DS research

This thesis explores engagement between autistic children and donkeys during Equid Assisted Activity (EAA) sessions. I present the blurred position of EAA in Human-Animal Research that results in unreliable methodology and understanding about the equids’ perceived abilities. I argue that ‘benefits of EAA’ explored in other research is a problematic concept, because of the heterogeneous nature of autism and the individual character differences between donkeys. Using narrative analysis and narrative ethology showed that autistic children and their donkey partners demonstrate diverse and complex engagement behaviours that cannot be reduced to an entity of benefits that applies to all individuals. Qualitative stories about autistic children and donkey interactions offered a broader understanding of who each participant was, resulting in their caretakers forming new accountabilities and making informed decisions about their participants’ wellbeing.

I questioned the quality of engagement in 15 reported studies on EAA and the methodological preference of only measuring and reporting human responses. In order to measure the quality of engagement between autistic children and donkeys I designed and tested a Quality of Engagement Tool (QET) that was reliable enough to be used in a number of research designs.

The QET identified that engagement behaviour of one partner was correlated with that of the other partner in the same session. Individuals (children or donkeys) engaged differently when interacting with a conspecific as opposed to a heterospecific. The stories presented through narrative analysis and narrative ethology, coupled with the findings from the QET are important for future research. Measuring outcomes for children would be highly dependent on their 3relationship with their equid partner or indeed if they had the same partner for the duration of the research therefore; equids and humans should be considered as equal participants. The thesis concludes with a summary of findings from this project and signposts future research directions.

Eighty normal cheek teeth and 26 normal incisors extracted from 14 donkeys (median age 19 years) at post mortem were anatomically examined including grossly and by computerised axial tomography (CAT) imaging. Decalcified histology was performed on 54 sections from 18 teeth (8 donkeys), undeclacified histology on 16 sections from 7 donkeys and scanning electron microscopy on 10 sections from 10 teeth (3 donkeys). The dental formulae and tooth number was found to be the same as in horses with a higher prevalence (17 %) of canine teeth in female donkeys. A decrease in tooth length, pulp horn length and pulp horn width with age was illustrated, as was an increase in occlusal secondary dentine depth with age, although not all these age changes were statistically significant. Normal histological and ultrastructural features of donkey teeth were identified and found to be similar to equine findings. Enamel was found to be thicker buccally in both maxillary and mandibular cheek teeth. Quantitative measurements of transverse dentine thickness around pulp cavities, dentinal tubule diameters and densities, and enamel prism diameters were made. Left lower incisors (301) were extracted from 7 donkeys and 6 horses for micro-hardness determination of enamel, primary and secondary dentine using a Knoop Hardness indenter. No significant difference between donkey and horse incisor microhardness was demonstrated.

Examination of 19 donkey skulls at post mortem examination showed donkeys to have a higher degree of anisognathia (27%) compared to horses (23%). Post mortem dental examination of 349 donkeys (median age 31) demonstrated a high prevalence of dental disease (93%) and in particular cheek teeth diastemata (85%). Furthermore, age was associated with increasing prevalence of dental disease and diastemata. Diastemata were also associated with the presence of other dental disorders and with colic-related death in affected donkeys. Quantitative measurements of 45 diastemata from 16 donkeys showed no difference in the medial and lateral width of diastemata but periodontal pockets were deeper laterally. The definition of valve and open diastemata were confirmed. Pulp exposure, dental caries and periodontal disease were examined in detail (54 skulls) at post mortem. A total of 19 teeth were extracted for further detailed examination as performed in normal anatomy.

Clinical dental examinations were performed on 357 donkeys in the U.K. that were selected for age distribution, and the prevalence of dental disease in different age groups was found to increase from 28% in the youngest group (age 0-10 years) to 98% in the oldest group (age > 35 years). An increased prevalence of most dental disorders with age was demonstrated as was an association between dental disease and weight loss, poor body condition score, supplemental feeding and previous episodes of colic. Clinical dental examination of 203 working donkeys in Mexico showed similar types of dental disorders as found in the U.K. study, with dental disease present in 62%, of which 18% required urgent dental treatment. There was a significant association between age groups and dental disease, and age groups and body condition score, but there was no association between dental disease and body condition score. However, body condition score was not associated with supplemental feeding or faecal egg counts either.

The purpose of this study was to compile practical feeding guidelines for donkeys in the UK. Current guidelines are to feed 0.75 of horse feeding recommendations on a body weight basis. However, the superior digestive efficiency of donkeys, compared to horses, may render the use of horse recommendations inappropriate. The formulation of guidelines specific to donkeys would enable owners to calculate their donkey‟s requirements with greater accuracy and prevent overfeeding.

A postal survey, used to gain information on the body condition score of donkeys in the UK, and the husbandry and feeding practices used to manage them, indicated that approximately 24% of donkeys in the UK are overweight. Feeding practices indicated that although owners were aware of their donkey‟s requirement for fibrous forages, the practice of feeding unnecessary concentrates, chaffs and high energy forages, in addition to grazing, was the likely cause of donkeys becoming overweight. The finding that the majority (85 – 90%) of donkeys were kept as non-working companion animals also reduced the need for owners to feed higher energy foods to their donkeys. Results also suggested that owners were unsure of how to adjust their donkey‟s diet to account for seasonal changes in requirements and pasture availability, as most owners‟ adjusted grazing access, and not the feeding of supplementary feeds.

