Journal of Mathematical Biology

, Volume 75, Issue 3, pp 521–541 | Cite as

Hybrid Markov chain models of S–I–R disease dynamics

  • Nicolas P. RebuliEmail author
  • N. G. Bean
  • J. V. Ross


Deterministic epidemic models are attractive due to their compact nature, allowing substantial complexity with computational efficiency. This partly explains their dominance in epidemic modelling. However, the small numbers of infectious individuals at early and late stages of an epidemic, in combination with the stochastic nature of transmission and recovery events, are critically important to understanding disease dynamics. This motivates the use of a stochastic model, with continuous-time Markov chains being a popular choice. Unfortunately, even the simplest Markovian S–I–R model—the so-called general stochastic epidemic—has a state space of order \(N^2\), where N is the number of individuals in the population, and hence computational limits are quickly reached. Here we introduce a hybrid Markov chain epidemic model, which maintains the stochastic and discrete dynamics of the Markov chain in regions of the state space where they are of most importance, and uses an approximate model—namely a deterministic or a diffusion model—in the remainder of the state space. We discuss the evaluation, efficiency and accuracy of this hybrid model when approximating the distribution of the duration of the epidemic and the distribution of the final size of the epidemic. We demonstrate that the computational complexity is \({\mathcal {O}}(N)\) and that under suitable conditions our approximations are highly accurate.


Markov population processes Epidemiology Fluid approximation Diffusion approximation 

