Abstract
There is limited knowledge about the factors that affect the prevalence of Coxiella burnetii in livestock. This study aims to explore the macroecological aspects of C. burnetii occurrence in different livestock species, across 15 different farms throughout the Republic of Korea, from October 2014 to June 2023. A total of 63 out of 540 collected blood samples were PCR positive for C. burnetii DNA. We used a binomial generalized linear mixed model to elucidate patterns of C. burnetii occurrence at the spatial and temporal scales. PCR positivity for C. burnetii was positively associated with PC5, an environmental principal component, which primarily represented the environmental variable “Proportion of forest area within a 50 m radius of farms”. The estimated variance between sampling years was smaller than the variance between the farm sites. This result may be useful in guiding livestock management strategies to reduce C. burnetii occurrence in livestock industry.
Keywords: Coxiella burnetii, Macroecology, Spatial, Temporal, Host, Livestock, Disease
Highlights
-
•
Coxiella burnetii was detected in 63 of 540 livestock blood samples.
-
•
Forest cover within 50 m of farms positively associated with C. burnetii occurrence.
-
•
Estimated between-farm variation exceeded between-year variation.
-
•
Local landscape features may influence C. burnetii occurrence in livestock.
1. Introduction
Disease macroecology studies large-scale patterns regulating the distribution, dynamics, occurrence and prevalence of diseases [1]. It focuses on how disease risk and transmissibility are influenced by spatial, temporal and biological factors such as climate, habitat and host species [2].
Coxiella burnetii (C. burnetii) is a globally widespread bacterium with a broad host range that includes wild and domestic animals as well as humans [3]. Previous studies have found clear spatiotemporal and host factor patterns connected to the occurrence of C. burnetii, linking it to certain meteorological conditions [4], vegetation structure [5], temporal variation [5] and host species [6]. However, few studies have explored the combined effect of spatial, temporal and host factors on the occurrence of C. burnetii in livestock.
In the Republic of Korea (ROK), C. burnetii was designated as a notifiable infectious disease in 2006, however it is classified as a neglected disease, and its macroecological patterns remain poorly understood. Clarifying these patterns could help identify where, when, and which hosts have a higher likelihood of C. burnetii occurrence [1], thereby supporting livestock management and disease-risk mitigation. To our knowledge, this study is the first to explore the effects of spatiotemporal and host factors on the occurrence of C. burnetii in livestock. We hypothesized that C. burnetii occurrence in livestock is affected by spatiotemporal and host factors, as transmission of the bacterium can differ due to 1) the proximity of forests, which affects the potential for wildlife-livestock interactions; 2) extreme weather events, such as the heatwaves of 2018 [7], which may affect livestock behavior and the bacterium itself and; 3) the occurrence of the bacterium being highest in cattle and goats.
2. Materials and methods
2.1. Ethics approval
All procedures were performed according to the ethical guidelines for the use of animal samples and were approved by the Institutional Animal Care and Use Committee (IACUC) of Jeonbuk National University (Approval No. JBU 2014-00026). Informed consent was obtained from the livestock owners prior to sample collection, and all samples were collected by an experienced, practicing veterinarian.
2.2. Sample size
The epidemiological dataset analyzed in this study contains 488 samples collected from 2014 to 2018, obtained from the nationwide surveillance data previously reported by Cho et al. (2021) [8], with an additional 52 samples included, which were collected separately in 2023. This study was designed as a retrospective macroecological analysis; therefore, no sample size calculation was performed for the present analysis. Blood samples were originally collected from livestock farms located in multiple geographic regions of the ROK through a nationwide tick-borne disease surveillance program [8]. Blood samples were collected from livestock selected by the farm owner or manager, regardless of their sex and age. For blood collection, both housed and pasture-raised cattle were individually restrained in a stanchion. All goats were physically restrained by at least two trained personnel during blood collection. All eligible blood samples with corresponding PCR results and complete epidemiological, geographic, and environmental information were included in the analysis. Records lacking complete farm-location or environmental-variable information were excluded, resulting in a final analytical dataset of 540 blood samples from 15 farms across multiple regions and from various livestock species, including beef cattle, dairy cattle, Korean native goats (KNG), Boer goats, horses and foals, from 2014 until 2023 (Fig. 1, Table S1, Table S2). All blood samples were transferred to Kyungpook National University for DNA extraction using the DNeasy blood kit (Qiagen Inc., Valencia, CA, USA) and stored at −20 °C until use.
