Talk:COVID-19 pandemic in Sweden/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Why infections started to decline in march/april?

Is there any good analysis what would try to explain why infections started to decline? Initial explanation was that there started to be enough people with immunity, but it doesn't seem to be case based on serology testing. Another explanation could be that the impact from social distancing and voluntary rules would be enough to turn the tide or there would be a crossimmunity or there is seasonal change how easily it will infect people, but is somebody tried to validate the explanations? --Zache (talk) 09:58, 26 June 2020 (UTC)

I read somewhere that the virus is gradually mutating to turn less deadly over time, since killing the host bodies makes it less efficient at transmitting to new ones. Hence, despite more people getting infected, the death rate continues to go down. David A (talk) 11:57, 26 June 2020 (UTC)

@David A: We are talking infections, not deaths, as per the title of the thread.

@Zache: What are the results of serology testing that pertain to reduced rate of infections? What article, and what numbers? --Dan Polansky (talk) 09:07, 27 June 2020 (UTC)

@Zache Same as in every European country. Self-isolating if you have any symptoms, improved hand hygiene, social distancing. And common sense really. There's never been any voluntary rules, everyone is expected to follow the advice from the agencies (see note 2). Besides, there's been several new laws. Regarding the decline, according to the Health Agency the decline began in April. Your second explanation is the same as theirs, see the weekly reports here. bladjur (talk) 21:41, 27 June 2020 (UTC)

Having looked again at Talk:COVID-19 pandemic in Sweden#New cases, I am not sure what this thread is about: I see no decline in daily cases in Sweden in March and April. But should we look for a relating figure that started to decline, we can consider the ratio of daily positive cases to daily tests, from FOHM report[1], Figur 1A. Andel positiva fall bland provtagna individer per veck. There, the peak is on week 15, which per one week numbering is the week ending on April 18 ("gcal -K Apr 2020"). The question then would be, what made the ratio stop increasing in week 15? Since, if the epidemic were increasing exponentially or at least superlinearly, there would be no such stop. I don't think the above response gives us a correct answer, let alone testable correct answer. As an aside, the ratio is more useful than the raw daily case count for some purposes since it removes the confounding effect of increased testing from the figure. --Dan Polansky (talk) 15:52, 29 June 2020 (UTC)

As sweden didn't do extensive testing before june the hospitalizations, ICU (and deaths in lesser extend) are a good proxy for how widespread the infections are. Even with extensive testing hospitalizations may be better proxy for the trends than testing alone. --Zache (talk) 21:01, 29 June 2020 (UTC)

What do you mean by "infections started to decline in march/april"? Do you mean that daily covid-relating ICU admissions started to decline? What observable event are you referring to? --Dan Polansky (talk) 11:03, 30 June 2020 (UTC)

They mean that the infection rate (number of new cases/people with an ongoing infection) peaked in April. This is based on several observable events, new/active ICU cases is only one of them. Surveillance is another (at a point in May, only 0.3% had an ongoing Covid-19 infection, compared to 0.9% at a point in April). The number of new ICU cases seems to have eached a peak in April with a seven-day average of >40 new daily hospitalizations and 500 active hospitalizations. Note that average time between onset of disease and ICU is 10.8 days and average length of ICU stay probably is several weeks. Source. See combined bar/line chart below the section "Patienter i intensivvården med Covid-19". You can change view (Ändra vy) between daily new cases (Nya) and total active cases (Antal).bladjur (talk) 17:53, 30 June 2020 (UTC)

Okay. If we replace the question "Why infections started to decline in march/april?" with "Why daily confirmed cases stopped growing in the beginning of April", consistent with the graphs in #New cases, one at least somewhat plausible hypothesis is that the achieved degree of immunization is what slowed the spread. This is in view of #Seroprevalence and prevalence in Stockholm, which cites a preprint indicating that, in May, about 30% people in Stockholm could have covid-specific T-cells (but please double check). Admittedly, it would be less than 30% in early April. Be it as it may, our understanding of the ways of immunization may be incomplete, and the slowdown-via-immunization hypothesis should not be outright dismissed. --Dan Polansky (talk) 09:13, 1 July 2020 (UTC)

As for "[...] the hospitalizations, ICU (and deaths in lesser extend) are a good proxy for how widespread the infections are": Neither daily new hospitalizations nor deaths are necessarily a good proxy for true infection growth if mild cases grow much faster than severe cases, and that kind of growth pattern seems to be the case in weeks 15-25 as per FOHM report[2], Figur 1C. Antal bekräftade fall av covid-19 per vecka i Sverige fördelat på lindriga och allvarliga fall. As per that figure, let us point out that while new daily infections did stop growing in early April, their growth resumed in week 23 (as a result of increased testing); at the same time, the serious new daily cases did stop growing and started to decline in week 15, mid April. --Dan Polansky (talk) 07:59, 4 July 2020 (UTC)

I lost you. You seriously try to explain that there is now more mild cases now than in mid april (and not that there were more infected people in mid-april including mild ones but they were only testing the severe ones)?--Zache (talk) 13:26, 4 July 2020 (UTC)

Given the data reported in Figur 1C, it is reasonable to think that the actual new daily severe cases are going down, whatever is true of the actual mild cases. To think that daily ICU admissions are a good proxy of the real new daily infections, both mild and severe, is to assume that there is no increase of measures to protect the vulnerable, that those vulnerables who can are not increasing their self-protection in response to publicly reported news in the media, and that there is no way for the infection to run out of easily infectable vulnerables such as those in nursing homes. I don't see a reason to jump to these assumptions, and therefore, I am wary of thinking that new ICU admissions are necessarily a good proxy. --Dan Polansky (talk) 18:06, 4 July 2020 (UTC)

All-cause deaths per 100 000 pop in Jan-May chart

All-cause deaths per 100 000 pop in Jan-May chart calculated from SCB[3][4][5]:

Inspired by a tweet by HaraldofW.

Data sources and calculation:

• The population is from http://www.statistikdatabasen.scb.se/pxweb/sv/ssd/START__BE__BE0101__BE0101A/BefolkningR1860/.
• The deaths per month up to 2019 are from http://www.statistikdatabasen.scb.se/pxweb/sv/ssd/START__BE__BE0101__BE0101G/ManadFoddDod/.
• Daily deaths for 2020 are from https://scb.se/hitta-statistik/statistik-efter-amne/befolkning/befolkningens-sammansattning/befolkningsstatistik/pong/tabell-och-diagram/preliminar-statistik-over-doda/.
• Months Jan-May are included, that is, Jan 1 to May 31. June is not included since 2020 data are incomplete for it.
• For deaths in months in, say, 2005, the population for 2004 was taken as reference.
• One might object that a whole infection season should be taken, say, Nov previous year to May current year. That seems to be a valid objection. Still, at worst, by selecting Jan-May we are doing year 2020 a disfavor: its worst death count in 2019/2020 season is in April.
• One might object that relating the deaths to population size would mislead if population were significantly aging or becoming significantly more young, or if the structure of vulnerability to infections would change in a different way. Addressing this objection would require more work.

For reference, Sweden's population[6]:

And here are deaths in Sweden in Jan-May not related to population:

The advantage of the above is that it addresses any objections concerning relating the deaths to population size. However, dealing with the absolute death numbers in disregard of population growth does not give the perfect picture either.

--Dan Polansky (talk) 11:19, 4 July 2020 (UTC)

To give a picture for the complete infection season, deaths per 100 000 pop in Oct-May in Sweden from the same sources as above:

Notes:

• The year indicated above is the main year of the season, that is, the year for which Jan-May data is taken.
• The starting month was chosen to be Oct rather than Nov, following CDC source[7] that includes Oct in U.S. flu season. The CDC source shows that Apr and May were never seasonal flu activity peaks in 1982-2018.
• Inspired by tweet by HaraldofW, which has the season start in Nov.

And here are deaths in Sweden in Oct-May not related to population:

--Dan Polansky (talk) 12:07, 4 July 2020 (UTC)

We can put the above in contrast to Spanish flu by plotting different years for Oct-May not related to population:

We can see that the impact in season 2018/2019 (2019 on x-axis) of what we would think is Spanish flu is very pronounced.

