pith. sign in

arxiv: 2604.24684 · v1 · submitted 2026-04-27 · 🧮 math.PR

Waning Immunity Fails to Restore a Positive Epidemic Threshold on Power-Law Networks

Zihao He , Souvik Dhara , Debankur Mukherjee This is my paper

Pith reviewed 2026-05-08 01:50 UTC · model grok-4.3

classification 🧮 math.PR
keywords SIRS epidemicpower-law networksconfiguration modelepidemic thresholdwaning immunityrandom graphshierarchical star
0
0 comments X

The pith

The epidemic threshold for the SIRS process remains zero on power-law networks

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

The paper establishes that introducing waning immunity in the SIRS model does not produce a positive epidemic threshold on random graphs with power-law degree distributions. This directly counters conjectures from the physics literature that expected only polynomial-time survival below some critical rate. Instead, the infection persists for an exponentially long time for any fixed positive infection rate. The result extends the zero-threshold behavior known for the SIS model and shows that temporary immunity does not change the long-term survival regime on these heterogeneous networks.

Core claim

For the SIRS epidemic process on the configuration model with power-law degree distribution with exponent tau greater than 2, the epidemic threshold is zero. The infection survives for an exponentially long time with high probability for any fixed positive infection rate. This is shown by identifying and analyzing a hierarchical star of order 2, a novel bottleneck structure that sustains the SIRS dynamics.

What carries the argument

The hierarchical star of order 2, a bottleneck structure embedded in the configuration model that sustains SIRS infection for exponentially long times with high probability.

If this is right

  • The critical infection rate separating polynomial and exponential survival regimes is zero for SIRS.
  • Infection survives exponentially long for any fixed positive infection rate on these networks.
  • Waning immunity does not restore a positive threshold, contrary to prior conjectures.
  • The hierarchical star replaces earlier star-based arguments used for SIS and works for the SIRS case.

Where Pith is reading between the lines

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

  • Mean-field approximations may systematically miss rare hierarchical structures that control long-term epidemic survival on heterogeneous networks.
  • Other models with temporary states, such as SEIRS, could exhibit similar zero thresholds on power-law networks.
  • Finite-size simulations of epidemics may need larger networks or targeted sampling of high-degree clusters to observe the predicted exponential survival.

Load-bearing premise

The configuration model with power-law degrees tau greater than 2 allows embedding of the hierarchical star of order 2, and this structure persists without depending on specific lengths of the immune or infectious periods.

What would settle it

Simulations or analysis showing that infection dies out in only polynomial time with high probability for arbitrarily small positive infection rates on large power-law networks would falsify the zero-threshold claim.

read the original abstract

In a seminal work, Chatterjee and Durrett (2009) established that for the SIS epidemic process on random graphs with power-law degree distributions, the infection survives for an exponentially long time (in the network size) for any fixed, positive infection rate. Equivalently, the critical infection rate separating polynomial and exponential survival regimes is zero. In contrast, a substantial body of work in the physics literature conjectures, based primarily on numerical evidence and heuristic mean-field arguments, that introducing waning immunity (as in the SIRS process) yields a strictly positive critical infection rate on random graphs with power-law degrees; see, e.g., Pastor-Satorras et al. (2015), Ferreira et al. (2016), Silva et al. (2022). In particular, below this threshold, the epidemic is expected to persist only for a polynomial duration. A recent work by Friedrich et al. (2024) reinforces this perspective by proving polynomial survival for the SIRS process on star graphs, which is in contrast to the exponential survival in the SIS case that underpins Chatterjee and Durrett's arguments. In this paper, we disprove this conjecture and show that the epidemic threshold is also zero for the SIRS process on the configuration model with power-law degree distribution with exponent $\tau>2$. Our proof uncovers a novel bottleneck structure for the SIRS dynamics, which we term a "hierarchical star" of order 2, and show that it sustains the infection for an exponentially long time with high probability.

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

2 major / 2 minor

Summary. The manuscript claims that the epidemic threshold for the SIRS process on the configuration model with power-law degree distributions (exponent τ > 2) is zero, disproving conjectures from the physics literature that waning immunity would produce a positive threshold. This is established by constructing a novel 'hierarchical star of order 2' subgraph that sustains infection for an exponentially long time with high probability, extending the Chatterjee-Durrett (2009) SIS result to SIRS via direct probabilistic analysis of this bottleneck structure.

