Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The use of animals in science : trends and public attitudes Ormandy, Elisabeth Helen 2012

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
24-ubc_2012_fall_ormandy_elisabeth.pdf [ 1.89MB ]
Metadata
JSON: 24-1.0073306.json
JSON-LD: 24-1.0073306-ld.json
RDF/XML (Pretty): 24-1.0073306-rdf.xml
RDF/JSON: 24-1.0073306-rdf.json
Turtle: 24-1.0073306-turtle.txt
N-Triples: 24-1.0073306-rdf-ntriples.txt
Original Record: 24-1.0073306-source.json
Full Text
24-1.0073306-fulltext.txt
Citation
24-1.0073306.ris

Full Text

THE USE OF ANIMALS IN SCIENCE: TRENDS AND PUBLIC ATTITUDES by Elisabeth Helen Ormandy B.Sc., The University of Edinburgh, 2002 M.Sc., The University of Edinburgh, 2005 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Animal Science)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  October 2012 © Elisabeth Helen Ormandy, 2012  Abstract Given the recent shift towards democratization of science, public engagement (including exploration of public attitudes) on issues related to animal research is important. This thesis explores public attitudes to changing practices in the use of animals in research. Chapter 1 provides a critical review of the existing research related to this topic. Chapter 2 presents a bibliometric analysis of changing patterns in animal use, and documents the increasing use of genetically modified (GM) animals, especially mice and zebrafish. Chapters 3 and 4 describe two online engagement experiments investigating how acceptance of animal-based research is affected by genetic modification, regulation, invasiveness, and the species used. Chapter 3 shows that support for the use of pigs in research decreased when the research involved an invasive procedure or GM animals. Support for invasive research increased when regulation was in place, but regulation had little effect on acceptance of GM animal use. Chapter 4 shows that participants who were willing to support biomedical research on zebrafish were equally willing to support the same research on mice. Participants expressed low levels of support for research involving ethyl-N-nitrosourea (ENU) mutagenesis. Some participants expressed a preference for the use of GM animal models over ENU mutagenesis based on the belief that the former causes less pain, and improves accuracy and efficiency when creating the animal model. Chapter 5 describes an interview study that examined the views of researchers, research technicians, and members of public toward the creation and use of genetically modified animals in biomedical science. The creation and use of GM animals for biomedical research purposes was generally well supported provided that this was associated with tangible human health benefits. However, it was recognized there are obstacles to Three Rs (replacement, reduction, refinement) implementation, and that there should be more effort placed on engaging the public on animal research. Chapter 6 concludes with key policy recommendations: 1) improve scientific reporting, 2) improve data and animal sharing, 3) improve recording of national animal statistics, 4) improve animal welfare assessment, and 5) supplement the Three Rs.  ii  Preface A version of Chapter 1 has been submitted for publication. Ormandy, E. H. and Schuppli, C. A. Public attitudes towards animal-based research: A review. The main ideas for the manuscript were developed and researched by Elisabeth H. Ormandy. A version of Chapter 2 has been published: Ormandy E. H., Schuppli C. A. and Weary D. M. (2009) Worldwide Trends in the Use of Animals in Research: The Contribution of Genetically Modified Animal Models, Alternatives to Laboratory Animals 37: 63-68. This paper was co-authored by C. A. Schuppli and D. M. Weary. These co-authors acted in the role typical of supervisory members. The main ideas for the manuscript were developed and researched by Elisabeth H. Ormandy. Co-authors supervised, helped interpret material, and edited drafts. A version of Chapter 3 has been accepted for publication: Ormandy E. H., Schuppli C. A. and Weary D. M. (in press) Public attitudes toward the use of animals in research: Effects of invasiveness, genetic modification and regulation. Anthrozoös. This paper was co-authored by C. A. Schuppli and D. M. Weary. These co-authors acted in the role typical of supervisory committee members. The main ideas for the manuscript were developed and researched by Elisabeth H. Ormandy. The research was developed in collaboration with Dr. Peter Danielson and the Norms Evolving in Response to Dilemmas (NERD) team at the Centre for Applied Ethics at UBC. Co-authors supervised, helped interpret material, and edited drafts. The project received research ethics board approval from UBC under certificate number: H06-H0532. A version of Chapter 4 has been accepted for publication: Ormandy, E. H., Schuppli, C. A. and Weary, D. M. (in press) Factors affecting people’s acceptance of the use of zebrafish and mice in research. Alternatives to Laboratory Animals. This paper was co-authored by C. A. Schuppli and D. M. Weary. These co-authors acted in the role typical of supervisory committee members. The main ideas for the manuscript were developed and researched by Elisabeth H. Ormandy. The research was developed in collaboration with Dr. Peter Danielson and the Norms Evolving in Response to Dilemmas (NERD) team at the Centre for Applied Ethics at UBC. Coauthors supervised, helped interpret material, and edited drafts. The project received research ethics board approval from UBC under certificate number: H06-H0532. Check the first pages of these chapters to see footnotes with similar information.  iii  Table of Contents Abstract ......................................................................................................................................... ii Preface .......................................................................................................................................... iii Table of Contents..........................................................................................................................iv List of Tables.................................................................................................................................vi List of Figures ............................................................................................................................. vii Acknowledgements .................................................................................................................... viii 1. Introduction ...............................................................................................................................1 1.1. Background ............................................................................................................................................ 1 1.2. Public attitudes towards animal research ............................................................................................... 2 1.3. Personal and cultural characteristics ...................................................................................................... 3 1.4. Animal characteristics............................................................................................................................ 7 1.5. Research characteristics ......................................................................................................................... 9 1.6. Critique of existing methods of public attitudes assessment ............................................................... 10 1.7. Addressing the gaps: thesis aims ......................................................................................................... 