--Dan Polansky (talk) 08:16, 5 July 2020 (UTC)

You can view more relevant and revealing charts and visuals, including interesting map of Swedish population density, here:

• Guest Blogger @HaraldofW – All you ever wanted to know about Corona Sweden, July 7, softwaredevelopmentperestroika.wordpress.com

The above is not a reliable source for Wikipedia's purposes. Nonetheless, we can plot the graphs ourselves from SCB and other sources, if desired; the graphs indicate their data sources as URLs to ease reproducibility, although they usually do not link the population data source, which would be this[8]. --Dan Polansky (talk) 08:11, 8 July 2020 (UTC)

Sweden's strategy failed: Much higher death rate than in Norway or Denmark, per capita

<<<Sweden never locked down during the coronavirus pandemic>>> 

An editor who has worked on this article should update it. (With some clear-cut information instead of statistical jargon.) Some articles from the past week:

Scientist admits Sweden could have battled virus better |The country's former state epidemiologist, Annika Linde, said that in retrospect she believes an early lockdown could have saved lives   

Mortality rate: 8th highest number of coronavirus-related deaths per capita in the world ... four times more than neighbouring Denmark and 10 times higher than Norway

Their plan was based on Herd Immunity:  BUT  fewer Swedes have developed immunity to the disease than had been hoped  

Sources: https://www.ctvnews.ca/world/sweden-didn-t-lock-down-but-economy-to-plunge-anyway-1.4973195 

https://www.ctvnews.ca/health/coronavirus/scientist-admits-sweden-could-have-battled-virus-better-1.4967014 

https://www.politico.com/news/2020/06/11/sweden-coronavirus-312838 Peter K Burian (talk) 13:47, 14 June 2020 (UTC)

Did the strategy save their economy? No

Sweden Has Avoided a Coronavirus Lockdown. Its Economy Is Hurting Anyway. (Wall Street Journal)   https://www.wsj.com/articles/sweden-has-avoided-a-coronavirus-lockdown-its-economy-is-hurting-anyway-11588870062

Sweden Says Covid Strategy Was Never About Shielding the Economy Sweden’s softer lockdown has resulted in one of the world’s highest death rates, relative to the population. At the same time, Andersson said in May that her country was facing a “very deep economic crisis,” with GDP set to sink about about 7%, despite the softer lockdown. ... Sweden’s recession is likely to be roughly as bad as that in the European Union https://www.bloomberg.com/news/articles/2020-06-09/sweden-says-covid-strategy-was-never-about-shielding-the-economy Peter K Burian (talk) 13:55, 14 June 2020 (UTC)

@Peter K Burian: Regarding the economy and recession, See Finance and the economy. Also see sixth paragraph in Strategy. Btw I looked at your links, it left me curious how they think "soft" or "hard" lockdown could matter much to an economy which relies on exports, or when people stayed at home without someone physically forcing them to do so. But I was also reminded that the narrative in the US is very different. Regarding "plan based on herd immunity", see sixth paragraph in Strategy and also Media coverage. Regarding immunity, I agree Monitoring and modeling needs to be updated. But it would probably be a waste of time to do it before Thursday. Regarding mortality, I'm not sure what you're after? The introduction informs the reader that "the number of deaths with confirmed COVID-19 has been significantly higher in Sweden compared to most of Europe, including other Scandinavian countries". Sourcing "8th highest" is likely to be impossible. Regarding Linde, although she received much coverage with her comments on the strategy, so did a lot of others. The Debate and criticism section exists for a reason, but WP:WEIGHT makes it quite a challenge to edit. However I think WP:BOLD is a good way of tackling it. bladjur (talk) 21:53, 16 June 2020 (UTC) (feedback welcome)

Just a comment about the economy: It was already in quite bad shape compared to the rest of Europe before this crisis hit, and our experts issued warnings that we would hit between 20% and 40% unemployment if we had a widespread shutdown for several months, so although the country is still hit hard, it would have been hit much harder otherwise. David A (talk) 12:57, 18 June 2020 (UTC)

Can you add this to the article? bladjur (talk) 10:58, 20 June 2020 (UTC)

I am not good at figuring out how to properly incorporate new information into preexisting articles, but here are two references, so you might be able to handle it yourself instead: [9] [10] David A (talk) 22:01, 20 June 2020 (UTC)

Here is more information about that Sweden was already in a quite bad position financially before the pandemic hit: [11] [12] David A (talk) 22:11, 20 June 2020 (UTC)

Is somebody willing to incorporate this into the article? David A (talk) 09:04, 24 June 2020 (UTC)

That updated table must be wrong, a jump of 10,000 in 10 days?Slatersteven (talk) 13:05, 18 June 2020 (UTC)

Testing numbers was made into politics a couple of weeks ago. As a result, Sweden went from only testing those with severe illness and healthcare workers to "free antibody tests for everyone" bladjur (talk) 22:21, 18 June 2020 (UTC)

The title of this section is not an accurate reflection of facts, to say the least. The Swedish strategy was a mitigation strategy, that is, one aiming at preventing healthcare overload. The strategy met its key objective of preventing healthcare overload well, and is therefore at least a moderate success. Excess death charts and related charts on this very talk page show that the health outcomes in Sweden, while unfortunate, are nowhere near the catastrophic forecasts made by some pseudo-experts. In retrospect, some elements of the strategy appear worthy of improvement: Sweden could have banned visits to nursing homes earlier and it could have been much more aggressive in nursing home protection.

There appears nothing analytically honest about comparing Sweden only to Scandinavian countries; if you want to assess the efficacy of a certain kind of strategy, you must not cherry pick items to include into your analysis, and you must include both poor-outcome heavy lockdowners and good-outcome non-lockdowners. We might hope that a proper peer-reviewed scientific article would do a better job of causal analysis than the articles linked above. As for economic impact, let us recall that the purpose of financial forecasting is to make astrology look respectable; let's wait and see what the actual outcomes are going to be. It would not be surprising to learn that e.g. the level of additional debt that Sweden has to incur would be much lower since the losses of small businesses would be lower. Furthermore, the proper economic assessment of the Swedish strategy would have to ask how that strategy would perform if significant number of European countries adopted the strategy: from the point of view of that strategy, other European countries defected in the prisoner's dilemma in which they found themselves. If the Swedish strategy were adopted across Europe, its true economic performance could be seen. Let us also note that economic performance cannot be divorced from mid-term health outcomes. The adverse health consequences of the economic impact of the lockdown policies are likely to be felt especially in the 3rd world. --Dan Polansky (talk) 10:00, 19 June 2020 (UTC)

This is a display of bad faith. But we are on Wikipedia, so keep your propaganda for your friends. Give us primary or secondary sources, or don't spam here. Alcyon007 (talk) 17:00, 20 June 2020 (UTC)

Let's discuss substance and subject matter. --Dan Polansky (talk) 09:09, 21 June 2020 (UTC)

One could say that if the method of preventing healthcare overload is not to give healtcare then it is not success. Ie. one can prevent the healtcare overload just by limiting the access to the health care and increasing the capasity of undertakers which was something what we saw in sweden. --Zache (talk) 19:52, 20 June 2020 (UTC)

As far as I know, there was no healthcare overload in Sweden, but I may be wrong. This Wikipedia article does not have the word "overload". Does anyone have any sources indicating that Sweden had healthcare overload during the covid pandemic? Let's look e.g. at Stockholm shuts field hospital as pandemic slowly eases grip on capital, June 4, reutes.com, which indicates about a field hospital built in Sweden to support Stockholm healthcare the following: "Eriksson said the field hospital had not needed to take in patients since the region’s regular hospitals had been able to increase capacity more than healthcare officials had initially thought feasible".

My reason for posting here is the belief that "Sweden's strategy failed" is not a factually correct statement belonging in an encyclopedia (it could be okay for a polemic journalistic article, but even that is not very clear), and that comparing Sweden only to select few countries while controlling for no confounding factors is not a sound scientific practice. --Dan Polansky (talk) 09:09, 21 June 2020 (UTC)

Ok, I will be more clear. In example, they had room in hospitals because were not taking eldery patients from care homes. In care homes however, there was no resources or methods or legal right to give care needed. With care i mean things like supplementary oxygen. (Bioedge is not good enough to be used as reference but it is good summary and there is link to paywalled WSJ article: [13]) Second example is that they had official protocol to transfer patients with likely covid-19 infections from hospitals to care homes without testing. This also helped to keep the hospital load lower, but spreaded the infections in care homes.[14]. Technically this can be on paper that there was no overload in hospitals, but at same time it leaded to dead people. Zache (talk) 10:41, 21 June 2020 (UTC)

Could you point at where it says that denying care for nursing home patients was the reason that hospitals didn't run out of beds? Because I couldn't find anything in in your source that supports your claim. — Preceding unsigned comment added by Blådjur (talk • contribs) 21:22, 21 June 2020 (UTC)

