Sustainable Dog Health Programs Are Possible: West Australian Experiences in Remote Management and Service Delivery

Transcription

1 Sustainable Dog Health Programs Are Possible: West Australian Experiences in Remote Management and Service Delivery By Kathryn Wilks In 1992 the Dog Health Program in Aboriginal Communities was initiated in the Kimberley region of Western Australia as a pilot program to assess the efficacy of a canine breeding and parasite control strategy. The program was borne out of concern by members of several Kimberley communities about the state of health of dogs in their communities. Concern by health professionals regarding the impact of dogs on environmental health and the potential role of dogs in zoonoses transmission was also considered in the development of the project. As a result, a thorough investigation into effective means to provide a canine parasite and breeding control program was undertaken by staff of Murdoch University in conjunction with health staff of the Kimberley region. The pilot program ran for four years and in 1995 was handed back to the people. In 2000, the program is still working and is considered by many environmental health personnel of the area to be one of the most successful programs undertaken in Aboriginal communities. The following will present a broad outline of the Dog Health Program including the historical context of the association between dogs and Aboriginal people, the present understanding of the health status of dogs in remote communities and the approaches to implementing dog management strategies in remote communities that are acceptable and effective. 1. THE HISTORY OF ASSOCIATION OF DOGS, DINGOES AND ABORIGINAL PEOPLE Asian seafarers are considered to have introduced the dingo into Australia as a result of their many voyages to Australia as part of the strong trade and sea links with the mainland people (Flood, 1995; Corbett, 1995). Archaeological evidence points to the entry of the dingo in the north of Australia approximately 4000 years ago, with about 500 years of lag time before the dingo colonised the southern regions of the mainland (Gollan, 1984). The dingo never reached Tasmania due to the severance of Tasmania some 8000 years ago by the rising seas of the Bass Strait (Gollan, 1984; Berndt & Berndt, 1992). Aboriginal people are believed to have quickly adopted the dingo into their lives, although domestication of the dingo by Aboriginal people was only partial (Meggitt, 1965) with very little evidence of selective breeding. Some archaeological evidence of a physical variation from the classical uniform dingo morphology comes from the remains of dingo burials in New South Wales and Victoria which date back about 1000 years (Gollan, 1984), suggesting that some selection did occur, but without any continuity. The interest by the mainland Aboriginal people in the dingo may have been due to the difference of dingoes to any other animal in Australia. Dingoes were avid hunters and carnivores and were the only animals to give birth in a manner similar to humans (Tonkinson, 1984). For these and many other reasons, dingoes were often accepted as members of the family by being given subsection (skin) names that automatically positioned dingoes into society by giving them status such as parent, grandparent, child and so on (Kolig, 1978; Tonkinson, 1984; Ross, 1987). Dingoes were also often the only animals to have special rights to attend treasured rituals and in some cases were considered to be qualified to attend as fully-fledged lawmen (a title given to men expert in religious lore and, as it seems here, dogs too) (Kolig, 1973; Kolig, 1978).

2 Traditionally dingoes were considered useful to Aboriginal people as hunting aides, companions, protectors and spiritual necessities. Some of these uses are evident today. Much controversy surrounds the usefulness of dingoes (and dogs today) as hunting aides, despite this being the most obvious reason for any domestication of such an efficient killer (Meggitt, 1965). Remarks by anthropologists and observers vary from dogs being a liability during hunting due to undisciplined howling and barking (Basedow, 1925; Gould, 1969; Hamilton, 1972; Kolig, 1978) to the hunting dogs being the primary providers of all food (White, 1972). As dingoes were the only animals on the mainland that were trained to any degree (Giles, 1889; Meggitt, 1965; White, 1972; Hayden, 1975), it is generally accepted that dingoes did serve some function in providing food for people. The role of dingoes as companions is still evident. Great affection for dingoes and dogs has been recorded such as the observation by Lumholtz (1889, p. 179) that... the master never strikes, but merely threatens it [the dog]. He caresses it like a child, eats the fleas off it, and then kisses it on the snout. Similarities between children and puppies may have been a reason for this closeness and unreasonable amount of petting and pampering (Basedow, 1925). Dingoes were often taken from the bush as puppies to be reared as substitute children for childless women (Hamilton, 1972). Occasionally the forelegs of the puppies were broken to prevent their return to the wild (Meggitt, 1965) and women have been observed to breastfeed puppies (Berndt & Berndt, 1942; Meggitt, 1965). The dingo (and dogs ) role as a nurturing tool and emotional outlet is considered to be very important (Hamilton, 1972). One of the most important uses of dingoes traditionally and dogs today is as a protector. Dingoes and dogs are considered to be expert at warning against the physical as well as the spiritual. With a keen sense of smell, acute hearing and alertness, dingoes and dogs were very important in protecting small family groups from secret revenge expeditions (Hamilton, 1972; Hayden, 1975) or supernatural threats (Hamilton, 1972; Kolig, 1978). Of the most complex and least understood uses of the dingo is the dingoes role in the creation and dreaming. Roles of the dingo vary from creator on a par to the water snake (Howe, 1993) to asocial marauder (Hamilton, 1972; Kolig, 1978; Tonkinson, 1984; Berndt & Berndt, 1992) and lethal spirit-being (Kolig, 1978). Either way, dingoes did feature heavily in the dreaming for some Aboriginal groups, and severe retribution by spiritual dingoes is still used as a threat to misbehaving people (personal observation). Domestic dogs quickly replaced dingoes after the colonisation of Australia by Europeans. Dogs were occasionally given as placatory gifts to Aboriginal people by the early explorers (Meggitt, 1965), and because of their relative ease to train, better assimilation to people s living arrangements (Kolig, 1978) and perceived social stature (Meggitt, 1965) they were often accepted willingly. Acceptance of the European dog did result in some problems that are evident today. Firstly, European dogs breed twice a year, whereas dingoes only breed once a year (Corbett, 1995). Also, dingoes brought from the bush often returned to it eventually (Lumholtz, 1889; Meggitt, 1965; Kolig, 1978) and very little breeding (or selection) was done in the camps. Secondly, dingoes were self-sufficient and economical. Whereas dogs expected to be fed and cared for by people (a trade off for domestication), dingoes would often hunt for the people, accept leftover food and complement their diet by foraging (Meggitt, 1965; Hamilton, 1972; Hayden, 1975). The rapid transition of Aboriginal people from nomadic lifestyles to permanent residence after colonisation is often considered a factor in the poor environmental health standards experienced by Aboriginal communities today (Gracey, 1992). Concomitant is the perceived overpopulation and

