Stortorget square in Stockholm old town, Sweden

The progress of the COVID-19 epidemic in Sweden: an update

Reposted from Dr. Judith Curry’s Climate Etc.

By Nic Lewis

I thought it was time for an update of my original analysis of 28 June 2020. As I wrote then, the course of the COVID-19 pandemic in Sweden is of great interest, as it is one of very few advanced nations where no lockdown order that heavily restricted people’s movements and other basic freedoms was imposed.

Unfortunately, some of the comment on how the COVID-19 epidemic has developed in Sweden has been ill-informed. Indeed, a shadowy group of academics, opinion leaders, researchers and others who are upset about Sweden’s strategy and are actively seeking to influence it has been unmasked. They have been coordinating efforts to criticize media coverage of Sweden’s strategy and to damage both the image of Sweden abroad and the reputation of  individuals who work in this field.

I present here updated plots of weekly new cases and deaths, with accompanying comments.[1]

Key Points

  • Despite Swedish Covid cases falling to low levels in the summer, they resurged in the autumn
  • This second wave, which was very likely a seasonal effect, now appears past its peak
  • Excess deaths in East Sweden were high in the first wave and low in the second; for South Sweden the opposite is true. This suggests that population immunity and/or the remaining number of frail old people are key factors in the severity of the second wave.
  • Excess deaths in Sweden to end 2020 were modest, particularly for 2019 (when deaths were abnormally low) and 2020 combined. They appear to be much lower relative to the population than in England, despite far harsher restrictions being imposed there.
  • Only 3% of recorded 2020 COVID-19 deaths in Sweden were of people aged under 60

Overall development of the epidemic

Figure 1 shows the overall picture for confirmed weekly total new COVID-19 cases, intensive care admissions and deaths in Sweden, up to data released on 9 February 2021. The criteria for testing were widened during the early months, so case numbers up to June 2020 are not comparable with subsequent ones. Weekly new cases have been divided by 50 in order to make their scale comparable to that for ICU admissions and deaths.  Death numbers for the two final weeks will be noticeably understated due to delays in death registrations.

Fig. 1 Total weekly COVID-19 confirmed cases, intensive care admissions and deaths in Sweden

In late summer 2020 it looked as if the epidemic had burnt itself out, however a strong second wave developed during September to December. Although start of school and university term, along with more relaxed behaviour, may have started the second wave off, over the period as a whole the primary driver was almost certainly a seasonal increase in the virus’s transmission and hence reproduction number. Studies that indicated a lack of substantial seasonality in transmission[2] [3] have been proven wrong.

Analysis by age group

The changing age composition of new cases over time is shown in Figure 2. Case numbers before and after June 2020 are not comparable because of the major widening of testing during June 2020. However, it is clear that the second wave has been dominated by infections of people aged 10 to 59 years.

Fig. 2 Weekly COVID-19 confirmed cases by age group in Sweden

After falling to very low levels in late July 2020, weekly COVID-19 recorded deaths rose strongly from late October on, across all age groups (Figure 3). The data show the number of people with confirmed COVID-19 who died, regardless of the cause of death. In total, about 50% of deaths occurred up to and after 30 September 2020, that is in the first wave and in the second wave (which is, however, not over yet). During the second wave, a slightly higher proportion of deaths have been of people aged 80+ years than in the first wave.

Fig. 3 Weekly COVID-19 recorded deaths by age group in Sweden

Regional analysis

I turn now to regional analysis. Figure 4 shows weekly confirmed new cases for each of the 21 regions in Sweden. Although widening of testing (mainly in the second quarter of 2020) varied between regions, it is evident that Stockholm and Västra Götaland, which dominated cases during the first wave, were also two of the three regions dominating the second wave, with Stockholm region leading both waves. However, while Skåne had relatively few first wave cases, it broadly matched Stockholm in the second wave, albeit with a delay. Cases have fallen quite sharply in almost all regions since the turn of the year.

Fig. 4 Weekly COVID-19 confirmed cases by region in Sweden

Regions have varying populations, so confirmed cases per 100,000 head of population give a better picture of relative disease incidence (Figure 5). There is negligible correlation between the regions that had the highest incidence of COVID-19 cases during the first wave (including or excluding June to August) and the post-September 2020 period. In the absence of growing population immunity having an effect, one might expect that in those regions in which the virus spread most easily prior to September 2020 (by which time it was well ensconced in all regions) – for instance, due to greater urbanisation – it would also have spread most easily in the second wave, in the absence of changes in other factors. A lack of correlation between cases in the first and second waves is consistent with greater population immunity in those regions that were harder hit in the first wave counteracting, during the second wave, the greater ease with which infections spread there when population immunity was low.

Fig. 5 Weekly COVID-19 confirmed cases per 100,000 head of population by region in Sweden

As for cases, it is difficult to discern an obvious relationship across regions between COVID-19 deaths per 100,000 people in the first and the second waves, and the correlation between them is negligible. The non-identity between recorded COVID-19 deaths and those actually caused by the disease may be one reason for this. A somewhat clearer picture comes from examining weekly excess deaths in geographical super-regions, as shown in a recent Swedish report.[4]

Fig. 6 Weekly deaths (purple line) up to week 3 of 2021 in Sweden compared with the expected normal death toll (solid green line) and its 95% confidence interval (dashed green line)

Figure 6 shows the position for Sweden as a whole. Data go up to week 3 2021; data for more recent weeks are incomplete. Peak excess deaths were higher in the first wave than in the second wave, the opposite relationship to that for recorded COVID-19 deaths. While this likely partly reflects an element of undercounting of deaths caused by COVID-19 at the peak of the first wave, it appears to be mainly due more to a considerably larger over counting of COVID-19 deaths throughout the second wave. While the second wave is not over yet, it does appear that excess deaths have peaked.

Figure 7 shows deaths for East Sweden, the population of which is dominated by Stockholm region.  Excess deaths in the first COVID-19 wave were further above normal than for Sweden as a whole, but excess deaths during the second wave peaked at a level not much above that in the 2017/18 flu and pneumonia season, and fell back within the 95% confidence interval by the end of 2020 (and to close to normal for Stockholm region alone).