Little bit longer quote from WSJ The triage recommendations were intended to prevent hospitals from being overwhelmed by a surge of sick people and to focus medical resources on those most likely to benefit from them. Many Western countries have similar guidelines, but they have only been implemented on a large scale in northern Italy when the regions ran out of hospital beds in March. ... In Sweden, critics say, these guidelines have too often resulted in older patients being denied treatment, even when hospitals were operating below capacity. Occupancy in the country’s intensive-care units, for instance, has yet to exceed 80%, according to government officials.. --Zache (talk) 05:15, 22 June 2020 (UTC)

Thank you. The bioedge.org[15] article is interesting and disturbing, e.g. 'A consistent theme [in complaints] is that nursing home residents with suspected Covid-19 were immediately placed on palliative care and given morphine and denied supplementary oxygen and intravenous fluids and nutrition. For many this was effectively a death sentence' and '“Older people [in nursing homes] are routinely being given morphine and midazolam, which are respiratory-inhibiting,” [...]'. If this is true, some of the people are likely to have been killed by the intervention, i.e. the respiratory inhibitants, rather than the covid alone. Nonetheless, I still maintain that moderate success is a fair description for the Swedish strategy as a whole, and it is a pity some elements of the strategy contributed to far too many excess deaths and prevented the strategy from being a complete success. Either way, it does not seem proper for an encyclopedia article to declare the strategy a failure or success at this stage; rather, let the article report facts closer to raw observation. --Dan Polansky (talk) 11:09, 21 June 2020 (UTC)

I think that Dan seems to make sense. David A (talk) 11:45, 21 June 2020 (UTC)

Also, Karolinska University Hospital in Stockholm said today (because they didn't have enough staff, equipment etc) : Vi har upprepade gånger tvingats säga nej till patienter vi normalt skulle ha accepterat, säger Michael Broomé till Dagens Nyheter ie We have repeatedly been forced to say no to patients we would normally have accepted, says Michael Broomé to Dagens Nyheter [16] --Zache (talk) 10:31, 22 June 2020 (UTC)

And just to be clear, I totally agree with this: Either way, it does not seem proper for an encyclopedia article to declare the strategy a failure or success at this stage; rather, let the article report facts closer to raw observation --Zache (talk) 10:36, 22 June 2020 (UTC)

@Dan Polansky

Sources speaks about "several cases" and "many complaints", not of a nationwide strategy. The decision of admitting someone to a hospital is always taken by the doctor. And the triage guidelines were written for what to do in a situation where hospitals has become completely full, not for how to prevent hospitals from becoming full.[17] Also, the guidelines makes it clear prioritizing should be based on biological age (how long an individual is expected to survive) as opposed to chronological age.[18] Regarding nursing home residents not being put on ventilators specifically, a common explanation by doctors is that many nursing home COVID patients are simply to frail to benefit of intensive care.[19][20] You have to be really sick to get into a Swedish nursing home. According to stats from a few years ago, one of five who moved into a nursing home died within one month. Swedish elderly care is centered around helping people to live at home as long as possible, and your municipality will rather give you around-the-clock home care than give you room at a nursing home. (note: the situation is of course different if you suffer from dementia). But in the latter article they've also talked to doctors suggesting they've prioritized too hard, one doctor speaks about how the guidelines created a "mood". Perhaps we have to wait for IVO to learn if this has been common. Also note that the coverage as usual is Stockholmcentric. bladjur (talk) 23:06, 27 June 2020 (UTC)

Just note. If I understood correctly problem was not getting on the intensive care and being put on ventilators, but getting even to the normal ward so they could get oxygen via breathing masks which is less intrusive or getting nutrients via drip if they cannot eat. --Zache (talk) 07:07, 29 June 2020 (UTC)

There was initial comments from Sweden's Healthcare Inspectorate review in thelocal.se: Sweden's Healthcare Inspectorate (IVO) was asked in June to review how the care sector had coped with the coronavirus after an earlier review exposed "serious flaws" in one in ten care homes. This included sick care home residents not being given an individual doctor's assessment regarding their diagnosis and treatment, meaning that some were not given the appropriate care, according to IVO. ... One problem the watchdog raised was that some municipalities may have introduced unnecessarily strict measures in an early stage. As regions prepared for a heavy strain on hospitals and the medical sector, some issued instructions that the requirement for an individual doctor's assessment be removed for people living in care homes. "In some cases these were never implemented, others took them away quite quickly, but it happened, and [authorities] went quite far at an early stage," said Wallström. In some cases, this meant that decisions were made on a general basis, and were not the best option for the individual patient. ... In the 40 worst-hit municipalities, 40 percent of the municipalities said they were unable to give patients an individual doctor's assessment. In other municipalities, this figure was lower at around 30 percent. Wallström named lack of individual assessment as one of the serious flaws. (If i understand correctly this is different review than review started by goverment of Sweden?)--Zache (talk) 15:53, 7 July 2020 (UTC)

Just a brief comment about that the fatality rate graph for Sweden has quickly declined enormously over time, as the virus has mutated to turn less deadly, so declaring what Sweden did a failure, compared to the far greater economic devastation that we would suffer from otherwise, seems extremely misleading: https://www.worldometers.info/coronavirus/country/sweden/ David A (talk) 13:22, 9 July 2020 (UTC)

As for "the virus has mutated to turn less deadly", there is no evidence of that, certainly not in the covid-coded death data and all-cause death data. --Dan Polansky (talk) 13:36, 9 July 2020 (UTC)

Well, despite that the virus is far more widespread in Sweden now, we have a statistically shown massively decreased number of deaths per day here. Please see the "Daily New Deaths" graph in the link that I provided. David A (talk) 14:06, 9 July 2020 (UTC)

There is no doubt daily covid-caused deaths have declined, as per official FOHM data (from which worldometers is sourced) and as per the all-cause death graph in COVID-19 pandemic in Sweden#Excess mortality or as per File:All-cause weekly deaths - Sweden.png. But the decline is not evidence that "the virus has mutated to turn less deadly"; that would be a wild jump to a conclusion that does not follow from the premise by any stretch. --Dan Polansky (talk) 14:16, 9 July 2020 (UTC)

Okay. What are the current hypotheses regarding the reasons for this then? David A (talk) 18:26, 9 July 2020 (UTC)

I can give you some hypotheses that come to mind. One possible explanation is the analogue of the economic law of diminishing returns: once the virus runs its way through places with highly vulnerable people where it can spread real fast such as nursing homes in Stockholm, it starts to have an increasingly harder time "finding" vulnerables to infect at the rate at which it originally did. Another possible factor may be increase of protective measures for the vulnerables. Another possibility is that the true daily rates of infection have been decreasing as well, as would be weakly suggested by the decline of reported daily rates for severe cases in the FOHM report. These explanations may be incorrect or not give the complete picture. We don't know. In any case, if you want to continue this conversation that does not really belong to this thread, please create a dedicated thread and post your reliable sources for "the virus has mutated to turn less deadly" or compelling analysis supporting that claim. --Dan Polansky (talk) 07:14, 10 July 2020 (UTC)

Okay. Never mind then. David A (talk) 07:28, 10 July 2020 (UTC)

Seroprevalence and prevalence in Stockholm

Per Swedish antibody study shows long road to immunity as COVID-19 toll mounts, May 20, reuters.com, " A Swedish study found that just 7.3 percent of Stockholmers developed COVID-19 antibodies by late April, [...]."

Thus, the covid seroprevalence in Stockholm was about 7.3% by late April. However, to determine how many people have had the virus, we need to know the degree to which covid infection tends to produce specific antibodies. Furthermore, to determine how many people have immunity against the virus, we need to know how many people were already immune before coming in contact with the virus. For that purpose, Coronavirus likely five times more common and less deadly than assumed, swprs.org, is relevant, and it traces to this pdf, which seems near-identical to Antikörpertests weisen nur ein Fünftel der Fälle nach, 02.06.2020, thunertagblatt.ch. These are not the kind of reliable sources that we need, but they do remind us that the actual population immunization degree may be much higher than the seroprevalence suggests. --Dan Polansky (talk) 10:10, 30 June 2020 (UTC)

This new study from Karolinska might interest you: Immunity to COVID-19 is probably higher than tests have shown

"New research from Karolinska Institutet and Karolinska University Hospital shows that many people with mild or asymptomatic COVID-19 demonstrate so-called T-cell-mediated immunity to the new coronavirus, even if they have not tested positively for antibodies. According to the researchers, this means that public immunity is probably higher than antibody tests suggest." bladjur (talk) 18:01, 30 June 2020 (UTC)

Super interesting. Let me quote: '“Moreover, roughly 30 per cent of the blood donors who’d given blood in May 2020 had COVID-19-specific T cells, a figure that’s much higher than previous antibody tests have shown.”' Caveat: refers to a preprint that is not peer reviewed. --Dan Polansky (talk) 09:05, 1 July 2020 (UTC)

