Michael P Ward

University of Sydney, Sydney, New South Wales, Australia

Are you Michael P Ward?

Claim your profile

Publications (135)315.52 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Pig farming is a common practice among smallholder farmers in Nusa Tenggara Timur province (NTT), eastern Indonesia. To understand their production systems a survey of smallholder pig farmers was conducted. Eighteen villages were randomly selected across West Timor, Flores and Sumba islands, and 289 pig farmers were interviewed. Information on pig management, biosecurity practices, pig movements and knowledge of pig health and disease, specifically classical swine fever was collected. The mean number of pigs per herd was 5.0 (not including piglets) and total marketable herd size (pigs ≥ two months of age) did not differ significantly between islands (P = 0.215). Chickens (71%) and dogs (62%) were the most commonly kept animal species in addition to pigs. Pigs were mainly kept as a secondary income source (69%) and 83% of farmers owned at least one sow. Seventy-four percent (74%) of pigs were housed in a kandang (small bamboo pen) and 25% were tethered. Pig feeds were primarily locally sourced agricultural products (93%). The majority of farmers had no knowledge of classical swine fever (91%) and biosecurity practices were minimal. Forty-five percent (45%) reported to consuming a pig when it died and 74% failed to report cases of sick or dead pigs to appropriate authorities. Sixty-five percent (65%) of farmers reported that a veterinarian or animal health worker had never visited their village. Backyard slaughter was common practice (55%), with meat mainly used for home consumption (89%). Most (73%) farmers purchased pigs in order to raise the animal on their farm with 36% purchasing at least one pig within the last year. Predominantly fattener pigs (34%) were given as gifts for celebratory events, most commonly for funerals (32%), traditional ceremonies (27%) and marriages (10%). For improved productivity of this traditional low-input system, research incorporating farming training and improved knowledge on pig disease and biosecurity needs to be integrated with greater access to extension services.
    Preventive Veterinary Medicine 12/2014; · 2.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Knowledge of live animal movement through markets and from farm-to-farm is needed to inform strategies for control of trans-boundary animal diseases (TADs) in south-east Asia, particularly due to consumer preference for fresh meat. In eastern Indonesia a TAD of principal interest for control is classical swine fever (CSF) due to its impacts on smallholder farmers. Pig movement is considered a contributor to failure of current CSF control efforts but pig movement patterns are not well understood. This study investigated movement of live pigs in West Timor, Flores and Sumba islands during 2009-2010, with the aim of informing CSF control policies for Nusa Tenggara Timor province. A market survey of 292 pig sellers and 281 pig buyers across nine live pig markets and a farmer survey across 18 villages with 289 smallholder farmers were conducted and information collected on pig movements. The data obtained was used for social network analysis (SNA) on formal (via a market) and informal (village-to-village) movements using information on trading practices, source and destination locations, and the number of pigs being moved. Both inter- and intra-island movements were identified, however inter-island movement was only observed between Flores and Sumba islands. West Timor and Sumba had highly connected networks where large numbers of villages were directly and indirectly linked through pig movement. Further for West Timor, both formal and informal pig movements linked the capital Kupang, on the eastern end of the island to the western districts bordering East Timor connecting all five districts and demonstrating that informal movement transports pigs over distances similar to formal movement on this island. Sumba had a higher potential for pigs to move to a greater number of sequential locations across the entire island. Flores was found to have a more fragmented network, with pig movements concentrated in its eastern or western regions, influenced by terrain. Markets were confirmed as high-risk locations for the introduction and spread of disease, having over 20 contacts (based on in- and out-degree values) depending on operational day. Villages considered high-risk for CSF spread via informal movements were characterised by higher volume of pig exits and/or linkage to higher numbers of other villages. These findings demonstrate that informal movement (often related to cultural practices) can be extensive and the high level of connectivity dictates that control strategies for CSF and other highly transmissible diseases must be formulated at the provincial level and in collaboration with East Timor. Copyright © 2014 Elsevier B.V. All rights reserved.
