pith. sign in

arxiv: 1906.08631 · v1 · pith:72BQOMKAnew · submitted 2019-06-18 · 🧬 q-bio.PE · physics.soc-ph· stat.AP

On the use of multiple compartment epidemiological models to describe the dynamics of influenza in Europe

Inbar Seroussi , Nir Levy , Daniela Paolotti , Nir Sochen , Elad Yom-Tov This is my paper

Pith reviewed 2026-05-25 20:12 UTC · model grok-4.3

classification 🧬 q-bio.PE physics.soc-phstat.AP
keywords influenzaSIR modelEuropehuman mobilityepidemiologyparameter estimationdisease surveillancestrain analysis
0
0 comments X

The pith

Multiple compartment SIR models show inter-country mobility plays a negligible role in influenza spread across Europe.

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

The paper develops a multiple compartment Susceptible-Infected-Recovered model to analyze spread of infectious diseases through different geographic areas and proposes a data-quality sensitive optimization framework for fitting the model to observed data. Fitting this model to incidence data for six influenza strains over seven seasons in Europe estimates inter-country and intra-country infection and recovery rates for each strain and season. The results indicate that disease parameters remain relatively stable with correlation greater than 0.5 over seasons and strains, with clustering by parameters consistent with genome sub-types. The analysis suggests inter-country human mobility has a negligible role in spread, while the model can estimate disease load in countries with poor data from adjacent countries.

Core claim

Fitting a multiple-compartment SIR model to European influenza incidence data over seven seasons reveals stable disease parameters across seasons and strains with correlations above 0.5, consistency with genomic sub-types in parameter-based clustering, negligible contribution from inter-country mobility to spread, and the capacity to infer disease burden in data-limited countries from neighboring ones.

What carries the argument

The multiple compartment SIR model incorporating separate intra-country and inter-country infection and recovery rates, fitted using a data-quality sensitive optimization framework.

If this is right

  • The model provides estimates of disease parameters usable for disease surveillance and control of influenza and other pathogens.
  • The stability of parameters supports consistent application across different seasons and strains.
  • The model enables estimation of disease load in countries lacking data by leveraging data from adjacent countries.
  • Clustering of strains by inferred parameters aligns with their genome sub-types.

Where Pith is reading between the lines

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

  • Travel restrictions between European countries would likely have limited effect on influenza dynamics.
  • The same modeling approach could be applied to other pathogens in similar multi-country settings.
  • Parameter stability over seasons opens the possibility of using partial data for real-time forecasting.

Load-bearing premise

The observed incidence data combined with the compartment structure can distinguish intra-country transmission rates from inter-country mobility effects without major confounding.

What would settle it

Demonstrating that models excluding inter-country mobility terms achieve comparable or superior fits to the data, or that the estimated mobility parameters are not consistently small.

Figures

Figures reproduced from arXiv: 1906.08631 by Daniela Paolotti, Elad Yom-Tov, Inbar Seroussi, Nir Levy, Nir Sochen.

Figure 1
Figure 1. Figure 1: An histogram of the Pearson correlation between the predicted number of infected [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: A box plot of the Spearman correlation between the infection rates and recovery [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: A box plot of the Spearman correlation between the infection rates and recovery [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Hierarchical plots of clustering 6 influenza strains according to the on-diagonal [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Estimated infection rates in the case of missing data. In red is the mcSIR [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Histogram of the correlations between the actual infection rate in the removed [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: An histogram of the Pearson correlations achieved between the predicted infection [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: A box plot of the Spearman correlation between the infection rates and recovery [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: A box plot of the Spearman correlation between the infection rates and recovery [PITH_FULL_IMAGE:figures/full_fig_p014_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Linkage plots of the average Euclidean distance of the on diagonal coefficients in [PITH_FULL_IMAGE:figures/full_fig_p014_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: (a-b) The Pearson correlation between the noisy simulated data of 5 countries [PITH_FULL_IMAGE:figures/full_fig_p015_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Results of the model fitting to mcSIR simulation over 6 countries. The data [PITH_FULL_IMAGE:figures/full_fig_p016_12.png] view at source ↗
read the original abstract