The Karolinska study is now mentioned in Coronavirus: Immunity may be more widespread than tests suggest, July 1, bbc.com. --Dan Polansky (talk) 11:04, 3 July 2020 (UTC)

A new study on T cell immunity for the covid, by researchers from Singapore:

• SARS-CoV-2-specific T cell immunity in cases of COVID-19 and SARS, and uninfected controls, 2020, nature.com

--Dan Polansky (talk) 08:45, 16 July 2020 (UTC)

Cautionary tale or not

More for reference:

• Cautionary tale article: Sweden Has Become the World’s Cautionary Tale, July 7, nytimes.com
• Rebuttal of the above: Multiple errors in the New York Times article about Sweden’s corona strategy, July 10, emanuelkarlsten.se

More on the same theme in #Sweden's strategy failed: Much higher death rate than in Norway or Denmark, per capita. --Dan Polansky (talk) 11:13, 11 July 2020 (UTC)

Take this from the NYT article: "Here is one takeaway with potentially universal import: It is simplistic to portray government actions such as quarantines as the cause of economic damage. The real culprit is the virus itself." This is plainly false. It is true that if a lone country does moderate interventions and other countries defect in the prisoner's dilemma by blocking international travel and trade, that country that did not defect is going to suffer the economic consequences of other countries' actions. But it is above all the lockdowns and restrictions that cause the huge economic damage, not the virus. Admittedly, without the virus there would be no lockdowns and restrictions, but the virus does not make them inevitable, just like the 1968 Hong Kong flu virus did not make lockdowns and restrictions inevitable in 1968. --Dan Polansky (talk) 11:41, 11 July 2020 (UTC)

Something that really needs mentioning is that when comparing deaths in Norway, Denmark and Sweden an important factor is how many sick people they started with. Come the 15th of March the borders are closed all over Europe, and you have to deal with the sick people that you have inside. If there was a reason to believe that Sweden started with an order of magnitude more sick people, then it would not be surprising that they end up with an order of magnitude more dead people. Now, without the sort of testing that Sweden never did in March, because they were only testing the very sick, because there wasn't enough tests to go around you cannot prove such a thing, but it would be surprising if this is not the case. The significant question to ask your population was 'where were you the last week in February and the first week of March. Could you have been out getting sick and bringing the disease home?' So far I have been unable to get accurate numbers out of Norway or Denmark, despite asking for now 3 months. But the hand-waving answer is 'about 1% of our population were travelling out the the country these weeks'. For Sweden the number is more than 10%, more than a million people. See: https://www.thelocal.se/20200611/public-health-agency-head-coronavirus-came-to-sweden-from-countries-that-were-under-our-radar The week that Stockholm takes it's one week winter-sport vacation -- various regions take this different times in a month long period -- is a top time for international travel in Sweden, and this winter sport week is a custom that is not shared by Denmark and Norway. We still don't know how many people returned to Sweden, sick, and quite likely asymptomatic so in no way feeling sick. But the claim that the most significant difference between Sweden and the other Nordic countries was the severity of the lockdown should not pass unchallenged. And if you have accurate travel numbers for Norway or Denmark for the last week of February and the first week of March I would be very grateful if you could forward them to me. Lacreighton (talk) 16:59, 11 July 2020 (UTC)

Just to not that there is winter sports week (ie. winter break) in Denmark, Norway and Finland too. Another note is that initially Norway had more (tested) infections than the sweden and afaik denmarks situation was worse than in sweden when they did a lockdown. If i remember correctly denmark did a lockdown because there were infections in care homes and they tried to get things under the control. --Zache (talk) 08:02, 13 July 2020 (UTC)

We need more solid sources, solid data and solid analysis, and less "If i remember correctly", I would say. The above may be true, but how do I verify it easily? --Dan Polansky (talk) 12:09, 13 July 2020 (UTC)

Case Denmark vs Sweden

• Winter break Copenhagen: 10.2.2020 - 14.2.2020[21] (all regions between 10.2. - 28.2. ?)
• Winter break Stockholm: 24.2.2020 - 28.2.2020[22] (all regions between 11.2. - 6.3. [23] ?)
• Denmark: Outbound tourism 2018 avg 131€, 3.4 nights, top 3 countries: Spain (18.1%), Germany (4,9%), Italy (3.3%) (eurostats)
• Sweden: Outbound tourism 2018 Avg 89€, 3.3 nights, top 3 countries: Spain, (9%), USA (4.6%), Thailand (3.2%) (eurostats)
• Denmark: Total international departures, Overnight visitors (tourists) 2018: 7 475 429 ( 1.29 per million POP (ocdc))
• Sweden: Total international departures, Overnight visitors (tourists) 2018: 18 855 000 ( 1.86 per million POP (ocdc))
• Arrivals to Austria Jan-Apr 2020 from Denmark: 171,2 (in 1.000) [24]
• Arrivals to Austria Jan-Apr 2020 from Sweden: 72,6 (in 1.000)

So I would say that there is no significant difference between Denmark and Sweden in terms of the total number of travel. Ie, even if one country would do a double number of traveling it would not make a ten-fold difference in sickness. However, there is some truth in the winter break vacations. The winter break in Copenhagen may have been early enough that it missed the initial wave of infections around Europe. In the level of the number of the tested infections in early march, Norway, Sweden, and Denmark were in pretty much in line (graph) and the difference becomes how the countries were able to turn the tide down after the lockdown. One can argue that there were a lot more infected people initially in Sweden and that was the reason why infections are declining lot slower. Explanation however doesn't explain why countries like Austria or Switzerland which were higher in number of tested infections were able to take down the spreading faster than Sweden. (graph in aatishb). --Zache (talk) 07:59, 14 July 2020 (UTC)

Perhaps you can find it at Eurostat? bladjur (talk) 14:45, 13 July 2020 (UTC)

Thank you. And yes, the travel bit that matters is were you travelling when people were sick. The regions of Sweden, such as Skåne, which had their winter sport week earlier did not cause a lot of infections, so presumably when they travelled they did not encounter sickness. But the last week of February seems to be a particularly bad time to be travelling. Lacreighton (talk) 21:14, 21 July 2020 (UTC)

Somali deaths in Stockholm Region

As for "At a point, it was reported that a disproportionate number of those that had died by then were Somali (6)[365] out of 89[366] deaths being members of the Somali community in the Stockholm Region.[365]": The Stockholm region now has a long report on the matter (up to the middle of June). https://www.sll.se/globalassets/1.-halsa-och-vard/bilagor---nyhet/bilagor-nyheter-2020/region-stockholm-rapport-ces-nr-1-om-covid-19-i-stockholms-lan.pdf (you want page 12). Indeed, a disproproportionate number of deaths are among those born in Somalia -- also for those born in Syria, Lebanon and Turkey. But the calculation is a bit involved, in that proportionally it is the Estonians and Finns who were most likely to die. But they were also older than the reference population of people born in Sweden. So they adjusted for age and sex, and here the Somalis came out significantly worse than people born in other countries. But then you adjust again for 'how much sickness was there in the neighbourhood where you live' and end up with a smaller, but still significant excess risk remains for those born in Somalia, Syria, Lebanon and Turkey. Certain areas which were among the worst hit also are neighbourhoods where a disporportionate number of Somalia-born live, but people with other lands of birth who lived in the badly effected areas aso got disproportinately sick if I am reading the report correctly. Other neighbourhoods where many Somalia-born live weren't as badly effected. This is a difficult bit of statistical adjusting to explain, but I thought I would leave the link to the paper in case somebody wants to do something with it. Lacreighton (talk) 21:14, 21 July 2020 (UTC)

I changed the section title from a sentence to "Somali deaths in Stockholm Region" and made the sentence the 1st item of the paragraph. I hate to edit other people's posts, but long sentences in section titles are very impractical and not the recommended practice. --Dan Polansky (talk) 06:23, 22 July 2020 (UTC)

Thank you. Lacreighton (talk) 09:18, 22 July 2020 (UTC)

death statistics

Take care there is something strange going on in the reporting of the death statistics.

https://ourworldindata.org/coronavirus/country/sweden?country=~SWE 14 – 10 15 – 9 16 – 27 17 – 21 18 – 26 19 – 0 20 – 0 21 – 20 22- 7 22-7 TOTAL 5646

https://www.coronatracker.com/country/sweden/ 14 – 9 15 – 27 16 -21 17- 26 18 - 0 19 – 0 20 – 20 21 – 7 22-21 TOTAL 5667

https://www.worldometers.info/coronavirus/country/sweden/ 14 – 7 15 - 6 16 - 3 17 - 2 18 - 2 20 - 2 21 -1 Total 5657 — Preceding unsigned comment added by 119.18.17.241 (talk) 23:04, 22 July 2020 (UTC)

