Analysis and commentary
by Catherine Riva, Serena Tinari – Re-Check.ch
February 19, 2021
The Swiss National COVID-19 Science Task Force, a group of experts that is supposed to inform decision-makers about the scientific evidence in the coronavirus crisis, continues to enjoy extraordinary visibility and influence in the daily lives of the Swiss in an unprecedented way. A debate on its legitimacy and its role appears more necessary than ever. Above all, it is high time to critically examine the quality of its scientific work.
In less than a year, the Swiss media have elevated some researchers to the rank of VIP. Whereas in January 2020, the names of Marcel Salathé, Jacques Fellay, Christian Althaus, Didier Trono, Samia Hurst-Majno, Richard Neher, Nicola Low, Martin Ackermann or Matthias Egger were virtually unknown to the general public, today almost everyone in Switzerland knows who these experts members of the Task Force are.
According to the Horizons magazine of the Swiss National Science Foundation (SNSF), the Swiss media database SMD has more than 1,400 entries for Marcel Salathé and around 700 entries for Christian Althaus (who left the Task Force in January 2021) for the period January to June 2020 alone. These record figures illustrate the findings of the report “Quality of the Media, Yearbook 2020”, published by the Center for Public Domain and Society Research (fög) of the University of Zurich: since February 2020, the media have become highly dependent on experts, and when it comes to the COVID-19 crisis, the first to be called upon in Switzerland are the members of the Task Force.
The official mission of these researchers is to advise the Federal Council, the cantons and the federal administration in the COVID-19 crisis by means of recommendations that are supposed to be based on scientific evidence. However, the Task Force rarely limits itself to providing advice. Since February 2020, not a month goes by without one of its members, or even the entire collective, actively giving voice to the media (as for instance here in June, here July, here in August, here in September, here in October, here in November and here in December 2020).
Most often, it is to paint a threatening picture of the situation, criticize the measures decreed by the authorities and demand stricter ones.
And even when the Federal Council initially decides not to follow its recommendations, the Task Force almost always ends up getting what it recommends: for example, the compulsory wearing of masks on public transport, or the closure of restaurants, bars and nightclubs).
As one of the most influential organizations in the country today, the Task Force’s legitimacy and the quality of its work deserve to be examined.
Which status? What legitimacy?
Let us begin by recalling that the Task Force is an unelected and unrepresentative body, not accountable to either the people or Parliament. It was self-established on the initiative, among others, of some of its members. The documents we obtained after a Freedom of Information (FOI) request indicate that Martin Ackermann, the current chairman of the Task Force, Yves Flückiger (swissuniversities), Michael Hengartner (President of the ETH) and Matthias Egger (President of the Swiss National Science Foundation) sent a Word file in three languages on March 24, 2020 to Lukas Bruhin, head of the Federal Council’s staff in charge of managing the coronavirus crisis. This document presented a “turnkey” concept for an ad hoc scientific task force. The comparison of this file and the document, through which the Federal Office of Public Health (FOPH) and the State Secretariat for Education, Research and Innovation (SERI) entrusted the mandate to set up a “Swiss National COVID-19 Task Force” to the Swiss National Science Foundation (SNSF), the ETH Board (Swiss Federal Institutes of Technology), Swissuniversities and the Swiss Academy of Sciences shows that the concept has been endorsed almost unchanged by the heads of the Crisis Staff, the FOPH, and the SERI. The Task Force was set up in a matter of a few e-mails, a few phone calls and a Power Point presentation. The question of the duration of its mandate was not addressed.
In addition to this lack of democratic legitimacy, the functioning of the Task Force is opaque, to say the least. Its members are co-opted. Appointments simply have to be approved by the Task Force’s proxies, which today are the Federal Department of the Interior (DFI) and the FOPH. There are no regulations that set out the criteria for appointment, election procedures or rotation. Pietro Vernazza, for example, head physician of the infectiology department of the cantonal hospital of St. Gallen, was invited to join the expert committee in March 2020. Before being uninvited without explanation.
Transparency regarding possible conflicts of interest of Task Force members was also not always on the agenda. In April 2020, when we asked SERI and the FOPH to send us the declarations of interest of these researchers, they pointed us to the website of the expert group, where, at the time, there were only links to the web page of each researcher in the institution where he or she worked. Such a page does not, in principle, contain a declaration of conflict of interest: it is above all a space where the scientist presents his work in the best possible light, with a list of his publications and distinctions. It was not until June 2020 that formal declarations of interest became available. However, the question of checking their accuracy remains.
