Design: Jean-Dominique Lavoix-Carli
Eleven months into the COVID-19 pandemic the second wave spreads. More than 50 million people were contaminated globally by 9 November 2020 (COVID-19 Dashboard by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University). More than 1.25 million people had died by then (ibid.).
On 10 November, Europe crossed the threshold of 300.000 deaths daily, i.e. almost a quarter of global daily deaths, when it only represents 10% of the world’s population (Reuters, COVID-19 Global tracker). On a 7-day average, Europe represents slightly more than 50% of global infections and slightly less than 50% of global deaths (Reuters, COVID-19 Global tracker).
The “geographic distribution of 14-day cumulative number of reported COVID-19 cases per 100 000 population, worldwide” for 10 November 2020 as given by the European Centre for Disease Prevention and Control also shows that Europe, with the U.S., are the worst hit regions in the world.
Before the pandemic, France, a member of the G7 with a permanent seat at the Security Council of the United Nations ranked 7th globally in terms of GDP (“World Economic Outlook Database, October 2020“, International Monetary Fund).
France also ranks high in the way the pandemic takes its toll on the country.
In November 2020, France faces record figures. On 10 November, it has the second highest daily rate of infection worldwide, after the U.S. (7-day average reported – Reuters). It also has the third highest daily number of deaths worldwide, after the U.S. and India, despite a much smaller population.
However, when compared with its population, France is in the middle range of countries both in terms of contamination and death, as shown in the comparative graph by Reuters.
Considering the worldwide place of France in terms of political, geopolitical and economic power and influence, what is happening and what shall we expect in the coming months?
Some of the critical factors to assess the multiple impacts of the pandemic in the future and thus reduce uncertainty are the casualties in terms of death, the capacity of the care system to resist the onslaught, as well as the suffering and long term impacts of those having contracted the disease. Thus, this assessment of the near future for France focuses on casualties and the strain on the hospital system. It builds upon official statistics.
First we shall briefly outline the pandemic situation in France and explain how we built the statistical forecasts.
Then we shall focus on three scenarios. In the first scenario, the pandemic is controlled and contagion is progressively lowered. The second scenario considers that the measures are less efficient than thought and contagion stagnates at a relatively high level. Finally, a last scenario looks at a situation where contagion increases by only 2% a day. We added this scenario to highlight what could happen if things were going only slightly wrong.
In a last part we shall present three more points and reflections that emerged out of this work and deserve being highlighted. We shall point out discrepancies in statistical series and possible impacts. We shall then underline two points that deserve further research and may lead to ethical debates as well as to future tensions. First, hospitalised patients may be increasingly sorted out for access to reanimation, without the population being fully conscious of this phenomenon. Finally, the fate of old-age pensioners in their last days facing COVID-19 sufferings may need to be debated.
France faces the COVID-19 second wave
From denial to mobilisation or explosion?
As we saw with the comparative figures above, France is far from being in control of the pandemic and even further away from being exemplary in its handling of the COVID-19. As so many other countries, it muddles through.
- The Red Team Analysis Weekly – 19 November 2020
- The Red Team Analysis Weekly – 12 November 2020
- France and 3 Scenarios for the COVID-19 Second Wave
- Adapting to the Burning World?
- The Red Team Analysis Weekly – 5 November 2020
- The Red Team Analysis Weekly – 29 October 2020
- What is an Issue in terms of Strategic Foresight & Warning and Horizon Scanning?
- The Red Team Analysis Weekly – 22 October 2020
Yet, in France, efforts abound to minimise the seriousness of the pandemic, to deny any severity to the disease, as well as to refuse measures attempting to control the COVID-19, sometimes violently (e.g. France 3 Auvergne Rhône-Alpes, “C’est ça la réalité, si vous ne voulez pas l’entendre, sortez d’ici” : le coup de gueule d’Olivier Véran à l’Assemblée“, 4 November 2020; “Nicolas Bedos remonté contre Olivier Véran : “J’ai envie de le buter’“, Femmes Actuelles, 8 November 2020).
The economy continues being opposed to sanitary security, when both must go hand in hand (e.g. franceinfo, “André Comte-Sponville : ‘Par souci de justice, on ne peut pas sacrifier toute l’économie pour des raisons sanitaires‘”, 10 November 2020; Valentin Deleforterie et AFP, “Reconfinement : qui plaide pour la mesure et qui s’y oppose ?“, RTL, 28 October 2020).
Considering the acceleration of the second wave, French President Emmanuel Macron and its government played their role as political authorities with as primary mission the security of those who are ruled. They imposed a second lockdown on 30 October 2020, however, milder than the previous one (French President Emmanuel Macron, Adresse aux Français, 28 October 2020).
