Content uploaded by Jim Hone
Author content
All content in this area was uploaded by Jim Hone on Sep 10, 2014
Content may be subject to copyright.
A preview of the PDF is not available
Estimation of the dynamics and rate of transmission of classical swine fever in wild pigs
Epidemiology and Infection
Abstract
Infectious diseases establish in a population of wildlife hosts when the number of secondary infections is greater than or equal to one. To estimate whether establishment will occur requires extensive experience or a mathematical model of disease dynamics and estimates of the parameters of the disease model. The latter approach is explored here. Methods for estimating key model parameters, the transmission coefficient (beta) and the basic reproductive rate (RDRS), are described using classical swine fever (hog cholera) in wild pigs as an example. The tentative results indicate that an acute infection of classical swine fever will establish in a small population of wild pigs. Data required for estimation of disease transmission rates are reviewed and sources of bias and alternative methods discussed. A comprehensive evaluation of the biases and efficiencies of the methods is needed.
... Commonly, models of the progress of disease from infected individuals through susceptible populations of con-specifics can be called SIR (Susceptible, Infectious Recovered) (e.g., Boccora and Cheong 1992) or SL (or E) IR (Susceptible, Latent (Exposed), Infectious, Resistant/Recovered) models (e.g., Ferguson et al. 1997). These fall into two basic categories: deterministic (e.g., Pech and Hone 1988) and stochastic (e.g., Hone et al. 1992). Deterministic models simply describe the progress of diseases through homogeneously mixed populations. ...
... Modeling is required to forecast the progression of diseases between feral goats and sheep. As previously mentioned, models of increasing complexity are possible, from simple deterministic (e.g., Anderson and May 1979;Pech and Hone 1988), through spatial deterministic (e.g., Pech and Mcilroy 1990), through stochastic (e.g. , Hone et al. 1992), through lattice spatial models (e.g. , Rhodes and Anderson 1997) through multispecies deterministic (e.g., Dobson and Meager 1996), multispecies stochastic and multispecies lattice models to combined GIS and epidemiology two-species models. Contact rates and transmission coefficients can be calculated using the contacts, £, determined in this paper and density estimates yet to be derived. ...
... This concept is sometimes referred to as the fundamental law of epidemiology, and it provides a quantitative approach to determining the host status of species for a given pathogen (Chapter1). A closely related epidemiological parameter of great importance is the threshold density for disease establishment, denoted K T (Kermack & McKendrick 1927 The situation is not always quite so bleak for diseases of domestic animals (e.g., see Nodelijk et al. 2000), though Ferguson et al. (1999) Dye et al. (1992); b Roberts (1996); c Barlow (2000); d Caley & Ramsey (2001); e Tompkins et al. (2000); f McCarty & Miller (1998); g Hone et al. (1992). Dobson & Meagher (1996); b Anderson & Trewhella (1985); c Pech & Hone (1988); d Pech & McIlroy (1990); e Caley (1993); f Dexter (1995); g Hone (1994b); h Swinton et al. (1998); i Keeling & Gilligan (2000); j ; k Rhodes et al. (1998). ...
This thesis is about making inference on the host status of feral ferrets in New Zealand for Mycobacterium bovis, the aetiological agent of bovine tuberculosis. The central question addressed is whether the rate of intra-specific transmission of M. bovis among ferrets is sufficient for the disease to persist in ferret populations in the absence of external, non-ferret sources of infection (inter-specific transmission). The question is tackled in three parts-firstly using model selection to identify suitable models for estimating the force of M. bovis infection in ferret populations; secondly applying statistical hypothesis testing to the results of planned manipulative field experiments to test the relationship between M. bovis infection in brushtail possums and that in ferrets; and thirdly using modelling to estimate intra-specific disease transmission rates and the basic reproductive rate (R o) of M. bovis infection in ferrets. The model selection approach clearly identified the hypothesis of oral infection related to diet was, as modelled by a constant force of infection from the age of weaning, the best approximation of how M. bovis infection was transmitted to ferrets. No other form of transmission (e.g., during fighting, mating, or routine social interaction) was supported in comparison. The force of infection (λ) ranged from 0.14 yr-1 to 5.77 yr-1 , and was significantly higher (2.2 times) in male than female ferrets. Statistical hypothesis testing revealed transmission of M. bovis to ferrets occurred from both brushtail possums and ferrets. The force of M. bovis infection in ferrets was reduced by 88% (λ=0.3 yr-1 vs. λ=2.5 yr-1) at sites with reductions in the population density of