Using heterogeneity in the population structure of U.S. swine farms to compare transmission models for porcine epidemic diarrhoea

Eamon B. O’Dea; Harry Snelson; Shweta Bansal

Losses were also apparent on a national economic scale. Producers had for the previous 8 years been making steady increases in the average litter size of about 0.16 head per year. 6 By November 2013, the average litter size had begun an abnormal downturn, 6 dropping 0.66 head by March 2014. 7 The virus also affected swine production in Asia and other parts of America. 8, 9 The mechanisms by which PEDV spread among farms are not yet clear. Transportation-associated transmission of PEDV has been supported by the observation at harvest facilities that it spreads among trailers used to transport swine, 10 and some experts believe that current resources of livestock trailers, trailer-washing facilities, and transport personnel are insufficient to allow for a standard 3-hour trailer cleaning between every load. 11 With such concerns in mind, some states responded to PED by requiring that imported swine be from PEDV-free premises. Transportation-independent mechanisms such as airborne particles 12 and contaminated feed [13] [14] [15] [16] have also been implicated. Detailed investigations of outbreaks on farms can be inconclusive regarding the mechanism of PEDV introduction. 3 Much of the research on PED involves detailed investigations on a small scale. For example, there have been epidemiological investigations of infected farms in North Carolina and a cluster of infected farms in Oklahoma and adjacent states. 17 Such work is effective for determining the biological plausibility of different routes, but the risk factors identified in a small-scale study may be specific to the small area of the study. Modelling studies based on large-scale surveillance data [18] [19] [20] can thus be a valuable complement to such work by quantifying the overall importance of a transmission route across a large population.