Fig. 7 As Figure 6 but for East Sweden (Stockholm, Uppsala, Södermanland, Östergötland, Örebro, Västmanland)

However, in Southern Sweden, the picture is quite different (Figure 8), with the second wave of COVID-19 excess deaths being considerably larger than the first, which was smaller than in the 2017/18 flu season.

Fig. 8 As Figure 6 but for South Sweden (Jönköping, Kronoberg, Kalmar, Gotland, Blekinge, Skåne, Halland, Västra Götaland)

The population of South Sweden is dominated by that of Västra Götaland in the north west and Skåne in the south, which contain respectively Sweden’s second and third largest cities (Gothenburg and Malmö). In southernmost Sweden, the first wave barely breached the upper bound of the 95% confidence interval for normal deaths, while peak excess deaths in the second wave were three times that level (Figure 9). In the remainder of South Sweden, excess deaths in the second peaked at a broadly similar level to in the first wave. The same is true for North Sweden, which is relatively sparsely populated and has few sizeable cities.

Fig. 9 As Figure 8 but for southernmost Sweden (Blekinge, Skåne) alone

In my view, the pattern of excess deaths in waves one and two in Stockholm-dominated East Sweden, compared to that in other parts of Sweden, suggests that much of the pool of people in East Sweden vulnerable to dying from COVID-19 had already succumbed by the end of wave one. On the other hand, although the level of previous infections and hence population immunity in Stockholm region at the end of wave one was more than adequate to inhibit large scale spread of COVID-19 during the summer, at the level of population mixing occurring then, with hindsight it was at that stage clearly insufficient to provide herd immunity in the winter, when transmission is higher, causing both the virus’s reproduction number (R0) and the herd immunity threshold to rise.[5]

Although it is too early to be certain, at present it appears that population immunity in both Stockholm region and Sweden as a whole is now adequate to prevent large-scale COVID-19 epidemic growth even in winter, at least at the current level of population mixing. However, there is a caveat in that the B.1.1.77 (UK-discovered) variant, which is estimated to be about one-third more transmissible[6] – and hence faster growing – only became apparent in Sweden during December. While present in 35% of all Swedish sequenced genomes during the first three weeks of 2021, it is not yet dominant, so transmission can be expected to rise as it achieves dominance over the next two or three months.

Total Swedish deaths due to COVID-19

Sweden had 10,082 deaths with confirmed COVID-19 infection for the 53 reporting weeks of 2020, ending 3 January 2021, including those reported subsequently.[7] On another measure[8], there were 9,432 deaths. Only 0.9% of deaths were of people under 50 years old, and only 3% were of under 60 year olds. People over 70 years old accounted for over 91% of COVID-19 deaths.

The definition of COVID-19 deaths imposed by the WHO is likely to over count deaths caused by COVID-19, since where there are multiple causes contributing to a death clinically-compatible with COVID-19 (normally respiratory failure or acute respiratory distress syndrome) it will be recorded as a death due to COVID-19 where SARS-CoV-2 infection is confirmed or suspected, even if COVID-19 is not considered to be the main cause of death.[9] Moreover, some countries have adopted even more over-pessimistic definitions of COVID-19 deaths. Others have likely undercounted COVID-19 deaths. And in many countries some deaths caused by COVID-19 at the start of the epidemic were probably not recognised as being such. Therefore, excess mortality over a normal level is usually thought to be the best measure of deaths due to COVID-19.

Excess mortality is primarily affected by the severity of respiratory disease (mainly influenza) winter seasons. A severe flu season, which may be caused by a new influenza virus strain, results in many more very frail unhealthy old people dying than a mild flu season. Severe flu seasons may occur in pairs in adjoining years, for instance due to widespread vulnerability to a new strain.

It follows that, other things being equal, fewer deaths will tend to occur in a flu season that follows a severe flu season, even more so where that is the second of a pair of severe flu seasons, as there will be fewer than normal very old and frail people alive. Correspondingly, more deaths than usual will tend to occur in a season following one or more mild flu seasons. This is known as the “dry tinder” effect. It has been shown, for example, that across 32 European countries there is a significant negative correlation (–0.63) between flu intensity in winter 2018/19 and 2019/20 combined and the COVID-19 mortality rate in the first wave (Figure 10).[10]

Sweden had unusually low mortality in 2019, which is largely a reflection of mild late 2018/19 and early 2019/20 flu seasons (the early and late part of each flu season falling in different calendar years). It thus had higher than usual “dry tinder” when the COVID-19 epidemic started.

A detailed analysis by a Danish researcher of the influence of “dry tinder” in Sweden, published by a US economic research institute, concluded that it accounted  for many COVID-19 deaths.[11]

Similarly, an analysis by an economics researcher at a US university[12], which looked at 15 factors apart from severity of government interventions that might explain the higher COVID-19 deaths in Sweden than in other Nordic countries, concluded that the “dry tinder” factor was the most significant one. That paper also considered it plausible that Sweden’s lighter government interventions accounted for only a small part of Sweden’s higher Covid death rate than in other Nordic countries.

Fig.10 Death rate from COVID-19 up to 10 June 2020 by total 2-year flu intensity for 32 countries. The R2 of 0.396 (r=−0.63) is significant at the 1% level. A reproduction of Figure 1 in reference 10.

A fair estimate of excess deaths in Sweden caused by COVID-19 in 2020 should reflect the unusually large number of very old and frail people who survived 2019. That can be done by comparing actual and predicted deaths for 2019 and 2020 combined.[13]

I calculated excess mortality in Sweden for each year from 2000 on, by 5-year age group and sex, as the difference between actual mortality and normal mortality predicted by a regression fit to actual mortality rates over either 2000–2018 or 2009–2018, and then used population data to derive the expected number of deaths in a normal year for 2019 and 2020.[14] Mortality rates have been declining since 2000 in all age groups, although more slowly over the last decade. However, the effect on overall mortality of declining mortality rates at each age is partially counteracted by the increasing average age of the population.

