pith. sign in

arxiv: 2606.07413 · v1 · pith:6BU67G6Dnew · submitted 2026-06-05 · 🧮 math.OC · q-bio.PE

A Nine-Compartment Nonlinear Epidemic Model with Spline-Based Identification of Time-Varying Transmission and Vaccination Dynamics: Application to the COVID-19 Third Wave in Italy

Lokman Rachid Melhani , Antonino Sferlazza , Lars Gr\"une , Dominique Persano Adorno , Filippo D'Ippolito , Omar Enzo Santangelo , Ivan Marchese , Antonino Lo Burgio
show 1 more author
Alberto Firenze
This is my paper

Pith reviewed 2026-06-27 21:07 UTC · model grok-4.3

classification 🧮 math.OC q-bio.PE
keywords epidemic modelCOVID-19time-varying parametersPCHIP splinestability analysisvaccination dynamicsItaly third wavereproduction number
0
0 comments X

The pith

A nine-compartment model with spline-fitted time-varying rates reproduces Italian COVID-19 third wave data and establishes conditions for epidemic decay.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper presents a nonlinear epidemic model with nine compartments that tracks two co-circulating strains, a super-spreader group, waning vaccine immunity, and explicit hospitalization and mortality flows. Time-varying transmission and vaccination rates are identified from January-May 2021 Italian data by parameterizing them with piecewise cubic Hermite interpolating polynomials, which reduces the fitting task to a fourteen-variable constrained optimization problem solved by sequential quadratic programming. The calibrated model matches active hospitalizations, cumulative fatalities, and cumulative vaccinations with high accuracy. Analytical work establishes well-posedness, supplies a closed-form basic reproduction number, proves local and global stability of the disease-free equilibrium, and gives a threshold condition guaranteeing epidemic decay whenever the effective reproduction number stays persistently below one. Sensitivity analysis ranks hospital throughput parameters above the transmission rate.

Core claim

After calibration the model reproduces active hospitalizations (R^2=0.966), cumulative fatalities (R^2=0.987), and cumulative vaccinations (R^2=0.999); the disease-free equilibrium is globally asymptotically stable when the effective reproduction number remains below one, and a sufficient threshold condition proves epidemic decay under persistent time-varying suppression; sensitivity consistently ranks hospital-related parameters above transmission rate, showing that reactive containment cannot prevent a hospitalization peak once the pre-existing latent viral load is fixed.

What carries the argument

PCHIP control-node parameterization of the time-varying transmission and vaccination rates, which converts their identification into a fourteen-variable SQP problem inside the nine-compartment structure.

If this is right

  • Whenever the effective reproduction number remains persistently below one the epidemic decays to the disease-free equilibrium.
  • Hospital throughput parameters exert stronger influence on outcomes than the transmission rate itself.
  • Reactive containment measures cannot avert an already-determined hospitalization peak driven by pre-existing latent infections.
  • The PCHIP parameterization converges to the true time-varying functions at rate O(h^2) as node spacing vanishes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same spline-based identification procedure could be applied to other countries or later epidemic waves with comparable data.
  • Policy emphasis could shift toward hospital capacity expansion rather than broad transmission suppression once the sensitivity ranking is accepted.
  • Direct testing of the threshold condition could be performed with data from subsequent waves or different interventions.

Load-bearing premise

The time-varying transmission and vaccination rates are accurately represented by the PCHIP control-node parameterization and the January-May 2021 Italian data accurately reflects the underlying dynamics without major reporting or testing biases.

What would settle it

If the epidemic fails to decay after the effective reproduction number remains persistently below one for a sustained interval, or if out-of-sample predictions from the fitted model diverge substantially from later observed hospitalization and fatality counts.

Figures

Figures reproduced from arXiv: 2606.07413 by Alberto Firenze, Antonino Lo Burgio, Antonino Sferlazza, Dominique Persano Adorno, Filippo D'Ippolito, Ivan Marchese, Lars Gr\"une, Lokman Rachid Melhani, Omar Enzo Santangelo.

