Published online Jan 11, 2024.
https://doi.org/10.4142/jvs.23185
Exploring preventive factors against insufficient antibody positivity rate for foot-and-mouth disease in pig farms in South Korea: a preliminary ecological study
Abstract
Background
Foot-and-mouth disease (FMD) is a highly contagious viral disease in livestock that has tremendous economic impact nationally. After multiple FMD outbreaks, the South Korean government implemented a vaccination policy for efficient disease control. However, during active surveillance by quarantine authorities, pig farms have reported an insufficient antibody positivity rate to FMD.
Objective
In this study, the spatial and temporal trends of insufficiency among pig farms were analyzed, and the effect of the number of government veterinary officers was explored as a potential preventive factor.
Methods
Various data were acquired, including national-level surveillance data for antibody insufficiency from the Korea Animal Health Integrated System, the number of veterinary officers, and the number of local pig farms. Temporal and geographical descriptive analyses were conducted to overview spatial and temporal trends. Additionally, logistic regression models were employed to investigate the association between the number of officers per pig farm with antibody insufficiency. Spatial cluster analysis was conducted to detect spatial clusters.
Results
The results showed that the incidence of insufficiency tended to decrease in recent years (odds ratio [OR], 0.803; 95% confidence interval [95% CIs], 0.721–0.893), and regions with a higher density of governmental veterinary officers (OR, 0.942; 95% CIs, 0.918–0.965) were associated with a lower incidence.
Conclusions
This study implies that previously conducted national interventions would be effective, and the quality of government-provided veterinary care could play an important role in addressing the insufficient positivity rate of antibodies.
INTRODUCTION
Foot-and-mouth disease (FMD) is a highly infectious disease that affects artiodactyl mammals, such as cattle, pigs, sheep, goats, and deer. It can cause hypersalivation, fever, reduced appetite, and the formation of blisters on the lips, tongue, gum, nose, and hoof. In severe infections, FMD can lead to death in young animals [1]. The FMD virus belongs to the genus Aphthovirus of the Picornaviridae family and consists of seven serotypes: O, A, C, Asia-1, SAT-1, SAT-2, and SAT-3 [2].
The first recorded incidence of FMD in South Korea occurred in 1933 and ceased in 1934 [3]. No further incidences occurred until 2000, when 15 cases (cattle) of serotype O infection occurred. This was followed by several other recorded incidences, including 16 cases (1 cattle, 15 pigs) of serotype O in 2002; 6 cases (cattle) of serotype A in January 2010; 11 cases (7 cattle, 4 pigs) of serotype O in April–May 2010; 153 cases (97 cattle, 55 pigs, 1 goat) of serotype O between November 2010 and April 2011; 3 cases (pigs) of serotype O in July–August 2014; 185 cases (5 cattle, 180 pigs) of serotype O between December 2014 and April 2015; 21 cases (pigs) of serotype O in January–March 2016; 9 cases (cattle) of serotype O simultaneously with serotype A in February 2017; 2 cases (pigs) of serotype A in March–April 2018; and 3 cases (cattle) of serotype O in January 2019. Subsequently, no new incidences were reported for four years until May 2023, when 11 cases (10 cattle, 1 goat) of serotype O FMD were reported [4, 5].
The control measures for FMD incidence in 2000 included mass culling and ring vaccination [6]. Vaccination was not implemented for the subsequent outbreaks in 2002 or throughout the early 2010s, and a disease-free status was maintained. However, the nationwide spread of FMD in 2010 in Andong-si of Gyeongsangbuk-do prompted the implementation of a large-scale national vaccination program, leading to a rapid decrease FMD incidence. Subsequently, this policy included periodic FMD vaccinations for cows, pigs and sheep to maintain a disease-free state [7].
