The Swedish COVID-19 experiment of not implementing early and strong measures to safeguard the population has been hotly debated around the world, but at this point we can predict it is almost certain to result in a net failure in terms of death and suffering. As of Oct. 13, Sweden’s per capita death rate is 58.4 per 100,000 people, according to Johns Hopkins University data, 12th highest in the world (not including tiny Andorra and San Marino). But perhaps more striking are the findings of a study published Oct. 12 in the Journal of the American Medical Association, which pointed out that, of the countries the researchers investigated, Sweden and the U.S. essentially make up a category of two: they are the only countries with high overall mortality rates that failed to rapidly reduce those numbers as the pandemic progressed.
Yet the architects of the Swedish plan are selling it as a success to the rest of the world. And officials in other countries, including at the top level of the U.S. government, are discussing the strategy as one to emulate—despite the reality that doing so will almost certainly increase the rates of death and misery.
Countries that locked down early and/or used extensive test and tracing—including Denmark, Finland, Norway, South Korea, Japan, Taiwan, Vietnam and New Zealand—saved lives and limited damage to their economies. Countries that locked down late, came out of lock down too early, did not effectively test and quarantine, or only used a partial lockdown—including Brazil, Mexico, Netherlands, Peru, Spain, Sweden, the U.S. and the U.K.—have almost uniformly done worse in rates of infection and death.
Despite this, Sweden’s Public Health Agency director Johan Carlson has claimed that “the Swedish situation remains favorable,” and that the country’s response has been “consistent and sustainable.” The data, however, show that the case rate in Sweden, as elsewhere in Europe, is currently increasing.
Average daily cases rose 173% nationwide from Sept. 2-8 to Sept. 30-Oct. 6 and in Stockholm that number increased 405% for the same period. Though some have argued that rising case numbers can be attributed to increased testing, a recent study of Stockholm’s wastewater published Oct. 5 by the Swedish Royal Institute of Technology (KTH) argues otherwise. An increased concentration of the virus in wastewater, the KTH researchers write, shows a rise of the virus in the population of the greater Stockholm area (where a large proportion of the country’s population live) in a way that is entirely independent of testing. Yet even with this rise in cases, the government is easing the few restrictions it had in place.
From early on, the Swedish government seemed to treat it as a foregone conclusion that many people would die. The country’s Prime Minister Stefan Löfven told the Swedish newspaper Dagens Nyheter on April 3, “We will have to count the dead in thousands. It is just as well that we prepare for it.” In July, as the death count reached 5,500, Löfven said that the “strategy is right, I am completely convinced of that.” In September, Dr. Anders Tegnell, the Public Health Agency epidemiologist in charge of the country’s COVID-19 response reiterated the party line that a growing death count did “not mean that the strategy itself has gone wrong.” There has been a lack of written communication between the Prime Minister and the Public Health Authority: when the authors requested all emails and documents between the Prime Minister’s office and the Public Health Authority for the period Jan. 1—Sept. 14, the Prime Minister’s Registrar replied on Sept. 17 that none existed.
Despite the Public Health Agency’s insistence to the contrary, the core of this strategy is widely understood to have been about building natural “herd immunity”—essentially, letting enough members of a population (the herd) get infected, recover, and then develop an immune system response to the virus that it would ultimately stop spreading. Both the agency and Prime Minister Löfven have characterized the approach as “common sense“ trust-based recommendations rather than strict measures, such as lockdowns, which they say are unsustainable over an extended period of time—and that herd immunity was just a desirable side effect. However, internal government communications suggest otherwise.
Emails obtained by one of the authors through Freedom of Information laws (called offentlighetsprincipen, or “Openness Principle,” in Swedish) between national and regional government agencies, including the Swedish Public Health Authority, as well as those obtained by other journalists, suggest that the goal was all along in fact to develop herd immunity. We have also received information through sources who made similar requests or who corresponded directly with government agencies that back up this conclusion. For the sake of transparency, we created a website where we’ve posted some of these documents.
One example showing clearly that government officials had been thinking about herd immunity from early on is a March 15 email sent from a retired doctor to Tegnell, the epidemiologist and architect of the Swedish plan, which he forwarded to his Finnish counterpart, Mika Salminen. In it, the retired doctor recommended allowing healthy people to be infected in controlled settings as a way to fight the epidemic. “One point would be to keep schools open to reach herd immunity faster,” Tegnell noted at the top of the forwarded email.
Salminen responded that the Finnish Health Agency had considered this but decided against it, because “over time, the children are still going to spread the infection to other age groups.” Furthermore, the Finnish model showed that closing schools would reduce “the attack rate of the disease on the elderly” by 10%. Tegnell responded: “10 percent might be worth it?”
