The failure of success: cyclic recurrences of a globally invasive pest.

In the six decades since 1960, the oriental fruit fly, Bactrocera dorsalis (Hendel), has been announced successfully eradicated in California by the U.S. Department of Agriculture a total of 564 times. This includes eradication declarations in one city a total of 25 different years, in 12 cities 8-19 different years, and in 101 cities 2-7 different years. We here show that the false negatives in declaring elimination success hinge on the easily achieved regulatory criteria, which have virtually guaranteed the failure of complete extirpation of this pest. Analyses of the time series of fly detection over California placed on a grid of 100-km2 cells revealed (1) partial success of the eradication program in controlling the invasion of the oriental fruit fly; (2) low prevalence of the initial detection in these cells is often followed by high prevalence of recurrences; (3) progressively shorter intervals between years of consecutive detections; and (4) high likelihood of early-infested cells also experiencing the most frequent outbreaks. Facing the risk of recurrent invasions, such short-term eradication programs have only succeeded annually according to the current regulatory criteria but have failed to achieve the larger goal of complete extirpation of the oriental fruit fly. Based on the components and running costs of the current programs, we further estimated the efficiency of eradication programs with different combinations of eradication radius, duration, and edge impermeability in reducing invasion recurrences and slowing the spread of the oriental fruit fly. We end with policy implications including the need for agricultural agencies worldwide to revisit eradication protocols in which monitoring and treatments are terminated when the regulatory criteria for declaring eradication are met. Our results also have direct implications to invasion biologists and agriculture policy makers regarding long-term risks of short-term expediency.

Comparison of enzootic risk measures for predicting West Nile disease, Los Angeles, California, USA, 2004-2010.

In Los Angeles, California, USA, 2 epidemics of West Nile virus (WNV) disease have occurred since WNV was recognized in 2003. To assess which measure of risk was most predictive of human cases, we compared 3 measures: the California Mosquito-Borne Virus Surveillance and Response Plan Assessment, the vector index, and the Dynamic Continuous-Area Space-Time system. A case-crossover study was performed by using symptom onset dates from 384 persons with WNV infection to determine their relative environmental exposure to high-risk conditions as measured by each method. Receiver-operating characteristic plots determined thresholds for each model, and the area under the curve was used to compare methods. We found that the best risk assessment model for human WNV cases included surveillance data from avian, mosquito, and climate sources.

Assessing alternative low pathogenic avian influenza virus surveillance strategies in a live bird market.

Surveillance, comprised of sampling and testing, of low pathogenic avian influenza virus (LPAIV) in a live bird market (LBM) may enable the detection of the virus, reducing its spread within the market to humans and birds and to other markets within the LBM system. In addition, detection of infected birds would also reduce the probability of reassortment and possible change from a LPAIV to a highly pathogenic avian influenza virus, which would have a devastating impact on the economy, trade, and society. In this paper we present results from a computer simulation model based on previously collected survey and experimental transmission data. Once we validated the model with experimental transmission data, we applied it to address some of the questions that need to be answered in order to create an efficient surveillance system in an LBM. We have identified effective sampling times, patterns, and sizes that would enhance the probability of an early detection of LPAIV if present and minimize the associated labor and cost. The model may be modified to evaluate different sized and structured LBMs. It also provides the basis to evaluate an entire LBM system for the United States or other countries.

Prevalence of low pathogenicity avian influenza virus during 2005 in two U.S. live bird market systems.

Oropharyngeal and cloacal swabs were collected from poultry sold in two live bird market (LBM) systems to estimate the prevalence of low pathogenicity avian influenza virus (LPAIV) shedding during the summer and fall of 2005. Random sampling was conducted in three LBMs in Minnesota where 50 birds were sampled twice weekly for 4 wk, and in three LBMs in a California marketing system. A stratified systematic sampling method was used to collect samples from Southern California LBMs, where LPAIV was detected during routine surveillance. No LPAIV was detected in the LBM system in Minnesota where realtime reverse transcription-PCR (RT-PCR) was conducted on oropharyngeal samples. RT-PCR was performed on swabs taken from 290 of 14,000, 65 of 252, and 60 of 211 birds at the three Southern California LBMs. The number of samples collected was based on the number of birds, age of the birds, and number of species present in the LBM. Virus isolation, subtyping, and sequencing of the hemagglutinin, neuraminidase, and other internal protein genes was performed on AIV-positive samples. The estimated prevalence of LPAIV in California was 0.345% in an LBM/supply farm with multiple ages of Japanese quail, 3% in an LBM with multiple ages and strains of chickens present, and 49.8% in an LBM with multiple species, multiple strains, and multiple ages. The positive virus samples were all LPAIV H6N2 and closely related to viruses isolated from Southern California in 2001 and 2004. Little or no comingling of poultry may contribute to little or no LPAIV detection in the LBMs.