12  2. Worldwide trends in the use of animals in research: The contribution of genetically modified animal models .........................................................................................................14 2.1. Introduction.......................................................................................................................................... 14 2.2. Methods................................................................................................................................................ 15 2.3. Results.................................................................................................................................................. 16 2.4. Discussion ............................................................................................................................................ 19 2.5. Conclusion ........................................................................................................................................... 21  3. Public attitudes toward animal-based research: Effects of invasiveness, genetic modification, and regulation..................................................................................................23 3.1. Introduction.......................................................................................................................................... 23 3.2. Methods................................................................................................................................................ 25 3.3. Results.................................................................................................................................................. 31 3.4. Discussion ............................................................................................................................................ 47 3.5. Conclusion ........................................................................................................................................... 51  4. Factors affecting people’s acceptance of the use of zebrafish and mice in research.........53 4.1 Introduction........................................................................................................................................... 53 4.2. Methods................................................................................................................................................ 55 4.3. Quantitative results .............................................................................................................................. 59 4.4. Qualitative results and discussion ........................................................................................................ 68  iv  4.5. Implications for policy ......................................................................................................................... 73 4.6. Conclusions.......................................................................................................................................... 73  5. The use of genetically modified animals in biomedical science: An exploration of stakeholder views ....................................................................................................................74 5.1 Introduction........................................................................................................................................... 74 5.2 Methods................................................................................................................................................. 75 5.3. Results.................................................................................................................................................. 77 5.4. Discussion ............................................................................................................................................ 85 5.5. Conclusions.......................................................................................................................................... 87  6. General discussion and recommendations ............................................................................89 6.1 Introduction........................................................................................................................................... 89 6.2. Limitations and successes of the thesis research ................................................................................. 90 6.3. Recommendations................................................................................................................................ 92 6.4. Conclusions.......................................................................................................................................... 96 6.5. Next Steps ............................................................................................................................................ 96  Bibliography.................................................................................................................................97  v  List of Tables Table 3.1. Independent variables manipulated in the experiment. ...............................................27 Table 3.2. Participant demographics ............................................................................................33 Table 3.3. An overview of codes that emerged from all stages of the qualitative data analysis. .42 Table 3.4. Results for the invasive question. ................................................................................44 Table 3.5. Results for the genetically modified (GM) animal question. ......................................45 Table 3.6. (a) Demographic variables and their effect on participant’s willingness to support animal-based research. ...............................................................................................46 Table 3.6. (b) Results from the multivariable logistic regressions for the invasive question and the GM animal question. ............................................................................................46 Table 4.1. Independent variables manipulated in the experiment. ...............................................57 Table 4.2. Participant demographics ............................................................................................61 Table 4.3. Additional questions given to participants ..................................................................63 Table 4.4. (a) Demographic variables and their effect on participant’s willingness to support ENU mutagenesis in zebrafish ...................................................................................65 Table 4.4 (b) Results from the multivariable logistic regression for the ENU mutagenesis in zebrafish question. ......................................................................................................65 Table 4.5. Quantitative results for the species treatment group. ..................................................66 Table 4.6. Quantitative results for the genetically modified (GM) animal treatment group. .......67  vi  List of Figures Figure 2.1. Trends in research animal use from 1983-2007.........................................................17 Figure 2.2. Trends in genetically modified (GM) animal species from 1983-2007.....................17 Figure 2.3. Trends in techniques used to genetically modify animals from 1983-2007 ..............18 Figure 3.1. An illustration of pathways through the questions.....................................................28 Figure
4.1.
