Hacker News Reader

Comment by msandford | Hacker News Reader

20k deaths for a population of 10M in Sweden is 0.2% right? USA has had just under 1M deaths for 330M people or closer to 0.3%.

I'm not saying that Sweden did the totally right 100% perfect thing. But it seems to have done much less catastrophically horrible than many predicted at least as far as I can see.

View on HN

I like what you’re doing here. Take a country like the United States with a horrible outcome and use them for comparison. More people died after vaccines than before in the United States. How does that happen?

One thing is absolutely clear: 1 million Americans never needed to die from coronavirus.

It's not clear to me the total deaths number is accurate in the US because the government was reimbursing hospitals for COVID treatment and paying for people's funerals who "died of covid". It incentivized people to list the cause of death as COVID even when covid didn't play a role in the death or was at best incidental to it.

Anecdotally this happened with the father of a friend of mine. He had cancer since before the pandemic and it became terminal after exhausting treatment options last year. As his condition deteriorated, they planned to move him to hospice, did a routine covid test and found he was positive. He died a few days later. He was going to die of cancer in days no matter what. He is counted among the million US covid deaths and the US government paid for a portion of the cost of the funeral.

What percentage of that million are cases like this? I don't know, but it is pretty clear it wasn't uncommon and in the context of your original comment I wouldn't call cases like this preventable deaths.

Look up excess deaths

You act like excess deaths is an easy analysis, but deaths vary significantly year or year. What you baseline are makes a big difference.

And considering we’re not even through the epidemic it seems like it’s a bit early to call winners and losers.

Nonsense. There are various reasonable ways to calculate a baseline and they all show large surges over the covid period (which correspond to reported covid deaths, btw).

Also, there are no winners, what are you even talking about?

Do you average deaths over the past 5 years? Or 10 years? How do you correct for population growth? How about age distribution? What if you had a flu pandemic and deaths were 5% higher 3 years? Correct for that or not?

Excess deaths could just have easily been caused by people postponing medical care for other conditions. Hospitals were canceling elective procedures for example.

> the government was reimbursing hospitals for COVID treatment

That's how all of medicare works.

I can't believe people have been regurgitating this talking point since the beginning of the pandemic with no thought whatsoever about what it means

They specifically boosted what Medicare paid for COVID treatment by 20% during the public health emergency.

This article has a good overview of the exact terms of the extra 20% payments made by Medicare.

https://www.hfma.org/topics/news/2020/04/increased-medicare-...

and? do you have any proof of your claim that "It incentivized people to list the cause of death as COVID even when covid didn't play a role in the death or was at best incidental to it"? Are you saying doctors are personally getting some sort of kickback?

Besides my direct experience? It doesn’t require fraud this is the result of the loose requirements to get reimbursed under these temporary measures and the fact that COVID tests are a routine part of hospital admissions. For example to be reimbursed for funeral costs COVID only need be listed as a contributing cause and doesn’t need to be the primary cause of death.

so you have no evidence.

How much influence do you think what you are mentioning has had on the numbers and what is your proof of that?

It would be a better benchmark to compare Sweden with Denmark, since societies and health care capabilities are more alike.

Since I looked it up out of curiousity:

https://www.statista.com/statistics/1104709/coronavirus-deat...

deaths per million inhabitants:

Sweden: 1,755.2

Denmark: 924.97

Finland: 521.89

Norway: 405.58

Maybe add Netherlands and Germany ...

And in DK we are now doing 20-50 corona dead pr. day - leaving us at 365*20-50 / 5.5 at 1,327 - 3.318 added for this year.

There's a recent large study in The Lancet, where they estimate 2020-2021 excess mortality, and compare these to reported Covid19 deaths. The study covers a large number of countries, but here I only look at your four.

              excess  excess/reported   excess/(1 million)
    Denmark:  10 400             308%                 1784
    Finland:    8780             503%                 1587
    Norway:      742              57%                  138
    Sweden:   18 100             118%                 1719

Either the study methodology to estimate excess mortality has some problems with Denmark and Finland. Or, if the study is correct, the official numbers from Finland and Denmark are bad in estimating their own Covid19 deaths: Denmark underestimating by 3x, Finland underestimating by 5x.

https://www.thelancet.com/journals/lancet/article/PIIS0140-6...

Direct link to results table:

https://www.thelancet.com/action/showFullTableHTML?isHtml=tr...

Denmark tested way more than any other country in the world. They most likely have the most accurate picture of how many people actually died with covid

I'd strongly question their method of calculating excess mortality, look at denmark on https://www.euromomo.eu/graphs-and-maps and compare it to other countries

I would assume methodological flaws. Just by looking at the number of authors and the publication date, the authors simply did not have enough time to control for area-specific factors for 191 countries and 252 subnational units.

Finland reported 53949 deaths from all causes in 2019, 55488 deaths in 2020, and 57343 deaths in 2021. (The figure for 2021 is preliminary.) The total increase from the 2019 level in 2020-2021 is ~5000 deaths, and a large part of that is explained by aging population.

Agree on using excess deaths data.

Also, it'll be years before knockoff effects can be more fully assessed.

"Better" in the sense that it makes them look worse, and some people prefer this version of events? Perhaps.

But if you compare across all of Europe by deaths per capita, they're near the best performers on the list (#13; #12 in continental Europe), and the only reason those specific countries look better is because they're the top three in continental Europe (they're also smaller countries; see below):

https://www.statista.com/statistics/1111779/coronavirus-deat...

In particular, the UK, France, Spain, Belgium, Italy and Portugal all did worse, despite far more significant impositions on society.

It's also important to know that worldwide, there's a clear dependence of per-capita deaths on population (greater population -> greater deaths per capita):

https://coronavirus.jhu.edu/data/mortality (second plot; deaths per 100k population tab)

And that Denmark has a population of ~5.8M, Norway 5.4M, and Sweden 10.4M. This alone should make you question any simple narrative involving interventions and outcome.

Societies and health care aren’t as similar as one might think in the Nordics. Sweden is surprisingly diverse, unequal, urbanized etc. It’s more “continental” than its neighbors in a lot of ways.

The US has like 5 times the obesity rate and obesity is a relatively strong predictor of adverse outcomes with C19.

Reply on news.ycombinator.com