When estimating normal mortality from trends in mortality over ,alternatively, 2000–2018 or 2009–2018, the excess combined 2019 and 2020 deaths were respectively 4,500 or 2,100, representing 0.043% or 0.020% of the mid-2020 Swedish population. For 2020 on its own, calculated excess deaths are 6,900 or 5,600 for the two regression bases (0.066% or 0.054% of the mid-2020 Swedish population).

Excess deaths for 2019 and 2020 combined were largely of men aged 65–79 and (to a lesser extent) aged 80–89 and 90+. Excess deaths of women were under 30% of those of men, based on mortality predicted by regressing over 2000-2018, and were actually negative based on regressing mortality over 2009-2018. On both regression bases and for both sexes, 2019 plus 2020 deaths of under 65 year olds were lower than predicted. And average overall mortality for 2019 and 2020 combined was lower than for any previous year this century (and very probably ever).

A more detailed analysis of Swedish mortality in 2020, but which used incomplete deaths data, was published a month ago by the blogger swdevperestroika; it is well worth reading.[15] That article made similar points, and reached similar conclusions, to my own analysis.

Comparison of Swedish and English excess deaths

I applied a similar analysis method to derive excess deaths in England for 2019 and 2020. The data published in England are rather less satisfactory than in Sweden, so the derived estimates should be regarded as approximate. I used data from Table 1 of the UK Office of National Statistics (ONS) monthly mortality analysis for December 2020, which spans 2001 to 2020.[16] Doing so gives best estimates for combined 2019 and 2020 excess deaths of 113,000 (0.20% of the population) when predicting normal deaths by regressing age-standardised annual mortality rates over 2001–2018, or 44,000 (0.08% of the population) when regressing over 2009–2018. The estimated excess deaths for 2020 alone were respectively 95,000 and 58,000. Other data published by the ONS suggests 2020 excess deaths in England were modestly below the average of these two estimates, and represented some 0.13% of the population.[17]


Whether the longer or shorter regression periods provide better estimates of normal mortality in 2019 and 2020, it seems clear that excess deaths, as a proportion of the population, were much higher in England than in Sweden. Excess deaths in England per 100,000 population were about four times those in Sweden for 2019 and 2020 combined, and about double those in Sweden for 2020 alone, .

Nicholas Lewis                                                                                       18 February 2021

Originally posted here, where a pdf copy is also available

5 26 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
February 18, 2021 2:33 pm

It looks worse than other Nordic countries, but not Apocalyptic like the governments would like us to think.

Reply to  Jeff in Calgary
February 18, 2021 2:44 pm

As Figure 10 shows, COVID takes the leftovers from influenza from the prior two years. This would explain at least part of the different outcomes among the Nordics.

Cyril Wentzel
Reply to  Jeff in Calgary
February 18, 2021 3:32 pm

Other Nordic countries – as Nic points out – did not have the mild flu seasons in the dry tinder range, although the graph shows not that big of a difference on the horizontal axis for Norway and Sweden.

I find it shocking to see what superficial and tribal intensity is on display by people and institutions who just want to ‘prove Sweden wrong’, merely to the effect that evade the inevitable conclusion that they all have it completely reversed.

It will be good to combine these insights with serology studies as well.
Good informative analysis Nick, thanks a lot.

Reply to  Jeff in Calgary
February 19, 2021 8:11 pm

“it is one of very few advanced nations where no lockdown order that heavily restricted people’s movements and other basic freedoms was imposed.”

This is completely wrong.
We in the Baltic states (just opposite) had no lockdown at all.

Estonians reckon the reason why mortality is so incredibly low, is because people are exposed to flu every winter, and have learnt to live with it, without fear.

Right now the (3rd wave actually) is going crazy up there, as confirmed by the sewage traces, – which confirm infection rates way above testing levels..
..still very low mortality, and people go to work normally.

The Swede
February 18, 2021 2:40 pm

It is also interesting to note that the seasonal Influensa has been nonexistent during the winter of 2020/2021 the fatalities from seasonal influenza that usually occurs during winter is completely absent due to social distancing. Usually sweden has between 400 to a 1000 fatalities each winter from seasonal flu. Now its five times worse due to COVID

Robert W Turner
Reply to  The Swede
February 18, 2021 5:49 pm

Could be due to anyone susceptible to dying from influenza already passed away from covid-19 last winter. I expect the following years are going to be very low in excess deaths everywhere, unless a new designer plague is released.

Paul Penrose
Reply to  The Swede
February 19, 2021 2:34 pm

It’s also possible that many of the the deaths that would have been attributed to influensa in the past have been attributed to Covid19 instead.

February 18, 2021 3:01 pm

I like this video.

Also, the following compares and contrasts Florida’s relatively open response vs. California’s strict lockdowns.

Pat Frank
Reply to  Scissor
February 19, 2021 8:15 am

I couldn’t independently confirm that bitchute video, Scissor. It’s likely not true.

Reply to  Pat Frank
February 20, 2021 6:03 am

You’re right, it’s suspicious. I looked too and didn’t find a source.

Reply to  Scissor
February 20, 2021 2:25 pm

Scissor, I don’t like the video at all. No names, no doucments, no lawsuit–are they really suing the CDC? Where and what are the charges–it seems bogus to me–who is he and why woudl the CDC give him samples of anything? If I asked the CDC for clean samples of covid–and they gave it to me, I would sue them! for being unsafe!

February 18, 2021 3:15 pm

I’m having trouble finding numbers I can believe but …

In England, blacks and asians are disproportionately affected by the wuflu. So, is ethnic makeup, by itself, responsible for the number of cases, hospitalizations, and deaths?

We seem to be seeing a world wide downtrend in the wuflu no matter what individual countries are doing. Is it possible that the lockdowns are completely ineffective?

Notwithstanding the above, why are several asian countries, mostly bordering China, doing so much better than everyone else?

Asians from the sub-continent aren’t the same as asians from the countries bordering China. Would I be correct in guessing that asians in England are by far mostly from the sub-continent?

