Any excess mortality experts out there?

I'm looking the cumulative excess mortality tracker table on @OurWorldInData

https://ourworldindata.org/grapher/excess-mortality-cumulative-deaths?tab=table&stackMode=absolute&time=earliest..2020-07-26&country=~ITA&region=World

As of July 26, which is last date we have excess mortality data for Canada, I see Canada has 3rd highest % excess mortality among peers.

The absolute number of excess deaths as of July 26 matches cumulative total #COVID19 deaths in #Canada reported by @covid_canada, so I suspect this means we're doing a good job of counting our COVID deaths, and that only a relatively small % of them would have died without COVID.

Other countries, such as USA, seem to be greatly under-reporting, since as of July 26, they had a cumulative excess mortality of nearly 250,000 deaths but are only reporting that number of COVID-19 deaths as of this week.

Ppl can see the whole list at the link above (move slider to July 26 to compare same dates as last available data for Cda).

But to me, this means #Canada has actually been extraordinarily hard hit by #COVID19. This would make sense given our high case fatality rate.

List of cumulative % excess mortality by country on Jul 26, from worst to best:

(England, Wales): +97,805%
+74,842%
+11,419%
+5,298%
+4,625%
+3,225%
+1,985%
+113%

There's no excess mortality data for the other high income countries w/ pop >20M I usually track.

I think it's pretty clear that #Canada is doing a good job of "counting the dead" for #COVID19, and reporting them.

But.....we either have an incredibly high true case fatality rate compared to other countries (i.e. we're inherently more likely to die of COVID--doubtful), or....

....we are missing an absolute ton of #COVID19 cases, and our epidemic has been much, much worse than we know and/or than we've been accustomed to thinking.

Here's the age breakdown of the excess mortality. I'm comparing to Germany in this graph, because generally they kept excess mortality very low during this period.

So, our excess mortality has almost exclusively been in people older than 65. It's high in the 85+ group, many of whom would be in nursing homes, but also in other groups that are less typically living in care homes. @LauraCRosella & colleagues saw this in Ontario cremation data.

Our excess deaths in older people are NOT just in longterm care settings, although these are the ones that may have been most visible due to media coverage etc.

OK. So let's do some thinking about case fatality rates and detected cases. Just back of the envelope estimates.

The world COVID-19 CFR to date is 2.4%. Canada is 3.6% (50% higher, 3rd only to Italy, UK, BTW).

The actual CFR is probably closer to 0.5-1%, according to experts.

This difference suggests that the size of the worldwide pandemic is likely 2.5-5 times larger than we know (very likely). Or the experts have greatly underestimated the true CFR (much, much less likely).

For those wanting to get a different sense of what this means, 1 in every 42 ppl diagnosed with #COVID19 worldwide dies. In #Canada it's 1 in every 28 diagnosed people who die.

Not good odds if you're Canadian.

ok. let's try to break this down a bit and see if we're more likely to die than other people in the world (disturbing if true), or if we've REALLY underestimated the scale of the epidemic in Canada.

Let's work with some data from Ontario.

As of the end of July, seroprevalence data from @PublicHealthON estimate that ~1.1 % of Ontarians have been infected.

https://www.publichealthontario.ca/-/media/documents/ncov/epi/2020/10/covid-19-epi-seroprevalence-in-ontario-july-31.pdf?la=en

ON 2019 pop was 14.57M. So about 160K ppl have been infected.

As of Jul 30, ON had 2809 reported #COVID19 deaths. Let's assume all deaths were reported.

2809 deaths/160K cases = a case fatality rate of 1.1%, which is in line w/ expert estimates of true CFR.

The more I think about all of this, the more convinced I am that we're MASSIVELY under-detecting the size of the epidemic in Canada, likely due to our testing strategies.

This is not good right now, when we badly need situational awareness to make public health decisions.