    Preventive Veterinary Medicine 12/2014; · 2.51 Impact Factor
  • Michael P Ward, Marta Hernández-Jover
    [Show abstract] [Hide abstract]
    ABSTRACT: The continued spread of rabies in Indonesia poses a risk to human and animal populations in the remaining free islands, as well as the neighbouring rabies-free countries of Timor Leste, Papua New Guinea and Australia. Here we describe the development of a generic risk assessment tool which can be used to rapidly determine the vulnerability of rabies-free islands, so that scarce resources can be targeted to surveillance activities and the sensitivity of surveillance systems increased. The tool was developed by integrating information on the historical spread of rabies, anthropological studies, and the opinions of local animal health experts. The resulting tool is based on eight critical parameters that can be estimated from the literature, expert opinion, observational studies and information generated from routine surveillance. In the case study presented, results generated by this tool were most sensitive to the probability that dogs are present on private and fishing boats and it was predicted that rabies-infection (one infected case) might occur in a rabies-free island (upper 95% prediction interval) with a volume of 1000 boats movements. With 25,000 boat movements, the median of the probability distribution would be equal to one infected case, with an upper 95% prediction interval of six infected cases. This tool could also be used at the national-level to guide control and eradication plans. An initial recommendation from this study is to develop a surveillance programme to determine the likelihood that boats transport dogs, for example by port surveillance or regularly conducted surveys of fisherman and passenger ferries. However, the illegal nature of dog transportation from rabies-infected to rabies-free islands is a challenge for developing such surveillance. Copyright © 2014 Elsevier B.V. All rights reserved.
    Preventive Veterinary Medicine 11/2014; · 2.51 Impact Factor
  • Source
    Salome Dürr, Michael P Ward
    [Show abstract] [Hide abstract]
    ABSTRACT: Disease transmission parameters are the core of epidemic models, but are difficult to estimate, especially in the absence of outbreak data. Investigation of the roaming behaviour, home range (HR) and utilization distribution (UD) can provide the foundation for such parameter estimation in free-ranging animals. The objectives of this study were to estimate HR and UD of 69 domestic dogs in six Aboriginal and Torres Strait Islander communities in northern Australia and to compare four different methods (the minimum convex polygon, MCP; the location-based kernel density estimation, LKDE; the biased random bridge, BRB; and Time Local Convex Hull, T-LoCoH) for investigation of UD and estimating HR sizes. Global positioning system (GPS) collars were attached to community dogs for a period of 1-3 days and positions (fixes) were recorded every minute. Median core HRs (50% isopleth) of the 69 dogs were estimated to range from 0.2 to 0.4ha and the more extended HR (95% isopleth) to range from 2.5 to 5.3ha, depending on the method used. The HR and UD shapes were found to be generally circular around the dog owner's house. However, some individuals were found to roam much more with a HR size of 40-104ha and cover large areas of their community or occasionally beyond. These far roaming dogs are of particular interest for infectious disease transmission. Occasionally, dogs were taken between communities and out of communities for hunting, which enables the contact of dogs between communities and with wildlife (such as dingoes). The BRB and T-LoCoH are the only two methods applied here which integrate the consecutiveness of GPS locations into the analysis, a substantial advantage. The recently developed BRB method produced significantly larger HR estimates than the other two methods; however, the variability of HR sizes was lower compared to the other methods. Advantages of the BRB method include a more realistic analytical approach (kernel density estimation based on movements rather than on locations), possibilities to deal with irregular time periods between consecutive GPS fixes and parameter specification which respects the characteristics of the GPS unit used to collect the data. The BRB method was therefore the most suitable method for UD estimation in this dataset. The results of this study can further be used to contact rates between the dogs within and between communities, a foundation for estimating transmission parameters for canine infectious disease models, such as a rabies spread model in Australia.
