Prediction of province-level outbreaks of foot-and-mouth disease in Iran using a zero-inflated negative binomial model.

To identify events that could predict province-level frequency of foot-and-mouth disease (FMD) outbreaks in Iran, 5707 outbreaks reported from April 1995 to March 2002 were studied. A zero-inflated negative binomial model was used to estimate the probability of a 'no-outbreak' status and the number of outbreaks in a province, using the number of previous occurrences of FMD for the same or adjacent provinces and season as covariates. For each province, the probability of observing no outbreak was negatively associated with the number of outbreaks in the same province in the previous month (odds ratio [OR]=0.06, 95% confidence interval [CI]: 0.01, 0.30) and in 'the second previous month' (OR=0.10, 95% CI: 0.02, 0.51), the total number of outbreaks in the second previous month in adjacent provinces (OR=0.57, 95% CI: 0.36, 0.91) and the season (winter [OR=0.18, 95% CI: 0.06, 0.55] and spring [OR=0.27, 95% CI: 0.09, 0.81], compared with summer). The expected number of outbreaks in a province was positively associated with number of outbreaks in the same province in previous month (coefficient [coef]=0.74, 95% CI: 0.66, 0.82) and in the second previous month (coef=0.23, 95% CI: 0.16, 0.31), total number of outbreaks in adjacent provinces in the previous month (coef=0.32, 95% CI: 0.22, 0.41) and season (fall [coef=0.20, 95% CI: 0.07, 0.33] and spring [coef=0.18, 95% CI: 0.05, 0.31], compared to summer); however, number of outbreaks was negatively associated with the total number of outbreaks in adjacent provinces in the second previous month (coef=-0.19, 95% CI: -0.28, -0.09). The findings indicate that the probability of an outbreak (and the expected number of outbreaks if any) may be predicted based on previous province information, which could help decision-makers allocate resources more efficiently for province-level disease control measures. Further, the study illustrates use of zero inflated negative binomial model to study diseases occurrence where disease is infrequently observed.

Factors associated with spatial clustering of foot-and-mouth disease in Nepal.

The purpose of this study was to quantify associations between hypothesized epidemiological factors and the spatial distribution of foot-and-mouth disease (FMD) in Nepal. Spatial clustering of reports of at least one FMD case by Village Development Committee (VDC) in 2004 was examined by use of the spatial scan statistic. A Bayesian Poisson multivariate regression model was used to quantify the association between the number of reports and 25 factors hypothesized to be associated with FMD risk. The spatial scan statistic identified (P < 0.01) two clusters of FMD reports. Large numbers of people, buffalo, and animal technicians in a district were associated with an elevated risk of a VDC reporting >or=1 FMD case. The knowledge of high-risk areas and factors associated with the risk of FMD in Nepal could be applied in future disease control programs.

A web-based system for near real-time surveillance and space-time cluster analysis of foot-and-mouth disease and other animal diseases.

Considerable attention has been given lately to the need for global systems for animal disease surveillance that support real-time assessment of changing temporal-spatial risks. Until recently, however, prospects for development of such systems have been limited by the lack of informatics tools and an overarching collaboration framework to enable real-time data capturing, sharing, analysis, and related decision-making. In this paper, we present some of the tools of the FMD BioPortal System (www.fmd.ucdavis.edu/bioportal), which is a web-based system that facilitates near real-time information sharing, visualization, and advanced space-time cluster analysis for foot-and-mouth disease (FMD). Using this system, FMD information that is collected and maintained at various data acquisition and management sites around the world can be submitted to a data repository using various mutually agreed upon Extensible Markup Language (XML) formats, including Health Level Seven (HL7). FMD BioPortal makes available a set of advanced space-time cluster analysis techniques, including scan statistic-based methods and machine learning-based clustering methods. These techniques are aimed at identifying local clusters of disease cases in relation to the background risk. Data and analysis results can be displayed using a novel visualization environment, which supports multiple views including GIS, timeline, and periodical patterns. All FMD BioPortal functionalities are accessible through the Web and data confidentiality can be secured through user access control and computer network security techniques such as Secure Sockets Layer (SSL). FMD BioPortal is currently operational with limited data routinely collected by the Office International des Epizooties, the GenBank, the FMD World Reference Laboratory in Pirbright, and by the FMD Laboratory at the University of California in Davis. Here we describe technical attributes and capabilities of FMD BioPortal and illustrate its functionality by analyzing and displaying information from a simulated FMD epidemic in California.

