Friday, 29 March, 2024
HomeTalking PointsSweden's COVID-19 experiment has divided the world

Sweden's COVID-19 experiment has divided the world

The world is divided about the Swedish “light touch”  approach to COVID-19, seeing it either a a triumph of good sense or disastrous misstep, writes Mallory Pickett for the New Yorker.

While lockdowns, curfews, and travel bans were being rolled out across the rest of the world, Swedish restaurants, stores, bars, museums, day cares, and elementary schools all remained open.  People were encouraged to work from home and to reduce travel, but both were optional. Masks were not recommended and remained rare. Households could mix; until the end of March, even parties of up to five hundred people were allowed.

The man behind Swedenʼs coronavirus response is Anders Tegnell, the countryʼs head epidemiologist. Tegnell worked in Zaire during the 1995 Ebola epidemic, and then served as an expert on infectious diseases for the European Union before being hired by the Swedish public-health agency, in 2013. The Swedish constitution gives government agencies extraordinary independence, so Tegnell and the public-health agency have led much of the coronavirus response, and, constitutionally, the government has little power to impose restrictions.

Tegnell has often said that lockdowns are not supported by science and that the evidence for mask-wearing is “weak”. His stance is a startling departure from the scientific consensus, but he maintains that if other countries were led by experts rather than politicians, more nations would have policies like Swedenʼs.

Within the country, Tegnell has become an icon of Swedish exceptionalism, believed to be excessively reasonable, level-headed, and rational. Supporters praised him for not giving in to political panic. Wearing a mask in Sweden was sometimes seen as a signifier of being anti-science.

Tegnell introduced the countryʼs light-touch approach to the pandemic in March, as cases were mounting. Internal e-mails obtained by the freelance journalist Emanuel Karlsten show that the government at least considered whether the strategy might lead to more widespread immunity.

In an e-mail exchange between Tegnell and the head of the Finnish public-health agency, on 14th and 15th March 2020, Tegnell suggested that keeping the schools open could help the young and healthy develop immunity sooner. His Finnish colleagues noted that their models found that closing schools would decrease the infection rate among the elderly by 10%. Tegnell responded, “10% might be worth it?” However, Sweden seems to have settled on its final strategy primarily through a mix of unorthodox scientific conclusions, pragmatism, and folkvett, a particularly Swedish notion of common sense.

The report says no epidemiologists thought that self-isolation should be a long-term solution. “Lockdown is a temporary measure that you do to buy you time,” David Nabarro, a long-time UN adviser and one of the World Health Organisationʼs special envoys on COVID-19, told me. But most countries were unable to get the virus under control and decided to leave lockdowns in place until a vaccine became available.

This seemed unsustainable to Tegnell. He also shirked masks. A year ago, there were few studies that directly measured the impact of masks on viral-particle emissions and community spread of the virus, and controlled experiments were difficult to conduct, for logistical and ethical reasons.

Over the course of the pandemic, evidence supporting masks has piled up, the report says. An analysis of mask mandates in German towns found that they may have reduced COVID-19 infections by about 45%. Another study examined US states with mask mandates and found that the daily growth of infections was 2% lower in states where a mask mandate had been in place for a few weeks.

By April, 2020, the WHO, the US Centres for Disease Control and Prevention, and other groups decided that there was enough evidence, and recommended face coverings. The CDC. now firmly concludes that “experimental and epidemiological data support community masking to reduce the spread of SARS-CoV-2.”

Yet Tegnell remained unsatisfied. In April, 2020, he wrote a letter to the European Centre for Disease Control urging against a mask recommendation, saying, “The argument for and evidence for an effect of face covering to limit the spread from asymptomatic persons is not clear… The arguments against are at least as convincing.”

As the Swedish strategy garnered attention, Johan Giesecke, Swedenʼs former head epidemiologist, told UnHerd, a British YouTube channel, that he expected the infection fatality rate to be “much, much lower” than was being reported in the media at the time. “I think it will be like a severe influenza rate, on the order of 0.1%.” (A study by the Swedish public health-agency later found that the rate was at least six times higher in Stockholm.)

