REPORTS

Cite as: H. Salje et al., Science

10.1126/science.abc3517 (2020).

Estimating the burden of SARS-CoV-2 in France

Henrik Salje1,2,3*, Cécile Tran Kiem1,4*, Noémie Lefrancq1, Noémie Courtejoie5, Paolo Bosetti1, Juliette Paireau1,6,
Alessio Andronico1, Nathanaël Hozé1, Jehanne Richet5, Claire-Lise Dubost5, Yann Le Strat6, Justin Lessler3,
Daniel Levy-Bruhl6, Arnaud Fontanet7,8, Lulla Opatowski9,10, Pierre-Yves Boelle11, Simon Cauchemez1†
1
Mathematical Modelling of Infectious Diseases Unit, Institut Pasteur, UMR2000, CNRS, Paris, France. 2Department of Genetics, University of Cambridge, Cambridge, UK.
3
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA. 4Collège Doctoral, Sorbonne Université, Paris, France. 5DREES,
Ministère des Solidarités et de la Santé, Paris, France. 6Santé Publique France, French National Public Health Agency, Saint-Maurice, France. 7Emerging Diseases
Epidemiology Unit, Institut Pasteur, Paris, France. 8PACRI Unit, Conservatoire National des Arts et Métiers, Paris, France. 9Epidemiology and Modelling of Antibiotic Evasion
Unit, Institut Pasteur, Paris, France. 10Anti-infective Evasion and Pharmacoepidemiology Team, CESP, Université Paris-Saclay, UVSQ, INSERM U1018, Montigny-le-
Bretonneux, France. 11Institut Pierre Louis d’Epidémiologie et de Santé Publique, Sorbonne Université, INSERM, Paris, France.
*These authors contributed equally to this work.
†Corresponding author. Email: simon.cauchemez@pasteur.fr

Downloaded from http://science.sciencemag.org/ on June 1, 2020

France has been heavily affected by the SARS-CoV-2 epidemic and went into lockdown on the 17 March
2020. Using models applied to hospital and death data, we estimate the impact of the lockdown and
current population immunity. We find 3.6% of infected individuals are hospitalized and 0.7% die, ranging
from 0.001% in those <20 years of age (ya) to 10.1% in those >80ya. Across all ages, men are more likely
to be hospitalized, enter intensive care, and die than women. The lockdown reduced the reproductive
number from 2.90 to 0.67 (77% reduction). By 11 May 2020, when interventions are scheduled to be eased,
we project 2.8 million (range: 1.8–4.7) people, or 4.4% (range: 2.8–7.2) of the population, will have been
infected. Population immunity appears insufficient to avoid a second wave if all control measures are
released at the end of the lockdown.

The worldwide pandemic of SARS-CoV-2, the coronavirus We present a suite of modeling analyses to characterize
which causes COVID-19, has resulted in unprecedented re- the dynamics of SARS-CoV-2 transmission in France and the
sponses, with many affected nations confining residents to impact of the lockdown on these dynamics. We elucidate the
their homes. Much like the rest of Europe, France has been risk of SARS-CoV-2 infection and severe outcomes by age and
hit hard by the epidemic and went into lockdown on the 17 sex and estimate the current proportion of the national and
March 2020. It was hoped that this would result in a sharp regional populations that have been infected and might be at
decline in ongoing spread, as was observed when China least temporarily immune (4). These models support
locked down following the initial emergence of the virus (1, healthcare planning of the French government by capturing
2). Following the expected reduction in cases, the French gov- hospital bed capacity requirements.
ernment has announced it will ease restrictions on the 11 May As of 7 May 2020, there were 95,210 incident hospitaliza-
2020. To exit from the lockdown without escalating infec- tions due to SARS-CoV-2 reported in France and 16,386
tions, we need to understand the underlying level of popula- deaths in hospitals, with the east of the country and the cap-
tion immunity and infection, identify those most at risk for ital, Paris, particularly affected (Fig. 1, A and B). The mean
severe disease and the impact of current control efforts. age of hospitalized patients was 68ya and the mean age of the
Daily reported numbers of hospitalizations and deaths deceased was 79ya with 50.0% of hospitalizations occurring
only provide limited insight into the state of the epidemic. in individuals >70ya and 81.6% of deaths within that age
Many people will either develop no symptoms or symptoms bracket; 56.2% of hospitalizations and 60.3% of deaths were
so mild they will not be detected through healthcare-based male (Fig. 1, C to E). To reconstruct the dynamics of all infec-
surveillance. The concentration of hospitalized cases in older tions, including mild ones, we jointly analyze French hospital
individuals has led to hypotheses that there may be wide- data with the results of a detailed outbreak investigation
spread “silent” transmission in younger individuals (3). If the aboard the Diamond Princess cruise ship where all passen-
majority of the population is infected, viral transmission gers were subsequently tested (719 infections, 14 deaths cur-
would slow, potentially reducing the need for the stringent rently). By coupling the passive surveillance data from
intervention measures currently employed. French hospitals with the active surveillance performed

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 1
aboard the Diamond Princess, we disentangle the risk of be- number of R0 = 3.0, it would require around 65% of the pop-
ing hospitalized in those infected from the underlying prob- ulation to be immune for the epidemic to be controlled by
ability of infection (5, 6). immunity alone. Our results therefore strongly suggest that,
We find that 3.6% of infected individuals are hospitalized without a vaccine, herd immunity on its own will be insuffi-
(95% CrI: 2.1–5.6), ranging from 0.2% (95% CrI: 0.1–0.2) in cient to avoid a second wave at the end of the lockdown. Ef-
females under <20ya to 45.9% (95% CrI: 27.2–70.9) in males ficient control measures need to be maintained beyond the 11
>80ya (Fig. 2A and table S1). Once hospitalized, on average May.
19.0% (95% CrI: 18.7–19.4%) patients enter ICU after a mean Our model can help inform the ongoing and future re-
delay of 1.5 days (fig. S1). We observe an increasing probabil- sponse to COVID-19. National ICU daily admissions have
ity of entering ICU with age—however, this drops for those gone from 700 at the end of March to 66 on 7 May. Hospital
>70ya (Fig. 2B and table S2). Overall, 18.1% (95% CrI: 17.8– admissions have declined from 3600 to 357 over the same
18.4) of hospitalized individuals go on to die (Fig. 2C). The time period, with consistent declines observed throughout
overall probability of death among those infected (the Infec- France (fig. S5). By 11 May we project 3900 (range: 2600–
tion Fatality Ratio, IFR) is 0.7% (95% CrI: 0.4–1.0), ranging 6300) daily infections across the country, down from between
from 0.001% in those under 20ya to 10.1% (95% CrI: 6.0–15.6) 150,000–390,000 immediately prior to the lockdown. At a re-