    Preventive Veterinary Medicine 07/2014; · 2.51 Impact Factor
  • Isabel Brazier, Mark Kelman, Michael P Ward
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this study was to describe the association between landscape and climate factors and the occurrence of tick paralysis cases in dogs and cats reported by veterinarians in Australia. Data were collated based on postcode of residence of the animal and the corresponding landscape (landcover and elevation) and climate (precipitation, temperature) information was derived. During the study period (October 2010-December 2012), a total of 5560 cases (4235 [76%] canine and 1325 [24%] feline cases) were reported from 341 postcodes, mostly along the eastern seaboard of Australia and from the states of New South Wales and Queensland. Significantly more cases were reported from postcodes which contained areas of broadleaved, evergreen tree coverage (P=0.0019); broadleaved, deciduous open tree coverage (P=0.0416); and water bodies (P=0.0394). Significantly fewer tick paralysis cases were reported from postcodes which contained areas of sparse herbaceous or sparse shrub coverage (P=0.0297) and areas that were cultivated and managed (P=0.0005). No significant (P=0.6998) correlation between number of tick paralysis cases reported per postcode and elevation was found. Strong positive correlations were found between number of cases reported per postcode and the annual minimum (rSP=0.9552, P<0.0001) and maximum (rSP=0.9075; P=0.0001) precipitation. Correlations between reported tick paralysis cases and temperature variables were much weaker than for precipitation, rSP<0.23. For maximum temperature, the strongest correlation between cases was found in winter (rSP=0.1877; P=0.0005) and for minimum temperature in autumn (rSP=0.2289: P<0.0001). Study findings suggest that tick paralysis cases are more likely to occur and be reported in certain eco-climatic zones, such as those with higher rainfall and containing tree cover and areas of water. Veterinarians and pet owners in these zones should be particularly alert for tick paralysis cases to maximize the benefits of early treatment, and to be vigilant to use chemical prophylaxis to reduce the risk of tick parasitism.
    Veterinary Parasitology 05/2014; · 2.55 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Estimates of time spent at migratory stopovers are often used to develop habitat conservation objectives for a variety of avian species, namely waterfowl. Because of limited previous research and a need for accurate conservation planning objectives, we estimated stopover duration and factors influencing stay of mallards (Anas platyrhynchos) in Illinois using radio telemetry during autumns 2009 and 2010. Total stopover duration of migrating mallards was approximately 68 days and was approximately twice that of previous studies which could have significant implications on habitat needs used for conservation planning purposes. Duration of stay post-capture did not vary by sex, body condition, year, or age but was inversely related to capture date. Our results suggest that wetland conservation objectives could increase 16.2% if our contemporary estimates were included in current planning models of the Upper Mississippi River and Great Lakes Region Joint Venture. © 2014 The Wildlife Society.
    Journal of Wildlife Management 04/2014; 78(4). · 1.64 Impact Factor
  • Michael P. Ward
    [Show abstract] [Hide abstract]
    ABSTRACT: Rabies continues to spread through the Indonesian archipelago. During the past 20 years, several islands – including Flores, Ambon and Bali – that had historically been free of rabies have become infected. However, the Dutch East Indies (a Dutch colony that became modern Indonesia following World War II) had been infected since the 1880s. The spread of rabies is a lesson in the emergence of an infectious disease. Reports of human cases treated for rabies and livestock rabies cases from the 1880s to 1917 were compiled. The spatial and temporal distribution of these cases was analysed using maps, spatial statistics and time-series techniques. The first confirmed case of rabies was reported in 1889 from the Batavia [Jakarta] district (although disease suspicion was reported as early as 1884). During the 1890s rabies was already commonly reported from Java and the east coast of Sumatra, and by the late 1890s, from Celebes [Sulawesi]. Between 1900 and 1916, cases were reported from other parts of Java, Sumatra and Sulawesi, and from Borneo, the Moluccas and other outlying islands. Between 1897 and 1916, a total of 8,826 human cases treated for rabies were reported and between 1908 and 1917, 1,033 livestock cases were reported. Most (97.5%) human cases treated were attributed to rabid dogs. Increasing numbers of reports were observed during the period. Between 1908 and 1916 the correlation between human and livestock case reports was 64.2%, and at the district level it was 75.9%. Moderate correlations (>40%) were found between human cases and livestock cases reported up to six months previously. Based on year of first report from each district, human cases were strongly clustered (Moran's autocorrelation 0.47, P = 0.005). The mostly likely spatio-temporal cluster of reported cases of humans treated for rabies originated from the west coast of Sumatra between 1899–1905, and other clusters were identified in west Java (1898–1899), the district of Batavia and in east Java (1910–1911), east Java (1910–1911), Nusa Tengarra Barat (1912), Borneo (1914) and the east coast of Sumatra (1903–1906). Rabies was probably first introduced to the colonial capital of the Dutch Indies, Batavia [Jakarta] in the 1880s. It then spread rapidly throughout most of archipelago during the next two to three decades because of the movement of dogs via the military forces, for trade and as pets, despite government regulations designed to control the epidemic. Such a history suggests that further emergence and reemergence of rabies in rabies-free islands will occur based on an island's location and position within the complex social, trade and transport network that represents the Indonesian archipelago. Targeted surveillance and enforcement of quarantine regulations remain critical, to prevent history repeating itself.
