Comparison of sampling strategies to detect porcine reproductive and respiratory syndrome virus in a simulated pig producing plant

The effectivity of different sampling schemes for the early detection of the introduction of porcine reproductive and respiratory syndrome virus into a pig herd was evaluated using Monte Carlo simulation. Within a theoretical breeding herd of 300 animals, disease transmission was simulated using a stochastic SEIR model incorporating actual animal movement data. The following parameters were evaluated for different sample sizes, sampling frequencies and diagnostic procedures (ELISA, PCR): the time from virus introduction until detection, the daily detection probability and the number of holdings to which infected animals are shipped before the disease is detected. The results show that the sample size has an influence on early detection. The biggest effects are, however, achieved by shortening the sampling intervals. The median detection time is approximately ten days shorter for PCR than for ELISA. If, however, the sampling intervals are chosen too wide there is a chance of overlooking the disease using PCR alone.