3 poor health of dogs, much of which is also a result of this transition. Population regulation, traditionally controlled by the capture of puppies from the wild without breeding within camps, became difficult after the introduction of the European dog. As a result, the health of dogs suffered due to overcrowding and difficulty in feeding large numbers of dogs. Aboriginal Peoples Attitudes to Canine Sickness and Health Historical literature relating to Aboriginal people s attitude to the welfare of their dogs varies according to region and opinion. Chewings (1936) noted that the dogs were usually emaciated and fed more on affection than food. By contrast, Lane (1928), Lumholtz (1889) and Berndt and Berndt (1942) believed dogs to be well fed. Others have observed that the amount fed to dogs varied according to the amount of food available to the owners (Nind, 1931; Buley, 1905). The apparent lack of feeding of dogs may be a remnant of the days when dingoes roamed the camps and foraged for themselves. Howe (1993), during a biosociological study of the relationship between dogs and Aboriginal people, found that attitudes toward the care of dogs varied according to the status of the dog (owner status in community, sex and age of dog and uniqueness) and the experiences of the owner. Older owners were noted to respond to the needs of their ill dogs and this was considered, in part, to be a traditional cultural imperative on the part of the dog owner. Younger people, raised in contemporary settings, did not have the perception of usefulness of dogs in a traditional sense, and Howe considers that this may reflect less attention to dog care by younger community members. Overall, for much of the post-contact period, Aboriginal people were neither encouraged nor equipped to take responsibility for the health of their dogs, their major alternatives being euthanasia or neglect (Howe, 1993). Euthanasia, particularly by shooting, is also shunned by many groups (Meggitt, 1965; Ross, 1987), but considered to be an unpleasant necessity by some (Howe, 1993). Them doggers, whitefella blokes, they come here and shoot m in the old days, them dogs... They say gimme money or I kill the dog. I say I got no money. So they take m... so I go hide m out bush. That s how we save m. Make them doggers bloody mad! Like they wanna shoot us, too! (George Wallaby, quoted in Arden, 1995, pp. 81.) 2. DOG HEALTH IN ABORIGINAL COMMUNITIES Most accounts of the health status of dogs in Aboriginal communities relate to the zoonotic pathogens. Given the very close association of Aboriginal people and dogs and the frequent accounts of close cohabitation (Ross, 1987; Berndt & Berndt, 1992), the potential for transmission of canine disease to humans is high. Most historical accounts of disease amongst dogs relate to mange (Morris, 1889; Mathew, 1889, cited in Hamilton, 1972), with one author (Mathew) even correlating this condition to similar conditions in the dogs owners. Contemporary observers also note mange (scabies) and parasites to be common ailments (Hamilton, 1972; Ross, 1987), although viral diseases such as distemper have also been noted in communities (personal observation). Ross (1987) stated that few people acknowledge any health hazard resulting from dogs, though beds are shared with dogs and the ground around some camps is strewn with faeces. This and similar comments has resulted in surveys of dogs in Aboriginal communities for zoonotic parasites. Studies into the health status of wild dogs in other countries have also centred on the prevalence of parasites (Pence, Windberg, Pence & Sprowls, 1983; Philips & Scheck, 1991), although antibody titres to a variety of canine viruses and rickettsae were found in wild dogs of the Kruger National Park (Van Heerden, Mills, Van Vuuren, Kelly & Dreyer, 1995). The results of the Kruger National Park study indicated that the wild dogs were exposed to these agents, but links to mortality could not be determined. In the case of dogs in Aboriginal communities, no studies have been done on the seroprevalence of non-zoonotic viral (or rickettsial) agents. Canine distemper, canine parvo-,

4 corona- and rotavirus are regarded as the most significant viral pathogens of domestic dog puppies between the ages of three and six months (Van Heerden et al, 1995). Considering the lack of vaccination of dogs in communities against these agents (personal observation), these agents may be responsible for epidemics from time to time. 3. THE DOG HEALTH PROGRAM The Dog Health Program in Aboriginal Communities project was established in 1992 as a pilot program to assess the benefits of a canine parasite and breeding control program in Aboriginal communities of the Kimberley Region, Western Australia. The pilot program was introduced to 14 communities, divided into three regions (coastal, central and eastern). Figure 1: Communities Involved in the Dog Health Program in the Kimberley Region of Western Australia The objectives of the program were to: * assess the health status of dogs in communities * implement a parasite control program to reduce the potential for zoonoses transmission * implement a breeding control program to reduce the number of dogs in communities, and * use the control program as a vehicle for education regarding the wider environmental health concerns in communities Dogs from each of the pilot communities were identified using a microchip system (Trovan Passive Transponder System ID-100). Signalment data, including sex, age (estimations based on dentition when necessary; see Harvey, 1985), owner name and place of residence, was collected from each newly registered dog. Clinical examinations and dermatological assessments were done on each dog and information regarding reproductive status was recorded. Faecal samples, blood samples and skin scrapings (where applicable) were collected. Each dog (apart from collie dogs and unweaned pups) was then treated with ivermectin (Ivomec, antiparasitic injection for cattle MSD AgVet, 10g/L; 200mg/kg subcutaneous). Mature female dogs (over 6 months of age) were given

5 proligestone injections (Covinan, Intervet (Australia) Pty Ltd, 100mg/mL; 10-30mg/kg subcutaneous) at the request of the owner for control of oestrus. These treatments and samplings were continued every three months for the duration of the program (3.5 years). The general methodology for the study is described elsewhere (Wilks, Williamson & Robertson, 1996). A telephone survey of 90 households of the urban centres of the Kimberley region was also undertaken to provide comparison data for the dog ownership and population characteristics of the Aboriginal communities of the program. The Dog Health Program Results The data collected during the pilot program provided valuable information about the dog population characteristics in communities as well as the effectiveness of the control program in reducing the parasite prevalence and reproduction in female dogs. Preliminary Survey of Dog Health The program focused on parasitic diseases and the results showed variation in prevalence of parasites across the Kimberley Region. The most common parasites found during the pre-treatment testing of dogs were Sarcoptes scabiei, Ancylostoma caninum, Giardia duodenalis, Dirofilaria immitis and ticks (see figures 2 4). Parasitic diseases were expected to be at a high level due to the tropical climate of the region and the lack of treatment of dogs for these diseases. a) Sarcoptes scabiei Sarcoptes scabiei is a skin burrowing mite capable of causing intense pruritus and resultant skin damage and alopecia in many host mammals. The species is subdivided into variants (Fain, 1978), which are mostly host specific. If close contact is maintained between host species (such as between humans and dogs), some cross-transmission of mites can occur (Emde, 1961; Beck, 1965; Newton & Gerrie, 1966; Sauer & Koch, 1967; Smith & Claypoole, 1967; Thomsett 1968; Elgart & Higdon, 1972; Charlesworth & Johnson, 1974; Agbede, 1978; Scott & Horn, 1987; Burton, 1997), despite the genetic separation between dog and human scabies (Walton et al, 1999). Chronically and severely affected dogs may become anorectic, depressed and develop weight loss and secondary bacterial pyoderma (Scott & Horn, 1987). Sarcoptic mange is also considered a major cause of mortality amongst natural red wolf (Philips & Scheck, 1991) and coyote populations (Pence et al, 1983). The percentage of dogs with lesions indicative of scabies in the Kimberley region varied from 14% [95% CI: 7, 21] to 52% [CI: 42, 63]. Dogs with more than 75% of hair lost were evident at all locations except the coastal region prior to treatment. Figure 2: Pre-treatment Prevalence of Parasites in Dogs from the Kimberley Region Coastal Region SKIN PARASITES n=123 Sarcoptes scabiei Rhipicephalus sanguineus GASTROINTESTINAL PARASITES n=50 Ancylostoma caninum Giardia duodenalis Sarcocystis Spirocerca Toxocara Hymenolepis spp. lupi canis nana Strongyloides stercoralis Spirometra Hammondia erinacei heydorni Isospora ohioensis Isospora canis

6 Figure 3: Pre-treatment Prevalence of Parasites in Dogs from the Kimberley Region Central Region SKIN PARASITES n=154 Sarcoptes scabiei Rhipicephalus sanguineus GASTROINTESTINAL PARASITES n=83 Ancylostoma caninum Giardia duodenalis Sarcocystis Spirocerca Toxocara Hymenolepis spp. lupi canis nana Strongyloides stercoralis Spirometra Hammondia erinacei heydorni Isospora ohioensis Isospora canis BLOOD PARASITES n=123 Dirofilaria immitis Percentage of Dogs Infected Figure 4: Pre-treatment Prevalence of Parasites in Dogs from the Kimberley Region Eastern Region SKIN PARASITES n=127 Sarcoptes scabiei Rhipicephalus sanguineus GASTROINTESTINAL PARASITES n=68 Ancylostoma caninum Giardia duodenalis Sarcocystis Spirocerca Toxocara Hymenolepis spp. lupi canis nana Strongyloides stercoralis Spirometra Hammondia erinacei heydorni Isospora ohioensis Isospora canis BLOOD PARASITES n=102 Dirofilaria immitis Percentage of Dogs Infected A small report of the effects of ivermectin treatments in four communities of Northern Queensland found scabies infestation in 21 out of 39 (54%) dogs (as determined by clinical presentation) (Presson, Palmer & Leach, 1989). Another report from Palmer and Presson (1990) describes signs of mite infestation in 42% (n=106) of dogs from selected communities in the Northern Territory. No other data from non-aboriginal communities of Australia exist for comparison, although the average prevalence of infection in Britain is considered to be 1% (Baxter & Leck, 1984). b) Hookworm Hookworm disease can be a severe anaemic disease for young puppies and dogs on a low plane of nutrition due to the blood sucking activities of the worm (Bowman, 1992). Ancylostoma caninum and Ancylostoma braziliense are also responsible for cutaneous larva migrans in humans who come in contact with the infective larvae (L3) in the ground (Robinson, Thompson & Lindo, 1989). The result is an intensely pruritic skin lesion due to the immunological response of the host. In Queensland Ancylostoma caninum has also been found to be the etiological agent for eosinophilic enteritis (Prociv & Croese, 1990), which may cause severe abdominal cramps prompting surgical

7 intervention (Croese, Loukas, Opdebeeck & Prociv, 1994). Both of these human manifestations of the disease have not been reported in Aboriginal communities. Hookworm (Ancylostoma caninum) was the most common parasite infecting dogs from each community in the Kimberley region. Overall, the average prevalence of hookworm was 65% [CI: 61, 74]. Comparison of the present study prevalence of hookworm with other locations is difficult due to the differences in sampling and diagnostic testing. Results of similar studies in Brisbane, Queensland, have shown that the prevalence of hookworm infection ranges from 17% in summer sampling of pet animals (Prociv et al, 1994) to 87% of stray dogs over one year of age (Boreham & Capon, 1982). c) Heartworm Dirofilaria immitis infection in dogs in tropical climates is very common and of considerable clinical significance for infected dogs (Dunsmore & Shaw, 1990). Dirofilaria immitis is also a rare zoonosis (Dunsmore & Shaw, 1990) although the exposure rate of Aboriginal people to D. immitis is almost 10 times higher than in Caucasians (Jenkins & Andrew, 1993). The highest prevalence was found in the eastern region at a remote inland community in semi-arid country (65% [CI: 48.2, 82.8]), and the lowest prevalence in the central region. The true prevalence of heartworm could not be determined from this survey, as adequate numbers of dogs for post mortem examination were not obtained. Heartworm is most prevalent at the coastal regions of Australia, but is also common in the tablelands of Queensland. The prevalence of heartworm in north-eastern coastal Australia, determined by circulating microfilariae, varies from 37% (Carlisle, 1969) to 68% (Aubrey & Copeman, 1972). In Brisbane, 36% of dogs were found with microfilariae, with a further 12 % having occult infections (Welch, Dobson & Freeman, 1979). These results are similar to those found in this study (10% to 65%) and reflect the tropical climate and abundance of mosquitoes. Reasons for the slightly lower prevalence in the Kimberley relate to the lower numbers of dogs in fairly remote and isolated communities. Most of the infected dogs were born within the communities and did not travel to other areas indicating that heartworm was endemic, but at a low level in some areas (Fitzroy Crossing). d) Ticks Rhipicephalus sanguineus ticks (brown dog ticks) are common canine parasites of the Kimberley region (prevalence varied from 19% [12, 26] to 24% [16, 31]) and are considered by many owners to be repugnant (Howe, 1993). Although Rhipicephalus sanguineus ticks are not of zoonotic importance, they are capable of causing dermatitis and anaemia in heavy infestation in dogs and are also an intermediate host for Babesia canis (not yet detected in the Kimberley region) (Dunsmore & Shaw, 1990). e) Giardia The main features of infection with Giardia in humans and animals is diarrhoea and malabsorption, although some animals (and humans) may remain symptomless (Casemore, 1990). In Australia, Giardia is considered to be the most important parasitic disease in terms of human morbidity and is one of the most common causes of diarrhoea in children with a failure to thrive (Swan & Thompson, 1986). Much controversy has arisen over the zoonotic nature of Giardia. Recent studies using molecular techniques have shown similarities of strains of Giardia isolated from humans and animals (Meloni, Lymbery & Thompson, 1989: Thompson, Lymbery & Meloni, 1990), although comparison of SSU-rRNA sequences of Giardia from dogs and people residing at the same location