sympatric brushtail possum populations. A smaller decline in the force of infection resulting from the lethal cross-sectional sampling of the ferret populations was also demonstrated. The modelling approach estimated the basic reproductive rate (R o) of M. bovis infection in ferrets in New Zealand to vary from 0.17 at the lowest population density (0.5 km-2) recorded to 1.6 at the highest population density (3.4 km-2) recorded. The estimates of R o were moderately imprecise, with a coefficient of variation of 76%. Despite this imprecision, the R o for M. bovis infection in ferrets was significantly less than unity for all North Island sites surveyed. Hence it is inferred ferrets are spillover hosts (0<R o <1) for M. bovis infection in these environments. That is, M. bovis infection will progressively disappear from these ferret populations if the source of inter-specific transmission is eliminated. The estimates of R o for M. bovis infection in South Island ferret populations were above one (the level required for disease establishment) for a iii number (5/10) of populations, though the imprecision made it impossible to ascertain whether R o was significantly greater than one. The estimated threshold population density (K T) for disease establishment was 2.9 ferrets km-2. It is inferred that, given sufficient population density (>K T), the rate of intra-specific transmission of M. bovis among ferrets is sufficient for the disease to establish in ferrets in the absence of inter-specific transmission. In these areas, ferrets would be considered maintenance hosts for the disease. Active management (e.g., density reduction or vaccination) of ferrets would be required to eradicate M. bovis from ferret populations in these areas, in addition to the elimination of sources of inter-specific transmission, particularly brushtail possums. v
... Referring to reported values of R 0 in CSF among Eurasian wild boar outside Japan, which ranged from 1.1 to 2.8 [27,50,51], we assumed that the plausible value of R 0 in CSF in Japanese wild boar can range between 1.0 and 3.0. Focusing on the cumulative impact of four vaccine campaigns that were completed in week 50, we estimated the required vaccination effort at week 28 compared to the cumulative vaccination effort until week 50 [V eff,i (50)] by varying the value of R 0 . ...
Understanding the impact of vaccination in a host population is essential to control infectious diseases. However, the impact of bait vaccination against wildlife diseases is difficult to evaluate. The vaccination history of host animals is generally not observable in wildlife, and it is difficult to distinguish immunity by vaccination from that caused by disease infection. For these reasons, the impact of bait vaccination against classical swine fever (CSF) in wild boar inhabiting Japan has not been evaluated accurately. In this study, we aimed to estimate the impact of the bait vaccination campaign by modelling the dynamics of CSF and the vaccination process among a Japanese wild boar population. The model was designed to estimate the impact of bait vaccination despite lack of data regarding the demography and movement of wild boar. Using our model, we solved the theoretical relationship between the impact of vaccination, the time-series change in the proportion of infected wild boar, and that of immunised wild boar. Using this derived relationship, the increase in antibody prevalence against CSF because of vaccine campaigns in 2019 was estimated to be 12.1 percentage points (95% confidence interval: 7.8–16.5). Referring to previous reports on the basic reproduction number ( R 0 ) of CSF in wild boar living outside Japan, the amount of vaccine distribution required for CSF elimination by reducing the effective reproduction number under unity was also estimated. An approximate 1.6 (when R 0 = 1.5, target vaccination coverage is 33.3% of total population) to 2.9 (when R 0 = 2.5, target vaccination coverage is 60.0% of total population) times larger amount of vaccine distribution would be required than the total amount of vaccine distribution in four vaccination campaigns in 2019.
... These models can be further refined but the problems will remain, particularly in the estimation of the transmission coefficient (β). An alternative approach for estimating β with greater robustness would be to use a method similar to that of Hone et al. (1992) or the approach of Chapter 6 with a more suitable virus. Different methods of estimating the transmission coefficient are outlined by Hone et al. (1992). ...
This thesis is concerned with studying aspects of the ecology of feral pigs in the wet-dry tropics of the Northern Territory. The data are needed for use in the management of feral pigs to reduce their agricultural and potential epidemiological impact. Particular emphasis is placed on collecting data needed for modelling foot-and-mouth disease in feral pigs, estimating agricultural damage caused by pigs and evaluating control techniques. All fieldwork was conducted in the Douglas Daly district of the Northern Territory.
