How The Mitigation Strategy Tempted Northern Europe And Convinced Only Sweden
In mid-March of this year, while Southern Europe was experiencing a fast spread of COVID-19, Northern Europe was witnessing the spread of the herd immunity idea, since then irredeemably associated with the COVID-19 discussion.
Herd immunity means that, when the majority of the population becomes infected with a virus, those who do not die from the disease become immune to it. As a consequence, the effect of subsequent outbreaks should be significantly reduced due to the absence of a sufficient number of viable hosts for the virus to infect. Although herd immunity used to be a term generally applied to the immunity obtained with population-wide vaccination campaigns, Britain’s chief science advisor, Sir Patrick Vallance, believed that in the case of the COVID-19 pandemic, it could be achieved via natural infection.
The main rational behind Vallance’s reasoning was the natural course of the century-old Spanish flu. In mid-1918, the second wave of the pandemic hit the world with a virulence that was superior to that of the first wave. Modern epidemiologists tend to believe that the reason why the second wave of the Spanish flu was so devastating was that not enough people became immune to the H1N1 virus during the first wave. According to Sir Vallance’s initial stance, herd immunity would prevent a similar outcome for SARS-CoV2, “a virus with comparable lethality to H1N1 influenza in 1918”.
Worldwide, much of the COVID-19 response was built mimicking the response to the 1918 pandemic. When the Spanish flu hit, with no available vaccine to protect against influenza infection and no antibiotics to treat secondary bacterial infections, control measures were limited to non-pharmaceutical interventions (NPI) such as isolation, quarantine, good personal hygiene, use of disinfectants, and limitations of public gatherings. Just like with COVID-19 today.
According to the British natural herd immunity concept, the best approach to the 2020 SARS-COV2 pandemic would not be a pure herd immunity strategy, which would unleash the virus without any containment, but rather a “soft” mid-way, that would involve implementing some, but not all, available NPI measures against COVID-19. This would allow the virus to infect the population, but at a low pace, so that the peak of infections could be reduced and the health care system would not be overloaded with an excessive number of severely sick cases. This approach is referred to as the mitigation strategy.
THE MOST INFLUENCIAL NON PEER-REVIEWED PAPER OF ALL TIME?
The mitigation strategy is described in detail in a report from the Imperial College London (ICL); it notably contains a model of the pandemic, mostly created by the group of professor Neil Ferguson, government advisor on the Scientific Advisory Group for Emergencies (SAGE) committee. The model predicts the effects of NPI measures on the load of COVID-19 on the health care system in the U.K. Johan Giesecke, a senior epidemiologist involved in the Swedish COVID-19 response, called this study “one of the most influential non-peered reviewed papers ever.” So what does the report say?
With a pure herd immunity, do-nothing, strategy (black line in figure 1) no NPI are implemented, and the virus is free to spread in the population. This is predicted to result in an extremely high peak in the first three months, exceeding by nearly 40 times the number of available beds in UK (red line in figure 1). On the other hand, the fast built immunity is expected to prevent a second wave of infection.
In a mitigation strategy scenario (orange line in figure 1) , some public health measures are taken, such as case isolation, quarantine of household members, and social distancing for the elderly (over 70). Schools and businesses stay open and travel is not restricted. Although restricting social gatherings would be an option in this strategy, they can be allowed because, as Neil Ferguson writes:
“Stopping mass gatherings is predicted to have relatively little impact (results not shown) because the contact-time at such events is relatively small compared to the time spent at home, in schools or workplaces and in other community locations such as bars and restaurants.” (Ferguson et al., 2020)
Mitigation is a restrained herd immunity strategy, that to some extent contains the spread of a virus in a population without completely suppressing it, thus building immunity without causing a complete healthcare system collapse.
“The aim of mitigation is to reduce the impact of an epidemic by flattening the curve, reducing peak incidence and overall deaths. Since the aim of mitigation is to minimize mortality [in the long-term], the interventions need to remain in place for as much of the epidemic period as possible. Introducing such interventions too early risks allowing transmission to return once they are lifted (if insufficient herd immunity has developed)“ (Ferguson et al., 2020)
With the mitigation scenario described in the ICL model, the load on the health care system is reduced to “only” a 2-fold excess as compared to the availability, during the first three months, making it possible for the system to at least partly compensate by creating new critical care beds. The most attractive aspect of this strategy is, however, the expectation that the second wave of the epidemic would be less disastrous, as compared to the suppression scenario (green line in figure 1).
