Reply to post: Re: So they know it’s damaging...

UK government puts IR35 tax reforms on hold for a year in wake of coronavirus crisis

Anonymous Coward
Anonymous Coward

Re: So they know it’s damaging...

Citation ? I'd take mortality figures with a very very VERY large handful of salt - especially if they've come from a source (e.g. most of the mainstream media, most of social media) which relies on sensationalism/fear to sell copies/page impressions.

It sounds simple - divide number of deaths by number of infections, right ? Wrong. Because we don't know the latter figure, only the number of clinically confirmed cases which is itself only a tiny fraction of all cases. To put this in perspective, on Monday we got a briefing that at the site I'm based in, there had been two clinically confirmed cases, but another couple of dozen assumed cases that were self-isolating - and that the local hospital is no longer testing people due to the time needed and the numbers involved. So a 7.5% figure derived from "clinically confirmed" cases would amount to somewhere between that, and only around 0.5%. Obviously on such a small sample, the numbers are meaningless, but do illustrate that the figure for "number of infections" is really hard to get a handle on, and even small changes in guesstimate can have a big impact on the calculated mortality rate.

If you want to be silly, I could say that last year during flu season, of people I know who had it and were hospitalised - the mortality rate was 50%. That's on the basis that I know two people (both family members) and only one came home. Again, it's a silly metric derived from a small sample size and invalid numbers - but I hope it demonstrates the challenge in providing such statistics.

So until you can get sensible numbers for all the people who caught it but didn't get symptoms (they'll be some), all those who simply stopped at home till the sniffles were gone (but weren't counted anywhere), etc, etc, etc, it's impossible to actually calculate a mortality rate.

So you have to guesstimate. And don't forget that the mortality rate for the normal seasonal flu is not insignificant. And that many (most ?) of the deaths are of people who are otherwise compromised - so a healthy person without any health problems should get a week or two of "feeling like crap" and then be fine, with the added bonus of having developed a natural immune response to further exposure.

And for anyone who's not read between the lines or otherwise had it explained to them. The current advice is not based on preventing people getting it at all - it's about slowing the spread so that the numbers needing medical treatment at any point in time stay low enough to deal with. Anyone not exposed at all will still be at risk next time it comes around - such as part of next year's seasonal flu.

PS - part of the day job (in engineering, not in the health area) involves collating numbers and assessing "do these look plausible ?". So I am usually sceptical of statistics (especially headlines) unless/until I can see where they came from - knowing how easy it is for an "inaccurate" number to feed through into a nonsense calculation result.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon