Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. The veterinarian’s role in end-of-life management of animals: An exploration of veterinary training and the perspectives of New Zealand cat owners A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Veterinary Science at Massey University, Manawatū, New Zealand Katherine Littlewood 2021 ii Abstract Domestic cats are living longer and more of them are living with chronic conditions. It is up to the owner(s) to make the decision about when, and how, to end the life of their cat. Owners may find such end-of-life decisions difficult, as they are influenced by many factors relating to their cat and to themselves. Owners often involve their veterinarian in the decision-making process; therefore, it is essential that veterinarians understand the role they play in end-of-life management of animals. To ensure veterinarians are effectively supporting owners to make end-of-life decisions, there is also a need to know how veterinary students are taught relevant topics and skills. The central question this research aimed to answer was: 'what is the veterinarian's role in end-of-life management of older and chronically ill cats in New Zealand?' Study 1 investigated how aspects of end-of-life management – technical euthanasia skills, end-of-life decision-making, and grief management – were taught to Australasian veterinary students. Study 2 then explored the role veterinarians play in end-of-life management from the perspective of owners of older and chronically ill cats, as well as how those owners made the decision to end their cat’s life. Study 1 demonstrated some gaps that, if filled, could improve veterinary training in end-of-life management of animals. Technical aspects of euthanasia were not taught consistently for companion animals and this needs to be improved to ensure new graduates meet client expectations – that is, they are as competent as my cat owner participants assumed. There also appeared to be gaps in teaching iii end-of-life decision-making relative to what was important to cat owners. Owners expected their veterinarian to be the professional or ‘expert’ when it came to knowledge of animal health and welfare, but not all veterinary students were taught how to assess animal welfare or quality of life in the context of end-of-life decisions. In contrast, teaching of grief-related topics left only a few gaps to fill. Grief management teaching best reflected many of the features my cat owner participants wanted from their veterinarian, and particularly the human-centred themes taught to students. Most of this grief management teaching was performed by student counsellors and psychologists in earlier (preclinical) years. This means these human-centred themes may not have been explicitly linked to the decision- making process, and, more importantly, to the veterinarian’s role in end-of-life management. Without explicit alignment, veterinary students may be left thinking that veterinarians have only a limited role, or even no role at all, in managing their clients’ emotions and that this should be left to trained professionals. However, my cat owner participants emphasised the important role their veterinarian had played in the end-of-life process, suggesting that training, in New Zealand at least, is effective in this regard. Significant methodological developments in this research include designing and conducting in-depth interviews using social science methodology, in addition to the analysis and interpretation of qualitative data. Future studies should compare owner and veterinary perceptions of the same euthanasia event to obtain a detailed picture of the veterinarian’s role in end-of-life management of animals. iv Acknowledgements Firstly, I would like to sincerely thank my supervisors: Associate Professor Ngaio Beausoleil, Professor Kevin Stafford, and Professor Chris Stephens. I could not have asked for a better group of people to support and guide me throughout this process. Our discussions have always been enjoyable, enlightening, and (often!) off-topic. They have provided me many opportunities that I could never repay them for. But most of all, their continued belief in my abilities has been unwavering, and often, necessary! I particularly want to thank my main supervisor, Ngaio. Her embodiment of a strong woman in leadership and academia is something I aspire to. I will never forget this wonderful quote that she presented me with one day (when I needed it most!) and which has stayed with me: “Strong women don’t have attitudes, they have standards” (Author Unknown) From the bottom of my heart, thank you Ngaio. My cheerful Irish supervisor also deserves special mention. I spent many long hours in Kevin’s office waxing lyrical about a variety of wonderful topics. He was always on hand to remind me of my place in the world and that it’s never that serious ‘gal’. Of course, thanks to Chris for introducing me to social science research. It was hard to change a traditional scientist into a social science researcher, but I hope I have done you proud. v Also, thanks to the other members of the Animal Welfare Science and Bioethics Centre (AWSBC), particularly David Mellor, Craig Johnson, and Nikki Kells, as well as Sophia Holdsworth, Morgan Heslop, Heidi Lehmann (and her dogs!), and Izzy Norris. Their inspiration and support have made the process much less arduous and (very often!) enjoyable. Others who deserve mention for helping me when I needed it most, and for bringing a little light into our tiny windowless postgraduate office, include Adrienne French, Megan Scholtens, Hilary Webb, Rebecca Lucas- Roxburgh, and Sara Azarpeykan. I would also like to thank a large number of people and groups for their invaluable assistance during the completion of this PhD. At Massey University: the anonymous veterinary educators who agreed to be interviewed for my first study, and Liz Norman for her veterinary social science research advice. For administrative support and being a general doer-of-all-things, Debbie Hill deserves significant acknowledgement. I do not know how she does so much for everyone and yet is still one of the loveliest people I have ever had the pleasure of knowing. And at universities across Australia, thanks to my representatives: Teresa Collins (Murdoch University), Susan Hazel (The University of Adelaide), Janice Lloyd (James Cook University), Catherine Mallia (Charles Sturt University), Leonie Richards (The University of Melbourne), Nicole Wedler (Murdoch University), Anne Quain (The University of Sydney), and Sarah Zito, for their time and efforts interviewing veterinary educators at their respective Universities, and thanks to the Australia veterinary educator participants. Also, thanks to the New Zealand vi Veterinary Association and Helen Beattie as well as The Veterinary Council of New Zealand and Seton Butler, for their support and for sharing details of my studies to the veterinary community. And a big thanks to New Zealand cat owners for their participation and enthusiasm for my research. I hope I have done your words justice. During my PhD, I was supported financially by a Doctoral Scholarship and the Muriel Caddie Scholarship in Veterinary Science (2016 and 2018) at Massey University, for which I am truly grateful. This PhD research was also made possible by funding from The Waltham Foundation. Travel to scientific conferences, to present the results of this research, was made possible by grants from Nestlé Purina, Massey University School of Veterinary Science (formerly the Institute of Veterinary, Animal, and Biomedical Sciences), and the AWSBC. I would like to thank my mum, dad, and furry family for cheering me on. Thanks also to my incredible daughter, Evie, who arrived part-way through my doctoral journey. We have had some challenging times navigating this together, but I am so lucky to be able to call myself her mum. She has made me laugh, and cry, but most importantly, she has taught me to focus on what matters most. And finally, this research would not have been possible without the unwavering support of my husband, David. No words can adequately describe how much you mean to me. You are my favourite. vii Dedication This thesis is dedicated to the many animals I have had the pleasure of interacting with during my life… …with a special dedication to Pearl. She was my first cat and the best cat. She taught me how special cats are and how hard it is to say goodbye. She made me want to become a veterinarian. But most of all, she was my friend. To the companion animal veterinarians out there, this thesis is also for you. You have one of the toughest jobs in the world, but also one of the most important. And to everyone who interacts with animals: “Do the best you can until you know better. Then when you know better, do better” (Maya Angelou) viii Publications Publications related to thesis research Littlewood, K. E., Beausoleil, N. J., Stafford, K. J., Stephens, C., Collins, T., Fawcett, A., Hazel, S., Lloyd, J. K. F., Mallia, C., Richards, L., Wedler, N. K., & Zito, S. (2018). Exploring how end-of-life management is taught to Australasian veterinary students. Part 1: technical euthanasia. Veterinary Record, 183(691), 1-10. https://doi.org/10.1136/vr.104775 (Chapter 4) Littlewood, K. E., Beausoleil, N. J., Stafford, K. J., Stephens, C., Collins, T., Quain, A., Hazel, S., Lloyd, J. K. F., Mallia, C., Richards, L., Wedler, N. K., & Zito, S. (2021). How decision-making about euthanasia for animals is taught to Australasian veterinary students. Australian Veterinary Journal, 1-10. https://doi.org/10.1111/avj.13077 (Chapter 5) Littlewood, K. E., Beausoleil, N. J., Stafford, K. J., Stephens, C., Collins, T., Fawcett, A., Hazel, S., Lloyd, J. K. F., Mallia, C., Richards, L., Wedler, N. K., & Zito, S. (2020). How management of grief associated with ending the life of an animal is taught to Australasian veterinary students. Australian Veterinary Journal, 98(8), 356-363. https://doi.org/10.1111/avj.12960 (Chapter 6) Littlewood, K. E., Beausoleil, N. J., Stafford, K. J., & Stephens, C. (2021). “What would you do?”: How cat owners make end-of-life decisions and implications for veterinary-client interactions. Animals, 11(4), 1-20. https://doi.org/10.3390/ani11041114 (Chapter 7) https://doi.org/10.1136/vr.104775 https://doi.org/10.1111/avj.13077 https://doi.org/10.1111/avj.12960 https://doi.org/10.3390/ani11041114 ix Other publications completed in parallel with thesis research Gates, M.C., Littlewood, K., Kongara, K., Odom, T.F., & Sawicki R.K. (2020) Guidelines for implementing a low-cost volunteer desexing skills training programme for veterinary and veterinary technology students. Journal of Veterinary Medical Education, 47(1), 27-38. https://doi.org/10.3138/jvme.0418-047r1 Gates, M., Littlewood, K., Kongara, K., Odom, T., & Sawicki, R. (2020). Cross- sectional survey of surgical techniques used to perform dog and cat spays in New Zealand veterinary practice. New Zealand Veterinary Journal, 68(1), 46-53. https://doi.org/10.1080/00480169.2019.1665594 Gates, M., Littlewood, K., Kongara, K., Odom, T., & Sawicki, R. (2020). Cross- sectional survey of anaesthesia and analgesia protocols used to perform routine canine and feline ovariohysterectomies. Veterinary Anaesthesia and Analgesia, 47, 38-46. https://doi.org/https://doi.org/10.1016/j.vaa.2019.06.008 Gates, M., Littlewood, K., Kongara, K., Odom, T., & Sawicki, R. (2020). Experience of practicing veterinarians with supervising final-year students and new graduates in performing desexing surgeries. Journal of Veterinary Medical Education, 47(4), 465-474. https://doi.org/10.3138/jvme.0918-100r Kells, N. J., Beausoleil, N. J., Godfrey, A. J. R., Littlewood, K. E., Ward, R. N., & Johnson, C. B. (2020). Effect of