How is New Zealand doing?

There are so many questions. It’s not obvious to me that we can put the differences between Sweden and England down to policy questions. There are simply too many lurking variables.

Reply to  commieBob
February 18, 2021 3:40 pm

Good questions. People with darker skin at northern latitudes tend to be deficient in vitamin D. Lockdowns at the very least do more damage from job loss, business failure, depression, substance abuse, suicide, etc.

Corona viruses tend to originate in Asia, so people living there have had exposure to similar viruses and hence some immunity from this.

This is a good site re: global/country CV cases

Robert W Turner
Reply to  Scissor
February 18, 2021 5:52 pm

Exactly. It’s quite simple really. It’s amazing how so many in the medical community continue to say it’s unknown why the cold and flue season is worse in winter.

Michael in Dublin
Reply to  Scissor
February 19, 2021 2:20 am

The vitamin D point is worth examining carefully because of the discrepancy between the low numbers of deaths in numerous African countries and blacks in the UK. Where the numbers of deaths are much higher, and from a younger age group, like in South Africa, I suspect that AIDS and TB are key factors. However, this information is hidden from the public.

Reply to  commieBob
February 18, 2021 6:00 pm

Another correlation I’d like to see is wuhan flu cases versus density of household population. I’ve seen hints at the flu being worst where large, extended families live together in close proximity.

Reply to  Bruce
February 19, 2021 12:30 am

Yes, cultural differences are generally ignored ( that would racist , right : it’s OK to suggest they are “victims” of being “people of colour” but NEVER suggest it is an effect of their customs are practices, or family size ).

That these communities tend to have more cohesive, extended families is likely a factor. This is recognized in Italy but that’s OK to say because they are white and Catholic.

Asian and african countries are less affected because they are not scared of hydroxychloroquine and it is cheap and freely available not BANNED.

Adam Gallon
Reply to  commieBob
February 19, 2021 12:54 am

In the UK, people with origins in the Indian subcontinent, tend to live in multigenerational families & many in relative poverty in overcrowded housing. There’s also quite a large number, especially of the older women, who have limited command of English, so the Government messages of social distancing & mask wearing, didn’t hit home as well as they could. Thus the younger family members, who often have poorly paid jobs, in hospitality, public & private transport, can’t afford to stop working, don’t qualify for furlough & if they’ve contracted Covid-19, are asymptotic or only have mild symptoms, bring the disease into the family home, where it strikes their grandparents, etc, with predictable results.

Gerry, England
Reply to  Adam Gallon
February 19, 2021 12:55 pm

This would seem to have been backed up by an otherwise truly awful Channel 4 programme called ‘Is Coronavirus Racist’ which blamed white people for the death rate in ethnic minorities. They found that there is no physiological difference that would cause increased deaths. So as Adam says, their jobs, living arrangements etc. contributed. Something that has disappeared from mention is the concept of ‘multiple infections’, where if you keep coming into contact with Covid from different people it could prove more deadly. This could explain lower age group deaths of healthy health sector workers, etc.

Xinnie the Pooh
Reply to  commieBob
February 20, 2021 5:15 am

Blood Type also has statistically significant effects on mortality

Mike O
February 18, 2021 3:20 pm

There have been many analyses of excess deaths which attribute the excess entirely to Covid. We may never know how much was Covid and how much restrictions on other types of health care. I saw a statistic a while back that there were only about 40% as many call for ambulances for severe cardiac issues as there were prior to the pandemic.

Reply to  Mike O
February 18, 2021 3:39 pm

And recently it was reported that a total of 7 (seven) cases of flu have been recorded in Sweden this winter. So while measures taken in Sweden have proven ineffective to stop Covid-19 they have been completely effective in stopping transmission of flu.

So apparently Covid-19 is considerably more infectious than flu.

Robert W Turner
Reply to  tty
February 18, 2021 5:55 pm

Perhaps anyone that would go to the hospital due to influenza already expired due to covid 19 last year. A dearth of international travel may play a role as well since influenza spread relies on mutation and exposure to new hosts without antibodies.

Intelligent Dasein
Reply to  Robert W Turner
February 19, 2021 5:10 am

Or perhaps they are just renaming flu cases as Covid cases. Think.

Reply to  tty
February 18, 2021 11:37 pm

Ivor Cummings has a good discussion of the phenomenon where a disease that is prevalent causes a decline in the incidence of other diseases. Apparently the effect is well documented if not well understood. Personally I think that the medical community downplays the innate imune response. If you get sick but are still robust, your innate immune response goes into overdrive for up to several weeks which decreases the likelihood of you getting a second disease while recovering from the first. If our bodies didn’t work that way, every major illness would begin a death spiral of diseases taking advantage of individuals weakened by previous illness. The adaptive immune response provides long term protection but the innate response provides short term and front line protection. This, incidentally, is why the vaccine trials are weak-as far as I can tell, none of them controlled for the innate immune response caused by the vaccine by using a placebo that also caused an innate immune response. There are consequences to medical researchers not knowing what a controlled trial is. They routinely control for the real but minor placebo effect but then miss controlling for a major confounding factor like innate immune response. Sadly, most medical researchers think the nil treatment is a control, whatever the heck they mean by that, but they don’t seem to understand the concept of a controlled experiment. Just one of many reasons to read Big Pharma publications with a jaundiced eye.

Reply to  BCBill
February 19, 2021 4:11 am

innate immune response goes into overdrive”

Vitamin D Suppresses the Cytokine Storm – YouTube

That overdrive is called a cytokine storm.

