Thirty eight breeding herds across four cohorts were followed from September 2008 to January 2012. The odds of a new PRRSV infection before filtration were 7.97 times higher than the odds after filtration was initiated, with a p value below 0.01. Median time to a new infection was 30 months in filtered herds against 11 months in non filtered herds, also at p below 0.01.
This is the primary source for the eightfold figure that circulates through industry material, and the study design is what gives the number its weight. Herds were compared against their own history rather than against a separate control group, which removes much of the difference between farms that filter and farms that do not. What it does not remove is uncertainty about what counts as a new infection, and Alonso and colleagues wrote their 2013 epidemiological study partly to address that, testing an approximately 80 percent risk reduction across a range of phylogenetically based case definitions. The two studies belong together in any citation and are not interchangeable.
The median time figures translate the odds ratio into something a producer can plan against. A herd going 30 months between introductions rather than 11 has a different production cycle, a different vaccination and stabilisation calendar, and a different capital position. That is the practical form the benefit takes: not the elimination of PRRSV introductions, but a substantial lengthening of the interval between them, with all the recovery that longer interval allows.