Significance. If the central claim holds, the result is significant for epidemic modeling on heterogeneous networks: it shows that the zero-threshold phenomenon persists under waning immunity, challenging mean-field heuristics and numerical conjectures in Pastor-Satorras et al. (2015) and related works. The explicit construction of the hierarchical star of order 2, with its direct proof from the configuration model definition and no reduction to fitted parameters, provides a valuable new tool for analyzing survival times and distinguishes the order-2 case from single-star polynomial survival.

major comments (2)
  1. [Analysis of hierarchical star survival time] The coupling arguments and tail bounds establishing exponential survival (with high probability) for the hierarchical star of order 2 require additional detail and verification. Potential gaps in these bounds would directly undermine the load-bearing claim that the epidemic threshold remains zero for any fixed positive infection rate.
  2. [Construction and embedding of hierarchical star of order 2] The embedding of the hierarchical star of order 2 into the configuration model and its analysis should explicitly address independence from specific choices of immune period length and recovery rates. The current treatment leaves open whether variations in these parameters could alter the exponential survival bound, weakening the generality of the zero-threshold result.
minor comments (2)
  1. Clarify the precise definition of 'order 2' in the hierarchical star and how it differs quantitatively from the single-star case in terms of survival time scaling.
  2. Add a brief comparison table or statement contrasting the SIRS hierarchical-star survival with the known SIS star results to highlight the novelty.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive feedback on our manuscript. We address each major comment in detail below and will revise the manuscript accordingly to improve clarity and rigor.

read point-by-point responses

  1. Referee: [Analysis of hierarchical star survival time] The coupling arguments and tail bounds establishing exponential survival (with high probability) for the hierarchical star of order 2 require additional detail and verification. Potential gaps in these bounds would directly undermine the load-bearing claim that the epidemic threshold remains zero for any fixed positive infection rate.

    Authors: We agree that the presentation of the coupling arguments and tail bounds can be strengthened for greater transparency. In the revised manuscript, we will expand the relevant section by providing a more detailed step-by-step description of the coupling between the hierarchical star process and the full epidemic dynamics, along with explicit derivations of the tail bounds and the resulting exponential survival probability. These additions will verify the bounds without changing the underlying probabilistic arguments. revision: yes

  2. Referee: [Construction and embedding of hierarchical star of order 2] The embedding of the hierarchical star of order 2 into the configuration model and its analysis should explicitly address independence from specific choices of immune period length and recovery rates. The current treatment leaves open whether variations in these parameters could alter the exponential survival bound, weakening the generality of the zero-threshold result.

    Authors: The construction of the hierarchical star of order 2 is embedded directly in the configuration model and the survival analysis holds for arbitrary fixed positive values of the infection rate, recovery rate, and waning immunity rate. The reinfection cycles sustained by the order-2 structure are robust to these parameters as long as they remain positive and finite. To make this explicit, we will add a dedicated remark in the revised manuscript stating the uniformity of the exponential survival bound with respect to the recovery and waning rates, together with a short justification that variations in their specific values (while keeping them positive) do not affect the exponential lower bound. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained

full rationale

The paper defines a novel 'hierarchical star of order 2' subgraph explicitly from the configuration model with power-law degrees and proves its exponential survival for SIRS dynamics using direct probabilistic embedding and coupling arguments. No fitted parameters are renamed as predictions, no ansatz is smuggled via citation, and the central threshold-zero claim does not reduce to any self-citation or self-definitional loop. Cited prior works (Chatterjee-Durrett, Friedrich et al.) provide contrast but are not load-bearing for the new construction, which stands on its own model definitions.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The claim rests on standard configuration-model properties and the newly defined hierarchical star; no free parameters are fitted and no new physical entities are postulated beyond the proof device.