The majority of the rest of Sweden’s policymakers seemed to have agreed: the country never closed daycare or schools for children under the age of 16, and school attendance is mandatory under Swedish law, with no option for distance learning or home schooling, even for family members in high risk groups. Policymakers essentially decided to use children and schools as participants in an experiment to see if herd immunity to a deadly disease could be reached. Multiple outbreaks at schools occurred in both the spring and autumn.
At this point, whether herd immunity was the “goal” or a “byproduct” of the Swedish plan is semantics, because it simply hasn’t worked. In early May, the Public Health Agency predicted that 40% of the Stockholm population would have the disease and acquire protective antibodies by the end of the month. According to the agency’s own antibody studies published Sept. 3 for samples collected up until late June, the actual figure for random testing of antibodies is only 11.4% for Stockholm, 6.3% for Gothenburg and 7.1% across Sweden. As of mid-August, herd immunity was still “nowhere in sight,” according to a Journal of the Royal Society of Medicine study. That shouldn’t have been a surprise. After all, herd immunity to an infectious disease has never been achieved without a vaccine.
Löfven, his government, and the Public Health Agency all say that the high COVID-19 death rate in Sweden can be attributed to the fact that a large portion of these deaths occurred in nursing homes, due to shortcomings in elderly care.
However, the high infection rate across the country was the underlying factor that led to a high number of those becoming infected in care homes. Many sick elderly were not seen by a doctor because the country’s hospitals were implementing a triage system that, according to a study published July 1 in the journal Clinical Infectious Diseases, appeared to have factored in age and predicted prognosis. “This likely reduced [intensive care unit] load at the cost of more high-risk patients”—like elderly people with confirmed infection—dying outside the ICU.” Only 13% of the elderly residents who died with COVID-19 during the spring received hospital care, according to preliminary statistics from the National Board of Health and Welfare released Aug.
In one case which seems representative of how seniors were treated, patient Reza Sedghi was not seen by a doctor the day he died from COVID-19 at a care home in Stockholm. A nurse told Sedghi’s daughter Lili Perspolisi that her father was given a shot of morphine before he passed away, that no oxygen was administered and staff did not call an ambulance. “No one was there and he died alone,” Perspolisi says.
In order to be admitted for hospital care, patients needed to have breathing problems and even then, many were reportedly denied care. Regional healthcare managers in each of Sweden’s 21 regions, who are responsible for care at hospitals as well as implementing Public Health Agency guidelines, have claimed that no patients were denied care during the pandemic. But internal local government documents from April from some of Sweden’s regions—including those covering the biggest cities of Stockholm, Gothenburg and Malmö—also show directives for how some patients including those receiving home care, those living at nursing homes and assisted living facilities, and those with special needs could not receive oxygen or hospitalization in some situations. Dagens Nyheter published an investigation on Oct. 13 showing that patients in Stockholm were denied care as a result of these guidelines. Further, a September investigation by Sveriges Radio, Sweden’s national public broadcaster, found that more than 100 people reported to the Swedish Health and Care Inspectorate that their relatives with COVID-19 either did not receive oxygen or nutrient drops or that they were not allowed to come to hospital.
These issues do not only affect the elderly or those who had COVID-19. The National Board of Health and Welfare’s guidelines for intensive care in extraordinary circumstances throughout Sweden state that priority should be given to patients based on biological, not chronological, age. Sörmlands Media, in an investigation published May 13, cited a number of sources saying that, in many parts of the country, the health care system was already operating in a way such that people were being denied the type of inpatient care they would have received in normal times. Regional health agencies were using a Clinical Frailty Scale, an assessment tool designed to predict the need for care in a nursing home or hospital, and the life expectancy of older people by estimating their fragility, to determine whether someone should receive hospital care and was applied to decisions regarding all sorts of treatment, not only for COVID-19. These guidelines led to many people with health care needs unrelated to COVID-19 not getting the care they need, with some even dying as a result—collateral damage of Sweden’s COVID-19 strategy.
Dr. Michael Broomé, the chief physician at Stockholm’s Karolinska Hospital’s Intensive Care Unit, says his department’s patient load tripled during the spring. His staff, he says, “have often felt powerless and inadequate. We have lost several young, previously healthy, patients with particularly serious disease courses. We have also repeatedly been forced to say no to patients we would normally have accepted due to a lack of experienced staff, suitable facilities and equipment.”
