Table 4.
Probability of PRRSV detection as a function of within pen prevalence
| Within pen prevalence | Probability (%) of detection with one oral fluid sample (95% CI)a | Serum samples to match oral fluid probabilityb | ||
|---|---|---|---|---|
| RT-PCR | ELISA | RT-PCR | ELISA | |
| 5% | 31% (9, 67) | 17% (6, 38) | 8 | 5 |
| 10% | 79% (48, 94) | 59% (37, 77) | 11 | 5 |
| 20% | 98% (88, 100) | 94% (82, 98) | 13 | 10 |
| 30% | 100% (96, 100) | 99% (93, 100) | 12 | 10 |
| 40% | 100% (98, 100) | 100% (97, 100) | 10 | 9 |
| 50% | 100% (99, 100) | 100% (98, 100) | 9 | 8 |
aProbability of detection in oral fluids estimated by logistic regression (pen as random effect) from data reported in Olsen et al. (2013b). Within pen prevalence established by placing PRRSV-positive pigs (14 days after MLV vaccination) in pens of PRRSV-negative pigs to achieve 25 pigs per pen [89]
bNumber of serum samples required to match the probability of detection for one oral fluid sample was estimated using a hypergeometric distribution