This is why these respiratory diseases are seen as seasonal

Paul C
Reply to  BCBill
February 19, 2021 6:59 am

I recall seeing that theory on innate immunity being primed by a competing virus. The analysis focused of SARS-1 arrival in France during a flu wave. Apparently it unexpectedly did not spread until after the flu season was over. Similarly it has been observed that people don’t get a cold and flu at the same time, though an opportunist bacterial pneumonia is more likely. So the theory is the innate immune response targets one category of infectious agent, but the weakened body is vulnerable to different categories of infection. Hence the antibiotic included in many reportedly effective early COVID-19 therapies.
One of the theories why lockdowns have been so ineffective is the suppression of other respiratory virus infection (colds/flu) may have increased susceptibility to the Whan coronavirus. It also theorises that deliberate introduction of a relatively benign cold virus to an outbreak location may prime people’s immune systems to a viral response and suppress the more dangerous virus. Whether an adaptive immune response (antibody) is activated against the suppressed virus conferring lasting immunity is the unanswered question.

Peta of Newark
February 18, 2021 3:32 pm

Not afraid of their animal products and fat..
Quote from the wiki
“”Swedish cuisine could be described as centred around cultured dairy products, crisp and soft (often sugared) breads, berries and stone fruits, beef, chicken, lamb, pork, eggs, and seafood.””
Big tradition of oily fish too

Big regional variations too…..

Not so strong on the veggies: ##
“”Sweden’s long winters explain the lack of fresh vegetables in many traditional recipes“”

## Wonders if that’s why kids around the world are immune to Covid, they universally hate eating vegetables?

Michael E McHenry
February 18, 2021 3:42 pm

The population of New Jersey USA at 8.9 million is close to that of Sweden. Here is the mortality curve to date. New Jersey went for strict measures. I don’t think you can compare countries. I also think COVID transmission is very poorly understood

Confirmed_Probable Dashboard.png
Reply to  Michael E McHenry
February 18, 2021 8:38 pm

New Jersey has a population density of 1224/sq.m. Sweden is 64/sq.m.

To get a better picture you would need to compare across the main population centres. However spread of infection will be increased by crowding.


Michael E McHenry
Reply to  RickWill
February 18, 2021 8:56 pm

That’s what I said you can’t compare countries

Reply to  Michael E McHenry
February 19, 2021 9:03 am

You can’t even compare counties in the US, by the same measure.

Reply to  RickWill
February 18, 2021 11:55 pm

Population density across the entire country is the wrong metric – not many people to contract or pass on a virus in a remote northern forest. A more suitable one would be percentage living in a urban environment – in Sweden it’s 88% and New Jersey 95%, so not that wildly different.

Michael E McHenry
Reply to  Fenlander
February 19, 2021 7:47 am

Good point

February 18, 2021 4:30 pm

It seems lockdowns didn’t help, since Sweden didn’t fare any worse than the lockdown-happy states.


I have issues with “confirmed cases.” I see this phrase used in many places, often without any specifics of the testing methodology. We know the PCR test is plagued with false positives, especially if the Cycle Threshold is above 25 or 30. Every positive PCR test result is called a “case” even if the individual shows mild or no symptoms. Death totals can be questionable as well, as there is often no distinction between dying from or dying with CV-19.


Keep in mind that according to the WHO, there are normally about 1 billion cases of influenza every year, resulting in 290,000 to 650,000 influenza-related respiratory deaths.



Joel O'Bryan
Reply to  PaulH
February 18, 2021 5:57 pm

Confirmed cases also simply reflects how aggressively that PCR testing was pursued.

Robert of Texas
February 18, 2021 4:38 pm

The lack of much FLU this season is very likely to come back to haunt us. Since people did not spread much FLU virus about in 2020, it now has much more time to mutate into forms harder for the immune system to recognize. When the lock downs and mask wearing stop and we get to a new FLU season, it may be a bad one as far as how many people get the FLU and how bad the symptoms are. One cannot really predict the death rate on any given year – it just depends on the mutations.

As for Sweden, one simply cannot say why COVID-19 was less severe there. Trying to use this as an argument that lock downs and masks do not work is simply cherry picking your example. There are far too many variables we still do not understand to point to just one of them. It could be as simple as fair skinned people who are outdoors a lot are less susceptible, or something in their diet, or just life style.

What scares me most is medical people now know we can almost eradicate the spread of the FLU through enforced lock downs and mask wearing (not sure mask wearing works against the FLU but something obviously does). We need to be developing highly effective long lasting vaccines against diseases, not hiding from them.

Reply to  Robert of Texas
February 18, 2021 8:08 pm

It might be cherry picking, but as Sherlock Holmes might have observed, it is curious just how many cherries there are to pick in the middle of winter…

Reply to  Robert of Texas
February 19, 2021 3:23 am

admit Im concerned at not getting my yearly colds/flu this yr
when they do come back as OS travel brings em into Aus theyre going to knock some of us sideways like the previous years did.
no immune system kick this yr makes for increased risk to me.

Abolition Man
February 18, 2021 5:46 pm

Thank you, Nicholas,
There has been a veritable cornucopia of pandemic panic and fear porn spread by our lying leaders and media regarding the ChiCom-19 virus! It is nice to see someone trying to cut through the hyperbole and present the real data in a concise and easily understood manner!

It is so frustrating to see many, seemingly rational people overreact to a serious, seasonal virus that could have been easily dealt with by the medical community if the politicians had not stuck their snouts into the mix! I believe there are now thirty studies showing lockdowns are ineffective at best and are likely far, far worse the the virus itself! I was wondering if you had heard any theories as to why flu deaths have dropped off in this time of corona virus? It is almost as if the viruses have a competition for hosts with the winner taking virtually all of the spoils!
I believe that, looking back on 2020, we will see we had all the information we needed from the Princess Cruiseline data and the nursing homes in Washington State, but chose to ignore the evidence and try something completely untested at the recommendation of our BBF; the completely honest and honorable Chinese Communist Party! It would appear that many of our leaders are more comfortable working for the CCP now than for the benefit of their own people! It must pay quite well!

Reply to  Abolition Man
February 19, 2021 1:30 am

It would appear that many of our leaders are more comfortable working for the CCP now than for the benefit of their own people! It must pay quite well!

Payment isn’t necessary. Our elites are more comfortable working with their elites than our elites are comfortable working with us.