Fluorescence-based multiplex real-time RT-PCR arrays for the detection and serotype determination of foot-and-mouth disease virus.

This study reports the development of distinct fluorescence-based multiplex rRT-PCR arrays, which are comprised of unique primer pairs that are statistically modeled from publicly available sequences, for overall foot-and-mouth disease virus and serotype-specific detection. Performance of the multiplex pan FMDV (MpFMDV) array was compared to the 5' UTR and two 3D(pol) assays, following tests on the Miniopticon (BioRad) and 7900HT SDS (ABI) platforms. The MpFMDV array and a recently developed alternative 3D(pol) assay correctly identified all 31 isolates that represented all seven serotypes, including 10 isolates reported previously to be undetectable by both OIE recommended assays (5' UTR and 3D(pol) assay). Statistically significant differences in mean C(T) values were observed with both platforms. With the Miniopticon, the MpFMDV assay detected FMDV at 9.86 and 2.67 cycles less than the 5' UTR and one of the 3D(pol) assay, respectively; whereas with the 7900HT SDS, the MpFMDV assay detected FMDV at 7.11, 4.71, and 2.33 cycles less than the 5' UTR and both 3D(pol) assays, respectively. The MpFMDV assay was more sensitive than the 5' UTR assay, which had a higher mean endpoint dilution of 1.3 log(10). With the exception of the serotype A and C multiplex arrays, the multiplex serotype-specific arrays correctly identified all isolates from five of seven serotypes (O, Asia 1, SAT 1, 2, and 3). Results presented in this study provide proof-of-principle and bench validation for the primer design model, and for the ability of multiplex arrays to accurately detect FMDV and to provide some degree of serotype discrimination of FMDV.

Variation in the VP1 gene of foot-and-mouth disease virus serotype A associated with epidemiological characteristics of outbreaks in the 2001 epizootic in Argentina.

A mixed binomial Bayesian regression model was used to quantify the relation between nucleotide differences in the VP1 gene of Foot-and-mouth disease virus (FMDV) serotype A, and epidemiologic characteristics of the outbreaks from which the viruses were obtained between January and December 2001 in Argentina. An increase in the probability of different nucleotides between isolates was associated with a longer time between isolation dates, a greater distance between isolation locations, an increase in the difference between attack rates, and an increase in the difference in outbreak durations. The farther apart the outbreak herds were in the southerly and easterly directions, the greater the difference in nucleotide changes. The model accurately predicted genetic distances of isolates obtained in 2001 compared with a 2002 isolate (P < 0.01), which suggested that the predictive modeling approach applied in the present study may be useful in understanding the epidemiology of evolution of FMDV and in forensic analysis of disease epidemics.

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.

Results of a survey of owners of miniature swine to characterize husbandry practices affecting risks of foreign animal disease.

OBJECTIVE: To characterize husbandry practices that could affect the risks of foreign animal disease in miniature swine. DESIGN: Survey study. STUDY POPULATION: 106 owners of miniature swine. PROCEDURES: An online survey of owners of miniature swine was conducted to obtain information about miniature pig and owner demographics; pig husbandry; movements of pigs; and pig contacts with humans, other miniature swine, and livestock. RESULTS: 12 states, 106 premises, and 317 miniature swine were represented in the survey. More than a third (35%) of miniature swine owners also owned other livestock species. Regular contact with livestock species at other premises was reported by 13% of owners. More than a third of owners visited shows or fairs (39%) and club or association events (37%) where miniature swine were present. More than 40% of owners fed food waste to miniature swine. Approximately half (48%) of the veterinarians providing health care for miniature swine were in mixed-animal practice. CONCLUSIONS AND CLINICAL RELEVANCE: Results of this study indicated that miniature swine kept as pets can be exposed, directly and indirectly, to feed and other livestock, potentially introducing, establishing, or spreading a foreign animal disease such as foot-and-mouth disease. In addition, the veterinary services and carcass disposal methods used by miniature swine owners may reduce the likelihood of sick or dead pigs undergoing ante- or postmortem examination by a veterinarian.