Identifying errors in avian influenza virus gene sequences and implications for data usage of public databases.

Public gene sequence databases have become important research tools to understand viruses and other organisms. Evidence suggests that the identifying information for some of the sequences in these databases might not belong to the sequences they are associated with. We developed two tests to conduct a comprehensive analysis of all published sequences of the hemaglutinin and neuramidase genes of avian influenza viruses (AIVs) to identify sequences that may have been misclassified. One test identified sequence pairs with highly similar nucleotide sequences despite a difference of several years between their sampling dates. Another test, which was applied to samples sequenced and deposited more than once, detected sequences with more nucleotide differences to their own than to their closest relatives. All sequences identified as misclassified were further traced to relevant publications to assess the likelihood of contamination and determine if any conclusions were associated with the use of these sequences. Our results suggested that among 4040 published gene sequences examined, approximately 0.8% might be misclassified and that publications using these sequences may include inaccurate statements. Findings from this report suggest that using laboratory-adapted strains and handling multiple samples simultaneously increases the risk of contamination. The tests reported here may be useful for screening new submissions to public sequence databases.

Distinctive egg-laying patterns in terminal versus non-terminal periods in three fruit fly species.

The specific objective of this study was to use a logistic regression model for determining the degree to which egg laying patterns of individual females at the end of life (i.e., terminal segments) in each of three different fruit fly species could be distinguished from the egg-laying patterns over a similar period in midlife (i.e., non-terminal segments). Extracting data from large-scale databases for 11-day terminal and 11-day non-terminal segments in the vinegar fly (Drosophila melanogaster), the Mexican fruit fly (Anastrepha ludens) and the Mediterranean fruit fly (Ceratitis capitata) and organizing the model's results in a 2 × 2 contingency table, we found that: (1) daily egg-laying patterns in fruit flies can be used to distinguish terminal from non-terminal periods; (2) the overall performance metrics such as precision, accuracy, false positives and true negatives depended heavily on species; (3) differentiating between terminal and non-terminal segments is more difficult when flies die at younger ages; and (4) among the three species the best performing metrics including accuracy and precision were those produced using data on D. melanogaster. We conclude that, although the reliability of the prediction of whether a segment occurred at the end of life is relatively high for most species, it does not follow precisely predicting remaining life will also be highly reliable since classifying an end of life period is a fundamentally different challenge than is predicting an exact day of death.

Potential impact of introduction of foot-and-mouth disease from wild pigs into commercial livestock premises in California.

OBJECTIVE: To estimate potential spread of foot-and-mouth disease (FMD) if introduced from wild pigs in California and to evaluate efficacies of various control strategies. SAMPLE POPULATION: Data for California livestock and from hunter surveys on wild pigs in California. PROCEDURES: A spatial, stochastic simulation model was used to simulate FMD epidemics that might occur if a dairy or beef herd were infected from contact with a wild pig. Index herd location and type were examined, in addition to different statewide movement ban (SWMB) durations, to determine their effect on extent of the epidemic. RESULTS: Duration, number of infected premises, size of simulated outbreak, number of culled animals, and spatial distribution of infected herds resulting from the simulated outbreaks varied considerably among geographic regions, depending on index case type and location. Outbreaks beginning in the southern region of California were consistently longest, whereas those beginning in the northern region were shortest. The largest outbreaks resulted from index cases located in the southern and valley regions, whereas outbreaks were smallest when originating in the Sonoma or northern regions. For all regions, when the index herd was a dairy herd, size and duration of the outbreak were consistently reduced with implementation of an SWMB >or= 3 days. CONCLUSIONS AND CLINICAL RELEVANCE: Introduction of FMDV from wild pigs into a dairy or beef herd could result in a large and rapidly spreading outbreak, potentially affecting large numbers of herds. Size and duration of the outbreak might be reduced with an SWMB; however, the impact is highly dependent on the index herd type and location.

Invasions by Eurasian avian influenza virus H6 genes and replacement of the virus' North American clade.