Fig. 1.
Geographic distribution of livestock farms included in the final analytical dataset in the Republic of Korea. Yellow circles indicate the locations of the 15 farms included in the analysis, and numbers correspond to the farm IDs used throughout the article and Table S1. The final dataset comprised 540 blood samples collected from multiple livestock species between October 2014 and June 2023. In the land-cover map, urban areas, agricultural, forest, and water are shown in red, orange, green, and blue, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
2.3. Genetic analysis
Occurrence of C. burnetii, was determined by screening blood samples for C. burnetii DNA by a nested PCR using a specific primer targeting the IS1111 transposase insertion element [9]. The primary PCR was performed using the 1F (5′-TATGTATCCACCGTAGCCAGTC-3′) and 1R (5′-CCCAACAACAACCTCCTTATTC-3′) primers, followed by nested PCR using the 2F (5′-GAGCGAACCATTGGTATCG-3′) and 2R (5′-CTTTAACAGCGCTTGAACGT-3′) primers. PCR conditions were 93 °C for 3 min, followed by 30 cycles of 93 °C for 30 s, annealing at 54 °C for 30 s, and 72 °C for 1 min. Distilled water was used as a negative control in all runs. The amplified fragment was 204 bp, and secondary PCR products were separated by electrophoresis on a 1.5% agarose gel and visualized after staining with ethidium bromide. All PCR products were purified using AccuPower PCR Purification kits (Bioneer, Daejeon, ROK), and then used for direct sequencing (Macrogen, Daejeon, ROK). The nucleotide sequences obtained in this study were aligned with the ClustalW algorithm using the BioEdit Program.
2.4. Collection of meteorological and environmental variables
Meteorological and environmental variables were collected to determine their association with the occurrence of C. burnetii. Meteorological variables were obtained from the Meteorological Administration (https://data.kma.go.kr/cmmn/main.do), and land cover data related to land use was obtained from the Ministry of Environment of Korea (https://egis.me.go.kr/) and the Korea Forest Service (https://www.forest.go.kr) (Table S3). Meteorological variables were analyzed using inverse distance weighting [10], while distance to the nearest instance of each land use and the proportion (%) of each land use within 50, 500, 1000, and 1500 m radius were calculated in ArcGIS [[10], [11]].
2.5. Defining characteristics using principal component analysis
The dimensionalities of the environmental and meteorological characteristics of the farm locations were reduced using varimax-rotated principal component analysis (PCA) executed through the psych package in R [12]. The varimax rotation allows for variables defining farm location to be identified more easily [13]. To avoid the effects of multicollinearity on the principal components, only variables with a variance inflation factor ≤3 [14], identified using the “vifstep” function in the usdm package in R [14], were used in the generalized linear mixed model.
2.6. Identifying spatiotemporal and host effects using Generalized Linear Mixed Models
Binomial Generalized Linear Mixed Models (GLMM), with a logit link function, were utilized to analyze the effects of spatiotemporal factors on C. burnetii occurrence using the lme4 package in R [15]. The livestock species were not included in the GLMM and were examined separately, due to being confounded with farm sites. A candidate model set was created from the global GLMM using the “dredge” function from the MuMIn package in R [16]. Candidate models were ranked using corrected Akaike information criterion (AICc) [[17], [18]]. Models within ΔAICc <2 were considered to have substantial support. Among models with substantial support, the model with the fewest parameters was selected for inference and prediction. The random-effect variance was also calculated using the “VarCorr” function from the lme4 package in R [15].
The explanatory power of the final model was evaluated using the area under the receiver operating characteristic curve (AUC), with the optimal cut-off value calculated using the R package pROC [19], and the true skill statistic (TSS) [20]. Model convergence was assessed using the optimizer status and convergence messages from the fitted lme4 model, while singularity was assessed using the “isSingular” function from lme4 [15]. Model diagnostics were also assessed using simulated residuals, in which the model convergence, overdispersion, residual uniformity and the residual spatial autocorrelation were checked using the DHARMa package in R [21]. Prior to assessing spatial autocorrelation, the geographic coordinates (longitude and latitude) were transformed into a projected coordinate system (EPSG:5179; meters). The spatial autocorrelation of the DHARMa residuals, which were aggregated to unique spatial locations, was evaluated using the “testSpatialAutocorrelation” function [21].
