Hacker News Reader

Comment by marvin | Hacker News Reader

Sweden: 1,937 deaths. 50 fewer seasonal deaths than expected.

Norway: 187 deaths. No increase in seasonal deaths.

Denmark: 384 deaths. Seasonal deaths unknown to me.

Finand: 149 deaths. Seasonal deaths unknown to me.

Sweden has approximately twice the population of the other Nordic countries, so per capita they have between ~2.5x and ~6x the Covid19 mortality of their neighbors, who have stricter preventive measures.

Honestly, I can't see how Anders Tegnell and his colleagues can sleep at night with the uncertainties they are powering through. By rights they should be terrified that they've gotten one of their assumptions wrong.

View on HN

Denmark has ~2.5x the per-capita covid-19 mortality of Finland, should Danish authorities be terrified that they've gotten their assumptions wrong?

The UK has ~2x the per-capita covid-19 mortality of Sweden, should English authorities be terrified that they've gotten their assumptions wrong?

You are cherry-picking like crazy!

There is clearly no evidence that stricter measures have any effect. If that was true, then there should at least be a linear correlation between strictness of measures vs. excess deaths, and that's simply not the case. You're extrapolating a linear correlation using a couple of data-points that happen to support your theory, i.e. Sweden vs other Nordic countries. Sweden vs. all of Europe paints a completely different picture, so you're not looking at that, because you're cherry-picking.

There are so many uncertainties around this whole situation, and I think the only thing we can say with certainty is that if a country has no or a low rate of excess deaths, that country is doing "ok". Not doing great, not doing shit, just doing ok.

> There is clearly no evidence that stricter measures have any effect

Working for us in NZ, although I think the timing of our national lockdown is strongly significant to our success, we managed to do so before it was widespread in the community.

Our new case rate is literally bottom of the FT Covid graph: https://www.ft.com/coronavirus-latest

And we have 50% the population of Sweden, and 0.72% the deaths. If we were to be per-capita equivalent with Sweden, we'd need about another 964 people to die.

Look, I agree with you that stricter measures probably have an effect. All countries have some forms of restrictions, all countries have flattened their curves, no-one is really seeing crazy exponential growth, common sense dictates that it's working.

But a lot of people in this thread are arguing that Sweden should have enforced stricter measures because that should have decreased the deaths further. And that position is unsupported, because there are comparable countries with stricter measures than Sweden that have a higher per-capita death count, for example the UK or Belgium. Looking at excess mortality the picture is similar. Sweden is absolutely not doing great, but it's not doing bad either.

If stricter measures always resulted in less deaths, as you argue, why are so many countries doing worse than Sweden? Why are more people dying in those countries? Is their population stupider? Unluckier? More defiant towards government? Clearly there are multiple other unknown factors at work here, which is why there's no evidence that stricter measures in Sweden would have an effect.

> for example the UK or Belgium

It depends when each country locked down relative to the number of cases and deaths at the time.

The UK is widely acknowledged (here anyway) to have acted too late in its lockdown as the were trying to pursue herd immunity.

Every country was widely acknowledged to be too late in its lockdown. In the ones that have seen disasters the accusation stuck, while in the ones that haven't it's been quietly forgotten.

I perhaps didn't emphasise this enough in my original post, but I believe it answers your questions.

> I think the timing of our national lock down is strongly significant to our success, we managed to do so before it was widespread in the community.

> Why are more people dying in those countries?

Timing is a very significant factor, I believe - I couch that as a belief, because I haven't had the time to compare all lockdowns, but you'll note that on this FT graph[1], NZ's lockdown began near immediately by the metric being used, whereas the UK's took a full two weeks.

We know that the UK government dithered, initially focused on herd immunity (and worth noting that Johnson and his advisor Dominic Cummings who was hugely influential in pushing for it both went on to contract the illness), hired social media bots(?!)[2] to spin the party line, and only after all that, then went into lockdown. A veritable case of closing the barn door after the infectious horse has bolted.

Of course there's a bunch of other factors in death rates - the comparative proportion of your population which is especially vulnerable, medical care capacity, but it seems obvious to me that the largest factor will always be the number of infections in the first instance.

[1]: https://www.ft.com/__origami/service/image/v2/images/raw/htt...

[2]: https://www.thecanary.co/trending/2020/04/21/fake-nurses-pro...

> Timing is a very significant factor, I believe - I couch that as a belief, because I haven't had the time to compare all lockdowns

Fair enough, and I agree. It's clear that countries who clamped down super early have fared much better than everyone else.

But going by that FT graph, Spain went into lockdown a week before the UK, and still did worse? Germany later than the UK, and is doing better? There's a crazy amount of uncertainty about what good lockdowns are once you've passed a certain point of community spread.

Sweden obviously missed the point in time where lockdowns would have worked very well, while NZ, Australia, Norway didn't.

The UK obviously missed it as well, but while they were debating herd immunity, it might have been too late anyway? We might never know.