Population ecology: Mark-recapture techniques were employed to monitor trends in pig population density in a site containing cereal crops over a period of 18 months. Over the duration of this period there was no significant trend for increase or decrease but the population density fluctuated seasonally between 2.2 pigs km-2 and 3.5 pigs
km-2. Breeding of feral pigs occurred year round with a peak coinciding with the start of the wet season and the burst of primary productivity associated with it. Mortality was highly seasonal, being highest during the late dry season and lowest during the wet season. A numerical response model was developed relating observed exponential rate of increase of feral pigs to antecedent rainfall. The model estimated the maximum exponential rate of increase (r_max) of feral pigs occupying a predominantly savanna woodland habitat to be 0.065 (S.E. ± 0.017) per month or 0.78 (S.E. ± 0.21) per year. Comparison of indices of feral pig population density in cropping and non-cropping areas showed that the presence of intensive cereal cropping increased feral pig population density almost four-fold. In the absence of cereal crops, population density was estimated to be 0.8 pigs km-2. I interpret the response to rainfall and the variation in density as evidence that feral pig populations are limited primarily by food availability.
Movements and habitat use: Studies of feral pig movements in the Douglas River study site showed them to concentrate on the riparian vegetation strip bordering the Douglas River and cereal crops (sorghum, maize) and crop residues remaining after harvesting (stubble). The importance of cereal crops is shown by the large reduction in population density in their absence. Aggregate, seasonal and daily movements of feral pigs were measured using a combination of mark-recapture and radio-tracking techniques. The mean aggregate home-range was 31.2 km2 for boars and 19.4 km2 for sows. The mean daily home-range of 1.0 km2 for boars and 0.7 km2 for sows was small in comparison to the mean aggregate home-range, showing that feral pigs do not utilise all their home-range simultaneously. A comparison of home-range, body-weight, and population density of feral pigs published in the literature shows the three variables to be significantly related. Home-range size of feral pigs is positively related to body-weight and inversely proportional to population density. Pigs were sedentary, with no tendency to disperse great distances from their initial home-range.
Disease modelling: Modelling of the spatial and temporal dynamics of a hypothetical foot-and-mouth disease (FMD) outbreak yielded predictions important for exotic disease contingency planning in the Northern Territory. Minimum estimates of the threshold density for the establishment of FMD range from 0.6 pigs km-2 to 2.0 pigs km-2. Minimum estimates of the threshold isolated group size for the establishment of FMD range from 52 to 192 pigs. Based on these estimates, the establishment of an outbreak of FMD on the major floodplain habitats of the Northern Territory is much more likely than in a woodland habitat due to the much higher density of pigs. The model predicts that where pig distribution is continuous, the velocity of spread of FMD virus will be at between 0.2 to 2.3 km day-1. Some problems remain with the estimation of key parameters in the epidemiological model used, in particular the transmission coefficient. Not enough is known about the continuity of feral pig distribution in the 'Top End' to accurately predict the broad-scale rate of spread of an FMD outbreak.
The potential for using viruses already occurring in feral pigs to estimate parameters needed for modelling exotic diseases such as foot-and-mouth disease was examined using porcine parvovirus. The applicability of the results obtained to modelling foot-and-mouth disease in feral pigs were unclear due to the different modes of transmission of the two viruses. Despite this, the basic idea showed promise, and other viruses that should occur in feral pigs and could potentially provide more useful information due to them having more similar modes of transmission to foot-and-mouth disease were identified.
Control techniques (Trapping): Factors affecting the success rate of traps for catching feral pigs were identified by constructing a generalised linear model relating capture rates of pigs to environmental variables. The most significant variable was time of year, with capture rates being highest during the late dry season and lowest during the late wet season. The next most important variables were the presence of fresh pig tracks at the trap site before construction and vegetation type, with capture rates being higher in closed forest, open forest and woodland compared to open woodland and low open woodland. Other variables identified as significant in influencing capture rates were whether pigs had previously eaten bait at the trap site; presence of rooting; bait type and distance from water. The model developed provides a useful framework for planning and conducting trapping programs of feral pigs in the Northern Territory.
Control techniques (Hunting): The effectiveness of a small team of hunting dogs for removing feral pigs was examined in relation to group size of feral pigs encountered. Hunting dogs were highly successful (90% success rate) when encountering solitary pigs. Percentage success rate rapidly declined as the group size of pigs increased. Hunting dogs did not affect the broad-scale movements of pigs. It was concluded that hunting with dogs is an effective technique for removing residual pigs after densities have been reduced by other forms of control.
Impact on agricultural crops and cost-effectiveness of control techniques: The effect of feral pigs on a range of cereal crops was measured and the cost-effectiveness of different control techniques was evaluated. Feral pigs were found to cause significant economic losses by reducing the harvestable yields of both maize and sorghum crops. The reduction in yield attributable to pigs ranged from 7% to 50% depending on the size and yield of the crop and the number of pigs depredating on it. The destructive potential per pig per crop was estimated to be 100 kilograms of maize or sorghum, or $28.5 per pig in monetary terms. No significant damage occurred in a crop protected by pig netting with an electric outrigger. Control of feral pigs by trapping, poisoning, helicopter shooting and exclusion fencing was evaluated and found to be cost beneficial in all cases with exclusion fencing estimated to be the most cost beneficial. A 76% reduction in the number of feral pigs resulted in a 71% reduction in the extent of crop damage. The relatively large home-ranges of feral pigs require that control efforts against feral pigs are coordinated with adjoining landholders.
