A flurry of criticism prompted Jonas Ludvigsson, of the Karolinska Institute, to end the COVID-19 research.
Alexander Donka
By Gretchen Vogel
Science‘s COVID-19 reports are supported by the Heising-Simons Foundation.
Pediatrician and epidemiologist Jonas Ludvigsson, of Sweden’s Karolinska Institute, has been a staunch supporter of his country’s unorthodox policies against the coronavirus. Among them was the decision in the spring of 2020 to keep preschools and schools open to children until the ninth grade, despite limited understanding of the virus and with little care to prevent school outbreaks. But Ludvigsson’s research, which suggested that the policy was relatively safe – and has been widely cited in arguments against closing schools – has been repeatedly criticized for the Swedish approach.
The most recent example is a research letter, published online by The new English medical journal (NEJM) on January 6, which analyzed serious illnesses and deaths among children and teachers in Sweden between March and June 2020. Critics – including the authors of two letters NEJM published on March 1 – said that the study is beside the point and is a distraction. It is well known that children are less likely to be hospitalized or die of COVID-19; instead, schools around the world have closed to slow the spread of the virus in the wider community.
Related
But Science I heard that another complaint sent to NEJM makes a more serious claim: that the authors have deliberately omitted important data that contradicted their conclusion.
The complaint comes from Bodil Malmberg, a private citizen from Vårgårda, Sweden. She used the country’s open registration law to get email correspondence between Ludvigsson and Swedish chief epidemiologist Anders Tegnell, the country’s pandemic policy architect, who clarified how the article came about. Malmberg says he requested the e-mails because the data on the NEJM role “made no sense.” Ludvigsson does not dispute the content of the emails, but defends the study’s conclusions. However, he says the flood of criticism and personal attacks made him decide to abandon the COVID-19 survey.
Ludvigsson, whose pre-pandemic research focused on gastroenterology, was one of the 47 original signatories to the Great Barrington Declaration, a controversial document published in October 2020 that argued that pandemic policies should focus on protecting the vulnerable while the rest of population builds immunity through natural infection.
Ludvigsson’s research seemed to support these ideas. In a review of the role of children in the pandemic, published in Acta Paediatrica in May 2020, he reported that there was “no major school outbreak in Sweden”, which he attributed to Tegnell’s “personal communication”. But, as critics noted, the Swedish media reported several outbreaks in schools so far, including one in which at least 18 of the 76 employees were infected and a teacher died. (The children were not tested.)
His NEJM The letter sounded another reassuring note. He reported that, in all of Sweden, only 15 children, 10 preschool teachers and 20 school teachers were admitted to intensive care units for COVID-19 complications between March and June 2020. The authors noted that 69 children with ages 1 to 16 died of any cause in Sweden during the same period, compared with 65 between november 2019 and february 2020, suggesting that the pandemic did not lead to an increase in infant deaths.
But the emails obtained by Malmberg show that, in July 2020, Ludvigsson wrote to Tegnell that “unfortunately, we see a clear indication of excess mortality among children aged 7 to 16, the ages when ‘children went to school ‘”. For the years 2015 to 2019, an average of 30.4 children in this age group died in the four months of spring; in 2020, 51 children in this age group died, “= excessive mortality + 68%”, wrote Ludvigsson. The increase may be a fluke, he wrote, especially since the numbers are small. Deaths in children aged 1 to 6 years were below average during the same period, so the combination of age groups helped to balance the increase, he noted.
The emails “cast a serious shadow” in the research letter, Malmberg wrote in an email to NEJM. (The newspaper declined to comment on his complaint.) Epidemiologist Jonas Björk of Lund University agrees that the time comparison used in the newspaper was unusual. “I don’t see any good reason to compare it with the previous months,” he says. “It is standard to compare with the same period in previous years” to account for seasonality and reduce statistical uncertainty.
Ludvigsson said Science a peer reviewer for NEJM suggested comparing the deaths from November to February, and that he combined the numbers of children of preschool and school age because of NEJMlength requirements. As part of his March 1 response to published criticisms, he updated the newspaper’s supplementary data with monthly deaths from 2015 to 2020, but did not signal a 68% increase in school-age children.
The increase in mortality is unlikely to be due to COVID-19 alone, notes Björk. Although Sweden tested very few children in the first months of the pandemic, fatal cases of COVID-19 probably would have appeared somewhere in the health care system. “Of course, this should be ruled out by examining the causes of death and medical records in more detail,” he says. And Björk agrees that the increase may be due to chance.
Mortality issues aside, critics say that NEJM The letter was simply not very helpful in the heated debate over the closure of schools. The main concern is not that children end up in intensive care, but that schools can accelerate the spread of the community, says Antoine Flahault, a specialist in global health at the University of Geneva and co-author of one of the criticisms in NEJM. Nisha Thampi, a pediatric infectious physician at the CHEO pediatric health and research center in Ottawa, Canada, agrees. The newspaper “doesn’t really tell me much about what is happening to children in schools,” she says.
Discussing whether schools are “safe” has lessened efforts to find ways to reduce the risk of viral spread in classrooms and corridors, says Flahault – which is key to keeping schools open and the virus under control.