Pith. sign in

REVIEW 5 major objections 4 minor 29 references

Critical Thresholds in Non-Pharmaceutical Interventions for Epidemic Control

T0 review · 5 major / 4 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper derives a critical threshold R = k̄βG(τ) < 1 for epidemic containment and shows contact tracing alone can contain diseases with R0 < 2.12, rising to R0 < 7.82 when paired with social distancing.

desk verdict A useful operational framework with a genuinely new critical-line result, but the headline thresholds are calibrated on one outbreak and rest on simplifications that need independent testing before policy use. read the letter →

arxiv 2512.08339 v3 pith:RVOLJWJS submitted 2025-12-09 physics.soc-ph

Jinghui Wang , Yutian Zeng , Cong Xu , Xiyun Zhang , Zhanwei Du , Jiarong Xie , Jiu Zhang , Sen Pei
show 2 more authors
Zijian Feng Yanqing Hu
This is my paper · ORCID
classification physics.soc-ph
keywords epidemiccontainmentcontacttracingsocialdistancingreproductionnumbercriticalthresholdnon-pharmaceuticalinterventionsphaseplaneCOVID-19
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper introduces a probabilistic framework that links the speed of contact tracing (measured by a tracing period τ) and the intensity of social distancing (measured by average close contacts k̄) to a single containment condition: the effective reproduction number R must stay below 1. Using high-resolution data from Shenzhen's 2022 Omicron outbreak, the authors validate this condition and show that rapid tracing alone can contain diseases with a basic reproduction number below 2.12, while adding moderate social distancing extends containment to diseases with R0 up to 7.82. These thresholds translate into practical guidance: for a given pathogen, the critical tracing period τc can be read off, and the phase plane of k̄ versus τ tells policymakers whether an outbreak is in a pandemic or contained phase. The work offers a budget-friendly alternative to mass PCR testing by quantitatively mapping when tracing and targeted distancing suffice.

What carries the argument

The central object is the function G(τ), the expected effective infectious period of a secondary case conditioned on two transmission-chain survival events: the secondary must be infected before the primary is quarantined, and the secondary must become infectious before it is traced. G(τ) is derived from a recursive probabilistic iteration over transmission generations, assuming independence of latent period, infectious period, and time-to-transmission, with an exponential distribution for the time-to-transmission approximated linearly as β. Together with k̄ and β, G(τ) defines the critical line R = 1 in the k̄–τ plane, which is the paper's main analytical tool for classifying containment ou

What would settle it

Collect high-resolution contact tracing data from an outbreak where the measured (k̄, τ) sits on the contained side of the critical line but the observed effective reproduction number R exceeds 1; this would falsify the model. Alternatively, simulate transmission with a gamma-distributed generation interval with coefficient of variation greater than 1 and check whether the predicted thresholds R0 ≈ 2.12 and 7.82 still hold; if the simulated containment boundary shifts by more than 20%, the exponential/linear approximation is the culprit.

Watch

Extended reading notes

Core claim

The paper claims that epidemic containment is equivalent to the inequality R = k̄βG(τ) < 1, where G(τ) is the expected effective infectious period under a test-trace-quarantine protocol with tracing delay τ. From this condition, the authors construct a critical line in the k̄–τ plane separating pandemic (R > 1) from contained (R < 1) phases. Fitting the framework to Shenzhen's 2022 Omicron data (1,187 cases, 86,451 contacts), they estimate that contact tracing alone can contain pathogens with R0 < 2.12 (95% CI 2.07–2.16) and that combining tracing with social distancing that reduces k̄ from 177.64 to 51.14 extends the bound to R0 < 7.82 (95% CI 7.70–7.93). The same model yields critical trac

Load-bearing premise

The derivation assumes the time from a case becoming infectious to transmitting is exponentially distributed with a constant per-day rate β (and linearly approximates it as β), and that every individual shares the same average contact number k̄ and the same tracing delay τ; if real transmission timing is non-exponential or heavily overdispersed, or if tracing delays vary widely across the population, the computed critical thresholds will shift.