However, the vaccination policy did not successfully increase seropositivity rates in pig farms [8], possibly due to the implementation of a single-dose vaccination, although the recommended FMD vaccination regimen for pigs aged 8–12 weeks involved two doses administered four weeks apart [4, 9]. Vaccination by non-veterinarians may be another reason. In South Korea, farm owners, who are usually not veterinarians, can legally administer vaccines to pigs. Governments have exclusively provided veterinary care to small-scale pig farms with fewer than 1,000 pigs. Not only the lack of competency [10], but also the low compliance of owners with the vaccination program could result in an insufficient seropositivity rate. As owners consider the side effects of vaccination, such as abnormal meat, their motivation to follow the vaccination program may be low. A previous study found that large-scale farms, where vaccination was commonly administered by non-veterinarians, faced a higher risk of FMD [11].
Strengthening policies to motivate proper administration and expanding opportunities to receive veterinary care from authorities could be practical solutions to cope with the lower positivity rate of antibodies. However, relevant studies providing quantitative evidence of the effectiveness of these solutions have been lacking. This study aimed to preliminarily investigate the effects of relevant policies and levels of veterinary care implemented by the authorities. In terms of the effects of these policies, we focused on two revisions that motivated appropriate vaccination in 2018. First, a revision of the legislation titled “Notification on the FMD Vaccination, Clinical Test and Certification” (No. 2018-8; Date of Enactment: February 1, 2018) mandated the pigs to be vaccinated twice, in line with the specific regulations, instead of the customary single vaccination. Second, in the provisory clause of Article 4, Paragraph 1 of the 'Notification of FMD Vaccination, Clinical Test and Certification' (No. 2018-8; Date of Enactment: February 1, 2018), the intention to reinforce preventive measures on FMD in 2018 is expressed as follows: “The farms that received tests on a higher number of animals than stated in the criteria of the confirmatory test sample collection may be exempt from a confirmatory test.” With the revised notification, an antibody titer test for FMD vaccination was performed using blood collected from 16 or more pigs at the release of fed pigs from the farm to the slaughterhouse. If the result fell below the specified criteria, the farms could be directly charged a fine without a confirmatory test. As two revisions were implemented in 2018, we examined temporal trends in the incidence of a low FMD antibody positivity rate after 2018 to assess the effects of the revision. To examine the effects of veterinary care, we analyzed the association of the number of government veterinary officers with the farm-level seropositivity rate, exploring whether a higher number of vets can lead to improved veterinary care quality.
MATERIALS AND METHODS
Study design and data acquisition
This study was conducted to examine the effects of policy revisions and the number of government veterinary officers on the low seropositivity rate on pig farms. The definition of farms with the low seropositivity rate in this study was farms of which seropositivity level were below 60% and 30% for breeding and fattening pigs, respectively. The criteria were following the legislation titled as ‘Notification of FMD Vaccination, Clinical Test and Certification.’ In addition to the major analysis, a descriptive analysis was implemented to overview spatial and temporal trends. The quantitative analysis utilized data on pig farms with a low seropositivity rate, the number of government veterinary officers, and the number of local pig farms, which were used as the denominators to calculate the rate of pig farms with a low seropositivity rate and density of veterinary officers.
Data on the pig farms with a low seropositivity rate for FMD were obtained from the Korea Animal Health Integrated System (KAHIS) [5]. The KAHIS is a repository of a large amount of data, including reports on infectious diseases in domestic livestock, monitoring logs, and disease assessment results. The recorded information included the location, date of disease confirmation, and cause of the low seropositivity rate on pig farms that were confirmed to show a low seropositivity rate in the nationwide assessment of FMD serotype monitoring. This assessment involves collecting blood from pigs on the farm or from those released into a slaughterhouse.
Data on the number of government veterinary officers per year were obtained via an application to disclose information to the Ministry of the Interior and Safety. Publicly available licensing data from the local administration database were used for the regional distribution of pig farms. A shape file acquired from the Statistical Geographic Information Service (SGIS) was used for the background map of Korea used for visualization. The data sources and resolutions used in this study are summarized in Table 1.
