I wish this article went into more detail about the relative death rates.
The fact that Sweden’s death rate “far outstripped that of other nordic” countries (all else being equal) would tend to indicate the Swedish strategy failed, no?
Are there important confounding variables that would explain the higher death rate?
Yes, here's a Norwegian site comparing the different Nordic countries per capita[1]. Other than Iceland being an outlier due to it being tiny, Sweden outstrips all other countries on pretty much any measure by a wide margin.
It seems like the number of patients under intensive care is slowly falling for Sweden but they're not anywhere close to where the other Nordics were months ago.
You can't quite compare countries like that and claim that Sweden would have had the same rate as the other. The world is full of examples where neighboring countries have radically different caseloads. I don't know enough to meaningfully discuss the details but I suspect you can't just claim Norway is the same as Sweden.
What is most remarkable is the trajectory the epidemic took within Sweden. No comparison is necessary to another country. That they have achieved a substantial reduction of their caseload without any of the invasive measures. Nobody would have ever believed that to be possible!
Other countries should learn from Sweden rather than being in denial - it demonstrates that other measures can be just as effective, and we need to understand how that was possible and whether the measures are transferable.
But if their economy did worse than their neighbour, what was the point of avoiding invasive measures? They have more deaths, more cases, and worse economic output, what exactly is successful here?
Quality of life. Additionally "our neighbours are doing something wrong and bad so we should too" is not a justification.
Lockdowns cannot stop COVID deaths and in the beginning nobody claimed they could. The idea that Sweden has performed worse than Norway is a massive goalpost shift that is rightly being called out as nonsense. Especially because it's clear how much corruption there is in these figures. Did the UK really do so badly compared to its neighbours, or is that a result of the health body defining a COVID death as anyone who ever tested positive and then died at any later point, of anything at all (i.e. all COVID infections are fatal according to UK stats).
Sweden has a population of ten million people and an economy very tightly tied to the rest of the EU, so exactly what measures we did I don’t think made a huge difference for the economy.
If transmission is from cough and sneeze droplets this would make sense, as it spreads fastest when transmitter has covid and they have a cold or the onset of the flu.
Yeah, Sweden has a remarkably high death rate due to very old populations like 80+ or 85+. I mean, really high, it's been pretty terrible. So would probably be quite different and more "normal" numbers if it hadn't hit our nursing homes like it did. Death rate is very low in comparison if you instead look at figures for age 70 and lower. This problem has also been acknowledged as probably the greatest problem
and a failure during the coronavirus crisis by Swedish epidemiologists despite "Protect the elderly!" being an early, daily mantra.
It's been asked why Sweden then had this problem and not our Nordic friends. Some answers might be understaffing or underestimated equipment needs. When the crisis hit them, there was already a global shortage of disinfectants etc.
It's also been asked how the virus entered the nursing homes to begin with. A too liberal stance for visitors a bit too long? Or brought in by staff using too crowded public transports?
Many questions and the investigations are still ongoing.
Compounding the death rate problem in Sweden was probably also a lack of centralized detail management and confusing directives that even led to palliative care for elderly in nursing homes not even being covid-19 victims. In Sweden, nursing homes work quite independently from each other, and even hospitals are strongly regionally controlled rather than directly state controlled even in a crisis like this. (there are some exceptions though, in these catastrophies, staff can be flewn across the country to help man understaffed hospitals etc)
tl;dr I think nursing home infections are the single outstanding factor for the abnormal Swedish death rates. If they are infected, they already have particularly weak immune systems that can often not even begin to fight this beast off. Hell, they probably often already have respiratory diseases and/or use immunosuppressants.
Many other countries (especially the US) also screwed up their handling of old folks homes.
Also, I don’t know what data you are citing, but note that many data sets are normalized by number of estimated or confirmed of cases, not by total population size.
It's too early to say the Swedish strategy failed, because the Swedish strategy was "do badly at the start to do better later", and "later" isn't finished yet.
On the flip side it would be possible to say it succeeded, if it did so much better that they are already better off than the rest of the world, but I don't think we're really there yet either (and the epidemiologists point is really just that there are signs that they might be there in the future).
> It's too early to say the Swedish strategy failed, because the Swedish strategy was "do badly at the start to do better later", and "later" isn't finished yet.