Figure 1
Figure 1. Figure 1: Transition diagram of the nine-compartment epidemic model. Solid arrows [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: COVID-19 pandemic trajectory in Italy (February 2020 to January 2025). [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Calibration results for the Italian Third Wave (January to May 2021). Top-left: [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Time evolution of all nine compartments over the January to May 2021 window. [PITH_FULL_IMAGE:figures/full_fig_p017_4.png] view at source ↗
read the original abstract

We develop a nine-compartment nonlinear epidemic model incorporating two co-circulating viral strains (ancestral I1 and the Alpha variant B.1.1.7 I2, which is 43-90% more transmissible, c2=1.5), a super-spreader subpopulation, partial vaccine-induced immunity with waning, and explicit hospitalization dynamics with differentiated mortality. Transmission and vaccination rates are treated as time-varying control inputs and identified from Italian COVID-19 data (January-May 2021) via a Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) control-node parameterization, reducing calibration to a fourteen-variable Sequential Quadratic Programming (SQP) problem with monotonicity and box constraints. A parametric bootstrap (n=1000) quantifies parameter uncertainty. The calibrated model achieves R^2=0.966 for active hospitalizations, R^2=0.987 for cumulative fatalities, and R^2=0.999 for cumulative vaccinations. Well-posedness, the basic reproduction number in closed form, and local and global stability of the disease-free equilibrium are established analytically. An L-infinity approximation error bound shows that the PCHIP control-node parameterization converges to the true time-varying parameters at rate O(h^2) as the node spacing vanishes. Local identifiability and a noise stability bound are established via the Fisher information matrix. A sufficient threshold condition proves epidemic decay under time-varying suppression whenever the effective reproduction number remains persistently below one. Sensitivity analyses consistently rank hospital throughput parameters above the transmission rate, providing a mathematical basis for the observation that reactive containment measures cannot prevent a hospitalization peak already driven by the pre-existing latent viral load.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript develops a nine-compartment nonlinear epidemic model incorporating two co-circulating strains (ancestral and Alpha variant with c2=1.5), a super-spreader subpopulation, waning vaccine-induced immunity, and differentiated hospitalization/mortality compartments. Transmission and vaccination rates are treated as time-varying inputs identified from Italian COVID-19 data (January-May 2021) via PCHIP control-node parameterization, reducing calibration to a 14-variable constrained SQP problem. The calibrated model reports R²=0.966 for active hospitalizations, R²=0.987 for cumulative fatalities, and R²=0.999 for cumulative vaccinations. Analytical contributions include well-posedness, closed-form basic reproduction number, local/global stability of the disease-free equilibrium, an L∞ error bound of O(h²) for the PCHIP approximation, Fisher-information-based local identifiability and noise stability, a sufficient threshold condition for epidemic decay when the effective reproduction number stays below one, and sensitivity analysis ranking hospital-throughput parameters above transmission rate.

Significance. If the analytical derivations and calibration hold, the work supplies a mathematically rigorous multi-strain compartmental framework with explicit time-varying control identification, stability thresholds, and identifiability results that could support policy analysis of containment timing and hospital capacity. The sensitivity ranking and threshold condition offer concrete, testable implications beyond standard models.

major comments (3)
  1. [Abstract/Results] Abstract and calibration/results sections: The reported R² values (0.966, 0.987, 0.999) are obtained by optimizing the 14 SQP parameters directly to the same dataset used for evaluation; this renders the metrics tautological by construction and does not establish out-of-sample predictive validity or robustness to data biases.
  2. [Model formulation/Calibration] Model formulation and calibration sections: The claim that the PCHIP parameterization with 14 variables and monotonicity/box constraints accurately recovers the underlying time-varying rates rests on the unverified assumption that Italian reporting data (January-May 2021) contain no major testing or ascertainment biases; no cross-validation or alternative parameterization comparison is described to support this.
  3. [Stability/Threshold analysis] Stability and threshold sections: The local/global stability proofs and the sufficient condition for decay (effective reproduction number persistently below one) are derived in terms of the fitted time-varying parameters; it is unclear whether the parametric bootstrap (n=1000) uncertainty propagates to alter the threshold or stability conclusions.
minor comments (2)
  1. [Model formulation] Clarify the precise mapping from the stated 43-90% transmissibility increase to the fixed value c2=1.5 and whether this parameter is held constant or allowed to vary.
  2. [Error bound/Identifiability] The L∞ error bound and Fisher-information steps are stated as routine but would benefit from explicit equation references or a short appendix derivation to aid verification.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the thorough and constructive comments. We address each major comment below, clarifying the manuscript's scope and indicating where revisions will strengthen the presentation.

read point-by-point responses

  1. Referee: [Abstract/Results] Abstract and calibration/results sections: The reported R² values (0.966, 0.987, 0.999) are obtained by optimizing the 14 SQP parameters directly to the same dataset used for evaluation; this renders the metrics tautological by construction and does not establish out-of-sample predictive validity or robustness to data biases.