An illustration of pathways through the questions. ....................................................56  vii  Acknowledgements I am greatly indebted to my supervisor, Dan Weary, and my PhD committee, David Fraser, Gilly Griffin, and Michael McDonald. Thanks are especially due to Cathy Schuppli, who has guided me through this work with limitless energy and kindness. Warm thanks also to Nina von Keyserlingk, Susan Cox, Peter Danielson, and Michael Burgess for graciously sharing their expertise. It is with recognized privilege that I consider you all not only my supervisors and colleagues, but my friends also. Thank you to the Canadian Council on Animal Care (CCAC), the Charles River Animal Welfare Award, the Leonard Klinck fellowship, The University of British Columbia, Genome British Columbia, and Genome Canada for providing various fellowships, scholarships, and funding. In particular, thank you to the CCAC for welcoming me to Ottawa as their 4th Research Fellow in Animal Policy Development. My time at the Animal Welfare Program has been shared with some extraordinary fellow students and colleagues – your friendship is written in every page. Thanks in particular to Chris McGill, for getting me to and from various corners of the world, often at very short notice. Appreciation also to my new colleagues and friends in Political Science and Cultural Geography who have refined and expanded my thinking – in particular, Darren Chang, Afsoun Afsahi, Laura Janara, and Gail Davies. There are a great many people outside my academic day-to-day to whom I owe gratitude for inspiring me, and for making me laugh on days when I really didn’t feel like it. Thank you to my dear friends, old and new: Travis Vakenti, Britt Gallpen, Helen Jones, Skyler Punnett, Megan Prenty, Sara Ciantar, Christopher Neufeld, Marcel Thalen, Louise Grieve, Gilly Griffin, Amelia Griffin, Jacqui Kotyk, Lauryn Kronick, Erika Doyon, Emily and Shawn Chamberlain, Jeff Hamilton, Christina David, Kim Horne, Paul Macgee, Marianne Farish, and Emma Baxter. Extra special thanks to: Mike and Veronica McGhee – I couldn’t have done this without your support, thank you for being such a wonderful part of my Canadian life. Mara Long – (my ‘anam cara’) thank you for mopping up my tears, spontaneous dance therapy, and epic travels. Scott Bell – I love and miss you, thank you for bursting my heart open and for challenging me to revolutionize my thinking and living. I also have many animal friends – Blue, Domino, Jarvis, viii  Josie, Rae, Midget Gem, Niko, Frank Zappa, Cooper, and Cloe – who all deserve special thanks for enriching my life, and for teaching me more about the complexity of human-animal relationships than any book could ever hope to. Finally, and most importantly, thank you to my amazing parents (Elizabeth and David) and my brother (Benjamin) – to whom this work is dedicated – for raising me in such a loving home, for not only allowing but actively encouraging my flights of fancy and whimsical adventures, and for laying down the roots of the person that I have become. Elisabeth - Vancouver, October 2012  ix  Dedication  For my wonderful family: Elizabeth, David, and Benjamin Ormandy.  “In teaching me independence of thought they had given me the greatest gift you can give a child, besides love, and they had given me that also” Bryce Courtenay, 1992  x  1. Introductionφ 1.1. Background Community engagement is widely acknowledged to be an important part of the governance of controversial research (Gaskell et al., 2003; Sherwin, 2001; Burgess & Tansy, 2008). Public engagement generally seeks to democratize science policy–making by expanding expert or stakeholder-driven conversations to include lay citizens (Rowe and Frewer, 2005). However, despite the attention that this issue has received, and the recognition of a need for public accountability in science generally (Mayer, 2003), little consensus has developed on how to better engage members of the public on issues related to animal-based research. According to Burgess (2012), there are several key principles that may be helpful in guiding the shift to improved public engagement on animal research. First, the principles of accountability, reasonableness and justification hold that animal use must be clearly justified and whoever makes that justification must be held accountable. Second, the principles of legitimacy, diversity and transparency require increased participant and community involvement. Accordingly, there should be greater openness and transparency of animal research. Finally, principles of constructive and dynamic deliberation require adaptive governance that is responsive to new issues, knowledge (particularly scientific knowledge) and technological advancements. As in many other jurisdictions, the current mechanism in Canada for taking into account societal values in decisions about how animals are used in research includes the requirement that a community representative is present on animal care committees (ACCs). However, ACC community representatives may feel isolated or inadequate because of a lack of technical understanding of the science; they may have been recruited for convenience rather than with a mandate to represent community values; and the chairperson may not make specific efforts to empower them to provide a community viewpoint (Schuppli & Fraser, 2007). However, it remains important to understand the diversity societal attitudes towards animal research in order to legitimize policy decisions. Given the shortcomings of relying on community representation on ACCs, other forms of public attitudes assessment, such as empirical research designs, are useful. The following chapter aims to review public-attitudes research regarding animal research, to  φ  A version of this chapter has been submitted for publication. Ormandy, E. H. & Schuppli, C. A. Public  attitudes towards animal research: A review.  