Which hogs, and how many, you sample for serology diagnostics depends on what you want to do.

For example, if the infection rate is at least 10 percent, and you test 29 animals, you can be 95 percent confident that you will identify at least one positive animal in a group of any size. A 50 percent prevalence would mean you only need to test five animals.

Those rates don’t apply, however, if you’re looking for agents or pathogens that you know are well established in the herd and are causing problems. In that case, the most simple and direct approach is to sample a few hogs that are sick or have just recovered.

What about sampling sick and healthy pigs and comparing titers? That rarely works because pigs that appear healthy are often subclinically infected.

The best results are usually produced from herd-health profiling. Two approaches are common. In cross-sectional sampling, several blood samples are taken at the same time from pigs of different age groups. If there is a clear difference in seroprevalance for a particular disease between age groups, you can be fairly certain at what age infection is occurring.

Longitudinal serological sampling is another approach, but it requires planning. Individual animals within groups are identified and serially bled as they move through the system. This provides a baseline should disease occur. Testing can identifiy any number of disease agents. When you add in clinical observations, you can make necessary correlations. 

Items in this column are drawn from Swine Practitioner, a sister publication of PORK’99.