Modeled detection time for surveillance for foot-and-mouth disease virus in bulk tank milk.

OBJECTIVE: To estimate when foot-and-mouth disease virus (FMDV) would first be detected in bulk tank milk of dairies after exposure to FMDV. SAMPLE POPULATION: Hypothetical dairy herds milking 100, 500, or 1,000 cows. PROCEDURES: For each day after herd exposure to FMDV, infection, milk yield, and virolactia were simulated for individual cows with low and high rates of intraherd transmission to estimate when a PCR assay would detect virus in bulk tank milk. Detection limits were based on assumptions for the number of virus genomes per milliliter of milk and for analytical sensitivity of a PCR assay. RESULTS: A mean of 10% of the cows was predicted to have FMD lesions from 7 to 8 days and from 13.5 to 15 days after herd exposure for herds with high and low intraherd transmission rates, respectively. Herd bulk milk volume decreased by 10% by 8.5 to 9.5 days and by 15 to 16.5 days after herd exposure for herds with high and low transmission rates, respectively. Mean times by which FMDV would be first detected in bulk milk were 2.5 days and 6.5 to 8 days after herd exposure, which were extended for 10 to 11 days and 17 to 18 days for herds with high and low transmission rates, respectively. CONCLUSIONS AND CLINICAL RELEVANCE: PCR screening of bulk milk for FMDV would likely detect FMDV in dairy herds several days sooner than might be expected for owner reporting of clinical signs and thus should be worthy of consideration for regional, national, or global FMD surveillance.

Spatial distribution of foot-and-mouth disease in Pakistan estimated using imperfect data.

We estimated the spatial distribution of foot-and-mouth disease (FMD) in Pakistan; we used a probability co-kriging model and the number of FMD outbreaks reported between 1996 and 2000 by Pakistan to the Office International des Epizooties. We used a k-Bessel model and small-ruminant and human densities as surrogate covariates for the population at risk and for livestock markets and movements, respectively. Compared to no or only one covariate, the co-kriging model with both densities provided the best fit to independently obtained data on the spatial distribution of virus isolations (P=0.57). The estimated probability of an FMD outbreak per 25km(2) cell ranged from 0.017 to 0.812, with the maximum relative probability of 47.8 (0.812/0.017). Areas with the highest relative probability of having an FMD outbreak were located in the Punjab region; this is a major animal-production area located along a traditional international animal-trade route.

Diagnosis using predictive probabilities without cut-offs.

Standard diagnostic test procedures involve dichotomization of serologic test results. The critical value or cut-off is determined to optimize a trade off between sensitivity and specificity of the resulting test. When sampled units from a population are tested, they are allocated as either infected or not according to the test outcome. Units with values high above the cut-off are treated the same as units with values just barely above the cut-off, and similarly for values below the cut-off. There is an inherent information loss in dichotomization. We thus develop a diagnostic screening method based on data that are not dichotomized within the Bayesian paradigm. Our method determines the predictive probability of infection for each individual in a sample based on having observed a specific serologic test result and provides inferences about the prevalence of infection in the population sampled. Our fully Bayesian method is briefly compared with a previously developed frequentist method. We illustrate the methodology with serologic data that have been previously analysed in the veterinary literature, and also discuss applications to screening for disease in humans. The method applies more generally to a variation of the classic parametric 2-population discriminant analysis problem. Here, in addition to training data, additional units are sampled and the goal is to determine their population status, and the prevalence(s) of the subpopulation(s) from which they were sampled.

Use of the scan statistic on disaggregated province-based data: foot-and-mouth disease in Iran.

