Incubation periods impact the spatial predictability of cholera and Ebola outbreaks in Sierra Leone

Abstract

Forecasting the spatiotemporal spread of infectious diseases during an outbreak is an important component of epidemic response. However, it remains challenging both methodologically and with respect to data requirements, as disease spread is influenced by numerous factors, including the pathogen’s underlying transmission parameters and epidemiological dynamics, social networks and population connectivity, and environmental conditions. Here, using data from Sierra Leone, we analyze the spatiotemporal dynamics of recent cholera and Ebola outbreaks and compare and contrast the spread of these two pathogens in the same population. We develop a simulation model of the spatial spread of an epidemic in order to examine the impact of a pathogen’s incubation period on the dynamics of spread and the predictability of outbreaks. We find that differences in the incubation period alone can determine the limits of predictability for diseases with different natural history, both empirically and in our simulations. Our results show that diseases with longer incubation periods, such as Ebola, where infected individuals can travel farther before becoming infectious, result in more long-distance sparking events and less predictable disease trajectories, as compared to the more predictable wave-like spread of diseases with shorter incubation periods, such as cholera.

Early detection of cholera epidemics to support control in fragile states: estimation of delays and potential epidemic sizes

“…Transmission started with a seed case(s) which generated secondary cases from a negative binomial distribution Z~NegB(R E , k) with a mean equivalent to the reproduction number (R E , 2.5 [13,30], reflecting early and high transmission potential among an unvaccinated population) and heterogeneity introduced by a dispersion parameter (k, 4.5, reflecting low overdispersion in R E ) [31]. Each new infection drew a time of infection from a serial interval distribution Sg amma(shape = 0.5, rate = 0.1) with a median of 5 days [4,13,32]. We assumed that at the time of outbreak detection there were 3 seed cases and that all resulting infectious persons were symptomatic.…”

Trip duration drives shift in travel network structure with implications for the predictability of spatial disease spread

“…Methods which account for uncertainty in human mobility and resulting disease dynamics therefore become especially important in order to make robust inferences about spatial transmission. For example, Kahn et al 69 showed that pathogens with longer incubation periods have patterns of spatial spread that are less predictable, which has important implications for vaccination campaigns. While the authors did not explicitly look at trip duration and network topology as we do here, these results lend support to our findings that the biological properties of pathogens that determine their speed of spatial propagation interact with the travel behavior of humans in a pathogen-specific manner, and suggests that this general relationship may be more widely applicable.…”

Discrete simulation analysis of COVID-19 and prediction of isolation bed numbers

“…To contain an outbreak, early detection of suspected cases is critical 17 . Some studies have described that a longer incubation period may be beneficial for epidemic control 18 , as this allows the Centers for Disease Control and Prevention (CDC) to have more time to deal with the overall epidemic. This conclusion may be more applicable to some known diseases, but for unknown diseases, we believe that a longer incubation period represents a more dangerous signal, making the development of the epidemic uncontrollable 18,19 .…”