Editorial extensions

If this is right

  • If the critical-line condition holds, public health agencies can set a target tracing period τc for any known R0 and check whether their operational speed is sufficient for containment.
  • The R0 < 2.12 threshold for tracing alone implies that for nearly half of major infectious diseases, mass PCR screening may be unnecessary if rapid contact tracing is in place.
  • Combining tracing with social distancing (reducing k̄) lifts the threshold to R0 < 7.82, covering roughly 87% of major pathogens, offering a practical alternative to full lockdowns.
  • Regional variations in tracing miss rates (from Shenzhen's 18.6% to Hong Kong's 73%) shift the thresholds downward, so the framework provides a data-driven way to assess whether supplementary testing is needed.
  • The k̄–τ phase plot can be used in real time to monitor whether an ongoing outbreak is drifting toward or away from the contained phase, enabling proactive policy adjustment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same G(τ) machinery could be adapted to other interventions such as routine testing frequency or isolation of suspected cases, where τ would represent a detection-to-isolation delay rather than a tracing delay, likely yielding similar threshold formulas.
  • The linearization of the transmission-time distribution (fs ≈ β) may break down for pathogens with overdispersed or superspreading transmission; if real serial intervals are highly variable, the numerical thresholds (2.12 and 7.82) could shift, so they should be treated as operational benchmarks for fairly regular transmission settings rather than hard biological constants.
  • A testable extension is to measure the distribution of effective infectious periods in real tracing data and compare it to the model's predicted G(τ); systematic deviations would indicate where the independence or exponential assumptions need revision.
  • The phase-plane approach could be generalized to compare the resilience of different cities by plotting their measured (k̄, τ) points against the critical line, revealing which regions have operational slack and which are precariously close to the pandemic boundary.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The paper proposes a probabilistic framework in which the effective reproduction number is written as R = k̄₊βG(τ) (Eq. 1), where k̄₊ is the average number of close contacts and τ is the contact-tracing delay. It derives critical curves R=1 in the (k̄₊, τ) plane, reports critical tracing periods τc for SARS-CoV-2 variants and other pathogens, and uses data from Shenzhen's 2022 Omicron outbreak (1,187 cases, 86,451 contacts) to estimate parameters and to validate the framework. The headline quantitative claims are that contact tracing alone can contain diseases with R₀ < 2.12 and that adding social distancing extends containment to R₀ < 7.82, implying that 43.33% and 86.67% of a list of major infectious diseases are containable under these policies.

Significance. If the derivation and parameter estimates are sound, the paper would provide a compact, operational rule linking tracing speed, social contact intensity, and R₀, with direct policy relevance. The Shenzhen dataset is unusually detailed, and the idea of a critical line in the (k̄₊, τ) plane is intuitive and useful. The paper also makes concrete, falsifiable threshold predictions. However, the central quantitative claims rest on strong distributional assumptions and on parameters estimated from the same outbreak used for validation; the statement of the model and all supporting derivations are in an unprovided supplementary file. These issues must be resolved before the thresholds can be accepted as generally applicable.