... Based on previous modelling studies of CSF in wild boar populations (21,29) and feral pigs (28,36), the transmission dynamics of CSF can be written as follows: ...
Understanding the morbidity and lethality of diseases is necessary to evaluate the effectiveness of countermeasure against the epidemics (e.g., vaccination). To estimate them, detailed data on host population dynamics are required; however, estimating the population size for wildlife is often difficult. We aimed to elucidate the morbidity and lethality of classical swine fever (CSF) currently highly prevalent in the wild boar population in Japan. To this end, we estimated lethality rate, recovery rate, and case fatality ratio (CFR) of CSF without detailed data on the population estimates of wild boar. A mathematical model was constructed to describe the CSF dynamics and population dynamics of wild boar. We fitted the model to the (i) results of the reverse transcription polymerase chain reaction (RT-PCR) test for the CSFV gene and the (ii) results of the enzyme-linked immunosorbent assay (ELISA) test for the antibody against CSFV in sampled wild boar. In the 280 wild boar sampled from September 2018 to March 2019 in the major CSF-affected area in Japan, the lethality rate and recovery rate of CSF per week were estimated as 0.165 (95% confidence interval: 0.081–0.250) and 0.004 (0–0.009), respectively. While the estimate of lethality rate of CSF was similar with the estimates in previous studies, the recovery rate was lower than those reported previously. CFR was estimated as 0.959 (0.904–0.981) using our estimate of recovery rate. This study is the first to estimate lethality rate of CSF from the dynamics of CSF epidemics in the wild boar population. Since the value of CFR is sensitive to the value of recovery rate, the accuracy in the estimate of recovery rate is a key for the accurate estimation of CFR. A long-term transmission experiment of moderately virulent strains may lead to more accurate estimation of the recovery rate and CFR of CSF.
... Bait vaccines for wild boar were employed during CSF outbreaks in Germany and France [15,16]. The estimations of the ideal vaccination rate in wild boar for the control of CSF were reported as 41% using a deterministic model, or from 9% to 52% using a stochastic model based on an outbreak of CSF in Pakistan [17,18]. ...
The prolongation of the classic swine fever (CSF) outbreak in Japan in 2018 was highly associated with the persistence and widespread of the CSF virus (CSFV) in the wild boar population. To investigate the dynamics of the CSF outbreak in wild boar, spatiotemporal analyses were performed. The positive rate of CSFV in wild boar fluctuated dramatically from March to June 2019, but finally stabilized at approximately 10%. The Euclidean distance from the initial CSF notified farm to the farthest infected wild boar of the day constantly increased over time since the initial outbreak except in the cases reported from Gunma and Saitama prefectures. The two-month-period prevalence, estimated using integrated nested Laplace approximation, reached >80% in half of the infected areas in March–April 2019. The area affected continued to expand despite the period prevalence decreasing up to October 2019. A large difference in the shapes of standard deviational ellipses and in the location of their centroids when including or excluding cases in Gunma and Saitama prefectures indicates that infections there were unlikely to have been caused simply by wild boar activities, and anthropogenic factors were likely involved. The emergence of concurrent space–time clusters in these areas after July 2019 indicated that CSF outbreaks were scattered by this point in time. The results of this epidemiological analysis help explain the dynamics of the spread of CSF and will aid in the implementation of control measures, including bait vaccination.
Rabies is a serious microparasitic disease due to the transmission of a virus between hosts. One of the most important animal hosts is the red fox (Vulpes vulpes). Write down an SI model for this mammal assuming that demography is logistic, transmission is density-dependent and no infected animal can recover from the disease.
In Japan, classical swine fever re-emerged in September 2018, after an absence of 26 years, and spread among wild boar populations. This study was conducted to understand the transmissibility of classical swine fever virus in wild boar populations, limited by the impossibility of a correct case count because of the nature of wildlife. Wildlife PCR test results between September 2018 and February 2019 were collected from the homepages of Gifu, Aichi, and Mie Prefectures. In descriptive epidemiology, the geographical spread in the three prefectures, and temporal patterns of PCR test positivity in Gifu Prefecture were observed. Using the Susceptible (S ) ‐ Exposed (E ) ‐ Infectious (I ) ‐ Recovery (R ) model, assuming the weekly test positivity follows beta distribution, parameters were estimated in a Markov Chain Monte Carlo simulation, and the basic reproduction number (R0) was calculated. As a result, the mean infectious period was as long as 100 ‐ 145 days, and R0 was estimated to be 4.2 ‐ 5.1. In this analysis, however, population dynamics considering births and deaths and population density were not considered. In the future, a more detailed study based on wild boar field data is necessary.