    Authors: The reported R² values are in-sample goodness-of-fit measures obtained directly from the SQP calibration on the January-May 2021 data. The manuscript's primary aim is to show that the nine-compartment model with constrained PCHIP parameterization can reproduce the observed trajectories and to derive its analytical properties; no out-of-sample predictive claims are made. We will revise the abstract and results sections to state explicitly that these are calibration fits and to distinguish them from predictive validation. revision: yes

  2. Referee: [Model formulation/Calibration] Model formulation and calibration sections: The claim that the PCHIP parameterization with 14 variables and monotonicity/box constraints accurately recovers the underlying time-varying rates rests on the unverified assumption that Italian reporting data (January-May 2021) contain no major testing or ascertainment biases; no cross-validation or alternative parameterization comparison is described to support this.

    Authors: The PCHIP control-node approach with monotonicity and box constraints was chosen to produce smooth, plausible rate trajectories while keeping the inverse problem tractable. The calibration uses the reported Italian data without explicit bias correction, an assumption common to compartmental studies. No cross-validation or alternative spline comparisons were performed. We will add a limitations paragraph in the calibration section acknowledging the data-fidelity assumption and the absence of cross-validation. revision: partial

  3. Referee: [Stability/Threshold analysis] Stability and threshold sections: The local/global stability proofs and the sufficient condition for decay (effective reproduction number persistently below one) are derived in terms of the fitted time-varying parameters; it is unclear whether the parametric bootstrap (n=1000) uncertainty propagates to alter the threshold or stability conclusions.

    Authors: The stability proofs and sufficient decay condition are established analytically for the deterministic system evaluated at the calibrated point estimates. The parametric bootstrap quantifies uncertainty in the 14 nodes but was not propagated through the threshold analysis. We will revise the stability section to clarify that the results apply to the fitted trajectories and that bootstrap-based robustness checks of the threshold remain future work. revision: partial

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper's core analytical results (well-posedness, closed-form R0, local/global stability of DFE, L-infinity PCHIP error bound, Fisher-information identifiability, and threshold condition for decay when Reff<1) are derived from the model structure and standard Lyapunov/Fisher techniques once the 9-compartment ODE system and PCHIP parameterization are accepted; these steps do not reduce to the fitted values by construction. The reported R2 values are explicitly calibration metrics obtained by SQP optimization to the January-May 2021 Italian data and are not presented as out-of-sample predictions or first-principles results. No self-citation chain, self-definitional loop, or renaming of known results is load-bearing for the claimed theorems. The derivation chain remains self-contained against external mathematical benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 2 invented entities

The central claim rests on standard ODE assumptions for epidemic dynamics plus the specific choice of PCHIP for time-varying rates and the accuracy of the input data. The 14 SQP variables are fitted parameters. New entities like the super-spreader subpopulation are introduced without independent evidence beyond the model fit.

free parameters (2)
  • 14 SQP optimization variables
    Calibration is reduced to a 14-variable Sequential Quadratic Programming problem with monotonicity and box constraints; these are fitted to match the Italian data.
  • c2=1.5
    Relative transmissibility of the Alpha variant is stated as 43-90% more transmissible, with c2=1.5 used as the value in the model.
axioms (2)
  • domain assumption The epidemic dynamics can be accurately described by a system of nonlinear ordinary differential equations with the specified nine compartments and transitions.
    Invoked as the foundation for the entire model construction and stability analysis.
  • standard math The PCHIP control-node parameterization with monotonicity constraints approximates the true time-varying functions with O(h^2) error as node spacing vanishes.
    Used to justify the spline-based identification method and its convergence properties.
invented entities (2)
  • Super-spreader subpopulation no independent evidence
    purpose: To capture heterogeneous transmission dynamics not explained by uniform mixing.
    Introduced as part of the nine-compartment structure; no independent evidence provided beyond the model fit.
  • Differentiated hospitalization and mortality compartments no independent evidence
    purpose: To model explicit hospital dynamics and variant-specific outcomes.
    Added to track differentiated mortality; based on observed data but treated as model structure.