Downloaded from http://science.sciencemag.org/ on June 1, 2020

in those >80ya (Fig. 2D and table S2). Our estimate of overall gional level, we estimate that 58% of infections will be in Ile-
IFR is similar to other recent studies that found values of be- de-France and Grand Est combined. We find that the time
tween 0.5 and 0.7% for the Chinese epidemic (6–8). We find people spend in ICU appears to differ across the country,
men have a consistently higher risk than women of hospital- which may be due to differences in health care practices (ta-
ization (RR 1.25, 95% CrI: 1.22–1.29), ICU admission once hos- ble S5).
pitalized (RR: 1.61, 95% CrI: 1.56–1.67) and death following Using our modeling framework, we are able to reproduce
hospitalization (RR: 1.47, 95% CrI: 1.42–1.53) (fig. S2). the observed number of hospitalizations by age and sex in
We identify two clear subpopulations in those cases that France and the number of observed deaths aboard the Dia-
are hospitalized: individuals that die quickly upon hospital mond Princess (fig. S6). As a validation, our approach is also
admission (15% of fatal cases, mean time to death of 0.67 able to correctly identify parameters in simulated datasets
days) and individuals who die after longer time periods (85% where the true values are known (fig. S7). As cruise ship pas-
of fatal cases, mean time to death of 13.2 days) (fig. S3). The sengers may represent a different, healthier population than
proportion of fatal cases who die rapidly remains approxi- average French citizens, we run a sensitivity analysis where
mately constant across age-groups (fig. S4 and table S3). Po- Diamond Princess passengers are 25% less likely to die than
tential explanations for different subgroups of fatal cases French citizens (Fig. 4 and fig. S8). We also run sensitivity
include heterogeneous patterns of healthcare seeking, access analyses with longer delays between symptom onset and hos-
to care, underlying comorbidities, such as metabolic disease pital admission, missed infections aboard the Diamond Prin-
and other inflammatory conditions. A role for immunopath- cess, equal attack rates across all ages, reduced infectivity in
ogenesis has also been proposed (9–12). younger individuals, a contact matrix with unchanged struc-
We next fit national and regional transmission models to ture before/during the lockdown and one with very high iso-
ICU admission, hospital admission, and bed occupancy (both lation of elderly individuals during the lockdown. These
ICU and general wards) (Fig. 3, A to D, fig. S5, and tables S4 different scenarios result in mean IFRs from 0.5 to 0.9%, the
to S6), allowing for reduced age-specific daily contact pat- proportion of the population infected by the 11 May 2020
terns following the lockdown and changing patterns of ICU ranging from 1.7–8.9%, the number of daily infections at this
admission over time (fig. S17). We find that the basic repro- date ranging from 1700 to 9600 and a range of reproductive
ductive number R0 prior to the implementation of the lock- numbers post lockdown of 0.62–0.73 (Fig. 4, figs. S8 to S15,
down was 2.90 (95% CrI: 2.80–2.99). The lockdown resulted and tables S7 to S12).
in a 77% (95% CI: 76–78) reduction in transmission, with the A seroprevalence of 3% (range: 0–3%) has been estimated
reproduction number R dropping to 0.67 (95% CrI: 0.65– among blood donors in Hauts-de-France, which is consistent
0.68). We forecast that by the 11 May 2020, 2.8 million (range: with our model predictions (range: 1–3%) for this population
1.8–4.7, when accounting for uncertainty in the probability of if we account for a 10-day delay for seroconversion (13, 14).
hospitalization given infection) people will have been in- Future additional serological data will help further refine es-
fected, representing 4.4% (range: 2.8–7.2) of the French pop- timates of the proportion of the population infected.
ulation (Fig. 3E). This proportion will be 9.9% (range: 6.6– While we focus on deaths occurring in hospitals, there are
15.7) in Ile-de-France, which includes Paris, and 9.1% (range: also non-hospitalized COVID-19 deaths, including >9000 in
6.0–14.6) in Grand Est, the two most affected regions of the retirement homes in France (15). We explicitly removed re-
country (Fig. 3F and fig. S5). Assuming a basic reproductive tirement home population from our analyses as transmission