    Preventive Veterinary Medicine 04/2014; · 2.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Nesting success and avian communities were compared between tilled and no-till soybean fields in Illinois. No-till had greater densities of birds than tilled fields, and the overall community in no-till was of greater conservation value due to more grassland birds using no-till compared with tilled fields. Nesting density was greater in no-till (4.5 nests/100 ha) than in tilled (1.6) fields. The most common nesting species were American robins, vesper sparrows, and mourning doves. Nest success, as estimated from daily survival rates, was 19.4% in no-till and 9.4% in tilled fields. Predation was the main cause of nest failure, but 24.4% of failures were caused by farm machinery. The authors propose that the previous year's crop residue and greater abundance of weedy plants in no-till resulted in increased nesting and foraging activity in no-till and greater nest success because of increased opportunity to conceal nests in no-till compared to tilled fields. No-till provides greater benefits to birds than tilled fields, and the large amount of acreage in row crops dictates that we understand the contribution of no-till fields to grassland bird populations.
    Agriculture Ecosystems & Environment 03/2014; 185:59–64. · 3.20 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Most birds are socially monogamous, diurnally active and have small home ranges. These birds occasionally undertake extraterritorial forays, presumably to seek extrapair copulations. We used automated radiotelemetry to examine nocturnal forays and activity of a diurnal, socially monogamous passerine, the yellow-breasted chat, Icterina virens. Males and females forayed during both day and night, but night forays were more common. In addition to varying between day and night, there were sex- and breeding-stage-specific differences in foraying behaviour. Males forayed more than females and more frequently when their female was not fertile. Conversely, females primarily forayed when fertile. We suggest that females foray at night to avoid detection, because their mate's knowledge of potential extrapair copulations may result in harassment or decreased parental care. Both sexes were nocturnally active; however, the activity of males peaked when their mates were fertile, probably to prevent their mates from foraying and guarding them from foraying males. The nocturnal behaviour of chats suggest that extraterritorial forays may be more common than previously thought and that nocturnal behaviours may be crucial for understanding the strategies males and females use to acquire extrapair copulations.
    Animal Behaviour 02/2014; 88:175–184. · 3.07 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Since the global spread of highly pathogenic avian influenza H5N1 during 2005–2006, control programs have been successfully implemented in most affected countries. HPAI H5N1 was first reported in Bangladesh in 2007, and since then 546 outbreaks have been reported to the OIE. The disease has apparently become endemic in Bangladesh. Spatio-temporal information on 177 outbreaks of HPAI H5N1 occurring between February 2010 and April 2011 in Bangladesh, and 37 of these outbreaks in which isolated H5N1 viruses were phylogenetically characterized to clade, were analyzed. Three clades were identified, 2.2 (21 cases), 2.3.4 (2 cases) and (14 cases). Clade 2.2 was identified throughout the time period and was widely distributed in a southeast–northwest orientation. Clade appeared later and was generally confined to central Bangladesh in a north–south orientation. Based on a direction test, clade 2.2 viruses spread in a southeast-to-northwest direction, whereas clade spread west-to-east. The magnitude of spread of clade was greater relative to clade 2.2 (angular concentration 0.2765 versus 0.1860). In both cases, the first outbreak(s) were identified as early outliers, but in addition, early outbreaks (one each) of clade 2.2 were also identified in central Bangladesh and in northwest Bangladesh, a considerable distance apart. The spread of highly pathogenic avian influenza H5N1 in Bangladesh is characterized by reported long-distance translocation events. This poses a challenge to disease control efforts. Increased enforcement of biosecurity and stronger control of movements between affected farms and susceptible farms, and better surveillance and reporting, is needed. Although the movement of poultry and equipment appears to be a more likely explanation for the patterns identified, the relative contribution of trade and the market chain versus wild birds in spreading the disease needs further investigation.