pith-pipeline@v0.9.1-grok · 5892 in / 1890 out tokens · 33481 ms · 2026-06-27T21:07:34.417137+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

35 extracted references · 26 canonical work pages

  1. [1]

    Audoly, G

    S. Audoly, G. Bellu, L. D’Angiò, M. P. Saccomani, and C. Cobelli, Global identifi- ability of nonlinear models of biological systems, IEEE Transactions on Biomedical Engi- neering, 48 (2001), pp. 55–65, https://doi.org/10.1109/10.900248

    work page doi:10.1109/10.900248 2001

  2. [2]

    Birkhoff and C

    G. Birkhoff and C. de Boor, Piecewise polynomial interpolation and approximation, in Approximation of Functions, H. L. Garabedian, ed., Elsevier, Amsterdam, 1968, pp. 164– 190

    1968

  3. [3]

    Brauer and C

    F. Brauer and C. Castillo-Chávez, Mathematical Models in Population Biology and Epi- demiology, vol. 40 of Texts in Applied Mathematics, Springer, New York, 2nd ed., 2012, https://doi.org/10.1007/978-1-4614-1686-9

    work page doi:10.1007/978-1-4614-1686-9 2012

  4. [4]

    K. P. Burnham and D. R. Anderson , Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, Springer, New York, 2nd ed., 2002, https: //doi.org/10.1007/b97636

    work page doi:10.1007/b97636 2002

  5. [5]

    M. Can, B. Othman, and E. Tekin, Mathematical modeling and analysis of COVID-19 pandemic with vaccination and mutant variants, Applied Mathematics and Computation, 465 (2024), p. 128418, https://doi.org/10.1016/j.amc.2023.128418

    work page doi:10.1016/j.amc.2023.128418 2024

  6. [6]

    de Boor, A Practical Guide to Splines, vol

    C. de Boor, A Practical Guide to Splines, vol. 27 of Applied Mathematical Sciences, Springer, New York, 1978, https://doi.org/10.1007/978-1-4612-6333-3

    work page doi:10.1007/978-1-4612-6333-3 1978

  7. [7]

    Diekmann and J

    O. Diekmann and J. A. P. Heesterbeek,Mathematical Epidemiology of Infectious Diseases: Model Building, Analysis and Interpretation, vol. 5 of Wiley Series in Mathematical and Computational Biology, John Wiley & Sons, Chichester, 2000

    2000

  8. [8]

    A. Endo, S. Abbott, A. J. Kucharski, and S. Funk, Estimating the overdispersion in COVID-19 transmission using outbreak sizes outside China, Wellcome Open Research, 5 (2020), p. 67, https://doi.org/10.12688/wellcomeopenres.15842.3

    work page doi:10.12688/wellcomeopenres.15842.3 2020

  9. [9]

    N. M. Ferguson, D. Laydon, G. Nedjati-Gilani, N. Imai, K. Ainslie, M. Baguelin, S. Bhatia, A. Boonyasiri, Z. Cucunubá, G. Cuomo-Dannenburg, et al., Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand, tech. report, Imperial College London COVID-19 Response Team, 2020, https: //doi.org/10.25561/77482

    work page doi:10.25561/77482 2020

  10. [10]

    Flaxman, S

    S. Flaxman, S. Mishra, A. Gandy, H. J. T. Unwin, T. A. Mellan, H. Coupland, 21 C. Whittaker, H. Zhu, T. Berah, J. W. Eaton, et al., Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe, Nature, 584 (2020), pp. 257– 261, https://doi.org/10.1038/s41586-020-2405-7

    work page doi:10.1038/s41586-020-2405-7 2020

  11. [11]

    F. N. Fritsch and R. E. Carlson, Monotone piecewise cubic interpolation, SIAM Journal on Numerical Analysis, 17 (1980), pp. 238–246, https://doi.org/10.1137/0717021

    work page doi:10.1137/0717021 1980

  12. [12]