8 in the Kimberley Region have revealed differences (Hopkins, Meloni, Groth, Wetherall, Reynoldson & Thompson, 1997). This latest work suggests that dogs are not a significant reservoir of Giardia infection for humans In the Kimberley region, 15% [CI: 10, 20] of dogs were found to be shedding Giardia cysts (pooled data), which is similar to findings by Thompson et al in 1993, who examined dogs from inland Kimberley (central) region communities (16% of 282 dogs). Comparatively, other studies have found the prevalence amongst dogs from a variety of backgrounds in Perth, to be 21% (n=333) (Swan & Thompson, 1986), with even higher rates amongst dogs from refuge shelters (30%, n=30) (Swan & Thompson, 1986). These rates are higher than those of the Kimberley communities with the risk of infection in the general canine population being 1.8 times more likely in Swan and Thompson s (1986) study than the present (Chi squared test, P=0.0001). f) Other diseases of dogs in Aboriginal communities Other parasitic agents that were found during the initial survey in the Kimberley region included Toxocara canis, Strongyloides stercoralis, Hymenolepis nana, but at low prevalences that were unlikely to cause any concern to dog health. Toxocara canis is mostly found in dogs under one year of age (Seah, Hucal & Law, 1975; Visco, Corwin & Selby, 1977; Kirkpatrick, 1988) and is capable of causing visceral larva migrans in humans (particularly children) who have accidentally ingested infective eggs from the environment (Seah, Hucal & Law, 1975; Gridwood, 1986). The eggs take several weeks for maturation, but are capable of surviving for several years (Glickman & Schantz, 1981). Symptoms in humans vary according to the organs that are invaded by the larvae. Control of this parasite is important due to the potential zoonotic effects. Lice (Trichodectes spp.) and fleas (Ctenocephalides spp.) were also found on dogs to varying degrees depending on the location within the Kimberley region. Both of these parasites are capable of causing disease in dogs, although their effects in the present study were probably masked by the skin lesions caused by Sarcoptes scabiei. Another study found Trichuris vulpis, Dipylidium caninum and Echinococcus granulosus to infect dogs of Aboriginal communities in New South Wales (Jenkins, Meek, Ardler & Hawksby, 1993). Trichuris vulpis in large burdens can cause typhlitis and anaemia in infected dogs (Dunsmore & Shaw, 1990) and has been reported to infect humans (Kenney & Yermakov, 1980). Dipylidium caninum, also a zoonosis, is of minor importance to both humans and dogs (Dunsmore & Shaw, 1990). Echinococcus granulosus, on the other hand, is a serious threat to human health. E. granulosus is responsible for hydatid disease in humans, which manifests from accidental ingestion of the eggs of the tapeworm. Dogs shed the eggs after infection from ingesting the viscera of infected intermediate hosts, usually sheep or macropods (Dunsmore & Shaw, 1990). Human hydatidosis may be life-threatening due to the development of cysts in vital organs. Incidences of 6.9 cases and 1.4 cases per 100,000 Aboriginal people have been reported in Western Australia which are considerably higher than the levels found in Caucasians (Stein & Mc Cully, 1970; Holman & Hicks, 1981). During the initial survey for the Dog Health Program project, faecal samples were tested for the presence of Campylobacter and Salmonella bacteria. Both Campylobacter and Salmonella are transmitted faecal-orally and can cause enteritis in humans. The role of these bacteria as canine zoonoses is considered minimal (Prescott & Munroe, 1982; Willard, Sugarman & Walker; 1987), although cases of people developing limited diarrhoeal disease after handling diseased puppies have been reported (Blaser, Cravens, Powers & Wang, 1978). The prevalence of Campylobacter ranged from 3% [0, 8] (coastal communities) to 7% [1, 13] (central region). Salmonella isolates were more common, with the prevalence ranging from 3% [0, 8] to 65% [50, 80]. These results are within the

9 ranges found in other Australian studies of normal and clinically affected dogs (McOrist & Browning, 1982; Malik & Love, 1989). Dog Population Characteristics in Communities a) Dog population structure The number of dogs in each community varied from nine to 155 across the Kimberley region and was dependent on the community size (number of people and households). Overall the population of dogs was young (estimated average age two years old) and the proportion of puppies in communities was higher than found during a survey of the urban centres of the Kimberley (with a largely non-aboriginal population). Overall the urban centres were 2.7 (CI: 1.5, 4.8) times more likely to have adult (greater than one year old) dogs than the Aboriginal communities. A young dog population has implications for zoonoses control as many of the zoonotic parasites (such as Toxocara canis and Ancylostoma caninum) were often found in younger animals. The sex distribution was skewed towards male dogs (59% males and 41% females). Bias toward male dog ownership has occurred in other populations such as those of Yolo County in California (56%) and South Africa (54%), but not to such an extent as in the present study. Explanations for male dog popularity in these overseas reports have not been provided, but one possible reason for the bias in the Kimberley is the selective killing of female newborn puppies to help control dog population numbers (personal observation; Howe, 1993). It is also often commented that males are more favourable than females for protection and hunting. Sex ratios for litters of wild dingoes in central Australia are biased toward male offspring (1.1 males to 1 female) as are domestic dogs, as determined from a survey in Alice Springs (1.7:1) (Corbett, 1995). The average ratio of the present study (1.44:1) is similar to these figures, which may indicate that the ratios are a natural phenomenon rather than a result of human intervention. Sterilisation of dogs by surgery was rare for the dogs of the Kimberley communities. The urban townships had dog sterilisation rates (33.3% of females and 66.3% of males) similar to those of Perth (33.8% of females and 69.4% of males) (Robertson et al, 1990). This increased rate compared with the communities may, among other factors, indicate the greater availability of veterinary services, a higher disposable income and increased education on dog population control in the urban centres. b) Dog population distribution The distribution of dogs within communities varied across the Kimberley and may have reflected community history and acceptance of dogs. Approximately 80% of houses in the eastern region of the Kimberley had at least one dog present, compared with the other regions that had rates (approximately 50%) that were similar to the urban centres of the Kimberley. Likewise, the average number of dogs per owner and household were greater in the eastern region than the coastal and central areas of the Kimberley region. The number of dogs per household and dog-owning-household were quite low in the coastal region and were almost in accordance with other non-aboriginal community studies (0.42 to 0.74 dogs per household and 1.24 to 1.7 dogs per dog-owning-household) (Franti & Kraus, 1974; Schneider & Vaida, 1975; Griffiths & Brenner, 1976; Nassar & Mosier, 1980; Nassar, Mosier & Williams, 1984; Nassar & Mosier, 1984; Robertson, Edwards, Shaw & Clark, 1990; Ticman & Carlos, 1992; Odendaal, 1994; Leslie, Meek, Kawah & McKeon, 1994; Patronek, Beck & Glickman, 1997). The communities of the coastal region are of mission background and have had a relatively long period of European contact, which may have altered previous canine ownership patterns. These communities are also sea-based and may not have had the necessity for dingo assistance in hunting