When a suppression strategy is applied, strict NPI are enacted, such as closure of all schools and universities, timely case isolation, household quarantine, and compulsory population-wide social distancing (for instance via temporary lock-downs), as well as travel bans and bans against social gatherings. Suppression was expected to work extremely well in the short run, resulting in an availability of critical care beds in Britain that would have actually exceeded the demand (Figure 1). The second wave of infections, though, is expected to be disastrous, with an estimated demand of critical care beds that would be exactly as high as that expected at baseline if no restrictions were implemented (see black line in figure 1).
The ICL also presented an estimate of the number of deaths that would result from the choice of each strategy. In that report, the seemingly sound mitigation strategy was estimated to result in 250,000 deaths in the UK and 1.1–1.2 million in the US.
THE U.K. RESPONSE
Although Prime Minister Boris Johnson and his team appeared very confident in the soundness of the mitigation approach, notwithstanding the expected high death toll, public opinion reacted to his “many lives lost” speech on March 12th (Box 1); and when ICL estimates went public on March 16th, the debate became heated. As rates of COVID-19 infections and deaths kept escalating in Europe, the UK government made a U-turn and entered a delayed lock-down, in line with the suppression strategy, on March 23th. For many weeks though, the consequences of the initial loose mitigation strategy kept claiming victims, making the UK a country with one of the highest COVID-19 related death tolls in Europe.
Since the publication of the report, Ferguson’s model has been described as “unreliable and a buggy mess” and “the most devastating software mistake of all time.” But the ICL report has nevertheless been considered valuable. Although it is now obvious that the model over-estimated COVID-19 mortality, it has provided the reasoning behind the mitigated herd immunity approach, and has allowed other countries to evaluate it, in fact mostly scaring them away.
THE NORWEGIAN DILEMMA
A conceptual framework similar to the British one is found in Norway’s early projections on the COVID-19 epidemic. On the website of the Norwegian Folkehelseinstituttet (FHI, the Public Health Institute), a very informative report describes, in detail, the decisional process that Norway underwent when planning its COVID-19 strategy. The document was published March 24th, the day after the UK went into lockdown.
The information provided by the FHI is an excellent complement to the model published by the Imperial College, as they both describe the reasoning and background knowledge at the moment the decisions that influenced the COVID-19 response in Northern Europe were taken.
The Norwegian report notably shows (Table 1) how the release strategy (do-nothing strategy) is expected to have a spreading rate or RE, between 2 and 3. This means that for each infected person, 2–3 other people will be infected, resulting in a fast exponential growth of the number of disease-ridden people in the population. The mitigation strategy has an expected RE of 1.3, which means that for everyfour infected persons there will be only 1 new case, with a resulting progressive reduction and levelling off of the infection rate, without stopping the spread. The suppression strategy, finally, is expected to bring the replication rate RE of the virus below 1 and progressively to 0, when the epidemic finally stops. It is also noteworthy, that the FHI report assesses the public health consequences to be least dramatic when a suppression strategy is implemented (Table 1).
The FHI also describes the different NPIs implemented in each strategy (Table 2), making it possible for anyone to assess which strategy a country follows by examining what measures are being implemented. This is useful when looking at countries that never declared their strategy endpoints, such as Sweden.
After a period of uncertainty, Norway entered into an early lock-down (suppression strategy) on March 12. As stated in the Norwegian report, initially the lock-down was aimed at “buying time” and work on preparedness, while gathering more information that would have served the decision of what strategy to follow long-term. Only in the following report, Norway stated that a suppression strategy was decided upon. In practice, however, this strategy was enacted from the day of the lockdown, as there has never been a factual attempt at implementing a mitigation strategy.
DENMARK: THE 5-HOUR LONG MITIGATION STRATEGY
Denmark reasoned in the same way as Norway and considered for a time a mitigation strategy. On March 11, it looked like the Danish people would join the Brits in the mitigation experiment, and at 3 pm Bolette Søborg, Senior Consultant at the Danish Health Authority, in a press conference stated that the shift would be imminent. This may have been the shortest mitigation intervention in history, as at 8 pm, that same day, the government announced that the nation was entering lock-down, a suppression strategy similar to the one enacted a few hours later by Norway.
