Abstract
The design of optimized non-pharmaceutical interventions (NPIs) is critical to the effective control of emergent outbreaks of infectious diseases such as SARS, A/H1N1 and COVID-19 and to ensure that numbers of hospitalized cases do not exceed the carrying capacity of medical resources. To address this issue, we formulated a classic SIR model to include a close contact tracing strategy and structured prevention and control interruptions (SPCIs). The impact of the timing of SPCIs on the maximum number of non-isolated infected individuals and on the duration of an infectious disease outside quarantined areas (i.e. implementing a dynamic zero-case policy) were analyzed numerically and theoretically. These analyses revealed that to minimize the maximum number of non-isolated infected individuals, the optimal time to initiate SPCIs is when they can control the peak value of a second rebound of the epidemic to be equal to the first peak value. More individuals may be infected at the peak of the second wave with a stronger intervention during SPCIs. The longer the duration of the intervention and the stronger the contact tracing intensity during SPCIs, the more effective they are in shortening the duration of an infectious disease outside quarantined areas. The dynamic evolution of the number of isolated and non-isolated individuals, including two peaks and long tail patterns, have been confirmed by various real data sets of multiple-wave COVID-19 epidemics in China. Our results provide important theoretical support for the adjustment of NPI strategies in relation to a given carrying capacity of medical resources.
Similar content being viewed by others
Data availability
The data sets were obtained from the National Health Commission of the People’s Republic of China and Province Municipal Health Commissions and are available from their websites http://sxwjw.shaanxi.gov.cn; http://wjw.xizang.gov.cn; http://wjw.xinjiang.gov.cn.
References
Bootsma M, Ferguson N (2007) The effect of public health measures on the 1918 influenza pandemic in U.S. cities. Proc Natl Acad Sci 104(18):7588–7593
Contreras S, Villavicencio H et al (2020) A multi-group SEIRA model for the spread of COVID-19 among heterogeneous populations. Chaos Solitons Fractals 136:109925
Corless R, Gonnet G et al (1996) On the Lambert W function. Adv Comput Math 5(1):329–359
Eames K (2007) Contact tracing strategies in heterogeneous populations. Epidemiol Infect 135(3):443–454
Ferguson N, Laydon D et al (2020) Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19 mortality and healthcare demand. Imperial College COVID-19 Response Team
Fetzer T, Graeber T (2021) Measuring the scientific effectiveness of contact tracing: evidence from a natural experiment. Proc Natl Acad Sci 118(33):1–4
Flaxman S, Mishra S et al (2020) Report 13: estimating the number of infections and the impact of non-pharmaceutical interventions on COVID-19 in 11 European countries. Imperial College COVID-19 Response Team
Fraser C, Riley S et al (2004) Factors that make an infectious disease outbreak controllable. Proc Natl Acad Sci 101(16):6146–6151
Fujiwara N, Onaga T et al (2022) Analytical estimation of maximum fraction of infected individuals with one-shot non-pharmaceutical intervention in a hybrid epidemic model. BMC Infect Dis 22(1):1–11
Health Commission of Tibet Autonomous Region (2022) Available from http://wjw.xizang.gov.cn
Health Commission of Xinjiang Uygur Autonomous Region (2022) Available from http://wjw.xinjiang.gov.cn
Ji Y, Ma Z et al (2020) Potential association between COVID-19 mortality and healthcare resource availability. Lancet Global Health 8(4):480
Joint Prevention and Control Mechanism of the State Council (2022) COVID-19 Disease Prevention and Control Guideline. Available from http://www.gov.cn/xinwen/2022-06/28/content_5698168.html
Julia E, Nicole L et al (2007) Non-pharmaceutical public health interventions for pandemic influenza: an evaluation of the evidence base. BMC Public Health 7(1):208
Keeling MJ, Rohani P (2011) Modeling infectious diseases in humans and animals. Princeton University Press, Princeton
Kretzschmar M, Rozhnova G et al (2020) Impact of delays on effectiveness of contact tracing strategies for COVID-19: a modelling study. Lancet Public Health 5(8):452–459
Lauro F, Kiss I et al (2021) Optimal timing of one-shot interventions for epidemic control. Plos Comput Biol 17(3):1–25