analgesic strategies on pain behaviour associated with combined ring castration and hot iron tail docking in https://doi.org/10.3138/jvme.0418-047r1 https://doi.org/10.1080/00480169.2019.1665594 https://doi.org/https:/doi.org/10.1016/j.vaa.2019.06.008 https://doi.org/10.3138/jvme.0918-100r x Merino lambs. Applied Animal Behaviour Science, 222, 1-9. https://doi.org/10.1016/j.applanim.2019.104914 Mellor, D. J., Beausoleil, N. J., Littlewood, K. E., McLean, A. N., McGreevy, P. D., Jones, B., & Wilkins, C. (2020). The 2020 five domains model: Including human–animal interactions in assessments of animal welfare. Animals, 10(10), 1-24. https://doi.org/10.3390/ani10101870 https://doi.org/10.1016/j.applanim.2019.104914 https://doi.org/10.3390/ani10101870 xi Table of Contents Abstract ii Acknowledgements iv Dedication vii Publications viii List of tables xv List of figures xvii CHAPTER 1: Introduction 1 1.1. Thesis aims and specific research questions 3 1.2. References 6 CHAPTER 2: Literature review 7 2.1. Older and chronically ill cats in New Zealand 7 2.2. Making the decision to euthanase a companion cat 11 2.3. The veterinarian’s role in managing the end of animals’ lives 27 2.4. References 58 CHAPTER 3: Methodology 75 3.1. Rationale for research approach 76 3.2. Issues of trustworthiness and rigour 82 3.3. My background 83 3.4. Conclusion 87 3.5. References 89 CHAPTER 4: Technical euthanasia skills teaching 91 xii Exploring how end-of-life management is taught to Australasian veterinary students. Part 1: technical euthanasia 91 4.1. Abstract 92 4.2. Introduction 92 4.3. Materials and methods 97 4.4. Results 103 4.5. Discussion 114 4.6. References 124 CHAPTER 5: Decision-making teaching 129 How decision-making about euthanasia for animals is taught to Australasian veterinary students 129 5.1. Abstract 130 5.2. Introduction 131 5.3. Materials and Methods 136 5.4. Results 141 5.5. Discussion 157 5.6. References 163 CHAPTER 6: Grief management teaching 169 How management of grief associated with ending the life of an animal is taught to Australasian veterinary students 169 6.1. Abstract 170 6.2. Introduction 171 6.3. Materials and Methods 174 xiii 6.4. Results 178 6.5. Discussion 186 6.6. References 195 CHAPTER 7: Cat owner perspectives 200 “What would you do?”: How cat owners make end-of-life decisions and implications for veterinary- client interactions 200 7.1. Simple summary 201 7.2. Abstract 202 7.3. Introduction 203 7.4. Materials and Methods 208 7.5. Results 212 7.6. Discussion 231 7.7. Conclusions 242 7.8. References 243 CHAPTER 8: General discussion 251 8.1. Major findings 251 8.2. Integrative findings 254 8.3. Overall role of veterinarians in EoL management of older and chronically ill cats and relationship to reported teaching 269 8.4. Methodological developments and considerations 275 8.5. Future research 287 8.6. Conclusions and recommendations 290 xiv 8.7. References 292 Appendix I: Chapter 3 Statement of contribution I Appendix II: Chapter 4 Statement of contribution II Appendix III: Chapter 5 Statement of contribution III Appendix IV: Chapter 6 Statement of contribution IV Appendix V: Study 1 Information sheet V Appendix VI: Study 1 Interview guide IX Appendix VII: Study 2 Information sheet XIV Appendix VIII: Study 2 Interview guide XVIII xv List of tables Table 2.1 Longevity and 'geriatric' ages for various cat breeds in the United Kingdom 11 Table 4.1 Veterinary science degree structures at eight Australasian Universities 99 Table 4.2 Demographic information for 111 participants interviewed at eight Australasian Universities 104 Table 4.3 Number (percentage) of eight representatives reporting some teaching of euthanasia techniques in pre-clinical and clinical years of the veterinary science degree at eight Australasia universities. 106 Table 4.4 Categorisation of euthanasia techniques taught by 111 participants interviewed at eight Australasian Universities. 108 Table 4.5 Categorisation of opportunity for euthanasia practise facilitated by 100 participants interviewed at eight Australasian Universities over all animal categories. 113 Table 4.6 Number (percentage) of representatives reporting students are given a chance to practise euthanasia techniques in the veterinary science degree at eight Australasia universities. 113 Table 5.1 Number of students and interview participants at eight Australasian Universities. 136 Table 5.2 Number of representatives reporting various modalities used for teaching euthanasia decision-making to students in the veterinary science degree at eight Australasian universities. 143 Table 5.3 Categorisation of euthanasia decision-making factors and indicators reported by 111 participants interviewed at eight Australasian Universities. 147 Table 5.4 Number of representatives reporting teaching factors and indicators for euthanasia decision-making to students in the veterinary science degree at eight Australasian universities. 150 Table 5.5 Whether or not students were taught a formal tool or protocol for assessing animal welfare status or quality of life in the context of making euthanasia decisions, and whether such a standard protocol or formal tool existed in the veterinary school clinic at eight Australasian universities. 154 Table 6.1 Veterinary science degree structures at eight Australasian Universities. 178 xvi Table 6.2 Categorisation of subject areas relating to grief management taught by 111 participants interviewed at eight Australasian Universities. 182 Table 7.1 Final template for owner interview themes 213 Table 8.1 Validity procedures within social science lens and related paradigm assumptions. From Creswell & Miller, (2000) 281 xvii List of figures Figure 2.1 The tripartite communication relationship between animal, veterinarian, and owner; QoL = quality of life. 30 Figure 4.1 Frequency plot showing the year since graduation for 109 of 111 participants interviewed at eight Australasian Universities as of January 2016. 100 Figure 8.1 Key research findings as they relate to the phase of a veterinary professional’s training and early career: The role of veterinarians in end-of-life management of animals with a focus on older and chronically ill cats. CA = companion animal-specific findings. 274 Figure 8.2 The findings of the current body of research have important implications for developing veterinary curricula and competency, of new graduates, as well as the welfare of the animals they treat and the wellbeing of the clients they serve. EoL = end-of-life 279 Figure 8.3 The tripartite communication relationship between animal, veterinarian, and owner; QoL = quality of life. 289 Figure 8.4 Recommendations for veterinary training and practising veterinarians from the findings of this research as they relate to the phase of a veterinary professional’s training and early career 291 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622205 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622205 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622206 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622206 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622206 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622209 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622209 file://///Users/klittlew/Dropbox/Kat%20Littlewood/A%20KAT%20@%20MASSEY/A%20KAT%20@%20MASSEY%20-%20My%20PhD/KLittlewood%20PhD%20thesis%202022-2-1.docx%23_Toc94622209 xviii CHAPTER 1: Literature review 1 CHAPTER 1: Introduction This research centres around the veterinarian’s role in end-of-life (EoL) management of animals, with a particular focus on end-of-life decision-making for domestic cats in New Zealand. Two key aspects of this topic are addressed in the research comprising this thesis: (1) teaching of topics relating to EoL management for various animal species to veterinary science students and (2) clinical practice relating to EoL care by veterinarians interacting with owners of older or chronically ill cats. The thesis begins with a brief overarching literature review chapter which situates various subtopics including quality of life and its assessment, human-animal relationships, models of veterinary-client relationships, legal and professional obligations of veterinarians, veterinary competency, technical euthanasia skills, decision-making, and grief management, within the broader context of this field of research (Chapter 2). The chapter that follows is dedicated to describing the overall methodology of the research in this thesis in relation to published literature (Chapter 3). Each of the data chapters is formatted for publication and begins with a comprehensive introduction focussed on the specific topic to be addressed. Likewise, each data chapter contains a detailed discussion of the results achieved, including the relationships of the findings to existing knowledge of the subject, CHAPTER 1: Introduction 2 limitations of interpretation based on the methodology employed, implications of the research in terms of the veterinarian’s role in EoL management, and proposals for future study. Accordingly, the general discussion (Chapter 8), is relatively brief, primarily integrating the results of the different studies to draw overall conclusions about how owners of cats manage EoL decisions, how EoL management is taught to veterinarians, and the veterinarian’s role in the process. A list of the relevant references is provided at the end of each chapter. Domestic cats are the most popular companion animal in New Zealand (Anonymous, 2020). These cats are living longer and more of them are living with chronic conditions (O'Neill et al., 2013, 2015). This may mean that cats experience less than optimal quality of life for longer. Cats are considered property in many Western jurisdictions, including New Zealand (Anonymous, 1999). Therefore, it is up to the owner to make the decision about when, and how, to end the life of their cat. Owners may find such EoL decisions difficult, as they are influenced by many factors relating to their cat and to themselves. Owners often involve their veterinarian in the decision-making process; therefore, it is essential that veterinarians understand the role they play in EoL management of animals. To ensure veterinarians are effectively supporting owners to make EoL decisions, there is also a need to know how veterinary students are taught relevant topics and skills. Research on the veterinarian’s role in EoL management of animals has primarily been conducted in the Northern Hemisphere, and on dogs. The situation CHAPTER 1: Introduction 3 may be different in New Zealand and for cat owners meaning that specific research in this context is required. Likewise, there is little research on veterinary training on topics relating to EoL management in Australasia. Accordingly, the following chapter reviews the literature relevant to EoL decision-making for owners of cats in New Zealand as well as to how EoL management is taught to veterinary students. 