    Giordano, F

    G. Giordano, F. Blanchini, R. Bruno, P. Colaneri, A. Di Filippo, A. Di Matteo, and M. Colaneri, Modelling the COVID-19 epidemic and implementation of population- wide interventions in Italy, Nature Medicine, 26 (2020), pp. 855–860, https://doi.org/10. 1038/s41591-020-0883-7

    2020

  13. [13]

    Giordano, M

    G. Giordano, M. Colaneri, A. Di Filippo, F. Blanchini, P. Bolzern, G. De Nico- lao, P. Sacchi, P. Colaneri, and R. Bruno, Modeling vaccination rollouts, SARS- CoV-2 variants and the requirement for non-pharmaceutical interventions in Italy, Nature Medicine, 27 (2021), pp. 993–998, https://doi.org/10.1038/s41591-021-01334-5

    work page doi:10.1038/s41591-021-01334-5 2021

  14. [14]

    Grüne and J

    L. Grüne and J. Pannek, Nonlinear Model Predictive Control: Theory and Algorithms, Communications and Control Engineering, Springer, Cham, 2nd ed., 2017, https://doi. org/10.1007/978-3-319-46024-6

    work page doi:10.1007/978-3-319-46024-6 2017

  15. [15]

    Guan, Z.-Y

    W.-J. Guan, Z.-Y. Ni, Y. Hu, W.-H. Liang, C.-Q. Ou, J.-X. He, L. Liu, H. Shan, C.-L. Lei, D. S. C. Hui, et al., Clinical characteristics of coronavirus disease 2019 in China, New England Journal of Medicine, 382 (2020), pp. 1708–1720, https://doi.org/10. 1056/NEJMoa2002032

    2019

  16. [16]

    X. Hao, S. Cheng, D. Wu, T. Wu, X. Lin, and C. Li , Reconstruction of the full transmission dynamics of COVID-19 in Wuhan, Nature, 584 (2020), pp. 420–424, https: //doi.org/10.1038/s41586-020-2554-8

    work page doi:10.1038/s41586-020-2554-8 2020

  17. [17]

    H. W. Hethcote, The mathematics of infectious diseases, SIAM Review, 42 (2000), pp. 599– 653, https://doi.org/10.1137/S0036144500371907

    work page doi:10.1137/s0036144500371907 2000

  18. [18]

    W. O. Kermack and A. G. McKendrick, A contribution to the mathematical theory of epidemics, Proceedings of the Royal Society of London A, 115 (1927), pp. 700–721, https: //doi.org/10.1098/rspa.1927.0118

    work page doi:10.1098/rspa.1927.0118 1927

  19. [19]

    H. K. Khalil, Nonlinear Systems, Prentice Hall, Upper Saddle River, NJ, 3rd ed., 2002

    2002

  20. [20]

    J. S. Lavine, O. N. Bjørnstad, and R. Antia, Immunological characteristics govern the transition of COVID-19 to endemicity, Science, 371 (2021), pp. 741–745, https://doi.org/ 10.1126/science.abe6522

    work page doi:10.1126/science.abe6522 2021

  21. [21]

    J. C. Lemaitre, J. Perez-Saez, A. S. Azman, A. Rinaldo, and J. Fellay, Assessing the impact of non-pharmaceutical interventions on SARS-CoV-2 transmission in Switzerland, Swiss Medical Weekly, 150 (2020), p. w20295, https://doi.org/10.4414/smw.2020.20295

    work page doi:10.4414/smw.2020.20295 2020

  22. [22]

    J. O. Lloyd-Smith, S. J. Schreiber, P. E. Kopp, and W. M. Getz, Superspreading and the effect of individual variation on disease emergence, Nature, 438 (2005), pp. 355–359, https://doi.org/10.1038/nature04153

    work page doi:10.1038/nature04153 2005

  23. [23]

    M. D. Morris, Factorial sampling plans for preliminary computational experiments, Techno- metrics, 33 (1991), pp. 161–174, https://doi.org/10.1080/00401706.1991.10484804

    work page doi:10.1080/00401706.1991.10484804 1991

  24. [24]