Almost two weeks into the lockdown, at long last, French policy-makers, as well as the media, appear to start trying to find a middle ground between being too reassuring and being too frightening (e.g. Franceinfo, “Covid-19 : ‘Le pic de l’épidémie est devant nous’, prévient Jérôme Salomon“, 9 November 2020; Le Figaro avec AFP, “Covid-19 en France : il y a un «frémissement» dans les chiffres“, 8 November 2020; FranceInter, “Olivier Véran constate un “ralentissement de la progression de l’épidémie“, bilan en fin de semaine, 8 November 2020). To reassure people too much would favour dangerous careless behaviour. Frightening people too much would create hopelessness and thus generate collective defeatism and depression.
Indeed, as we explained at the end of March, political authorities must fully ensure security, be it sanitary, economic, or related to civil disorders, to terrorist attacks and to foreign policy risks (Helene Lavoix, COVID-19 Antiviral Treatments and Scenarios, The Red Team Analysis Society, 30 March 2020). This is not a choice. An either/or situation is not an option. People cannot be asked to choose between one or the other type of security. An incapacity to deliver on all fronts will only lower the legitimacy of political authorities and then lead to even more difficulties in ensuring security. It is thus crucial to find the proper way to mobilise people in these difficult times.
In order to be able to deliver and mobilise, it is vital to know what lays ahead. This is true actually for all actors, as each needs also to develop its own response to the pandemic, while evaluating the decisions and actions taken by political authorities. It is here, in that space, that mobilisation develops.
Thus, are we truly seeing first signs of improvement in France? Will the lockdown truly end before Christmas? Which decisions should the French government take for the holiday season? Will they lead to an improvement of the situation and thus strengthen legitimacy or, on the contrary contribute to degrade both the overall security outlook and legitimacy?
Statistical trends and forecasts
Using the daily official statistics given by the French health authority (Prime Minsiter office and the data revised over time of Santé publique France and its observatory GEODES), we built a simple model that allows us to track data over time and forecast the trends ahead.
This is not an epidemiological model considering all factors. We only sought to focus on the most severe cases of the disease and on deaths, as well as on the impact on the French care system, while preparing the ground to look at the possible impact of what is now called Long Covid.
We built our model starting from the positive COVID19 cases, to which we apply first a rate to differentiate between asymptomatic and symptomatic cases. Then we apply the timeframe for incubation we identified in the scientific literature in a previous article (see Dynamics of contagion and the COVID-19 Second Wave). This gives us the daily number of symptomatic cases. Then, we apply rates we calculate out of the historical series as communicated by GEODES. We obtain first severe cases of COVID-19, i.e. those needing hospitalisation, which gives us the daily number of hospitalisation. Using the same system we work out the daily number of Intensive Care Units (ICUs), as well as daily deaths. As far as death outside the hospital system is concerned, i.e. in retirement home (EHPAD and EMS), we fully rely on the Prime Minister Office statistics and only project real death rates in the future.
This forecasting model allows us to capture broad trends. We used it to create three scenarios.
Three Scenarios for France, the COVID19 second wave and Christmas 2020
The true meaning of Christmas – Scenario 1
The first and most favourable scenario would be a world where a new spirit enters the population and where successfully fighting the COVID-19 becomes a shared objective. Meanwhile, innovation considering all aspects of security is actively sought. In that case, for one month we would have a 5% daily decrease of contagion. Seeing the positive results of their efforts, the population would feel vindicated in making even more efforts to further reduce contagion. As a result, the next month would see a daily 10% decrease of the number of positives cases.
The results for this scenario are shown on the three following charts.
The mass of people infected and then becoming symptomatic would still be very large.
The peak of the second wave, in daily terms would be reached between 13 and 19 November. Note that the peak on 13 November comes from a very high rate of contagion on 7 November (more than 86.000 cases), but is uncertain. Indeed we do not know if this rate resulted from data flows’ problems or reflected reality.
In cumulative terms, hospitalisations would peak around 19 November at approximately 38.500 beds and ICUs around 21 November at 5.846. The total expected cumulative deaths would reach 64.734 on 29 December 2020.
The French hospital system would thus, in general terms, cope. Nonetheless, these national estimates hide widely different situations across regions, which may negatively impact the overall situation, as shown in the animated graph below by BFMTV.
In this first scenario, and assuming there is no worsening of the situation in hospitals, and no increase in severity of the illness, it appears as possible to soften slightly the lockdown before Christmas. Yet, the French population would need to continue being extremely careful. Indeed it must not risk seeing again a worsening of the situation. Considering the characteristics of our scenario in terms of new collective understanding of the pandemic and related behaviour, this should not be a problem.