axioms (2)
  • standard math The configuration model generates a simple graph with the prescribed power-law degree sequence for τ>2.
    Invoked to embed the hierarchical star and apply concentration results.
  • domain assumption Infection, recovery, and waning rates are positive constants independent of network size.
    Required for the exponential survival time to hold uniformly.
invented entities (1)
  • hierarchical star of order 2 no independent evidence
    purpose: Bottleneck structure that sustains SIRS infection for exponentially long time
    Newly defined subgraph consisting of a central hub connected to secondary hubs whose leaves maintain the infection chain despite waning immunity.

pith-pipeline@v0.9.0 · 5585 in / 1438 out tokens · 62848 ms · 2026-05-08T01:50:58.591349+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

30 extracted references · 30 canonical work pages

  1. [1]

    R. M. Anderson and R. M. May.Infectious Diseases of Humans: Dynamics and Control. Oxford University Press, 1991

    work page 1991

  2. [2]

    Andersson and T

    H. Andersson and T. Britton.Stochastic Epidemic Models and Their Statistical Analysis, volume 151 ofLecture Notes in Statistics. Springer, New York, NY, 1 edition, 2000. ISBN 978-0-387- 95050-1. doi: 10.1007/978-1-4612-1158-7

    work page doi:10.1007/978-1-4612-1158-7 2000

  3. [3]

    F. Ball, T. Britton, M. Deijfen, et al. Central Limit Theorems for SIR Epidemics and Percola- tion on Configuration Model Graphs.Annals of Applied Probability, 31(5):2235–2272, 2021

    work page 2021

  4. [4]

    Berger, C

    N. Berger, C. Borgs, J. T. Chayes, and A. Saberi. On the Spread of Viruses on the Internet. InProceedings of the Sixteenth Annual ACM–SIAM Symposium on Discrete Algorithms (SODA 2005), pages 301–310, New York, NY, USA, 2005. ACM. electronic

    work page 2005

  5. [5]

    Bhamidi, D

    S. Bhamidi, D. Nam, O. Nguyen, and A. Sly. Survival and Extinction of Epidemics on Random Graphs with General Degrees.The Annals of Probability, 49(1):244–286, 2021. doi: 10.1214/20-AOP1451

    work page doi:10.1214/20-aop1451 2021

  6. [6]

    Bhamidi, D

    S. Bhamidi, D. Nam, O. Nguyen, and A. Sly. The Contact Process on Random Graphs and Galton–Watson Trees.Annals of Probability, 2023. Preprint arXiv:1902.03263. 59

    work page arXiv 2023

  7. [7]

    Chatterjee and R

    S. Chatterjee and R. Durrett. Contact Processes on Random Graphs with Power Law Degree Distributions Have Critical Value 0.Annals of Probability, 37(6):2332–2356, 2009. doi: 10. 1214/09-AOP471

    work page 2009

  8. [8]

    Dhara, R

    S. Dhara, R. van der Hofstad, J. S. H. van Leeuwaarden, and S. Sen. Heavy-tailed Configuration Models at Criticality.arXiv preprint arXiv:1612.00650, 2019. URLhttps: //arxiv.org/abs/1612.00650. v3

    work page arXiv 2019

  9. [9]

    Draief and L

    M. Draief and L. Massouli ´e.Epidemics and Rumours in Complex Networks. London Mathe- matical Society Lecture Note Series. Cambridge University Press, 2009

    work page 2009

  10. [10]

    R. Durrett. Ten lectures on particle systems, 1995. Lecture notes (Saint-Flour 1993)

    work page 1995

  11. [11]

    S. C. Ferreira, R. S. Sander, and R. Pastor-Satorras. Collective versus hub activation of epidemic phases on networks.Physical Review E, 93(3):032314, 2016. doi: 10.1103/PhysRevE. 93.032314

    work page doi:10.1103/physreve 2016

  12. [12]

    D. A. Freedman. On tail probabilities for martingales.The Annals of Probability, 3(1):100–118,

    work page

  13. [13]

    doi: 10.1214/aop/1176996452

    work page doi:10.1214/aop/1176996452

  14. [14]