Our old buddy, Noam Chomsky, observed that press censorship wasn’t necessary during the early part of the Viet Nam war era because the owners of the media were members of the establishment. They didn’t need to be explicitly told what to print. It really is just a question of who folks identify with. After that’s established, they don’t need to be told what to do, they’re smart enough to figure it out for themselves.

Joel O'Bryan
February 18, 2021 5:51 pm

I find it amazing that the “shadowy group of academics, opinion leaders, researchers …” actually think they could get away with that subterfuge in this age.

The Climate Scammers at UEA CRU and in the US, and the US Democratic National Committee members have learned their lesson the hard way that if they coordinate their dirty-deeds online in emails or social media in their own name, they will be uncovered/un-masked.

After Climate Gate, Part Deux the climate scammers have undoubtedly adopted using direct phone calls, unrecorded group calls, and other more secure means not subject to FOIA or to coordinate their dirty deeds. Then after the 2016 fiasco of their emails being released, the US DNC and Leon Panetta learned to take their corruption coordination to other means.

Reply to  Joel O'Bryan
February 18, 2021 7:47 pm

I have mentioned before that India experienced as bad a case rise up to September as the US (and Sweden) but since then has declined steeply in deaths and cases. This needs to be examined more closely. see
Just like the pitiful information in the US, the Indian press ranges all over the map on the possible explanations. Most are patently ridiculous, such as “it’s herd immunity.” Can’t be, the % of the population as “cases” is much, much less than the US, with no herd immunity.
Vaccines? (India Big Pharma actually works) – not in September.
The best explanation so far is wide acknowledgement of the importance of Vitamin D3, supplemental zinc and Ivermectin. Some say HCQ.
The Sweden case is obvious. Most Indians live in naturally ventilated, often warm houses, where the virus is ventilated out. In the Winter, Swedes seal up their houses and trap the small aerosol particles inside. Also, low winter humidity (indoors) dries virus particles, shrinking them and making them more likely to stay airborne.

Reply to  Enginer01
February 18, 2021 8:23 pm

Life expectancy in India is 69.7 years. Sweden was 82.4. It is probably a bit less in Sweden now.

My son worked in Covid wards in Victoria at the peak of the second wave. Many 70+yo who presented were medicated to ease difficulties rather than treated in ICU. People over 70+yo with other issues were unlikely to make it out of ICU in a good way. His youngest death was 72yo although there were a few in their 50s who had long recoveries.

There is no doubt lockdown can be effective. Victoria was slow to lockdown for the second wave and suffered badly as a result. Now it is much more like Taiwan. As soon as it is in the community stop socialising. Done and dusted in a few days.

Rory Forbes
Reply to  RickWill
February 18, 2021 9:20 pm

This whole family of viruses (corona), of which there are 7 in humans, is far more virulent in Summer than colder months. Australia is in Summer.

Reply to  Rory Forbes
February 18, 2021 9:55 pm

Australia’s first wave was March/April. The second wave in Victoria was Jul to Sep. Season not a lot to do with it.

People proximity is the main factor. Population density, family size, house size, propensity for social gatherings and so on. Victorians learnt a lot about the issues or poor assimilation of new Australians when they immigrate in large numbers.

The curfew and travel restrictions were probably the most effective in limiting the rapid spread. The curfews enabled police to more easily find the large gatherings and suspicious gatherings like brothels and drug dealing. (Although there was no restrictions on going to the local drug clinic for a fix. That was one reason a non-essential worker could be more than 5km from home) Any vehicle out on the road after 9pm was fair game for police. There were more police cars than private cars on the roads after 9pm. Many streets were completely devoid of traffic for the entire night.

Rory Forbes
Reply to  RickWill
February 18, 2021 10:54 pm

In other words you imposed police state measures, overrunning UN recognized human rights and freedoms, almost to the point of Nazis and Soviet levels. Good for you. You quarantined everyone, including uninfected people. You employed the most idiotic use of the fallacious precautionary principle at a terrible cost to your economy and people’s mental and economic health.

Everyone (except China) had a “first wave” in March/April. All corona viruses are seasonal. Australia is still under lock-down, curfews and restrictions. Stop patting yourselves on the back. You accomplished nothing. Clearly Victoria is STILL in the 2nd wave. This entire plandemic has been a model of politically directed over reaction to an infection that barely rates a Severity Index level of 2 (0.1–0.5% mortality) on a global scale. It as reached a low 3 in some badly managed locations.

Reply to  Enginer01
February 19, 2021 6:23 pm

Autopsies, many well after the fact, established that most of the sickest, likely to enter ICU, and quickest to die suffered from the morbidity of poor (crappy) diet. Low zinc, very poor Vitamin D levels. Just like we have been saying since March. LISTEN, IDIOTS!

Abolition Man
Reply to  Joel O'Bryan
February 18, 2021 8:37 pm

Having all the government agencies and the Big Tech Nazis working with you to suppress dissent, and help accustom the populace to regular, required vaccinations; certainly makes it more likely that they can get away with their crimes and corruption! I’m just not sure how smart it is to assume that the ChiComs will go along with the program of reducing human population drastically!
Our ruling elite could well find themselves facing off against an army of tens of millions of Chinese who are fanatic in their support for their hegemony. Since China’s one-child policy has led to millions of men without women, I’m sure they would happily accept Western women as part of their tribute! Think of the treatment of Asian women by the Imperial Japanese Army expanded to an international scale! Perhaps Zhou Bai Den’s statement about a dark winter was his last semi-intelligent remark; now Kamala can lead the way to a socialist New World Order under the benevolent ministrations of our ChiCom overlords!

February 18, 2021 10:42 pm

My first thought on this report was to question the source (Sveriges Radio, not Nic) as I’m always a little suspicious when a news service agrees with something I’ve been thinking about for a long time.

In the case of COVID, I’ve been saying to family and friends since the very start, we should be following Sweden if we don’t want to ruin our economies.

I know people here may disagree with me but I’ve also said as a precautionary measure we should wear masks. A mask isn’t about protecting me from others, it’s about protecting others from a potentially deadly virus I could be carrying without my knowledge.

I checked Sveriges Radio to see if the newspaper was some right-wing nut job type of place.

