The Final Size of a Serious Epidemic
 506 Downloads
Abstract
In an epidemic of a serious disease, there is likely to be behavioral response that decreases the epidemic size considerably, and taking this into account may lead to estimates of the final epidemic size that are much smaller and more realistic than estimates that do not take this into account.
Keywords
Epidemic models Final size relations Behavioral response1 Introduction
It has been standard practice in analyzing disease outbreaks to formulate a dynamical system as a deterministic compartmental model, then to use observed early outbreak data to fit parameters to the model, and finally to analyze the dynamical system to predict the course of the disease outbreak and to compare the effects of different management strategies. In general, such models predict an initial stochastic stage (while the number of infectious individuals is small), followed by a period of exponential growth. Measurement of this early exponential growth rate is an essential step in estimating contact rate parameters for the model. A thorough description of the analysis of compartmental models may be found in Hethcote (2000).
However, instances have been noted where the growth rate of an epidemic is clearly slower than exponential, especially for diseases which are viewed as very serious. For example, the initial apparently exponential spread of the 2013–2015 Ebola epidemic in West Africa can be viewed as a composition of locally asynchronous outbreaks at local level displaying subexponential growth patterns during several generations (Chowell et al. 2015). One of the earliest examples (Colgate et al. 1989) concerns the growth of HIV/AIDS in the USA, and a possible explanation might be the mixture of shortterm and longterm contacts. This could be a factor in other diseases where there are repeated contacts in family groups and less frequent contacts outside the home. There may be more than one mechanism influencing the early epidemic growth rate, and it may not be possible to estimate the effects of different mechanisms from incidence data.
A variety of epidemiological situations in which slower than exponential epidemic growth might be possible have been described (Chowell et al. 2016b, a). Ultimately, the challenge for epidemiological modeling would be to determine which of these situations allow slower than exponential growth by deriving and analyzing mechanistic models to describe each of these situations. This is an important new direction for epidemic modeling. Some suggestions include metapopulation models with spatial structure including crosscoupling and mobility, clustering in spatial structure, dynamic contacts, agentbased models with differences in infectivity and susceptibility of individuals, and reactive behavioral changes early in a disease outbreak (Chowell et al. 2016a). It may well turn out that slower than exponential growth may be ruled out in some cases but is possible in others. For example, heterogeneity of mixing in a single location can be modeled by an autonomous dynamical system and the linearization theory of dynamical systems at an equilibrium shows that early epidemic growth for such a system is always exponential. On the other hand, metapopulation models may well allow many varieties of behaviors. It is likely that more than one mechanism is at work in decreasing the growth rate, and this would complicate disentangling the mechanisms involved form the incidence data.
We suggest that an assumption of a contact rate that decreases in time, because of behavioral response to an outbreak of a disease that is often fatal, is a very plausible assumption. For example, during the SARS epidemic of 2002–2003 many people began to wear face masks to try to prevent disease spread through contact with airborne particles, even though this may have in fact been counterproductive.
Another example of a behavioral response is given by the Great Plague in the village of Eyam near Sheffield, England, which suffered an outbreak of bubonic plague in 1665–1666, the source of which is generally believed to be the Great Plague of London (Raggett 1982). The Eyam plague was survived by only 83 of an initial population of 350 persons.
In Eyam, the rector persuaded the entire community to quarantine itself to prevent the spread of disease to other communities. He assumed that infection was transmitted directly between people. While this is possible, it was thought that bubonic plague is transmitted mainly by rat fleas. When an infected rat is bitten by a flea, the flea becomes extremely hungry and bites the host rat repeatedly, spreading the infection in the rat. When the host rat dies, its fleas move on to other rats, spreading the disease further. As the number of available rats decreases, the fleas move to human hosts, and this is how plague starts in a human population. It is believed now that fleas were the main agents of the spread of plague (Dean et al. 2018).
One effect of this policy was to increase the infection rate in the village by keeping fleas, rats, and people in close contact with one another, and the mortality rate from bubonic plague was much higher in Eyam than in London. Further, the quarantine could do nothing to prevent the travel of rats and thus did little to prevent the spread of disease to other communities.
2 A VariableContactRate Epidemic Model
Another direction that would be well worth further exploration would be contact rates decreasing in time because of individual behavioral changes in response to a disease outbreak. A contact rate which is a decreasing function of time can certainly lead to early epidemic growth slower than exponential. A step in this direction has been initiated in a discrete model (Fisman et al. 2013) that has been applied to an Ebola model in Fisman et al. (2014).
Our goal is to develop a simple approach to the question of making estimates early in an outbreak of a dangerous disease for the eventual size of the epidemic, considering the total size of the epidemic over a country rather than the more complicated growth in separate villages. Study of the local structure of an epidemic is essential for attempts to control the epidemic; our intent is only to obtain early estimates of the full extent of an epidemic in order to judge the resources necessary for management.
3 Heterogeneity in epidemic models
Another possible explanation for epidemic sizes smaller than expected is heterogeneity of mixing of individuals, and we will extend our model to epidemics which include heterogeneity.
It has been observed that in many disease outbreaks, the phenomenon of superspreading events, situations in which a small fraction of the population causes more than its share of disease cases, has been significant (LloydSmith et al. 2005). For example, superspreading events were important in the SARS outbreak of 2002–2003 (Riley et al. 2003), in which there was a cluster of at least 125 cases apparently infected by a single index patient as well as another cluster of perhaps 300 cases. For a given basic reproduction number, the number of disease cases is generally fewer if there is superspreading than if the population mixing is homogeneous.
In the 2014–2015 West Africa Ebola epidemic, there is also evidence of superspreading (Lau et al. 2017). It is suggested in Galvani and May (2005) that commonly roughly 20% of the individuals in a population are responsible for 80% of the disease cases.
We have described a simple superspreading epidemic model in Brauer (2018). Here, we extend this model to include a timedependent contact rate and disease deaths.
We consider an SEIR epidemic model in a population divided into two groups. Group 1 members (superspreaders), forming a fraction \(\rho \) of the total population, have a contact rate \(\sigma a\) and Group 2 members have a contact rate a, with \(\sigma > 1\). Mixing between groups is proportionate. We take \(\sigma = 16,\rho = 0.2\) in order to conform to the suggestion in Galvani and May (2005) that commonly roughly \(20 \%\) of the individuals in a population are responsible for \(80 \%\) of the disease cases.

