I'm appalled that @theEIU and @theEconomist consider this acceptable. It's in no way a meaningful analysis, it says nothing about the response and the conclusions are hilarious at best. This is poor journalism and even worse science. A thread. 1/n

https://www.eiu.com/n/campaigns/oecd-countries-responded-to-the-coronavirus-crisis/
2/n Let's look at the methodology. They take 3 indicators:
- number of tests
- the amount of cancelled cancer surgery
- excess mortality
and 3 risk factors:
- obesity prevalence
- share of 65+ population
- international arrivals

All scored 1-4 based on aribtrary breakpoints.
3/n Of those, only 2 are related to response, the first 2. But somehow they decided (god knows why) to weight excess deaths 4 times more than all the other components. This is in itself weird, and becomes problematic if you look at how death rate is looked at.
4/n Compare Belgium to the US. Belgium has a score of 1, the US a score of 3. They got this data from FT. You see excess deaths in Belgium peak pretty high. But Belgium has a population that's smaller than New York, and its peak was gone half may.

https://www.ft.com/content/a26fbf7e-48f8-11ea-aeb3-955839e06441
5/n Belgium was particularly hard hit early on due to influx from Italy (ski holidays). Now compare Belgium to individual states. You see immediately that reported deaths were higher in the hardest hit states. And that's not even excess mortality.

https://www.statista.com/statistics/1109011/coronavirus-covid19-death-rates-us-by-state/
6/n Overall, BE excess mortality was back to normal half may. Over the entire US, it was still 8-12% higher by may 23. New York is still at +16-28%. Yet, the US somehow "responded far better" than Belgium?

https://epistat.wiv-isp.be/momo/ 

https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm#dashboard
7/n Compare BE to Chili (which has a score of 4). Keep in mind Belgium report also suspected deaths, while Chili's reporting is problematic. Anyone who wants to argue that Belgium responded far worse than Chili?

https://ig.ft.com/coronavirus-chart/?areas=bel&areas=chl&areasRegional=usny&areasRegional=usnj&cumulative=0&logScale=1&perMillion=0&values=deaths
8/n By the way, @TheEIU can't even get their childish 1-4 score table right. BE cancelled less than 30% of cancer surgery according to their own (wrongly cited) source, but yet got a score of 3. Go figure.

https://bjssjournals.onlinelibrary.wiley.com/doi/full/10.1002/bjs.11746
9/n So while Belgium got a maximum score for the only 2 parts that are actually related to response , it still comes out as the worst country. And that's only because of the insane weight placed on a problematically naive way of looking at mortality. How do you defend that?
10/10 If you really think that averaging 1-4 scores with arbitrary breakpoints on a few ill defined "factors" can somehow translate to "covid response", you should look for another job and leave this to journalists like @jburnmurdoch .

Dammit.

https://ig.ft.com/coronavirus-chart/?areas=bel&areas=chl&areasRegional=usny&areasRegional=usnj&cumulative=0&logScale=1&perMillion=0&values=deaths
You can follow @JorisMeys.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.