Influenza A virus (IAV) infection in pigs is a concern to producers, veterinarians and the general public. This study presents models to estimate the sensitivities (Se) and specificities (Sp) of respiratory clinical signs (RCS), and real-time reverse transcription polymerase chain reaction (RRT-PCR) resulted from oral fluid (OF) and nasal swab (NS) samples in the absence of a gold standard. In addition, the models estimated an average prevalence of IAV infection in the Midwestern United States (US) growing pig populations. Bayesian model provided estimates under scenarios where IAV vaccination reduced only clinical manifestations, but not infection (basic model), or where vaccination reduced both. By the basic model, the Se and Sp of RCS from posterior distributions were 0.38 (95%Cridible interval (CrI): 0.28, 0.48) and 0.66 (95%CrI: 0.61, 0.71). The Se and Sp of of RRT-PCR were 0.84 (95%CrI: 0.87, 0.90) and 0.93 (95%CrI: 0.82, 0.97), and those of NS RRT-PCR were 0.79 (95%CrI: 0.71, 0.89) and 0.97 (95%CrI: 0.90, 0.99) respectively. The true prevalence estimate of IAV infection in the Midwestern US growing pig populations was 0.24 (95%CrI: 0.16, 0.30). In the second scenario, the Se and Sp of RCS were reduced by vaccination whereas those of NS and OF-RRT-PCR were not reduced by vaccination. Depending on the prior knowledge of vaccination, the model (in the second scenario) estimated that vaccination reduced the true prevalence of IAV in growing pigs, and thereby this has broader implications for the control and perhaps eradication of IAV in growing pigs.
Key words: Bayesian estimation, test accuracy, prevalence, influenza A virus, swine.
Copyright © 2019 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0