So, at this moment we are only celebrating that Sweden succeeded at the "do badly at the start" part?
To me it seems that Sweden's strategy has been celebrated since the first day, and the celebrations never stopped. At this moment I wonder what would it actually take to make the celebrations stop.
Right. The denominator in rate is per "unit of time" and we dont know the period. The two ways you could compare them are equal points on a bell curve, or equal time periods. What you can't do is compare two instantaneous speeds and then argue that one car has a faster average speed.
I found it interesting that the general consensus seems to be that Sweden failed, and yet NY state was just hailed as "doing it correctly" by Dr. Fauci [0] (despite one of the very highest per-capita death rates in the world). We didn't do much better here in MA in terms of deaths, although like NY we're only seeing a very small (and importantly flat) number of cases here.
Meanwhile Australia, a country that was previously hailed as having done it right and contained COVID-19 has now locked down 6.6 million people [1], closed state borders between Queensland and NSW [2] and currently seeing record numbers of infections [3].
In Australia, there has been an increase due to failings of the state government of Victoria in poor practices by guards of returned travellers in forced quarantine resulting in its spread, particularly to public housing apartment complexes[1]
As this jump occurred during school holidays, it looks like Melbournians have spread it to NSW as well which is why there is a jump in NSW. [2]
Australia would still be in a relatively good position if it wasn't for the incompetence of the Andrews Government.
Although to get slightly meta about this, the core blame lies with the 25 years trend of Australian governments outsourcing or selling off every possible previously-government-owned service they can (mostly to China-owned corporations) for a short term sugar hit of revenue for the currently incumbent politicians, to make themselves look good, without caring about the negative long term consequences to government revenue.
Profitable government monopolies like the Land Titles Registries have been sold to private international owners, who then raise costs for citizens, in order to recoup their investment faster.
Politicians strongly favour awarding outsourced contracts to their donors. In this case without opening the contract up for tender or competitive bidding, and obviously without much due diligence
e.g. I read that one of the security firms donated $76k to Dan Andrews party last year.
So government outsources work to contractors who subcontract the work to independent ABN holders, who in turn can subcontract the work out to their employees, who in turn might be subcontractors themselves to the last company in the chain.
Politicians and public servants are normally happy with this arrangement because they can then wash their hands of blame for any mistakes, and pin it on the outsourcers who stuffed up.
In this case though, the blame is serious enough that it has jumped the subcontractor firewalls and burned its way all the way back up to the top (Dan Andrews).
Right, but its a bit of role reversal since Victoria is run by a Labour government which outsourced this to poorly trained security guards, whereas NSW with its Liberal (read conservative for Americans) government got the police to do it.
Keep in mind Sweden's most populous city (Stockholm) has a population of 1 million. NYC has a population of 8.4 million. Not to mention Sweden is culturally homogenous while NY is extremely diverse.
I was merely saying that the consensus seems to be that Sweden has failed in it's experiment and been irresponsible in allowing a higher per-capita death rate compared to it's nordic neighbors. However it seems that NY (state) is now being held up as an example of a good response -- specifically with Dr. Fauci called Gov. Cuomo's response "doing it correctly" (despite 32K deaths many of which it seems might have been directly related to a policy of discharging COVID patients into nursing homes [0]).
> found it interesting that the general consensus seems to be that Sweden failed, and yet NY state was just hailed as "doing it correctly" by Dr. Fauci [0]
You should read the sources you quote. Dr. Fauci's comments are on NY's approach on "how to successfully bring down cases of coronavirus." It says nothing about what enabled the epidemic to spread like wildfire, which is Sweden's case.
Taken right from your own source:
"Again, we have a problem. We need to admit it and own it. But we have got to do the things that are very clear that we need to do to turn this around, remembering we can do it," Fauci said."
I did read it. My point was that it seems to me that Sweden and NY are at about the same place now. Sweden basically just let it happen and NY locked down and imposed significant measures. Both had significant deaths in the elderly population (although significantly more in NY than Sweden) and both now have very few new cases & deaths, so the outcome has been similar.
Also interesting is that CA took very similar measures to NY with very similar timing and now are seeing the exact opposite to NY (although the media seem to be a bit quiet on that!).