1.1. Thesis aims and specific research questions In attempting to balance the needs of an animal with the owner’s wishes, the veterinarian has a challenging role when involved in EoL management. This is particularly the case for older and chronically ill cats in New Zealand. Very little is known about New Zealand cats, owner considerations, and the role, or roles, of the veterinarian in managing the process of ending these animals’ lives. Given the important roles veterinarians can play in EoL management of animals, the overarching aim of research presented in this thesis is to explore how owners make the decision to end the life of their cat and the role of their veterinarian in the decision-making process. I also sought to understand how veterinarians are taught to manage the end of animals lives during their training. Specifically, the questions this thesis seeks to answer are: ➢ How do New Zealand cat owners make the decision to end the life of their cat? ➢ From the owner’s perspective, what is the veterinarian’s role in managing the end of their cat’s life? CHAPTER 1: Introduction 4 ➢ How are veterinary students taught to manage aspects of the end of lives of the animals under their care and how well does this training appear to align with what New Zealand cat owners expect of their veterinarian? Scope of research Two studies were used to answer these questions. Study 1 (Chapters 4, 5, 6) focuses on how EoL management is taught to Australasian veterinary students, while study 2 (Chapter 7) is client focused. Study 1 focuses on when, how, and by whom, subjects relating to EoL management were taught to students in Australasian veterinary schools. The aim was to describe teaching in three areas: euthanasia techniques (Chapter 4); EoL decision-making (Chapter 5); and grief management (Chapter 6). All eight Australasian veterinary schools were included, and a representative at each school asked questions of staff who teach EoL management to veterinary students using a structured interview guide. Most animal species encountered in veterinary practice were included in this study to achieve a holistic understanding of how EoL management is taught to veterinary students, particularly as some topics were expected to be generic. This was followed by a second study which retrospectively explored the perspectives of owners following the euthanasia of their companion cat (Chapter 7). Owners who had recently euthanased their cat were recruited for interview and their veterinarian was also separately interviewed about the same event. Questions sought to explore: the factors involved in EoL decision-making; how owners CHAPTER 1: Introduction 5 assessed feline quality of life (QoL); what ‘good’ QoL for a cat meant to owners; and what the veterinarian’s role was in the decision-making process. This thesis presents the results of the interviews with cat owners. The data from the parallel interviews with veterinarians was determined to be beyond the scope of this thesis. Separate publications are planned for: (1) results of veterinary interviews; and (2) a comparison of owner and veterinary decision-making relating to the same euthanasia event. Together, these two studies provided rich information about the veterinarian’s role in EoL management by examining how veterinary students were taught (Chapters 4, 5, 6) and how this role was enacted in clinical practice in the management of older and chronically ill cats (Chapter 7). Value of this research This research is unique in applying an exploratory inductive approach to investigating the roles of the veterinarian in EoL management using interviews with cat owners and with educators (Charles et al., 1997). The outcomes of this research will benefit the veterinary profession by providing insights into the perspectives of New Zealand cat owners when making EoL decisions. A better understanding of the veterinarian’s role in the process, and what is taught to veterinary students, will inform the evolution of training, and therefore competency, of veterinarians. Anticipating owners’ concerns and further understanding their needs will lead to improved outcomes and better assistance CHAPTER 1: Introduction 6 with EoL decision-making. In doing so, the veterinary-client relationship will be strengthened. Ultimately, better veterinary-client relationships will improve the welfare of older and chronically ill cats. 1.2. References Anonymous. (1999). Animal Welfare Act 1999. Ministry for Primary Industries. Anonymous. (2020). Companion animals in New Zealand 2020. Companion Animals New Zealand. Charles, C., Gafni, A., & Whelan, T. (1997). Shared decision-making in the medical encounter: What does it mean? (or it take at least two to tango). Social Science & Medicine, 44(5), 681-692. O'Neill, D. G., Church, D. B., McGreevy, P. D., Thomson, P. C., & Brodbelt, D. C. (2013). Longevity and mortality of owned dogs in England. Veterinary Journal, 198(3), 638-643. https://doi.org/10.1016/j.tvjl.2013.09.020 O'Neill, D. G., Church, D. B., McGreevy, P. D., Thomson, P. C., & Brodbelt, D. C. (2015). Longevity and mortality of cats attending primary care veterinary practices in England. Journal of Feline Medicine and Surgery, 17(2), 125-133. https://doi.org/10.1177/1098612x14536176 https://doi.org/10.1016/j.tvjl.2013.09.020 https://doi.org/10.1177/1098612x14536176 CHAPTER 2: Literature review 7 CHAPTER 2: Literature review This chapter reviews the literature relevant to end-of-life (EoL) decision-making for owners of cats in New Zealand and the potential roles of veterinarians in that process, as well as to how EoL management is taught to veterinary students. By using a narrative review style, I aim to situate various subtopics within the broader context of this field of research including: quality of life and its assessment, human- animal relationships, models of veterinary-client relationships, veterinary communication skills, legal and professional obligations of veterinarians, veterinary competency, technical euthanasia skills, decision-making, and grief management. I begin by providing the rationale for focussing my research on EoL decision-making for older and chronically ill cats in New Zealand. 2.1. Older and chronically ill cats in New Zealand My research focuses on cats for several reasons. Domestic cats, as companions or pets, have received relatively little attention in the animal welfare or veterinary literature thus far, and yet are the most popular companion animal (CA) in New Zealand (Anonymous, 2020c). The relative paucity of information on cats in the literature could be due to the limited knowledge of their behaviour and needs, or to a misguided belief that they do not have as many needs as dogs (Robertson, 2008; Rodan, 2010; Smith et al., 1996). In terms of EoL decision-making, the CHAPTER 2: Literature review 8 overwhelming focus is on assessing quality of life (QoL) and is on dogs (e.g., Lavan, 2013; Mullan & Main, 2007; Yeates et al., 2011), with few references to feline studies (e.g., Freeman et al., 2016). In addition, most of these studies pertain to the situation in North America (e.g., Freeman et al., 2016; Lavan, 2013), or the Northern Hemisphere (e.g., Mullan & Main, 2007). However, 41 % of New Zealanders share their house with a cat (Anonymous, 2020c) and cats, on average, live longer than dogs (O'Neill et al., 2013, 2015). Furthermore, New Zealanders are likely to view their CAs in a different way to their Northern Hemisphere counterparts (Kellert, 1997). The popularity of companion cats in New Zealand is probably the result of human lifestyle changes that restrict the keeping of dogs (e.g., renting properties). Therefore, there is a growing need to learn more about the life-long welfare of CA cats in New Zealand. This need includes their welfare at the end of their lives. As veterinary care has improved, many companion animals are living longer (Bonnett & Egenvall, 2010; Egenvall et al., 2009). For example, the American Veterinary Medical Association estimated that in 2011 more than 20% of cats were over 11 years of age. The Companion Animals New Zealand (2020c) report does not mention feline ages, but it is reasonable to assume a similar proportion of New Zealand-owned cats are also over this age. This, coupled with the increasing popularity of cats as companions, has led to an increase in the number of older cats CHAPTER 2: Literature review 9 requiring care in New Zealand (Anonymous, 2020c; Gardiner, 2014; Ormerod, 2008). To complicate matters, the term ‘old’, when used to describe cats, has diverse definitions. A cat over the age of 8 years may be classified as mature (Bellows et al., 2016; Vogt et al., 2010), senior (Fortney, 2012), or geriatric (Carter et al., 2014), with variation in this age threshold (e.g., 9 years: Finch et al., 2013; 10 years: Ghys et al., 2015; 15 years: Vogt et al., 2010). Others use ‘75-80% of anticipated lifespan’ as the point when a cat, or dog, becomes ‘geriatric’ (Baetge & Matthews, 2012; Dodman et al., 1984; Hughes, 2008). This percentage definition makes including the variable life expectancies of different dog breeds more practical (Hughes, 2008), but necessitates an accurate estimate of lifespan. It is important to understand what an old CA is to get a sense of when EoL decisions might become relevant to owners. For cats, some authors suggest some “well-cared for” cats can live into their early twenties (Bellows et al., 2016). A nationwide study using data obtained from veterinary practices in the United Kingdom (UK) found a median feline longevity of 14.0 years (range 0.0 - 26.7), with variation among breeds (Table 2.1). Mortality was bi-modally distributed and peaked at years 1 and 16. This suggests the presence of two subpopulations: those that suffered an early death because of trauma (typically road traffic accidents), and cats that survived to an older age, when renal disorders, non-specific illness, and neoplasia were the most common causes of mortality (O'Neill et al., 2015). It is worth noting that more CHAPTER 2: Literature review 10 than 85% of deaths in this UK study involved euthanasia. This, in addition to the bi-modally distributed population, makes interpretation of these data difficult in terms of estimating actual feline lifespan, and for classification as an ‘old cat’. Breed differences in feline life expectancy will also impact upon what is considered ‘old’ or ‘geriatric’ for different cat breeds. In New Zealand, 80% of owned cats are mixed/non pure bred (Anonymous, 2020c). Given the longer lifespan of crossbred cats (Table 2.1), this could mean that the median lifespan of New Zealand cats is higher than in other countries with more purebred cat ownership. Therefore, age at ‘geriatric’ might likewise be older for New Zealand cats. For these reasons, my research will allow participants to judge what a cat’s age may mean for its life expectancy and what they perceive as ‘old’. Chronically ill cats have also been included in my research because EoL decisions for these animals may be similarly complicated. Cats are susceptible to a range of chronic diseases, for example, renal disorders and neoplasia (O'Neill et al., 2015; Rollin, 2007). The extended duration of these kinds of illnesses and the, often, slow decline contributes to the difficulty in making QoL assessments and EoL decisions. That is, the point at which the animal’s life is no longer ‘worth living’ is not always clear (Anonymous, 2009; Yeates, 2011). These old and chronically ill cats in New Zealand are at risk of animal welfare compromise if they are kept alive beyond a reasonable QoL. CHAPTER 2: Literature review 11 Table 2.1 Longevity and 'geriatric' ages for various cat breeds in the United Kingdom Breed Median age (range) at death in years* Age calculated as ‘geriatric’ in years† Birman 16.1 (1.0-20.7) 12.1-12.9 Burmese 14.3 (0.7-20.7) 10.7-11.4 Siamese 14.2 (0.9-21.1) 10.7-11.4 Persian 14.1 (0.0-21.2) 10.6-11.3 Crossbred 14.0 (0.0-26.7) 10.5-11.2 British shorthair 11.8 (0.0-21.0) 8.9-9.4 Maine Coon 11.0 (0.2-19.0) 8.3-8.8 Ragdoll 10.1 (0.1-17.9) 7.6-8.1 Abyssinian 10.0 (1.0-20.8) 7.5-8.0 Bengal 7.3 (0.6-13.7) 5.5-5.8 Overall 14 (0.0-26.7) 10.5-11.2 *Based on data from O’Neill et al. (2015) †Calculated according to the definition of 75 - 80% of anticipated lifespan (Baetge & Matthews, 2012; Dodman et al., 1984; Hughes, 2008) 2.2. Making the decision to euthanase a companion cat This research topic represents an under-appreciated area for animal welfare concern. In New Zealand, cats are considered the property of their owners (Anonymous, 1999). This has implications for EoL management, because