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 2
dynamics may be different in these closed populations. This (COVID-19) using age-adjusted data from the outbreak on the Diamond Princess
means our estimates of immunity in the general population cruise ship, February 2020. Euro Surveill. 25, 2000256 (2020).
doi:10.2807/1560-7917.ES.2020.25.12.2000256 Medline
are unaffected by deaths in retirement homes, however, in
the event of large numbers of non-hospitalized deaths in the 8. K. Mizumoto, K. Kagaya, G. Chowell, Early epidemiological assessment of the
transmission potential and virulence of coronavirus disease 2019 (COVID-19) in
wider community, we would underestimate the proportion of Wuhan City: China, January-February, 2020. medRxiv 2020.02.12.20022434
the population infected. Analyses of excess death will be im- [Preprint]. 13 March 2020. https://doi.org/10.1101/2020.02.12.20022434.
portant to explore these issues.
9. L. Peeples, News Feature: Avoiding pitfalls in the pursuit of a COVID-19 vaccine.
This study shows the massive impact the French lock- Proc. Natl. Acad. Sci. U.S.A. 117, 8218–8221 (2020).
down had on SARS-CoV-2 transmission. Our modeling ap- doi:10.1073/pnas.2005456117 Medline
proach has allowed us to estimate underlying probabilities of 10. D. Ricke, R. W. Malone, Medical Countermeasures Analysis of 2019-nCoV and
infection, hospitalization and death, which is essential for the Vaccine Risks for Antibody-Dependent Enhancement (ADE), 27 February 2020;
interpretation of COVID-19 surveillance data. The forecasts https://ssrn.com/abstract=3546070.
we provide can inform lockdown exit strategies. Our esti- 11. J. Yang, Y. Zheng, X. Gou, K. Pu, Z. Chen, Q. Guo, R. Ji, H. Wang, Y. Wang, Y. Zhou,
mates of a low level of immunity against SARS-CoV-2 indi- Prevalence of comorbidities and its effects in patients infected with SARS-CoV-2:
cates that efficient control measures that limit transmission A systematic review and meta-analysis. Int. J. Infect. Dis. 94, 91–95 (2020).