> Also interesting is that CA took very similar measures to NY with very similar timing and now are seeing the exact opposite to NY
CA did not act with similar timing relative to disease progression, whether on initiating the lockdown or reopening; particularly, it began quickly reopening when cases had barely flattened, earlier than New York either on the calendar or with regard to the progression of the epidemic, whereas NY didn't start reopening till they had dropped way down and still hasn't opened indoor restaurants in NYC, while California had reopened bars and restaurants statewide before beginning reclosing.
> although the media seem to be a bit quiet on that!
California's woes resulting from the premature reopening are heavily covered in the national (and even international) media. If you mean that the false claim that California policy closely tracked New York’s and got different outcomes isn't getting media play, you are right, but that's only because it's completely false.
New York was somewhat slower on the calendar and even moreso with regard to cases in closing, which was a big initial failure relative to California, but California bungled even worse with rushed reopening.
Was going by the information on rt.live regarding closing and re-opening for NY and CA, it shows the following dates (and may well be incorrect):
| Shelter Start | Shelter End | Reopening Start
-------+---------------+-------------+------------------
CA [0] | Mar 19 | ? | May 25
NY [1] | Mar 22 | May 28 | May 15
Agreed that NYC itself didn't re-open at the same rate up but my understanding is that it was similar in CA (county by county and the more populous counties such as the SF bay area did not re-open as much).
For sure though, when NY locked down the community spread was likely far greater than it was in CA which I think shows in the C19 related deaths.
> Was going by the information on rt.live regarding closing and re-opening for NY and CA, it shows the following dates (and may well be incorrect):
For California, that's definitely wrong. California moved into Stage 2 reopening on May 8.
> Agreed that NYC itself didn't re-open at the same rate up but my understanding is that it was similar in CA (county by county and the more populous counties such as the SF bay area did not re-open as much).
That's loosely, qualitatively, true, but even most of the more populated counties had reopened more than NYC, and several counties had variances to go ahead of the statewide reopening.
Thanks for clearing that up. Hmmm ... seems to be a lot of misinformation out there even from "good" sources, such as Johns Hopkins and the NYT. For example regarding the situation in Mass:
In Sweden immigrants from Iraq, Syria, Somalia and other countries form a large part of deaths and ICU patients.
It will be interesting to see why this is eventually.
There’s language barriers, people living in close quarters and perhaps the vitamin D thing as well?
Poorer populations are dying of COVID at higher rates throughout the world. In the US, at least, the lockdown is mostly protecting the rich.
I wonder if Sweden’s approach is more economically and racially equitable than full lockdowns. I’d love to see a study of this that’s backed by quality data.
Sure they are but significantly more people deficient in vitamin D are dying as well for reasons that have little to do with Poverty. Complete racial equity is likely impossible.
People always jumps to “it must be poverty!” when the Vitamin D thing comes up.
There is a strong correlation between Vitamin D levels and COVID-19 outcome, regardless of skin color. Independently, there is a strong correlation between having low Vitamin D levels and living in an environment to which dark skin is not adapted to.
That alone should be enough to raise some caution, there is no need to wait for better data because Vitamin D supplementation is in order either way.
Unfortunately, the media is mostly silent on this, likely because it wants yet another social justice narrative instead.
> Evidence was from 5 published studies in peer-reviewed journals. One observational cohort study (D'Avolio et al. 2020), 3 observational prognostic studies involving published data sets using correlation or regression (Hastie et al. 2020, Ilie et al. 2020 and Laird et al. 2020) and 1 case-control survey (Fasano et al. 2020) looked retrospectively at the association between vitamin D status and development of COVID‑19. None of the studies were intervention trials of vitamin D supplementation for the prevention or treatment of COVID‑19.
> Four of the studies found an association or correlation between a lower vitamin D status and subsequent development of COVID‑19. However, confounders such as body mass index (BMI) or underlying health conditions, which may have independent correlations with vitamin D status or COVID‑19, were not adjusted for (D'Avolio et al. 2020, Fasano et al. 2020, Ilie et al. 2020 and Laird et al. 2020). Vitamin D status was based on serum 25‑hydroxyvitamin D (25(OH)D) levels in 3 studies and the proportion of participants taking a vitamin D supplement in 1 study. The largest UK study (Hastie et al. 2020) found an association between vitamin D status and COVID‑19 only in a univariable analysis (with this single potential causative factor). Importantly, no causal relationship between vitamin D status and COVID‑19 was found after adjustment for confounders such as comorbidity, socio-demographics, ethnicity, BMI and other baseline factors.