The spatial scan statistic was applied to density-smoothed data that approximated the spatial distribution within the area and reduced the potential bias produced when location data have been aggregated for large areas. The method is illustrated, using data on the location of foot-and-mouth disease (FMD) outbreaks in Iran. Data examined were 4477 FMD outbreaks reported on a per province basis between June 1996 and September 2003. A kernel density of the outbreak locations was estimated, using a fixed radius and the centroid of each province as the designated location of all cases reported for the province. The radius that produced a density map with the highest correlation with expert opinion was 4 degrees (latitude/longitude). Livestock density was used as a proxy for the underlying population at risk of acquiring FMD. Livestock and outbreak density maps were overlain to obtain the number of outbreaks and livestock in each of 15,599 cells covering the mapped surface of the country. A spatial scan statistic was applied to the density-smoothed data assuming that the outbreaks had a Poisson distribution. Results were compared with those obtained using a spatial scan statistic on provincially aggregated data. Application of the spatial scan statistic on the density-smoothed data allowed identification of clusters (P<0.01) related more to the actual geographic distribution of cases (expert opinion) and of animals at risk, than to the distribution of the provinces. Significant clusters of FMD were identified that coincided with roads, neighboring countries, and high-density population areas, suggesting that the region may represent a route for cross-continent transmission of FMD.

Evaluation of Mycobacterium avium subsp paratuberculosis infection of dairy cows attributable to infection status of the dam.

OBJECTIVE: To estimate the extent to which infection with Mycobacterium avium subsp paratuberculosis (MAP) of cows in a large dairy was attributable to the infection status of their dams. DESIGN: Retrospective longitudinal study. ANIMALS: 625 dam-daughter pairs of Holstein cows. PROCEDURE: Serologic test results were compared between cows and their dams. Logistic regression was used to assess whether a cow's serologic status was associated with its dam's serologic status. Infection with MAP attributable to being born to a seropositive dam was estimated for individual cows and for the herd. RESULTS: Cows with seropositive dams were 6.6 times as likely to be seropositive, compared with cows of seronegative dams. For seropositive cows born to seropositive dams, 84.6% of seropositivity was attributable to being born to a seropositive dam and 15.4% to other exposures, including exposure as calves to flush water that contained feces of adult cattle. For the herd as a whole, the seropositive status in 34% of seropositive cows was attributable to being born to a seropositive dam. CONCLUSIONS AND CLINICAL RELEVANCE: For dairy herds that breed seropositive cows, subsequent transmission of MAP to their daughters, either congenitally or via exposure to feces and colostrum of the dam shortly after birth, can contribute substantially to maintaining prevalence of MAP in a herd. Removal of seropositive, clinically unaffected cows and their daughters would be necessary to reduce infection with MAP attributable to congenital or peri-parturient transmission from dam to daughter.

Influence of age and production type on liver copper concentrations in calves.

A retrospective study of necropsy reports was undertaken to assess the relation between liver copper concentration and age, sex, and production class for calves that did not have a history compatible with copper toxicosis or with copper deficiency. Results of a generalized least squares regression analysis of 601 records found that copper concentration was as much as 60 mg/kg wet weight higher for dairy calves than for beef calves (P < 0.00001, R2 = 0.172). For both beef and dairy calves, copper concentration also was significantly associated with linear (P = 0.001) and nonlinear (age2, P = 0.002; age3, P = 0.008) functions of age, where the concentration increased for the first 2 months, then declined until 9 months of age, after which it began to increase. Sex was not associated with copper concentration (P = 0.22). The 95% confidence intervals of the mean concentration for a group of calves and the 95% prediction intervals of a single concentration value for an individual animal are presented. As a consequence of the significant influence of both production class (dairy or beef) and age on liver copper concentration of calves with no evidence of copper toxicosis or copper deficiency, diagnosis of copper imbalances based on liver copper concentration in calves should consider the diagnostic covariates of age and production class.

Mycoplasma otitis in California calves.