owners are legally entitled to decide if, and when, to end the life of their animal and often do so in consultation with their veterinarian. The way in which EoL is managed and euthanasia decisions made are important to three groups: the animal, the owner/client, and the veterinarian. For cats, negative experiences due to age-related changes (e.g., pain caused by arthritis), or chronic CHAPTER 2: Literature review 12 illness, reduce the quality of these animals’ lives (Robertson, 2008). As cats are unable to verbally communicate their affective (mental) experiences (McMillan, 2005; McMillan, 2007; Taylor & Mills, 2007), the EoL decision falls to their owner, often in consultation with a veterinarian. However, there appears to be a growing ethos, particularly among urban CA owners, that any life is preferable to a humane death. In support, many CA owners place a high value on keeping animals alive (Anonymous, 2009; Sandøe & Christiansen, 2007; Wensley, 2008), and there is potential for this attitude to have significant detrimental impacts on feline welfare. This is particularly the case for older cats and those with chronic disease, whose slow deterioration may result in them being kept alive beyond what is optimal for their QoL. Therefore, separating the owner’s perceptions of their animal’s QoL from their personal situation is challenging and tends to result in an overlap between animal and owner considerations for EoL decisions. With this overlap in mind, seven factors that influence euthanasia decision-making have been articulated in the literature. The first three factors relate to the animal itself, while the latter four are owner-related factors. Of the three animal-related factors, ‘overall QoL’, or an animal that is ‘suffering’, is the most suggested reason for euthanasia in several studies (e.g., Ireland et al., 2011; McGowan et al., 2012; Slater et al., 1996). Secondly, pain, either caused by an incurable disease, that which is chronic and recurrent, or acute severe pain, is another factor (e.g., Ireland et al., 2011; McGowan et al., 2012). Third, a poor or ‘hopeless’ prognosis is a frequent finding (e.g., Ireland et al., 2011; McGowan et al., 2012; Slater et al., 1996). CHAPTER 2: Literature review 13 In terms of owner-related factors, the owner’s relationship with their animal is often cited (e.g., McGowan et al., 2012), as is the fifth factor: veterinary advice (e.g., Ireland et al., 2011; McGowan et al., 2012). Sixth, anticipated burden or cost to the owner can also be a cause for euthanasia (e.g., Ireland et al., 2011; McGowan et al., 2012; Slater et al., 1996), as can the seventh factor, their previous experience with the disease or with the death of a pet and their stance towards euthanasia (e.g., Dickinson et al., 2014; Slater et al., 1996). Each of these factors is discussed in more detail below. Factors influencing EoL management that relate to the animal Three factors that influence how the end of their life is managed, and when the euthanasia decision is made, relate primarily to the animal itself: overall quality of life, pain, and prognosis. Quality of life and its assessment As we cannot directly measure the mental experiences of animals, EoL management typically relies on clinical, and other, indicators of QoL or animal welfare status (Fraser, 2008; McMillan, 2000; Mellor & Stafford, 2009). In particular, the mental experience of pain and its palliation in animals near the end of their lives is one of the cornerstones of veterinary hospice care (Bishop et al., 2016; Epstein et al., 2005; Goldberg, 2016). CHAPTER 2: Literature review 14 Despite its meaning being intuitively understood (McMillan, 2000; McMillan, 2005), there is currently no universally accepted definition of, or assessment methodology for, QoL in animals (McMillan, 2000; Taylor & Mills, 2007). This is due to the continual evolution of knowledge about the mental experiences of animals, changes in how society expects animals to be treated, and developments in the scientific assessment of the outcomes of such treatment (Fraser, 2008; Rollin, 2007). QoL is conceptualised by some as the animal’s welfare state, or mental experiences, assessed over an extended period (Taylor & Mills, 2007; Yeates, 2012). Several authors consider QoL and animal welfare to be synonymous or see QoL as simply representing a part of animal welfare (e.g., Broom, 2007; Fraser & Duncan, 1998; Mellor & Stafford, 2009). However, because QoL tends to be more intuitively understood by lay animal carers and CA owners, I have chosen to use this term, rather than animal welfare status, throughout this thesis. The term ‘health-related quality of life (HRQoL)’ is also gaining traction and appears to be used as a measure of the success of medical (e.g., Iliopoulou et al., 2013; Lynch et al., 2010) or surgical (e.g., Milner, 2006; Snelling & Edwards, 2003) treatments for individual animals (Freeman et al., 2005). HRQoL assessment may be used to discriminate between individuals or to assess the extent of changes in a particular animal’s status over time, for example, when evaluating a treatment protocol, monitoring disease progression, or considering prognosis (Reid et al., 2013). CHAPTER 2: Literature review 15 However, HRQoL is poorly defined, and there tends to be an overlap in meaning with the broader concept of QoL (Broom, 2007; Iliopoulou et al., 2013; McMillan, 2000). For example, in veterinary oncology, HRQoL has been described as “the effect of cancer and its treatment on body function and well-being” (Lynch et al., 2010, p. 172). To assess the effect of disease on an animal’s overall QoL, health needs to be recognised as one component of QoL. However, if other aspects of an animal’s life are not included in QoL assessments, there is a risk that sources of negative experiences not directly related to their disease or treatment will be ignored (e.g., boredom), and potential sources of pleasure not considered (e.g., affectionate sociability resulting from affiliative interactions) (Hewson et al., 2007; Mellor et al., 2020). There is currently no gold standard protocol for QoL assessment in small animals (Hewson et al., 2007; McMillan, 2000; Taylor & Mills, 2007). Formal tools may be used to assess QoL, and there are several scales designed by veterinarians, using their own experiences of EoL decision-making, to indicate when euthanasia may be appropriate (e.g., Hilst, 2013; Villalobos, 2004). However, these scales have not been validated, nor is there evidence of science-based reasoning for inclusion of the individual items contributing to the scales (Belshaw et al., 2015). There is also a need for an improved understanding of the different perspectives of the potential users of these tools, namely, the owner and their veterinarian (Weary et al., 2016). CHAPTER 2: Literature review 16 Pain and its assessment Pain is another factor identified as influencing euthanasia decisions for CAs. Pain features in most QoL assessment protocols, and particularly in those designed for EoL decisions (e.g., Hilst, 2013; Villalobos & Kaplan, 2007). For example, one such protocol assigns 10 of the available 80 QoL assessment points to “low Ouch or Pain” when assessing for “a healthy and happy pet” (Hilst, 2013). Current definitions of pain emphasise its subjective nature, distinguishing the mental/emotional component of pain from its simple sensory function, and clarifying its adaptive role and impact on QoL. For example: “An unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage.” (Raja et al., 2020, p. 1977) These kinds of definitions emphasise the importance of painful experiences for the person, or animal, experiencing them. Feline pain and its management have been recently reviewed (e.g., Merola & Mills, 2016; Monteiro & Steagall, 2019; Steagall & Monteiro, 2019), and guidelines developed (e.g., Epstein et al., 2015; Mathews et al., 2014), by experts in veterinary anaesthesia and analgesia. However, the absence of pain does not imply that an animal is experiencing good QoL. Pain is only one kind of unpleasant mental experience that an animal may have and is only one factor among many that can impact upon the animal’s QoL, and thus the decision to end their life. Pain is also an emotional (mental) state; therefore, attention CHAPTER 2: Literature review 17 should be given to other aspects of an animal’s QoL (i.e., welfare) that have the potential to impact on its mental experiences, and thus its overall welfare (Taylor & Robertson, 2004). For example, physical discomfort and limits on threat avoidance may add to the negative experiences, and potentially worsen the pain experience, of a hospitalised cat (Taylor & Robertson, 2004). For this reason, and because feline pain and its assessment were not the primary focus of this research, a brief review of feline pain management, recognition, and assessment follows. A pain response is a multifactorial event with sensory components (i.e., nociception; the neural processing of noxious stimuli) and perceptive components (i.e., pain experience; a sensory and emotional experience with an unpleasant quality) (Epstein et al., 2015; Reid et al., 2013). The emotional experience of pain, which is of most direct relevance to welfare or QoL, arises only due to activity at cognitive levels within the central nervous system (Johnson, 2018). Because animals are unable to directly communicate their mental experiences (e.g., the emotional component of pain that is unpleasant), pain is assessed by proxy using physiological and behavioural indicators (Beausoleil & Mellor, 2017; Epstein et al., 2015; Murrell & Johnson, 2006). This is still challenging, because such indicators may reflect nociception, rather than the subjective emotional component of pain experienced by the animal, and may be influenced by various factors other than pain. For instance, no single physiological measurement can be reliably used as a sensitive indicator of pain CHAPTER 2: Literature review 18 outside of a research setting (Murrell & Johnson, 2006; Taylor & Robertson, 2004). In a clinical setting, physiological indicators such as blood pressure, heart rate, and respiratory rate can be affected by factors such as fear and anxiety (Monteiro & Steagall, 2019; Murrell & Johnson, 2006; Paul-Murphy et al., 2004; Taylor & Robertson, 2004). In a veterinary clinical sense, behaviour-based indicators of pain in cats include posture, activity level, demeanour, interaction with people, attention to the wound, vocalisation, facial expression, response to treatment, and response to touch, pressure, or palpation (Epstein et al., 2015; Merola & Mills, 2016; Waran et al., 2007). Algometers are useful for mechanical threshold testing of pain responses and can indicate hyperalgesia – an extreme response to noxious stimulation. In clinical settings, the use of algometers is often replaced with physical palpation of animals to gauge their responses to touch (Epstein et al., 2015; Monteiro & Steagall, 2019; Taylor & Robertson, 2004). Less specific behaviours such as changes in appetite are not sensitive clinical indicators of pain; some cats experiencing pain will eat and there are many other potential causes for inappetence or anorexia other than pain (Monteiro & Steagall, 2019). These challenges mean that there is no gold standard technique for assessing pain in cats (Epstein et al., 2015; Murrell & Johnson, 2006). In an attempt to address some of these challenges, composite scoring methods, or ‘pain scales’, have been developed that use a combination of techniques to evaluate CHAPTER 2: Literature review 19 pain experience (e.g., Brondani et al., 2013; Finka et al., 2019; Reid et al., 2017). For example, the refined Glasgow Composite Measures Pain Scale uses multiple animal-based indicators