    Ndairou, I

    F. Ndairou, I. Area, J. J. Nieto, and D. F. M. Torres, Mathematical modeling of COVID-19 transmission dynamics with a case study of Wuhan, Chaos, Solitons & Fractals, 135 (2020), p. 109846, https://doi.org/10.1016/j.chaos.2020.109846

    work page doi:10.1016/j.chaos.2020.109846 2020

  25. [25]

    Nocedal, S

    J. Nocedal and S. J. Wright, Numerical Optimization, Springer, New York, 2nd ed., 2006, https://doi.org/10.1007/978-0-387-40065-5

    work page doi:10.1007/978-0-387-40065-5 2006

  26. [26]

    A. Pan, L. Liu, C. Wang, H. Guo, X. Hao, Q. Wang, J. Huang, N. He, H. Yu, X. Lin, S. Wei, and T. Wu, Association of public health interventions with the epidemiology of the COVID-19 outbreak in Wuhan, China, JAMA, 323 (2020), pp. 1915–1923, https: //doi.org/10.1001/jama.2020.6130

    work page doi:10.1001/jama.2020.6130 2020

  27. [27]

    https://github.com/pcm-dpc/COVID-19,

    Presidenza del Consiglio dei Ministri – Dipartimento della Protezione Civile, COVID-19 Italia — Monitoraggio situazione. https://github.com/pcm-dpc/COVID-19,

  28. [28]

    A. M. Ramos, M. R. Fernández, M. Vela-Pérez, A. B. Kubik, and B. Ivorra,A simple but complex enough θ-SIR type model to be used with COVID-19 real data. Application to the case of Italy, Physica D: Nonlinear Phenomena, 421 (2021), p. 132839, https://doi. org/10.1016/j.physd.2021.132839

    work page doi:10.1016/j.physd.2021.132839 2021

  29. [29]

    A. Raue, C. Kreutz, T. Maiwald, J. Bachmann, M. Schilling, U. Klingmüller, and J. Timmer, Structural and practical identifiability analysis of partially observed dynamical models by exploiting the profile likelihood, Bioinformatics, 25 (2009), pp. 1923–1929, https: //doi.org/10.1093/bioinformatics/btp358

    work page doi:10.1093/bioinformatics/btp358 2009

  30. [30]

    Riou and C

    J. Riou and C. L. Althaus, Pattern of early human-to-human transmission of Wuhan 2019 novel coronavirus (2019-nCoV), December 2019 to January 2020, Eurosurveillance, 25 22 (2020), p. 2000058, https://doi.org/10.2807/1560-7917.ES.2020.25.4.2000058

    work page doi:10.2807/1560-7917.es.2020.25.4.2000058 2019

  31. [31]

    Saltelli, M

    A. Saltelli, M. Ratto, T. Andres, F. Campolongo, J. Cariboni, D. Gatelli, M. Saisana, and S. Tarantola, Global Sensitivity Analysis: The Primer, John Wi- ley & Sons, Chichester, 2008

    2008

  32. [32]

    B. Tang, X. Wang, Q. Li, N. L. Bragazzi, S. Tang, Y. Xiao, and J. Wu, Estimation of the transmission risk of the 2019-nCoV and its implication for public health interventions, Journal of Clinical Medicine, 9 (2020), p. 462, https://doi.org/10.3390/jcm9020462

    work page doi:10.3390/jcm9020462 2019

  33. [33]

    van den Driessche and J

    P. van den Driessche and J. Watmough, Reproduction numbers and sub-threshold endemic equilibria for compartmental models of disease transmission, Mathematical Biosciences, 180 (2002), pp. 29–48, https://doi.org/10.1016/S0025-5564(02)00108-6

    work page doi:10.1016/s0025-5564(02)00108-6 2002

  34. [34]

    O. J. Watson, G. Barnsley, J. Toor, A. B. Hogan, P. Winskill, and A. C. Ghani, Global impact of the first year of COVID-19 vaccination: a mathematical modelling study, Lancet Infectious Diseases, 22 (2022), pp. 1293–1302, https://doi.org/10.1016/ S1473-3099(22)00320-6

    2022

  35. [35]

    World Health Organization, Coronavirus disease (COVID-19) pandemic, WHO situa- tionreport, WorldHealthOrganization, 2020, https://www.who.int/emergencies/diseases/ novel-coronavirus-2019. 23

    2020