Porcine reproductive and respiratory syndrome (PRRS) continues to affect the swine industry with 20 to 30% of herds breaking annually, costing $664 million every year. The industry needs the ability to better predict risk of a PRRS outbreak as well as to evaluate the level of biosecurity on their farms. The objective of this SHIC-funded study conducted at Iowa State University was to measure and benchmark the relationship between key biosecurity aspects and PRRS outbreaks in breeding herds, while validating a short biosecurity screening survey (44 questions). The analysis offers a flexible, shortened approach to screen breeding herd’s PRRS biosecurity vulnerability. Plus, this study highlights the value of using data to build upon the understanding of biosecurity risk in an operation.
The biosecurity screening survey captured data on herd demographics, PRRS outbreak history, frequency of high-risk events, surrounding swine density, transport biosecurity practices, carcass disposal, and people movement. Three machine-learning algorithms were used to identify key biosecurity factors and practices associated with PRRS outbreaks. The most accurate, based on ability to explain variability between farms in the frequency of reported PRRS outbreaks, was selected. Moreover, positive predictive value (PPV) from the models were utilized as a biosecurity score. The correlation between PPV and the reported frequency of PRRS outbreaks was investigated.
In the study, 13 production systems from 15 states were enrolled (n=188 sow herds). The best machine learning algorithm predicted PRRS occurrences with an accuracy of 78.5%. The 44 biosecurity measures were ranked according to their contribution to the model prediction. The first four most important variables were devotion to breeding genetic replacements, number of swine premises within a three-mile radius, number of breeding females on the premises, and distance to the nearest public road. The probability that a herd had reported an outbreak was higher in farms that had raised breeding animals as genetic replacements. The higher the number of swine facilities within a three-mile radius and the closer the farm was to a public road, the more likely the facility would be expected to break with PRRS.
Finally, the PPVs were correlated with the number of reported PRRS outbreaks (correlation of 72%). The majority of farms that had reported no outbreaks in the past five years had low (<50%) probability of having a PRRS outbreak (i.e., low score) while a larger number of PRRS cases were expected for farm who had reported at least one outbreak (i.e., high score), with few exceptions.
Swine producers continue to offer efforts to increase biosecurity with the intention to provide the best welfare for their pigs and produce a wholesome product for consumers. Recently, PRRS 1-4-4, lineage 1C has been creating issues in the upper Midwest, underscoring the need for a tool like the one being studied. The Swine Health Information Center (SHIC) co-hosted a webinar and posted a podcast on PRRS 1-4-4, lineage 1C.
As the world deals with the COVID-19 pandemic, SHIC continues to focus efforts on prevention, preparedness, and response to novel and emerging swine disease for the benefit of US swine health. As a conduit of information and research, SHIC encourages sharing of its publications and research. Forward, reprint, and quote SHIC material freely. SHIC is funded by America’s pork producers to fulfill its mission to protect and enhance the health of the US swine herd. For more information, visit https://www.swinehealth.org or contact Dr. Sundberg at [email protected].