It appears not. Even Wiki states “The public’s trust in the company, along with its Public Service counterparts in Sweden, may have decreased slightly during the 2000s. The decrease is most significant among right wing citizens.”

February 19, 2021 12:22 am

Thanks to Nic for all the time and expertise he puts into this.

One question:
“This second wave, which was very likely a seasonal effect, now appears past its peak”

What is the basis for that conclusion? Was this same genome, or a “new strain” as happened in UK and France. 

There seems to be an erroneous, default assumption that sars-cov-2 is seasonal “because flu”.

It has been shown in vitro that beta corona viruses including this one do NOT respond in the same way to humidity and temperature as does influenza.

Are we supposed to conclude that UK “second wave” +ve tests peaked on 29th Dec 2020 because that is when winter ended in UK ???

No, it was not the vaccine either which had bare even started deployment at that time let alone had time to produce an immune response. The second wave has already run its course. Vaccines *may* be helping if fall a little fast in the last week or two.

Reply to  Greg
February 19, 2021 1:35 am

Strain development effects

Feb13th Viral Reality Update: “A Wake-Up Call to the World” – YouTube

Suppression may be accelerating the evolution of the virus, especially the spike proteins. This could be a big problem considering the fact that these new vaccines only target the spike proteins.

Reply to  Greg
February 20, 2021 4:00 am

Greg, why do you believe that Winter ends on 29th December in the UK? We classify Winter starting on 21st December & Spring from 21st March. What we do have, is weather!

It’s an almost balmy 55°F in London & 48°F in Scotland, Which could all change by next week! Westerlies generally predominate, but we also get what are known as “lazy” Easterlies from the colder continent of Europe. Those winds from the east are lazy, because they can’t be bothered to go around people; instead they cut right through the body.

February 19, 2021 1:05 am

European all cause mortality by country

Graphs and maps — EUROMOMO

February 19, 2021 1:23 am

I live in the Stockholm area and I am happy that we still have most of our freedom. I work from home now and I go shopping as I usually did. We have some restrictions concerning the number of persons in a store at the same time. Screens are put up and markings on the floor for distances to be kept. We do not need to wear masks. Fact is that the mortality in 2012 (0,97%) was higher in Sweden than in 2020 (0,95%). So nobody can claim that our measures are not efficient enough or that too many people are dying. There is no scenario were nobody will die and as long as we are in the “normal” range compared to previous years, we do not need harder restrictions.

February 19, 2021 1:42 am

from the article:
“Only 0.9% of deaths were of people under 50 years old, and only 3% were of under 60 year olds. People over 70 years old accounted for over 91% of COVID-19 deaths”

That accounts for 94%. Does that mean that 6% of deaths were of people between 60 and 70 years old?

Paul C
Reply to  AndyHce
February 19, 2021 8:59 am

From the wording it could be either 2.1% or 3% 50-59, and 5.1% or 6% 60-69/70, but either way, the best way to have a long life expectancy is to be young – unfortunately, the one thing that is completely beyond our control. Percentages like this are skewed by populations within each category. In countries with low life expectancy, few die of COVID-19 because their young population survives this infection, but nevertheless have died before they reach old age.

Reply to  AndyHce
February 19, 2021 9:12 am

Yes, 6%

very old white guy
February 19, 2021 3:22 am

Tell me what are excess deaths and what are they in excess of? About 56 million people die every year from all causes, cov2 did not increase that, so just what are excess deaths?

Steve Keppel-Jones
Reply to  very old white guy
February 22, 2021 10:33 am

Canada had no measurable “excess” deaths last year. The USA appeared to show some (a few percent higher than normal), but it’s not clear whether that was because of Covid or other factors related to lockdowns, e.g. fear, depression, suicide, loss of income and consequent poverty, inability to access normal medical services, etc.

Paul in uk
February 19, 2021 4:53 am

I’ve been plotting my own graphs for various locations around UK and for Los Angeles Co US using my own simple equation to adjust raw cases based on number of tests and % positive. Am I the only person doing this, or does everyone know about this or is my method wrong? I’ve not seen any graphs like mine, but I find mine much more useful than the usual graphs of cases. I know it’s an approximation and it goes a little wrong when testing is very high and % positive very low. But it still to me seems a lot more useful than the raw data or random sampling. Using this method; better comparison with earlier waves, better calculation of cumulative cases (e.g. to look for herd immunity) takes out misleading trends, peaks and troughs as testing changes. I’m sure there’s loads of information in there that needs to be urgently found and analysed before the vaccine confuses the issue.

Using this method some UK locations are over 20% cumulative and I wonder if they may now have a kind of herd immunity. I’m following LA co US as well I can easily get the data I need in tabular form and it is close to some UK locations on cumulative cases but I believe lockdown was eased as I think the UK should have done since early February for locations that have high cumulative cases. So I’m interested to see how LA progresses. Although the vaccine is now confusing the issue.

I wanted to do same analysis for Sweden, Africa, India, various other US locations but I’m having trouble getting the data in easy to understand and use tabular form including testing and % positive. I think this could give useful information about the herd immunity question and impact of lockdown, new variants etc.

If the graph below posts ok: Grey line is raw cases, black adjusted so you can see my equation has adjusted the first wave up massively, but the shape matches surprisingly well with deaths (pink line) and hospitalisations (green). Deaths and hospitalisations I have offset the date (and used a different offset between 1st and current waves – was there a difference in delay between reporting?) and scaled to fit so don’t read off values against the Y axis. I think the deaths data is showing a ski jumping effect relative to cases as it goes over peaks, but not evident yet on descent from last peak as I presume some of the data has yet to be added.

Paul in uk
Reply to  Paul in uk
February 19, 2021 5:35 am

Same for Los Angeles county, US. I am still playing around with my adjustment equation, e.g. I can improve fit at the summit of the December LA peak, but that makes mortality between first and second waves on England graph look slightly out. You can see the adjustment has compensated for a drop in testing over Christmas.