    Preventive Veterinary Medicine 01/2014; · 2.51 Impact Factor
  • Preventive Veterinary Medicine 01/2014; 114(1). · 2.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Australia experienced its first ever outbreak of equine influenza in August 2007. Horses on 9359 premises were infected over a period of 5 months before the disease was successfully eradicated through the combination of horse movement controls, on-farm biosecurity and vaccination. In a previous premises-level case-control study of the 2007 equine influenza outbreak in Australia, the protective effect of several variables representing on-farm biosecurity practices were identified. Separately, factors associated with horse managers' perceptions of the effectiveness of biosecurity measures have been identified. In this analysis we applied additive Bayesian network modelling to describe the complex web of associations linking variables representing on-farm human behaviours during the 2007 equine influenza outbreak (compliance or lack thereof with advised personal biosecurity measures) and horse managers' perceptions of the effectiveness of such measures in the event of a subsequent outbreak. Heuristic structure discovery enabled identification of a robust statistical model for 31 variables representing biosecurity practices and perceptions of the owners and managers of 148 premises. The Bayesian graphical network model we present statistically describes the associations linking horse managers' on-farm biosecurity practices during an at-risk period in the 2007 outbreak and their perceptions of whether such measures will be effective in a future outbreak. Practice of barrier infection control measures were associated with a heightened perception of preparedness, whereas horse managers that considered their on-farm biosecurity to be more stringent during the outbreak period than normal practices had a heightened perception of the effectiveness of other measures such as controlling access to the premises. Past performance in an outbreak setting may indeed be a reliable predictor of future perceptions, and should be considered when targeting infection control guidance to horse owners and managers.
    Preventive Veterinary Medicine 12/2013; · 2.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The continued spread of rabies through the eastern islands of Indonesia poses a risk of rabies introduction to Timor Leste. To prepare for such an incursion and to undertake surveillance activities, the size and distribution of the roaming dog population needs to be estimated. We present the results of the first such surveys ever undertaken in Timor Leste. Roaming dog surveys were undertaken in each capital of the 13 districts of Timor Leste, including the national capital, Dili. Within these locations, local urban areas (aldeias) were targeted and sight-re-sight counts were undertaken on consecutive days. Estimated dog populations were adjusted for the sampling fraction. Overall, counts were performed in a total of 53 of 131 (40.5%) sucos and in 192 of 797 (24.1%) aldeias in these selected sucos. Within the surveyed urban areas, there were an estimated 21.2 people per roaming dog, a ratio substantially higher than the World Health Organization's average global estimate of 10 people per dog. The highest populations of dogs were estimated in the cities of Dili (4919), Baucau vila (3449) and Lospalos (2536). The latter two are important because of their location in the northeast of Timor Leste, where the risk of rabies incursion from recently infected islands in eastern Indonesia, is likely greatest. The sight-resight method of estimating roaming dog populations is practical in developing countries; more use of photography to aid resighting of dogs could increase the accuracy of this method.
    Preventive Veterinary Medicine 12/2013; · 2.51 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A growing body of work shows the benefits of applying social cognitive behavioural theory to investigate infection control and biosecurity practices. Protection motivation theory has been used to predict protective health behaviours. The theory outlines that a perception of a lack of vulnerability to a disease contributes to a reduced threat appraisal, which results in poorer motivation, and is linked to poorer compliance with advised health protective behaviours. This study, conducted following the first-ever outbreak of equine influenza in Australia in 2007, identified factors associated with horse managers' perceived vulnerability to a future equine influenza outbreak. Of the 200 respondents, 31.9% perceived themselves to be very vulnerable, 36.6% vulnerable and 31.4% not vulnerable to a future outbreak of equine influenza. Multivariable logistic regression modelling revealed that managers involved in horse racing and those on rural horse premises perceived themselves to have low levels of vulnerability. Managers of horse premises that experienced infection in their horses in 2007 and those seeking infection control information from specific sources reported increased levels of perceived vulnerability to a future outbreak. Different groups across the horse industry perceived differing levels of vulnerability to a future outbreak. Increased vulnerability contributes to favourable infection control behaviour and hence these findings are important for understanding uptake of recommended infection control measures. Future biosecurity communication strategies should be delivered through information sources suitable for the horse racing and rural sectors.