Variation in C. burnetii PCR positivity among livestock species was described by calculating the observed positivity rates for each livestock species. The differences in detection frequencies among the livestock species were assessed using the Fisher's exact test. This analysis was considered explanatory, and not adjusted for farm site clustering, due to the uneven distribution of livestock species among the farms. Therefore, the results were not interpreted as independent livestock species effects.
3. Results
Out of the 540 blood samples collected from 15 farms, 63 samples (11.67%) were positive for C. burnetii by PCR. Farm-level positivity rates ranged from 0% to 57.14%, with the highest positivity observed at Farm 11 (20/35, 57.14%), followed by Farm 8 (7/28, 25%) and Farm 12 (2/8, 25%), whereas C. burnetii was not detected at Farm 14. Detailed information on the detection of C. burnetii according to farm, livestock species, region, and year is provided in Supplementary Table S1.
3.1. Effects of spatial, temporal and host variables on C. burnetii occurrence
A total of 255 possible model combinations (Table S4.1) were evaluated, PCA components (Appendix B) included as fixed effects and sampling year and farm location included as random effects. Among the candidate models with substantial support (ΔAICc <2, Table S4.2), the most parsimonious final model indicated that the occurrence of C. burnetii was positively associated with the environmental PCA component PC5, which primarily represented the environmental variable “Proportion of forest area within a 50 m radius” (Fig. 2, β = 0.16, p < 0.05). The final model estimated that random-intercept variance was greater among farms (0.23, SD = 0.48, Fig. 2A, C), compared to across sampling years (0.01, SD = 0.11, Fig. 2B, D).
Fig. 2.
Predicted probability of Coxiella burnetii occurrence across PC5, an environmental PCA component primarily representing the variable “Proportion of forest area within a 50 m radius of livestock farms”. Predictions for C. burnetii DNA were obtained from the final selected binomial GLMM, containing PC5 and the random intercepts “Farm site” and “Sampling year”. Panels A and B showcase farm-specific and year-specific fitted curves, respectively. Panels C and D show the corresponding predictions separately for each farm site and sampling year. Solid lines indicate the fitted probabilities, whilst the dashed lines (C, D) indicate the 95% confidence intervals. Colours identify the same farm and sampling year in the corresponding panels.
The final most parsimonious model had an AUC of 0.75 and a TSS of 0.42 on the model-fitting data [20]. The final model converged successfully, and was not singular, and showed no evidence of overdispersion (dispersion ratio = 1.01, p > 0.05) or problematic residual patterns in uniformity (Kolmogorov-Smirnov D = 0.05, p > 0.05), zero inflation (observed-to-simulated ratio = 1, p > 0.05) and significant outliers (1/540 observations, p > 0.05). The Moran's I test conducted on the aggregated DHARMa residuals, detected no significance spatial autocorrelation (Moran's I = 0.099, p = 0.056).
The observed PCR positivity for C. burnetii differed among the livestock species (Fisher's exact test, p < 0.05). KNGs exhibited the highest observed positivity rate, with 23 out of 55 sampled individuals testing positive for C. burnetii DNA (41.82%), with the other livestock species having lower observed positivity rates (Table S1). However, as livestock categories and farm sites were strongly confounded in this study, these findings should be interpreted as descriptive observations rather than an evaluation of the independent contribution of livestock species.
4. Discussion
Among the spatiotemporal and host variables examined, the selected final most parsimonious GLMM model identified a positive association between PCR positivity for C. burnetii and PC5, an environmental principal component associated with the variable “Proportion of forest area within a 50 m radius of farms”. The estimated residual variance was greater for farms than among sampling years after accounting for PC5. The positive association between PC5 and C. burnetii occurrence suggests that farms located in forest-associated landscapes may have a higher probability of C. burnetii detection in livestock.
4.1. Effect of forest on C. burnetii occurrence
Forests are important habitats for many wildlife hosts of C. burnetii [3]. In the ROK, forest edges adjacent to livestock farms are commonly used by birds and wild mammals, including water deer (Hydropotes inermis), roe deer (Capreolus pygargus), wild boar (Sus scrofa), and other medium-sized mammals [[3], [22], [23]]. A greater proportion of forest cover within a 50 m radius of farms may increase opportunities for interactions between wildlife and livestock, potentially facilitating direct or indirect transmission of C. burnetii.