major comments (5)
  1. [Materials and Methods, Eq. (6)] The expression for G(τ) is derived under two assumptions: Ts is exponentially distributed and its density is replaced by the constant β because β is 'typically small' (βSZ = 0.038). This linearization is uncontrolled: the shape of the serial-interval distribution is precisely what determines how sensitive the critical line is to τ. For pathogens with larger β or non-exponential, overdispersed serial intervals, the thresholds R₀ < 2.12 and R₀ < 7.82 can shift. The paper should either give the exact expression without the fs ≈ β approximation, or provide a sensitivity analysis showing that the thresholds are robust to realistic departures from exponentiality. As written, the approximation is load-bearing for every reported threshold.
  2. [Materials and Methods, Eqs. (2)–(6)] Eq. (6) integrates the latent-period density fe(te) against kernels involving t and t+τ−te without explicitly conditioning on the chain-continuation conditions in Eqs. (2)–(3). The distribution f(t;τ) in Eq. (5) is conditional on the two events, but it is not clear that the same conditioning is applied to fe(te) and to the joint distribution of Ts, Te, and T_l. The current text appears to mix conditional and unconditional densities, which would make G(τ) incorrect even under the exponential assumption. The derivation in Supplementary Sec. 8 must be shown explicitly and the conditioning made precise.
  3. [Fig. 2c and Supplementary Sec. 6] The validation of the model is not independent. The theoretical critical line in Fig. 2c uses βSZ and k̄₊SZ estimated from Shenzhen's Omicron outbreak, and the trajectory points are computed from the same outbreak. Likewise, R₀(SZ) in Table 1 is estimated from the same data, so the reported τc = 13 h for Omicron is not an out-of-sample prediction. The paper should be explicit that Fig. 2c demonstrates model fit rather than independent validation, and ideally should test the framework on a separate outbreak or provide a genuinely prospective prediction.
  4. [Supplementary material (Secs. 1, 5, 6, 8, 9, 10)] The main text repeatedly refers to supplementary sections for the derivation of R, the estimation of k̄₊ and β, the computation of confidence intervals, the simulation model, and the R₀ thresholds. None of this material is included with the manuscript, so the central derivation and the numerical results cannot be verified. At minimum, the key equations and simulation protocols must be presented in the main text or in a complete supplementary file before the claims can be evaluated.
  5. [Results, 'tracing alone' thresholds and Fig. 3] The thresholds R₀ < 2.12 and R₀ < 7.82 are derived from simulations calibrated to Shenzhen's k̄₊, β, and miss rate q = 0.186. The generalization to other diseases and regions assumes that the same contact structure and tracing efficiency apply. The Discussion acknowledges that thresholds are optimistic for regions with higher miss rates, but the paper does not quantify how sensitive the headline percentages are to plausible variation in q, k̄₊, or the serial-interval distribution. A formal sensitivity analysis over these parameters is needed to support the 43.33% and 86.67% claims.
minor comments (4)
  1. [Fig. 2 caption] Typo: 'Theoritical' should be 'Theoretical'.
  2. [Materials and Methods, paragraph after Eq. (4)] Typo: 'scenerios' should be 'scenarios'.
  3. [References] Reference 20 has a spacing issue: 'B Y ang' should be 'B Yang'.
  4. [Fig. 2c caption] The caption says 'solid curve represents the theoretical result,' but the curve uses parameters estimated from the same data set. Please rephrase to indicate that this is a model fit, or provide an independent validation.

Circularity Check

2 steps flagged · score 6.0 of 10

Shenzhen validation is in-sample: the theoretical critical line and the Omicron τc are computed from parameters fitted to the same Shenzhen dataset, so the claimed empirical confirmation is partly built into the fit.

  1. fitted input called prediction [Fig. 2c caption; Materials and Methods B]
    "The solid curve represents the critical line obtained by R = ¯kβG(τ) and setting R = 1. Note that while the solid curve represents the theoretical result, the points are based on real data. ... Please refer to Supplementary Sec. 6 for the computation of the time-varying τt and ¯k+t, as well as ¯k+SZ and βSZ."

    The theoretical critical line in Fig. 2c is not a parameter-free prediction; it is drawn using βSZ and ¯k+SZ estimated from the same Shenzhen outbreak (Supplementary Sec. 6). Plotting the in-sample Rt trajectory against this curve and reporting that 'Shenzhen's data align closely with our model' is a goodness-of-fit statement, not an independent validation. The apparent agreement is partly forced by the fitted parameters.

  2. fitted input called prediction [Results opening paragraph; Table 1; Supplementary Sec. 6C]
    "The asterisks represent the estimated R0 (95% confidence interval) of the Omicron transmission in Shenzhen (see Supplementary Sec. 6C). ... We estimate the critical contact tracing period, τc, at 26 hours (95% CI 24.75–28.07) for Delta and 13 hours (95% CI 11.63–14.58) for Omicron. Given Shenzhen has the reaction times at 17 hours in 2021 (12) and 11 hours in 2022 (13), explaining the success against Delta and subsequent adaptation for Omicron. These alignments validate our theory."