1  discuss some of the major shortcomings of public-attitudes research to date, and to identify areas for improvement and further work. 1.2. Public attitudes towards animal research People’s views toward animal research range from a desire for complete abolition to strong support of the use of animals for this purpose (e.g. MORI, 2010; Gallup Poll, 2010; Eurobarometer, 2010; Hagelin et al, 2003). However, as Knight et al. (2009) point out, fundamental arguments over the use of animals in research have shifted little over time, with those involved in research (i.e. scientists, researchers) tending to frame arguments around the benefits of their work and the lack of alternatives to animal models, compared to opponents of animal use who tend to frame arguments around animal welfare and the suffering of the animals involved (Baldwin, 1993; Paul, 1995). The term ‘attitude’ has been used to refer to “the evaluation of an object, concept, or behaviour along a dimension of favour or disfavour, good or bad, like or dislike” (Ajzen & Fishbein 2000, p. 3). Attitudes are distinct from, but related to people’s beliefs and values. It is postulated in the expectancy-value model (Fishbein 1963; 1967) that attitudes are formed through a person’s accessible beliefs about an object, where a belief is defined as “the subjective probability that the object has a certain attribute” (Ajzen & Fishbein 2000, p. 4). Ajzen and Fishbein (2000, p. 4) give an illustrative example: “a person may believe that exercise (the attitude object) reduces the risk of heart disease (the attribute).” An important implication of the expectancy-value model is that attitudes towards an object are formed automatically and inevitably as we acquire new (and pertinent) information about an object’s attributes, and as the subjective values of these attributes become linked to the object (Fishbein, 1967). So, assessing people’s attitudes towards animal research can tell us more about whether different types of animal research are normatively considered ‘good’ or ‘bad’ at both a personal and societal level, as well as providing a starting place to understand the underlying beliefs on which such attitudes are formed, and to understand the behaviours that are influenced by certain beliefs and attitudes. Previous public-attitudes studies have involved the use of survey style methods, and have assumed that attitudes towards animal use are uni-dimensional (that is, an individual will hold the same attitudes towards all types of animal use) (Knight & Barnett, 2008). In addition, some studies do not disclose all the methodological details of the survey (Herzog et al., 2001), and in some cases the questions that make up these surveys are worded in biased ways, thus compromising the value of the results. 2  The following review of public-attitudes literature is presented in terms of 3 broad categories of factors known to influence people’s willingness to support animal-based research (as identified by Knight & Barnett, 2008): personal and cultural characteristics, animal characteristics, and research characteristics. The paper then goes on to discuss shortcomings associated with survey style methods in more depth. Finally, in light of this critique, the paper makes recommendations on how gaps in this growing literature can be addressed to move towards more sound models for public attitudes research and public engagement. 1.3. Personal and cultural characteristics In order to understand different attitudes towards the use of animals, and their use in research specifically, many studies have focused on personal characteristics (i.e. things about a person that may influence their decision on whether to support or oppose the use of animals in research). The personal characteristics discussed below include: age, gender, rural versus urban background, experience with animals/pet ownership, and religion. Also discussed are factors that are based more around a person’s beliefs and potentially shaped by characteristics: vegetarianism, and belief in animal mind. 1.3.1. Age It has generally been reported that moral acceptance of the use of animals in research is positively correlated with age (Hagelin et al., 2003). In their 1981 study, Kellert and Berry suggest that younger people are more opposed to animal use than older people. The authors go on to describe how older males presented a more utilitarian view towards animals, suggesting that older people tend to emphasize the practical value of animals. Some recent research (Driscoll, 1992; Furnham & Pinder, 1990, MORI, 1999) echoed this finding, but other studies have found that younger participants are more supportive of animal-based research than older participants (Schuppli & Weary, 2010). The effect of age on attitudes towards animals may be a result of a cohort effect, where people with a shared history are more likely to share beliefs and attitudes (Kendall et al., 2006), or may also be related to attitudinal changes with age (Kellert, 1996). 1.3.2. Sex Identity Sex identity has been consistently found to relate to attitudes towards the treatment of research animals (and animals in general), with virtually all studies reporting that females are more likely to object to animal use (e.g. Broida et al., 1993; Driscoll, 1992; Gallup & Beckstead, 1988; Kellert & Berry, 1981, Matthews & Herzog, 1997). A lower proportion of females accept the use of animals in research compared to men (e.g. Furnham & Pinder, 1990; Rajecki et al., 3  1993; Plous, 1996; Wells & Hepper, 1997; Navaro et al., 2001; Swami et al., 2008) and most studies of the animal protection movement have found that female activists outnumber males by a ratio of two or three to one (Herzog, 1993; Jasper & Nelkin, 1992; Plous, 1991). The effects of sex identity on attitudes towards the use of animals in research are consistent across many studies, with differences between males and females extending to at least 15 different countries (Pifer et al., 1994). Pifer (1996) reported that, among a range of predictors, sex identity was the strongest correlate of opposition to animal research. It might be that females are less supportive of animal use because they are more likely to attribute mental states to animals, and more likely to have a sympathetic reaction if they believe that animal use will cause some kind of pain or distress to animals (Knight et al., 2003). Indeed, males have been shown to present lower levels of ‘belief in animal mind’ (BAM) compared to females (Herzog & Galvin, 1997) (see later paragraph for a discussion of BAM). In addition, Kellert (1980a) reported that males exhibited more “dominionistic” attitudes towards the environment, while females exhibited more “moralistic” attitudes, a difference that might also explain differences in attitudes towards animal use. Rather than characterizing people strictly by sex identity, others have examined sex role orientation (SRO) in relation to attitudes towards the use of animals in research, (Herzog et al., 1991; Peek et al., 1997). Herzog et al. (1991) suggest that differences in attitudes are associated with feminine versus masculine SRO, with people who identify as more feminine being generally less supportive. However, Peek et al. (1997) speculate that sex differences differ not as a result of SRO, but because of the structural location of females in society (i.e. females may perceive themselves and animals to have similar positions in society; Adams, 1994). Similarly, the social position of females may also lead to greater concern for animals. For example, Kendall et al. (2006) have argued that females are primary family caretakers (and so are more likely to take on nurturing roles), and may be more likely to engage in household tasks that put them in more direct contact with animals. 