Researchers from the École des Hautes Études de la Santé Publique (EHESP) presented an article on their website estimating that a little over 60,000 deaths could have been avoided thanks to containment. This work has been extensively commented on in the press and has not failed to raise certain questions, even reservations. However, it is on the basis of these elements that the Prime Minister justified France's policy of containment and progressive deconfinement before the National Assembly, giving this study official approval. However, two scientific articles appeared simultaneously on 2 May and produced results that were diametrically opposed to the official discourse: containment has not spared lives, it has not been able to slow the course of the epidemic. It has been applied without any scientific measurement of its proven effectiveness, and has undesirable side effects, not only economic ones
The sudden emergence of the Covid-19 epidemic in our daily lives posed a major challenge to governments. Faced with an unexpected and unprecedented situation, the leaders of most countries had to take urgent decisions and show unprecedented responsiveness to protect their populations. Most countries were not prepared for such a pandemic and found themselves faced with a lack of screening, resuscitation and protection capabilities. These shortcomings led them to adopt social distancing measures and, depending on the country, to implement a wide range of public measures, ranging from total containment, sanctioned by law enforcement, to more flexible practices, encouraging social distancing without locking the population in.
The choice of confinement has affected half of humanity. The French have been confined since 17 March; they are forbidden to leave their homes without a derogatory certificate, a modern form of pass. Confinement ordered, without ever saying the word, by the President of the Republic in his speech of 16 March. A presidential decision taken on the basis of the opinion of a Scientific Council specially set up for this purpose.
After several weeks of confinement, and as the deconfinement date approaches, the validity of this decision is still being debated between those who believe that this measure, despite all its harmful consequences, was necessary, and those who believe that it could be avoided. Indeed, studies are accumulating to provide answers to these questions. Two of them were published on May 2, one in pre-publication on the specialized platform medRXiv (Thomas Meunier), the other in the International Journal of Medicine - JIM (Jean-François Toussaint et al.) They clearly challenge the methods and conclusions of another study, carried out by the EHESP (Jonathan Roux et al.) confirming the French political choice of containment.
Before going into the details of the scientific controversy, it is useful to consider a situation that has gone unnoticed. While new medical treatments proposed to cure Covid-19 must be validated by specific medical protocols, including double-blind studies, the benefits and risks of containment strategies are not subject to any comparative testing. The ethical and methodological framework of a randomized study is, of course, impossible to establish at the population level (yet it is the basis for establishing the scientific validity of therapeutic trials) and is therefore lacking to determine the benefit of measures that also have significant and lasting side effects". These modelling studies, due to the very specific context of the pandemic and the urgency of the situation, have disrupted the normal course of any scientific study: carrying out the study, submission to a peer-reviewed scientific journal, expert appraisal, publication after possible revision or rejection of the article by the journal. "write the authors of the JIM.
These measures of total containment, such as those decided in Italy, France, Spain and the United Kingdom, have not been tested in the countries of Western Europe for centuries; they come from the depths of the ages and the great fears of mankind.
Why not enjoy unlimited reading of UP'? Subscribe from €1.90 per week.
Their effects on the mental and physical health of the contemporary population are largely unknown. It has already been demonstrated that the epidemic episode of COVID-19 has, in itself, affected mental health, with anxiety and depression syndromes, and that the consequences of isolation could aggravate these pathologies. In the absence of a control group, the impacts on the population of Western Europe will not be measurable for several months. Nevertheless, an increase in mortality due to difficulties in accessing basic health care, a worsening of mental conditions linked to isolation, particularly among the elderly, and the social and human consequences of the economic recession are to be expected.
Containment measures are therefore only appropriate if their effect in limiting the spread of the epidemic is to save more lives than the number of deaths they cause." Attempting a real-time evaluation of the effectiveness of total containment policies therefore seems crucial to help public policy decisions in the coming weeks. "wrote Thomas Meunier in the preamble to his article of the medRXiv.
However, the EHESP study is a reference for the government
The study led by Jonathan Roux was published on April 23rd on the website of the École des Hautes Études de la Santé Publique. It presents a model of the number of hospitalizations, resuscitation patients and deaths potentially avoided by containment in France since March 17th. Based on the available knowledge on viral transmission, the authors estimate the effect of a lack of containment on the number of events in the three situations. To do so, they observe the data from March 20th to 28th and verify that their model reproduces the results for the 13 metropolitan regions. They then extend this model, which is supposedly validated, until April 19th and then measure the difference between the model and the results observed at that date.
They conclude that containment prevented 61,739 deaths in one month, not adding to the 12,069 hospital deaths observed to that date. In this model, mortality reached 10,000 in-hospital deaths on April 19 and doubled every 4-5 days, as indicated on the website of the École des Hautes Études de la Santé Publique." The study shows that one month of containment would have prevented up to 60,000 deaths and that without containment, more than 100,000 resuscitation beds would have been needed by 20 April 2020. " it says.
It was on the basis of these conclusions that the Prime Minister justified in hindsight the French containment policy implemented since 17 March, before the Members of Parliament on 28 April:
However, Professor Toussaint and his colleagues believe that this EHESP paper poses several methodological problems: the confidence intervals around the modelled curves are large, which could call into question the models themselves; on the other hand, the curve diverges from the real points used to validate the model and overestimates the number of events that could have occurred without containment; and finally, since the model values increase exponentially over time, the modelling systematically amplifies the results as time passes. In short, " the model significantly overestimates the positive effects of containment ".
