4 ms·
I'm submitting this because there are people on HN who think we're over-counting deaths to Covid-19. This article shows that in some countries we're under-coun
by DanBC 7y ago
I'm submitting this because there are people on HN who think we're over-counting deaths to Covid-19.
This article shows that in some countries we're under-counting covid-19 deaths by thousands.
In the UK we have two counts. There's the daily count from PHE. That's the number used for sources like Worldometers. It only counts people who died in hospital who were tested positive for covid-19. https://coronavirus.data.gov.uk/ https://coronavirus.data.gov.uk/
We also have the ONS numbers. These are based on death certification and then coding by statisticians. There's some lag involved. Here's the most recent version, issued on 3rd April. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredweeklyinenglandandwalesprovisional/weekending3april2020 https://www.ons.gov.uk/peoplepopulationandcommunity/birthsde...
- mytailorisrich 7y agoI have never heard anyone claiming that deaths from covid-19 are over-counted. It is clear that, if anything, they are indeed under-counted. Edit: this is getting ridiculous being downvoted for reasonable comments. Is this HN? No serious source claims that deaths are over-counted. Ignore fake news from conspirationists. The US seem to be sinking if that what is happening on your side of the pond.
- moron4hire 7y agoFind some Republicans on Facebook the day after NYC updated their estimate of COVID-19 deaths to include unconfirmed cases.
- overthemoon 7y agoJust my own perspective, but claims of over-counting appear to be getting more common. There were several demonstrations (granted, over-covered in the media) over the weekend in the US about opening the country up again, and accusations of inflating death toll numbers.
- michaelbuckbee 7y agoIt's becoming a conservative media talking point (Tucker Carlson, Brit Hume): https://www.theguardian.com/world/2020/apr/09/coronavirus-conspiracy-theory-death-overcount-anthony-fauci https://www.theguardian.com/world/2020/apr/09/coronavirus-co...
- docdeek 7y agoThere were many suggestions that Italy was over-counting because anyone who died WITH COVID-19 was counted as having died OF COVID-19.
- rsynnott 7y agoIt’s very much conspiracy theorist stuff, but it’s definitely a thing.
- lolc 7y agoEven if you test every single death, there is the problem of determining who dies with Corona versus who dies of Corona. In the current situation, these numbers are hard to keep apart. A lot of bodies are disposed of before viral spread and cause of death are assessed. We will be able to guess at the numbers, but the error bars will remain big.
- DanBC 7y agoIn this thread we have at least one person saying deaths are over-counted, and there are plenty of similar comments on HN. I agree about the down voting.
- mytailorisrich 7y agoThis is noise. Ignore it, don't amplify it.
- pixelrevision 7y agoThe unfortunate reality is that anyone who is claiming that at this point likely has made up their mind on this. We’ve got huge problems right now with media trust.
- makomk 7y agoThe reality is that we're probably doing both. We're undercounting Covid-19 deaths due to the headline figures in the UK mostly only counting deaths that occur in hospitals, and also overcounting because not all of those deaths are caused by Covid-19: https://www.bbc.co.uk/news/health-51979654 https://www.bbc.co.uk/news/health-51979654 Also, this is a terrible, clickbaity headline. When they say that the care home deaths are "far higher" than official figures, they mean that they're far higher than the figures for the number of deaths two weeks ago. For something that's growing exponentially. As the analysis at the bottom points out, "the hospital figures show deaths have been increasing at a similar rate", so this is sad but not exactly surprising. (It might not be exactly the same rate of growth as hospital deaths, but the numbers given in the article are only very approximate.)