He said that the difference between Sweden and other countriesʼ policies was that “the Swedish government decided early, in January, that the measures we should take against the pandemic should be evidence-based. And when you start looking around at the measures that are being taken now by other countries, you find that very few of them have a shred of evidence.”

Tegnell is quoted in the New Yorker as saying that he had been in close contact with his counterparts in the UK, who were planning similarly light restrictions. But cases in the UK were increasing rapidly. On 16 March, scientists at Imperial College London published a paper, based on an epidemiological model, predicting that, unless some form of lockdown was imposed, more than five hundred thousand Brits would die from preventable COVID-19 infections.

A week later, the UK Prime Minister Boris Johnson, announced that his government would be closing schools, bars, and restaurants, falling in step with the rest of Europe. “It was slightly frustrating,” Tegnell is quoted in the report as saying last August. “We were really hoping we could take us through this crisis together.”

Sweden is a country of 10m people, about the size of Los Angeles County, and almost 20% of Swedes live in Stockholm, the largest city. In April, 2020, a widely-circulated pre-print version of a paper by researchers at Uppsala University, adapting the Imperial College model, predicted that, under the Swedish strategy, 50% of susceptible Swedes would be infected within thirty days, resulting in over 80,000 deaths by July.

That spring, the virus began spreading unchecked in Sweden. “It just kept adding up,” Tegnell said. “I mean, youʼre always kind of hopeful and think that, OK, this is something thatʼs going to pass over.” Soon, the per-capita death toll was among the highest in Europe.

Slight modifications were made to the Swedish advice: visitors to nursing homes were banned on 30 March; people were not allowed to gather in groups larger than fifty.

The report quotes Tegnell as saying that the death toll weighed on him. “I think this was a big frustration and feeling of failure for us,” he said. But he remained steadfast, often saying, in interviews, “Judge me in a year.”

In the summer of 2020, case counts began to drop in Sweden. And speaking at the end of August, Tegnell felt that the worst was behind him. Just a few weeks later, he and other international experts had a special audience with Boris Johnson, to give advice on how Sweden had brought numbers down with their light-touch approach. “I think we are reasonably optimistic,” he said. “Our prognosis is, No, we donʼt really see a huge second wave coming on.”

However, the New Yorker reports, this did not last. By December, cases and hospitalisations were higher than theyʼd been since the earliest days of the pandemic. Intensive-care units in Stockholm and Malmö, the countryʼs third biggest city, were full.

“It was just this development we did not want to see,” Björn Eriksson, Stockholmʼs director of health and medical care, said. Confidence in the public-health agency had fallen from 68% in October, to 52% in December. The government appointed an independent commission to investigate the Swedish response to the pandemic, and passed a new law temporarily allowing the state to close businesses as needed.

Tegnellʼs prediction of a tapering epidemic curve and quickly-attained immunity never came to pass. Swedenʼs per-capita case counts and death rates have been many times higher than any of its Nordic neighbours, all of which imposed lockdowns, travel bans, and limited gatherings early on.

Over-all in Sweden, 13,000 people have died from COVID-19. In Norway, which has a population that is half the size of Swedenʼs, and where stricter lockdowns were enforced, about 700 people have died.

The New Yorker reports says itʼs likely that some simple policy changes – especially shutting down visitations to nursing homes sooner and providing more PPE and testing to nursing-home staff – would have saved lives. And, the report says, the strategy doesnʼt seem to have helped the economy much: the Swedish GDP fell by around 3%, better than the European average, but similar to the drop in other Nordic countries.

Fredrik Elgh, a virologist at Umeå University and one of Tegnellʼs former bosses, wishes that Sweden had implemented restrictions like those used by other countries in the region. “Why donʼt they go the same route as our neighbours that have been so successful?” he said. “We could have done that, too, if we had followed their path.”