Total initial population size is N. Group 1 initial population size is \(N_1 = \rho N\). Group 2 initial population size is \(N_2 = (1  \rho )N\).

Group i is divided into susceptibles (\(S_i\)), exposed members (\(E_i \)), infectives (\(I_i\)), and recovered members (\(R_i\)).
 Mixing between groups is proportionate. The fraction of contacts of susceptibles with groups 1 and 2, respectively, are$$\begin{aligned} p_1 = \frac{\rho \sigma }{\rho \sigma + (1  \rho )}, \quad p_2 = \frac{1  \rho }{\rho \sigma + (1  \rho )}. \end{aligned}$$

Infections in each group recover at rate \(\gamma \).

There is a disease death rate f.

There are no demographic effects (births, deaths, migration) on the population.
4 Guinea: An Example
Simulation, using Maple, of the superspreader model (3), with \(S_1(0) = 2{,}117{,}792, S_2(0) = 8{,}491{,}168\), lead to \(S_1(\infty ) = 2{,}099{,}272, S_2(\infty ) = 8{,}486{,}508\) corresponding to 23, 220 disease cases. a slightly better estimate.
5 Sierra Leone: An Example
Simulation, using Maple, of the superspreader model (3), with \(S_1(0) = 106{,}919, S_2(0) = 4{,}276{,}784\), leads to \(S_1(\infty ) =1{,}058{,}093, S_2(\infty ) = 4{,}273{,}995\) corresponding to 13, 912 disease cases. a slightly better estimate.
6 Liberia: An Example
The Ebola epidemic in Liberia began somewhat later than the epidemics in Guinea and Sierra Leone. There was a minor outbreak in April 2014, and the first cases in the full epidemic were reported on June 16, 2014. Fisman et al. (2014) found no decrease in the reproduction number by early September 2014. The data after early September are incomplete and unreliable, and the only available estimate of a decrease in the reproduction number for Ebola in Liberia is in WHO Ebola Response Team (2014) which leads to a decrease rate of \(\kappa = 0.0018\), probably smaller than the actual rate.
Simulation, using Maple, of the superspreader model (3), with \(S_1(0) = 757{,}396, S_2(0) = 3{,}029{,}584\), leads to \(S_1(\infty ) =724{,}293, S_2(\infty ) = 3{,}021{,}121\) corresponding to 41, 636 disease cases, a considerably better estimate.
7 Conclusions
In order to plan management strategies for an outbreak of an epidemic, it is important to have an estimate early in the epidemic of how serious the epidemic will be. For an epidemic that may spread very widely, an estimate based on a simple SIR or SEIR model may predict a very large number of disease cases. If the disease is viewed as very dangerous, there may be substantial behavioral changes in the affected population even before official control measures are begun. These may lead to a decrease in the reproduction number of the epidemic and a much smaller number of disease cases than an initial prediction. We suggest that it would be very useful to estimate both the (initial) basic reproduction number and an effective reproduction number a little later in the epidemic in order to estimate how the reproduction number changes over the course of the epidemic. The estimate may be improved further by assuming heterogeneity of mixing, which appears to be an extremely common phenomenon. Reasonably accurate initial estimates are useful in planning the overall management of an epidemic, which may be quite distinct from the problems of management in individual regions.
There is a possible contradiction in this procedure. Behavioral response to an epidemic may be much larger if the public estimates of epidemic size are estimates that do not assume behavioral response. It would be important to emphasize how large the effect of even small decreases in reproduction number can be on the size of an epidemic.
Notes
Acknowledgements
The author is grateful to an anonymous reviewer who supplied several useful comments and references, as well as for the funding support received from the Natural Sciences and Engineering Research Council (Canada) Grant OGPIN 20390199
References