A retrospective study of Mycoplasma otitis in California calves submitted for necropsy between 1993 and 2002 was conducted to characterize the demographic features of the disease and the pathologic findings associated with infection. Sixty-one confirmed cases of Mycoplasma otitis were identified among 20,525 necropsied cattle. All affected animals were calves, ranging in age from 2 weeks to 4 months and with a median age of 1.5 months. Ninety-two percent of the cases were dairy breeds. A higher percent of necropsied calves with Mycoplasma otitis were males (0.45%) than females (0.23%). The proportion of cases that had Mycoplasma otitis increased from 1993 to 2002, and there was a significant (P < 0.05) seasonal distribution, with the highest proportion in the spring and the lowest in the summer months. Infections involved both the middle and inner ear and were characterized by a suppurative inflammatory response with extensive bony involvement. Three species of Mycoplasma were isolated from the ears: M. bovis, M. bovirhinis, and M. alkalescens. Concurrent pneumonia occurred in 47 cases (77%), and Mycoplasma was isolated from the lungs of 30 of those cases. The increasing proportion of Mycoplasma otitis cases in the past 10 years emphasizes the importance of identifying risk factors that could be modified to lower the incidence of this disease in calves.

Influence of age, sex, and production class on liver zinc concentration in calves.

Determination of zinc concentrations in the liver of calves and young stock is commonly requested by practitioners and nutritionists to assess whether they receive an appropriate amount of zinc in their diet. However, interpretation of liver zinc concentrations is currently based on information reported for adult cattle for which the health status was unknown and irrespective of production class, sex, and age. A retrospective study of necropsy reports was undertaken to assess the relationships between liver zinc concentrations and age, sex, and production class for calves that did not have a history compatible with zinc toxicosis or zinc deficiency. Results of a generalized least squares, polynomial regression analysis of 474 records found that zinc concentration was not affected by sex (P = 0.29) or production class (P = 0.50). Zinc concentration was significantly associated with linear (P < 0.00001) and nonlinear (quadratic, P = 0.0039) functions of age (r2 = 0.1503), where the concentration decreased from 93 mg/kg wet weight at 30 days of age to 57 mg/kg wet weight at 9 months of age, after which it began to increase. The age-specific 95% confidence limits of the mean concentration for a group of calves and the 95% prediction limits of a single concentration value for an individual animal estimated in this study suggest reconsideration of the recommended limits for liver zinc concentration in calves. As a consequence of the significant influence of age on liver zinc concentration of calves presumably not experiencing zinc toxicosis or deficiency, diagnosis of zinc imbalances based on liver zinc concentration needs to consider age as a diagnostic covariate.

Effect of multiple sampling on diagnostic sensitivity.

Overall diagnostic sensitivity is the probability that a diagnostic procedure will detect an agent if the tested animal is indeed infected. The overall or effective sensitivity is a function of both the probability that the assay will detect the agent if it is present in the sample tested and the probability that the agent will be present in the sample tested if the animal is infected with the agent. Thus, even with a highly sensitive assay, the probability of detecting an infected animal may be low or nil if the sampling procedure failed to capture the agent in samples tested by the assay. In this article, it is demonstrated how increased frequency of testing, such as testing multiple subsamples, can have a profound effect on increasing the overall sensitivity of a diagnostic procedure.

Effect of bovine viral diarrhea virus infection on fertility of dairy heifers.

A prospective field study in heifers from birth to first breeding was undertaken on two commercial dairies to assess the effect of bovine viral diarrhea virus (BVDV) congenital and post-natal infection (PNI) on fertility. A high BVDV Type 2 antibody titer (1:4096) at 10 months of age was associated with 32 more days to conceive, compared with a low titer (1:128). Conversely, infection with BVDV by 5-6 months of age and high BVDV Type 2 titers 1 month before conception or breeding was associated with improved fertility. Heifers with evidence of congenital BVDV infection had lower fertility than non-infected heifers (15-42 days longer time-to-first AI), which depended on BVDV Type 2 titers at 10 months of age. Neospora caninum infection was associated with additional services per conception (SPC) and Leptospira interrogans infection was associated with a delay in the time-to-first breeding. It appears that under field conditions, the effect of subclinical BVDV infection on subsequent heifer fertility may be due to a complex of interrelationships among multiple BVDV infections that depend on the type and timing of infection relative to reproductive development and events.