A SPECIAL CASE: THE NETHERLANDS
Like the U.K., Norway and for a short while Denmark, the Netherlands were also tempted by a mitigation strategy. On March 16th, the Dutch Prime minister, Mark Rutte, announced that the country would build herd immunity by allowing the virus to spread to the population in a controlled manner.
The implemented measures indeed combined a stay-at-home-recommendation and social distancing, appealing to people’s own sense of responsibility and self-discipline, in a similar way as in the Swedish mitigation strategy (which is described next). However, a closer look shows that the Dutch measures were in fact more substantial: on the one hand, as in a typical mitigation strategy, good hygiene and stay/work at home measures were recommended, large gatherings were forbidden and universities and colleges were asked to go into distance teaching on March 12th; but on March 15th, i.e. only few days later, schools, cafés, restaurants and sport clubs were also asked to close. In fact, the strategy was quickly re-labeled an “intelligent lockdown” (Box 1). Interestingly, testing/tracing measures were also implemented early, but only in the northern part of the country, and thanks to the influential Prof Alex Friedrich, a virologist and head of microbiology and infection control at the University Medical Center in Groningen (UMCG). Friedrich, who disagreed with the loose strategy, announced that from March 18th the northern region would enact stricter measures such as systematic testing, case isolation, and tracing, thereby following WHO recommendations. A local test center was built, and the local health authorities were asked to direct people there for screening. As a result, during the period March 16th-April 5th, 2020, the region observed only a 4% increase in their death toll as compared to previous years, whereas there was a 100% increase in the worst-hit region of North Brabant, in the south of the Netherlands. This special form of mitigation and suppression strategy put the Netherlands in between countries that opted early for suppression, like Norway, and countries that entered into a delayed lock-down like U.K. or never went into lock-down, like Sweden.
Consequently, COVID-19 spread has not been fully avoided, and the Netherlands currently have a total death toll per capita that is significantly higher than that of Norway, but still lower than that of Sweden and the U.K. and it has now levelled off, as opposed to the situation in the other two countries (Box 1, Figure 2). However, nowadays, in contrast to the relaxed mitigation strategy initially praised, the current strategy is in favor of stricter measures favoring a suppression strategy. The shift in view in the Netherlands is notably supported by economists who believe that maximum control and containment of coronavirus would benefit not only public health but also the economy. Klynveld Peat Marwick Goerdeler (KPMG) group and Vrije University (VU) researchers estimated that sharpening up the contact tracing system and making sure the key R or reproduction number remains below 1 would generate €123bn for the economy by 2022.
SWEDEN AND ITS ONE-WAY TICKET TO MITIGATION
Differently from the other European countries reviewed above, Sweden’s COVID-19 strategy was never clearly stated. No decisional framework was ever presented by the Swedish Public Health Agency (Folkhälsomyndigheten, FHM), nor by the government. There has been short-term projections in terms of cases, health care burden and mortality that are calculated based on the updated regional data, but these projections do not give indications on the overall strategy used to curb the spread of SARS-CoV2 in Sweden.
The only conceptual framework that can be found among the published documents of the Swedish FHM is the pamphlet Pandemiberedskap (pandemic preparedness). The pamphlet was drafted in December 2019 as a mean to prepare for a generic “pandemic flu”. The document contains a simplified version of the model later presented by ICL and applies to a generic flu pandemic. It simply shows two curves, one steeper and of shorter duration, where no NPI are applied, and one less steep but of longer duration, where some NPI are put in place. According to the report, the pandemic cannot be stopped and NPI will only prolong the duration of the pandemic rather than reduce the number of victims. NPI are considered valid only to a certain extent as means to avoid healthcare system overload, by distributing the cases over a longer period.