The spread of highly pathogenic avian influenza virus (AIV) (H5N1) underlines the potential for global AIV movement through birds. The phylogenies of AIV genes from avian hosts usually separate into Eurasian and North American clades, reflecting limited bird migration between the hemispheres. However, mounting evidence that some H6 sequences from North America cluster with Eurasian subtype H6 sequences calls the strict hemispheric divide into question. We conducted a comprehensive phylogenetic analysis of the extent and timing of cross-hemisphere movements by the H6 gene. Results suggested that Eurasian H6 subtype has invaded North America several times, with the first invasions occurring 10 years before the first detection of invading isolates. The members of the North American clade decreased from 100% in the 1980s to 20% in the 2000s among H6 isolates from North America. Unraveling the reasons for this large-scale gene movement between hemispheres might identify drivers of global AIV circulation.

Transmission of low-pathogenicity avian influenza virus of subtype H6N2 from chickens to Pekin ducks and Japanese quail (Coturnix coturnix japonica).

In this experiment we evaluated the transmission characteristics of a chicken-adapted low-pathogenicity avian influenza virus (LPAIV) of subtype H6N2, from infected chickens to Japanese quail and Pekin ducks, which are commonly sold in live bird markets located in Southern California. The layout of the cages and bird-handling practices were similar to those found in Southern California live bird markets. Five out of 20 chickens were inoculated with LPAIV H6N2, and placed in direct contact with five chickens and in indirect contact with 10 chickens, 10 Japanese quail and 10 Pekin ducks. Transmission of LPAIV was followed in each bird daily for 15 days post inoculation by real-time reverse transcriptase-polymerase chain reaction testing of oropharyngeal and cloacal swab samples. This strain of H6N2 LPAIV, isolated from commercial poultry in California, was transmitted to chickens, quail, and ducks from chickens. An antibody response was detected in ducks by haemagglutination inhibition tests, but avian influenza virus was only detected by reverse transcriptase-polymerase chain reaction in one duck. Avian influenza virus was detected in quail (5 and 7 days post inoculation) before chickens (8 and 9 days post inoculation), all of which were in indirect contact with infected chickens; however, this difference was not statistically significant.

Simulation of an early warning system using sentinel birds to detect a change of a low pathogenic avian influenza virus (LPAIV) to high pathogenic avian influenza virus (HPAIV).

The placement of sentinel birds in a commercial poultry flock infected with low pathogenic avian influenza virus (LPAIV) may be an effective way of detecting subsequent change in the isolate to a high pathogenic avian influenza virus (HPAIV). Data collected from the 2002 Chilean HPAIV outbreak, along with information from a literature review of laboratory studies involving A/chicken/Chile/176822/02 (H7N3/LP) and A/chicken/Chile/184240-1/02 (H7N3/HP) viruses, were used to construct a computer simulation model. Mortality rates of the original LPAIV-infected population and the sentinel population were compared to detect the presence of HPAIV. A total of 12 increased mortality threshold scenarios were examined, using one-day absolute (2, 3, or 4 birds) or relative (0.5, 1.0, or 1.5%) mortality thresholds, and two-day absolute (1, 2, or 3 birds) or relative (0.25, 0.50, or 1.00%) mortality thresholds, to indicate the change from LPAIV to HPAIV in the sentinel and original populations, respectively. Results showed that following a one-day approach, threshold mortalities occurred on average at 7.35, 7.82, and 8.17 (0.5, 1.0, or 1.5%) and 6.21, 6.38, and 6.45 (2, 3, or 4 birds) days after the first infectious case for the original and sentinel populations, respectively. The two-day approach delayed the occurrence of threshold mortalities, on average, to 7.64, 8.05, and 8.62 (0.25, 0.50, or 1.00%) and 6.86, 6.78, and 7.23 (1, 2, or 3 birds) days after the first infectious case for the original and sentinel populations, respectively. Although, significant (p<0.10) differences were observed among different combinations of detection times for the original and sentinel populations, the use of sentinel birds has a maximum mean advantage, over monitoring mortality exclusively in the original population, of 1.96 and 1.84 days for one- and two-day threshold moralities, respectively. Additionally, the early warning system based on a sentinel vs. original population presented a decrease of the probabilities of a false alarm, from 0.04-0.45 to <0.01-0.10%. These findings may be used by decision makers to evaluate the risk of not depopulating a flock infected with a H5 or H7 LPAIV strain and the benefit of using sentinel birds as an early warning system of a change to HPAIV.

