PT  - JOURNAL ARTICLE
AU  - Eamon B. O’Dea
AU  - Harry Snelson
AU  - Shweta Bansal
TI  - State-level transport flows are predictive of the dynamics of porcine epidemic diarrhea virus
AID  - 10.1101/017178
DP  - 2015 Jan 01
TA  - bioRxiv
PG  - 017178
4099  - http://biorxiv.org/content/early/2015/03/27/017178.short
4100  - http://biorxiv.org/content/early/2015/03/27/017178.full
AB  - More than a year after the emergence and rapid spread of porcine epidemic diarrhea virus (PEDV) in the U.S. swine herd, the extent to which the virus has spread through pathways associated with the transportation of swine remains unclear. We analyze counts of state-level, laboratory-confirmed infections to better discern the pathways by which the virus has propagated. In particular, we aim to establish and quantify any large-scale association of swine movements with the spread of PEDV. To that end, we find that the similarity of the dynamics of cases in a pair of states increases with transport flows. We find with stability selection that balance sheet variables and the number of farms in a state are likely be relevant predictors of PEDV burdens. Fitting a time series susceptible-infected-recovered model by maximum likelihood, we reject the hypothesis that flows have no effect on the transmission rate. We show with simulation how our state-level analyses may be affected by farm-level variation in risk relations. Overall, the results are consistent with the common belief that transmission is associated with swine movement and provide quantification of the strength of association.