RE: UK Government doing the right thing22 Mar 2020 03:59
The apparrent rate of growth of the infected might be a useful measure, and it might not. Take the UK, where to begin with everyone who presented with a 'reasonable' risk of being infected was tested. What 'reasonable' is might vary by area. Anyway, policy was changed so that people who were in a serious condition ( admitted to hospital ) were tested. As a response to the testing system being overloaded. Now the WHO have ticked off Westminster and the NHS and now more testing is being done. And even more 'soon'. The figures and the data from such variable criteria become meaningless. Most testing in most countries is pretty meaningless. Firstly testing has to be administered in a consistent way, the limitations of testing have to be acknowledged. We only know through UK testing the ratio of seriously infected people to the number of dead ones. When health services become overwhelmed then those that triage separates after being assessed into those rooms where they are left to die, the testing these people may not be a priority from a clinical point of view but useful for understanding the epidemic. We will never know the number being infected with the virus until we test people at random. Which I have not seen data on so far. Even random testing would not be the whole picture until we have a random test to identify people that have 'had' the virus.
Death rates? Unless we know the number infected then the number dying is just that. By itself it will tell us little. Death rates around the world have been quoted in some places as 0.5%, while in Italy I saw one news service quote it as 8.5%. This is the number of those tested positive that ultimately die. Not the number of people who catch it that die. We need an antibody random test plus an infection random test.
The data is extremely poor and by the time we understand the numbers, it may be too late for many. I have been modelling the spread for the last six weeks and today was the second day my numbers were more than 4% wrong. I think I failed to include a different Ro factor for Sundays. Just a possibility. The solution is deterministic after best guess.
The UK is on course for Italy style numbers in 8 days, 10 days behind where they are, only with a health service that was much worse than theirs. Initiatives can change the future, but there is a time delay involved circa 14 days.
Now where did I put that Hydroxychloroquine....