The hospitals will nonetheless still have to cope with a relatively heavy burden.
Many families will mourn over Christmas. Meanwhile the prospects for long covid illness need to be appraised. However, awareness to have once more overcome a pandemic wave, to have succeeded as a collective body creates the right conditions for mobilisation. French society can thus now boldly works towards truly innovating to live with the COVID-19, with other pandemics and even with the impacts of climate change.
A dreary never-ending struggle – Scenario 2
The second scenario considers that the lockdown is only partly effective considering the amount of denial and opposition in the country. In other terms, if a large majority of people respects the lockdown and the various sanitary measures, too many do not. That minority thus destroys the efforts of the majority.
To portray these behaviours, we estimated the level of continuing infections as equal to the average number of positive cases between 28 October and 10 November 2020. Because we are in the case of a pandemic where the natural growth of contagion – i.e. without any measures – is exponential, this scenario still shows a measure of success as well as the efforts of the population.
These efforts are, unfortunately, not sufficient, as shown on the three graphs below.
The mass of people infected and then becoming symptomatic would be very large and stabilise at a high level.
Here we do not have a peak for the second wave even though the oscillations of hospitalisations, ICUs and deaths are reduced over time around continuous lines at high levels. Daily, for as long as the contagion goes on, we would need to expect these high levels of hospitalisations, ICUs and deaths.
In cumulative terms, considering exits, hospitalisations would stabilise around approximately 35.000 to 36.000 beds, permanently used for COVID-19 patients. Between 5.000 and 5.200 ICUs beds would permanently be used for COVID-19 patients.
The system would be stretched thin but no explosion would take place during the period considered. Nonetheless, the total expected cumulative deaths would increase and reach 72.368 on 29 December 2020. What matters to understand here, compared with the previous scenario is that deaths would not stop but would continue rising with time, reaching approximately more than 80.000 deaths in mid-January.
In that case, the lockdown could not be relaxed before Christmas. It could not be eased either after Christmas. Indeed, the constant high pressure on hospitals, in terms of ICUs, would make very hazardous indeed any further relaxation of any measure.
This scenario may not be unlikely. It is also a very dangerous scenario.
Indeed, both a large part of the population and the medical staff would have to consent a long and sustained effort, with no end in sight. Meanwhile, the only reward would be deaths and even more deaths. More families would be bereaved with time, while increasingly more people would know and be surrounded by people grieving.
Most probably the resulting tension, grief, and disappointment too would make it increasingly unfeasible to sustain the type of efforts demanded. After a while, it is likely that people would give up making efforts, or become violent against those who do not. The medical staff would increasingly become unable to cope. As a result, contagion would spread again while the care system would start showing signs of breaking down. We would slide towards a situation looking increasingly more like the third scenario.
Soylent Green* – Scenario 3
This scenario looks at a situation where contagion is increasing by only 2% a day. There, those opposing the measures against the COVID-19 – whatever their rationale – have started taking the upper hand. Political authorities are either too weak to oppose creeping disobedience or finally decide to follow self-interested short-term lobbies with a policy of laissez-faire. Protective measures are still applied but increasingly less stringently. As a result, contagion slowly rises by what is thought initially to be something meaningless, 2% a day.
However, with time, 2% a day leads to a high daily level of new positive cases.
The mass of people infected and then becoming symptomatic would be very large and continue rising.
Here again, obviously, we do not have a peak for the second wave. The wave swells and lengthens becoming a never ending tide. Daily hospitalisations rise accordingly, as the number of ICUs and deaths, for as long as people refuse to consider the pandemic and the need for continuous protective measures.
In cumulative terms, hospitalisations never stabilise, but climb higher and higher. At the end of December around approximately 70.000 beds are used for COVID-19 patients, which represents 18% of the overall hospital beds in France (OECD (2020), Hospital beds (indicator). doi: 10.1787/0191328e-en – accessed on 11 November 2020). And the numbers keep rising.
More than 8.400 ICUs are occupied by COVID-19 patients. Hospitals struggle and desperately try to find new beds, while other pathologies cannot be handled anymore. Everyone knows that it is now a question of weeks before the whole system collapses. Harsh and stringent measures need to be taken, but, even in that case, the damage will be terrible and the care system may not survive.
The total cumulative deaths for the COVID-19 is expected to reach 78.724 on 29 December 2020. However, it only tells part of the story as deaths stemming from all other pathologies now have to be added to the count. What matters to understand here, compared with the two previous scenarios, is that first COVID-19 related deaths would not stop but would continue rising with time, while, second, pathologies that had been so far cured would become deadly again, adding to the overall number of deaths.
