Evaluation of current foot-and-mouth disease sero-surveillance system in South Korea, 2020–2024

Despite nationwide vaccination, sporadic Foot-and-Mouth Disease (FMD) outbreaks continue to occur in South Korea, highlighting persistent vulnerabilities in the national serological surveillance system. We evaluated the performance of the national FMD surveillance system by integrating spatio-temporal outbreak clustering using SaTScan (2010–2019), nationwide structural protein (SP) and non-structural protein (NSP) serosurveillance data (2020–2024), hypergeometric estimation of herd- and component-level surveillance sensitivity, Bayesian latent class modeling of true seropositivity, and statistical assessment of outbreak risk and repeated low seropositivity. Ten spatio-temporal outbreak clusters were identified, including five statistically significant clusters, with major clusters concentrated in northern and central regions. Component-level surveillance sensitivity approached 100% in every year, but because it rises with the number of herds tested it saturates at national sampling volumes and cannot distinguish between species or years or demonstrate adequate within-herd detection; herd sensitivity showed substantial temporal and species-specific heterogeneity, with pigs exhibiting the highest sensitivity, cattle the lowest, and goats showing intermediate but variable performance. Bayesian latent class estimates of true SP seroprevalence were high and stable over the study period, with cattle (99.4–99.7%) consistently above pigs (92.8–97.8%) and goats (87.5–94.2%); the estimates for pigs and goats depended on the assumed assay sensitivity by up to 7.7 percentage points but remained above 80% under all prior assumptions examined. Counties with FMD outbreaks were more likely to exhibit repeated low seropositivity, while livestock density and farm numbers were significantly higher in outbreak areas. In a separate county-level analysis of outbreak occurrence, high total animal density (incidence rate ratio 2.28) and high livestock movement (incidence rate ratio 1.48) were associated with a higher incidence of outbreaks, whereas species-specific animal density was not. These findings indicate that extensive nationwide testing and high system-level sensitivity may mask important within-herd and species-specific detection gaps. Transitioning from predominantly uniform surveillance toward a statistically informed, risk-based framework incorporating species differences, herd-level sensitivity, spatial heterogeneity, livestock density, animal movement, and localized immunity gaps could improve early detection and resource allocation and strengthen alignment with WOAH standards for FMD-free status recognition.