All statistics in this article comes from official sources only:

Public Health Agency

National Board of Health and Welfare

bladjur (talk) 10:22, 25 July 2020 (UTC)

Until now that is. It seems like one of the websites in your comment just got added back to the article in an us-centric edit.bladjur (talk) 16:45, 25 July 2020 (UTC)

Disruptive edits

One account has repeatedly deleted entire sections in this article. The only explanation given is "article was pretty lengthy". That might be the case, but the solution isn't to randomly delete entire sections when they definitely include information vital to the subject. Please see WP:CRV for Wikipedia's guidelines on content removal. — Preceding unsigned comment added by Blådjur (talk • contribs) 10:39, 25 July 2020 (UTC)

I didn't remove any material; I simply moved it to another article that's still standing: Swedish government response to the COVID-19 pandemic. I already mentioned this on your talk page and provided a link to that article. Per WP:SIZERULE, articles that are more than 100 kB in length "[a]lmost certainly should be divided". The article as it stands now, as of this writing, is 239 kB. At least one split is necessary. The article about the pandemic should summarize the main points, while more specific details can go to the government response article. Love of Corey (talk) 03:28, 27 July 2020 (UTC)

For a comparison, after I moved the content to the new article again, the article is now 149 kB. That's still over 100 kB, but at least the article is now at a slightly more manageable length. Love of Corey (talk) 08:07, 27 July 2020 (UTC)

To provide more context I should mention that template Template:COVID-19_pandemic_data/Sweden_medical_cases alone hase size of 35 657 bytes, while section COVID-19_pandemic_in_Sweden#Statistics code takes up to 45 349 bytes - so from formal perspective even this too parts together "should be splitted" looking on the issue only from "size of code" perspective. So it looks like this size rule should be relaxed here, at least taking into account that fact that this article is statistics-heavy and references-heavy 84.47.179.91 (talk) 18:21, 27 July 2020 (UTC)

The notabilty of the subject justifies a long article. Some parts obviously suffer from too much detail, which is understandable as the article depicts an ongoing event and information is constantly being added to it. But of course this should be fixed by cleaning up the existing article, not by deleting random parts. And as this being the main article, a broad overview is necessary, with separate articles with more detail if necessary. And this "split" is likely very confused to someone going here looking for information on how the Swedish society has responded to the pandemic. — Preceding unsigned comment added by Blådjur (talk • contribs) 20:55, 27 July 2020 (UTC)

As there are obviously differing views on whether to delete entire sections of the article or not, I'm afraid a concensus has to be reached before you delete half of the article. Until then, I undid your edit where you removed almost 40k characters with the summary "Oh my god". — Preceding unsigned comment added by Blådjur (talk • contribs) 21:19, 27 July 2020 (UTC)

You're either misinterpreting my point completely or dodging it. I MOVED THE CONTENT TO A NEW ARTICLE. Just because it's not in this article doesn't mean it's been deleted altogether. You can find the material in the new article.

You also just proved my point: "And as this being the main article, a broad overview is necessary, with separate articles with more detail if necessary." That's why the government response article is there: to provide more detail on the government response to the pandemic. This article should be primarily focused on the pandemic in Sweden. All other topics should be summarized as much as possible, with the finer details being split off to individual articles if they're too plentiful. As you said, this is an ongoing event, so splits are necessary at the moment.

As for this comment: "And this 'split' is likely very confused to someone going here looking for information on how the Swedish society has responded to the pandemic." That's why there's a link to the new article at the top of the government response section, as well as a link to the article in the lede. People will know to click links for further information if they're curious or wondering why the topic is being summarized.

And yes, I welcome a community consensus on the topic. Love of Corey (talk) 21:22, 27 July 2020 (UTC)

For reference, the dispute of this thread (split or not split) continues below in #RfC: Splitting content on government response to separate article. --Dan Polansky (talk) 12:46, 28 July 2020 (UTC)

Comparison of per capita death rate between Sweden and the United States

While statistic comparisons between Sweden and its neighbouring countries are relevant at least in part, the comparison between Sweden and the US is irrelevant and arguably misleading. To begin with, the proclamation that Sweden's deaths per capita is 25% higher than in the US comes out of nowhere in the article, and seems purely to be some form of propaganda. Sweden as a country has very few similarities with the United States, and it would make more sense to make comparisons to other European countries. Furthermore, there is a huge difference of both the population levels and the total Covid-19 death rates between Sweden and the United States, why a death per capita comparison on a national level seems even more forced and unnatural. Incidentally, as of July 30th, there are 11 individual states in the U.S. that have higher Covid-19 Death per capita rates than Sweden. Some of these states have a larger population than Sweden, some have a smaller one. This however is not mentioned in the article, although it's probably slightly more relevant. In short, I suggest comparisons between the United States and Sweden are removed from the article, since the U.S. has nothing to do with Sweden's response to the Covid-19 pandemic. — Preceding unsigned comment added by 192.36.34.249 (talk) 08:25, 30 July 2020 (UTC)

I too suspect the editor wants to make some kind of point, and either way it's an obvious USA-centric edit. But let's for a second assume it is relevant. If that would be the case, the article should naturally explain why, and the information should be sourced. As of now, there are no context whatsoever. The same goes for the comparision with "other Scandinavian countries such as Denmark and Norway". I'm familiar with those countries as well as the repeated use of the comparision in debates on social media and elsewhere, but if I read a single sentence comparing the numbers of Guatemala with El Salvador and Honduras, I would have no clue about what to make of this information without any information about why the comparision is relevant. I would however be interested to know to what extent Guatemala has been affected by the pandemic. The introduction previously said something like "the number of deaths has been significantly higher than in most European countries, including neighboring Scandinavian counties", with additional information in main section. And we definitely shouldn't assume that the information "5,702 confirmed deaths[note 1] related to COVID-19 in Sweden" is sufficient enough (I would never have guessed Guatemala has a population of 17.2m). And sooner or later someone will add something like "but lower than Belgium and Britain", which won't make things more clear. Secondly, third, etc are several clear WP:MOSLEAD issues. As of now, we might as well add a table and a photograph of Donald T. to the introduction.

However I for sure won't touch their edits. I once moved a similar edit of theirs to the statistics section. Didn't change a single word iirc. This resulted in them going to the Teahouse writing a long text accusing me of being a shill and being a threat to Wikipedia. Almost made me delete my account. — Preceding unsigned comment added by Blådjur (talk • contribs) 19:40, 30 July 2020 (UTC)

I went ahead and removed the comparison in diff. The above objections to the comparison seem valid; I would emphasize that there are no obvious criteria by which the countries were selected for the comparison. A Wikipedia reader can do such a comparison easily by looking at File:COVID-19 Outbreak World Map Total Deaths per Capita.svg, a map that is included in other Wikipedia articles; in the map, the reader can pick any group of countries to compare against rather than being presented with an arbitrary selection. --Dan Polansky (talk) 09:50, 31 July 2020 (UTC)

Regarding relevance of comparison. Suggested "original researchers" little bit provoke me to suggest some other view of those numbers. If one look at Eurostat tables Causes of death - standardised death rate by residence (hlth_cd_asdr2) [25] and Population by age group (tps00010) [26] one could compare countries by *death per year per 100 000 residents*, and most interesting - wich impact covid pandemy potentially has on that numbers. For example let's consider Finland vs Sweden.

• Sweden 2020 covid death per 100K: 55.7
• Finland 2020 covid death per 100K: 5.9
• Sweden 2017 yearly death per 100K: 915.53
• Finland 2017 yearly death per 100K: 959.56
• Sweden 2020 ESTIMATED (2017+covid) yearly death per 100K: 971.23
• Finland 2020 ESTIMATED (2017+covid) yearly death per 100K: 965.46

(see [27] for shorter countries list)

So even taking into account comparably significant excess mortality in Sweden it's yearly death per 100K residents number will be probably inline with other neighboring countries (and will be still probably better then in Germany, where even without pandemy that number was in range 1000+).