In June, Dagens Nyheter reported a story of one case showing how disastrous such a scenario can be. Yanina Lucero had been ill for several weeks in March with severe breathing problems, fever and diarrhea, yet COVID-19 tests were not available at the time except for those returning from high risk areas who displayed symptoms, those admitted to the hospital, and those working in health care. Yanina was only 39 years old and had no underlying illnesses. Her husband Cristian brought her to an unnamed hospital in Stockholm, but were told it was full and sent home, where Lucero’s health deteriorated. After several days when she could barely walk, an ambulance arrived and Lucero was taken to Huddinge hospital, where she was sedated and put on a ventilator. She died on April 15 without receiving a COVID-19 test in hospital.
Sweden did try some things to protect citizens from the pandemic. On March 12 the government restricted public gatherings to 500 people and the next day the Public Health Agency issued a press release telling people with possible COVID-19 symptoms to stay home. On March 17, the Public Health Agency asked employers in the Stockholm area to let employees work from home if they could. The government further limited public gatherings to 50 people on March 29. Yet there were no recommendations on private events and the 50-person limit doesn’t apply to schools, libraries, corporate events, swimming pools, shopping malls or many other situations. Starting April 1, the government restricted visits to retirement homes (which reopened to visitors on Oct. 1 without masks recommended for visitors or staff). But all these recommendations came later than in the other Nordic countries. In the interim, institutions were forced to make their own decisions; some high schools and universities changed to on-line teaching and restaurants and bars went to table seating with distance, and some companies instituted rules about wearing masks on site and encouraging employees to work from home.
Meanwhile Sweden built neither the testing nor the contact-tracing capacity that other wealthy European countries did. Until the end of May (and again in August), Sweden tested 20% the number of people per capita compared with Denmark, and less than both Norway and Finland; Sweden has often had among the lowest test rates in Europe. Even with increased testing in the fall, Sweden still only tests only about one-fourth that of Denmark.
Sweden never quarantined those arriving from high-risk areas abroad nor did it close most businesses, including restaurants and bars. Family members of those who test positive for COVID-19 must attend school in person, unlike in many other countries where if one person in a household tests positive the entire family quarantines, usually for 14 days. Employees must also report to work as usual unless they also have symptoms of COVID-19, an agreement with their employer for a leave of absence or a doctor recommends that they isolate at home.
On Oct. 1, the Public Health Authority issued non-binding “rules of conduct” that open the possibility for doctors to be able to recommend that certain individuals stay home for seven days if a household member tests positive for COVID-19. But there are major holes in these rules: they do not apply to children (of all ages, from birth to age 16, the year one starts high school), people in the household who previously have a positive PCR or antibody test or, people with socially important professions, such as health care staff (under certain circumstances).
There is also no date for when the rule would go into effect. “It may not happen right away, Stockholm will start quickly but some regions may need more time to get it all in place,” Tegnell said at a Oct. 1 press conference. Meanwhile, according to current Public Health Agency guidelines issued May 15 and still in place, those who test positive for COVID-19 are expected to attend work and school with mild symptoms so long as they are seven days post-onset of symptoms and fever free for 48 hours.
Sweden actually recommends against masks everywhere except in places where health care workers are treating COVID-19 patients (some regions expand that to health care workers treating suspected patients as well). Autumn corona outbreaks in Dalarna, Jönköping, Luleå, Malmö, Stockholm and Uppsala hospitals are affecting both hospital staff and patients. In an email on April 5, Tegnell wrote to Mike Catchpole, the chief scientist at the European Center for Disease Control and Prevention (ECDC): “We are quite worried about the statement ECDC has been preparing about masks.” Tegnell attached a document in which he expresses concern that ECDC recommending facemasks would “imply that the spread is airborne which would seriously harm further communication and trust among the population and health care workers” and concludes “we would like to warn against the publication of this advice.” Despite this, on April 8 ECDC recommended masks and on June 8 the World Health Organization updated its stance to recommend masks.
Sweden’s government officials stuck to their party line. Karin Tegmark Wisell of the Public Health Agency said at a press conference on July 14 that “we see around the world that masks are used in a way so that you rather increase the spread of infection.” Two weeks later, Lena Hallengren, the Minister of Health and Social Affairs, spoke about masks at a press conference on July 29 and said, “We don’t have that tradition or culture” and that the government “would not review the Public Health Agency’s decision not to recommend masks.”
All of this creates a situation which leaves teachers, bus drivers, medical workers and care home staff more exposed, without face masks at a time when the rest of the world is clearly endorsing widespread mask wearing.
On Aug. 13, Tegnell said that to recommend masks to the public “quite a lot of resources are required. There is quite a lot of money that would be spent if you are going to have masks.” Indeed, emails between Tegnell and colleagues at the Public Health Agency and Andreas Johansson of the Ministry of Health and Social Affairs show that the policy concerns of the health authority were influenced by financial interests, including the commercial concerns of Sweden’s airports.