Epidemiologic investigation of the re-emergence of infectious salmon anemia virus in Chile.

Reports from surveillance activities were analyzed to determine the epidemiology of the re-emergence of infectious salmon anemia virus (ISAV) in Atlantic salmon farms in Chile. The epidemic and spatial and spatiotemporal patterns were described, taking into account commercial compartmentalization of the farms. During the 64 wk study period, 76 ISAV-infected salmon farms, representing 17 companies, were reported in 65% of the management geographic zones of the 10th region in southern Chile. Approximately 20% of the farms at risk became infected, with the incidence rate increasing slightly over time. Results from epidemic analyses and observed spatial and spatiotemporal patterns suggested an initial dispersal and subsequent clustering of cases around the index case (IC) in a propagated epidemic mode. Results suggested that delayed depopulation may have been a key factor in the spread and persistence of ISAV. Clustering of cases supported the assumption that passive transmission in seawater from ISAV-infected farms (proximity) is a critical factor in controlling disease. The re-emergence of ISAV in Chile has resulted in one of the largest ISAV epidemics reported in the world and this study generates new hypotheses and provides useful information for spatial disease control planning in salmon farming areas.

Results of a survey to estimate cattle movements and contact rates among beef herds in California, with reference to the potential spread and control of foot-and-mouth disease.

OBJECTIVE: To estimate contact rates and movement variables for shipments of beef cattle to and from producer premises in California. DESIGN: Cross-sectional survey. SAMPLE POPULATION: 583 beef producers in California. PROCEDURES: Questionnaires were developed and distributed to beef cattle producers in California. The study period was from April 20, 2005, through September 7, 2006. Data from completed questionnaires were entered manually into an electronic format. Descriptive statistical analyses were performed. RESULTS: Median number of interstate animal movements (movement of cattle into or out of California) was 0.17/mo; on the basis of this median, beef cattle were moved interstate > 2 times/y. Respondents kept beef cattle at up to 5 locations throughout the year. More than 40% of the movements from the respondents' premises were to a sale yard or auction facility. CONCLUSIONS AND CLINICAL RELEVANCE: Frequency of animal movements in this statewide study differed from values in another study of monthly shipments of animals to and from beef operations in 3 counties of California. The survey reported here revealed more frequent movements of animals to and from beef operations of all sizes. In addition, there were more high-risk indirect contacts on beef operations than has been reported previously. However, the number of low-risk indirect contacts was similar for small beef operations but less for large beef operations than has been reported elsewhere. Epidemic simulation models for California based on data in earlier studies likely underpredicted disease transmission involving beef herds.

An economic analysis of a contingency model utilising vaccination for the control of equine influenza in a non-endemic country.

BACKGROUND: Equine influenza (EI) is an infectious respiratory disease of horses that has never been reported in New Zealand (NZ). However, the 2007 EI outbreak in Australia, previously EI free, spurred the NZ government and stakeholders into evaluating alternative EI control strategies in order to economically justify any future decision to eradicate or manage EI. To build on the policy debate, this paper presents an epinomic (epidemiologic and economic) modelling approach to evaluate alternative control strategies. An epidemiologic model to determine how alternative EI control strategies influence the distribution of EI. Model results were then input into a cost-benefit analysis framework, to identify the return and feasibility of alternative EI eradication strategies in NZ. METHODS: The article explores nine alternative eradication scenarios and two baseline strategies. The alternative scenarios consisted of three vaccination strategies (suppressive, protective or targeted) starting at three time points to reflect the commercial breeding-cycle. These alternatives were compared to two breeding-cycle adjusted baselines: movement restriction in the breeding season (August to January) or non-breeding season (February to July). The economic loss parameters were incursion response, impact to the commercial racing industry (breeding, sales and racing), horse morbidity and mortality, and compensation to industry participants. RESULTS AND CONCLUSIONS: Results suggest that the economic viability of the EI eradication programme is dependent on when within the breeding-cycle the EI outbreak occurs. If an outbreak were to occur, the return on each dollar invested for protective or suppressive vaccination strategies would be between NZD$3.67 to NZD$4.89 and between NZD$3.08 to NZD$3.50 in the breeding and non-breeding seasons, respectively. Therefore, protective or suppressive vaccination strategies could be prioritised, regardless of season. As multiple industry stakeholders benefit from these strategies, the study will enable policy development and to better formulate a user-pays eradication programme.