Downloaded from http://science.sciencemag.org/ on June 1, 2020

doi:10.1016/j.ijid.2020.03.017 Medline
risk will have to be maintained beyond the 11 May 2020 to
avoid a rebound of the epidemic. 12. M. Bolles, D. Deming, K. Long, S. Agnihothram, A. Whitmore, M. Ferris, W.
Funkhouser, L. Gralinski, A. Totura, M. Heise, R. S. Baric, A double-inactivated
severe acute respiratory syndrome coronavirus vaccine provides incomplete
REFERENCES AND NOTES protection in mice and induces increased eosinophilic proinflammatory
pulmonary response upon challenge. J. Virol. 85, 12201–12215 (2011).
1. M. U. G. Kraemer, C.-H. Yang, B. Gutierrez, C.-H. Wu, B. Klein, D. M. Pigott, L. du doi:10.1128/JVI.06048-11 Medline
Plessis, N. R. Faria, R. Li, W. P. Hanage, J. S. Brownstein, M. Layan, A. Vespignani,
H. Tian, C. Dye, O. G. Pybus, S. V. Scarpino; Open COVID-19 Data Working Group, 13. A. Fontanet, L. Tondeur, Y. Madec, R. Grant, C. Besombes, N. Jolly, S. F. Pellerin,
The effect of human mobility and control measures on the COVID-19 epidemic in M.-N. Ungeheuer, I. Cailleau, L. Kuhmel, S. Temmam, C. Huon, K.-Y. Chen, B.
China. Science 368, 493–497 (2020). doi:10.1126/science.abb4218 Medline Crescenzo, S. Munier, C. Demeret, L. Grzelak, I. Staropoli, T. Bruel, P. Gallian, S.
Cauchemez, S. van der Werf, O. Schwartz, M. Eloit, B. Hoen, Cluster of COVID-19
2. H. Tian, Y. Liu, Y. Li, C.-H. Wu, B. Chen, M. U. G. Kraemer, B. Li, J. Cai, B. Xu, Q. in northern France: A retrospective closed cohort study. medRxiv
Yang, B. Wang, P. Yang, Y. Cui, Y. Song, P. Zheng, Q. Wang, O. N. Bjornstad, R. 2020.04.18.20071134 [Preprint]. 23 April 2020.
Yang, B. T. Grenfell, O. G. Pybus, C. Dye, An investigation of transmission control https://doi.org/10.1101/2020.04.18.20071134.
measures during the first 50 days of the COVID-19 epidemic in China. Science
368, 638–642 (2020). doi:10.1126/science.abb6105 Medline 14. L. Grzelak, S. Temmam, C. Planchais, C. Demeret, C. Huon, F. Guivel, I. Staropoli,
M. Chazal, J. Dufloo, D. Planas, J. Buchrieser, M. M. Rajah, R. Robinot, F. Porrot,
3. J. Lourenço, R. Paton, M. Ghafari, M. Kraemer, C. Thompson, P. Simmonds, P. M. Albert, K.-Y. Chen, B. Crescenzo, F. Donati, F. Anna, P. Souque, M. Gransagne,
Klenerman, S. Gupta, Fundamental principles of epidemic spread highlight the J. Bellalou, M. Nowakowski, M. Backovic, L. Bouadma, L. Le Fevre, Q. Le Hingrat,
immediate need for large-scale serological surveys to assess the stage of the D. Descamps, A. Pourbaix, Y. Yazdanpanah, L. Tondeur, C. Besombes, M.-N.
SARS-CoV-2 epidemic. medRxiv 2020.03.24.20042291 [Preprint]. 26 March Ungeheuer, G. Mellon, P. Morel, S. Rolland, F. Rey, S. Behillil, V. Enouf, A.
2020. https://doi.org/10.1101/2020.03.24.20042291. Lemaitre, M.-A. Creach, S. Petres, N. Escriou, P. Charneau, A. Fontanet, B. Hoen,
4. L. Bao, W. Deng, H. Gao, C. Xiao, J. Liu, J. Xue, Q. Lv, J. Liu, P. Yu, Y. Xu, F. Qi, Y. T. Bruel, M. Eloit, H. Mouquet, O. Schwartz, S. van der Werf, SARS-CoV-2
Qu, F. Li, Z. Xiang, H. Yu, S. Gong, M. Liu, G. Wang, S. Wang, Z. Song, W. Zhao, Y. serological analysis of COVID-19 hospitalized patients, pauci-symptomatic
Han, L. Zhao, X. Liu, Q. Wei, C. Qin, Reinfection could not occur in SARS-CoV-2 individuals and blood donors. medRxiv 2020.04.21.20068858 [Preprint]. 24
infected rhesus macaques. bioRxiv 2020.03.13.990226 [Preprint]. 14 March April 2020. https://doi.org/10.1101/2020.04.21.20068858.
2020. https://doi.org/10.1101/2020.03.13.990226. 15. French Government website, Info Coronavirus Covid 19 (in French);
5. J. Lessler, H. Salje, M. D. Van Kerkhove, N. M. Ferguson, S. Cauchemez, I. www.gouvernement.fr/info-coronavirus/carte-et-donnees.
Rodriquez-Barraquer, R. Hakeem, T. Jombart, R. Aguas, A. Al-Barrak, D. A. T. 16. H. Salje, C. Tran Kiem, Code and data for: Estimating the burden of SARS-CoV-2
Cummings; MERS-CoV Scenario and Modeling Working Group, Estimating the in France, Version 1.0, Zenodo (2020); http://doi.org/10.5281/zenodo.3813815.
severity and subclinical burden of Middle East respiratory syndrome coronavirus
infection in the Kingdom of Saudi Arabia. Am. J. Epidemiol. 183, 657–663 (2016). 17. Field Briefing: Diamond Princess COVID-19 Cases, 20 Feb Update (2020);
doi:10.1093/aje/kwv452 Medline www.niid.go.jp/niid/en/2019-ncov-e/9417-covid-dp-fe-02.html.
6. R. Verity, L. C. Okell, I. Dorigatti, P. Winskill, C. Whittaker, N. Imai, G. Cuomo- 18. Ministry of Health, Labour and Welfare, About new coronavirus infection (in
Dannenburg, H. Thompson, P. G. T. Walker, H. Fu, A. Dighe, J. T. Griffin, M. Japanese);
Baguelin, S. Bhatia, A. Boonyasiri, A. Cori, Z. Cucunubá, R. FitzJohn, K. www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000164708_00001.html.
Gaythorpe, W. Green, A. Hamlet, W. Hinsley, D. Laydon, G. Nedjati-Gilani, S.
Riley, S. van Elsland, E. Volz, H. Wang, Y. Wang, X. Xi, C. A. Donnelly, A. C. Ghani, 19. H. Nishiura, D. Klinkenberg, M. Roberts, J. A. P. Heesterbeek, Early
N. M. Ferguson, Estimates of the severity of coronavirus disease 2019: A model- epidemiological assessment of the virulence of emerging infectious diseases: A
based analysis. Lancet Infect. Dis. 10.1016/S1473-3099(20)30243-7 (2020). case study of an influenza pandemic. PLOS ONE 4, e6852 (2009).
doi:10.1016/S1473-3099(20)30243-7 Medline doi:10.1371/journal.pone.0006852 Medline