Table 1
Data utilized in this study
Descriptive and qualitative analysis
The spatial distribution of farms with a low positivity rate for FMD antibodies is shown by a map using QGIS v3.24.3. To check whether clusters could be formed in the spatial distribution of the farms with a low seropositivity rate, SaTScan v10.1.2 was used, and the results were expressed on a map. In terms of the spatial cluster analysis, Bernoulli model was employed to detect spatial clusters using both the occurrence data and distribution of local pig farms. It was purely spatial analysis and no attribution regarding time was incorporated. Maximum cluster size was designated as 50% of total farms by default. The temporal trend of the rate was suggested for each city/province and visualized using Microsoft Excel. Suspected causes of low seropositivity identified by field investigations were also described for qualitative assessment. The data for the qualitative analysis were restricted to the Chungnam region due to data accessibility.
Quantitative analysis
Using logistic regression models, the association between the incidence of farms with a low seropositivity rate between 2018 and 2022 (5 years) as the outcome variable and two potential preventive factors was examined. The two potential factors were 1) year and 2) number of veterinary officers per local pig farm. The analysis unit was defined as in each year and in each district, and statistical analysis was performed at the aggregate level. Metropolitan cities (including Seoul, Busan, and Incheon) were excluded from the analysis owing to a lack of data on the number of pig farms. Districts without pig farms were excluded. Hence, the analysis included 147 districts over five years, resulting in a total of 735 analysis units (= 5 × 147).
Logistic regression models were utilized rather than Poisson models in this study because multicollinearity arose when a Poisson regression model was used, owing to the strong correlation between the number of pig farms as the offset and the number of veterinary officers. Using a logistic regression model, binary variables for the incidence proportion of farms with a low seropositivity rate were used as outcome variables, and two thresholds were used to create the binary variable: 1) median value of the incidence proportion, and 2) third quantile value of the incidence proportion. For example, if district “A” has 100 pig farms and 3 pig farms with low seropositivity rate, the incidence proportion of district A is 3%. Assuming that the average of the incidence proportion for all study units (= 735 units) is 2.5% and third quartile value of the incidence proportion is 4%, the district “A” is categorized as case district (district with higher incidence proportion) in the model with first threshold. But the district “A” is categorized as control district (district with lower incidence proportion) in the model with second threshold.
Both univariate and multivariate analyses were conducted. For the logistic regression analysis, the glm function of R v.4.2.1 was used.
RESULTS
The spatial distribution of pig farms and those with a low positivity rate to FMD antibodies was shown in Fig. 1. The number of cases within the study period in Chungnam Province was higher than that in other regions (n = 254), followed by Gyeonggi (n = 208), Jeonnam (n = 175), and Gyeongnam (n = 167). Eight spatial clusters were detected in the cluster analysis, as shown in Fig. 2. The centroids of these clusters were located in Chungnam (n = 3), Jeonnam (n = 2), Gyeongnam (n = 1), Gyeongbuk (n = 1), and Gyeonggi (n = 1). Descriptive temporal analysis showed that the number of cases and incidence tended to decrease. Although sporadic peaks in incidence were observed in 2019, the incidence decreased over time (Fig. 3).
Fig. 1
Spatial distribution of main variables. Characters A, B, and C indicate the number of pig farms with a low positivity rate to foot-and-mouth disease antibodies, local pig farms, and government veterinary officers per local pig farm, respectively.
Fig. 2
Detected spatial clusters of low positivity to foot-and-mouth disease antibodies in pig farms. The occurrence data of a low positivity rate to foot-and-mouth disease antibodies on pig farms between 2018–2022 were used. The Bernoulli model was employed to detect spatial clusters using both the occurrence data and the distribution of local pig farms. Blue rounds indicate cluster locations (n = 8).
Fig. 3
Temporal trends of a low positivity rate to foot-and-mouth disease antibodies in pig farms by province (2018–2022). Each line indicates a province. The number of cases and incidence (farm level) are shown in A and B, respectively.