1.3.3. Rural versus urban background Some studies have shown that people with a rural background have a greater acceptance of animal use than urban people, and greater support for animal experimentation (Hills, 1995; Pifer et al., 1994, Kalof et al., 1999). This finding suggests that rural and urban places provide distinct opportunities for contact and relationships with animals, as well as diverse cultural experiences that shape and strengthen people’s attitudes about animals (Kendall et al., 2006). Animal use often differs in urban and rural regions (Jasper & Nelkin, 1992). The utilitarian relationships with animals that are associated with rural locations may shape an individual’s attitudes to animals in 4  different settings, including animal research. A cross-cultural study of public attitudes to the use of animals in research (Pifer et al., 1994) found a link between a nation’s level of industrialization and urbanization and attitudes towards animal research. For instance, the two least industrialized countries within the European Community had the highest level of support for animal research. Crettaz von Roten (2012) also found differences in acceptance of animal research between European countries, with industrialized countries displaying higher level of approval of animal research than post-industrial countries. Pifer et al. (1994) suggest that countries that have a closer relationship with the land have more pragmatic and utilitarian attitudes toward animals, so using animals for human ends is not seen as a contentious issue. In developed countries urban people may never come into contact with the animals they eat; instead animals are more likely to be companions and part of the family (Jasper & Nelkin, 1992). Perhaps for this reason, urban residence has been found to be related to greater concern for animal well-being (Kellert, 1996; Ohlendorf et al., 2002; Hills, 1995). 1.3.4. Experience with animals Attitudes towards the human use of animals can also be shaped by a person’s previous or existing experience of animals (Wells & Hepper, 1997; Knight & Barnett, 2008); for example, Driscoll (1992) found that pet owners rated animal-based research as less acceptable than did nonpet owners. This finding is also echoed in other studies that showed that pet owners form an attachment with their animals, and that this strengthens a general positive attitude towards other animals (Blackshaw & Blackshaw, 1993; Furham & Heyes, 1993; Paul & Serpell, 1993; Hagelin et al., 2002). According to ‘contact theory’ (e.g. Allport, 1954), contact with members of an ‘outgroup’ (e.g. non-human animals) can lead to a mutual understanding and decreased prejudice towards that group. Contact may also allow emotional attachment and empathy towards animals to develop (Boogaard et al., 2006; Daly & Morton, 2006; Furnham et al., 2003; Serpell, 1996). This may explain why experience of animals promotes affection and positive attitudes towards animals in general, which is in conflict with utility or instrumental uses of animals, such as animal research (Serpell, 2004). Thus pet ownership, or other positive experiences of animals may increase people’s opposition to animal research. Conversely, a negative encounter with an animal may equally shape people’s view, making them more supportive of animal use (Knight et al., 2004). 1.3.5. Religion Religion can influence how people view and relate to animals. For example, Christianity has been shown to be positively associated with support for the use of animals in research (Bowd 5  & Bowd, 1989). Driscoll (1992) found differing views across different Christian denominations: persons reporting no religious affiliation or an affiliation with the Catholic church rated various examples of animal-based research as less acceptable than did persons reporting a traditional Protestant affiliation. There are, of course, also specific animal species that are either revered (e.g. cows in Hinduism) or avoided (e.g. pigs in Judaism) in different religious traditions. This may in turn affect people’s willingness to support or oppose the use of certain species for research purposes. 1.3.6. Vegetarianism and animal or environmental advocacy Demand for particular types of food is influenced primarily by social and psychological factors such as beliefs, attitudes, norms and values (Kalof et al., 1999). In western countries vegetarianism, in particular, is related to value orientations such as an increase in altruistic values and a decrease in traditional values (Dietz et al., 1995). Moreover, vegetarianism is likely to relate to a wider ideological perspective in terms of the ‘world view’ or ‘ethical ideology’ held by people (Buss et al., 1986; Furnham & Pinder, 1990; Herzog & Golden, 2009). So, rather than being a predictor of attitudes towards animals per se, vegetarianism is an action or behaviour that results from a particular attitude towards animals. This attitude may be generalized into a broader concern with animal rights, protection or welfare, due to underlying beliefs. It follows that vegetarianism has been associated with lower acceptance of the use of animals in research compared to non-vegetarianism (Furnham & Heyes, 1993; Schuppli & Weary, 2010). Studies have shown that people who are politically left-wing-oriented are less supportive of animal experimentation. This finding may also be explained by differences in people’s ‘world views’ or ethical ideologies (Buss et al., 1986; Eurobarometer, 2001; Herzog et al., 2001), because attitudes towards animals are closely related to attitudes towards other political and social matters (Furnham & Pinder, 1990). In a similar vein, concern over environmental issues (which may also be linked to vegetarianism) is negatively related to support of animal research (Broida et al., 1993). 