"The largest UK study (Hastie et al. 2020) found an association between vitamin D status and COVID‑19 only in a univariable analysis (with this single potential causative factor). Importantly, no causal relationship between vitamin D status and COVID‑19 was found after adjustment for confounders such as comorbidity, socio-demographics, ethnicity, BMI and other baseline factors."
A couple of problems with interpreting that study:
First, it only looks at infections, not course of infection. The million-dollar question is not whether Vitamin D prevents an infection altogether, but how high the risk for a severe course of COVID-19 is, given a Vitamin-D deficiency.
Secondly, it adjusts for ethnicity and the "Townshend deprivation index" as a proxy for socio-demographics.
You have to look at the huge impact of being classified as "black" (i.e. very dark skin color), and the large impact of being classified as "south asian" (likely moderately dark skin color). No other factor besides "poor health" even comes close. These people are largely Vitamin-D deficient, so by controlling for that factor, you are implicitly lowering the impact of Vitamin-D. It's like saying "Oh yeah your Vitamin-D levels are low but really the problem is that you're black". If you look at the controlled numbers, being insufficient in Vitamin-D becomes protective (OR ~= 0.9), which doesn't make any sense.
"Townshend deprivation" measures household overcrowding and lack of access to a car, which are obviously factors that increase the risk of contracting any viral infection. Again, this shouldn't be controlled for in order to answer the question.
Of course a single-variate analysis is not enough, because many diseases are causal for low Vitamin-D levels, but a multi-variate analysis with too many variables isn't meaningful.
Is it valid to treat ethnicity as a confounder for vitamin D given that the belief is dark skin colour in northern climates results in lower vitamin D?
> Independently, there is a strong correlation between having low Vitamin D levels and living in an environment to which dark skin is not adapted to.
Honest question: what skin color is actually adapted to western Europe or much of North America? Virtually everyone I know with a lighter skin tone needs to apply sunscreen for any outdoor activity to avoid rapid sunburns, and virtually everyone I know with a darker skin tone (jokingly or not) suspects they have a Vitamin D deficiency.
There is no one such skin color for the latitude range you are describing.
Just look at the “average” skin color of whatever ethnicity has been in a particular environment for tens of thousands of years. Native Americans clearly have darker skin than Anglosaxons.
Also, much of western Europe has been colonized by peoples that previously spent thousands of years up further north. The weather in Britain is different from France, and so forth.
I believe one of the Swedish epidemiologists who helped engineer Sweden's COVID response said not to compare the death rates today, but a year from today and see how they compare then.
This is not really true. For one thing both cases and deaths have been strongly clustered so that population level rates are not so interesting. What matters is whether your ancestors adapted to life in northern latitudes, what your weight, insulin resistance, and vitamin D levels are, and how densely urbanized your environment is. Another factor is that Sweden aggressively labeled all hospitalizations and deaths involving a COVID-19 diagnosis as being COVID-19 cases and deaths even if the patients were very old and had chronic health problems. This is a major difference between both Denmark and Finland which tended to be very cautious about labeling hospitalizations and deaths as COVID-19 related and had low testing rates as well. Looking at all cause mortality figures yields very different results.
Sweden saw the same level of damage to their export-oriented industries as their neighbours did, but less damage to the service industries that are directly affected by a lockdown and less of a decrease in mobility: https://www.imf.org/en/News/Articles/2020/06/01/na060120-swe... You can't just lump the two together and claim it shows that not locking down causes about the same level of economic damage as locking down, especially given that those export losses are presumably directly attributable to their neighbours' lockdowns.
This indicates Sweden wound up with relatively similar de-facto restrictions (a social distancing strictness of ~50 versus ~60 for the other Nordic nations) and a substantially higher death rate.
It also says "Sweden did not record an economic contraction in Q1 of 2020", which is a misleading claim, given the lockdowns largely came in the last few weeks of that quarter, and later goes on to acknowledge "Forecasters agree that Sweden will face a severe recession in 2020".