7. T. W. Russell, J. Hellewell, C. I. Jarvis, K. van Zandvoort, S. Abbott, R. Ratnayake, S. 20. G. Béraud, S. Kazmercziak, P. Beutels, D. Levy-Bruhl, X. Lenne, N. Mielcarek, Y.
Flasche, R. M. Eggo, W. J. Edmunds, A. J. Kucharski; Cmmid Covid-Working Yazdanpanah, P.-Y. Boëlle, N. Hens, B. Dervaux, The French connection: The first
Group, Estimating the infection and case fatality ratio for coronavirus disease large population-based contact survey in France relevant for the spread of

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 3
 infectious diseases. PLOS ONE 10, e0133203 (2015). permits unrestricted use, distribution, and reproduction in any medium,
doi:10.1371/journal.pone.0133203 Medline provided the original work is properly cited. To view a copy of this license, visit
https://creativecommons.org/licenses/by/4.0/. This license does not apply to
21. K. Mizumoto, K. Kagaya, A. Zarebski, G. Chowell, Estimating the asymptomatic figures/photos/artwork or other content included in the article that is credited
proportion of coronavirus disease 2019 (COVID-19) cases on board the Diamond to a third party; obtain authorization from the rights holder before using such
Princess cruise ship, Yokohama, Japan, 2020. Euro Surveill. 25, (2020). material.
doi:10.2807/1560-7917.ES.2020.25.10.2000180 Medline
SUPPLEMENTARY MATERIALS
22. Stan Development Team, RStan: the R interface to Stan (2020); https://mc- science.sciencemag.org/cgi/content/full/science.abc3517/DC1
stan.org/. Materials and Methods
23. Z. Du, X. Xu, Y. Wu, L. Wang, B. J. Cowling, L. A. Meyers, Serial interval of COVID- Supplementary Text
19 among publicly reported confirmed cases. Emerg. Infect. Dis. Figs. S1 to S17
10.3201/eid2606.200357 (2020). doi:10.3201/eid2606.200357 Medline Tables S1 to S12
References (17–32)
24. Q. Bi, Y. Wu, S. Mei, C. Ye, X. Zou, Z. Zhang, X. Liu, L. Wei, S. A. Truelove, T. MDAR Reproducibility Checklist
Zhang, W. Gao, C. Cheng, X. Tang, X. Wu, Y. Wu, B. Sun, S. Huang, Y. Sun, J.
Zhang, T. Ma, J. Lessler, T. Feng, Epidemiology and Transmission of COVID-19 in 20 April 2020; accepted 11 May 2020
Shenzhen China: Analysis of 391 cases and 1,286 of their close contacts. Published online 13 May 2020
medRxiv 2020.03.03.20028423 [Preprint]. 4 March 2020. 10.1126/science.abc3517