Bacteria shed in the bodily fluids of infected wild animals can persist in the environment for extended periods [[24], [25]] which, coupled with C. burnetii's high resistance to physical stress, allows for the continuous persistence of the bacterium in the environment [26]. This can result in high levels of environmental contamination, increasing the likelihood of livestock exposure through indirect contact. Livestock may also be directly exposed to contaminated bodily fluids from wildlife while grazing, potentially leading to infection. Moreover, previous studies have also reported higher seroprevalence of C. burnetii among wildlife living in close proximity to livestock [27], suggesting the potential for transmission between wildlife and livestock.
Additionally, wildlife can support tick populations, which may act as vectors for C. burnetii and contribute to the maintenance of its natural transmission cycle. Transmission by ticks can occur transstadially, transovarially, and horizontally, through bites and feces, after they become infected with C. burnetii during blood meals [28]. Infected ticks can shed large numbers of C. burnetii in their feces, potentially contaminating the environment. Once dried, contaminated fecal material may become aerosolized, and C. burnetii in tick feces has been reported to remain infectious for at least 1 year [29]. Studies into the prevalence of C. burnetii in ticks throughout the ROK have identified the two most abundant tick species, Haemaphysalis longicornis and H. flava [30]. However, because wildlife and ticks were not sampled in the present study, the observed forest effect should be interpreted as an ecological association rather than direct evidence of wildlife-mediated transmission.
4.2. Effects of livestock species on C. burnetii occurrence
KNGs showed the highest observed PCR positivity for C. burnetii DNA among the livestock species examined. This may reflect the differences in management practices, stocking density, host susceptibility or environmental exposure [31]. However, due to the livestock species being strongly confounded with farm sites, it was not possible to distinguish the effects of livestock species on the occurrence of C. burnetii. Consequently, the observed differences should be interpreted cautiously and viewed as descriptive patterns rather than independent livestock species effects.
4.3. Spatiotemporal factors affecting C. burnetii occurrence
The estimated residual heterogeneity was greater among farms than among sampling years after accounting for PC5, which was primarily associated with the variable “Proportion of forest area within a 50 m radius of the farm”. This finding indicates that unexplained variation remained greater among farm sites than among sampling years. Although farm management practices were not explicitly evaluated in the present study, differences in management, housing conditions, livestock density, and other farm-level characteristics may have contributed to this residual variation. Most livestock sampled in this study originated from open farms that generally operate under extensive production systems. Compared with intensive farming systems, such systems may reduce close contact among animals and exposure to contaminated bodily fluids [32]. Previous studies have also reported higher levels of C. burnetii aerosol contamination in indoor facilities than in outdoor environments [33]. However, higher prevalences of C. burnetii have also been reported in livestock raised under extensive production systems [34]. Collectively, these findings suggest that the relationship between farm management and C. burnetii occurrence is likely context-dependent and may vary according to local environmental and husbandry conditions.
Although comparisons of prevalence among farms and livestock species may provide useful information, the relevance of estimating prevalence separately by year is less clear. This may be due to the inconsistent sampling across farms and years, as some farmers not providing consent for repeated sampling, resulting in some sites being sampled more than others. Consequently, the temporal variation in the detection of C. burnetii may not have been captured for farms which may have been sampled only once over the sampling period. In addition, the observed temporal variation may also reflect unmeasured fluctuations in livestock management practices [31], environmental conditions [35] or livestock population dynamics [31]. Furthermore, the ability to assess the potential influence of some farm-level characteristics including approximate herd size and livestock density (Table S2) on C. burnetii occurrence, may have been limited because these data were unavailable. As a result, the unmeasured farm-level and temporal differences may have introduced residual variation.
Declaration of generative AI use
During the preparation of this manuscript, authors used ChatGPT and Microsoft Copilot to help improve readability and clarity. The authors reviewed, edited and verified content, and take full responsibility for the content of the manuscript.