10 in the past, compared with desert communities. Also, it appears that many dogs at these communities have companionship roles, as small dogs, such as Chihuahuas, are popular. In contrast, the eastern (desert) communities of the study have had less European contact and may have more traditional views to dog ownership. The number of dogs per owner for the eastern region (2.7), was much lower than the number of dogs per household (4.1), suggesting that there were several dog-owners living at the same household. In most non-aboriginal communities, the number of dogs per owner and dogs per household are much the same, as only one person per house takes the role of owner. In the Kimberley communities, many owners may live at the same house due to the crowded living conditions. (The average number of Aboriginal people per private community dwelling in 1991 was 7.7 compared with 2.6 per dwelling for non-aboriginal people [Census, 1993].) This has implications for animal management programs, as workers requiring owner cooperation must ensure they address all owners at each household. Overall the ratio of people to dogs was found to be an average of 4.9 to 1, which is within the range found in many other studies in non-aboriginal societies (range of 3.9:1 to 7.3:1) (Franti & Kraus, 1974; Schneider & Vaida, 1975; Griffiths & Brenner, 1976; Nassar & Mosier, 1980; Nassar, Mosier & Williams, 1984; Nassar & Mosier, 1984; Robertson, Edwards, Shaw & Clark, 1990; Ticman & Carlos, 1992; Odendaal, 1994; Leslie, Meek, Kawah & McKeon, 1994; Patronek, Beck & Glickman, 1997). Since the number of dogs per household and owner was high, but the number of dogs per person was not, this indicates that the perceived dog overpopulation may not be due to too many dogs, but not enough households. The same number of people, owners and dogs located in another community with more houses could result in figures similar for non-aboriginal communities of other studies. There also was a disproportionately high number of dogs per owner for older age groups of the human population although 31% of the Aboriginal population was under the age of 15 at the 1991 census. That is, there was an overdispersion of dog ownership in the older age groups although these groups contributed less to the human population census. To outsiders, there may appear to be an overpopulation of dogs in Aboriginal communities, but this was really only the case for certain owner-age groups in the present study. c) Dog population dynamics One year after the start of the program, 47% of the original dogs had died or gone missing during that year. Consequently, over 50% of dog-owning-households had lost a dog in the same period, which was four times the rate of similar households in the urban communities of the Kimberley region. Overall, dogs that died or went missing were 1.58 times more likely to come from multipledog households than from the general population, indicating that crowding of dogs at households may have contributed to the high mortality rate, possibly from low nutrition. Very few people could provide explanations for missing or dead dogs, which is in accordance with other difficulties encountered regarding information retrieval in Aboriginal communities. Only 9.2% of the dog deaths were of a known cause when voluntary euthanasia was excluded. The causes of death ranged from drowning to snake envenomation. Many of the reported deaths were season related as 33.3% of deaths of known cause were due to drowning. The voluntary euthanasia rates during the wet seasons were higher than normal (20% of deaths) due to the poor condition of some dogs after widespread flooding across the region. Other reasons for voluntary euthanasia included overpopulation, aggressive behaviour and disease. The percentage of dog-owning households that acquired new dogs in the previous 12 months were 86.5% and 93.9% for June 1993 and June 1994, respectively. Seventy-four percent of newly acquired dogs were from communities or townships that were not part of the program

11 Although some of the dog population characteristics from some of the Kimberley Aboriginal communities were similar to those found in other studies, the rates of ownership per household and per owner were higher in Aboriginal communities. Mortality rates and rates of introduction of animals into the communities were also much higher than for non-aboriginal centres. The ownership patterns indicate that older people are more likely to own large numbers of dogs, and communities with more traditional customs are more likely to have higher numbers of dogs per capita than those with more contemporary experiences with dogs. The Effectiveness of the Parasite and Breeding Control Program The success of the parasite and breeding control program was reliant on the support and assistance provided by the communities in encouraging owners to cooperate and present dogs for treatment. Initially 99% of dogs were re-presented at the next visit, but this decreased over time to 81% at the end of the pilot program. With these capture rates, the effect of the program on some parasites was considerable. Parasite Control Ivermectin is a potent endo- and ectoparasiticide which, when used at 200mg/kg S/C in dogs, can kill a wide variety of parasites, except tapeworms. The program was mainly targeted at reducing the prevalence of scabies and hookworm in dogs a) Sarcoptes scabiei. Repeated skin appraisals of dogs every three months before treatment revealed a consistent reduction in scabies prevalence and severity across all of the communities of the pilot program. Ivermectin is very effective at resolving Sarcoptes scabiei infections in dogs when given at 200mg/kg twice at a day interval (Thimmappa Rai & Yathira, 1988). Despite the treatments in the present study only being given every three months, there was an eventual reduction in prevalence of scabies to a level below 5% at all communities. Resolution of scabies-induced skin lesions may also take at least 30 days (Paradis, de Jaham & Pagé, 1997). It is likely that much longer than this is required if secondary skin infections and extensive hyperkeratotic lesions are evident initially, as was the case for approximately 10% of pretreatment dogs (score 4) in the present study. Treated dogs were also exposed to untreated dogs that may have had scabies, which would have made re-infection possible in the present study. Figure 5: Percentage of Dogs from the Eastern Region with Scabies PERCENTAGE OF DOGS WITH MANGE st Treatment Mar- 92 Jun- 92 Sep- 92 Dec- 92 Mar- 93 Jun- 93 Sep- 93 Dec- 93 Mar- 94 Jun- 94 Sep- 94 Score 4 Score 3 Score 2 Score 1 MONTH AND YEAR OF VISIT

