In this study, we sought to identify factors that were associated with PEDv outbreaks on a sample of 190 sow farms from within the Swine Vet Center practice. Additionally, we investigated the role of PRRSv in these infections. The objective of this study was to identify factors that were associated with either of these diseases, and use this information to help producers guide the decision making process on their farms in an effort to reduce the frequency of these (and likely other) diseases.
We began by plotting the locations of al farms on the map. Using a spatial temporal scan statistic, we identified 5 areas, or disease clusters, where PEDv and/or PRRSv infected farms occurred 4.5 – 20 times in excess of the expected based on a random distribution of the cases (p < 0.05). Of particular interest was an area in South West Minnesota, and North West Iowa where PRRSv and PEDv occurred together more commonly than expected. This is a similar area to what has been previously reported to have had a higher than expected amount of PRRS.
Using this information, we tested the data for associations between being positive for PRRSv and/or PEDv with specific farm management practices including production company the farm belonged to, production type (commercial or genetic), filtration status, frequency of previous PRRSv infections and PRRSv vaccine use in the past 4 years, feed mill type (toll or owned and size), carcass disposal (external rendering or onsite method) and whether a particular herd owned or contracted (exclusively or not) any trucking, manure pumping, or facilities and high biosecurity (aerosol filtration). Additionally, we included herd size, county density, and whether or not the farm was located in one of the disease clusters.
In this model, we found that after controlling for a farm being within a disease cluster, increasing county density was associated with increased odds, whereas high biosecurity (aerosol filtration) was associated with decreased odds of reporting and infection of PRRSv and/or PEDv on farms (P < 0.05).
In a second model, we tested what factors were associated with being a diseased farm inside of a diseased cluster versus a diseased farm outside of a diseased cluster. Here we identified that contracted trucking was 30 times more common (P < 0.05) among diseased farms within diseased cluster.
In general, these findings might support the importance of extremely high biosecurity, such as is common among filtered farms. This is especially important if the farm is located in a high risk disease cluster area, with high swine density. Additionally, if contracted trucking is used, strict biosecurity practices that minimize cross over contamination at the farm would be warranted. Additionally, efforts should be made to work with contracted trucking companies to develop mutually beneficial biosecurity practices which might include route sequencing, and washing and disinfecting protocols.
The consistency in location of disease clusters in South West Minnesota and North West Iowa might also suggest the need to intensify disease control measures in the region. This might include Regional Control Projects (RCPs), which aim to share disease status in the effort of disease control and management. Several RCPs have gained early success and have shown, at least anecdotally, success at reducing disease pressure within a region. These efforts have the potential to control not only diseases such as PRRSv and PEDv, but may also help to prevent the spread of Foreign Animal Diseases.
In conclusion, these results support the ongoing need for high levels of biosecurity in areas where diseases have been documented to occur in higher than expected levels. This may be especially true in swine dense regions. Additionally, contracted trucking may pose a threat to the health of swine herds, and strict biosecurity practices may prove to be useful in minimizing disease spread.
The entire manuscript for this study has been submitted to PLOS ONE and is currently under review. Additionally, this study comprises a chapter in Dr. Steve Tousignant’s PhD dissertation thesis which will be made publicly available through the University of Minnesota Libraries.
