Hacker News new | ask | show | jobs
by CogentHedgehog 2037 days ago
Key point:

> The latest figures from the Office for National Statistics (ONS) show the number of deaths registered in England and Wales in the week ending 6 November 2020 was 14.3% higher than the five-year average.

But it is good to know that the measures to control COVID-19 are also suppressing flu and pneumonia (though not enough to make up for COVID-19 mortality).
4 comments
dmichulke 2037 days ago
Also, they didn't write the standard deviation, so that 14% is kind of useless
link
himinlomax 2037 days ago
Over a stable population of 10s of millions of people, it's unlikely to be notable.
link
dmichulke 2037 days ago
That misses the point. I want to know whether +14% is a 1,3, or 5 sigma event.

For that calculation stddev is in the denominator and whether it's 2%, 4% or 6% makes a huge difference, notable or not.

link
mjbeswick 2037 days ago
It's still very subjective. The thing is that people that are dying of COVID-19 would very likely have died in the new few years, so if you compare deaths over the the last few years to a previous time period their would be very little difference.
link
JoshuaDavid 2037 days ago
I really hope that is "14.3% higher than the 5 year average mortality rate for the first week of November from 2015 - 2019" not "14.3% higher than the 5 year average mortality rate for the entire time period from the start of 2015 to the end of 2019." Because mortality is pretty strongly seasonal.

Edit: "So far this year, up to 6 November, 517,650 people have died from any cause in England and Wales. This is 58,555 more than the five-year average." So the definitely seasonality adjusted excess mortality for the year to date is about 12.7% (or 12.4% possibly, depending on how the leap day was handled).

link
lvass 2037 days ago
A whole week of data!?
link
renewiltord 2037 days ago
Oh god just click through, dude. Why are so many hn commenters obsessed with these banalities https://www.ons.gov.uk/peoplepopulationandcommunity/birthsde...

I should warn you that you might have to scroll a little, lest you find yourself writing a comment in response before you read section 2.

link
CogentHedgehog 2037 days ago
Yup, the data is right there. HN commenters just want to "but aktsualllly..."

The spike in deaths from the UK's first wave is pretty visible in the graphs, and the spike from cases we're seeing now will probably take a few weeks to show up in the data.

link