    For the Shenzhen Omicron variant, R0 = 20.17 is itself estimated from the same February–April 2022 Shenzhen outbreak (Table 1 footnote, Supplementary Sec. 6C). The corresponding τc = 13.07 h is then obtained by feeding this fitted R0 into the model's Eq. S3, so the 'critical tracing period' is a deterministic transformation of the fitted reproduction number. Comparing that derived value with Shenzhen's observed 11-hour reaction time and calling it validation uses the same dataset twice: once to estimate R0 and once to confirm the resulting τc.

full rationale

The analytic core of the paper—Eq. 1, R = ¯k+βG(τ), and the derivation of G(τ) in Eq. 6—is a self-contained probabilistic argument given the stated assumptions (exponentially distributed Ts, the fs≈β linearization, independence of Ts, Te, Tl). No load-bearing self-citation chain or imported uniqueness theorem is present, and the thresholds for other variants/diseases are calculated from published R0 values, which are external inputs. However, the paper's empirical validation is partially circular. The critical line in Fig. 2c uses βSZ and ¯k+SZ fitted from the same Shenzhen trajectory it is compared against, and the Omicron τc = 13 h is computed from an R0 estimate taken from the same outbreak. These are fitted-inputs-called-predictions, not independent confirmations. The headline R0 thresholds (2.12 and 7.82) are also calibrated simulation outputs based on Shenzhen parameters, so they are scenario-dependent extrapolations rather than parameter-free universal predictions. Because the central theoretical relationship is independently derived but its main empirical support reduces to in-sample fitting, the appropriate circularity score is 6: partial circularity in the validation, not a fully definitional derivation.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The thresholds depend on parameters measured from a single outbreak (β, k̄+, q, and the Omicron-SZ R0) plus strong distributional assumptions. The framework is not parameter-free, and the 'universal' phrasing overstates the independence of the result.

free parameters (5)
  • β (per-day transmission probability) = βSZ = 0.038
    Estimated from Shenzhen Omicron data (Supplementary Sec 6A); multiplies G(τ) in R = k̄+βG(τ), so all thresholds scale with it.
  • k̄+ without social distancing = 177.64 (95% CI 155.40–199.88)
    Baseline average effective contacts per positive case in Shenzhen; used for the tracing-only R0 threshold.
  • k̄+ with social distancing = 51.14 (95% CI 37.82–64.46)
    Average effective contacts under Shenzhen's distancing measures; drives the R0<7.82 threshold.
  • q (contact-tracing miss rate) = 0.186
    Fraction of cases detected by PCR rather than contact tracing; used in simulations to set tracing incompleteness.
  • R0 for Omicron in Shenzhen = 20.17 (95% CI 14.39–25.64)
    Estimated by the authors from the same Shenzhen data (Table 1 footnote), then used to compute τc=13.07h, so the validation for this variant is not independent.
assumptions (6)
  • domain assumption Tl, Ts, and Te are independent (Methods, Eq 5).
    Required to write the recursive PDF f(t;τ) and to factor expectations in Eq 6; no biological justification in main text.
  • ad hoc to paper Transmission times are exponential and β is small enough that fs(ts)≈β (Methods, Eq 6, βSZ=0.038).
    This linearization makes G(τ) tractable and determines the shape of the critical line; if not valid, all thresholds shift.
  • domain assumption The population is homogeneous with a single average contact number k̄+ and uniform transmission probability β.
    Used to write R = k̄+βG(τ) for a whole population; authors acknowledge this in limitations.
  • domain assumption A single tracing period τ applies to all contacts, with a fixed miss rate q.
    The critical line is a function of one τ; real contact tracing has distributed delays and variable completeness.
  • domain assumption Imported cases are excluded.
    Discussion says imported cases overwhelmed policies in Singapore and Australia; exclusion makes thresholds optimistic for open economies.
  • domain assumption Literature R0 estimates for the 30 diseases and for variants are accurate point values.
    The 43.33%/86.67% coverage claims and Table 1 τc values inherit these external estimates; Omicron-SZ R0 is instead estimated from the same data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Critical Thresholds in Non-Pharmaceutical Interventions for Epidemic Control." pith.science (2026). https://pith.science/paper/RVOLJWJS

@misc{pith2026251208339,
  author       = {Pith},
  title        = {Pith review of: Critical Thresholds in Non-Pharmaceutical Interventions for Epidemic Control},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RVOLJWJS}},
  note         = {Machine review of arXiv:2512.08339}
}
abstract