Other observation could be made, when selecting different age groups [28]

• Sweden 2017 yearly, age 65+, yearly death per 100K residents of that age: 4 123.16
• Sweden year 2019, age group 65 - 79: 14.8 % , 80+ : 5.1 % , 65+ : 19.9 %
• Sweden 2017 ESTIMATED age 65+ yearly death per 100K all residents: 820.50
• Sweden __,,__ + covid 2020: 876.2
• Finland 2017 yearly, age 65+, yearly death per 100K residents of that age: 4 185.77
• Finland year 2019, age group 65 - 79: 16.3 % , 80+ : 5.5 % , 65+ : 21.8 %
• Finland 2017 ESTIMATED age 65+ yearly death per 100K all residents: 912.49
• Finland __,,__ + covid 2020: 918.39

So here one could see that with simple numbers "trickery" one could came to conclusion that "Sweden will be doing even better than Finland (in year 2020)" basing on that kind of numbers (While trick there is in that fact that Swedish population is bit younger than Finnish, and weight of 1 old men death is higher in Finland than in Sweden)

And of course it would be good to make similar calculations on more fresh data (taken for year 2019). But still I believe it will show not much difference with what is calculated above. 84.47.179.91 (talk)

Yes, the trick is that the population in Finland is older. Also argument is always to use Norway as comparison and see how it goes.  :) --Zache (talk) 05:02, 2 August 2020 (UTC)

RfC: Splitting content on government response to separate article

Should most of the content regarding the Swedish government's response to the COVID-19 pandemic be split off into an individual article called Swedish government response to the COVID-19 pandemic? Love of Corey (talk) 21:56, 27 July 2020 (UTC)

• Comment - There is an ongoing conflict between myself and Blådjur over spinning off a lot of the content about the Swedish government response to the aforementioned individual article, which I created. I believed it should be, based on: 1) WP:SIZERULE, considering the article, when including all of the content on the government response, measured at 239 kB, and even a compromise attempt stood the article at 188 kB; and 2) the fact that the government response topic seemed to be a very notable topic on its own, with a plethora of extensive information on not only how the response carried out, but national and international reaction. Blådjur disagreed, saying, "The notabilty [sic] of the subject justifies a long article." They suggested rectifying the size issue by cleaning up the existing article, and they claimed a split would be very confusing to readers. You can read our full discussion here. I'd like to hear thoughts on this from the community. Love of Corey (talk) 21:56, 27 July 2020 (UTC)
• Comment - The disagreement between me and Blådjur culminated in the latter saying a community consensus should be reached, which I readily agreed to, hence this RfC, which I felt was the appropriate avenue for this sort of thing. Love of Corey (talk) 22:02, 27 July 2020 (UTC)

I'm not sure what pointing at my spelling mistakes has to do with this, because it seems like you understood me perfectly well. English isn't my native language, but I do my best. Maybe you think it makes your point more valid than mine? Spelling is of course important, and I wouldn't mind you fixing my spelling mistakes silently. — Preceding unsigned comment added by Blådjur (talk • contribs) 23:59, 27 July 2020 (UTC)

• Comment The first sentence in your article reads "Sweden's response to the COVID-19 pandemic has been met with international scrutiny.". If you feel this is how to start the introduction to this what you call "very notable topic", perhaps you really should go for an "International criticism to Sweden's response to the COVID-19 pandemic" article instead of an article covering the entire response the pandemic in Sweden. Ö And if you believe the content could be "split" into two separate articles, please be more specific about which parts of the subject should be covered in each article. If you believe it's possible to avoid confusion, you need to be very specific. Mind that your criteria needs to be obvious for all readers not familiar with the subject. I'm familiar with the subject, but it's very confusing to me, especially since most of the content you removed has nothing to do with the response from the government. And for someone looking for that information, I'm not sure it's obvious to find it through a link in the middle of an article under the headline "Government". Not at all. Like what exactly do you mean with "This article should be primarily focused on the pandemic in Sweden"? bladjur (talk) 23:58, 27 July 2020 (UTC)

Maybe you should take this to Talk:Swedish government response to the COVID-19 pandemic. This is supposed to be a community vote discussion. I'll continue this discussion there. Love of Corey (talk) 00:33, 28 July 2020 (UTC)

• Comment What exactly are the community voting on? You want us to vote on a "split", but we need you to clearly specify what each of the two articles are supposed to cover. You propose that the article COVID-19 pandemic in Sweden should be split into two articles, one about the "Swedish government response to the COVID-19 pandemic" and one article "primarily focused on the pandemic in Sweden". Exactly what is each article supposed to cover, and how can we see that the readers easily find this information, preferably on their first try? — Preceding unsigned comment added by Blådjur (talk • contribs) 01:16, 28 July 2020 (UTC)

Uhhhhh...you just answered the first part of your own question. As for the second part, we simply use link templates like {{main}}, like what the article is currently doing. Love of Corey (talk) 01:25, 28 July 2020 (UTC)

The question that begins this RfC discussion also pretty much summarizes what the problem is. I can't see how anyone would have a hard time understanding it. Love of Corey (talk) 01:28, 28 July 2020 (UTC)

• Comment Yabberbot asked me to comment on this and its an obvious coat rack article - have nominated it for deletion -----Snowded ^TALK 04:40, 28 July 2020 (UTC)

That said, do you think the content should be split off into a new article, considering the amount of WP:RS around it? Or do you think a significant trim is possible? That's the question I'm asking. Love of Corey (talk) 05:06, 28 July 2020 (UTC)

Radical trim to content in both cases and one article is more than enough at this stage - the content in both cases reads like an essay rather than an encylopedia entry -----Snowded ^TALK 06:25, 28 July 2020 (UTC)

• Comment - I personally think that the current version of the article seems fine, and that it should not be significantly cut down to omit highly relevant information. David A (talk) 07:04, 28 July 2020 (UTC)

The current version of the article measures at 188 kB, though. I think we can do better than that, length-wise. Love of Corey (talk) 09:19, 28 July 2020 (UTC)

• Relevant observations: 1) In the WP:SIZERULE, the suggested size limit is specified in terms of "Readable prose size" rather than wikitext size; I copy-and-pasted text from Blådjur's latest revision (diff) into a plain text file and counted 101 530 bytes, which approximately matches the size guideline; the guideline Wikipedia:Article size has the following specification/definition: "Readable prose size: the amount of viewable text in the main sections of the article, not including tables, lists, or footer sections", and while I did not copy footer section, I did copy tables, so without table data, it would be even shorted. 2) Even if the article were longer, that could be justified by usefulness: there was no actual problem that needed to be solved. 3) It is unfortunate that there is now a low-added-value discussion wasting time of people who actually made substantive contribution to the article, such as Blådjur. And there is more relevant: Wikipedia:Article size#No need for haste: "As browsers have improved, there is no need for haste in splitting an article when it starts getting large. Sometimes an article simply needs to be big to give the subject adequate coverage. If uncertain, or with high profile articles, start a discussion on the talkpage regarding the overall topic structure. Determine whether the topic should be treated as several shorter articles and, if so, how best to organize them. If the discussion makes no progress consider adding one of the split tags in order to get feedback from other editors." The split was made without a previous discussion and was then reinforced by a revert when opposition arose; both actions violate Wikipedia:Article size as quoted. --Dan Polansky (talk) 12:40, 28 July 2020 (UTC)

Wait, how were you able to get 101 kB even with the tables? I keep getting over 200 kB. What is this footer section that you left out? Love of Corey (talk) 21:43, 28 July 2020 (UTC)

• Agree with arguments of Dan Polansky. From my point of view, it could be perceived differently if this splitting attempt would happen few years after pandemy end. But making this now could be perceived at least like "Premature optimization". 84.47.179.91 (talk) 19:50, 28 July 2020 (UTC)

• Snowded and David A: I think it can be significantly (but perhaps not radically) trimmed without cutting relevant information. Most stuff is relevant, but some parts reads more like news (well, it is) and has way too much detail. One example:

The government has issued progressively stricter advisories against travel. Beginning on 17 February, the Swedish Ministry of Foreign Affairs advised against all trips to Hubei, China, as well as non-essential travel to the rest of China apart from Hong Kong and Macao.

On 2 March the Ministry for Foreign Affairs advised against trips to Iran, due to the uncontrolled spread of the COVID-19 in the country. The Swedish Transport Agency also revoked Iran Air's permit for Iranian flights to land in Sweden from the same date. According to the foreign ministry, there were several thousand Swedish citizens in Iran at the time of the ban, many of them with difficulties getting back to Sweden. On 6 March, the Ministry for Foreign Affairs advised against all non-necessary trips to northern Italy, specifically the regions of Piemonte, Liguria, Lombardia, Emilia-Romagna, Trentino-Alto Adige, Valle d'Aosta, Veneto, Friuli Venezia Giulia, Marche and Toscana. Turin, Milan, Venice, Verona, Trieste and Florence are large cities in these regions. The Public Health Agency of Sweden, who initiated the recommendation for the Ministry for Foreign Affairs, stated that the decision was based solely on the strain of the Italian health care system. On similar grounds, the foreign affairs ministry also advised against all non-necessary travel to the city of Daegu and the province of Gyeongbuk in South Korea. The advice regarding travel to Italy was extended 10 March to include all of its regions. Finally, all international travel was discouraged on 14 March. The advice was to be in place for one month, after which it would be up for review. Travel from non-EU/EEA member states was stopped on 17 March and unnecessary travel within Sweden was advised against on 19 March.