Li H, Zhang H (2023) Cost-effectiveness analysis of COVID-19 screening strategy under China’s dynamic zero-case policy. Front Public Health 11(1829):1099116
Morgan A, Woolhouse M et al (2021) Optimizing time-limited non-pharmaceutical interventions for COVID-19 outbreak control. Philos Trans R Soc B 376(1829):20200282
Nicola M, Alsafi Z et al (2020) The socio-economic implications of the coronavirus pandemic (COVID-19): a review. Int J Surg 78:185–193
Noor M, Raza A et al (2022) Non-standard computational analysis of the stochastic COVID-19 pandemic model: an application of computational biology. Alex Eng J 61(1):619–630
Onishchenko G, Sizikova T et al (2022) The omicron variant of the Sars-Cov-2 virus as the dominant agent of a new risk of disease amid the COVID-19 pandemic. Herald Russ Acad Sci 92:381–391
Rawson T, Brewer T et al (2020) How and when to end the COVID-19 lockdown: an optimization approach. Front Public Health 8:262
Shaanxi Municipal Health Commission (2022) Available from: http://sxwjw.shaanxi.gov.cn
Tang B, Bragazzi N et al (2020a) An updated estimation of the risk of transmission of the novel coronavirus (2019-nCov). Infect Dis Modell 5:248–255
Tang B, Wang X et al (2020b) Estimation of the transmission risk of the 2019-nCoV and its implication for public health interventions. J Clin Med 9(2):1–13
Tang S, Wang X et al (2023) Threshold conditions for curbing COVID-19 with a dynamic zero-case policy derived from 101 outbreaks in China. BMC Public Health 23(1):1–12
Wang X, Zhang X et al (2020) Challenges to the system of reserve medical supplies for public health emergencies: reflections on the outbreak of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) epidemic in China. Biosci Trends 14(1):3–8
World Health Organization Writing Group (2006) Nonpharmaceutical interventions for pandemic influenza, national and community mensures. Emerg Infect Dis 12(1):88–94
Wu F, Zhao S et al (2020) A new coronavirus associated with human respiratory disease in China. Nature 579:265–269
Zhai S, Luo G et al (2021) Vaccination control of an epidemic model with time delay and its application to COVID-19. Nonlinear Dyn 106(2):1279–1292
Zhang W, Huang L et al (2022) Vaccine booster efficiently inhibits entry of SARS-CoV-2 omicron variant. Cell Mol Immunol 19(3):445–446
Zhao J, Yuan Q et al (2020) Antibody responses to SARS-CoV-2 in patients with novel coronavirus disease 2019. Clin Infect Dis 71(16):2027–2034
Zhou W, Bai Y et al (2022) The effectiveness of various control strategies: an insight from a comparison modelling study. J Theor Biol 549:111205
Zhou P, Yang X et al (2020) A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature 579:270–273
Zu J, Li M et al (2020) Transmission patterns of COVID-19 in the mainland of China and the efficacy of different control strategies: a data-and model-driven study. Infect Dis Poverty 9(1):1–14
Funding
This work was supported by the National Natural Science Foundation of China (Grant No. 12031010 and No. 12126350). The Fundamental Research Funds for the Central Universities (Grant No. 2021CBLY002).
Author information
Authors and Affiliations
Contributions
All authors designed and conducted the research. HZ and ST did the theoretical analyses. SH and HZ did the data analysis and numerical calculations. HZ, ST and RAC were the lead writers of the manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A: Analysis of the maximum number of non-isolated infected individuals with respect to \(T_{on}\) and \(T_{off}\).
In Theorem 4.1, we summarize the analytical forms of the maximum number of non-isolated infected individuals for cases (\(a_{1}\))-(\(a_{2}\)) and (\(a_{4}\))-(\(a_{6}\)) with respect to the number of susceptible individuals at the onset \(S_{T_{on}}\) and the end \(S_{T_{off}}\) of the second intervention. It follows from the first equation of system (4.1) that \(S_{T_{on}}\) and \(S_{T_{off}}\) are monotonically decreasing with respect to \(T_{on}\) and \(T_{off}\), respectively, i.e. \(\frac{\partial S_{T_{on}}}{\partial T_{on}}<0\) and \(\frac{\partial S_{T_{off}}}{\partial T_{off}}<0\). In the following, we analyse the impact of the start and end timings of the second intervention on the maximum number of non-isolated infected individuals in detail.
- (i):
-
The maximum number of non-isolated infected individuals occurs in the region \((a_{1})\), i.e. case \((a_{1})\).
