True prevalence and spatial distribution of antibodies to Brucella spp. in goat populations in Hubei Province, People's Republic of China: Progress toward disease elimination

Prev Vet Med. 2024 Dec 28:235:106414. doi: 10.1016/j.prevetmed.2024.106414. Online ahead of print.

Abstract

Caprine brucellosis, mainly caused by Brucella melitensis, remains a significant zoonotic threat worldwide, affecting animal productivity, welfare, and public health. This study aimed to estimate the true prevalence (TP) and spatial distribution of antibodies to Brucella spp. among goat populations in Hubei Province, China. In 2021, approximately 1.4 million serum samples were collected from 23,126 goat flocks across 82 counties of 16 municipal regions of Hubei Province. A combination of the Rose Bengal Test and Serum Agglutination Test in series was used to detect antibodies against Brucella spp. A hierarchical Bayesian Latent Class Model was used to account for imperfect diagnostic sensitivity and specificity of the tests, conditional dependence between the two tests, and hierarchical data structure to estimate the TP and the probability of achieving a 95 % probability of having a TP below 0.1 % for each county and municipal region. Apparent prevalence was 0.051 % and 0.536 % at the animal and flock level, respectively. The median animal level TP in the 82 counties was 0.0088 % (Range: 0.0008 %, 9.3730 %), with 76.8 % of counties showing a median TP estimate below 0.1 %. Counties containing positive goats were mainly clustered in Huanggang and Huangshi, and counties bordering positive counties had a higher risk of seropositivity. Notably, 52.4 % of counties achieved a 95 % probability with a TP below 0.1 %. Sensitivity analyses confirmed the robustness of these findings across prior distributions. It was concluded that Hubei Province has achieved remarkable progress in caprine brucellosis elimination programs, and priority interventions should be given to positive counties and their bordering counties.

Keywords: Brucella spp; Disease freedom; Goat; Hierarchical Bayesian latent class model; True prevalence.