I don't think this really supports the case you're trying to make.
The problem is they had a higher death rate than the neighbors on the same timeline. Oh and their economy also suffered just as bad as neighbors who locked down. Not sure how it could be a success with more deaths and everything else being the same.
I'm not trying to predict the future, I'm just saying it's possible.
At the same time, it's not possible at this time to declare "sweden's strategy has failed" - there's plenty of time for other countries to catch up on excess mortality.
I do think there's a very real chance that, assuming hospitals are not overwhelmed, there won't be a lot of difference in overall mortality between developed countries. This disease is, apparently, super infectious after all, and eradication seems unlikely.
I wasn't saying it was a failure, but that it is hard to call it a success. Again I think we will need to wait a while to determine true successes or failures (although I think here in the US it is trending towards being a tremendous failure).
> Before the introduction of measles vaccine in 1963 and widespread vaccination, major epidemics occurred approximately every 2–3 years and measles caused an estimated 2.6 million deaths each year.
> Herd immunity was first recognized as a naturally occurring phenomenon in the 1930s when A. W. Hedrich published research on the epidemiology of measles in Baltimore, and took notice that after many children had become immune to measles, the number of new infections temporarily decreased, including among susceptible children.
Notably there are substantial excess in 2018 or 2019 during the same weeks. Not quite sure what to make of that.
If you were subtracting 2020 from 2018 then it looks like you would end up with 80 thousand fewer deaths in the first 10 weeks of 2020. What is that about.
The death numbers is not counted the same way in Sweden and Norway, in Sweden they count every one that had covid-19 even if they died of something else after they recovered from corvid-19, they do not count it that way in Norway.
This theory can be dismissed by looking at excess death counts. A very clear spike in overall, country-wide deaths - of all kinds - can be seen in Sweden that is not seen in Norway.
But Sweden has roughly double the population of Finland and Norway, so Sweden (1000 weekly deaths) is comparable to Finland (500 weekly deaths) and not that far off of Norway (200 weekly deaths).
Irrelevant. It's how far the spikes go above normal per-capita that matters, which is why each chart has a different Y axis that fits that particular country.
The pictured charts clearly show Sweden and the US seeing substantially more deaths than normal, while Finland and Norway stayed at roughly their normal rates.
On the other hand as of ~3 months ago it wasn't enough to show all the typical covid-19 symptoms to be allowed to get tested at all (in the Stockholm region at least).
They have since expanded testing, so in principle anyone can get tested now (if they can manage to book a time slot at least..).
I know for sure that a lot of the deaths are from nursing care homes (but I don't know in how large percentage), something seen also elsewhere in the world.
> The fact that Sweden’s death rate “far outstripped that of other nordic” countries (all else being equal) would tend to indicate the Swedish strategy failed, no?
Why are we only comparing Sweden to other nordic countries?
They pushed their deaths forward by dealing with the pandemic with a light hand. I have a feeling two years from now the total damage will be comparable to other countries in Europe.
The comparison with other "nordic" countries seems arbitrary. If we go by latitude, why is there such a huge variance in death rates between Belgium, France, Netherlands and Germany?
The biggest factor is probably outbreaks in retirement homes. I think you can make the argument that the strategy failed to protect these elderly people, but it might also be the conditions in these facilities that are to blame.
Yes, in Sweden the problem was the nursing homes and they were run (and still are run) badly. The other measures that could have been done (within the covid response itself) would probably not have helped this, goes the argument (but certainly a major overhaul of how the nursing homes are staffed and run would have helped - that would never have been possible in the short timeframe though).
We don't know. If there is no vaccine ever, and no long term immunity from getting it, then it doesn't matter what you do every country will eventually be infected and have the same death rate. Sweden just happens to be there first but everyone will get there.
Of course my assumptions are big. If there is a vaccine then it is personally worth a lock down to not get it. If there is herd immunity I want everyone else to get it while I personally lock down.
Lockdown kills too. In Australia there were about 700 excess deaths from January to April 2020 due to people being unwilling or unable to access normal medical care:
The fact that Sweden’s death rate “far outstripped that of other nordic” countries (all else being equal) would tend to indicate the Swedish strategy failed, no?
Are there important confounding variables that would explain the higher death rate?