For case (\(a_{1}\)), the maximum number of non-isolated infected individuals is given as
which depends on \(S_{T_{on}}\) and \(S_{T_{off}}\). It implies that \(I^{c}_{max}\) depends not only on \(T_{on}\) but also on \(T_{off}\). Taking the derivative of equation (A.1) with respect to \(T_{off}\), yields
Correspondingly, taking the derivative of Eq. (A.1) with respect to \(S_{T_{on}}\), yields
where \(\frac{\partial S_{T_{off}}}{\partial S_{T_{on}}}=\frac{\beta _{1c}S_{T_{off}}I_{T_{off}}}{\beta _{10}S_{T_{on}}I_{T_{on}}}>0\). It follows from Eq. (A.2) that if \(\beta _{1c}>\beta _{10}\), then \(\frac{\partial I^{c}_{max}}{\partial T_{off}}<0\). Thus, when \(\beta _{1c}>\beta _{10}\), \(I^{c}_{max}\) is monotonically decreasing with respect to \(T_{off}\). However, it is difficult to calculate the sign of Eq. (A.3), so we cannot determine the monotonicity of \(I^{c}_{max}\) with respect to \(T_{on}\) for case (\(a_{1}\)).
- (ii):
-
The maximum number of non-isolated infected individuals occurs in the region (\(a_{2}\)), i.e. case \((a_{2})\).
For case (\(a_{2}\)), the maximum number of non-isolated infected individuals is given as
which only depends on \(S_{T_{on}}\). This implies that \(I^{c}_{max}\) only depends on \(T_{on}\) for case (\(a_{2}\)). Taking the derivative of equation (A.4) with respect to \(T_{on}\), yields
Thus, if \(\beta _{1c}>\beta _{10}\), then \(\frac{\partial I^{c}_{max}}{\partial T_{on}}>0 \). Therefore, when \(\beta _{1c}>\beta _{10}\), \(I^{c}_{max}\) is monotonically increasing with respect to \(T_{on}\) for case (\(a_{2}\)).
- (iii):
-
The maximum number of non-isolated infected individuals occurs in the region \((a_{4})\), i.e. case \((a_{4})\).
For case (\(a_{4}\)), the maximum number of non-isolated infected individuals is given as
which only depends on \(S_{T_{on}}\). This implies that \(I^{c}_{max}\) also only depends on \(T_{on}\). Taking the derivative of the above equation (A.6) with respect to \(T_{on}\), we have
Thus, \(I^{c}_{max}\) is monotonically increasing with respect to \(T_{on}\) for case (\(a_{4}\)).
- (iv):
-
The maximum number of non-isolated infected individuals occurs in the region \((a_{5})\), i.e. case \((a_{5})\).
For case (\(a_{5}\)), the maximum number of non-isolated infected individuals is given as
If \(I^{c}_{max}=I_{2}\), then \(\frac{\partial I^{c}_{max}}{\partial T_{off}}=0\), and when \(\beta _{1c}>\beta _{10}\), \(\frac{\partial I^{c}_{max}}{\partial T_{on}}<0\). Thus, when \(\beta _{1c}>\beta _{10}\), \(I^{c}_{max}\) is monotonically increasing with respect to \(T_{on}\). If \(I^{c}_{max}=I_{1}\), it follows from Eq. (A.2) that \(\frac{\partial I^{c}_{max}}{\partial T_{off}}<0\). Thus, in this case, \(I^{c}_{max}\) is monotonically decreasing with respect to \(T_{off}\). Therefore, for case (\(a_{5}\)), \(\frac{\partial I^{c}_{max}}{\partial T_{off}}\le 0\) and \(I^{c}_{max}\) is quasi-monotonically decreasing with respect to \(T_{off}\). However, it is difficult to calculate the value of Eq. (A.3), and we cannot determine the monotonicity of \(I^{c}_{max}\) with respect to \(T_{on}\) for case (\(a_{5}\)).
- (v):
-
The maximum number of non-isolated infected individuals occurs in the region \((a_{6})\), i.e. case \((a_{6})\).