We develop a multiple compartment Susceptible-Infected-Recovered (SIR) model to analyze the spread of several infectious diseases through different geographic areas. Additionally, we propose a data-quality sensitive optimization framework for fitting this model to observed data. We fit the model to the temporal profile of the number of people infected by one of six influenza strains in Europe over $7$ influenza seasons. In addition to describing the temporal and spatial spread of influenza, the model provides an estimate of the inter-country and intra-country infection and recovery rates of each strain and in each season. We find that disease parameters remain relatively stable, with a correlation greater than $0.5$ over seasons and stains. Clustering of influenza strains by the inferred disease parameters is consistent with genome sub-types. Surprisingly, our analysis suggests that inter-country human mobility plays a negligible role in the spread of influenza in Europe. Finally, we show that the model allows the estimation of disease load in countries with poor or none existent data from the disease load in adjacent countries. Our findings reveal information on the spreading mechanism of influenza and on disease parameters. These can be used to assist in disease surveillance and in control of influenza as well as of other infectious pathogens in a heterogenic environment.

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 develops a multi-compartment SIR model with explicit inter-country coupling terms and a data-quality-weighted optimization procedure. It fits the model to weekly incidence time series for six influenza strains across European countries over seven seasons, estimates intra- and inter-country transmission and recovery rates, reports parameter correlations >0.5 across seasons and strains, concludes that inter-country mobility effects are negligible, and demonstrates imputation of incidence in data-sparse countries from neighboring data.

Significance. If the separation of mobility from local transmission proves robust, the work would supply quantitative evidence on the stability of influenza parameters across strains and seasons together with a practical framework for spatial imputation, both of which could inform surveillance and intervention planning in heterogeneous European settings.

major comments (2)
  1. [Abstract / Model fitting] Abstract and model-fitting section: the central claim that inter-country human mobility plays a negligible role rests on the fitted inter-country transmission coefficients being small relative to intra-country ones, yet no identifiability diagnostics (profile likelihood, Hessian eigenvalues, or synthetic-data recovery experiments) are reported; without these it remains possible that mobility influxes are absorbed into adjustments of the local β or the data-quality weight.
  2. [Results] Results and validation: the manuscript provides no comparison of the multi-compartment model against single-compartment SIR baselines on the same incidence series, nor any out-of-sample prediction on held-out seasons; this absence prevents assessment of whether the added inter-country terms and data-quality weighting actually improve explanatory power or merely overfit the observed curves.
minor comments (2)
  1. [Abstract] Abstract: typographical errors (“stains” for “strains”, “none existent” for “non-existent”) should be corrected.
  2. [Abstract / Methods] The abstract states that the model “provides an estimate” of rates but supplies neither the governing equations nor the explicit form of the data-quality weight; these should appear in the main text or supplementary material for reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below. We agree that the suggested analyses would strengthen the claims and will incorporate them.

read point-by-point responses

  1. Referee: [Abstract / Model fitting] Abstract and model-fitting section: the central claim that inter-country human mobility plays a negligible role rests on the fitted inter-country transmission coefficients being small relative to intra-country ones, yet no identifiability diagnostics (profile likelihood, Hessian eigenvalues, or synthetic-data recovery experiments) are reported; without these it remains possible that mobility influxes are absorbed into adjustments of the local β or the data-quality weight.

    Authors: We acknowledge that formal identifiability diagnostics were not reported. The high parameter correlations (>0.5) across seven independent seasons and six strains provide indirect evidence that the estimates, including the consistently small inter-country coefficients, are robust rather than artifacts of non-identifiability. Nevertheless, to directly address the concern, we will add synthetic-data recovery experiments and profile likelihood analysis in the revision. revision: yes

  2. Referee: [Results] Results and validation: the manuscript provides no comparison of the multi-compartment model against single-compartment SIR baselines on the same incidence series, nor any out-of-sample prediction on held-out seasons; this absence prevents assessment of whether the added inter-country terms and data-quality weighting actually improve explanatory power or merely overfit the observed curves.