And with all that uncertainty, I think it's wrong for people to argue that Sweden "obviously" would have saved lives by going into a hard lockdown earlier. Maybe, probably, perhaps. Would a stricter lockdown now do anything in Sweden? Unlikely.

> Spain went into lockdown a week before the UK, and still did worse? Germany later than the UK, and is doing better?

Not relative to number of cases: Germany locked down later than UK, but a week earlier on the exponential growth curve.

I'll post this link again, because it's the best articulation of this point that I've found: https://medium.com/@tomaspueyo/coronavirus-act-today-or-peop...

Timing, with respect to where a country/region is on the exponential growth curve, is critical. In almost every place you see the same outcome, it's crazy.

> Not relative to number of cases: Germany locked down later than UK, but a week earlier on the exponential growth curve.

Not according to the Financial Times graph in the post I was commenting on. Germany went into lockdown at 3000 cases per day, while the UK went into lockdown at 700 cases per day, and Spain at 600 cases per day, according to that.

Is the FT graph wrong? Do you have a better data source that proves your point?

You're missing an implicit assumption - that each country's testing regimes and levels and criteria were comparable - for an example from the top of my head, some countries only test if you show symptoms.

So if you'll allow me to rephrase:

> Germany went into lockdown at 3000 known cases per day, while the UK went into lockdown at 700 known cases per day, and Spain at 600 known cases per day.

As you said earlier, so many damn factors in this.

To add more data, Germany locked down on Mar 17 at 24 known deaths. The UK locked down on Mar 23 at 336 known deaths. Germany reached 342 known deaths on Mar 27. That's 10 days of lead time on the deaths metric.

Germany's new deaths per day (smoothed) stopped increasing at 215 per day on Apr 9 (2.5 weeks later). The UK's death rate stopped increasing at 850 on Apr 15 (2.5 weeks later).

https://91-divoc.com/pages/covid-visualization/?chart=countr...

Deaths tend to be more accurate than just confirmed cases.

One way to account for testing rates is to look at the test positive rate, and use that as a proxy. If the test positive rate is 50% (as it was in Spain at some point), you can be pretty sure you’re missing a lot of cases.

So if Spain has “600 cases” but only did 1200 tests, they probably have way more real cases than Germany’s “3000 cases” out of 30,000 tests.

FWIW, Germany had 9 deaths on March 17, while UK had 10 deaths on March 14. By that metric, the UK led Germany by 3 days. There’s more to it than just the one national lockdown date — in particular, Germany is federal and I think different states went into lockdown at different times, as in the US.

But meanwhile, South Korea and Taiwan have avoided lockdowns entirely, so it’s clearly not only about lockdown timing — availability of testing, as in Germany, helps a lot with keeping the sick isolated.

Counterpoint: Santa Clara county locked down 3 days earlier than Los Angeles, but a week later on the epidemic curve. LA right now has had nearly twice the death rate per capita, and unlike Santa Clara which is almost at the end of the curve, may not even have peaked.

Lockdowns aren't everything..

Santa Clara has less population density and air pollution.

Santa Clara’s official lockdown, and the tech world’s “work from home” mandates were separated by about a week, which could account for some of this.

> and worth noting that Johnson and his advisor Dominic Cummings who was hugely influential in pushing for it both went on to contract the illness

Not really worth noting. It's an interesting anecdote, and I'm sure those who take the pro-lockdown position revel in the schadenfreude… but it has no relevance to the debate.

I does when Boris was going around boasting that he was "shaking hands with everyone" (in regards to a hospital visit). Then ends up in intensive care himself using up stretched resources.

I will give leaders who's job it is to meet lots of people to run a country a break, IF they take a reasonable position. Boris flaunting the risks to build up his public persona and then getting egg on his face is absolutely relevant. If you care in any way how a country is run.

Dominic is not so much in the public eye, but I bet he had a lot to do with Boris's early bravado.

It shows how widely it spread in the UK before they went into lockdown, which underlies my point that the efficacy of a lockdown is strongly tied to the timing thereof.

> If stricter measures always resulted in less deaths, as you argue, why are so many countries doing worse than Sweden?

The UK acted too late. Compare Sweden to Australia. Australia has 2.5 times the population of Sweden (25.6 million versus 10.3 million). Sweden has had 1,937 COVID-19 deaths and Australia has had 74. Sweden's deaths per million population is 192 and Australia's deaths per million population is 3. Sweden's rate of testing is 9,357 tests per million population and Australia's is 17,743:

https://www.worldometers.info/coronavirus/

https://coronavirus.jhu.edu/map.html

Australia's had fewer cases, fewer deaths, and is doing more testing than Sweden.

> we have 50% the population of Sweden, and 0.72% the deaths

After google, the percentage of elder in Swedden is like 20% and in New Zealand like 12%. If you have half of population, you would have half of elder with the same percentage, but is minor. You should have like 6 elder people by each 20 elder in Swedden.

Even if the covid would perform exactly the same in all countries and the measures taken would be exactly the same, the deathcount by million of people would be different.