Twenty-one cases with low seropositivity were included in the qualitative analysis. The identified suspected cause of the low seropositivity rate in the Chungnam region revealed that 14 cases were attributed to inadequate vaccination, leading to a failure to administer the prescribed dose of vaccines (Table 2).
Table 2
SeroPositivity rates to Foot-and-Mouth Disease and causes of the low seropositivity in pig farms in the Chungnam region
The results of the logistic regression models are shown in Fig. 4. All examined associations, including both univariate and multivariate analyses, were statistically significant. In the univariate analysis, the results showed an annual decrease of 20.7% (odds ratio [OR] and 95% confidence interval [95% CI] were 0.793 [0.713–0.880] and 0.661 [0.581–0.751] for the models using the threshold median and 3rd quantile of the incidence, respectively). The density of government-owned veterinary officers was also negatively associated with incidence (OR and 95% CI were 0.940 [0.916–0.963] and 0.961 [0.933–0.988] for the models using the threshold median and 3rd quantile of incidence, respectively). In the multivariate analysis, the results showed a trend of a 19.7% annual decrease (OR and 95% CI were 0.803 [0.721–0.893] and 0.668 [0.586–0.759] for the models using the threshold median and 3rd quantile of the incidence, respectively). The density of government-owned veterinary officers was also negatively associated with incidence (OR and 95% CI were 0.942 [0.918–0.965] and 0.967 [0.939–0.994] for the models using the threshold median and 3rd quantile of incidence, respectively).
Fig. 4
Univariable and multivariable associations between a low positivity rate of Foot-and-Mouth Disease antibodies and explanatory variables. All analyses were conducted using univariable (model 1-4) and multivariable (model 5-6) logistic regression models incorporating the two explanatory variables. The outcome variable was the incidence of low seropositivity in farms and was categorized by either the median (model 3, 4, and 6) or 3rd quantile value (model 1, 2, and 5) of the incidence.
DISCUSSION
The study results provide an overview of the spatial distribution of a low positivity rate to FMD antibodies in pig farms between 2018–2022. Spatial analysis showed that pig farms with low seropositivity rates were concentrated in certain regions and formed eight spatial clusters. Associations between incidence and three potential preventive factors, including year, number of veterinary officers per pig farm, and number of veterinary clinics for industrial animals per pig farm, were examined. Temporal trend analysis revealed that the incidence of antibody insufficiency decreased significantly each year. In addition, the associations with the other two factors were weakly significant.
The detected spatial clusters indicated that the regions within these clusters had similar geographical attributes [12]. These attributes need to be further investigated through epidemiological studies to identify potential risk factors [13]. Although there is no confirmed explanation for the spatial autocorrelation of antibody levels, social factors may be a candidate. Considering the unique features of pig farming, in which communication and exchange are active within community units, incorrect information on FMD vaccines can be shared within a network. Subsequently, this might lead to neighboring pig farms conducting vaccinations incorrectly, resulting in insufficient antibody levels. These spatially clustered patterns have also been reported for human vaccinations [14, 15].
Field epidemiological investigations identified inadequate vaccination as the major suspected cause for the low seropositivity rate. Considering that pig farms in South Korea suffer from a lack of experienced workers and a high turnover rate [16], the provision of high-quality education and motivation to workers on pig farms could effectively reduce the incidence of low seropositivity rates. The significant association of the year variables with incidence, which was supposed to be a proxy for the effect of policy revision, also implied the importance of education and motivation. Although these associations could not be confirmed from these preliminary results, the revision of policies, including mandatory two-dose vaccination and strengthening surveillance, could motivate farm workers to follow an adequate regimen. Furthermore, a higher density of veterinary officers was correlated with a lower number of pig farms with a low seropositivity rate. Veterinary officers play a key role in livestock disinfection. As an essential human resource in the livestock industry, veterinary officers are experts who work in close contact with farms regarding the disinfection of livestock. An increase in the number of such experts would lead to an increase in the number of pig farms with a better understanding of and compliance with government policies. An increased number of veterinary officers would facilitate more frequent contact with farms, enabling the delivery of high-quality veterinary services and strengthening livestock disease control efforts. Previous studies also suggested that clear information regarding vaccination and educational experiences can improve compliance among farm workers [17, 18]. Therefore, governments must exert greater efforts to increase the number and competence of veterinary officers.