A simulation model of intraherd transmission of foot and mouth disease with reference to disease spread before and after clinical diagnosis.

Intraherd transmission of foot and mouth disease virus (FMDV) was examined using a simulation model for a hypothetical 1,000-cow dairy, assuming clinical diagnosis was made when at least 1% (10 cows) or 5% (50 cows) had clinical signs of FMD, I index case cow, and transition state distributions for the latent, subclinically infectious, and clinically infectious periods of FMD calculated from published data. Estimates assumed for the number of animal-to-animal contacts (k) adequate for transmission ranged from 0.6 to 9.0 per hour (13.7-216.0 per day). A total of 40,000 iterations (5,000 for each scenario, assessing 4 adequate contact rates and 2 detection criteria) were run. The model predicted that FMD would not be diagnosed in the herd until 10.0-13.5 days after the index case cow had become infected, at which time between 65% and 97% of the cows (646-967 cows) to nearly 100% (978-996 cows) would already have become infected with the virus, if the number of cows showing clinical signs of FMD at the time of diagnosis were 10 or 50, respectively. At the time of diagnosis, the simulated number of infectious cattle varied substantially from 82-472 to 476-537 cows, depending on adequate contact rate and whether the diagnosis was made when 10 or 50 animals were showing clinical signs, respectively. The simulated number of infectious cows increased rapidly during the first few days after diagnosis. In the scenario where at least 10 cows showing clinical signs was necessary before a clinical diagnosis was made, each day after diagnosis, the number of infectious animals increased by nearly 100 to more than 200 cases per day up to day 5, assuming 0.57-9.0 animal-to-animal contacts per hour, respectively. Results obtained when it was assumed that at least 50 clinical cases were present at the time of diagnosis showed smaller relative increases because nearly one-half of the herd was projected to be infected at the time of diagnosis. From these results, it is clear that once an individual in a herd becomes infected with FMDV, herd infectivity is not static, rather it accelerates as would be expected as long as there are sufficient susceptible animals to sustain the increasing transmission rate, after which time the rate at which new infections occurs will diminish. Results indicate that biosecurity strategies aimed at minimizing both intraherd and interherd contact will be critical in minimizing the spread of FMD before the initial diagnosis is made. In addition, simulations suggest that very early clinical diagnosis of FMD and effective isolation or depopulation and disposal will be critical in limiting the number of infectious animals capable of transmitting the virus to other herds and thus in timely control of an epidemic. Early diagnosis will rely on early virus detection from animals in the preclinical phase of infection, rather than waiting for clinical signs to manifest in sufficient numbers to be noticed and to warrant investigation.

Conceptual foundations for infectious disease surveillance.

The purpose of this report is to offer concepts for consideration in developing infectious disease surveillance systems, defined here as active, formal, and systematic processes intentionally directed to rapidly seek out and identify infectious disease agents or disease. Performance of surveillance systems can be judged by their accuracy (sensitivity and specificity), precision (repeatability), timeliness, multiple utility, and value. Surveillance system operation and function necessary to achieve high performance are defined in part by characteristics of the specific infectious disease, including disease transition state dynamics, that define probabilities of being in the latent, infectious, or clinical phase of disease. Two key components of surveillance are the sampling scheme, which is intended to maximize the probability of capturing an infected animal or specimen as soon as possible after the herd has been exposed, and the diagnostic assays, which should maximize the probability of detecting the agent, or evidence of the agent, if it is present in the specimen, while minimizing the likelihood of a false-positive result. Proportional risk sampling, targeted sampling, and repeated sampling are strategies that can improve overall surveillance system accuracy and particularly the temporal sensitivity related to early detection. Hierarchical sampling schemes and multiplexed assays can maximize efficiency and improve utility by serving multiple surveillance systems and purposes. Development of the surveillance systems needed to address emerging and foreign animal diseases will necessarily require design and architecture that are highly probability-driven to maximize surveillance sensitivity and specificity and to minimize cost.