Disentangling the influence of livestock vs. farm density on livestock disease epidemics

Abstract

Susceptible host density is a key factor that influences the success of invading pathogens. However, for diseases affecting livestock, there are two aspects of host density: livestock and farm density, which are seldom considered independently. Traditional approaches of simulating disease outbreaks on real‐world farm data make dissecting the relative importance of farm and livestock density difficult owing to their inherent correlation in many farming regions. We took steps to disentangle these densities and study their relative influences on epidemic size by simulating foot‐and‐mouth disease outbreaks on factorial combinations of cattle and farm populations in artificial county areas, resulting in 50 unique cattle/farm density combinations. In these simulations, increasing cattle density always resulted in larger epidemics, regardless of farm density. Alternatively, increasing farm density only led to larger epidemics in scenarios of high cattle density. We compared these results with simulations performed on real‐world farm data from the United States, where we initiated outbreaks in U.S. counties that varied in county‐level cattle density and farm density. We found a similar, but weaker relationship between cattle density and epidemic size in the U.S. simulations. We tested the sensitivity of these outcomes to variation in pathogen dispersal and farm‐level susceptibility model parameters and found that although variation in these parameters quantitatively influenced the size of the epidemic, they did not qualitatively change the relative influence of cattle vs. farm density in factorial simulations. By reducing the correlation between farm and livestock density in factorial simulations, we were able to clearly demonstrate the increase in epidemic size that occurred as farm sizes grew larger (i.e., through increasing county‐level cattle populations), across levels of farm density. These results suggest livestock production trends in many industrialized countries that concentrate livestock on fewer, but larger farms have the potential to facilitate larger livestock epidemics.

Influence of contract commitment system in reducing information asymmetry, and prevention and control of livestock epidemics: Evidence from pig farmers in China

“…In response to an increasing population, urbanization, rising income, and an emerging middle class, the demand for livestock products has increased, leading to an increase of livestock epidemics such as African swine fever [ 1 , 2 ]. According to the World Health Organization report, it is documented that the livestock industry's output caused by epidemics decreased up to more than 20% each year, and posed severe challenges to meat-food safety, and further lead to the outbreak of zoonotic infectious disease [ 3 ].…”

Predictors of bovineSchistosoma japonicuminfection in rural Sichuan, China

“…japonicum infection in China (9,45), and suggests that being in close proximity to higher densities of bovine hosts may correspond with increasing infection risk, as has been found for other bovine pathogens (47,48). However, it is worth noting that household-level bovine ownership was not among the top predictors in any of our RF models, highlighting that the larger-scale lens (i.e.…”

Challenges to the Application of Spatially Explicit Stochastic Simulation Models for Foot-and-Mouth Disease Control in Endemic Settings: A Systematic Review

“…The Warwick model has been used to understand predictors of FMD transmission risk [ 53 ], identify high-risk areas [ 54 ], understand spatiotemporal process [ 55 ], evaluate mitigation strategies [ 56 , 57 ], determine optimal control strategies [ 58 , 59 ], guide policymakers [ 60 ], assist in real-time policy-making [ 61 ], understand the effect of vaccine availability constraints on epidemiologic and economic outcomes [ 62 ], estimate prevalence of asymptomatic carriers [ 63 ], understand the effect of livestock density vs. farm density [ 64 ], assess agreement between model outputs and epidemic data [ 65 ], understand the impact of the resolution of spatial data to inform control policies [ 66 ], and determine the predictor of final epidemic size [ 67 ] and computational advancement [ 68 ].…”