1.3.7. Belief in animal mind “Belief in animal mind” is the term used to describe people’s belief in the mental abilities of animals. Do we believe that animals are self-aware, capable of solving problems, or experiencing emotions such as fear, sadness, happiness and pleasure? (Herzog & Galvin, 1997; Knight et al., 2003). BAM is a relatively consistent predictor of attitudes towards the human use of animals (Hills, 1995; Herzog & Galvin, 1997; Knight et al., 2003; Schuppli, 2011), and appears to explain more of the variation in people’s attitudes than personal characteristics, such as gender 6  (Knight & Barnett, 2008). BAM negatively correlates with support for animal use and positively correlates with concern for animal welfare and humane behaviour towards animals (Broida et al., 1993; Knight & Barnett, 2008), and empathy towards other humans and animals (Hills, 1995). If one believes that certain species are likely to experience internal thoughts and feelings, then subjecting them to discomfort as part of animal-based research may seem unacceptable. This line of reasoning would suggest that people should be less accepting of research using species rated highly in BAM, particularly primates. However, a study by Knight et al. (2009) showed that more support was expressed for the use of monkeys in medical research compared to other animals, such as dogs, cats, rabbits, guinea pigs, rats and mice. In this study it was scientists (rather than lay persons or animal welfarists) that indicated strong support for the use of monkeys in research. Knight et al. (2009) show that, despite attributing ‘animal mind’ to monkeys, the scientists involved in their study perceive monkeys as being appropriate animal models for medical research practice. This finding indicates that, in some cases, BAM may be trumped by other factors (such as perceived benefit or necessity of research). 1.4. Animal characteristics While most studies have focused on personal and cultural characteristics to explain variation in attitudes, factors relating to animal characteristics also influence people’s views on this subject. The animal characteristics discussed below include species, sentience, neoteny/appeal and genetic modification. 1.4.1. Species, sentience and appeal People hold different attitudes toward animal use depending on the species involved (Driscoll, 1992; Driscoll, 1995; Herzog & Galvin, 1997). People tend to rate animals classed as pets (e.g. dogs and cats) or primates as having higher mental abilities compared to other species such as fish or mice (Eddy et al., 1993; Herzog & Galvin, 1997). People are more supportive of using smaller-brained animals such as mice and rats (Eddy et al., 1993), and less supportive of using animals classed as pets (Driscoll, 1992), and animals believed to have ‘higher’ mental abilities such as tool use, problem solving, and self awareness (Herzog & Galvin 1997; Knight & Barnett, 2008). Therefore, a given person may support the use of mice and rats for dissection purposes but not support the use of chimpanzees, cats and dogs. In a recent study involving interviews with members of Animal Care Committees (who are responsible for the ethical review of research proposals involving the use of live animals), Schuppli (2011) reported that committee members were less comfortable with research using non-human primates and companion animals. 7  Different views regarding species may be due to belief in the mentality of different species as well as other factors such as, a) personal affection for particular kinds of animals or individual animals (Arluke, 1988), b) the special consideration given to certain species based on the relationship we typically have with those animals (Schuppli et al., 2004; Wells & Hepper, 1997), c) their ‘cuteness’ or attractiveness (Herzog & Galvin, 1997; Hagelin et al., 2003; Knight & Barnett, 2008), or d) where a species falls on the phylogenetic scale (Hagelin et al., 2000). From literature on public attitudes towards species conservation, it has also been shown that animals that retain a neonatal appearance (neoteny) are more likely to be supported in conservation efforts (Batt, 2009; Gunnthorsdottir, 2001). Attitudes towards the use of different species in research may also change as we learn more about animal behaviour and welfare. For example, recent research suggests that fish (that are often considered an acceptable replacement for mammals in research; CCAC, 2005a; DeTolla et al., 1995; Fabacher & Little, 2000) have the capacity to feel pain (Braithwaite & Huntingford, 2004; Chandroo et al., 2004). 1.4.2. Genetic modification Public views towards the genetic modification of animals tend to be complex, but predominantly negative (Birke et al., 2007). Some members of the public express grave concern for the ‘unnaturalness’ of genetic modification and its potential to lead to unknown consequences (Macnaghten, 2001; Eurobarometer, 2001). In his 2001 study, Macnaghten found considerable concern about the genetic modification of animals and the uses to which genetically modified (GM) animals might be put. Participants in his focus-group study showed a “reaction against the proposed technology as intrinsically a violation of nature and transgressive of so-called natural parameters” (Macnaghten, 2001: p.25). Another primary concern that has emerged is that genetic modification will lead to unexpected (and potentially bad) consequences; indeed one aspect of the unease about GM animals is a fear that nature might ‘bite back’ (Macnaghten, 2004; Birke et al., 2007). In addition to these main arguments in opposition to genetic modification technology as applied to animals, a more recent study by Macnaghten (2004) shows an emerging concern from the public about the increase in the numbers of animals used in research due to the currently inefficient and unpredictable nature of the genetic modification process. This sentiment also emerges in a recent paper by Schuppli et al., (2004) who argue that the creation and use of GM animals challenges the Three Rs (replacement, reduction, refinement), particularly reduction.  