Hacker News Reader

Comment by chasd00 | Hacker News Reader

go here and scroll down to Cases by Country, normalized by country population then highlight Sweden. Across the different metrics Sweden seems to be doing about average.

http://91-divoc.com/pages/covid-visualization/

here's projections for Sweden but given how wildly these projections have fluctuated i'm not sure how valuable it is

https://covid19.healthdata.org/sweden

View on HN

Cases are a poor metric because they're a function of per-capita testing, which is variable across nations. You have to go by deaths. In that case you can see that they're much worse than their neighbors Norway and Finland, and that they're closer to regions with catastrophic death rates like Switzerland.

The earlier lockdown measures are imposed, the greater the impact, hence the comparatively excellent numbers from Finland. Sweden has long held denial of asymptomatic transmission (especially by Tegnell, even in this article), which we've learned is categorically wrong. The IHME model has been considered notoriously optimistic, as the US number of deaths has been rising but the US peak is reported as 1 week ago. The IHME model assumes a full lockdown in the US until August, which has already been violated by a number of US states this week.

To back this up with some data, this article has information about tests per capita: https://www.thelocal.se/20200414/understanding-swedens-figur...

If Sweeden had been testing at the rate of Norway or Germany, they would not be in the middle of that cases per capita graph. This is despite Sweeden's infection growth being ~15 days behind Italy's, e.g. enough time for a lockdown to have almost entirely prevented it.

So it seems clear that Sweeden's policies have made infections worse. The questions that remain are: 1) Will their peak infection rate go over hospital capacity? 2) Will they have economic benefits from having population immunity faster?

As Tegnell points out, the relevant comparison is death rates after the epidemic has run its course, say after a year or so. Lockdown just delays some of the infections to buy time for additional mitigations which will hopefully decrease the total number of deaths. The virus will still eventually spread through the population until herd immunity is reached.

This ignores hospital capacity, which is the most important consideration when projecting out fatality rates. Lockdowns are not just about buying time for additional mitigations.

And the ICU load has hovered around 80% the last couple of weeks. Which was the goal of the chosen strategy. Current goal is to keep things below that while things develop

One of those mitigations is increasing hospital surge capacity.

> Will their peak infection rate go over hospital capacity?

https://www.icuregswe.org/data--resultat/covid-19-i-svensk-i...

First bar graph is total number of patients with covid-19 in ICU care in Sweden. Blue is yesterday's numbers, orange is today's numbers, including backdating of cases.

So probably not, is the answer to your question.

Sweden's death counts are the most statistical precise ones so far, exactly because every death from someone tested positive for SARS-CoV-2 is counted as a death from CoViD-19. No other country is doing this, you can check that by following metrics such as "excess deaths" after seasonally adjusting deaths by country and adding their official death tolls [0].

[0] https://news.ycombinator.com/item?id=22945974

>>> Sweden's death counts are the most statistical precise ones so far, exactly because every death from someone tested positive for SARS-CoV-2 is counted as a death from CoViD-19.

So if the 95 years old who already withstood 3 heart attacks dies from the 4th one, and he tests positive for SARS-CoV-2 you then blame his death on CoVID-19, and not on his weak heart? Doesn't make much sense to me.

It makes sense when you think that this removes the individual variation per hospital/doctor writing a death certificate, you don't rely on professional opinion, you count all of them as it is easier to later clean this statistic with models based on historical data than the opposite (to figure out the missing data for deaths from CoViD-19 after the fact).

You need to be aware that Sweden has a long history of keeping demographical statistics [0] so there is a method to the apparent madness of counting.

[0] https://www.scb.se/en/About-us/main-activity/history-of-stat...

Why would you normalize by population in the early stages of an epidemic? If you assume a large and small country both have similar growth rate and number of initial cases, then their absolute case growth will be similar in the early stages. But the large country will appear "better" once you normalize for population. This is exactly the pattern I see when you compare Sweden to its smaller neighbors.

is this still the early stages? A number of countries have passed their peak.

A number of countries have most of their populations under what is effectively house arrest

Meanwhile, the virus still permeates, those same populations' collective immunity has barely changed, and we're only marginally closer to a vaccine

The fundamentals won't have changed when we do get around to removing the lockdowns

Whether the virus is permeating is the point. If the virus has broadly spread across society, and it sounds like we're agreed that it has, then you have to normalize by population to get useful numbers.

Why would you assume a large and small country both have similar number of initial cases? I'm under the impression a lot of the nordic initial cases were travelers coming home from ski trips in Italy. That ought to scale pretty linearly with population.

Reply on news.ycombinator.com