12 b) Hookworm. The reduction in prevalence of hookworm followed a seasonal pattern throughout the pilot study that varied according to the region within the Kimberley. Being a typical geohelminth, hookworm is reliant on environmental factors (such a humidity and temperature) for survival. Prevalence was lowest during the dry seasons, but increased during the wet seasons. It is likely that although small improvements in reduction in prevalence were noted, there would have been even more rapid reductions in the intensity of infection in the dogs. By reducing the intensity of infection (number of worms per host), the likelihood of hookworm disease in the dogs would have also decreased and contributed to the overall improvement in condition. c) Other Parasites. In one community, weekly samplings were conducted to determine the immediate effects of treatment on parasite levels. During the samplings, the prevalence and number of ticks on dogs had reduced from 53% (n=73) to 17% (n=69) within one week of treatment. Unfortunately the prevalence of ticks had increased to 47% (n=51) by five weeks after the treatment due to the availability of larval ticks from the dogs environment. Heartworm larval stages (L3 and L4) are extremely sensitive to ivermectin and this forms the basis of monthly prophylactic treatments for heartworm. During the Dog Health Program, ivermectin was only given at three monthly intervals, which was considered to be too sparse to ensure that the larvae did not mature. Ivermectin is potent against circulating microfilariae at doses higher than 50 mg/kg (Campbell & Benz, 1984), which explained a reduction in the prevalence of microfilariae positive dogs after several treatments. The overall effect of the program may be a reduction in the reservoir of microfilariaemic dogs and hence a reduction in heartworm prevalence over time. McCall, McTier, Ryan, Gross and Soll (1996) conducted a trial to assess the efficacy of ivermectin against heartworm of three and four months duration. Commencement of one monthly heartworm prophylaxis (ivermectin PO 6 mg/kg) in dogs that had established infections of three months duration resulted in 97.7% less total worm burden than in the controls. Although the ivermectin was continued every month for one year, the results do indicate that ivermectin at three monthly intervals may be of some effect in preventing further infection. It is still unknown exactly at which age D. immitis is completely refractory to prophylaxis (McCall et al, 1996). Breeding control Proligestone (Covinan) is recommended to be given three months after the first treatment, then four months later and then continued every five months thereafter for permanent control of oestrus in dogs. During the Dog Health Program, treatments were given every three months to ensure that any bitches that missed one treatment were only one month overdue for their treatment, rather than several months. The treatment regime did not result in any observable adverse side effects. The compliance rate for proligestone treatments was high. Between 62 and 100% of bitches presented for examination were treated with the contraceptive, which extrapolated to between 62 to 82% of all mature bitches receiving proligestone at each visit. The breeding program was effective. The conception rates during the period of treatment (less than 10%) were well below the conception rates before and after the treatment period (greater than 15%). The reproductive rate for entire bitches during the program was between approximately 40% and 80% less than Perth and the Kimberley urban centres, respectively. Although some bitches did breed during the program, this was not due to ineffective contraception treatments, as 51% of these bitches had not been treated with proligestone before their pregnancy.

13 The pregnancies throughout the program were either due to bitches entering the communities already pregnant or because they missed the treatments. Despite the effectiveness of the contraceptive treatments, the average rate of change in population for every three months was This extrapolated to an annual rate of change in population of 0.94 indicating a very low decline in population numbers. This overall annual rate of change of population (excluding June to September 1993) was similar to those found in Kansas and Las Vegas (0.975 and 0.96) (Nassar & Mosier, 1980; Nassar, Mosier & Williams, 1984), where there are higher reproductive rates (17 and 13%), but much lower rates of acquisition of dogs from outside the community. The dynamics of dog populations in these communities is different to those found in the present study, as they are more stable, with fewer inward migrating dogs. For adequate canine population control in the Kimberley communities, breeding control must therefore be coupled with reductions in importation of new dogs from other areas. This will bring the dynamics more in accord with centres where population control of dogs is considered to be more adequate. Conclusions of the Pilot Project The pilot project demonstrated that successful control of canine parasites and breeding is possible in Aboriginal communities if undertaken as a three monthly treatment regime with good capture rates of dogs. Although a seasonal pattern of prevalence was noted for hookworm, the rates of infection eventually declined indicating that the program needs to be conducted consistently for at least a year before dramatic results are seen. Likewise, scabies infection rates did not decline until after approximately nine months of treatment for most communities, possibly due to importation of infected dogs resulting in re-infection of treated dogs. The program also highlighted the differences between areas within the Kimberley region, especially with respect to the pattern of dog ownership and population characteristics. It is likely that the differences are related to differing historical experiences and association with dogs and the age/gender characteristics of community members. This in turn, to an extent, dictates the health status of dogs and possibly the motivation for involvement in the Dog Health Program. Overall, the effectiveness of the Dog Health Program in different communities cannot be viewed in isolation from the historical background of the people and cannot be considered uniform across all communities. 4. SUSTAINABLE DOG HEALTH PROGRAMS IN WESTERN AUSTRALIA The pilot program expanded from 14 communities to over 100 communities in the Kimberley and Pilbara regions of Western Australia. Programs in communities are now managed and operated by the community people and supervisors from Public Health Units of the Health Department of Western Australia and local shires of the two regions. The transition from pilot program to control by staff from the regions was carried out over a one-year period in Several Aboriginal Senior Environmental Health Workers * (SEHW) were interviewed to provide an update on the current workings of the Dog Health Program (DHP) in the Kimberley and Pilbara regions. The following outlines the experiences of the SEHWs in conducting the DHP. * The use of the title Senior Environmental Health Worker includes field officers and other coordinating personal who are at a senior level and have completed advanced EHW training.