Swedavia, the owner of the country’s largest airport, Stockholm Arlanda, told employees during the spring and early summer they could not wear masks or gloves to work. One employee told Upsala Nya Tidning newspaper on Aug. 24 “Many of us were sick during the beginning of the pandemic and two colleagues have died due to the virus. I would estimate that 60%-80% of the staff at the security checks have had the infection.”
“Our union representatives fought for us to have masks at work,” the employee said, “but the airport’s response was that we were an authority that would not spread fear, but we would show that the virus was not so dangerous.” Swedavia’s reply was that they had introduced the infection control measures recommended by the authorities. On July 1, the company changed its policy, recommending masks for everyone who comes to Arlanda—that, according to a Swedavia spokesperson, was not as a result of “an infection control measure advocated by Swedish authorities,” but rather, due to a joint European Union Aviation Safety Agency and ECDC recommendation for all of Europe.
As early as January, the Public Health Agency was warning the government about costs. In a Jan. 31 communique, Public Health Agency Director Johan Carlsson (appointed by Löfven) and General Counsel Bitte Bråstad wrote to the Ministry of Health and Social Affairs, cautioning the government about costs associated with classifying COVID-19 as a socially dangerous disease: “After a decision on quarantine, costs for it [include] compensation which according to the Act, must be paid to those who, due to the quarantine decision, must refrain from gainful employment. The uncertainty factors are many even when calculating these costs. Society can also suffer a loss of production due to being quarantined [and] prevented from performing gainful employment which they would otherwise have performed.” Sweden never implemented quarantine in society, not even for those returning from travel abroad or family members of those who test positive for COVID-19.
Not only did these lack of measures likely result in more infections and deaths, but it didn’t even help the economy: Sweden has fared worse economically than other Nordic countries throughout the pandemic.
The Swedish way has yielded little but death and misery. And, this situation has not been honestly portrayed to the Swedish people or to the rest of the world.
A Public Health Agency report published July 7 included data for teachers in primary schools working on-site as well as for secondary school teachers who switched to distance instruction online. In the report, they combined the two data sources and compared the result to the general population, stating that teachers were not at greater risk and implying that schools were safe. But in fact, the infection rate of those teaching in classrooms was 60% higher than those teaching online—completely undermining the conclusion of the report.
The report also compares Sweden to Finland for March through the end of May and wrongly concludes that the ”closing of schools had no measurable effect on the number of cases of COVID-19 among children.” As testing among children in Sweden was almost non-existent at that time compared to Finland, these data were misrepresented; a better way to look at it would be to consider the fact that Sweden had seven times as many children per capita treated in the ICU during that time period.
When pressed about discrepancies in the report, Public Health Agency epidemiologist Jerker Jonsson replied on Aug. 21 via email: “The title is a bit misleading. It is not a direct comparison of the situation in Finland to the situation in Sweden. This is just a report and not a peer-reviewed scientific study. This was just a quick situation report and nothing more.” However the Public Health Agency and Minister of Education continue to reference this report as justification to keep schools open, and other countries cite it as an example.
This is not the only case where Swedish officials have misrepresented data in an effort to make the situation seem more under control than it really is. In April, a group of 22 scientists and physicians criticized Sweden’s government for the 105 deaths per day the country was seeing at the time, and Tegnell and the Public Health Agency responded by saying the true number was just 60 deaths per day. Revised government figures now show Tegnell was incorrect and the critics were right. The Public Health Agency says the discrepancy was due to a backlog in accounting for deaths, but they have backlogged deaths throughout the pandemic, making it difficult to track and gauge the actual death toll in real time.
Sweden never went into an official lockdown but an estimated 1.5 million have self-isolated, largely the elderly and those in risk groups. This was probably the largest factor in slowing the spread of the virus in the country in the summer. However, recent data suggest that cases are yet again spiking in the country, and there’s no indication that government policies will adapt.
Health care workers, scientists and private citizens have all voiced concerns about the Swedish approach. But Sweden is a small country, proud of its humanitarian image—so much so that we cannot seem to understand when we have violated it. There is simply no way to justify the magnitude of lost lives, poorer health and putting risk groups into long-term isolation, especially not in an effort to reach an unachievable herd immunity. Countries need to take care before adopting the “Swedish way.” It could have tragic consequences for this pandemic or the next.
Correction, Feb. 5
The original version of this story misstated the date on which the Swedish Public Health Agency estimated that 40% of the Stockholm population would acquire protective antibodies by the end of May. It was early May, not April.