Non-pharmaceutical interventions, such as contact tracing and social distancing, are critical for controlling epidemic outbreaks, yet their dynamic interactions remain underexplored. We introduce a probabilistic framework to analyze the synergy between contact tracing speed, quantified by the contact tracing period $\tau$, and the average number of close contacts, $\bar{k}_+$, reflecting social distancing measures. We identify critical thresholds ($R=1$) that separate pandemic and contained phases in the $\bar{k}_{+}-\tau$ plane, validated using high-resolution data from Shenzhen's 2022 Omicron outbreak (1,187 cases, 86,451 contacts). Our findings show that contact tracing alone can contain diseases with $R_0 < 2.12$ (95% CI 2.07-2.16), covering 43.33% of major infectious diseases, while combining with social distancing extends control to $R_0 < 7.82$ (95% CI 7.70-7.93), encompassing 86.67% of pathogens. These results, supported by empirical data, highlight the efficacy of rapid tracing and targeted social distancing as alternatives to mass PCR testing. Our framework offers actionable insights for optimizing NPI strategies, though challenges in scaling to regions with higher tracing miss rates or weaker infrastructure underscore the need for adaptive, data-driven policies.

Figures

Figures reproduced from arXiv: 2512.08339 by the authors.

Figure 1
Figure 1. Epidemic Control in Shenzhen, 2022. (a) Contact trajectories of positive cases. Nodes are geolocated to cases’ residential addresses. Edges represent epidemiological links indicating close contact history between them. The color of the edge (olive green, warm peach, burnt orange) indicates three time periods divided equally according to the chronological order of close contact. (b) A real transmission chain (see Sup… view at source ↗
Figure 3
Figure 3. Policy with incomplete contact tracing. The olive green and burnt orange lines in (a)-(c) represent simulation results under the Shenzhen network without and with social distancing measures (w/o and w/ SDM), respectively. In each simulation, we miss a proportion q of close contacts for each positive case during contact tracing. Based on Shenzhen’s control data, we set q = 0.186 as the proportion of positive cases id… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references

  1. [1]

    With Shenzhen-level social distancing, Rc 0 rises to 5.42 (95% CI 5.18–5.63), 4.19 (95% CI 4.01–4.33), and 3.97 (95% CI 3.80–4.11) (Fig

    of 1.51 (95% CI 1.41–1.57), 1.19 (95% CI 1.13–1.23), and 1.13 (95% CI 1.08–1.17) in these regions, respectively. With Shenzhen-level social distancing, Rc 0 rises to 5.42 (95% CI 5.18–5.63), 4.19 (95% CI 4.01–4.33), and 3.97 (95% CI 3.80–4.11) (Fig. 3c). These findings suggest that, despite regional variations, the control policy offers a viable framework...

  2. [2]

    These results suggest that rapid tracing infrastructure, paired with adaptive distancing, can mitigate outbreaks in early phases, reducing reliance on costly mea- sures like mass lockdowns. Policymakers should invest in scalable tracing systems and public compliance campaigns, particularly for culturally diverse regions, to maximize NPI efficacy while pre...

    2022

  3. [3]

    RE Baker, et al., Infectious disease in an era of global change.Nat. Rev. Microbiol.20, 193–205 (2022)

    2022

  4. [4]

    F Wu, et al., A new coronavirus associated with human respiratory disease in china.Nature 579, 265–269 (2020)

    2020

  5. [5]

    SM Smith, et al., Use of non-pharmaceutical interventions to reduce the transmission of influenza in adults: A systematic review.Respirology20, 896–903 (2015)

    2015

  6. [6]

    D Bell, et al., Non-pharmaceutical interventions for pandemic influenza, national and community measures.Emerg. Infect. Dis.12, 88–94 (2006)

    2006

  7. [7]

    MU Kraemer, et al., The effect of human mobility and control measures on the covid-19 epidemic in china.Science368, 493–497 (2020)

    2020

  8. [8]

    Heal.8, e488–e496 (2020)

    J Hellewell, et al., Feasibility of controlling covid-19 outbreaks by isolation of cases and contacts.The Lancet Glob. Heal.8, e488–e496 (2020)

    2020

Show all 29 references
  1. [9]

    Dis.15, 204–211 (2015)

    S Merler, et al., Spatiotemporal spread of the 2014 outbreak of ebola virus disease in liberia and the effectiveness of non-pharmaceutical interventions: a computational modelling analysis.The Lancet Infect. Dis.15, 204–211 (2015)