Downloaded from http://science.sciencemag.org/ on June 1, 2020

https://doi.org/10.1101/2020.03.03.20028423.
25. L. Tindale, M. Coombe, J. E. Stockdale, E. Garlock, W. Y. V. Lau, M. Saraswat, Y.-
H. B. Lee, L. Zhang, D. Chen, J. Wallinga, C. Colijn, Transmission interval
estimates suggest pre-symptomatic spread of COVID-19. medRxiv
2020.03.03.20029983 [Preprint]. 6 March 2020.
https://doi.org/10.1101/2020.03.03.20029983.
26. S. Funk, socialmixr; https://github.com/sbfnk/socialmixr.
27. O. Diekmann, J. A. Heesterbeek, J. A. Metz, On the definition and the
computation of the basic reproduction ratio R0 in models for infectious diseases
in heterogeneous populations. J. Math. Biol. 28, 365–382 (1990).
doi:10.1007/BF00178324 Medline
28. N. Hens, G. M. Ayele, N. Goeyvaerts, M. Aerts, J. Mossong, J. W. Edmunds, P.
Beutels, Estimating the impact of school closure on social mixing behaviour and
the transmission of close contact infections in eight European countries. BMC
Infect. Dis. 9, 187 (2009). doi:10.1186/1471-2334-9-187 Medline
29. Décret n° 2020-260 du 16 mars 2020 portant réglementation des déplacements
dans le cadre de la lutte contre la propagation du virus covid-19, Legifrance;
www.legifrance.gouv.fr/affichTexte.do?cidTexte=JORFTEXT000041728476&ca
tegorieLien=id.
30. A. Gelman, J. B. Carlin, H. S. Stern, D. B. Rubin, Bayesian Data Analysis (CRC
Texts in Statistical Science, Chapman and Hall, ed. 2, 2004).
31. COVID-19 Community Mobility Report, www.google.com/covid19/mobility/.
32. J. Wallinga, M. Lipsitch, How generation intervals shape the relationship between
growth rates and reproductive numbers. Proc. Biol. Sci. 274, 599–604 (2007).
doi:10.1098/rspb.2006.3754 Medline
ACKNOWLEDGMENTS
Funding: We acknowledge financial support from the Investissement d’Avenir
program, the Laboratoire d’Excellence Integrative Biology of Emerging Infectious
Diseases program (Grant ANR-10-LABX-62-IBEID), Santé Publique France, the
INCEPTION project (PIA/ANR-16-CONV-0005) and the European Union’s
Horizon 2020 research and innovation program under grant agreement No.
101003589. HS acknowledges support from the European Research Council (No.
804744) and University of Cambridge COVID-19 Rapid Response Grant. Author
contributions: H.S., C.T.K. and S.C. conceived the study, developed the
methods, performed analyses and co-wrote the paper. N.L., N.C., N.H., A.A., P.B.,
J.P., J.R., C.-L.D., L.O., P.Y.B., A.F., J.L., D.L.B., Y.L.S. contributed to data
collection and analysis. All authors contributed to paper revisions. Competing
interests: The authors declare no competing interests. Data and materials
availability: Code for the paper is available at (16).This work is licensed under a
Creative Commons Attribution 4.0 International (CC BY 4.0) license, which

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 4
 Downloaded from http://science.sciencemag.org/ on June 1, 2020
Fig. 1. COVID-19 hospitalizations and deaths in France. (A) Cumulative
number of general ward and ICU hospitalizations, ICU admissions and
deaths from SARS-CoV-2 in France. The green line indicates the time when
the lockdown was put in place in France. (B) Distribution of deaths in
France. Number of (C) hospitalizations, (D) ICU and (E) deaths by age group
and sex in France.

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 5
 Downloaded from http://science.sciencemag.org/ on June 1, 2020
Fig. 2. Probabilities of hospitalization, ICU admittance and death.
(A) Probability of hospitalization among those infected as a function of
age and sex. (B) Probability of ICU admission among those hospitalized
as a function of age and sex. (C) Probability of death among those
hospitalized as a function of age and sex. (D) Probability of death among
those infected as a function of age and sex. For each panel, the black line
and grey shaded region represents the overall mean across all ages. The
boxplots represent the 2.5, 25, 50, 75 and 97.5 percentiles of the
posterior distributions.