For case (\(a_{6}\)), the maximum number of non-isolated infected individuals is given as
If \(I^{c}_{max}=I_{T_{on}}\), then \(\frac{\partial I^{c}_{max}}{\partial T_{on}}>0\) and \(\frac{\partial I^{c}_{max}}{\partial T_{off}}=0\). Thus, \(I^{c}_{max}\) is monotonically increasing with respect to \(T_{on}\). If \(I^{c}_{max}=I_{1}\), it follows from Eq. (A.2) that \(\frac{\partial I^{c}_{max}}{\partial T_{off}}<0\). Thus, in this case, \(I^{c}_{max}\) is monotonically decreasing with respect to \(T_{off}\). Therefore, \(\frac{\partial I^{c}_{max}}{\partial T_{off}}\le 0\) and \(I^{c}_{max}\) is quasi-monotonically decreasing with respect to \(T_{off}\) for case (\(a_{6}\)). However, it is difficult to calculate the sign of Eq. (A.3), so we cannot determine the monotonicity of \(I^{c}_{max}\) with respect to \(T_{on}\) for case (\(a_{6}\))
Appendix B: Analysis of the special case \(T_{off}=T^{c}_{end}\) in model (2.1) with (2.2)
In the following, we carry out the numerical analyses to understand the effect of the timing of the second intervention and close contact tracing strategy on the maximum number of non-isolated infected individuals and the time needed to realize the dynamic zero-case aim. Here we let \(N=763\), \(S_{0}=762\), \(I_{0}=1\), \(\beta =0.155\), \(c_{0}=10\), \(q_{0}=0.01526\), \(\gamma _{I}=0.1\) and \(\delta _{I}=0.2504\), and produce the contour plots of the maximum number of non-isolated infected individuals, and the time needed to realize the dynamic zero-case with respect to \(T_{on}\) and \(\Delta q\), \(T_{on}\) and \(\Delta c\), \(\Delta q\) and \(\Delta c\), respectively, as shown in Fig. 12a–f. In Fig. 12a and b, we let \(c_{c}=10\), i.e. the contact rate during the second intervention is the same as that without the second intervention. It follows from Fig. 12a that the earlier the second intervention, the stronger the contact tracking intensity during the second intervention, the smaller the maximum number of non-isolated infected individuals. In Fig. 12b, it is worth mentioning that \(T^{c}_{end}\) is non-monotonically dependent on \(\Delta q\) for a fixed small \(T_{on}\), which implies that even if a second intervention is implemented early, it may take longer to realize the dynamic zero-case policy due to the contact tracing intensity not being strong enough. Therefore, in this case, to minimize the maximum number of non-isolated infected individuals and shorten the duration, the best strategy is to intervene early and fully strengthen the contact tracing intensity. As shown in Fig. 12a and b, it is clear that when \((T_{on}, \Delta q)\) nears the point (1, 0.85), both \(I^{c}_{max}\) and \(T^{c}_{end}\) are minimized.
Comparing this with the results shown in Fig. 12a, we can conclude that the contact rate should be reduced to a very low level to get the same \(I^{c}_{max}\) in Fig. 12c. This implies that strengthening the contact tracing intensity is more effective than reducing the contact rate in minimizing the maximum number of infected individuals. It follows from Fig. 12b and d that strengthening the contact tracing intensity is also more effective than reducing the contact rate in shortening the time needed to realize the dynamic zero-case aim.
The results from Fig. 12e with \(T_{on}=3\) reveal that the stronger the contact tracing intensity, the smaller the contact rate during the second intervention, the more effective they are to minimize the maximum number of non-isolated infected individuals. In terms of the time needed to realize the dynamic zero-case aim, it follows from Fig. 12f that \(T^{c}_{end}\) is monotonically decreasing with respect to \(\Delta q\) for any fixed \(\Delta c\). It is worth mentioning that for small and intermediate \(\Delta q\), \(T^{c}_{end}\) is an upward function of \(\Delta c\). This means that when the quarantine rate (\(q_{c}\)) is not large enough, in order to shorten the time needed to realize the dynamic zero-case aim, it is necessary to reduce the contact rate to a very low level (i.e. \(\Delta c\) is large enough).
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhou, H., Sha, H., Cheke, R.A. et al. Model analysis and data validation of structured prevention and control interruptions of emerging infectious diseases. J. Math. Biol. 88, 62 (2024). https://doi.org/10.1007/s00285-024-02083-y
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00285-024-02083-y
Keywords
- Emerging infectious diseases
- Structured prevention and control interruptions
- Multiple peaks
- Optimal strategy