and has been validated for clinical use on cats with surgical, traumatic, and medical acute pain conditions (Reid et al., 2017). The ‘recommended analgesic intervention level’ for the Glasgow scoring system is 5/20 (Reid et al., 2017). However, the precautionary principle suggests that no cat should be denied analgesic therapy based only on the results of a pain score (i.e., if in doubt, analgesia should be provided), so these scores should be used with caution when it comes to therapeutic interventions (Monteiro & Steagall, 2019; Sneddon et al., 2014). Despite the existence of tools for pain scoring animals, research suggests there may still be bias in analgesic use amongst veterinarians (Dohoo & Dohoo, 1996; Williams et al., 2005). Female veterinarians and recent graduates are reported to use analgesics for companion animals more than male and earlier graduating veterinarians. The authors suggest this may be linked to gender differences in discerning pain in animals, as well as differences in empathy towards animals (Dohoo & Dohoo, 1996; Williams et al., 2005). However, Farnworth et al., (2014) did not find the same difference in analgesic use between genders, which suggests that this disparity may be reducing over time. Owners also have an important role to play in improving the management of chronic feline pain in their home environment. They know their cat well and are CHAPTER 2: Literature review 20 more likely than veterinarians to detect subtle changes in their cat’s behaviour (Epstein et al., 2015; Monteiro & Steagall, 2019; Reid et al., 2013). However, owners may not be able to recognise these behaviours as indicators of pain and, over time, chronic pain behaviours may be disregarded by owners as ‘normal’ or an inevitable part of the ageing process (Taylor & Robertson, 2004). Poorly recognised and managed feline pain can result in early euthanasia (Epstein et al., 2015; Taylor & Robertson, 2004). Therefore, educating owners about pain behaviours is recommended (Epstein et al., 2015; Monteiro & Steagall, 2019). Involving owners in pain management has the added advantage of increasing veterinarian-client interaction and shared decision-making, as well as improving the human-animal relationship (Epstein et al., 2015; Monteiro & Steagall, 2019). Prognosis and life-expectancy Many people feel the need to find an element of certainty in the uncertainty inherent in some EoL decisions for animals (Christiansen et al., 2013; Stoewen et al., 2019; Stoewen et al., 2014). To achieve this, many owners look to their veterinarian to understand what is happening with their animal. Veterinarians in turn make use of the prognostic and life-expectancy data that are emphasised in veterinary literature about chronic conditions and disease trajectories (Rollin, 2007). The result is that owners might then focus on prognosis and life-expectancy when making euthanasia decisions. CHAPTER 2: Literature review 21 To summarise, in terms of factors relating to the animal itself, the focus of EoL management and euthanasia decision-making for CAs has been on QoL and protocols designed to assess QoL, with a particular emphasis on HRQoL, including pain and prognosis. Assessments of these animal-related factors involve subjective judgements by owners and are thus influenced by factors relating to owners themselves. Factors influencing EoL management that relate to the owner Due to the complex nature of decision-making, and to the reliance on humans to make these decisions for the animals under their care, there is a need to explore the human and ‘other’ factors that also influence EoL management (Dickinson et al., 2014; O'Neill et al., 2015; Sanders, 1995). A better understanding of these factors may improve the decisions made, as well as reduce the emotional burden on owners and veterinarians involved in the decision-making process (Morris, 2012; O'Neill et al., 2015). In general terms, owner factors that may influence decisions about EoL management include the owner’s relationship with their animal, the anticipated burden or cost to the owner, the owner’s ethical stance towards euthanasia, and veterinary advice. Relationship with animal The decision to end an animal’s life may depend on the category of animal concerned and its use or perceived value to the persons involved (Brockman et al., 2008). For example, livestock are often considered differently to CAs. Even within CHAPTER 2: Literature review 22 species, a dairy cow in a commercial herd that is no longer producing adequately will be considered differently than if it is the farmer’s ‘pet’ with the associated attachments. In the same way, an indoor dog owned by a family may have a different value to one that is kept outdoors and used solely for farm work. Therefore, animal value, category, and use are all important factors in EoL decision-making (Brockman et al., 2008; Rollin, 2011). Linked to value, the relationship an owner has to a particular animal within a class will likely impact upon the decision to end its life (Dawson & Campbell, 2009; Williams et al., 2017). For example, some people may consider some CAs to be members of the family and be willing to go to extremes to maintain that relationship even if their QoL is poor (Brockman et al., 2008; Dawson & Campbell, 2009; Heuberger & Pierce, 2017; Rollin, 2011; Williams et al., 2017). Others may have a contractual view of animals and believe that the CAs under their care are used for the benefit of humans. Such owners may give less consideration to their animal’s overall wellbeing or death (Brockman et al., 2008; Rollin, 2002; Rollin, 2011). The term ‘human-animal bond’ is used frequently in North American literature to refer to a very strong attachment and commitment of a person to their (usually) companion animal (e.g., Brackenridge et al., 2012; Knesl et al., 2016; Ormerod, 2008; Rowan, 2008; Schneider et al., 2010; Shaw, 2006; Serpell, 2019). However, this term has been explicitly avoided in this thesis for several reasons. Firstly, it is CHAPTER 2: Literature review 23 inherently value-laden and risks ‘othering’ animal owners who do not share this same overtly close relationship with their animal(s) (e.g., Brackenridge et al., 2012; Schneider et al., 2010). Secondly, the human-animal bond, ordered as such, implies a unidirectional relationship between owner and animal. That is, the owner is bonded to the animal. This term does not support a dyadic relationship between owner and animal, and based on applied animal behavioural science, nor should it because attachment and relationships are structured differently for animals (Ines et el., 2021; Payne et al., 2015; Rehn et al., 2014; Rehn & Keeling, 2016). This unidirectional relationship carries risks for an animal’s welfare. For example, if the owner is unable to separate their needs in the bonded relationship from the animal’s QoL, they may delay severance of this relationship (i.e., euthanasia) (Grimm et al., 2018; Quain et al., 2021; Rollin & Rollin, 2001; Serpell, 2019; Wensley, 2008). And finally, as Rollin & Rollin (2001) so eloquently put it, companion animals are ‘used’ by humans (i.e., for companionship) and the promulgation of the term ‘human-animal bond’ can be detrimental to these animals, and veterinarians, when used indiscriminately: “…a companion animal is kept to give, and to receive, love, probably the ultimate human requirement…. we have a contractual relationship with all domestic animals, but most clearly so with those who are totally dependent on us, and for whom we have left no room to subsist, let alone thrive, on their own. If the human animal bond is to be more than a slogan for the very lucrative pet industry, more than a marketing ploy for veterinary services, we CHAPTER 2: Literature review 24 must face up to the fact that animals are doing fine holding up their end of the bargain; it is we who should be ashamed.” (Rollin & Rollin, 2001, pg. 7) To capture the full extent of this potential dyadic relationship, the term ‘human- animal relationship’ has been preferentially used as much as possible throughout this thesis. The strength and dynamics of the human-animal relationship are evident when the grief of owners over the loss of an animal is examined (Payne et al., 2015). Grief may be more significant if the animal is long-lived, has been an integral part of family life, or represents a human relationship – past or present. Anticipatory grief occurs when owners are not adequately prepared to sever their connection to their animal. This form of grief may result in CA owners keeping their animal alive beyond a reasonable QoL (Spitznagel et al., 2020). Consideration of how grief is managed leading up to and after euthanasia is an important aspect of EoL management but is not the main focus of this research, except in terms of the veterinarian’s role. Many studies have already concentrated on the grief experience for people who have recently euthanased their CA, that is, owner grief (veterinary grief is discussed in a later section) (e.g., Fernandez-Mehler et al., 2013; Slater et al., 1996). In Western social culture, grief relating to the loss of an animal is not always legitimised by peers (Adams et al., 1999; Chur-Hansen, 2010; Dickinson et al., 2010; Marton et al., 2019; Redmalm, 2015; Testoni et al., 2019). Owners who struggle to deal with grief at the loss of their animal are at risk of ‘complicated grief’ CHAPTER 2: Literature review 25 and depressive episodes, and anticipatory grief may cause them to delay euthanasia to avoid negative emotions (Adams et al., 1999; Testoni et al., 2017). There are obvious implications for animal welfare if animals experiencing poor QoL are not euthanased in a timely manner (Knesl et al., 2017; Rollin, 2011). Thus, veterinarians, as members of an animal care profession and animal welfare advocates, have an important role to play in supporting and validating their clients’ grief (Adams et al., 1999; Marton et al., 2019; Peck, 2005; Testoni et al., 2019). Burden or cost to owner Financial and time commitments are important considerations for many owners who are faced with managing the end of their animals’ lives (Brockman et al., 2008; Coe et al., 2007; Yeates, 2012). These owners are often forced to decide whether they will engage with therapy, or palliation, for their animal, along with the associated hardships for them (Brockman et al., 2008). Owners without pet insurance were seven times more likely to opt for euthanasia over surgery for their dog with gastric dilatation-volvulus (Boller et al., 2020). This decision is often based on them weighing up the owner- and animal-related factors alongside their resource capabilities (e.g., time, money, skill). If, on balance, the burden of keeping an animal alive is too great, a decision to euthanase may result. Stance towards animal euthanasia The owner’s relationship with their animal, their previous experience with death, and other personal factors (e.g., religion, culture) may impact on their overall CHAPTER 2: Literature review 26 stance towards animal euthanasia (Dickinson et al., 2014; Slater et al., 1996). This may result in different owners in similar circumstances coming to different conclusions about how to manage the end of their animals’ lives. All these owner-related personal or psychosocial factors may impact on EoL management of animals, but more information is needed to understand how (Spitznagel et al., 2020; Stoewen et al., 2019). Inevitably there will be some overlap between animal- and owner-related factors. Cats cannot verbally communicate, so it is up to their owner, with or without their veterinarian, to interpret behavioural and other indicators, in terms of QoL, to make EoL decisions. While some studies have explored aspects of EoL decision-making for CAs, no