    2014

  2. [10]

    S Lai, et al., Effect of non-pharmaceutical interventions to contain covid-19 in china.Nature 585, 410–413 (2020)

    2020

  3. [11]

    JA Polonsky, et al., Feasibility, acceptability, and effectiveness of non-pharmaceutical interventions against infectious diseases among crisis-affected populations: a scoping review.Infect. Dis. Poverty11, 8–26 (2022)

    2022

  4. [12]

    X Zhang, W Zhang, S Chen, Shanghai’s life-saving efforts against the current omicron wave of the covid-19 pandemic.The Lancet399, 2011–2012 (2022)

    2011

  5. [13]

    Q Zou, et al., Impact of kindergarten structures on the dynamics of hand, foot, and mouth disease and the effects of intervention strategies: an agent-based modeling study.BMC Medicine23, 357 (2025)

    2025

  6. [14]

    National health commission of the People’s Republic of China,Guidelines on Novel Coronavirus Prevention and Control., 8th edition (2021)

    2021

  7. [15]

    Shenzhen Headquarters for Prevention and Control of Pneumonia Epidemic Caused by Novel Coronavirus,The Handbook on Novel Coronavirus Prevention and Control Measures in Shenzhen., (2022)

    2022

  8. [16]

    X Wang, L Shi, Y Zhang, H Chen, G Sun, Policy disparities in fighting covid-19 among japan, italy, singapore and china.Int. J. for Equity Heal.20, 1–11 (2021)

    2021

  9. [17]

    A Stobart, S Duckett, Australia’s response to covid-19.Heal. Econ. Policy Law17, 95–106 (2022)

    2022

  10. [18]

    J Luyten, E Schokkaert, Belgium’s response to the covid-19 pandemic.Heal. Econ. Policy Law17, 37–47 (2022)

    2022

  11. [19]

    J Cumming, Going hard and early: Aotearoa new zealand’s response to covid-19.Heal. Econ. Policy Law17, 107–119 (2022)

    2022

  12. [20]

    Epidemiol Community Heal.74, 861–866 (2020)

    MJ Keeling, TD Hollingsworth, JM Read, Efficacy of contact tracing for the containment of the 2019 novel coronavirus (covid-19).J. Epidemiol Community Heal.74, 861–866 (2020)

    2019

  13. [21]

    RR Lash, et al., Covid-19 case investigation and contact tracing in the us, 2020.JAMA network open4, e2115850–e2115850 (2021)

    2020

  14. [22]

    B Y ang, et al., Universal community nucleic acid testing for coronavirus disease 2019 (covid-19) in hong kong reveals insights into transmission dynamics: a cross-sectional and modeling study.Clin. Infect. Dis.75, e216–e223 (2022)

    2019

  15. [23]

    S Wang, F Zhang, Z Wang, Z Du, C Gao, Reproduction numbers of sars-cov-2 omicron subvariants.J. T ravel. Medicine29, taac108 (2022)

    2022

  16. [24]

    Y Liu, AA Gayle, A Wilder-Smith, J Rockl¨ov, The reproductive number of covid-19 is higher compared to sars coronavirus.J. T ravel. Medicine(2020)

    2020

  17. [25]

    F Campbell, et al., Increased transmissibility and global spread of sars-cov-2 variants of concern as at june 2021.Eurosurveillance26, 2100509 (2021)

    2021

  18. [26]

    S Riley, et al., Transmission dynamics of the etiological agent of sars in hong kong: impact of public health interventions.Science300, 1961–1966 (2003)

    1961

  19. [27]

    X Tan, L Yuan, J Zhou, Y Zheng, F Y ang, Modeling the initial transmission dynamics of influenza a h1n1 in guangdong province, china.Int. J. Infect. Dis.17, e479–e484 (2013)

    2013

  20. [28]

    Infect.60, 26–35 (2010)

    SC Chen, CM Liao, Probabilistic indoor transmission modeling for influenza (sub)type viruses.J. Infect.60, 26–35 (2010)

    2010

  21. [29]

    (Princeton University Press), (2011)

    MJ Keeling, P Rohani,Modeling Infectious Diseases in Humans and Animals. (Princeton University Press), (2011)

    2011

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.