Assessing intrastate shipments from interstate data and expert opinion

Abstract

Live animal shipments are a potential route for transmitting animal diseases between holdings and are crucial when modelling spread of infectious diseases. Yet, complete contact networks are not available in all countries, including the USA. Here, we considered a 10% sample of Interstate Certificate of Veterinary Inspections from 1 year (2009). We focused on distance dependence in contacts and investigated how different functional forms affect estimates of unobserved intrastate shipments. To further enhance our predictions, we included responses from an expert elicitation survey about the proportion of shipments moving intrastate. We used hierarchical Bayesian modelling to estimate parameters describing the kernel and effects of expert data. We considered three functional forms of spatial kernels and the inclusion or exclusion of expert data. The resulting six models were ranked by widely applicable information criterion (WAIC) and deviance information criterion (DIC) and evaluated through within- and out-of-sample validation. We showed that predictions of intrastate shipments were mildly influenced by the functional form of the spatial kernel but kernel shapes that permitted a fat tail at large distances while maintaining a plateau-shaped behaviour at short distances better were preferred. Furthermore, our study showed that expert data may not guarantee enhanced predictions when expert estimates are disparate.

The Importance of Livestock Demography and Infrastructure in Driving Foot and Mouth Disease Dynamics

“…This data set remains the most geographically comprehensive cattle shipment data set available, and USAMM has been constructed specifically to scale up this limited sample of inter-state shipment data to complete nation-wide cattle movement networks consisting of both inter- and intra-state shipments. Two versions of USAMM have been published previously—USAMMv1.0 [ 11 ] and and USAMMv2.0 [ 16 ]. In USAMMv1.0, shipments are modeled using a state-level distance dependence together with state-level covariates consisting of historical inflow data, and county-level weights scaling linearly with the total number of premises in the county.…”

“…In USAMMv2.1, we built on the earlier USAMMv2.0 movement model of Brommesson et al [ 16 ] to improve the previous statistical model for the probability of observing a set of interstate cattle shipments given by the 2009 ICVI data. In this section, we describe this previous model, and in the next section outline the changes that were made to it in the present work.…”

“…However it is also possible for the model to disregard one or both of them for certain parameter values. The latter is equivalent to the base model in Brommesson et al [ 16 ] which can be viewed as a special case of USAMMv2.1 where each county has an effect of industry covariates equal to one (i.e., no effect of covariates), and where every premises has equal weight regardless of size and type. A full mathematical description and details on parameter estimation are given in Supplementary Section S1.3 .…”

“…Because livestock shipments in the U.S. are not tracked unless they cross a state line, the shipment-based transmission in USDOS is predicted by the United States Animal Movement Model (USAMM). USAMM estimates the full U.S. cattle shipment network based on partially observed data [ 11 , 16 ]. In this study, we add additional detail to these two models in order to asses the impact of structural uncertainty while preserving the ability to simulate FMD at the U.S. national scale.…”

Simulating contact networks for livestock disease epidemiology: a systematic review

“…Building on these, the USAMM model [25], which has been applied and modified extensively [76][77][78][79][80][81][82][83][84], uses a Bayesian kernel approach to reconstruct the US cattle trade network. Similarly to GMs, movement probabilities were modelled as a function of the number of cattle premises at the origin and destination, and the distance between them, while also incorporating data on historical state-level cattle inflows.…”