From a study of dry matter (DM) and digestible energy (DE) intakes by 20 mature donkeys maintaining weight during each UK season, the maintenance DE requirements of donkeys were calculated. Results showed no effect of sex on DM or DE intake. Season significantly (P<0.001) affected DM and DE intakes, implying increased requirements in winter compared to spring, summer and autumn. Dry matter intakes (DMI) increased from 51g/kg BW0.75 in spring, summer and autumn to 66g/kg BW0.75 in winter. Digestible energy requirements increased from 0.32MJ/kg BW0.75 in spring, summer and autumn to 0.43MJ/kg BW0.75 in winter. Comparison of results with horse recommendations showed considerably reduced requirements by donkeys. Horse recommendations overestimated DE requirements in summer and winter by 82 and 30%, respectively, making horse recommendations unsuitable for calculating donkey energy requirements.

Husbandry practices commonly used by owners to manage their donkeys grazing access (grazing time, grazing area, strip grazing), were assessed for their effect on DMI by grazing donkeys in summer and autumn, using n-alkanes. The effect of grazing time was assessed by restricting donkeys to 8, 12 or 23 hours grazing per day. Season significantly affected food intake with donkeys in the 8 and 23 hour grazing groups eating more during summer when pasture availability was greater. Donkeys responded to the poorer quality summer pasture by grazing more intensively but less selectively, increasing the rate at which food was consumed. Grazing time was only influential over grass intake in summer, when pasture was more abundant. Restricting donkeys to 12 hours or less grazing per day significantly (P<0.001) reduced their grass intake compared to that of donkeys with 23 hours access. When grazing sparse pastures (autumn), grazing time did not influence grass intake, indicating an effect of herbage mass on grazing behaviour. Herbage mass was the most influential factor over diet composition (percentage of grass and straw consumed) in a second grazing study assessing the affect of strip grazing and set stocking systems on intake by grazing donkeys during summer and autumn. Herbage mass per donkey was higher in the set stocking system during both seasons, resulting in higher grass intakes. Determining if either grazing system was more effective at regulating grass intake was prevented due to differences in pasture availability between study sites.

It is concluded that donkeys have lower DMI and maintenance DE requirements than horses, requiring donkey feeding guidelines to be formulated. Excess body weight in donkeys is caused in part, by the feeding of energy dense feeds in addition to low energy forages. Most owners place little nutritional importance on pasture, despite its potential to provide a large percentage of daily DM, DE and nutrient intake. Therefore nutritional guidelines must include advice on how to manage access to grazing, and how to feed donkeys with access to pasture. Restricting grazing time to 8 hours a day did reduce grass intake by donkeys, but was only effective when grazing abundant pastures. Providing ad libitum straw to grazing donkeys allows them to satisfy their DM and dietary fibre requirements without consuming excess energy.

The Donkey Sanctuary Devon has, like many equine units, faced problems with anthelmintic resistance within the cyathostomin parasite population. Cyathostomins are ubiquitous, and although they are to some extent tolerated by their hosts they are also linked to potentially serious disease processes, making their management an important part of equine husbandry. Over use of the available anthelmintics has been blamed for the current situation, with cases of drug resistance reported across all the classes of anthelmintic currently licensed for use in equidae. Alternative methods of controlling cyathostomins include pasture hygiene by removal of faeces, which theoretically reduces re-infection by removing the immature larval stages from the grazing area, and targeted selective treatment (TST), which works by treating those animals shedding the highest number of eggs and leaving other animals untreated. The aim of this thesis is to evaluate these methods of parasite control at The Donkey Sanctuary Devon over an entire grazing season.

The most widely used method of monitoring parasitism is the Faecal Worm Egg Count (FWEC), which is known to be highly variable and can be difficult to analyse appropriately using traditional statistical methods. Therefore, computationally intensive Bayesian Markov chain Monte Carlo (MCMC) methods were employed, to ensure that the conclusions made are robust. A faecal egg count reduction test (FECRT) was also used to evaluate anthelmintic drug efficacy at the end of the study. There was a study population of 667 donkeys, divided into 15 groups under four different management strategies across four farms of The Donkey Sanctuary Devon. Ninety pasture larval counts and nearly 3000 FWEC were conducted from May 2010 to November 2010, followed by a FECRT on a proportion of the study population.

The principal conclusion was that twice-weekly removal of faeces from pasture, combined with a relatively high TST threshold of 2000 EPG, provides control of cyathostominosis in the donkey population studied. Manual removal of faeces (rather than using an automated sweeper) was also found to significantly reduce the requirement for anthelmintic doses when using a treatment threshold of 2000 EPG. Where faeces removal from pasture is not practical, lower TST thresholds provide greater control of cyathostomin transmission than higher thresholds, at the cost of more frequent dosing. The groups with higher dosing rates showed reduced drug efficacy at the end of the grazing season, highlighting the necessity to reduce reliance on anthelmintic use.