The New Yorker reports that the fatalities in the elder homes, which account for about 50% of the COVID-19 deaths in Sweden, seem especially needless; if visits to these facilities had been banned sooner, if their workers had been advised to wear masks and get tested frequently, itʼs possible that thousands of lives could have been saved. People from both pro- and anti-Tegnell camps believe that this was an unnecessary tragedy. (Tegnell agrees, although he does not think that the responsibility lies solely with the health agency.)

In December, Swedenʼs King, Carl XVI Gustaf, criticised the countryʼs policies, saying, “The Swedish people have suffered enormously in difficult conditions,” and adding that, when it came to the strategy that the country employed, “I think we have failed.”

But, although the outcome in Sweden was a black spot in Scandinavia, “compared to other countries in Europe, itʼs not the worst off,” Patrick Heuveline, a professor of sociology at the University of California-Los Angeles who studies pandemic mortality rates, is quoted in the New Yorker as saying. “Itʼs not as bad as Italy, Spain, the UK, and Belgium for example.”

Tegnell holds up this statistic when defending his strategy, claiming that sparsely-populated Norway and Finland are the outliers, and that Sweden should be compared to the rest of Europe. Sweden has a larger foreign-born population than other Nordic countries, and its population is more concentrated in urban areas, Tegnell claims.

Other experts are sceptical of this argument. “I find no correlation between proportion of foreign-born and COVID death rate,” Heuveline wrote. “Norway has a higher proportion of foreign-born than Denmark, which has about the same proportion as Italy (about 10%), but COVID-19 mortality is much higher in Italy than in Denmark, and higher in Denmark than in Norway.”

Over-all, Elgh, Heuveline and other experts argue, Swedenʼs population is more similar to the other Nordic countries. Its first infections also came later than in other parts of Europe, giving its government more time to warn its citizens of the virusʼ severity. The report says for these reasons, comparisons to the rest of Scandinavia, which are less favourable to Sweden, may be more apt.

Even so, the New Yorker reports that Swedenʼs death toll was not as high as some predicted. But most high schools and universities in Sweden went online. Staying at home was optional rather than mandatory, but mobility data from cell phones show that Swedes did significantly reduce their movement. Lone Simonsen, an epidemiologist at Roskilde University, in Denmark, believes that two of Swedenʼs simplest restrictions – limiting the sizes of gatherings and regulating visits to nursing homes – did much of the work to limit spread.

“Most countries do a bunch of things at the same time, and you really canʼt find the effect of the individual ones,” she said. She also points out that Swedenʼs restrictions, while relatively relaxed, also remained fairly steady. “They didnʼt fluctuate like we did in other countries,” Simonsen said. “Weʼve been on a yo-yo diet with COVID over here. The numbers went down, we opened up, and then they went up.” There may also be factors that weʼre not yet aware of.

“Weʼve just got to be humble about what we know and what we donʼt know. And we still donʼt know a lot,” Howard Forman, a professor of public-health and management at Yale University, is quoted in the report as saying. Protections that seemed important may turn out, after long-term study, to have been less effective than we thought. “If you, one day, come to me and say that masks only reduce spread by 15% or even 10%, Iʼd be like, OK, well, thatʼs within range. Itʼs not a total shock. And, by the way, I would still say it was worth wearing the masks.”

The New Yorker reports that Sweden remains divided on its governmentʼs response to the pandemic. On 18 December, as hospitals were already braced for a post-Christmas surge, Tegnell and the public-health agency finally recommended wearing masks – but only on public transit and only during rush hour. (Tegnell sees distancing as the most important barrier to the virus – “Masking is not a golden bullet,” he said – and only recommends the use of masks where distance cannot be maintained.)

The report says days after this new rule was passed, Tegnell was asked whether he still thought the evidence for masks was “sketchy.” “Yes,” he responded. “Unfortunately, there is not much new evidence in place.” But he said that he had decided that it was better not to take chances. “Due to the developments we see, we need to use even measures where evidence and effect is low,” he said.

 

Full New Yorker report (Open access)

MedicalBrief — our free weekly e-newsletter

We'd appreciate as much information as possible, however only an email address is required.