Several limitations of this study should be noted for proper interpretation. First, the covariates were not included in the logistic regression analysis. Potential confounders may have affected these associations. Second, the sensitivity of surveillance for low seropositivity rates could differ by region because it is active surveillance whose frequency and sensitivity are affected by administrative budgets. Third, small-scale pig farms were not included in the local pig farm data. Finally, the results of the cluster analysis may be uncertain. There are various statistical approaches for detecting spatial clusters, and the results usually depend on the methodology used.
In this study, a logistic regression model was used to assess the effects of explanatory variables. The number of farms with a low seropositivity rate for FMD decreased over time, which can be attributed to the reinforced effects of legislation regarding FMD vaccination. The number of pig farms with a low FMD seropositivity rate decreased as the number of local veterinary officers increased. The study's findings are significant in the present context, where the shortage of veterinary officers has reached a critical level. They emphasized the urgent need to prioritize the development and support of professionals in relevant veterinary fields.
Conflict of Interest:The authors declare no conflicts of interest.
Author Contributions:
Conceptualization: Han D, Min KD.
Data curation: Han D.
Formal analysis: Min KD.
Investigation: Han D.
Methodology: Min KD.
Resources: Han D.
Software: Min KD.
Supervision: Min KD.
Validation: Ahn B.
Visualization: Han D.
Writing - original draft: Han D, Min KD.
Writing - review & editing: Min KD, Ahn B.
References
-
Bachrach HL. Foot-and-mouth disease. Annu Rev Microbiol 1968;22(1):201–244.
-
-
Park JH, Lee KN, Kim SM, Ko YJ, Lee HS, Cho IS. Geographical distribution and molecular epidemiology of the foot-and-mouth disease viruses of major groups. Korean J Vet Serv 2009;32(4):315–323.
-
-
Choe C, Jung YH, Do YJ, Cho A, Kang SJ, Kim UH, et al. Analysis of foot-and-mouth disease virus structural protein antibody positive rates according to ages in cattle and pigs. Korean J Vet Serv 2018;41(1):9–13.
-
-
Animal and Plant Quarantine Agency. Korea Animal Health Integrated System. Gimcheon: Animal and Plant Quarantine Agency; 2023.
-
-
Sur J, Shin J, Loubroth J, Yeh M, Ku B, Choi K, et al. In vivo characterization and transmission of Korean foot-and-mouth disease virus (FMDV). Korean J Vet Res 2000;40(4):719–727.
-
-
Ahn GH, Bae JG, Jung K, Wang YI, Jung JY, Kang SK, et al. Development of antibodies after foot and mouth disease vaccination in pigs. Korean J Vet Serv 2013;36(1):15–21.
-
-
Ko EY, Cho J, Cho JH, Jo K, Lee SH, Chung YJ, et al. Reduction in lesion incidence in pork carcass using transdermal needle-free injection of foot-and-mouth disease vaccine. Han-gug Chugsan Sigpum Hag-hoeji 2018;38(6):1155–1159.
-
-
Tobler WR. Cellular geography. In: Gale S, Olsson G, editors. Philosophy in Geography. Dordrecht: D. Reidel Publishing Company; 1979. pp. 379-386.
-
-
Nam I. Analysis of purpose, effectiveness and problem of HACCP system implementation according to scales of swine farm. Korean J Org Agric 2017;25(3):643–651.
-
MeSH Terms
Figures
1 / 4
Tables
1 / 2