Another problem is that neither the legislature nor the sovereign have the means to know how the 70 current members of the Task Force make their decisions: another FOI request filed by Mike Wyniger, film director in Bern, revealed that the Task Force does not keep minutes of its meetings. Gian Casutt, Head of Communications of the ETH Board, justified this lack of official documents as follows: “As you may know, the members of the Task Force work on a voluntary basis and are not remunerated, so the procedures remain simple and free of administrative burdens.” This willingness to operate in a non-bureaucratic way may have advantages for the Task Force members, but it poses a major problem in terms of transparency and the possibility of democratic control. Indeed, the federal law on archiving stipulates that “all documents of the Confederation that have legal, political, economic, historical, social or cultural value” must be archived. It is the documents thus archived that are accessed by those who file FOI applications. In fact, if the Task Force does not produce minutes of its meetings, an important part of its activities can never be subjected to a detailed examination, neither today nor in the future.
Finally, no text resolves the question of the suspension of the Task Force’s activities or its dissolution.
Non-compliance with rules and role confusion
Nevertheless, from the very beginning, its rules of procedure specify that its members are not supposed to communicate directly with the media or on social networks, unless they do so solely on their own behalf. Only the Task Force’s chairman is entitled to do speak to the media, and if such communication concerns new measures, it can happen only after they have been made public by the authorities.
The least that can be said is that some members of the Task Force do not always take these provisions into account, if we refer to the regularity of their public interventions in mainstream media and social networks.
The interview of Samia Hurst-Majno published by the Blick on February 7, 2021 is emblematic of what has been happening over the last ten months: when asked, these experts never speak only as researchers in a specialized field; the media always specify that they are “members of the Task Force”. Thus, in the Blick article, Samia Hurst-Majno is first presented as “vice-chair of the Task Force”, her specialty is only mentioned after seven questions (out of eleven). The researcher speaks at length on epidemiological issues, notably the threat of a third wave and the evolution of the number of “cases”, but also on the attitude of the country’s inhabitants (we would be “too careless”) and the need for new measures. These are all subjects that do not fall within her field (bioethics). This obviously does not prevent Samia Hurst-Majno, like other members of the Task Force, from feeling empowered to comment on these aspects, and to distribute the good and bad points to the population and the authorities.
The result of this confusion of roles has been the same for months: the statements of Task Force members make the headlines. As soon as they are picked up by other media, they become “facts” that stir up anxiety and tension, and end up influencing the executive. However, the ins and outs of the discussions between the executive and the Task Force are unknown. The FOI request that we submitted in mid-January 2021 should not be answered until the end of February 2021. We will not fail to keep our readers informed.
In the meantime, one thing is certain: the public statements of the members of the Task Force regularly determine the direction of the debate despite the fact that the rules of procedure of this group of experts stipulate that things should not be done this way.
But isn’t this independence positive? One might object. Doesn’t it guarantee that scientists can work without political pressure? The answer could be yes, if the scientific output of the Task Force was of high quality.
Unfortunately, this is not the case.
The scenarios, analyses and models produced by these researchers have regularly proved to be wrong. Starting with the figures that Christian Althaus, Emma Hodcroft, Richard Neher and Marcel Salathé presented in a first letter to federar councillor Alain Berset in February 2020. The spectacular rates of serious cases, hospitalizations and deaths they put forward were based, among other things, on the totally erroneous modelling of Imperial College London. Of course, in February 2020, it was difficult to be sure of anything about the SARS-CoV-2 virus. But the same is true for the scenarios that the Task Force and some of its members presented afterwards in the press briefings and through their interventions in the media: they were often and heavily mistaken, as illustrated by the following three examples.
1. Evolution of the epidemic in the summer of 2020: In April 2020, a study supervised by Jacques Fellay, associate professor at the EPFL and the CHUV and member of the Task Force, announced that the “gradual relaxation of anti-coronavirus measures in Switzerland could increase the number of infections and provoke a rebound during the summer”, with “5,000 to 15,000 deaths, or even more than 20,000”. Jacques Fellay was clear: “All the groups that model the epidemic in Switzerland come to very similar conclusions. The big unknown is the effect of the gradual relaxation of the measures currently in place. It is not by banning a few festivals during the summer that we will be able to keep the R0 below 1.2”. In mid-June 2020, the Task Force returned to the matter: its chairman announced that he feared a second wave in the summer and criticized the Federal Council’s measures as insufficient. In reality, between May 1 and October 1, 2020, without a lockdown, there were never more than 0.29 hospitalizations per 100,000 inhabitants per day with a maximum peak of 29 hospitalizations in one day, and never more than 0.08 deaths per 100,000 inhabitants attributed to COVID-19 per day with a maximum peak of 7 deaths in one day. These figures are far removed from the predictions that had been put forward by the experts. To the best of our knowledge, they have not been retracted, corrected, and nobody has apologized for the mistakes.