14 Coordination of Dog Health Programs in the Kimberley and Pilbara Regions The DHP in the remote north-west are based around the Health Department s Public Health Units of the Kimberley and Pilbara regions. In 1994, the Pharmaceuticals Services division of WA Health (in conjunction with Murdoch University, the Veterinary Surgeon s Board of Western Australia and the Public Health Units) devised a protocol for the effective and safe use of the restricted drugs of the program: ivermectin, proligestone and pentobarbitone. Personnel involved with the DHP are required to pass a competency-based course approved by the Veterinary Division of Murdoch University (please contact author for further details regarding course). The permits for the use of the S4 drugs are held by the medical officers of the Public Health Units and reports documenting the use of the drugs are required by Pharmaceutical Services every year (see appendix). Each operator is required to keep records of the location and name of the owner (or other identifier) of all dogs treated and the dosage given. Coordination in the Kimberley region is divided into three geographical areas: Broome/West Kimberley, Derby/West Kimberley and Kununurra/East Kimberley. Most of the Kimberley is managed through SEHWs under the umbrella of the Kimberley Public Health Unit, although the Derby/West Kimberley Shire assumes responsibility for the program in the Derby region. All areas report back to the Kimberley Public Health Unit on a regular basis to satisfy the requirements of the protocol. In the Pilbara region, two SEHWs coordinate the program. In the coastal areas, three DHP-certified EHWs conduct the program under the supervision of the SEHWs. As in the Kimberley region, the SEHWs report back to the Pilbara Public Health Unit. Each of the SEHWs manage the program and in some communities are the sole administrators of the program. Within most communities, support is provided by trained EHWs, Community Development Employment Program (CDEP) workers or interested community members. Generally it is the cooperation of the owners of dogs that is considered paramount, although the support staff provide a valuable service in acting as liaison personnel and educators of community members. Technical Components of the Dog Health Program Ivermectin and proligestone are administered to dogs every three months, as was the case with the pilot program. Several delivery systems for ivermectin have been tried at both regions, but the operators still prefer subcutaneous injections. In some cases, ivermectin is administered with poultry vaccinator guns attached to backpacks; in other communities, individual syringes are drawn up. Oral administration to dogs (using meat, bread, bread and fat) has not been successful due to the unpalatable nature of ivermectin and the time required to follow-up dogs that reject the dose. Pouron solutions are to be assessed in the near future (see Paradis, 1997, for effectiveness of pour-on ivermectin against S. scabiei). Treatments are timed to be given as close before the wet season as possible to aid in natural environmental control of hookworm. Euthanasia of dogs has become an important component of the program as the operators have gained the confidence of the community members. Consent is required from the owners of all dogs to be euthanased. Record-keeping has facilitated the identity of dogs and has reduced the confusion regarding dog ownership within communities. In the Kimberley region, most SEHWs euthanase dogs by intravenous overdose of pentobarbitone. In other areas, shire rangers are called in to

15 remove unwanted dogs. The cooperation between DHP operators and shire personnel has resulted in very effective DHPs and facilitates the shire s quest to encourage dog control (particularly in the townships). Initially one of the major drawbacks to the DHP was the capture of dogs for treatment. As the program has evolved, community and owner support has improved and the SEHWs no longer consider this a problem. Each owner is required to catch their dog/s and hold them during treatment. Use of restraining ropes or muzzles is not needed for most animals. Capture of dogs using tranquilliser guns has also not been necessary although two people from the Shire of Wyndham/East Kimberley have competency certificates to use the guns. The guns are used for restraining dangerous dogs around the townships. The major drawbacks to using the guns has been the short distances over which they work, the time taken for dogs to become sedate and the loss of needles with dogs that escape. Use of the guns in communities is not well accepted because of the reaction of community members to past attempts at population control using rifles to shoot dogs. As is the case with many health programs in remote areas, one of the major drawbacks to the DHP in the Kimberley region is the rapid turnover of environmental health staff in communities and the continual need for training programs for new members. User-Pay Systems for the Program In the East Kimberley, the DHP operates on a user-pay system where the community councils agree to pay for the drugs and consumables. The system has been in place since 1995 and, according to the SEHW field officer, has been very successful. The councils budget for the DHP, and if the costs are too high, report back to the community members and encourage euthanasia of dogs that cannot be cared for. In the Pilbara, some communities have used top-up money (equivalent to rent payments by community members) to employ private veterinarians to surgically sterilise dogs. The costs of the labour and travel for the program are carried by the Public Health Unit and shires, but are considered minimal as the program is an adjunct to the work already carried out by SEHWs in each region. In fact, the DHP is considered the main key to continuing other programs and work in communities. All SEHWs interviewed believed that without the DHP, many of the environmental health problems would not be brought to their attention as quickly. The DHP is also considered very important in introducing other environmental health programs such as pest control. In the East Kimberly, the DHP has been run concomitantly with parasite control programs for children in some communities and is considered an important aspect of education about parasites. Criteria for New Communities Joining the Program The demand for the DHP has outstripped the supply in some regions resulting in more EHWs requiring training to run the program. Overall, there are very few (if any) communities in the Kimberley and Pilbara that do not have a DHP, and the only requests that the SEHWs have of communities is support in the form of liaison volunteers and the cooperation of owners in handling their dogs. The involvement of virtually all communities of the regions has resulted in a dramatic decline in dog numbers and has probably facilitated the improvement in dog health for all communities. Ownership of the Dog Health Program Each community is individual and within each community, dog owners are individuals. Likewise, community attitudes to their ownership of DHPs vary. In the Kimberley region, payment for the program is suspected to add to a feeling of community ownership of the program, although complete community ownership is not considered feasible or even necessary. In the Pilbara region