Risk factors associated with human infection during the 2006 West Nile virus outbreak in Davis, a residential community in northern California.

We collected a total of 15,329 mosquitoes during weekly sampling in Davis, CA, from April through mid-October 2006 at 21 trap sites uniformly spaced 1.5 km apart over an area of approximately 26 km(2). Of these mosquitoes, 1,355 pools of Culex spp. were tested by multiplex reverse transcriptase-polymerase chain reaction, of which 16 pools (1.2%) were positive for West Nile virus (WNV). A degree-day model with a developmental threshold of 14.3 degrees C accurately predicted episodic WNV transmission after three extrinsic incubation periods after initial detection. Kriging interpolation delineated that Culex tarsalis were most abundant at traps near surrounding agriculture, whereas Cx. pipiens clustered within residential areas and greenbelt systems in the old portion of Davis. Spatial-temporal analyses were performed to test for clustering of locations of WNV-infected dead birds and traps with WNV-positive Cx. tarsalis and Cx. pipiens; human case incidence was mapped by census blocks. Significant multivariate spatial-temporal clustering was detected among WNV-infected dead birds and WNV-positive Cx. tarsalis, and a WNV-positive Cx. pipiens cluster overlapped areas with high incidences of confirmed human cases. Spatial analyses of WNV surveillance data may be an effective method to identify areas with an increased risk for human infection and to target control efforts to reduce the incidence of human disease.

A procedure to characterize geographic distributions of rare disorders in cohorts.

BACKGROUND: Individual point data can be analyzed against an entire cohort instead of only sampled controls to accurately picture the geographic distribution of populations at risk for low prevalence diseases. Analyzed as individual points, many smaller clusters with high relative risks (RR) and low empirical p values are indistinguishable from a random distribution. When points are aggregated into areal units, small clusters may result in a larger cluster with a low RR or be lost if divided into pieces included in units of larger populations that show no increased prevalence. Previous simulation studies showed lowered validity of spatial scan tests for true clusters with low RR. Using simulations, this study explored the effects of low cluster RR and areal unit size on local area clustering test (LACT) results, proposing a procedure to improve accuracy of cohort spatial analysis for rare events. RESULTS: Our simulations demonstrated the relationship of true RR to observed RR and p values with various, randomly located, cluster shapes, areal unit sizes and scanning window shapes in a diverse population distribution. Clusters with RR < 1.7 had elevated observed RRs and high p values. We propose a cluster identification procedure that applies parallel multiple LACTs, one on point data and three on two distinct sets of areal units created with varying population parameters that minimize the range of population sizes among units. By accepting only clusters identified by all LACTs, having a minimum population size, a minimum relative risk and a maximum p value, this procedure improves the specificity achieved by any one of these tests alone on a cohort study of low prevalence data while retaining sensitivity for small clusters. The procedure is demonstrated on two study regions, each with a five-year cohort of births and cases of a rare developmental disorder. CONCLUSION: For truly exploratory research on a rare disorder, false positive clusters can cause costly diverted research efforts. By limiting false positives, this procedure identifies 'crude' clusters that can then be analyzed for known demographic risk factors to focus exploration for geographically-based environmental exposure on areas of otherwise unexplained raised incidence.

Use of heterogeneous operation-specific contact parameters changes predictions for foot-and-mouth disease outbreaks in complex simulation models.

The role of contact parameters in a complex spatial simulation model of foot-and-mouth disease spread was determined by comparing predictions of number of infected premises, epidemic duration, and relative infection risk for different production sectors between a model that included the Full, heterogeneous (differing by production type) type-specific information about animal, vehicle and personnel movement between premises, and models that used partial and homogeneous (similar across production types) weighted-mean or proxy parameter sets for contacts between premises of all types. The model was run using a dataset of known premises locations in a three-county area in the Central Valley of California and categorized into 13 premises types and six production sectors. Results from models run with homogeneous contact parameters were always different from those obtained from the Full model, demonstrating that model predictions are affected by heterogeneity in contact parameters. Models simplified by using weighted-mean parameters predicted fewer infected premises. Models that were simplified by using medium dairy farm or large swine operation proxy parameters predicted longer epidemics with more infected premises, while those using small beef operation proxy parameters predicted shorter epidemics with fewer infected premises. Simplified-parameter models underestimated the impact on the economically important dairy sector, while overestimating the impact on beef and backyard operations. Results establish a need for heterogeneous, operation-specific contact parameters in complex stochastic simulation models that must be weighed against the cost of obtaining and coding premises type-specific contact information.