2. Saturation of hospital capacities in autumn 2020: during the months of October, November and December 2020, the Task Force constantly criticized the measures taken as insufficient. At no time, however, were Swiss hospitals and their intensive care units saturated (1) (2) (3). This did not prevent the major media from maintaining an atmosphere of panic for months with graphs and red lines interposed, without contextualizing or putting the data into perspective.
3. Evolution of the epidemic between mid-December 2020 and mid-January 2021: In mid-December 2020, the Task Force announced that the measures were not sufficient and that a lockdown was necessary. It feared the worst for the holidays. In fact, since mid-December, the number of hospitalizations and deaths attributed to COVID-19 has been steadily decreasing. Despite the fact that many Swiss have been frequenting ski resorts and eating in restaurants on the slopes in the cantons where they were open (a behavior that made fear the worst) (4) (5). Here again, none of the gloomy predictions have come true: since the beginning of November 2020, all the curves (“new cases”, hospitalizations, deaths) have shown a steady decline. And here again, there has been no correction, no retraction, no apologies from the experts.
The question of the British, South African and Brazilian “new variants” will probably be the next test of the robustness of the Task Force’s predictions. On December 26, it once again called for “urgent additional measures in Switzerland in the face of the pandemic” because of these new variants. On December 29, 2020, it presented a “Scientific Update” devoted to this topic, including variant B.1.1.7 (“British variant”). This text was accompanied by disturbing graphics. According to the scenarios considered, these predicted an explosion of incidence, which could peak in April at more than 20,000 “new cases” per day. This scenario was to come true if “containment measures” were to bring “to 0.9 the reproduction rate of the currently dominant SARS-CoV-2 strains in Switzerland”, reducing “the number of infections with these strains by about half every four weeks”. On the same day, Martin Ackermann gave a speech at the press briefing, at which he presented these same graphs.
What happened next felt like déjà vu. On December 30, the Federal Council announced that it would not tighten the measures. It finally changed course and introduced tougher measures on January 13, 2021. While all indicators continued to fall, the «variants» were being raised to argue that the worst was to be feared. The result was a five-week extension to the closure of restaurants, cultural institutions and sports and recreation facilities. But also the introduction of “new measures to drastically reduce contacts between people”: compulsory home office, closure of many stores, new restrictions for private events and gatherings.
Some clues suggest that here again, the Task Force may have needlessly painted the devil on the wall. According to its scenarios, even the most optimistic ones, the daily number of new “cases” should have risen in February already. So far, nothing of the sort has happened. On the contrary, since the beginning of February, the 7-day average is less than 1500 “new cases” per day. This same indicator is also decreasing in the United Kingdom where this new variant has been dominant since the last week of December 2020. As is the number of hospitalizations and deaths. Finally, a King’s College study published on February 1, 2021 concluded that there is no difference in terms of symptoms and severity of evolution with the new variant.
On February 8, 24 Heures asked the Task Force’s chairman Martin Ackermann “whether he still believed in the possibility of a 20,000 peak or whether he had revalued the ‘worst-case scenario’ downwards”. The article’s author noted: “No clear answer was given. The expert no longer gives any figures for March or April.”
What are the policy briefs worth?
Scenarios and other models are unfortunately not the only problematic aspect of the Task Force’s work. One could also question the quality of the policy briefs it has published. (“70+ papers!”, Matthias Egger, former chairman of the Task Force, recently jubilated in a tweet). The policy briefs are dossiers where the Task Force “address urgent issues regarding the COVID-19 crisis” and which “the thinking of the Taskforce on this topic at that time“. The Task Force further states that if “required, the policy briefs are updated in the light of new studies or other data”.
Let’s take two examples of policy briefs on measures that affect several million people in Switzerland every day: facemasks and PCR testing.
At the end of July 2020, InfoSperber noted that in the policy brief “Strategy to react to substantial increases in the numbers of SARS-CoV-2 infections in Switzerland” of July 1, 2020, the claim that a meta-analysis had shown that wearing a mask could reduce the transmission of the virus “by up to 80%”, for example, was based on methodologically untenable shortcuts and inaccurate quotes. But it gave the media the ammunition to put pressure on the cantonal authorities and the Federal Council. It was not corrected, despite the shortcomings highlighted by InfoSperber.