CRediT authorship contribution statement
Oliwia Uche-Eze: Visualization, Formal analysis, Writing – original draft. Kyoung-Seong Choi: Supervision, Investigation, Conceptualization, Writing – review & editing. Seunghyeok Kang: Data curation. Jinho Park: Investigation. Sungwon Hong: Supervision, Project administration, Methodology, Funding acquisition, Writing – review & editing.
Funding
This study was supported by the Sejong Science Fellowship, the National Research Foundation of Korea (NRF) [grant number NRF-2021R1C1C2004162], as well as the Korean Basic Science Institute (National Research Facilities and Equipment Center) grant funded by the Korea Government (MSIT) [grant number RS-2026-25471266].
Declaration of competing interest
No competing interests
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.onehlt.2026.101601.
Contributor Information
Kyoung-Seong Choi, Email: kschoi3@knu.ac.kr.
Sungwon Hong, Email: shong@knu.ac.kr.
Appendix A. Supplementary data
Supplementary data about the positivity rates of C. burnetii according to farm, livestock species, region and year (Table S1), characteristics of the livestock farms included in the study (Table S2), list of meteorological and environmental variables (Table S3), the GLMM model-selection outputs, predicting the probability of C. burnetii (Table S4.1), and top candidate GLMM models ΔAICc <2 (Table S4.2), supplementary excel table for Table S4.1 (full model-selection output of GLMM).
Results of principal component analysis.
Data availability
Data will be made available on request.
References
- 1.Stephens P.R., Altizer S., Smith K.F., Alonso Aguirre A., Brown J.H., Budischak S.A., Byers J.E., Dallas T.A., Jonathan Davies T., Drake J.M., Ezenwa V.O., Farrell M.J., Gittleman J.L., Han B.A., Huang S., Hutchinson R.A., Johnson P., Nunn C.L., Onstad D., Park A., Vazquez-Prokopec G.M., Schmidt J.P., Poulin R. The macroecology of infectious diseases: a new perspective on global-scale drivers of pathogen distributions and impacts. Ecol. Lett. 2016;19:1159–1171. doi: 10.1111/ele.12644. [DOI] [PubMed] [Google Scholar]
- 2.Paull S.H., Song S., McClure K.M., Sackett L.C., Kilpatrick A.M., Johnson P.T. From superspreaders to disease hotspots: linking transmission across hosts and space. Front. Ecol. Environ. 2011;10:75–82. doi: 10.1890/110111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Celina S.S., Cerný J. Coxiella burnetii in ticks, livestock, pets and wildlife: a mini-review. Front. Vet. Sci. 2022;9 doi: 10.3389/fvets.2022.1068129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Valiakos G., Giannakopoulos A., Spanos S.A., Korbou F., Chatzopoulos D.C., Mavrogianni V.S., Spyrou V., Fthenakis G.C., Billinis C. Use of geographical information system and ecological niche model to analyse potential exposure of small ruminants to Coxiella burnetii infection in central Greece. Small Rumin. Res. 2017;147:77–82. doi: 10.1016/j.smallrumres.2016.12.042. [DOI] [Google Scholar]
- 5.de Rooji M.M.T., Borlée Floor, Smit M., de Bruin A., Janse I., Heedrick D.J.J., Wouters I.M. Detection of Coxiella burnetii in ambient air after a large Q fever outbreak. PLoS One. 2016;11 doi: 10.1371/journal.pone.0151281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ruiz-Fons F., Astobiza I., Barandika J.F., Hurtado A., Atxaerandio R., Juste R.A., García-Pérez A.L. Seroepidemiological study of Q fever in domestic ruminants in semi-extensive grazing systems. BMC Vet. Res. 2010;6 doi: 10.1186/1746-6148-6-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Korean Meteorological Administration Annual Report Korea Meteorological Administration. 2018. https://www.kma.go.kr/download_01/Annual_Report_2018.pdf (Accessed 11 December 2024).