    BMC Veterinary Research 07/2013; 9(1):152. · 1.74 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Declines of migratory birds have led to increased focus on causative factors for these declines, including the potential adverse effects of habitat fragmentation on reproductive success. Although numerous studies have addressed how proximity to a habitat edge, patch size, or landscape context influence nest survival or brood parasitism, many have failed to find the purported effects. Furthermore, many have sought to generalize patterns across large geographic areas and habitats. Here, we examined evidence for effects of edge proximity, patch size, and landscape context on nest survival and brood parasitism of grassland birds, a group of conservation concern. The only consistent effect was a positive association between edge proximity and brood parasitism. We examined effects of patch size on nest survival (37 studies) and brood parasitism (30 studies) representing 170 and 97 different estimates, respectively, with a total sample size of > 14000 nests spanning eastern North America. Nest survival weakly increased with patch size in the Great Plains, but not in the Midwestern or Eastern United States, and brood parasitism was inversely related to patch size and consistently greater in the Great Plains. The consistency in brood parasitism relative to nest survival patterns is likely due to parasitism being caused by one species, while nest survival is driven by a diverse and variable suite of nest predators. Often, studies assume that predators responsible for nest predation, the main driver of nest success, either are the same or exhibit the same behaviors across large geographic areas. These results suggest that a better mechanistic understanding of nest predation is needed to provide meaningful conservation recommendations for improving grassland bird productivity, and that the use of general recommendations across large geographic areas should only be undertaken when sufficient data are available from all regions.
    Ecological Applications 06/2013; 23(4):879-87. · 4.13 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: On-farm biosecurity practices have been promoted in many animal industries to protect animal populations from infections. Current approaches based on regression modelling techniques for assessing biosecurity perceptions and practices are limited for analysis of the interrelationships between multivariate data. A suitable approach, which does not require background knowledge of relationships, is provided by Bayesian network modelling. Here we apply such an approach to explore the complex interrelationships between the variables representing horse managers' perceptions of effectiveness of on-farm biosecurity practices. The dataset was derived from interviews conducted with 200 horse managers in Australia after the 2007 equine influenza outbreak. Using established computationally intensive techniques, an optimal graphical statistical model was identified whose structure was objectively determined, directly from the observed data. This methodology is directly analogous to multivariate regression (i.e. multiple response variables). First, an optimal model structure was identified using an exact (exhaustive) search algorithm, followed by pruning the selected model for over-fitting by the parametric bootstrapping approach. Perceptions about effectiveness of movement restrictions and access control were linked but were generally segregated from the perceptions about effectiveness of personal and equipment hygiene. Horse managers believing in the effectiveness of complying with movement restrictions in stopping equine influenza spread onto their premises were also more likely to believe in the effectiveness of reducing their own contact with other horses and curtailing professional visits. Similarly, the variables representing the effectiveness of disinfecting vehicles, using a disinfectant footbath, changing into clean clothes on arrival at the premises and washing hands before contact with managed horses were clustered together. In contrast, horse managers believing in the effectiveness of disinfecting vehicles (hygiene measure) were less likely to believe in the effectiveness of controlling who has access to managed horses (access control). The findings of this analysis provide new insights into the relationships between perceptions of effectiveness of different biosecurity measures. Different extension education strategies might be required for horse managers believing more strongly in the effectiveness of access control or hygiene measures.
    Preventive Veterinary Medicine 03/2013; · 2.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: In a previous premises-level case-control study of the 2007 equine influenza outbreak in Australia, the protective effect of several variables representing on-farm biosecurity practices was identified. However, using logistic regression it was not possible to definitively identify individual effects and associations between each of the personal biosecurity measures implemented by horse premises owners and managers in the face of the outbreak. In this study we apply Bayesian network modelling to identify the complex web of associations between these variables, horse premises infection status and other premises-level covariates. We focussed this analysis primarily on the inter-relationship between the nine variables representing on-farm personal biosecurity measures (of people residing on the premises and those visiting), and all other variables from the final logistic regression model of our previous analysis. Exact structure discovery was used to identify the globally optimal model from across the landscape of all directed acyclic graphs possible for our dataset. Bootstrapping was used to adjust the model for over-fitting. Our final Bayesian graphic network model included 18 variables linked by 23 arcs, each arc analogous to a single multivariable generalised linear model, combined in a probabilistically coherent way. Amongst the personal biosecurity measures, having a footbath in place, certain practices of visitors (hand-washing, changing clothes and shoes) in contact with the horses, and the regularity of horse handling were statistically associated with premises infection status. The results of this in-depth analysis provide new insight into the complex web of direct and indirect associations between risk factors and horse premises infection status during the first 7 weeks of the 2007 equine influenza outbreak in Australia. In future outbreaks, unnecessary contact and handling of horses should be avoided, especially by those coming from off the premises. Prior to any such contact, persons handling horses should use a footbath (if present), change their clothes and shoes, and wash their hands.