 Althaus CL (2014) Estimating the reproduction number of Ebola virus (EBOV) during the 2014 outbreak in West Africa. PLoS Curr Outbreaks. 2014 Sept 2 Edition. https://doi.org/10.1371/currents.outbreaks.91afb5e0f279e7f29e7056095255b288
 Brauer F (2018) Early estimates of epidemic final sizes. J Biol Dynam. https://doi.org/10.1080/17513758.2018.1469792
 Chowell G, Hengartner NW, CastilloChavez C, Fenimore PW, Hyman JM (2004) The basic reproductive number of Ebola and the effect of public health measures. J Theor Biol 229:119–126CrossRefGoogle Scholar
 Chowell G, Viboud C, Hyman JM, Simonsen L (2015) The Western Africa Ebola virus disease epidemic exhibits both global exponential and local polynomial growth rates. PLoS Curr. https://doi.org/10.1371/currents.outbreaks.8b55f4bad99ac5c5db3663e916803261
 Chowell G, Sattenspiel L, Bansal S, Viboud C (2016a) Mathematical models to characterize early epidemic growth: a review. Phys Life Rev 18:66–97Google Scholar
 Chowell G, Viboud C, Simonsen L, Moghadas S (2016b) Characterizing the reproduction number of epidemics with early subexponential growth dynamics. J R Soc Interface 13:20160659. https://doi.org/10.1098/rsif.2016.0659
 Colgate SA, Stanley EA, Hyman JM, Layne SP, Qualls C (1989) Riskbehavior based modelof the cubic growth of acquired immunodeficiency syndrome in the United States. Proc Natl Acad Sci 86:4793–4797CrossRefGoogle Scholar
 Dean KR, Krauer F, Walløe L, Lingjaerde OC, Bramanti B, Stenseth NC, Schmid BV (2018) Human ectoparasites and the spread of plague in Europe during the Second Pandemic. Proc Natl Acad Sci 115:1304–1309CrossRefGoogle Scholar
 Fisman DN, Hauck TS, Tuite AR, Greer AL (2013) An IDEA for short term outbreak projection:nearcasting using the basic reproduction number. PLoS ONE 8:1–8CrossRefGoogle Scholar
 Fisman DN, Khoo E, Tuite AR (2014) Early epidemic dynamics of the West Africa 2014 Ebola outbreak: estimates derived with a simple twoparameter model. PLoS Curr. https://doi.org/10.1371/currents.outbreaks.89c0d3783f36958d96ebbae97348d571
 Galvani AP, May RM (2005) Dimensions of superspreading. Nature 438:293–295CrossRefGoogle Scholar
 Hethcote HW (2000) The mathematics of infectious diseases. SIAM Rev 41:599–653MathSciNetCrossRefzbMATHGoogle Scholar
 Lau MSY, Dalziel BD, Funk S, McClelland A, Tiffany A, Riley S, Metcalf CJE, Grenfell BT (2017) Spatial and temporal dynamics of superspreading events in the 2014–2015 West Africa Ebola epidemic. Proc Natl Acad Sci 114:2337–2342CrossRefGoogle Scholar
 LloydSmith JO, Schreiber SJ, Kopp PE, Getz WM (2005) Superspreading and the effect of individual variation on disease emergence. Nature 438:355–359CrossRefGoogle Scholar
 Pell B, Phan T, Rutter EM, Chowell G, Kuang Y (2018) Simple multiscale modeling of the transmission dynamics of the 1905 plague epidemic in Bombay. Math Biosci 301:83–92MathSciNetCrossRefzbMATHGoogle Scholar
 Raggett GF (1982) Modeling the Eyam plague. IMA J 18:221–226zbMATHGoogle Scholar
 Riley S, Fraser C, Donnelly CA, Ghani AC, Laith JA, Hedley A, Leung GM, Ho LM, Lam TH, Thach TQ, Chau P, Chan KP, Lo SV, Leung PY, Tsang T, Ho W, Lee KH, Lau EMC, Freguson NM, Anderson RM (2003) Transmission dynamics of the etiological agent of SARS in Hong Kong: impact of public health interventions. Science 300:1961–1965CrossRefGoogle Scholar
 Shanafelt DW, Jones G, Lima M, Perrings C, Chowell G (2018) Forecasting the 2001 footandmouth disease epidemic in the UK. EcoHealth 15:338–347CrossRefGoogle Scholar
 van den Driessche P, Watmough J (2002) Reproduction numbers and subthreshold endemic equilibria for compartmental models of disease transmission. Math Biosci 180:29–48MathSciNetCrossRefzbMATHGoogle Scholar
 Viboud C, Simonsen L, Chowell G (2016) A generalizedgrowth model to characterize the early ascending phase of infectious disease outbreaks. Epidemics 15:27–37CrossRefGoogle Scholar
 WHO Ebola Response Team (2014) Ebola virus disease in West Africa—the first 9 months of the epidemic and forward projections, New England. J Med 371:1481–1495Google Scholar
Copyright information
OpenAccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.