    Authors: We agree that explicit baseline comparisons and out-of-sample validation are needed to quantify the benefit of the inter-country terms. The original manuscript focuses on the multi-compartment model, parameter stability, and imputation utility. We will add single-compartment SIR comparisons on the same series and out-of-sample predictions on held-out seasons in the revised version. revision: yes

Circularity Check

0 steps flagged

No circularity: parameter estimates and their properties are direct outputs of fitting, not reductions by construction

full rationale

The paper defines a multi-compartment SIR model with intra- and inter-country coupling terms, fits all rates jointly to incidence time series via a data-quality-sensitive optimizer, and then reports descriptive statistics on the resulting parameter values (e.g., cross-season correlations >0.5 and relative size of inter-country coefficients). These statistics are computed directly from the fitted values; they are not presented as out-of-sample predictions, nor do any equations reduce the reported findings to the inputs by definition. No self-citations, uniqueness theorems, or ansatzes are invoked to justify the central claims. The procedure is a standard inference pipeline whose outputs can be externally validated or falsified against independent mobility or genomic data.

Axiom & Free-Parameter Ledger

4 free parameters · 2 axioms · 0 invented entities

The model introduces several fitted transmission and recovery rates per strain and season plus a data-quality weighting scheme; these are free parameters. No new physical entities are postulated. The compartment connectivity structure is a domain assumption.

free parameters (4)
  • intra-country infection rate
    Fitted per strain and season to incidence data
  • inter-country infection rate
    Fitted per strain and season; central to mobility conclusion
  • recovery rate
    Fitted per strain and season
  • data-quality weight
    Chosen to modulate the optimization objective
axioms (2)
  • domain assumption Incidence time series accurately reflect true infection dynamics without systematic reporting bias
    Required for any parameter inference from case counts
  • domain assumption The chosen multi-compartment topology captures the dominant spatial mixing patterns
    Structural assumption that enables separation of intra- and inter-country rates

pith-pipeline@v0.9.0 · 5774 in / 1352 out tokens · 18662 ms · 2026-05-25T20:12:35.871166+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

23 extracted references · 23 canonical work pages

  1. [1]

    Contributions to the mathematical theory of epidemics

    Kermack WO, McKendrick AG. Contributions to the mathematical theory of epidemics. II.The problem of endemicity. Proc R Soc Lond A. 1932;138(834):55–83

    work page 1932

  2. [2]

    Human mobility and the spatial transmission of influenza in the United States

    Charu V, Zeger S, Gog J, Bjørnstad ON, Kissler S, Simonsen L, et al. Human mobility and the spatial transmission of influenza in the United States. PLoS computational biology. 2017;13(2):e1005382

    work page 2017

  3. [3]

    Modeling Influenza-like illnesses through composite compart- mental models

    Levy N, Iv M, Yom-Tov E. Modeling Influenza-like illnesses through composite compart- mental models. Physica A: Statistical Mechanics and its Applications. 2018;494:288– 293

    work page 2018

  4. [4]

    Conjunction of factors triggering waves of seasonal Influenza

    Chattopadhyay I, Kiciman E, Elliott JW, Shaman JL, Rzhetsky A. Conjunction of factors triggering waves of seasonal Influenza. eLife. 2018;7

    work page 2018

  5. [5]

    On state-space reduction in multi- strain pathogen models, with an application to antigenic drift in influenza A

    Kryazhimskiy S, Dieckmann U, Levin SA, Dushoff J. On state-space reduction in multi- strain pathogen models, with an application to antigenic drift in influenza A. PLoS computational biology. 2007;3(8):e159

    work page 2007

  6. [6]

    The role of the airline transportation network in the prediction and predictability of global epidemics

    Colizza V, Barrat A, Barth´ elemy M, Vespignani A. The role of the airline transportation network in the prediction and predictability of global epidemics. Proceedings of the National Academy of Sciences. 2006;103(7):2015–2020

    work page 2006

  7. [7]

    Multiscale mobility networks and the spatial spreading of infectious diseases

    Balcan D, Colizza V, Gon¸ calves B, Hu H, Ramasco JJ, Vespignani A. Multiscale mobility networks and the spatial spreading of infectious diseases. Proceedings of the National Academy of Sciences. 2009; p. pnas–0906910106

    work page 2009

  8. [8]