[...]

On 7 April, the foreign ministry extended the advice against all non-essential travel abroad until 15 June, when it would again be reconsidered. On 9 May, Swedish foreign minister Ann Linde said that although a decision about an extension was yet to be made, she made clear that travel wouldn't return to normal after 15 June. On 13 May, the Foreign Ministry again extended the advice for non-necessary foreign travel to 15 July. From 30 June, the advice against non-essential travel were lifted for 10 EU countries, namely Belgium, France, Greece, Iceland, Italy, Croatia, Luxembourg, Portugal, Switzerland and Spain, as well as for Monaco, San Marino and the Vatican City. The advice against travel to other countries within the EU, EEA and the Schengen Area would remain in effect until 15 July, while advice were extended until 31 August for countries outside those areas.

This example could easily be significantly trimmed without losing relevant information IMO:

The government has issued progressively stricter advisories against travel. Beginning in February, the Swedish Ministry of Foreign Affairs advised against all trips to Hubei, China, as well as non-essential travel to the rest of Mainland China. In early March, the Ministry for Foreign Affairs advised against trips to Iran, due to the uncontrolled spread of the COVID-19 in the country. The Swedish Transport Agency also revoked Iran Air's permit for Iranian flights to land in Sweden, resulting in several thousand Swedish citizens finding themselves with difficulties getting back to Sweden from Iran. This was followed by advice against all non-necessary trips to parts of South Korea and northern Italy, later extended to the entire country, in order to avoid further strain on their health systems which were under high pressure due to a large number of cases. From mid-March, the Ministry discouraged all non-necessary international travel. The advice was initially for one month, but after several extensions it were to be in effect for most countries until 31 August, although restrictions had being lifted for countries within the EU, EEA the Schengen Area at various points during the summer.

Quick example, but easily cut to half its size. bladjur (talk) 12:57, 29 July 2020 (UTC)

• Oppose at present. The current state of Swedish government response to the COVID-19 pandemic is that of a WP:POVFORK (which the creator of the page has repeatedly denied was their intent, for the record), and splitting the article like this is just a recipe for a WP:POVFORK even if it isn't one at the time of the split. I tend to agree with the arguments above that it was not necessary at the time of the split (we have plenty of WP:Featured articles of comparable length, such as Macedonia (ancient kingdom), Byzantine Empire, Cleopatra, Elvis Presley, Taylor Swift, Paul McCartney, Michael Jackson, Lady Gaga, Bob Dylan, Mariah Carey, and Hillary Clinton), that splitting now is perhaps not the best timing considering that the pandemic is still ongoing, and that trimming is generally a better solution to overly lengthy content than splitting. I'll also note the existence of Wikipedia:Articles for deletion/Swedish government response to the COVID-19 pandemic, which makes this something which is discussed separately at two different locations.TompaDompa (talk) 00:18, 6 August 2020 (UTC)

• I also oppose splitting the page. There are other articles of comparable length, and given that Sweden has taken a different approach to the pandemic, it is highly relevant to make as much concise and reliable information as possible available to the public at large. David A (talk) 08:38, 6 August 2020 (UTC)

References

Comparing Sweden to New York

There is now at least one article comparing Sweden to New York as for covid-caused deaths:

• Why Sweden Succeeded in “Flattening the Curve” and New York Failed, July 15, fee.org

One should approach the above with great caution: it is not clear that Sweden and New York (state?) are comparable as for demographics, especially population interaction density. Unless that is controlled for and shown to be similar, the comparison is invalid. Intuitively, it might make more sense to compare New York state with Stockholm county, but that would have to be verified. New York state contains New York City, an international hub whose hub character seems much larger than that of Stockholm; furthermore, New York is home to sizeable black population, and due to some biological or cultural causes blacks are much more liable to die of covid per https://covidtracking.com/race, "Nationwide, Black people are dying at 2.5 times the rate of white people". Moreover, New York may have obesity rates different from Stockholm or Sweden, and obesity is a risk factor for covid-caused death. In general, comparisons that do not control for confounding factors are invalid. --Dan Polansky (talk) 08:24, 20 July 2020 (UTC)

Also one which could create difference in how mild and moderate cases are developing to deathly ones. The major reason for death with COVIC-19 is blood clotting which also correlates with lack of physical exercise which is problem if people are locked to inside. Not sure if there is any research but could be cited from multiple news pieces that doctors were alerted about this (example) In sweden there was no lockdown and when comparing to other nordic countries it's movement data was also affected but not so heavily [29]. Also from movement data when comparing walking in Finland and Sweden. In Finland goverments line in march was that people should have physical exercise and not just stay inside, but try not meet people outside of their household. --Zache (talk) 09:50, 21 July 2020 (UTC)

I recall a panel discussion TV show a while ago featuring, among others, Leif Pagrotsky and a Swedish/American doctor. They talked about New York having a significant number of people who belonged to one or more risk groups due to medical conditions, but wasn't under treatment for said conditions. This even included conditions like high blood pressure which is usually very easy to treat. If this is correct, it might be a part of the explanation. It's possible that people who belong to risk groups were more or less equally affected, but if people in risk groups in general are of poorer health, the risk of a worse outcome should increase.

There are some strange things in the article though. Like the sentence "It probably has much more to do with the fact that New York failed to protect the most at-risk populations: the elderly and infirm". If the writer suggest that Sweden succeeded, they would likely be seen as an alien from other space in Sweden. Another one is "Swedes fared so much better than New York, where authorities disempowered individual actors and prevented nursing homes from taking sensible precautions". Although I admit I have no clue what this is about, but suggesting that Swedish nursing homes were taking sensible precautions would be met with the same reaction. You don't have to look further than the edits from Zache on this page to learn that a big number of Swedish homes has been seen as everything but sensible (if you read this Zache, please consider making more contributions to the article).

And as you say, yet again a writer fails to explain why a specific comparision makes sense. In this case NY State with Sweden as a whole, including Lapland but excluding Alaska. Many would say it might be better to compare NY with the Stockholm region instead as the latter is densely populated and has a large metropolitan area (by Swedish standards that is), but I don't think that would necessarily be more logical. Just compare Skåne to other Swedish regions. To some extent, Skåne and Copenhagen could be seen as being in one metropolitan area. This group of writers includes the ones who always add comparisons with Sweden, Denmark and Norway to this article. So far the edits has offered no sourced explanations.

However, when we (not us specifically) discuss failures we pretty much put an equal sign between "belonging to a risk group" and "living at a nursing home". But most people in risk groups doesn't live in a nursing home and get by without home care. Perhaps old people and people with relevant health problems in general has been more protected in Sweden, and that they have been including this in the equation so to say.

I have to say the numbers for New York (1,728.7/M) is shockingly high. I get why the TV producers chose the sensationalist title "Döden i New York". bladjur (talk) 23:16, 29 July 2020 (UTC)

Sweden managed to flatten the curve without a lockdown, but they did it like almost every European country and indeed they have a lower rate of confirmed COVID cases per capita of national population than the USA (about 0.6% of Swedes vs a staggeringly high 1.6% of Americans). 2605:E000:100D:C571:8C36:F847:196F:592 (talk) 00:43, 7 August 2020 (UTC)

I'm not sure why 1.6% covid confirmed cases in the U.S. would be staggeringly high; it is in part a result of huge levels of testing in the U.S.[30] and in part a consequence of multiple U.S. states refusing to bleed their economies until Godot brings the vaccine, so they chose to partially reopen the economies. (The above does not seem particularly relevant to the topic of the thread, Comparing Sweden to New York, anyway.) For reference: Total confirmed COVID-19 cases per million people, ourworldindata.org, which shows 0.8% for Sweden and 1.44% for the U.S. --Dan Polansky (talk) 10:22, 7 August 2020 (UTC)

Comparing Sweden to Nordics

In diff, I again removed comparison of Sweden to Denmark and other Nordics. The comparison was traced to Worldometers, which only provides data and not country selection. It is not obvious that Nordics have similar characteristics relevant to the covid epidemic. One step to address the concern would be provision of solid source that does such a comparison; the source should demonstrate that Nordics have similar demographics, similar nursing home size, similar traffic density in urban centers, similar Spring holiday week, etc. Doing the comparison without tracing to solid source is not only original research, but also bad original research. --Dan Polansky (talk) 09:53, 6 August 2020 (UTC)

One defect of the removed comparison and its phrasing is the implication that lockdowns are effective in reducing deaths from the covid. For this, I know only one peer-reviewed scientific article:

• A country level analysis measuring the impact of government actions, country preparedness and socioeconomic factors on COVID-19 mortality and related health outcomes, July 21, 2020, thelancet.com

The above investigation included look at correlations of obesity, median population age, GDP, time to border closures from the first reported case, and more.