one has focused on the entire EoL management process. There has not been a holistic examination of the owner’s perspective on, and management of, EoL decision-making for cats. Retrospective interviews with owners following a euthanasia decision will provide valuable information about how owners and their veterinarians communicated about, and eventually settled upon, an EoL decision for the cat concerned. These interviews will also explore the methods used for QoL assessment as part of the EoL decision and how grief was managed. Owner understanding and assessment of QoL is also likely to have an impact on how EoL decisions are made. This makes the process of segregating EoL related factors into discrete survey questions challenging. A reflexive social science approach to understanding this complex situation is likely to provide a richer, more holistic, understanding of euthanasia decision-making (Charles et al., 1997). CHAPTER 2: Literature review 27 2.3. The veterinarian’s role in managing the end of animals’ lives Veterinary advice was cited in the previous section as an important factor that influences EoL decision-making for animals. The advice veterinarians give owners, and their role in EoL management of animals, is influenced by several factors, namely: their relationship with their client, legal and professional obligations, competency, communication skills, and how they are taught. Many of these factors sit under the umbrella of veterinary professionalism. Veterinary professionalism Professionalism is distinct from professional skills. Professional skills include communication, management, business, teamwork, and leadership, whereas professionalism has an emotional component and refers to appropriate behaviours and attitudes (Mossop & Cobb, 2013). Both professionalism and professional skills impact upon the veterinary-client relationship (Mossop & Cobb, 2013). Mossop (2012) asked veterinarians, veterinary nurses, clients, and representatives from the veterinary professional association in the United Kingdom to describe what they understood by the term ‘veterinary professionalism’. The attributes they listed included caring and empathy, honesty and trust, altruism, personal efficiency, communication skills, problem-solving, decision-making, self- confidence, self-regulation, knowing one’s limits, and maintaining technical competency (Mossop 2012). These attributes balance the interests of the client, the CHAPTER 2: Literature review 28 patient, the practice, and the wider veterinary profession and represents the core of professionalism (Mossop 2012). Tinga et al. (2001) surveyed veterinary students and new graduates about their perceptions of veterinary technical and professional skills. When describing the characteristics of a successful veterinarian, new graduates ranked interpersonal skills and veterinary-client communication skills higher than knowledge of veterinary sciences or accurate diagnostic skills (Tinga et al. 2001). Likewise, members of the Quebec Veterinary Association identified professional activities such as important for career success (Doucet & Vrins 2009). Skills in assessing animal welfare or quality of life were not explicitly evaluated by Tinga et al. (2001). Likewise, important attributes relating to veterinary-animal communication of mental experience (i.e., animal welfare assessment) were manifestations of disease in individual animals or populations, mechanisms of disease, mechanisms of pathogenicity, normal behaviour, physiology, nutrition, disease aetiology, and pathophysiology (Doucet & Vrins 2009). Neither study explicitly focused on the mental experiences of animals. These studies support the inclusion of professionalism training in veterinary curricula. In support, a systematic review of the veterinary literature to determine the importance of professional competencies to graduate success revealed that professional behaviour and communication skills were the two competencies CHAPTER 2: Literature review 29 recognised as most important (Cake et al. 2016). Communication is a professional skill very closely related to professionalism because effective communication skills are necessary to demonstrate a professional attitude (Mossop & Cobb, 2013). Health and welfare advocacy were less important for veterinary graduates. However, the importance of being able to assess animal welfare was not explicitly evaluated and ‘patient care’ was confounded with ‘effective communication with clients’ (Cake et al. 2016). Interestingly, communication skills were perceived to be important by veterinarians and employers but less so by clients, perhaps implying that clients have a different communication focus than veterinarians (Cake et al. 2016). The role of veterinary-client communication in EoL management of animals When a veterinarian is involved, the way in which the end of animal life is managed, and euthanasia decisions made, is important to three groups: the animal/patient, the owner/client, and the veterinarian. Therefore, these same three parties should be considered in veterinary communication strategies during EoL management of animals (Figure 2.1). This thesis focuses on communication between the veterinarian and the animal as well as between the owner and animal (Figure 2.1; blue and green arrows). That is, my focus is on how veterinarians and owners assess QoL (i.e., animal welfare) at the end of animal life (Figure 2.1; blue and green arrows), factors that might CHAPTER 2: Literature review 30 impact on these assessments (Figure 2.1; red boxes and arrows), and how this process is taught and enacted in practice. Animals ‘communicate’ their mental experiences (i.e., how they are feeling or their quality of life) via welfare indicators (Beausoleil & Mellor, 2017). These indicators must be correctly recognised and interpreted by humans for effective veterinarian-animal and owner-animal communication to occur (Figure 2.1; blue and green arrows). Figure 2.1 The tripartite communication relationship between animal, veterinarian, and owner; QoL = quality of life. The relationship between veterinarian and owner/client also factors into EoL management of animals (Persson et al., 2020; Rollin, 2007; Rollin, 2011), and can be enhanced by effective communication strategies (Adams & Frankel, 2007; Knesl, 2016; Main, 2013). Therefore, in this section, I will review the role of Animal/patient Owner/clientVeterinarian Indicators of QoL as perceived by the veterinarian (‘animal-veterinary communication’) Indicators of QoL as perceived by the owner/client (‘animal-owner communication’) Veterinary-client communication strategies Veterinary factors impacting on how QoL is assessed e.g., animal value Owner factors impacting on how QoL is assessed e.g., relationship with animal CHAPTER 2: Literature review 31 veterinary-client communication in EoL management of companion animals (Figure 2.1; yellow arrow). Effective communication for human medical professionals has been well described (e.g., Baile et al., 2000; Dickinson, 2006; Kurtz & Silverman, 1996; Rosenbaum et al., 2004; Tulsky, 2005; Weissman, 2004). Information and research about veterinary-specific communication skills is increasing (e.g., Adams & Kurtz, 2006; Adams & Frankel, 2007; Borden et al., 2019; Brown et al., 2021; Coe et al., 2008; Cornell & Kopcha, 2007; Cron et al., 2000; Da Costa, 2021; Dysart et al., 2011; Gordon et al., 2021; Janke et al., 2021a,b; Matte et al., 2020; Radford et al., 2006; Shaw, 2006; Shaw et al., 2008; Stoewen et al., 2014a,b; Stoewen et al., 2019) and conferences dedicated to veterinary communication are becoming commonplace (e.g., Anonymous, 2021b). There is now literature dedicated to aspects of veterinary EoL management communication such as delivering bad news (e.g., Radford et al., 2006; Shaw & Lagoni, 2007), clinical decision-making (e.g., Cornell & Kopcha, 2007; Janke et al., 2021a,b; Spitznagel et al., 2020), euthanasia discussions (e.g., Borden et al., 2019; Matte et al., 2020; Radford et al., 2006; Shaw & Lagoni, 2007), and grief management (e.g., Adams et al., 1999; Bussolari et al., 2018; Marton et al., 2019; Matte et al., 2020; Peck, 2005; Radford et al., 2006; Testoni et al., 2019). Therefore, despite historic arguments to the contrary (Shaw, 2006; Shaw et al., 2004), there now exists sufficient veterinary-specific communication literature to focus this brief review of veterinary communication skills (Adams, 2013). CHAPTER 2: Literature review 32 This section focuses on the veterinary-client communication for several reasons. Firstly, unlike much of human medicine, excluding paediatric care (Feudtner, 2007; Truog et al., 2006), the patient and client are not the same. Therefore, there are practical ‘communication’ constraints in interpreting ‘what animals want’ (i.e., their welfare) (Fraser et al., 1997; Taylor & Mills, 2007; Yeates, 2012). Veterinary clients are proxy decision-makers for their animals (‘the patient’), with the result that veterinarians may be torn between owner and animal needs, or welfare, when communicating and interacting with clients (Blackwell, 2001; Grimm et al., 2018; Quain et al., 2021). Secondly, euthanasia is an option for veterinary EoL management, whereas euthanasia, or ‘physician-assisted suicide’, is limited to a very small number of jurisdictions for human EoL management (e.g., The Netherlands; Borgsteede et al., 2007; Georges et al., 2007). Human palliative care literature could be a useful source of information to guide veterinary communication strategies during EoL discussions (e.g., Hamdoune & Gantare, 2021). However, having euthanasia as a viable ‘treatment’ option makes EoL discussions particularly challenging for veterinarians, as they are often faced with shifting their focus, and conversations, from heroic efforts to save an animal’s life to potentially ending it (Quain et al., 2021; Taylor, 2021). When used as a means for delaying euthanasia, palliation and hospice care increase the risk of overtreatment or dysthanasia, that is, ‘difficult death’, and have serious welfare implications for the animal involved (Quain et al., CHAPTER 2: Literature review 33 2021). A measured discussion about the ethics of animal hospice and palliative care is beyond the scope of this thesis. As such, literature focused on palliation (veterinary or human) will only be briefly covered in this decision-making section. Third, unlike human medicine in many countries, veterinary medicine is not state- funded. Owners are full fee-paying clients (and/or pay for pet insurance), therefore, cost-benefit discussions are differently weighted for animals than for humans, and euthanasia is an alternative to costly advanced veterinary care (Brockman et al., 2008; Brown et al., 2021; Coe et al., 2008; Quain et al., 2021). Finally, the relationship between owners and their animals dictates their treatment and how veterinarians communicate with different clients (Persson et al., 2020; Rollin, 2007; Rollin, 2011). Altogether, these factors are likely to influence how veterinarians communicate EoL management with their animal-owning clients when compared with their human medical counterparts (Adams, 2013). Models of veterinary-client relationships and communication styles The veterinarian’s relationship with their client can influence the role they play in EoL management for animals. In their clinical role, veterinarians may be tasked with caring for, treating, and subsequently euthanasing the same animal (Persson et al., 2020; Rollin, 2007; Rollin, 2011). Amid these potentially conflicting roles, veterinarians may need to advise owners on EoL decisions while also attending to the owner’s concerns and grief (Persson et al., 2020). Therefore, the owner- CHAPTER 2: Literature review 34 veterinary relationship is