Mathematics Subject Classification

60J27 60J28 92D30 60J22 60J60 


  1. Andersson H, Britton T (2000) Stochastic epidemic models and their statistical analysis, lecture notes in statistics, vol 151, 1st edn. Springer, New York. doi: 10.1007/978-1-4612-1158-7 Google Scholar
  2. Bailey NTJ (1950) A simple stochastic epidemic. Biometrika. doi: 10.1093/biomet/37.3-4.193 MathSciNetzbMATHGoogle Scholar
  3. Bailey NTJ (1957) The mathematical theory of epidemics. Griffin, LondonGoogle Scholar
  4. Ball F, Donnelly P (1995) Strong approximations for epidemic models. Stoch Proc Appl. doi: 10.1016/0304-4149(94)00034-Q MathSciNetzbMATHGoogle Scholar
  5. Ball F, Neal P (2010) Applications of branching processes to the final size of SIR epidemics. In: Workshop on branching processes and their applications, lecture notes in statistics, vol 197. Springer, Berlin, pp 207–223. doi: 10.1007/978-3-642-11156-3_15
  6. Barbour A (1975) The duration of the closed stochastic epidemic. Biometrika. doi: 10.1093/biomet/62.2.477 MathSciNetzbMATHGoogle Scholar
  7. Barbour AD (1974) On a functional central limit theorem for Markov population processes. Adv Appl Prob. doi: 10.2307/1426205 zbMATHGoogle Scholar
  8. Barbour AD (1976) Quasi-stationary distributions in Markov population processes. Adv Appl Prob. doi: 10.2307/1425906 zbMATHGoogle Scholar
  9. Barbour AD (1980a) Density dependent Markov population processes. In: Biological growth and spread, lecture notes in biomathematics, vol 38, Springer, Berlin, pp 36–49. doi: 10.1007/978-3-642-61850-5_4
  10. Barbour AD (1980b) Equilibrium distributions of Markov population processes. Adv Appl Prob. doi: 10.2307/1426422 zbMATHGoogle Scholar
  11. Bartlett MS (1949) Some evolutionary stochastic processes. J R Stat Soc 11:211–229MathSciNetzbMATHGoogle Scholar
  12. Bartlett MS (1956) Deterministic and stochastic models for recurrent epidemics. In: Proceedings of the third Berkeley symposium on mathematical statistics and probability, University of California Press, Berkeley, vol 4Google Scholar
  13. Black AJ, Ross J (2015) Computation of epidemic final size distributions. J Theor Biol. doi: 10.1016/j.jtbi.2014.11.029 Google Scholar
  14. Coulson T, Rohani P, Pascual M (2004) Skeletons, noise and population growth: the end of an old debate? Trends Ecol Evol. doi: 10.1016/j.tree.2004.05.008 Google Scholar
  15. Ethier SN, Kurtz TG (2008) Markov processes: characterisation and convergence. Wiley, Hoboken. doi: 10.1002/9780470316658
  16. Fox GA (1993) Life history evolution and demographic stochasticity. Evol Ecol. doi: 10.1007/BF01237731 Google Scholar
  17. Grenfell BT, Wilson K, Finkenstadt BF, Coulson TN, Murray S, Albon SD, Pemberton JM, Clutton-Brock TH, Crawley MJ (1998) Noise and determinism in synchronized sheep dynamics. Nature. doi: 10.1038/29291 Google Scholar
  18. Jenkinson G, Goutsias J (2012) Numerical integration of the master equation in some models of stochastic epidemiology. PLoS One. doi: 10.1371/journal.pone.0036160 Google Scholar
  19. Keeling MJ, Wilson HB, Pacala SW (2000) Reinterpreting space, time lags, and functional responses in ecological models. Science. doi: 10.1126/science.290.5497.1758 Google Scholar
  20. Kendall DG (1965) Mathematical models of the spread of infection. In: Mathematics and Computer Science in Biology and Medicine. H.M.S.O, London, pp 213–225Google Scholar
  21. Kermack WO, McKendrick AG (1927) A contribution to the mathematical theory of epidemics. Proc R Soc Lond A. doi: 10.1098/rspa.1927.0118 zbMATHGoogle Scholar
  22. Kurtz TG (1970) Solutions of ordinary differential equations as limits of pure jump Markov processes. J Appl Prob. doi: 10.2307/3212147 MathSciNetzbMATHGoogle Scholar
  23. Kurtz TG (1971) Limit theorems for sequences of jump Markov processes approximating ordinary differential processes. J Appl Prob. doi: 10.2307/3211904 MathSciNetzbMATHGoogle Scholar
  24. Lefèvre C (1990) Stochastic epidemic models for SIR infectious diseases: a brief survey of the recent general theory. In: Stochastic Processes in epidemic theory, lecture notes in biomathematics, vol 86. Springer, Berlin, pp 1–12. doi: 10.1007/978-3-662-10067-7_1
  25. Nagaev AV, Startsev AN (1970) The asymptotic analysis of a stochastic model of an epidemic. Theory Prob Appl 15(1):98–107. doi: 10.1137/1115007
  26. Rand DA, Wilson HB (1991) Chaotic stochasticity: a ubiquitous source of unpredictability in epidemics. Proc R Soc Lond B. doi: 10.1098/rspb.1991.0142 Google Scholar
  27. Safta C, Sargsyan K, Debusschere B, Najm HN (2015) Hybrid discrete/continuum algorithms for stochastic reaction networks. J Comput Phys. doi: 10.1016/ MathSciNetzbMATHGoogle Scholar
  28. Sazonov I, Kelbert M, Gravenor MB (2011) A two-stage model for the SIR outbreak: Accounting for the discrete and stochastic nature of the epidemic at the initial contamination stage. Math Biosci. doi: 10.1016/j.mbs.2011.09.002 MathSciNetzbMATHGoogle Scholar
  29. Scalia-Tomba G (1985) Asymptotic final-size distribution for some chain-binomial processes. Adv Appl Prob. doi: 10.2307/1427116 MathSciNetzbMATHGoogle Scholar
  30. Spagnolo B, Fiasconaro A, Valenti D (2003) Noise induced phenomena in Lotka-Volterra systems. Fluct Noise Lett. doi: 10.1142/S0219477503001245 zbMATHGoogle Scholar
  31. Watson R (1980) A useful random time-scale transformation for the standard epidemic model. J Appl Prob. doi: 10.2307/3213022 MathSciNetzbMATHGoogle Scholar
  32. Watson R (1981) An application of a martingale central limit theorem to the standard epidemic model. Stoc Proc Appl. doi: 10.1016/0304-4149(81)90023-5 MathSciNetzbMATHGoogle Scholar
  33. Waugh WAO (1958) Conditioned Markov processes. Biometrika. doi: 10.1093/biomet/45.1-2.241 MathSciNetzbMATHGoogle Scholar
  34. Yi S, Ulsoy A (2006) Solution of a system of linear delay differential equations using the matrix Lambert function. In: Proc. of the 25th American control conference, Minneapolis, pp 2433–2438. doi: 10.1109/ACC.2006.1656585

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  1. 1.School of Mathematical Sciences and the ARC Centre of Excellence for Mathematical and Statistical FrontiersUniversity of AdelaideAdelaideAustralia

Personalised recommendations