Herd sensitivity in relation to test sensitivity in swine vesicular disease serological tests

After the swine vesicular disease (SVD) outbreaks in 1992 in the Netherlands a national monitoring programme was initiated, testing 12 samples from every pig farm three times per year. In this monitoring a slightly higher cut-off was used than the cut-off agreed on within the European community. The author has analysed the effect of this higher cut-off on the percentage of false positive and false negative results, using information on SVD antibody titres in sera obtained from the monitoring programme and the outbreaks in 1992 and 1994. The number of false positive results was reduced by 63% when using the higher cut-off. On average the test sensitivity was reduced from 100% to 88%, resulting in a change of the average herd sensitivity from 91.7% to 91.5%, when testing 12 samples per farm. When three samples per farm were tested, the average herd sensitivity changed from 64.9% to 62.9%. The results further indicate that, in contrast to what is generally presumed, there is a relationship between test sensitivity and the prevalence of infection. The results clearly show that sample size is far more important in obtaining a high herd sensitivity than achieving a high test sensitivity.