Looking at the graph for whole world and quite a few other countries I think I see a similar dip in raw cases over Christmas, which perhaps is really a peak and if so, interestingly so many locations are showing the peak on the same few days at end of December. As I think I also saw on a headline today, the drop worldwide seems to have started before the vaccine could have a significant impact. My guess is some locations herd immunity may have kicked in, others lockdown working. 

Paul C
Reply to  Paul in uk
February 19, 2021 9:40 am

The big problem with your strategy of easing lockdown after a large number of cumulative cases is that it effectively rewards areas for poor performance, and condemns areas that have managed infection rates well to the permanent punishment of lockdown. While social distancing probably has a beneficial effect, and lockdowns and mask wearing probably do have an effect, we don’t even know the sign of that effect. Mask wearing could be comparable to cloud cover cooling daytime, but blanketing at nighttime preventing a frost. An infected person wearing a mask may well reduce the amount of virus expelled as spittle. However, that could be at the expense of retaining more virus within their own lungs, and having to breathe more deeply could force more virus-containing aerosols past the widely spaced fibres of the mask. Conversely, an uninfected person wearing a mask may suck more virus in, through having to breathe more deeply. Similarly, locking down one infected person within a household may increase the viral load to other members of that household, creating symptomatic cases which would have otherwise led to asymptomatic immunity. We still don’t know!

Paul in uk
Reply to  Paul C
February 19, 2021 4:42 pm

Thanks Paul C. I think I see what you mean, although I don’t mean to imply that the well behaved areas should necessarily remain locked down. Perhaps another way of looking at it is the places worst affected have been in lockdown a lot longer than other places and weathered a much more severe storm, so everyone in those areas has suffered regardless of their good behaviour or not. 

My feeling is worldwide this wave is possibly very much on the way out and fast even without vaccines, but my worry is a worse new wave due to mutations evading natural and vaccine immunity and perhaps the priority should be a system to test, track, trace (TTT) to prevent that wave, less focus on vaccines (except perhaps for the most vulnerable) because presumably we urgently need that TTT, vaccines or not.

If my analysis/graphs are right it suggests to me numbers I heard on the news today of 1 in 115 currently infected in England is several weeks out of date and we’re a lot lower than that now if I assume a case lasts 10 days. But this is just me, a non expert just spending a few minutes each day on it, no one to review or check, so I have low confidence in my conclusions and I mention my method here hoping people will confirm me right or wrong.

I don’t know how to calculate R, but just did a quick attempt at something approximating it with EXP(gradient of cases curve), the idea being I get 1 when gradient is zero, unfortunately when cases very low it tends to 1 again so presumably need to include cases somehow, (or how do you calculate/define R if you have almost no cases?) but even so I think it is showing something very interesting that for England R was above 1 in September, dropped to 1 about 5 days after lockdown (5th November, or was it dropping before lockdown anyway?), rose above 1 again early December, suddenly increased massively from 16th December to 25th. Then dropped immediately and has been low since 4th January.

For some reason I haven’t figured out yet excel is getting dates confused on this graph; not going to 2021, if I hover over lines it shows 21, but x axis displaying 20.

I wonder if maybe this kind of analysis applied to Sweden as well and compared with other places may show some very interesting things relating to lock downs, herd immunity R, etc.

February 19, 2021 7:28 am

Leopards have spots. Tigers have stripes. Epidemics show waves and hot spots over time and space…but the math is all the same..Laplacian Diffusion Equation…The pattern that shows depends on the values for rates & constants plugged into the equation…..Masks & social distancing etc have probably negligible influence on those values

February 19, 2021 8:27 am

This site is becoming unreadable with ads plastered over content and popping up all over.

Reply to  D.Anderson
February 19, 2021 2:46 pm

Get some ad blocker add-ons for your browser! AdBlock (Plus, +, whatever) and Ghostery. They’re free.

February 19, 2021 9:24 am

Peking Flu, back in 1992 , killed far more in Sweden, than, Covid, today.

Paul C
February 19, 2021 9:56 am

In addition to the UK, the Zoe symptom study has also been done in the USA
and in Sweden
just some more data to look through.

February 19, 2021 12:37 pm

The biggest confounding factor is, that it is the way people behave that decides what the transmission of the disease is. Whatever regulations are imposed what really matters is how people actually behave.

My observation is that people in UK and throughout the EU and USA behaved much the same as they did in Sweden, some took all precautions, some took no precautions at all, and some could not take any precautions as they were too poor.

Speaking personally as a UK citizen who was careful, the first wave left my family and contacts untouched, not so the second wave, deaths and serious illness from Sars Cov 2 invaded my life, it is horrible.

Gerry, England
February 19, 2021 1:17 pm

Something else to throw into the discussion is the role played by hospitals in spreading the infection. The UK NHS has been very successful in this by making the basic error of not isolating Covid cases from the start. Nosocomial deaths could easily be a quarter or more of the UK death toll, which is a badly distorted figure anyway since it records any death within 28 days of testing positive as a Covid death.

In the first outbreak, having grown tired of waiting for the UK’s ‘world beating track & trace’ system to show any functionality, some areas set up their own track and trace systems. This is what should have been done from the start but we are saddled with an idiot for Prime Minister who thinks he should lead on everything and sees himself as some sort of Winston Churchill Mark 2. To their surprise – but not mine – they traced almost half of infections back to an NHS source.

Moving to the current outbreak, it has been said that 80% of all infections in the community can be traced to contact with at least one healthworker. Of course the NHS trusts, directors and PHE deny this but then they also denied anything was wrong in a Shropshire hospital when the high number of infant deaths was questioned….until the police investigation started. I have read that one of my county hospitals rapidly built a Covid unit to separate the cases but I think this is rare and they still fail to understand the need to isolate Covid as evidenced by the failure to use the ‘Nightingale’ temporary hospitals.

February 19, 2021 4:45 pm

Here is what the Canadian government is trying to do to us. Talk about a hockey stick! It’s a wonder anyone believes anything they say, but many do.

Saved image from Tweetium (31).jpg
Verified by MonsterInsights