Simulation model for evaluation of testing strategies for detection of paratuberculosis in midwestern US dairy herds.

We developed a stochastic simulation model to compare the herd sensitivity (HSe) of five testing strategies for detection of Mycobacterium avium subsp. paratuberculosis (Map) in Midwestern US dairies. Testing strategies were ELISA serologic testing by two commercial assays (EA and EB), ELISA testing with follow-up of positive samples with individual fecal culture (EAIFC and EBIFC), individual fecal culture (IFC), pooled fecal culture (PFC), and culture of fecal slurry samples from the environment (ENV). We assumed that these dairies had no prior paratuberculosis-related testing and culling. We used cost-effectiveness (CE) analysis to compare the cost to HSe of testing strategies for different within-herd prevalences. HSe was strongly associated with within-herd prevalence, number of Map organisms shed in feces by infected cows, and number of samples tested. Among evaluated testing methods with 100% herd specificity (HSp), ENV was the most cost-effective method for herds with a low (5%), moderate (16%) or high (35%) Map prevalence. The PFC, IFC, EAIFC and EBIFC were increasingly more costly detection methods. Culture of six environmental samples per herd yielded >or=99% HSe in herds with >or=16% within-herd prevalence, but was not sufficient to achieve 95% HSe in low-prevalence herds (5%). Testing all cows using EAIFC or EBIFC, as is commonly done in paratuberculosis-screening programs, was less likely to achieve a HSe of 95% in low than in high prevalence herds. ELISA alone was a sensitive and low-cost testing method; however, without confirmatory fecal culture, testing 30 cows in non-infected herds yielded HSp of 21% and 91% for EA and EB, respectively.

The live bird market system and low-pathogenic avian influenza prevention in southern California.

Although live bird markets (LBMs) have been associated with outbreaks of avian influenza (AI), there are some LBM systems where AI outbreaks are extremely rare events. The California LBMs have not had any detected avian influenza viruses (AIVs) since December 2005. Responses to a detailed questionnaire on the practices and characteristics of the participants in the California low-pathogenic (LP) AI control program have been described to characterize possible reasons for the lack of AI outbreaks in LBMs. Compliance with an LPAI control program that contains active surveillance, prevention, and rapid response measures by those involved in the LBM system, rendering services to dispose of carcasses, no wholesalers, and few third-party bird deliveries was associated with the lack of LPAIV circulating in the Southern California LBM system.

Risk of introducing viral hemorrhagic septicemia virus (VHSV) to the Chilean South Pacific via sardine imports from Europe.

Chile imports from Spain 100s of metric tons of frozen sardine Sardina pilchardus fished in European oceans, which, with several other clupeids, are presumed susceptible to infection with viral hemorrhagic septicemia virus (VHSV). The frozen sardines are directly introduced into the sea as bait to catch southern hake Merluccius australis in the same areas where wild and pen-raised salmonids are present. A simulation model was therefore developed to evaluate the potential risk of infection of wild Chilean southern hake with VHSV from imported bait. The model indicated that VHSV-susceptible fish species present in Chilean waters, like southern hake, are not at immediate risk of infection. However, sensitivity analyses showed that infectious doses at lower concentrations of VHSV combined with higher VHSV-prevalence import scenarios could likely result in VHSV infections of a moderate number of indigenous southern hake (> or =54 fish yr(-1)).

Temporal-spatial epidemiology of foot-and-mouth disease outbreaks in Mongolia, 2000 - 2002.

Prior to 2000, foot-and-mouth disease (FMD) had not been observed in Mongolia since 1973; however, between April 2000 and July 2002, Mongolia reported 44 FMD outbreaks that affected cattle, sheep, goats, and camels. The objectives of this study were to describe the distributions of the 44 reported FMD outbreaks in Mongolia and to assess their spatial clustering and directions of movement. Official reports were collected to obtain the number and species of animals both affected and at risk, and the date and geographical coordinates of each outbreak. Significant global and local spatial clusters of reported FMD outbreaks were identified. Disease spread during the second epidemic moved 76 degrees northeast and the spread of the disease during the third epidemic moved 110 degrees northwest. FMD outbreaks were clustered intensely close to other FMD-positive counties. These findings can be used in the future to help plan prevention and control measures in high risk areas.