The above investigation does not support the notion that lockdowns are effective in reducing deaths from the covid. Does anyone have better article or another scientific article, whatever its conclusions?--Dan Polansky (talk) 10:22, 6 August 2020 (UTC)

I think that Dan Polansky makes sense as usual. David A (talk) 09:33, 8 August 2020 (UTC)

Relevant research by the Public Health Agency of Sweden

According to mathematical models by the Public Health Agency of Sweden, 26% of the population of Stockholm county (0.26 x 2.344 million = 609440 people) had been infected at some point on May 1 2020. I think that this seems very relevant to include in the page.

https://www.folkhalsomyndigheten.se/publicerat-material/publikationsarkiv/e/estimates-of-the-peak-day-and-the-number-of-infected-individuals-during-the-covid-19-outbreak-in-the-stockholm-region-sweden-february--april-2020/ David A (talk) 19:16, 18 May 2020 (UTC)

Blådjur would you be interested in adding this information and reference to the page where it is appropriate? David A (talk) 08:23, 19 May 2020 (UTC)

@David A: Where's appropriate.. I tried looking at some other "COVID-19 pandemic in..." articles but couldn't find much. I propose a new section for Monitoring and modeling. There's some stuff in the Response from the health-care sector section about studies on infection spread and it's out of place there anyway. What do you think? bladjur (talk) 10:25, 19 May 2020 (UTC)

That seems fine to me. Thanks for helping out. David A (talk) 11:43, 19 May 2020 (UTC)

Is anybody here willing to add a reference to this study to the page? It seems very relevant to me. David A (talk) 11:45, 21 May 2020 (UTC)

The study[31] (and pdf[32]) was published on 30 April 2020 and uses one of those epidemiological models that are too likely to lead to wrong results. I vaguely remember there are now better studies, based on testing for antibodies. --Dan Polansky (talk) 07:26, 22 May 2020 (UTC)

There may be a problem with using testing for antibodies to find out how many people have been infected. https://www.biorxiv.org/content/10.1101/2020.05.13.092619v2 indicates -- though beware of small sample size, again -- that people with mild covid-19 infections clear the infection without producing much in the way of antibodies. Thus the number of people who were infected, recovered, and still test negative in a serum test may be a whole lot higher than expected. I am waiting to see what happens to this paper after the peer review. Lacreighton (talk) 16:29, 22 May 2020 (UTC)

Interesting; thank you. --Dan Polansky (talk) 16:33, 22 May 2020 (UTC)

New information: https://unherd.com/thepost/swedish-doctor-t-cell-immunity-and-the-truth-about-covid-19-in-sweden/ Swedish epidemiologist who studies t-cell immunity says that 2x as many people have t-cell immunity as have covid antibodies. T-cell immunity generally is good for a decade or more, in other diseased, but of course we can only make a best guess at a new disease. This means that the doubts I mentioned above about testing for antibodies are legitimate. Antibodies are thus only one path to herd immunity, and not having antibodies does not necessarily mean that you are at risk to contract the disease again. Lacreighton (talk) 17:13, 11 July 2020 (UTC)

Indeed, you were right. There are other links about T-cell-based immunity in #Seroprevalence and prevalence in Stockholm thread below (or in talk page archive, later). --Dan Polansky (talk) 10:40, 8 August 2020 (UTC)

Excess mortality per mortality.org

Excess mortality for Sweden is available not only in EuroMOMO but also in mortality.org, which has visualization in shinyapps.io[33]. The vizualization shows deaths rather than EuroMOMO's z-scores and it allows to show deaths per sex and per age group. The age group can be seen e.g. by choosing "Deaths, Ages 15-64" in "Measure" field. As a baseline to compare to, they use averages over a time range, and the time range can be adjusted. I am not sure how to use this in the article; perhaps it could be added as an external link. --Dan Polansky (talk) 10:03, 31 July 2020 (UTC)

@Dan Polansky: ≤When reading this article in Dagens Medicin I thought it might be of interest to you: https://www.dagensmedicin.se/artiklar/2020/08/12/farre-avlidna-efter-var-med-overdodlighet/

"The number of deceased has decreased since the beginning of July, and according to Statistics Sweden (SCB), the number is now below average." bladjur (talk) 17:58, 13 August 2020 (UTC)

Also, seems like Statistics Sweden still publish new data on number of deaths weekly: https://www.scb.se/om-scb/nyheter-och-pressmeddelanden/folj-preliminar-statistik-om-dodsfall/ bladjur (talk) 18:07, 13 August 2020 (UTC)

@Blådjur: Thanks; I updated the excess death chart in the article. I now placed another kind of chart to Wikiversity:COVID-19/All-cause deaths/Sweden and Wikiversity now has charts for many other countries at Wikiversity:COVID-19/All-cause deaths, especially European counties but also Israel, U.S. and few selected U.S. states. --Dan Polansky (talk) 09:21, 15 August 2020 (UTC)

Backstory behind the politics

• Expressen, 2020-08-11, Mejlen som avslöjar Gieseckes inflytande över coronaplanen (Part 1: The email revealing Giesecke's influence over the corona plan)
• Expressen, 2020-08-12, ”Det låter som om man är villig att acceptera sjukdom och död” (Part 2: "It sounds like you are willing to accept illness and death")
• Thelocal.se, 2020-08-12, Why is everyone in Sweden talking about Anders Tegnell's emails? (mostly commenting part 1)
• Guardian, 2020-08-17, /Swedens covid 19 strategist under fire over herd immunity emails (based on Expressens story + deleted emails)

I created a topic for the sources related why/how Sweden chose its strategy. --Zache (talk) 19:46, 18 August 2020 (UTC)

Cases Past 100,000?

Sweden has now passed 100,000 cases. Somebody change the color on the world map. — Preceding unsigned comment added by 2600:1700:82E0:A690:75AB:917F:FB05:DCB6 (talk) 19:51, 14 October 2020 (UTC)

gatherings of fewer than 50 people should be preceded

Hey, Article reads "The agency also recommended that plans for events and gatherings of fewer than 50 people should be preceded by a risk assessment and, if necessary, followed by mitigation measures". But I check the source [34] , translate it and it says "the Government has decided to ban public events and public gatherings with over 50 participants". Could somebody please check. Manyareasexpert (talk) 17:21, 22 October 2020 (UTC)

Ok found that article also says "On 27 March the government announced that the ban on public gatherings would be lowered to include all gatherings of more than 50 people" , the topic could be closed now Manyareasexpert (talk) 13:49, 23 October 2020 (UTC)

Excess mortality - monthly chart

Are there any concerns about adding following chart (from official source) into Excess mortality section. 84.47.179.91 (talk) 18:18, 30 November 2020 (UTC)

source: Statistics Sweden

All-cause monthly deaths per 100 000 inhabitants 2015–2020

References

Timeline needs updating

Somebody with access to the info pls update the "Timeline of Daily Cases" graph. It has stopped at Jan 7, 2021. Titus III (talk) 00:31, 12 January 2021 (UTC)

Please explain your concerns regarding migrating Cases / ICU hospitalisations / Deaths charts to Tabular Data data source

@Cubeloadan: please explain your concerns regarding migration of chart from sections COVID-19_pandemic_in_Sweden#Statistics to Tabular Data data source (c:Data:COVID-19_Sweden daily cases hospitalisations deaths.tab). Also I need to pay your attention that I've asked for feedback for this intent yet 26 February 2021 in last comment of discussion #Updating_the_"Deaths_per_day"_graph? : "Are there any thoughts/suggestions/objections about going that way ?" So please explain the reasons behind your reverting of charts migration in your contribution: Special:Diff/1009669195
84.47.179.91 (talk) 20:26, 1 March 2021 (UTC)

@Cubeloadan: I've turned back tabular data based version of charts since I strongly believe that new version of charts has better maintainability then former one. And for now I literally see no reasons (no other editors thoughts or comments) to think that new version of charts is worse than previous. 84.47.179.91 (talk) 12:47, 2 March 2021 (UTC)