%0 Journal Article %T An ensemble model based on early predictors to forecast COVID-19 health care demand in France %+ Modélisation mathématique des maladies infectieuses - Mathematical modelling of Infectious Diseases %+ Direction des maladies infectieuses - Infectious Diseases Division [Saint-Maurice] %+ Centre de Recherches sur l'Action Politique en Europe (ARENES) %+ Recherche sur les services et le management en santé (RSMS) %+ Predict Services [Castelnau-le-Lez] %+ Modélisation en pharmacologie de population (XPOP) %+ Institut Polytechnique de Paris (IP Paris) %+ Institut Pierre Louis d'Epidémiologie et de Santé Publique (iPLESP) %A Paireau, Juliette %A Andronico, Alessio %A El Hoz, Nathanaël %A Layan, Maylis %A Crepey, Pascal %A Roumagnac, Alix %A Lavielle, Marc %A Boëlle, Pierre-Yves %A Cauchemez, Simon %Z 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/ANR16-CONV-0005), the European Union’s Horizon 2020 research and innovation program under grants 101003589(RECOVER) and 874735 (VEO), AXA, Groupama, and EMERGEN %< avec comité de lecture %@ 0027-8424 %J Proceedings of the National Academy of Sciences of the United States of America %I National Academy of Sciences %V 119 %N 18 %P e2103302119 %8 2022-05-03 %D 2022 %R 10.1073/pnas.2103302119 %M 35476520 %K COVID-19 %K ensemble model %K forecasting %Z Life Sciences [q-bio]/Santé publique et épidémiologieJournal articles %X Significance The COVID-19 pandemic is inducing significant stress on health care structures, which can be quickly saturated with negative consequences for patients. As hospitalization comes late in the infection history of a patient, early predictors—such as the number of cases, mobility, climate, and vaccine coverage—could improve forecasts of health care demand. Predictive models taken individually have their pros and cons, and it is advantageous to combine the predictions in an ensemble model. Here, we design an ensemble that combines several models to anticipate French COVID-19 health care needs up to 14 days ahead. We retrospectively test this model, identify the best predictors of the growth rate of hospital admissions, and propose a promising approach to facilitate the planning of hospital activity. %G English %2 https://pasteur.hal.science/pasteur-03690824v3/document %2 https://pasteur.hal.science/pasteur-03690824v3/file/Paireau_PNAS_2022_119_18.pdf %L pasteur-03690824 %U https://pasteur.hal.science/pasteur-03690824 %~ INSERM %~ PASTEUR %~ X %~ UNIV-RENNES1 %~ CNRS %~ INRIA %~ INRIA-SACLAY %~ CRAPE %~ SANTE_PUB_INSERM %~ X-CMAP %~ X-DEP %~ X-DEP-MATHA %~ OPENAIRE %~ INRIA_TEST %~ TESTALAIN1 %~ CMAP %~ IPLESP %~ INRIA2 %~ CAMPUS-AAR %~ AAI %~ UR1-HAL %~ EHESP %~ UR1-SHS %~ UNIV-PARIS-SACLAY %~ UR1-DROIT %~ EHESP-ARENES %~ TEST-UR-CSS %~ UNIV-RENNES %~ SORBONNE-UNIVERSITE %~ SORBONNE-UNIV %~ SU-MEDECINE %~ IP_PARIS %~ IP_PARIS_COPIE %~ UNIV-PARIS %~ UNIVERSITE-PARIS %~ TEST-HALCNRS %~ SU-TI %~ ANR %~ UR1-DEGSP %~ GS-COMPUTER-SCIENCE %~ ALLIANCE-SU %~ RSMS %~ TEST3-HALCNRS %~ TEST4-HALCNRS %~ MATH-MODEL-INFECT-DISEASES %~ PASTEUR_UMR2000 %~ TEST5-HALCNRS