8  1.5. Research characteristics The characteristics of the research that an animal will be involved in can also influence people’s decision about whether to support or oppose the research. The research characteristics discussed below are: the purpose of the research, the level of invasiveness (or harm) that the animal will experience, and the availability of non-animal alternatives. 1.5.1. Type of research One consistent theme in public-attitudes research is that participants often draw distinctions between different types of animal use in laboratories (Birke et al., 2007). It is very common that the use of animals in medical experiments is deemed more acceptable than the use of animals in cosmetics testing. For example, Aldhous et al. (1999) found that whether or not mice were subjected to pain, illness, or surgeries, people were more likely to disapprove if the experiment was designed to test the safety of a cosmetics ingredients than if it tested the safety and effectiveness of a drug or vaccine, and this result is echoed in other studies (Kane, Parsons & Associates, 1989; as cited in Pifer et al., 1994; Driscoll, 1992; Wuensch & Poteat, 1998; Knight & Barnett, 2008; Schuppli 2011). However, the use of animals in medical experiments is not always preferred. Schuppli and Weary (2010) found that participants in an online engagement study were more supportive of the use of pigs in environmental research (to reduce agricultural pollution) than in biomedical research (which aimed to decrease rejection rates in organ transplantation). However, in this case the purpose of the research may be trumped by other factors. For example, non-animal alternatives to the biomedical research scenario used in the study by Schuppli and Weary (2010) (e.g. increasing human organ donations) may be seen as a more viable option. It would appear that people’s attitudes towards experiments involving animals are likely to change depending on the beneficiary, purpose, or necessity of the research. As noted by Henry and Pulcino (2009), “the literature suggests that animal research that is viewed as providing tangible, meaningful benefits to humans is considered more acceptable than animal research that is viewed as less beneficial or necessary.” 1.5.2. Availability of alternatives The necessity of animal research ties into the availability of non-animal alternatives, with research that is deemed unnecessary being less favoured. For example, Stanistreet and Spofforth (1993) found that participants were less supportive of the use of animals in research that was viewed as “non-necessary” than research that was viewed as “necessary.” It seems that the availability of non-animal alternatives, or a belief that alternatives exist, may be particularly influential on people’s attitudes towards the use of animals in research (e.g. Hagelin et al., 2003). 9  Two studies in particular illustrate that when non-animal alternatives are available, there is a higher level of opposition. Research by Knight et al. (2003), showed that animal use was most likely to be supported when participants perceived there to be no other choice than using animals. However, Knight et al. (2003) also found that their participants (nine men, eight women) could seldom think of alternatives for animals in research and in teaching and so they believed that there was little choice other than using animals. In a follow-up study, Knight et al. (2009) showed that different attitudes towards animal experiments between scientists and animal welfarists could, in part, be explained by differing beliefs about the availability of non-animal alternatives. 1.5.3. Level of harm Invasiveness, or level of harm that the animals experience during a given experiment has also been shown to influence people’s support of animal-based research (Wells & Hepper, 1997; Plous, 1996). Richmond et al. (1990) found that the most common objection to animal experimentation is related to whether animals experience pain and suffering. In fact, a review by Hagelin et al. (2003) illustrated that survey respondents are less likely to support animal research if the words “pain” or “death” are used. In a more recent study (Henry & Pulcino, 2009), results indicated that participants were more opposed to biomedical research that resulted in harm to the animals. 1.6. Critique of existing methods of public attitudes assessment There is a growing body of literature related to public attitudes towards animal use in general, and animal research more specifically. However, there are potential shortcomings that should be addressed for future studies. Three primary shortcomings are discussed below: 1) use of college students as participant samples, 2) use of general questions about ‘animal use’ rather than specific questions about different types of animal use (or even different types of animal research), and 3) use of Likert scales or rating scales that do not allow for more qualitative reasoning. There may be different motivations for conducting research on public attitudes to animals. Some of the studies cited in this review clearly aimed to explore the attitudes of select groups of people, for example veterinary and medical students (Hagelin et al., 2000) or psychology students (Furnham & Heyes, 1993; Plous, 1996). Exploring the attitudes of such student groups may be of particular importance give that these individuals are likely to use animals in research at some point in their career. However, the primary aim of this thesis is to inform animal policy, as such the information provided by studies that restrict attitudes research to student groups is limited.  