PCR testing is also a very important issue. Positive PCR tests are in fact today the first “unit of measurement” of the seriousness of the epidemic, since they are the source of “new cases” that are reported day after day in the media and invoked by the authorities to justify new restrictions. For the person concerned, and often for her relatives, a positive PCR test is always fraught with consequences: quarantine, deprivation of the right to leave one’s home, prohibition of access to certain territories in case of travel, deprivation of contact with relatives, even hospitalization for moderately dependent people. In this context, one could legitimately expect the authorities and the Task Force to take great care to ensure that the conduct and interpretation of these tests are carried out according to solid criteria, duly screened and evaluated according to the best methodological standards.
However, in January 2021, the World Health Organization (WHO) updated its guidance on the handling and interpretation of PCR tests. In this document, the WHO states that when the prevalence of a disease decreases, the risk of false-positive results increases, regardless of the claimed specificity of the test; that PCR testing is an aid for diagnosis, not a diagnostic test; and that the number of amplification cycles needed to detect the virus by PCR testing is inversely proportional to the viral load of the person tested. For this reason, the WHO recommends that when the test results do not match the clinical presentation (e.g. a positive test in a symptom-free person), a new sample should be taken and the test repeated, and that the laboratory should report the number of amplification cycles (Ct value). Finally, the WHO recommends caution in the interpretation of weak positive results.
We asked the Task Force whether it would consider following the WHO and recommending that testing centers, doctors and hospitals re-test if the person tested positive was symptom-free, and that laboratories report the number of amplification cycles performed during the test. Didier Trono, virologist, professor at the EPFL and chairman of the Task Force’s Expert Group Diagnostics and testing, replied: “These two points have more to do with the FOPH than with the Task Force, because they are more legal than scientific.” The FOPH, for its part, has informed us that it does not intend to modify its directives in this area.
Trono’s answer raises a question: whether the issues here are really “more legal than scientific”. The question of the usefulness of the PCR test in terms of its ability to predict whether a person is infectious or not has been debated for several months in the medical-scientific literature, as shown by an ongoing review of 29 studies.
The preliminary conclusions of this review contrast with those of the policy brief “An update on SARS-CoV-2 detection tests”, published by the Task Force at the end of October 2020. This document states: “Commonly used machines perform 40 amplification cycles, and Ct values of 37 and below are generally considered unequivocally positive, but even higher values raise a strong suspicion of infection due to the extreme specificity of the technique.” We asked the Task Force and the FOPH whether it could be concluded from this statement that most of the results of PCR tests conducted in Switzerland had been obtained after 40 amplification cycles.
The FOPH answered yes.
Trono explained: “Most of the results reflect tests that become positive well before 37 cycles. Those at this level or above are exceptional, and are usually confirmed by a second test.” Unfortunately, there are no studies or documents to verify these claims, since not all laboratories systematically indicate the Ct value when reporting test results.
We also asked Trono on which references in the literature the Ct value of 37 advanced in the policy brief was based, as well as the statement that “even higher values raise a strong suspicion of infection due to the extreme specificity of the technique”, as we could not find any publications corroborating them.
Trono’s answer: “This comment (the policy brief, editor’s note) describes the attitude of laboratories involved in diagnosis in Switzerland”, it is therefore “not the result of a compilation of the literature or of a particular study”. The researcher further clarified: “It should be noted that two of the members of the TF (Task Force, editor’s note) are directors of diagnostic laboratories, namely Laurent Kaiser from Geneva and Alexandra Trkola from Zurich, and participated in the drafting of the related documents. Furthermore, since the beginning of the pandemic, the TF has regularly discussed with teams from other diagnostic units, including occasionally from the private sector.”
How about EBM?
Unfortunately, in Evidence-Based Medicine (EBM), expert opinions, however eminent, are not considered evidence until they are transparently supported by clinical evidence. The same applies to the practices of actors such as laboratories: it is not enough that they are commonplace, it must be possible to verify what type of evidence they are based on.
We asked Trono what were the references on which the experts based their assertions and on which laboratories based their practices. Trono replied as follows: “Please dispense me from the sermons on EBM: when the document quoted speaks of ‘strong suspicion’ it refers to the qualitative assessment of this type of result by those who perform these tests, in the same way that a radiologist would speak of ?strong suspicion of pulmonary embolism? on the basis of an evocative but not pathognomonic scintigraphy, leaving it up to the health care team to interpret this result in the context of the symptoms and clinical signs presented by the patient. The use of the term ‘strong suspicion’ is therefore not objectionable since it is not intended to be quantitative at all.”