Okay, so, to tweak it based on that, our most vulnerable populaton is 30% the size of Sweden's most vulnerable population, and we have 0.72% the deaths.

Other difference can be that you have passed the problem in summer-fall so far, whereas europe passed it in winter-spring. This means, for example, that water droplets in the air will dry much faster in your case

You are about to enter in flu season and the next months can be complicated, so take care for yourself and your family. This is not over.

Sweden unfortunately got a rather large injection of cases from tourists visiting northern italy in just the wrong week. NZ doesn't have the same close traffic with any early covid hotspot.

I'm not sure where you got that impression about the NZ cases.

A bit over half of New Zealand's early confirmed cases were directly linked to recent overseas travel. Although the data [1] is a bit awkward to analyze, in weeks #12-14 (15 March .. 4 April) 451 cases out of the 883 total (51%) were classified as "overseas travel".

The data only includes the country of origin of the final leg of their international flight. Since there aren't any direct Europe-NZ flights, some of the cases with origins in Australia/USA/Asia could be due to passengers returning from Europe via an intermediate airport.

[1] https://www.health.govt.nz/our-work/diseases-and-conditions/...

The number of Swedes traveling in general and especially traveling to virus hotspots like Italy is far higher than the number of Kiwis doing it.

You know, maybe, just maybe, being a remote island so far away from the rest of the world it's often forgotten on world maps helps with reducing transmission vectors in pandemics.

The specific assertion was that "NZ doesn't have the same close traffic with any early covid hotspot", but that's demonstrably false.

Perhaps there were more Swedes traveling in general, but that is neither here nor there. [There were an average of 16,300 people arriving in NZ each day in early March.] NZ did in fact have plenty of overseas arrivals returning from various early hotspots (such as New York and Italy) and this is reflected in the confirmed case data showing that a whopping 39% of all of our cases were infected during their overseas travel.

Furthermore, every single significant cluster of cases was kickstarted by an index case that had recently returned from overseas.

It wasn't distance or being an island that protected NZ from experiencing an epidemic outbreak, but rather the rapid action from our government: travel restrictions, isolation/quarantine of arrivals, and strict lockdown of the entire country.

New York was not a particularly early hotspot, it was much later than Italy.

We shut down flights from China in February for that reason, and China is a masssive source of tourism for us - we also had the plague ship Royal Princess dock here and spawn a cluster, all of our largest clusters came from overseas contact - either from tourists or returning Kiwis.

Ummm...

> There is clearly no evidence that stricter measures have any effect. If that was true, then there should at least be a linear correlation between strictness of measures vs. excess deaths, and that's simply not the case.

What? No, timing of measures is critical here. [0] The same exact strictness of measures, implemented a week later, can cause 4x more deaths (total, and thus per capita too) merely because of the delay.

Yes, English and Danish authorities should absolutely be terrified that they waited to long to act, just as the Swedish authorities should be.

[0] https://medium.com/@tomaspueyo/coronavirus-act-today-or-peop...

> Denmark has ~2.5x the per-capita covid-19 mortality of Finland

...and ~35% the per capita mortality of Sweden. Denmark is pretty much in the middle of the pack. That's clearly less reason for panic than if they were already in the top third and rising fast. And you accuse others of cherry-picking?

> Sweden vs. all of Europe paints a completely different picture

No, it really doesn't. Thanks for proving that this kind of jingoism isn't limited to the US.

It's not just that Sweden has a higher death rate but that Sweden's death shows a sharp exponential rise still. This effectively in the last few days, a week ago Sweden could be pointed to as a "miracle".

The other countries that have seen these exponential indeed effectively panic and impose any measure they can - social distancing and lock since that's all they have.

As people point out, there's no absolute evidence that quarantines, lock-downs or parachutes save lives yet common sense still dictates they do more than nothing and when a disaster is approaching, do you really want to say you failed to do that could have averted mass deaths?

Moreover, the Sweden versus the other Nordic countries does seem like a better comparison because these countries arguably have similar levels of relative isolation from the world, similar health care systems and similar cultures.

> is that if a country has no or a low rate of excess deaths, that country is doing "ok".

Where are you getting figure for all cause mortality for Sweden please?

> should English authorities be terrified that they've gotten their assumptions wrong?

There is wide-spread condemnation about the UK approach - too little, too late.

Compare countries of similar population:

pop 10.7e6 https://covid.observer/gr/#daily

pop 10.7e6 https://covid.observer/cz/#daily

pop 10.3e6 https://covid.observer/se/#daily

pop 10.3e6 https://covid.observer/pt/#daily

pop 9.8e6 https://covid.observer/hu/#daily

pop 8.9e6 https://covid.observer/at/#daily

pop 8.6e6 https://covid.observer/ch/#daily

I want to point out the high absolute numbers of deaths, the deficient recovery, and that the curve shows no decline in Sweden.

I would be completely unsurprised that when this is over, an angry mob captures Tegnell and tears him limb from limb while alive.

Reply on news.ycombinator.com