The unseen role of vehicle movements

Evaluating disease transmission pathways in swine farms

By Jason A. Galvis and Gustavo Machado, North Carolina State UniversityThe role of vehicles in spreading diseases between farms is emerging as a crucial but underrecognized pathway for transmitting infectious diseases, particularly affecting the swine industry. The link between vehicle movements and outbreaks of diseases such as porcine epidemic diarrhea and African swine fever has been established (Adedeji et al., 2022; Boniotti et al., 2018; Cheng and Ward, 2022; Garrido‐Mantilla et al., 2022; Lowe et al., 2014; Nigsch et al., 2013; Yoo et al., 2021), yet the precise risk these movements pose remains unclear. This uncertainty stems from an absence of information about pathogen viability on vehicle surfaces and the efficacy of cleaning and disinfection to eliminate such pathogens.This research introduces an innovative approach to reproduce the indirect contact network between farms by vehicle movements (Figure 1). This approach integrates the concept of pathogen stability, which decreases over time with variations in environmental temperatures (Nuanualsuwan et al., 2022). It also accounts for potentially eliminating pathogens through cleaning and disinfection procedures.Given the unknown efficacy of these cleaning measures, we simulated scenarios with different levels of cleaning effectiveness: 0%, 10%, 50%, 80%, 90% and 100%, activated whenever a vehicle underwent a cleaning process. To determine the risk of vehicles transmitting infections due to their proximity to farms, we utilized the Perimeter Buffer Area (PBA) data from RABapp^TM (Machado et al., 2023), defining a farm as being in contact with a vehicle if it entered a specified distance from the PBA. We analyzed vehicle movements from 823 vehicles affiliated with three swine production companies in two U.S. regions, categorizing these vehicles into six types based on their cargo: humans (crew), pigs-to-farms, pigs-to-markets, feed, undefined and a composite category of all vehicle types (combined-vehicles). For each category, we constructed a contact network. We assessed the farms within the infection chain, defined as the median number of farms at risk of infection over a year, based on the sequence of vehicle contacts in the network (Figure 1.)

Figure 1. Network reconstruction framework. A vehicle visited a farm if the vehicle’s latitude and longitude were inside a buffer distance and for a minimum time specified by the visit time (red box, top-left panel). An edge connecting different farms is recorded if all farms were visited by the exact vehicle in decreasing chronological order and if the edge weight (E), representing our pathogen stability, is higher than 0 (green box, bottom-right panel). In the example, a vehicle visited four different farms, creating edges from A1→ B1 and A1→C1, and B1→ C1, while no edges were recorded from A1, B1, and C1 to D1 because the vehicle stopped at a C&D before visiting D1 and the effectivity of cleaning d was high. The weight edges among the farms are calculated through an exponential distribution, where 𝜆 is the decay rate for each average temperature (⍵) from the source of the contact (e.g., farm A1, green dot) until the destination (e.g., farms B1 and C1, red dots). Similarly, Γ is the cumulative time from the source of the contact (e.g., farm A1, green dot) until the destination (e.g., farms B1 and C1, red dots) (Galvis and Machado, 2022).

In the first region, the analysis revealed that without effective cleaning (0% efficacy), the combined vehicles could potentially infect up to 2,157 farms through the contacts available in the vehicle contact network, a value comparable to the number of farms in the infection chain of feed transport vehicles (Figure 2).

Figure 2. The number of farms potentially infected via vehicle movements. Solid lines represent the median, while shadowed areas represent the interquartile ranges from ten simulations performed for each cleaning probability. R1 and R2 identify the different study regions. Combined vehicles represent the combination of all vehicles movements from region one: feed-vehicle are vehicles transporting feed to farms, pig-farm-vehicle are vehicles transporting live pigs to farms, pig-market-vehicle are vehicles transporting live pigs to market, crew-vehicle vehicles are vehicles transporting farm personnel, and undefined-vehicles are vehicles without a defined role (Galvis and Machado, 2022).

Galvis is a postdoctoral researcher and Machado is an assistant professor of population health and pathobiology, both at NC State.