Trono expressed himself at length in this e-mail, without however answering the question we asked him. So we still don’t know on which scientific references the Task Force bases its statement about a “strong suspicion of infection due to the extreme specificity of the technique”. One point is moreover surprising: the virology laboratory of the HUG directed by Laurent Kaiser, which Trono underlines collaborated in the drafting of the policy brief concerned, indeed specified in a circular of April 2020: “It should be noted that for all of these techniques, Ct values above 35 represent very low quantities of viral RNA.” And low amounts of viral RNA are not compatible with a “strong suspicion of infection”.
To recap: Despite the importance of PCR tests in the current crisis, the Task Force and the FOPH believe that it is sufficient to rely on the practice and that there is no need to present transparent evidence of a good risk/benefit ratio. The FOPH replied: 2The recommendations of the Task Force are clear and correspond to good laboratory practice in Switzerland. The FOPH agrees with the Task Force on this point.” And when the representative of the Task Force is asked for scientific evidence in the sense that the EBM defines it, he asks to be dispensed “from the sermons on EBM”.
Where are the studies?
More generally, it is regrettable that the Task Force did not recommend initiating studies that would have made it possible to establish whether the non-pharmaceutical measures (facemasks, social distancing, quarantine, restriction of the number of encounters, closure of infrastructures) that it recommended presented a good benefit/risk ratio.
However, in April 2020, the Task Force confirmed that the knowledge on which it based its recommendations was insufficient. And on its site, it underlines that part of its mission is: “Identifying fields and opportunities for research where the Swiss scientific community can make an important contribution to understanding and combating COVID-19.” The urgency of such work on non-pharmaceutical interventions was highlighted in September 2020 by the Deutsches Netzwerk Evidenzbasierte Medizin in Germany. And in November 2020, an analysis published in the medical journal BMJ recalled the need to address the potential harms of such measures.
Within ten months, members of the Task Force would have had time, for example, to design and launch cluster studies in Switzerland on the benefits of wearing masks, non-access to certain structures, closures and lockdowns. Admittedly, these experts do not receive any additional remuneration and their time is valuable, as the ETH Board spokesman mentioned above has explained. But then, why not allocate it to research that would identify measures that could make a difference without harming the population, instead of using this precious time to address the media and ignite new controversies by torpedoing the principles of subsidiarity and the functioning of institutions?
However, it should be pointed out that the Task Force has made a significant research effort on the consequences of measures that it has recommended and have been implemented in the country: a survey that recently highlighted the worrying state of mental health of a growing part of the population. But here too, the conclusion of the expert group published on January 20, 2021 leaves one wondering. Instead of being self-critical (if the mental health of the population is deteriorating, it may also be due to the measures taken by the authorities on the advice of this expert group), the Task Force recommends first of all to “reduce case numbers”. In other words, the number of positive tests. This amounts, in its logic, to recommending even tougher restrictions.
Nor can Switzerland expect answers to all these questions to come from the National Research Programme COVID-19 (NRP 78 “Covid-19”) of the SNSF, which was launched on the initiative of the same actors who advocated the establishment of the Task Force and whose Steering Committee has been chaired since November 2020 by Marcel Salathé. None of the projects selected, in fact, aims at assessing the benefit/risk ratio of non-pharmaceutical interventions imposed for months on millions of people in Switzerland.
Only two projects address these points indirectly, but their premise is that these measures (restrictions, above all) are appropriate and effective, and that the challenge is to develop instruments to convince people to comply with them. Here again, the research priorities of those who present themselves as “the Swiss scientific community” appear oddly oriented.
Do we want to go on like this?
This overview suggests that the Task Force is hardly acting as a scientific advisory committee, conscious of the limits of its mandate, concerned about the complexity of public health issues (where even the best-intentioned measures can prove deleterious), and eager to demonstrate the greatest possible methodological rigor, by being transparent about both the sources it is using and the unknowns that still accompany the current knowledge.
It is also somewhat disturbing that these alleged experts in public health policy seem to give little weight to avoiding inflammatory statements, which end up confusing and frightening the public. For more than a year now, the population has been in a permanent state of anxiety and must face every day both alarmist media coverage and the undeniable weight of far-reaching restrictions, which have a considerable impact on private and professional life.
The profile of the Task Force, with its public interventions, is increasingly resembling that of a pressure group, whose aim seems to be to promote a given strategy, even if it means imposing it through a selective reading of “science”, and which seems to have no intention either of acknowledging its past errors of appreciation or of learning from them.
The Task Force has had more than ten months to show what it is capable of. The time seems to have come for the legislative power – whose silence has been deafening since the beginning of this crisis – and the people to ask themselves whether they wish to see this group of experts continue to exist and work as it has done so far.
February 19, 2021 – Notice: we have added a screenshot of the e-mail showing the first contact proposing the foundation of the Task Force and added the names of the e-mail’s co-signatories.