an important consideration in EoL management of animals. Veterinarians can take various roles when they interact with their clients. Examples of these roles include: the paternalistic veterinarian (Emanuel & Emanuel, 1992), the information-provider, the garage mechanic, or paediatrician (Rollin, 2006), the veterinarian who influences their client (Christiansen et al., 2016; Yeates & Main, 2010), or who shares in the decision-making process (Charles et al., 1997; Christiansen et al., 2016; Rollin, 2002). The paternal model of healthcare is based on the traditional role of the veterinarian as an all-knowing father figure. A paternalistic veterinarian would have used their Aesculapian authority to dictate their client’s decisions (Charles et al., 1997; Emanuel & Emanuel, 1992; Rollin, 2002; Rollin, 2007). This is the unique authority attributed to medical professionals, including veterinarians, and is a vestige from when medicine and divinity were inseparable. This authority derives from traits such as specialised knowledge and a moral imperative to heal and do good by their patients, as well as the perception that medicine may still be related to magic by the scientifically naïve (Persson et al., 2020; Rollin, 2002). Rollin (2002) argues that Aesculapian authority is the most powerful type of authority and gives the example of kings and politicians being dictated to by their personal physicians. This paternal model has since gone out of favour in both CHAPTER 2: Literature review 35 human and veterinary medicine and has been replaced with concepts such as informed choice and client autonomy (Back et al., 2009; Charles et al., 1997; Emanuel & Emanuel, 1992; Rollin, 2002). However, Rollin (2002) argues that paternalism, despite being vilified, has benefits when used for EoL management of animals. Paternalist veterinary EoL decision-making and training would allow the veterinarian to answer a client who asks, “What would you do?” in a directive manner, rather than avoiding the question (Rollin 2002). This paternalistic style of communication may also be useful during the initial phases of emergency consultations when an urgent decision needs to be made (Bateman, 2007). The information-provider model of healthcare is at the opposite end of the decision-making spectrum to the paternal model. In this model, the veterinarian gives their client the information necessary for an informed choice but leaves the decision to the client and respects their autonomy (Heuberger & Pierce, 2017; Morgan & McDonald, 2007; Slater et al., 1996). Client autonomy is often touted in the veterinary literature as being Best Practice for owner-veterinarian interactions during EoL decision-making (Morgan & McDonald, 2007; Morton, 2010). The term regularly appears in North American-centric publications and is often defined by decisions based on informed consent (e.g., Morgan & McDonald, 2007). However, as Rollin (2002) argues, this veterinary-client communication style risks an owner keeping an animal alive if they are overly focused on their own interests (i.e., permanent severance of a relationship). Veterinarians have a duty to CHAPTER 2: Literature review 36 end suffering when an animal’s QoL is not worth living and some argue that they should exert their Aesculapian authority to achieve this (Rollin 2002). The ‘information provider’ communication model can be refined by using an interactive information-giving technique, such as the ‘chunk and check’ method (Silverman et al. 2005). This method encourages veterinarians to give small ‘chunks’ of information and check for understanding before giving more information (Silverman et al. 2005). Ascertaining a client’s communication and information preferences is another way of promoting owner participation during a clinical consultation and is included in frameworks promoting effective veterinary- client communication e.g., Calgary-Cambridge communication guides (Radford et al., 2006) and the SPIKES model (Baile et al., 2000). An alternative paradigm for owner-veterinarian interactions focusses on allowing the client to decide how much of a role their veterinarian plays in the decision- making process (Christiansen et al., 2016; Janke et al., 2021a; Yeates & Main, 2010). For this alternative form of autonomy to be achieved, there is an expectation that information will be given to clients in a direct and jargon-free manner and their participation preferences elicited (Stoewen et al., 2014). In line with this paradigm, Yeates and Main (2010) argue veterinarians can exert influence over their client’s decision-making in circumstances when the client has asked them to. In this case, veterinarians may lead the decision-making process, while still respecting client autonomy because the client has delegated the veterinarian this role (Rollin, 2011; CHAPTER 2: Literature review 37 Yeates & Main, 2010). Within this veterinary-client communication style, the veterinarian can answer the ‘what would you do?’ question and direct a client towards euthanasia of their animal (Rollin 2002). Veterinarians can also share in the decision-making process with their client (Charles et al., 1997; Christiansen et al., 2016; Rollin, 2002; Spitznagel et al., 2018). Shaw et al. (2006) described this shared communication style, found in 42% of appointments in a companion animal practice, as a partnership between veterinarian and client with a focus on creating a joint venture or ‘relationship’. The ‘chunk and check’ method is further refined in this approach by also taking the client’s perspective into account, that is, at part of the ‘checking’ process (Silverman et al. 2005). Shared decision-making may have a role in preventing caregiver burden, as it increases the sense of control owners have over the situation while simultaneously providing them with the veterinary expertise they seek (Charles et al., 1997; Spitznagel et al., 2018). Shared decision-making may be linked to client autonomy, as it involves a two-way exchange of information and an agreement on how to proceed, that is, an informed choice. Importantly, this decision-making relies on a shared understanding of preferences and therefore is likely only possible in an ongoing relationship between veterinarian and client (Charles et al., 1997; Christiansen et al., 2016). CHAPTER 2: Literature review 38 Veterinarians who share the EoL decision-making process are less likely to be accused of forcing clients to euthanase their animals, while still acknowledging their role as an animal care professional (Rollin 2002). However, this ‘relationship- centred’ model, despite sounding romantic, is an anthropocentric approach to veterinary communication. This approach emphasises the role of the animal in the owner-client’s life over considerations about the animal’s QoL. Shaw (2013) argues that this communication approach has implications for veterinary professionals, clients, and patients and yet the evidence for this statement is derived from physician-patient communication research. Therefore, the animal, and its QoL, is still not explicitly included in this communication paradigm. The roles veterinarians assume when interacting with their clients can be linked to their ethical concern towards animals. They can act as garage mechanics within a utilitarian ‘animal as object’ framework where their goal is just to diagnose and/or fix animals (Rollin, 2002; Rollin, 2007). This framework is likely to result in paternalistic veterinary-client communication styles. Alternatively, veterinarians can be akin to paediatricians and treat animals as direct objects of moral concern by taking on a more nurturing role (Rollin, 2002; Rollin, 2007). These veterinarians may use persuasive tactics to influence their client when their choices are not ‘reasonable’ (Christiansen et al., 2016; Yeates & Main, 2010). However, as Rollin (2002) argues, the major difference between paediatricians and veterinarians is the role of the ‘consensus social ethic’ when it comes to decision- making. In paediatric care, an obstructive third party (e.g., a parent refusing to CHAPTER 2: Literature review 39 allow necessary medical care) can be circumvented by a court of law. This is not the case for veterinarians and their animal-owning clients (Rollin 2002). The Strong Patient Advocate concept recognizes that veterinarians are obligated to their patient, that is, the animal under their care, and aligns with Codes of Professional Conduct for Veterinarians, but does not give due consideration to the legal status of animals as human property (Anonymous, 2011, 2012; Coghlan, 2018). Overall, it is apparent that veterinarians need to understand their client’s personal circumstances and assist them with decision-making for their animal (Christiansen et al., 2016; Heuberger et al., 2016; Spitznagel et al., 2020). The veterinarian may have to take on several different roles, dictated by the client, to achieve this. How they interact and their relationship would facilitate the veterinarian fulfilling the role desired by a particular client. Flexibility in approach and appreciation of context, coupled with clients’ needs, means that the veterinarian can adopt a style best suited for that veterinary-client interaction (Shaw et al., 2006; Shaw, 2013). Veterinary-client communication during euthanasia decision-making for companion animals This section is framed around communication strategies that can be used during a clinical consultation where an owner has presented their companion animal (CA) for diagnostics that later result in a terminal (or complex) diagnosis, or where the owner is unsure whether to euthanase their CA. CHAPTER 2: Literature review 40 Information can be gathered from a client using open- or closed-ended questions (Cornell & Kopcha, 2007; Dysart et al., 2011; Radford et al., 2006; Shaw, 2006). Both types of questions are useful during clinical discussions and can be used together to gather as much data as possible about an animal (Cornell & Kopcha, 2007; Dysart et al., 2011; Radford et al., 2006; Shaw, 2006). Shaw (2006) recommends a ‘funnel technique’, that is, starting broad by asking owners open- ended questions about their reasons for presenting their animal for veterinary evaluation. These questions are followed by targeted closed-ended questions to clarify details. This approach has the added benefit of promoting owner participation (Shaw, 2006). Reflecting an owner’s message back to them in one’s own words is another way of demonstrating understanding and listening skills and helps to develop a harmonious relationship, that is, rapport (Cornell & Kopcha, 2007; Knesl et al., 2016; Shaw, 2006). Another way of establishing rapport with clients is via acknowledging and demonstrating understanding of the client’s predicament. This can be achieved using empathy (Radford, 2013). In this context, Radford (2013) uses empathy as “…not only a personal recognition of how a client may be feeling but letting the client know of our awareness” (Radford, 2013, pg. 36). This description of empathy most closely aligns with the ‘psychological empathy, projection, or perspective-taking’ concept of empathy or the ‘imagine other’ perspective (Batson, 2009). To achieve this type of empathy, a veterinarian would use the strategy of ‘naming’ their client’s emotions with phrases such as “I can see you are very upset” (Radford, 2013, pg. 36). CHAPTER 2: Literature review 41 However, the limitation of this type of empathy is that it relies on us having a theory of what other people are likely to feel. Conversely, simulating or projecting oneself into the other’s situation or imaging one’s own feelings in the other’s place allows us to feel as the other feels (Batson, 2009), but may lead to feelings of moral distress (Fawcett, 2013; Moira & Van