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 6
 Downloaded from http://science.sciencemag.org/ on June 1, 2020
Fig. 3. Time course of the SARS-CoV-2 epidemic to 11 May 2020. (A) Daily
admissions in ICU in metropolitan France. (B) Number of ICU beds occupied
in metropolitan France. (C) Daily hospital admissions in metropolitan
France. (D) Number of general ward beds occupied in metropolitan France
(E) Daily new infections in metropolitan France (logarithmic scale).
(F) Predicted proportion of the population infected by 11 May 2020 for each
of the 13 regions in metropolitan France. (G) Predicted proportion of the
population infected in metropolitan France. The black circles in panels (A),
(B), (C) and (D) represent hospitalization data used for the calibration and
the open circles hospitalization data that were not used for calibration. The
dotted lines in panels (E) and (G) represent the 95% uncertainty range
stemming from the uncertainty in the probability of hospitalization following
infection.

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 7
 Downloaded from http://science.sciencemag.org/ on June 1, 2020
Fig. 4. Sensitivity analyses considering different modeling assumptions.
(A) Infection fatality rate (%). (B) Estimated reproduction numbers before
(R0) and during lockdown (Rlockdown). (C) Predicted daily new infections on 11
May. (D) Predicted proportion of the population infected by 11 May. The
different scenarios correspond to: Children less inf. – Individuals <20ya are
half as infectious as adults; No Change CM – the structure of the contact
matrix is not modified by the lockdown; CM SDE – Contact matrix after
lockdown with very high social distancing of the elderly; Constant AR –
Attack rates are constant across age groups; Higher IFR – French people
25% more likely to die than Diamond Princess passengers; Higher AR DP –
25% of the infections were undetected on the Diamond Princess cruise ship;
Delay Distrib – Single hospitalization to death delay distribution; Higher
delay to hosp – 8 days on average between symptoms onset and
hospitalization for patients who will require an ICU admission and 9 days on
average between symptoms onset and hospitalization for the patients who
will not. For estimates of IFR and reproduction numbers before and during
lockdown, we report 95% credible intervals. For estimates of daily new
infections and proportion of the population infected by 11 May, we report the
95% uncertainty range stemming from the uncertainty in the probability of
hospitalization given infection.

First release: 13 May 2020 www.sciencemag.org (Page numbers not final at time of first release) 8
Estimating the burden of SARS-CoV-2 in France
Henrik Salje, Cécile Tran Kiem, Noémie Lefrancq, Noémie Courtejoie, Paolo Bosetti, Juliette Paireau, Alessio Andronico, Nathanaël
Hozé, Jehanne Richet, Claire-Lise Dubost, Yann Le Strat, Justin Lessler, Daniel Levy-Bruhl, Arnaud Fontanet, Lulla Opatowski,
Pierre-Yves Boelle and Simon Cauchemez

published online May 13, 2020

Downloaded from http://science.sciencemag.org/ on June 1, 2020

ARTICLE TOOLS http://science.sciencemag.org/content/early/2020/05/12/science.abc3517

SUPPLEMENTARY http://science.sciencemag.org/content/suppl/2020/05/12/science.abc3517.DC1
MATERIALS

RELATED http://stm.sciencemag.org/content/scitransmed/12/541/eabb5883.full
CONTENT
http://stm.sciencemag.org/content/scitransmed/12/534/eabb1469.full
http://stm.sciencemag.org/content/scitransmed/11/499/eaat0360.full
http://stm.sciencemag.org/content/scitransmed/9/396/eaal3653.full
REFERENCES This article cites 15 articles, 4 of which you can access for free
http://science.sciencemag.org/content/early/2020/05/12/science.abc3517#BIBL

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Use of this article is subject to the Terms of Service

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement of
Science, 1200 New York Avenue NW, Washington, DC 20005. The title Science is a registered trademark of AAAS.
Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science.
No claim to original U.S. Government Works