The abundance of feral pigs in Australia has been estimated previously and been a topic of some debate. This study aims to update a previous estimate of abundance (13.5 million, 95% CI: 3.5 million to 23.5 million) of feral pigs in Australia. Abundance estimates for the 1970s, 1980s, 1990s, 2000s and 2010s were collated from published literature. Mean abundances in the middle decades were estimated using the ratio method. The average abundance of feral pigs varied from 4.4 million (95% CI: 2.4 million to 6.3 million) in the 1980s, to 3.0 million (95% CI: 2.3 million to 3.7 million) in the 1990s, to 3.2 million (95% CI: 2.4 million to 4.0 million) in the 2000s. Mean density across all 142 studies was 1.03 pigs km–2. The average abundance of feral pigs in Australia during the 1980s to 2000s was much lower and more precise than estimated previously, so scientists and managers should update their use of abundance estimates. Density estimates are above, and below, estimates of threshold host densities for infectious exotic disease establishment.
La maladie de l’œdème est une maladie connue depuis de nombreuses années chez le porc. Les premiers cas recensés dans une population de suidés sauvages sont apparus en 2013 en Ardèche. Un nouveau foyer de cette maladie est ensuite apparu en 2016 dans les Pyrénées-Orientales à la frontière entre la France et l’Espagne. Comprendre les facteurs permettant son apparition ainsi que sa transmission est nécessaire afin d’anticiper de futures mortalités dues à cette maladie. Dans cette thèse, une analyse épidémiologique de cette maladie chez le sanglier a été réalisée. Des clusters de mortalités sont alors apparus et ont permis de mettre en évidence une possible source de contamination unique et récurrente dans le temps. La mise en place d’une nouvelle méthode pour étudier la détectabilité des cadavres de sanglier a souligné la difficulté de retrouver des cadavres de sanglier en forêt. La dernière analyse épidémiologique à partir d’un modèle de type « Spatial point pattern » a mis en avant de possibles facteurs de risque d’apparition et de transmission qui ont ensuite été analysés plus précisément. L’analyse des données issus des tableaux de chasse en Ardèche a été réalisée afin de détecter des variations de la densité et du ratio J/A des populations de sanglier suggérant un stress alimentaire chez le sanglier, un prodrome ou une conséquence de la maladie. Aucun stress alimentaire ne fut détecté lors de cette analyse. Des hypothèses ont pu être émises pour expliquer certaines variations observées : i) la conséquence directe de la maladie, ii) un phénomène environnemental particulier et iii) un évènement pathogénique. La piste de l’événement pathogénique a été approfondie avec la découverte du SDRP (syndrome dysgénésique et respiratoire du porc). Les interactions porcs-sangliers, nombreuses en Ardèche, ont été déterminées comme potentiellement responsables du passage de la bactérie entre le compartiment domestique et sauvage. Une étude génétique a également été effectuée pour investiguer le gène alpha-1-fucosyltransferase associé à la sensibilité du porc à la maladie. Tous les sangliers analysés étaient sensibles à la maladie. D’autres analyses complémentaires sont nécessaires afin de comprendre au mieux cette maladie ainsi que les différents facteurs de risque pour l’apparition mais également la transmission.
Methods for surveying the extent of feral pig rooting, feral pig dung and feral pigs counts were investigated in Namadgi National Park, Australia. The repeatability, precision and accuracy of the methods were examined over twelve months. Repeatable measures od pig rooting and counts of dung pellets were obtained. Rooting was a more accurate indicator of past feral pig presence than dung and counts of feral pigs. The results and their implications for surveying populations of feral pig are discussed.
This chapter deals with various aspects of the humoral and cellular immune defense mechanisms as well as with some possible immunopathological events associated with hog cholera virus (HCV) infection in pigs.
Examines how a simple deterministic model of the interaction between foxes Vulpes vulpes and rabies can be used both to explain certain features of the temporal dynamics of the disease and to explore possible consequences of culling and vaccination of wild foxes in efforts to control or eliminate rabies. -from Author
The pig appears to be the only domestic animal species, which is naturally infected by the virus of hog cholera (HC). All breeds and ages are considered susceptible, although adults generally stand a better chance to survive the infection. Natural outbreaks of HC also may occur in European wild boar (Sus scrofa ferus).