    Friedrich, A

    T. Friedrich, A. G ¨obel, N. Klodt, M. S. Krejca, and M. Pappik. Analysis of the Survival Time of the SIRS Process via Expansion.Electronic Journal of Probability, 29:1–29, 2024. doi: 10.1214/24-EJP1140

    work page doi:10.1214/24-ejp1140 2024

  15. [15]

    doi: 10.1017/CBO9781316339831

    A. Frieze and M. Karo ´nski.Introduction to Random Graphs. Cambridge University Press, Cambridge, 2015. doi: 10.1017/CBO9781316339831

    work page doi:10.1017/cbo9781316339831 2015

  16. [16]

    T. E. Harris. Nearest-neighbor Markov interaction processes on multidimensional lattices. Advances in Mathematics, 9(1):66–89, 1972. doi: 10.1016/0001-8708(72)90030-8

    work page doi:10.1016/0001-8708(72)90030-8 1972

  17. [17]

    Huang and R

    X. Huang and R. Durrett. The contact process on random graphs and galton–watson trees. ALEA, Latin American Journal of Probability and Mathematical Statistics, 17:159–182, 2020. doi: 10.30757/ALEA.v17-07

    work page doi:10.30757/alea.v17-07 2020

  18. [18]

    S. Janson. The probability that a random multigraph is simple.Combinatorics, Probability and Computing, 18(1-2):205–225, 2009. doi: 10.1017/S0963548308009644

    work page doi:10.1017/s0963548308009644 2009

  19. [19]

    Janson, M

    S. Janson, M. Luczak, and P . Windridge. Law of Large Numbers for the SIR Epidemic on a Random Graph with Given Degrees.Random Structures & Algorithms, 45(4):726–763, 2014

    work page 2014

  20. [20]

    M. J. Keeling and P . Rohani.Modeling Infectious Diseases in Humans and Animals. Princeton University Press, 2008

    work page 2008

  21. [21]

    P . Lam, O. Nguyen, and I. Yang. Optimal Bound for Survival Time of the SIRS Process on Star Graphs. arXiv:2412.21138, 2024. URLhttps://arxiv.org/abs/2412.21138

    work page arXiv 2024

  22. [22]

    A. A. Lashari, A. Serafimovi ´c, and P . Trapman. The Duration of a Supercritical SIR Epidemic on a Configuration Model.Electronic Journal of Probability, 26:1–54, 2021

    work page 2021

  23. [23]

    T. M. Liggett.Interacting Particle Systems, volume 276 ofGrundlehren der mathematischen Wissenschaften. Springer-Verlag, New York, 1985. 60

    work page 1985

  24. [24]

    Mountford, J.-C

    T. Mountford, J.-C. Mourrat, D. Valesin, and Q. Yao. Exponential extinction time of the contact process on finite graphs.Stochastic Processes and their Applications, 123(11):4046–4073,

    work page

  25. [25]

    doi: 10.1016/j.spa.2013.05.007

    work page doi:10.1016/j.spa.2013.05.007 2013

  26. [26]

    Mountford, D

    T. Mountford, D. Valesin, and Q. Yao. Metastable densities for the contact process on random graphs.Annals of Applied Probability, 26(3):1483–1510, 2016. doi: 10.1214/15-AAP1128

    work page doi:10.1214/15-aap1128 2016

  27. [27]

    Pastor-Satorras, C

    R. Pastor-Satorras, C. Castellano, P . Van Mieghem, and A. Vespignani. Epidemic pro- cesses in complex networks.Reviews of Modern Physics, 87(3):925–979, 2015. doi: 10.1103/ RevModPhys.87.925

    work page 2015

  28. [28]

    J. C. M. Silva, D. H. Silva, F. A. Rodrigues, and S. C. Ferreira. Comparison of theoretical approaches for epidemic processes with waning immunity in complex networks.Physical Review E, 106(3):034317, 2022. doi: 10.1103/PhysRevE.106.034317

    work page doi:10.1103/physreve.106.034317 2022

  29. [29]

    J. A. Tropp. Freedman’s inequality for matrix martingales.Electronic Communications in Probability, 16:262–270, 2011. doi: 10.1214/ECP .v16-1624

    work page doi:10.1214/ecp 2011

  30. [30]

    van der Hofstad.Random Graphs and Complex Networks, volume 2

    R. van der Hofstad.Random Graphs and Complex Networks, volume 2. Cambridge University Press, Cambridge, 2024. 61

    work page 2024