- 8.Cho H.-C., Hwang S., Kim E.-M., Park Y.-J., Shin S.-U., Jang D.-H., Chae J.-S., Choi K.-S. Prevalence and molecular characterization of Coxiella burnetii in cattle, goats, and horses in the Republic of Korea. Vector Borne Zoonotic Dis. 2021;21:502–508. doi: 10.1089/vbz.2020.2764. [DOI] [PubMed] [Google Scholar]
- 9.Parisi A., Fraccalvieri R., Cafiero M., Miccolupo A., Padalino I., Montagna C., Capuano F., Sottili R. Diagnosis of Coxiella burnetii-related abortion in Italian domestic ruminants using single-tube nested PCR. Vet. Microbiol. 2006;118:101–106. doi: 10.1016/j.vetmic.2006.06.023. [DOI] [PubMed] [Google Scholar]
- 10.Johnston K., Environmental Systems Research Institute . Environmental Systems Research Institute; Redlands, Calif.: 2001. Using ArcGis Geostatistical Analyst.https://content.esri.com/support/documentation/ao_/geostatistical_analyst_tutorial.pdf (Accessed 30 August 2024) [Google Scholar]
- 11.Hong S., Kim D.-K., Do Y., Kim J.Y., Kim Y-M., Cowan P., Joo G.-J. Stream health, topography, and land use influences on the distribution of the Eurasian otter Lutra lutra in the Nakdong River basin, South Korea. Ecol.Indic. 2018;88:241–249. doi: 10.1016/j.ecolind.2018.01.004. [DOI] [Google Scholar]
- 12.Revelle W. Northwestern University; Evanston, Illinois: 2020. psych: Procedures for Psychological, Psychometric, and Personality Research.https://CRAN.R-project.org/package=psych (Accessed 17 November 2024). [Google Scholar]
- 13.Abdi H., Williams L.J. Principal component analysis. WIREs Comput. Stat. 2010;2:433–459. doi: 10.1002/wics.101. [DOI] [Google Scholar]
- 14.Naimi B., Hamm N.A.S., Groen T.A., Skidmore A.K., Toxopeus A.G. Where is positional uncertainty a problem for species distribution modelling? Ecography. 2013;37:191–203. doi: 10.1111/j.1600-0587.2013.00205.x. [DOI] [Google Scholar]
- 15.Bates D., Maechler M., Bolker B., Walker S. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 2015;67 doi: 10.18637/jss.v067.i01. (Accessed 20 September 2024) [DOI] [Google Scholar]
- 16.Bartoń K. Multi-Model Inference. 2020. https://CRAN.R-project.org/package=MuMIn (Accessed 18 November 2024).
- 17.Burnham K.P., Anderson D.R. Springer; New York, NY: 2004. Model Selection and Multimodel Inference. [Google Scholar]
- 18.Mac Nally R. Regression and model-building in conservation biology, biogeography and ecology: the distinction between – and reconciliation of – ‘predictive’ and ‘explanatory’ models. Biodivers. Conserv. 2000;9:655–671. doi: 10.1023/a:1008985925162. [DOI] [Google Scholar]
- 19.Robin X., Turck N., Hainard A., Tiberti N., Lisacek F., Sanchez J.-C., Müller M. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics. 2011;12 doi: 10.1186/1471-2105-12-77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Allouche O., Tsoar A., Kadmon R. Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS) J. Appl. Ecol. 2006;43:1223–1232. doi: 10.1111/j.1365-2664.2006.01214.x. [DOI] [Google Scholar]
- 21.Hartig F. DHARMa: Residual Diagnostics for Hierarchical (Multi-Level/Mixed) Regression Models. 2026. https://CRAN.R-project.org/package=DHARMa R package 0.5.0.