In that case, because of the choice made by people and elite groups, Christmas would probably not be spent under lockdown. It would however be everything save a celebration of hope, joy and love.
Hopefully, this scenario is unlikely. It is nonetheless useful as a scenario to remind people of why lockdown and other measures against the COVID-19 are necessary and must be embraced.
Further reflections and questions
Statistical updates and discrepancies on the number of COVID19 positive cases
Working on the statistics, we noticed there is a discrepancy between the figures given daily by the French officials, including the Prime Minsiter office and the data revised over time by Santé publique France and Géodes.
This is notably true on Sundays, when the numbers given by Geodes tend to be much smaller than those given by the PM office. We kept the data of Geodes for the sake of logic as we do not know how or why data are changed from one office to the other.
More interestingly in terms of steering policies, we noticed that Geodes revises its data for positive cases of COVID19 with time, probably as it gets the results of the tests. Data are updated sometimes over more than a month.
If changes are small, such as adding for example ten more positive cases on a specific day, this does not truly matter in terms of policy.
However, if changes are large, then the discrepancy can create similarly large problems.
We found out one such example. On 26 October 2020, 66.866 positive COVID cases were reported by Geodes, which showed a large increase (+23,28%) compared with the previous week, when the highest number of cases was 54.238 (on 23 October). After regular updates, these 66.866 cases became, on 9 November 2020, 68.453 cases.
Yet, on the PM website, the number of positive cases given for 26 October 2020 is 26.771.
The difference between 66.866 and 26.771 cases is immense. Such a huge discrepancy in figures makes it very difficult for external actors to steer their own policies.
If ever such discrepancies are also incorporated in the data given at the highest level of the French government, then policies may become inadequate.
Hopefully, this is only a problem of communication of data, but it does create doubts, which are unnecessary in these difficult times.
COVID19 patients die when they are not in ICUs
As we worked through our forecasts, considering what scientific papers, the WHO and medical doctors described about the course of the illness, we expected to see the daily number of deaths in hospital being more or less correlated with the number of patients having to be reanimated. We also expected that the number of deaths would be lower than the number of patients in ICUs. We knew, considering the variations in terms of length of intensive care that the correlation would most probably be imperfect.
Yet, allowing for a 8 days lag between the daily ICU numbers and daily deaths, we found rates of death compared with past daily ICUs numbers that were much higher than expected and also relatively erratic.
As a result, for the needs of a first crude statistical observation, we attempted to remove the lags between the different clinical events for which we had statistics. This would allow us to see for an hypothetical time the number of infections, the corresponding hospitalisations, reanimations and deaths. We obtain the following chart:
What we see is that, with time, as more patients enter the ICUs, around the beginning of October, the curve of deaths starts increasingly to match closely the curve of ICUs.
This would mean that all patients entering reanimation die. As this is not meant to be the case, one explanation we see for this phenomenon is that patients die as they are not in ICUs.
This means that patients either are not transferred in ICUs on time or that they are sorted out for access to reanimation. Because the French hospital system is, by large, not meant to have collapsed, then the probable explanation is that patients are sorted out.
Yet, despite the ethical considerations involved, there is no related debate in French society. Yet, the phenomenon was even more pronounced for the first wave.
Old-age pensioners die in retirement homes
As for the previous point this is an ethical concern that we feel compelled to highlight.
Throughout countries, the death of old-age pensioners in retirement homes because of the COVID-19 is an issue.
However, here, poring over statistics, time lags, the disease evolution and the sufferings it entails, we may only wonder about the way the physical sufferings of old-age pensioners are alleviated in retirement homes when these facilities were not designed to handle cases that in hospitals demand heavy treatment.
France, as most Western countries, upholds human rights as constitutive values. Certainly, wondering about and acting justly on the sufferings of its elders faced with a severe disease is part of human rights.
In one of the first articles we wrote on the COVID-19, we pointed out that a pandemic was not only about a virus but also about people and how they acted and reacted to the pandemic (Hélène Lavoix, “The Coronavirus COVID-19 Epidemic Outbreak is Not Only about a New Virus“, The Red Team Analysis Society, 12 February 2020). What increasingly emerges is that a pandemic also forces us to face who we are and who we want to be as individuals and, collectively, as a society. Only if we find pride in ourselves and in the justness of our actions and choices can we find the strength to live with the COVID-19.
*Soylent Green is a 1973 American dystopian film directed by Richard Fleischer, Produced by Walter Seltzer and Russell Thacher, starring Charlton Heston and Leigh Taylor-Young and distributed by Metro-Goldwyn-Mayer.