    Large-scale spatial-transmission models of infectious disease

    Riley S. Large-scale spatial-transmission models of infectious disease. Science. 2007;316(5829):1298–1301

    work page 2007

  9. [9]

    Spatial heterogeneity and the persistence of infectious diseases

    Hagenaars T, Donnelly C, Ferguson N. Spatial heterogeneity and the persistence of infectious diseases. Journal of theoretical biology. 2004;229(3):349–359. 16

    work page 2004

  10. [10]

    Combin- ing participatory influenza surveillance with modeling and forecasting: three alternative approaches

    Brownstein JS, Chu S, Marathe A, Marathe MV, Nguyen AT, Paolotti D, et al. Combin- ing participatory influenza surveillance with modeling and forecasting: three alternative approaches. JMIR public health and surveillance. 2017;3(4)

    work page 2017

  11. [11]

    The Geographic Spread of Infectious Diseases: Models and Applications: Models and Applications

    Sattenspiel L. The Geographic Spread of Infectious Diseases: Models and Applications: Models and Applications. Princeton University Press; 2009

    work page 2009

  12. [12]

    Multi-Season Analysis Reveals the Spatial Structure of Disease Spread

    Seroussi I, Levy N, Yom-Tov E. Multi-Season Analysis Reveals the Spatial Structure of Disease Spread. arXiv:190204073. 2019

    work page 2019

  13. [13]

    Modeling infectious diseases in humans and animals

    Keeling MJ, Rohani P. Modeling infectious diseases in humans and animals. Princeton University Press; 2011

    work page 2011

  14. [14]

    Mathematical methods of classical mechanics

    Arnol’d VI. Mathematical methods of classical mechanics. vol. 60. Springer Science & Business Media; 2013

    work page 2013

  15. [15]

    Available from: http://apps.who.int/flumart/Default?ReportNo=12

    Surveillance information provided by the World Health Organization (WHO), available from;. Available from: http://apps.who.int/flumart/Default?ReportNo=12

    work page

  16. [16]

    Detecting influenza epidemics using search engine query data

    Ginsberg J, Mohebbi MH, Patel RS, Brammer L, Smolinski MS, Brilliant L. Detecting influenza epidemics using search engine query data. Nature. 2009;457(7232):1012

    work page 2009

  17. [17]

    Available from: https://ecdc.europa.eu/en/seasonal-influenza/ surveillance-and-disease-data/flu-news-europe

    European Centre for Disease Prevention and Control, available from;. Available from: https://ecdc.europa.eu/en/seasonal-influenza/ surveillance-and-disease-data/flu-news-europe

    work page

  18. [18]

    The role of population heterogeneity and human mobility in the spread of pandemic influenza

    Merler S, Ajelli M. The role of population heterogeneity and human mobility in the spread of pandemic influenza. Proceedings of the Royal Society B: Biological Sciences. 2009;277(1681):557–565

    work page 2009

  19. [19]

    Multistrain Infections in Metapopulations

    Garmer S, Lynn R, Rossi D, Capaldi A. Multistrain Infections in Metapopulations. Spora: A Journal of Biomathematics. 2015;1(1):4

    work page 2015

  20. [20]

    Metapopulation epidemic models with heterogeneous mixing and travel behaviour

    Apolloni A, Poletto C, Ramasco JJ, Jensen P, Colizza V. Metapopulation epidemic models with heterogeneous mixing and travel behaviour. Theoretical Biology and Med- ical Modelling. 2014;11(1):3

    work page 2014

  21. [21]

    Available from: https://ec.europa.eu/eurostat/statistics-explained/ index.php/Passenger_transport_statistics

    Human mobility in Europe provided by the Eurostat Statistics Explained, available from;. Available from: https://ec.europa.eu/eurostat/statistics-explained/ index.php/Passenger_transport_statistics

    work page

  22. [22]

    MATLAB Statistics and Machine Learning Toolbox; 2017b

    work page

  23. [23]

    Particle swarm optimization

    Eberhart R, Kennedy J. Particle swarm optimization. In: Proceedings of the IEEE international conference on neural networks. vol. 4. Citeseer; 1995. p. 1942–1948. 17

    work page 1995