- 22.Park G.-S., Kwon W. Ju, Seo B.-J., Kim C. Infectious disease challenges in Korean native black goats (KNBGs): current status, control needs and strategic approaches. J. Infect. Dis. Diagn. 2025;10:1–5. doi: 10.35248/2576-389X.25.10.308. [DOI] [Google Scholar]
- 23.Lim S.J., Han S.H., Kim K.Y., Hong S., Park Y.C. Relative Abundance of Mammals and Estimation of Minimum Trapping Effort Using Camera Traps in Jangsudae, Seoraksan National Park. Mammal Stud. 2023;48 (3):171–179. doi: 10.3106/ms2022-0035. [DOI] [Google Scholar]
- 24.Lee Y.-J., Chae S.-J., Kim B., Shin H.S., Kwak S.M., Jeong H., Lee S., Kang Y.-M., Kwak D., Seo M.-G. First molecular typing of tick-derived Coxiella burnetii from wildlife in South Korea: surveillance and genetic characterization. Transbound. Emerg. Dis. 2025;2025 doi: 10.1155/tbed/2533438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kim S.G., Kim E.H., Lafferty C.J., Dubovi E. Coxiella burnetii in bulk tank milk samples, United States. Emerg. Infect. Dis. 2005;11:619–621. doi: 10.3201/eid1104.041036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Berri M., Rousset E., Champion J.L., Russo P., Rodolakis A. Goats may experience reproductive failures and shed Coxiella burnetii at two successive parturitions after a Q fever infection. Res. Vet. Sci. 2007;83:47–52. doi: 10.1016/j.rvsc.2006.11.001. [DOI] [PubMed] [Google Scholar]
- 27.Kersh G.J., Fitzpatrick K.A., Self J.S., Priestley R.A., Kelly A.J., Lash R.R., Marsden-Haug N., Nett R.J., Bjork A., Massung R.F., Anderson A.D. Presence and persistence of Coxiella burnetii in the environments of goat farms associated with a Q fever outbreak. Appl. Environ. Microbiol. 2013;79:1697–1703. doi: 10.1128/AEM.03472-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Enright J.B., Franti C.E., Behymer D.E., Longhurst W.M., Dutson V.J., Wright M.E. Coxiella burnetii in a wildlife-livestock environment. Am. J. Epidemiol. 1971;94:79–90. doi: 10.1093/oxfordjournals.aje.a121298. [DOI] [PubMed] [Google Scholar]
- 29.Kazar J. Coxiella burnetii infection. Ann. N. Y. Acad. Sci. 2005;1063:105–114. doi: 10.1196/annals.1355.018. [DOI] [PubMed] [Google Scholar]
- 30.Aitken I.D., Bögel K., Cračea E., Edlinger E., Houwers D., Krauss H., Rády M., Řeháček J., Schiefer H.G., Kazán J., Schmeer N., Tarasevich I.V., Tringali G. Q fever in Europe: current aspects of aetiology, epidemiology, human infection, diagnosis and therapy. Infection. 1987;15:323–327. doi: 10.1007/bf01647731. [DOI] [PubMed] [Google Scholar]
- 31.Truong A.-T., Yun B.-R., Lim J., Min S., Yoo M.-S., Yoon S.-S., Yun Y.-M., Kim J.-T., Cho Y.S. Real-time PCR biochip for on-site detection of Coxiella burnetii in ticks. Parasit. Vectors. 2021;14 doi: 10.1186/s13071-021-04744-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Tan T.S.E., Hernandez-Jover M., Hayes L.M., Wiethoelter A.K., Firestone S.M., Stevenson M.A., Heller J. Identifying scenarios and risk factors for Q fever outbreaks using qualitative analysis of expert opinion. Zoonoses Public Health. 2022;69:344–358. doi: 10.1111/zph.12923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Álvarez-Alonso R., Basterretxea M., Barandika J.F., Hurtado A., Idiazabal J., Jado I., Beraza X., Montes M., Liendo P., García-Pérez A.L. A Q fever outbreak with a high rate of abortions at a dairy goat farm: Coxiella burnetii shedding, environmental contamination, and viability. Appl. Environ. Microbiol. 2018;84 doi: 10.1128/AEM.01650-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hwang S., Cho H.-C., Shin S.-U., Kim H.-Y., Park Y.-J., Jang D.-H., Kim E.-M., Kim J.W., Park J., Choi K.-S. Seroprevalence and molecular characterization of Coxiella burnetii in cattle in the Republic of Korea. Pathogens. 2020;9 doi: 10.3390/pathogens9110890. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Yeh S.-W., Won Y.-J., Hong J.-S., Lee K.-J., Kwon M., Seo K.-H., Ham Y.-G. The record-breaking heat wave in 2016 over South Korea and its physical mechanism. Mon. Weather Rev. 2018;146:1463–1474. doi: 10.1175/mwr-d-17-0205.1. [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary data about the positivity rates of C. burnetii according to farm, livestock species, region and year (Table S1), characteristics of the livestock farms included in the study (Table S2), list of meteorological and environmental variables (Table S3), the GLMM model-selection outputs, predicting the probability of C. burnetii (Table S4.1), and top candidate GLMM models ΔAICc <2 (Table S4.2), supplementary excel table for Table S4.1 (full model-selection output of GLMM).
Results of principal component analysis.
Data Availability Statement
Data will be made available on request.