| dc.description.abstract | Controlling and preventing outbreaks of infectious diseases in human and non-human animals is a priority objective for public and animal health professionals, and failure to do so can result in a significant loss of life and have adverse effects on the economy and on the conservation of biodiversity. Free-ranging domestic dogs are reservoirs for several zoonotic diseases, and are currently the focus of two major public health programmes; the ‘Zero by 30’ strategic plan to eliminate dog-mediated rabies and the Guinea worm eradication programme. The management of these and other dog-mediated diseases would benefit from insights into the social and spatial ecology of free-ranging domestic dogs, and their relationships to the dynamics of disease transmission. In this thesis, I explore spatial and temporal dynamics in the contact rates and space use of free-ranging domestic dogs in rural Chad; a region where the risks to humans of infection with rabies and Guinea worm are high. I first describe static contact networks for free-ranging dogs, using novel proximity sensors, and then simulate the transmission of a communicable disease similar to rabies through the networks. My results show that there is considerable variation in the number and duration of contacts between individuals, and that communities in the network were defined by household membership. Disease simulations showed that dogs with a higher ranked degree (more first order contacts) have a higher probability of starting an epidemic should they become infected, while those with a higher ranked eigenvector centrality (more second order contacts) tend to cause larger epidemics. Furthermore, in one settlement ranked eigenvector centrality was positively correlated with home range size. I demonstrate that dogs are not equal in the epidemiological risks that they present and that there may be traits (e.g. drivers behind space use) that could identify individuals that present a higher risk. I then explore dog space use using GPS loggers and continuous time movement models to calculate the dogs’ home ranges and to investigate temporal patterns in their activity and space use. I find that dogs had larger ranges in the dry season, but that this was also when 70% of dogs can be more reliably found around their households. Owner activity explained some of the variation in dog space use, whereby dogs from households that went hunting had larger ranges in the dry season. These results demonstrate how dog spatial ecology can be used to inform disease management practices, such as vaccination campaigns, and highlight the potential to use targeted strategies at the household level. Next, I use the detail of dyadic-level interactions to investigate the spatial-temporal variations in contact rates between dogs. I find that the probability that individuals were ever observed in contact and the hourly probability and duration of contacts should they have interacted, were all negatively correlated with the distance between the individual’s households. Contact rates were higher around the household and village where they peaked between 5am-9am and again between 6pm-8pm. However, the duration of contacts were highest when they occurred outside the village. Evidence for seasonal preferential mixing among the sexes was found, whereby male-male and male-female dyads had higher contact rates in the dry season. Next, I quantify the exposure of free-ranging dogs to water sources in Chad; which are considered potential sources for Guinea worm infection. I use the activity of dogs (based on GPS data) as a proxy for variation in exposure. I find that 85% of dogs had visited at least one water source, but that dogs with larger ranges had higher activity levels around water sources. In all field sites and seasons, no more than 4 water sources accounted for 95% of dog visits, and these were all <0.5 km from a household with tracked dogs. These results can be used to prioritise the chemical treatment of water sources that have the highest dog ‘traffic’. For the last data chapter I investigate the risk factors for Guinea worm infection in dogs from rural villages in Ethiopia. I find that exposure of dogs to potential sources of infection was influenced by variations in dog ranging behaviour, owner behaviour and the characteristics of natural water bodies. This study suggests that the classical transmission pathway for Guinea worm (drinking water from a contaminated source) should be the focus for Guinea worm control in non-human animals in Ethiopia. Finally, I tie the results of this thesis together, and discuss the implications that they have for our understanding of the dynamics of disease transmission in free-ranging domestic dogs, and how this might be used to improve the management of dog-mediated diseases. I conclude that insights into the variations in dog ecology, particularly in their contact behaviour, can be useful in the management of dog-mediated diseases, and that relating contact behaviour to drivers of space use could inspire targeted strategies that are tailored to local socio-ecological contexts. | en_GB |
