignore this, is surely racist?

But there's good news. Only 346 people died yesterday of Covid-19 in the UK.  Good news? Only relative to the 980 per day we saw a few weeks ago. We're still in a pandemic.


  1. Charles Dytham9 May 2020 at 21:07

    There are some stats on testing here:
    Complicated by their "pillars"

    Home testing kits were controversially included in the daily tests when the (50k?) ones posted out were used to hit their 100k target. What I'd like to know is: of the total kits posted, how many have been processed (only +ve result number given)?

    I tried to use their awful system to register a received test kit - final hurdle was to match the kit barcode number to my details, but this failed saying my date of birth was wrong! I spent over an hour on their 10p per minute helpline where the poor chap was having to use the same IT system as I had. He eventually concluded that the kit barcode number I had received did not exist!

    He gave me another help number to call which was "unobtainable".

    So, nearly a week on from original request, I am no further forward, and need to re-order a test kit.

    At this rate, if I did have it, I may well test negative by the time I can get test done.

    Bring on the antibody tests.

  2. Sounds like a bargain British bungle.

  3. Charles Dytham11 May 2020 at 19:13

    Today's (Monday) daily stats at the link above has:
    Deaths in hospitals 229
    Deaths in all settings 210
    It does then have an explanation of why the daily figure for deaths in hospitals can be higher than the daily figure for deaths in all settings, but I couldn't get my head around it.

  4. Data precision and analysis was the topic of my PhD.