The possibility that the Swedish FHM was regarding COVID-19 just as an ordinary flu, and was therefore of the opinion that there was no real need to stop it, could shed some light on the extremely loose Sweden’s approach towards the SARS-CoV2 pandemic. In the beginning of the epidemic, the only recommendations implemented in Sweden revolved around hygiene and isolation of self-assessed cases (stay at home if you are sick), as in any seasonal flu containment campaign. When the mortality rates in other countries started to sky-rocket, Swedish FHM began to also prescribe social distancing of people older than 70, a limitation of social gatherings to a maximum of 500 participants (reduced to a maximum of 50 on March 27th), recommended distance work for those in a position to do so, and distance education for high schools and universities (see Box 1). Travelers from other countries could come to Sweden without being quarantined, but for a while Swedes could not leave the country, and could not travel farther than a 2-hour journey within Sweden.
Masks, however, were not recommended, not even in primary care or elderly care facilities. In fact there was a recommendation NOT to use them, unless working with a COVID-19 patient. Along the same line of reasoning, primary and secondary schools stayed open, as well as shops, cafés and restaurants. Most interestingly, testing and contact tracing was stopped early in March, and was restricted to only those COVID-19 cases severe enough to be hospitalized. Cases that stayed at home were not followed-up, and were encouraged to return to their daily activities after only two days without fever. A dry cough or a loss of smell were not considered adequate reasons to stay away from work or school. Moreover, not only was no tracing of the cases’ contacts performed, but household members of cases were told to socially interact as usual, and were not among those prioritized to be tested for COVID-19.
Even in the absence of clear statements about the Swedish strategy, in the light of all we have learned from the detailed British and Norwegian reports, as well as from the Swedish flu pandemic document, the containment measures put in place by Sweden unequivocally point to a loose mitigation strategy, very similar to the one initially chosen by the U.K. A further indication that Sweden has been following this strategy aimed at building immunity at a slow pace, is offered by the frantic search for data to prove that herd immunity against COVID-19 has indeed been achieved in the country. After the Minister for Health and Social Affairs, Lena Hallengren, and FHM criticized private immunity testing, arguing that it was unreliable and capitalized on people’s concerns during a pandemic, at the end of May FHM not only allowed, but also encouraged private immunity testing, after securing an agreement to receive immunity data from the private companies. There may not have been any clear statement about which approach was chosen, but actions often speak louder than words.
Why is there such a peculiar silence on the overall Swedish strategy? This is all the more puzzling as FHM has been holding daily press updates on the handling of the COVID-19 pandemic in Sweden, ever since early March. When questioned about the strategy, vague, generic answers are given by both FHM and the government’s representatives, including Prime Minister Stefan Löfven and Minister for Foreign Affairs Ann Linde. The answers range from “we are not doing anything different from any other country” or “our aim is to save lives like any other country” to “we are succeeding at flattening the curve and protecting the health care system”. Why is it so difficult to claim that the mitigation/herd immunity strategy is followed? The nearly 6000 victims of COVID19 in Sweden (Figure 3) may partly give the answer to this question…
CONTRIBUTORS: Barbara Caracciolo, M.Sc epidemiology, Ph.D aging research; Marie Gorwa, professor in applied microbiology; Lena Einhorn, MD Ph.D virology; Anders Jansson, MD in clinical physiology; Anders Vahlne, professor emeritus in clinical virology; Stefan Einhorn, professor in molecular oncology; Björn Olsen, professor in infectious diseases; Cecilia Söderberg-Nauclér, professor in microbial pathogenesis; Åke Gustafsson, clinical virologist; Stefan Hanson, clinical virologist, Ph.D international health; Åke Lundkvist, professor in virology; Anders Wahlin, professor emeritus in hematology; Andrew Ewing, professor in bioanalytical chemistry; Jens Sörensen, associate professor in peace and development research; Sigurd Bergmann, professor emeritus in religious studies; Olle G P Isaksson, professor in biomedicine; Marcus Carlsson, associate professor in mathematics; Jan Lötvall, professor in allergy, asthma, immunology; Emil Bergholtz, professor in theoretical physics; Jana Bergholtz, Ph.D in earth and planetary sciences, patients’ representative; Manuel Felices, MD, Ph.D endocrinology; Bo Lunbäck, professor in clinical epidemiology and lung diseases; Nele Brusselaers, associate professor in clinical epidemiology; Leif Bjermer, professor in respiratory medicine and allergology.
