WHO - Covid excess death rates for UK lower than Italy, Germany & Spain

[TSSeasider]

I thought this may have been given a bit more profile by HMG than it has been, but that may be due to Purdah.

On the BBC article, stating that 15m died from Covid around the world, not 5 million as reported by looking at excess death rates; I was pretty surprised to see, our excess death rate was lower than Germany's; given how much the actions of the UK Government (in England) were compared with the Germans, we've done significantly better than I thought we would.

To have done better than Italy, Spain and Germany should be recognised as a relative success.

Covid: World’s true pandemic death toll nearly 15 million, says WHO

These include deaths from Covid and from the indirect effects of the pandemic over two years worldwide.

www.bbc.co.uk

 

[Lost Seasider]

To have done better than Italy, Spain and Germany should be recognised as a relative success.

Looks to me like this is the Lancet study from March repackaged.

The important thing to note is that these results come from a model and not real world data, what the authors are trying to do is estimate the total global excess mortality including countries where no reliable data exists, results for individual countries should be treated with extreme caution.

The thing is that countries like Italy, Spain, Germany and the UK have reliable death data, so it's a simple matter to calculate the excess mortality using those, the results are somewhat different to the model, with The Economist and real world data we get:

• Italy: 335;
• Spain: 266;
• UK: 221;
• Germany 149;

so the model significantly underestimates the EM in Italy and vastly overestimates the EM in Germany, unless the Germans really have had another 100,000 Covid deaths and nobody has noticed.

I haven't checked (yet), but I imagine this model also says that Sweden suffered the same number of excess deaths as Finland and Norway, and no doubt the anti-vax brigade will latch onto this as proof that lockdowns don't work.

And for the life of me, I can't follow the logic in dividing the EM by two.


Edit:

I've checked, the figures for the Nordic countries are:

• Denmark: 32;
• Finland: 26;
• Norway: -1;
• Sweden: 56;

so at least we're spared another round of "proof" that lockdowns don't work.

 

[1966_and_all_that]

Looks to me like this is the Lancet study from March repackaged.

The important thing to note is that these results come from a model and not real world data, what the authors are trying to do is estimate the total global excess mortality including countries where no reliable data exists, results for individual countries should be treated with extreme caution.

The thing is that countries like Italy, Spain, Germany and the UK have reliable death data, so it's a simple matter to calculate the excess mortality using those, the results are somewhat different to the model, with The Economist and real world data we get:

• Italy: 335;
• Spain: 266;
• UK: 221;
• Germany 149;

so the model significantly underestimates the EM in Italy and vastly overestimates the EM in Germany, unless the Germans really have had another 100,000 Covid deaths and nobody has noticed.

I haven't checked (yet), but I imagine this model also says that Sweden suffered the same number of excess deaths as Finland and Norway, and no doubt the anti-vax brigade will latch onto this as proof that lockdowns don't work.

And for the life of me, I can't follow the logic in dividing the EM by two.


Edit:

I've checked, the figures for the Nordic countries are:

• Denmark: 32;
• Finland: 26;
• Norway: -1;
• Sweden: 56;

so at least we're spared another round of "proof" that lockdowns don't work.

Some decent points in there Lost. Of course there are a myriad of qualifying factors that the data doesn't address - it can't - such as Sweden having far fewer centres of population density than does the UK. I would be very wary of using the data on its own to draw conclusions as to the veracity of Government policy.
Edit: not that I'm saying you have btw.

 

[TSSeasider]

Looks to me like this is the Lancet study from March repackaged.

The important thing to note is that these results come from a model and not real world data, what the authors are trying to do is estimate the total global excess mortality including countries where no reliable data exists, results for individual countries should be treated with extreme caution.

The thing is that countries like Italy, Spain, Germany and the UK have reliable death data, so it's a simple matter to calculate the excess mortality using those, the results are somewhat different to the model, with The Economist and real world data we get:

• Italy: 335;
• Spain: 266;
• UK: 221;
• Germany 149;

so the model significantly underestimates the EM in Italy and vastly overestimates the EM in Germany, unless the Germans really have had another 100,000 Covid deaths and nobody has noticed.

I haven't checked (yet), but I imagine this model also says that Sweden suffered the same number of excess deaths as Finland and Norway, and no doubt the anti-vax brigade will latch onto this as proof that lockdowns don't work.

And for the life of me, I can't follow the logic in dividing the EM by two.


Edit:

I've checked, the figures for the Nordic countries are:

• Denmark: 32;
• Finland: 26;
• Norway: -1;
• Sweden: 56;

so at least we're spared another round of "proof" that lockdowns don't work.

This is the WHO explanation.

Seems comprehensive.

The true death toll of COVID-19: estimating global excess mortality

Estimates suggest the total number of global deaths attributable to the COVID-19 pandemic in 2020 is at least 3 million, representing 1.2 million more deaths than officially reported.

www.who.int

 

[tommytwojags]

So, on average, countries have declared only one third of covid deaths. Maybe the UK should have been a little more circumspect with its published figures like most others.

 

[rockontommy]

Lockdowns kill, Sweden got it right. The End

 

[Lost Seasider]

This is the WHO explanation.

Seems comprehensive.

The true death toll of COVID-19: estimating global excess mortality

Estimates suggest the total number of global deaths attributable to the COVID-19 pandemic in 2020 is at least 3 million, representing 1.2 million more deaths than officially reported.

www.who.int

Doesn't explain how they got to their estimates of excess mortality, this is apparently the model they used: https://www.who.int/data/technical-advisory-group/covid-19--mortality-assessment/membership

I think the point worth repeating is that Germany is an advanced first world country and their death statistics should be considered reliable, and it ought to be a simple task to calculate excess mortality from that, but what this study says is that Germany has had an extra 100,000 or so deaths, presumably from Covid, but nobody has noticed or bothered to record them, and the German death statistics, from Covid or otherwise, are not reliable.

That is simply not plausible.

To be fair, I don't think the authors are trying to estimate excess mortality on a country by country basis, their aim is to produce an estimate of global excess mortality, but the press have reported this in a way that implies that all first world mortality statistics are meaningless, and I don't think that argument flies.

 

[Lost Seasider]

BBC's More or Less on the subject.

Short version, the authors of the WHO report themselves think the Germany figures are wrong and are going to release revised figures, they also think the Sweden estimate is too low.

Longer version, the problem lies with the way they've calculated expected mortality, which they then compare to the actual mortality to calculate the excess.

Most excess mortality calculations are based on a simple average of deaths in the preceding 5 years or so, in this case the authors have created a more complicated model of expected mortality that includes factors such as HIV and diabetes rates, and some results look a bit odd.

For Germany, apparently the model placed too much weight on unusually low death numbers in 2019, which meant the expected mortality for 2020 and 2021 was too low, which made the excess mortality look too high.

So regardless of the authority of the source, it's always worth checking if what they're saying really makes sense.