den Brink, 2020; Rollin, 2011). These empathy constructs are limited to human-human relationships. Human empathy towards animals (‘animal-orientated empathy’) has been studied (Paul, 2000), however, this psychometric scale has questions that could be misinterpreted. For example, strongly agreeing with the statement “I get annoyed by dogs that howl and bark when they are left alone” (Paul, 2000, pg. 202), could be interpreted as not being empathetic towards the dog’s behaviour at being left alone, or strongly empathetic towards the dog’s plight at being left alone, that is, not empathetic towards the dog’s owners. Shaw (2006) presents three broad skills that contribute towards effective communication: Content skills relate to the information or ‘what’ is presented (e.g., clinical or technical details), Process skills are ‘how’ or the ‘style’ this information is presented, or gathered, and includes verbal and nonverbal methods (Adams & Kurtz, 2006; Radford et al., 2006), and Perceptual skills are dictated by a person’s cognitive and relationship skills, for example, by their personal attitudes and awareness of others (Shaw, 2006). CHAPTER 2: Literature review 42 The general communication strategies described above are often taught together, and this can be easily achieved by using a framework or model for effective communication, for example, Calgary-Cambridge communication guides (Radford et al., 2006) or the SPIKES model for delivering bad news (Baile et al., 2000). The Calgary-Cambridge Guides are widely used in veterinary communication training and provide a useful framework for developing ‘process’ communication skills, that is, they promote an effective style of communication (Adams & Kurtz, 2006; Radford et al., 2006). The veterinary Calgary-Cambridge model was developed based on the version for physicians (Kurtz & Silverman, 1996) and is designed to be taught and used in role play scenarios with veterinary students and actors (Radford et al., 2006). The authors recognised the need to modify the veterinary version to reflect the ‘tripartite nature’ of veterinary consults, that is, they include veterinarian, patient/animal, and client/owner (Radford et al., 2006). The original style of medical consultation was still retained in the modified (veterinary) Calgary-Cambridge guide by taking a “patient-centred and collaborative” approach (Radford et al., 2006, pg. 39). However, this approach has not translated to patient/animal-centred care, but instead focuses on client/owner-centred care with the result that the animal is secondary to the client throughout the modified guide. For example, the authors refer to a need to “attend to client’s and animal’s physical comfort” during the ‘Initiating the Consultation’ phase of the modified guide, but the animal is only sporadically considered CHAPTER 2: Literature review 43 throughout, usually after the client, and without a framework for such consideration (Radford et al., 2006, pg. 39). Physical comfort is not the only need an animal has. Instead, we now recognise animal needs in five dimensions, or ‘domains’ (Mellor et al., 2020). Therefore, limiting consideration of an animal’s welfare to its comfort level is insufficient for a truly ‘tripartite’ veterinary consultation. Despite acknowledging the importance of understanding ‘what the animal wants’, the veterinary Calgary-Cambridge guide does not discuss how this might be achieved, that is, how to assess animal welfare or QoL. Therefore, there is still a need for animal welfare assessment to be explicitly linked to communication skills training for veterinarians to reduce the risk of overtreatment or dysthanasia (i.e., ‘difficult death’) of animals resulting from ‘miscommunication’ about their QoL (Quain et al., 2021). Another communication framework that could be useful in EoL conversation is the six-step SPIKES model. This model was originally developed for human medics to deliver bad news (Baile et al., 2000). It has been used to train veterinary students to deliver bad news during consultations and as a structured approach to euthanasia decision-making discussions (Shaw & Lagoni, 2007). The six steps of the model’s mneumonic encourages users to consider: • Setting up the interview (i.e., consider private location and people involved in conversation, establishing rapport with open-ended questions), CHAPTER 2: Literature review 44 • Perception (i.e., client’s perspective of what is wrong with animal and desire for information), • Invitation (i.e., permission to share information), • Knowledge and information provision (i.e., providing a warning ‘shot’ of what is to come, delivering bad news in stages e.g., ‘chunk and check’ for understanding using open-ended questions), • Emotions and empathetic responses (i.e., acknowledging, validating, and normalising client’s emotions), • Summarise and strategise (i.e., summarise discussion, discuss plan). A veterinary-specific addition to the ‘Knowledge’ step for EoL discussions includes “provide accurate and detailed information about the animal’s condition” (Shaw & Lagoni, 2007, pg. 102). However, the model does not provide guidance on how this might be achieved or how veterinarians or animal-owning clients might decide when an animal’s life is no longer worth living (i.e., how to evaluate the animal’s QoL using welfare indicators). Additionally, the ‘Perception’ step invites veterinarians to consider the animal’s purpose for the client and defers decision- making to them. Shared veterinary-client decision-making is not possible without effective ways of sharing the information underpinning the decision in a structured and coherent way. Gray & Moffett (2013) advocate a similar approach to veterinary EoL communication using a seven-step tool that separates the client’s perspective CHAPTER 2: Literature review 45 (‘Perception’) of what is wrong with animal and their desire for information into two steps. Critically, ‘welfare concerns’ are discussed separately to the model and not explicitly integrated within its approach thereby focusing on the ‘how’ rather than the ‘what’ of human-animal-related communication (Gray & Moffett 2013). The need for veterinary-specific communication strategies has implications for how veterinary-client communication should be taught in veterinary schools. In their survey of veterinary students and recent graduates, Tinga et al. (2001) found that 37% of graduates felt uncomfortable dealing with demanding people, 25% about delivering bad news, and 20% about interpreting peoples’ non-verbal behaviour and conducting client-present euthanasia. This points to the importance of incorporating targeted veterinary-client EoL communication training in veterinary schools. Veterinary schools in the United Kingdom (UK) advocate approaches adapted from training medical practitioners and cite the Calgary-Cambridge consultation framework (Kurtz & Silverman, 1996) as one such example for teaching consultation structure, techniques, and delivery (Mossop et al. 2015). Interactive sessions using simulators, or trained actors, represent an experiential approach to this teaching and have improved communication outcomes (e.g., improved clinical outcomes and client information retention) over purely didactic delivery (Latham & Morris, 2007). At UK veterinary schools, communication skills are assessed using a mix of formative and summative approaches, such as student performance CHAPTER 2: Literature review 46 during a specific consultation scenario and reflective portfolios and diaries (Mossop et al. 2015). The focus of much of this training is on how to communicate with clients, rather than what to communicate. There is still a need to integrate welfare indicators into frameworks for veterinary EoL communication training, and “providing the correct amount and type of information” is an important communication skill necessitating assessment (Latham & Morris, 2007, pg. 182). Veterinary-client communication during grief management resulting from euthanasia of companion animals Once an animal is deceased, the veterinary communication strategy and relationship moves from consideration of three parties (veterinarian-client- patient) to two (veterinarian-client). Much has been written about veterinary and client grief management in EoL situations for veterinarians (e.g., Adams et al., 1999; Marton et al., 2019; Peck, 2005; Testoni et al., 2019). These approaches have been discussed in the preceding section of this chapter (see Models of veterinary- client relationships and communication styles). However, it is at this stage that the human medical communication literature could offer some insights into how veterinarians might effectively relate to (i.e., communicate) with their grief- stricken clients following the loss of their animal. To illustrate, Van der Geest et al. (2014) explored the influence of parents’ perceptions of their interactions (e.g., communication and parental involvement) with health care professionals during paediatric palliative care on their subsequent grief when their child died of cancer. Parents rated communication (e.g., clear information about prognosis and when CHAPTER 2: Literature review 47 delivering bad news, availability of health care professionals, opportunity to express their emotions), continuity of care, and parental involvement as important during the palliative phase (Van der Geest et al., 2014). Those parents who rated these factors highly experienced lower levels of long-term grief. Also of importance to these parents was how well their child’s symptoms were managed. Factors such as the child’s dyspnoea severity, anxiety, anger, and uncontrolled pain were significantly associated with higher levels of long-term grief for these parents (Van der Geest et al., 2014). These results point to the importance of effective interactions between clinicians and proxy decision-makers (in this case parents, but in the case of this thesis, animal-owning clients) during EoL management. They also draw attention to the importance of ‘symptom management’ (i.e., QoL management) in patients when viewed from the perspective of these proxies. Van der Geest et al.’s findings also show that it is not just how bad news is communicated (i.e., communication styles and skills), but what is communicated (e.g., clear information about prognosis) that is important to people managing end of life and palliative care. In summary, SPIKES is a useful protocol for delivering bad news in a veterinary setting. However, SPIKES focuses on ‘how’ or the ‘style’ of communication, that is, process communication skills (Shaw, 2006). SPIKES may be useful on its own to deliver the news that a veterinary client’s animal has died while hospitalised (e.g., under anaesthesia, or after a protracted illness), but it does not offer strategies for CHAPTER 2: Literature review 48 developing content skills associated with EoL management of animals. In other words, SPIKES does not provide guidance to veterinarians about ‘what’ to present to animal-owning clients during the EoL decision-making process. Overall veterinary communication during EoL management of animals Veterinary communication strategies during EoL management of animals should consider three parties: the animal/patient, the owner/client, and the veterinarian (Figure 2.1). Historically, veterinary communication frameworks have focused exclusively on the veterinarian-client relationship and/or briefly mentioned the animal/patient (e.g., Radford et al., 2006; Shaw & Lagoni, 2007; Gray & Moffett 2013). This anthropocentric approach may be a feature of the origins of veterinary communication in human-centred fields where the patient and client are typically the same individual. However, this approach does not represent the tripartite nature of veterinary