    Preventive Veterinary Medicine 03/2013; · 2.51 Impact Factor
  • Michael P Ward, Fraser I Lewis
    Preventive Veterinary Medicine 02/2013; · 2.51 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Rabies was first reported in the Sultanate of Oman is 1990. We analysed passive surveillance data (444 samples) collected and reported between 2006 and 2010. During this period, between 45 and 75% of samples submitted from suspect animals were subsequently confirmed (fluorescent antibody test, histopathology and reverse transcription PCR) as rabies cases. Overall, 63% of submitted samples were confirmed as rabies cases. The spatial distribution of species-specific cases were similar (centred in north-central Oman with a northeast-southwest distribution), although fox cases had a wider distribution and an east-west orientation. Clustering of cases was detected using interpolation, local spatial autocorrelation and scan statistical analysis. Several local government areas (wilayats) in north-central Oman were identified where higher than expected numbers of laboratory-confirmed rabies cases were reported. For fox rabies, more clusters (local spatial autocorrelation analysis) and a larger clustered area (scan statistical analysis) were detected. In Oman, monthly reports of fox rabies cases were highly correlated (r(SP)>0.5) with reports of camel, cattle, sheep and goat rabies. The best-fitting ARIMA model included a seasonality component. Fox rabies cases reported 6 months previously best explained rabies reported cases in other animal species. Despite likely reporting bias, results suggest that rabies exists as a sylvatic cycle of transmission in Oman and an opportunity still exists to prevent establishment of dog-mediated rabies.
    Preventive Veterinary Medicine 01/2013; · 2.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this study our aim was to value the benefits of rapid effective trace-back capability-based on a livestock identification system - in the event of a foot and mouth disease (FMD) outbreak. We simulated an FMD outbreak in the Texas High Plains, an area of high livestock concentration, beginning in a large feedlot. Disease spread was simulated under different time dependent animal tracing scenarios. In the specific scenario modeled (incursion of FMD within a large feedlot, detection within 14 days and 90% effective tracing), simulation suggested that control costs of the outbreak significantly increase if tracing does not occur until day 10 as compared to the baseline of tracing on day 2. In addition, control costs are significantly increased if effectiveness were to drop to 30% as compared to the baseline of 90%. Results suggest potential benefits from rapid effective tracing in terms of reducing government control costs; however, a variety of other scenarios need to be explored before determining in which situations rapid effective trace-back capability is beneficial.
    Preventive Veterinary Medicine 01/2013; · 2.51 Impact Factor

Publication Stats

1k Citations
315.52 Total Impact Points


  • 2008–2014
    • University of Sydney
      • Faculty of Veterinary Science
      Sydney, New South Wales, Australia
  • 2004–2014
    • University of Illinois, Urbana-Champaign
      • • Department of Natural Resources and Environmental Sciences
      • • Prairie Research Institute
      • • Department of Animal Biology
      Urbana, Illinois, United States
  • 2013
    • University of Melbourne
      • Faculty of Veterinary Science
      Melbourne, Victoria, Australia
  • 2012
    • Charles Sturt University
      Уогга Уогга, New South Wales, Australia
    • Ministry of Agriculture and Forests, Bhutan
      Thimbu, Thimphu, Bhutan
  • 2011
    • University of New South Wales
      • School of Biological, Earth and Environmental Sciences (BEES)
      Kensington, New South Wales, Australia
  • 2007–2010
    • University of Minnesota Morris
      • College of Veterinary Medicine
      Saint Paul, Minnesota, United States
  • 2005–2010
    • Texas A&M University
      • • Department of Veterinary Integrative Biosciences
      • • College of Veterinary Medicine and Biomedical Sciences
      • • Department of Veterinary Pathobiology
      College Station, TX, United States
  • 2009
    • United States Department of Agriculture
      Washington, Washington, D.C., United States
  • 2002–2005
    • Purdue University
      • Department of Veterinary Clinical Sciences (VCS)
      West Lafayette, Indiana, United States
    • Rosario National University
      Rosario, Santa Fe, Argentina