EHESP researchers no doubt wanted to demonstrate the effectiveness of widespread containment in France; was it a public commission to justify a difficult policy? We will never know.
However, the method used in this study does not provide a scientific answer to the question posed because many of the parameters estimated in the study are not scientifically valid. a priori Jean-François Toussaint and his colleagues affirm that the use of such models "... and still very uncertain to this day can change the trajectories and results of projections in an extremely consequential way. Jean-François Toussaint and his colleagues affirm that the use of such models "... must remain cautious about the conclusions to be drawn and no one can reasonably rely on this type of study to justify a measure whose health, economic and social impacts we have not finished measuring. "
Total containment with no effect
As if to drive the nail in, another international study conducted by Franco-American researcher Thomas Meunier and published on medRxiv confirms these reservations. The analysis report warns at the outset: " We show here that the available data show no effect of the total containment policies applied in Italy, Spain, France and the United Kingdom in the temporal evolution of the COVID-19 epidemic. "He continues:" no positive change is noted in the trend in the daily growth rate of deaths, doubling time or number of reproductions, weeks after the containment policies should have shown their effects. "
To fight against disinformation and to favour analyses that decipher the news, join the circle of UP' subscribers.
One of the interesting results of the study is the comparison of countries applying different containment measures. The figure below shows the evolution of daily growth rates of deaths (t), their linear trends, reproduction figures and doubling times for 10 countries. The researchers selected countries with more than 1000 deaths on 15 April 2020, and opportunely chose to exclude the data from China, given the growing doubts about their accuracy. The time reference was chosen as the day when the total number of deaths exceeded 100 in each country.
The evolution of the daily number of deaths shows a similar general downward trend in all countries. It is interesting to note that, while the linear trends in growth rates have similar slopes in almost all countries, they show a wide range of values, which shows that, although the slowing of the epidemic is following a similar trajectory, each country started at very different levels of growth rates.
This general downward trend is accompanied by a steady decline in the transmission rate in all countries, with similar slopes and, again, a wide range of initial reproduction rates. As would be expected with a declining growth rate and reproduction rate, the doubling time is increasing in all countries from the beginning of the time series. The figure therefore shows, the authors state, "... that the temporal evolution of the epidemic is homogeneous in Western Europe "In particular, countries with policies of social distancing but no home confinement, such as the Netherlands and Germany, have experienced a decline in the epidemic that is very similar in terms of growth rate, number of virus replications and doubling time to that of countries where home confinement is police-controlled. On the other hand, results from Sweden suggest that a total lack of action may lead to a more variable decline in the epidemic.
Furthermore, this observational study, using a generalized phenomenological method based on daily death records alone, demonstrates that the total containment policies of France, Italy, Spain and the United Kingdom did not have the expected effects in the evolution of the COVID epidemic19 .
In summary, the researchers are formal and go against the dominant theses: " Our results show a general downward trend in growth rates and reproduction numbers two to three weeks before total containment policies produce visible effects."
Researchers suggest that "the researchers suggest that no lives were saved by this strategy. " compared to policies of social distancing less restrictive than total containment. They continue: " Comparison of the evolution of the epidemic between countries that are totally locked out and neighbouring countries that only apply social distancing measures confirms the absence of any effect of home confinement."
The evolution of the epidemic in Sweden, however, indicates that in the absence of social distancing measures, the decline of the epidemic may be subject to greater fluctuations.
This work therefore suggests that social distancing measures, such as those applied in the Netherlands and Germany, or in Italy, France, Spain and the United Kingdom prior to total containment strategies, have approximately the same effects as coercively police-controlled home containment policies.
So far, the scientists who authored this study note that the reasons for the relatively steady decline of the epidemic remain largely unknown. While social distancing efforts may have contributed to this, environmental conditions may just as well have played a role (possible seasonality of the virus). in any case, while social distancing measures (barrier gestures, avoidance of concentrations of people, wearing of masks, etc.) have tangible effects on controlling the epidemic, population containment measures are clearly in question. they are justified on erroneous grounds and controversial methods, yet they are the basis for political decisions that directly impact the lives of tens of millions of people. this is how Jean-François Toussaint concludes: " it is probably best to avoid making such errors, especially if they alter the public decision"
At the moment, France is wondering about the end of confinement, while understanding that May 11th will not be a date of liberation, a return to old habits, because the epidemic is still there. So we invent a deconfinement that still confines us a little. We invent green and red maps of the territory, we define travel limits. We promise to confiscate holidays, beaches, stadiums. We deprive the population of almost everything because scientists know almost nothing about this virus, which has surprises in store for us every day. We continue with the idea of confinement. Errare humanum est, perseverare diabolicum est.
Containment is magical thinking while waiting for the vaccine. Whether it is ineffective, whether it produces deleterious side effects, matters little in terms of the sense of control it gives to the person who imposes it. This is a short-sighted calculation that other countries do not make, preferring to pay the sole price of the epidemic without adding the price of the damage caused by a policy that is out of step with the times. Imported from the ancestral epidemic in China, containment is not only ineffective, it is a revelation of our inadequacies and our distress.