10  Gallup and Beckstead (1988), Sieber (1986), and others, used undergraduate psychology students to make up their sample for survey or interview studies. In fact, one paper (Herzog & Dorr, 2000) examined 15 issues of Society and Animals published between 1993 and 1998. The authors reported that, “the data in 11 of these articles were obtained using undergraduates. Of these, one article did not specify the source of the students, one used education students as subjects and the other nine were based on students taking psychology classes” (p. 2). Notably, using a large national sample, Kellert (1980b) and Kellert and Berry (1981) reported that both education and age were related to knowledge and attitudes towards animals. This suggests that college students, being both young and educated, are likely to be more concerned about animals than the general public. Given that the regulation of animals in research was developed, in part, in response to public concerns, it is pertinent that public-attitudes research reflects a diversity of views, rather than limiting the breadth of studies by using convenience sampling of students, or sampling other select groups. As further pointed out by Herzog and Dorr (2000), “undergraduate psychology majors are a narrow source of information on human/animal relationships” (p. 2). This is echoed in a recent article in the Economist (2012), which highlights the challenges to using undergraduate students as a source of information, and explores the benefits of crowdsourcing (e.g. the use of Mechanical Turk to recruit survey participants). The primary benefit to crowdsourcing is the diversity of participants: there is less reliance on information provided by participants from western, educated, industrialized, rich, and democratic subsets of the world population. A second shortcoming is that most studies have asked rather general questions about ‘animal’ use. However, studies by Kellert and Berry (1981), Driscoll (1992) and Knight et al. (2003) illustrate that people have strong likes and dislikes for different kinds of animals, and multidimensional views regarding different types of animal use. To ask someone to agree or disagree with the statement such as “it is alright to do research on animals” is ambiguous. It may be that only people with more extreme views will disagree with this statement because it does not specify what kind of research, or perhaps more importantly, what kind of animal is involved. Research animal use is changing, particularly as a result of increasing use of technologies such as genetic modification (via pronuclear microinjection or other more targeted techniques) and ethyl-N-nitrosourea (ENU) mutagenesis (a commonly used method of chemically inducing mutations, particularly in mice – de Angelis et al., 2000, and zebrafish – de Bruijn et al., 2010) to create animal models of disease. So far, research exploring public attitudes to the genetic modification of animals has mostly focused on farm animals, rather than laboratory animals that 11  are used in much greater numbers. Despite the prevalence of ENU mutagenesis, no research to date has explored people’s attitudes to this procedure. In addition new developments in areas of personalized medicine, particularly oncology, may pose new challenges. For example, a patient with a tumour may be able to have tumour samples taken and implanted into animal hosts (e.g. mice) so that a range of treatments can be tested, and a better targeted therapeutic treatment for the patient developed (M. Bally, personal communication). Such procedures will likely increase animal numbers and may also require alterations to the current process of animal protocol review and approval, as well as perhaps introducing a more personal, direct involvement in the public’s role in animal use. A third shortcoming is that many of the studies cited above were performed using methods that asked participants to respond on a scale (e.g. Likert scale, rating of preference scale), or asked questions requiring a simple “Yes” or “No” response, without any insight into the reasoning that led to these responses. Participants are constrained in their choice of answers by the limited options that are provided by the researcher (which may lead to research bias) (Cummins & Gullone, 2000), and are unable to provide any qualification to explain their response. The exploration of people’s reasons for their “Yes/No” or Likert scale responses is important. Restricted response options do not allow for consideration of what people’s concerns are (e.g. why they might be opposed to certain types of research), thus making it difficult for policy makers to make progress in addressing societal concerns. 1.7. Addressing the gaps: thesis aims Changes in societal attitudes and opinions often result in a push to improve animal-related regulation and public policy (Kirkwood & Hubrecht, 2001). However, mechanisms for including public opinion in animal research policy are lacking. One recent article highlights the secrecy surrounding animal research (Holmberg & Ideland (2010), while another (Lyons, 2011) draws attention to some of the problems that may be encountered if decisions about animal research are not opened up to a wider community. A case study by Lyons (2011) warns against the formation of policy communities with exclusive membership that “tend(s) to produce outcomes that consistently favour network members at the expense of excluded groups” (Lyons, 2011, p. 357). In the article, Lyons described a specific area of research in the UK (xenotransplantation between pigs and primates) in which, to the detriment of the animals involved, decisions were made without input from experts or stakeholders outside the policy community, and without wider public engagement. Such activities buck the current trend towards democratization of science and 12  science policy (Irwin, 2001; Elam & Bertlisson, 2003; Schilele, 2008), and highlight the need for wider expert and public engagement, especially for research that is considered to be contentious. Therefore, it is important to assess public opinion about animal-based research, and to engage a variety of different stakeholders, including the public, when developing animal policy. One approach to improving public engagement on animal research is to conduct empirical studies that explore public attitudes towards animal research in ways that correct for some of the criticisms outlined previously in this paper. For example, studies that, 1) avoid reliance on convenience sampling of students and ensure that participants reflect a diversity of views, 2) use a well-planned experimental framework that allows exploration of not only where people draw the line in terms of what they are willing to accept, but also why, and 3) focus on gaining a better understanding of public attitudes towards specific (rather than general) aspects of animal research. For example, attitudes towards emerging technologies (like genetic modification or other genetic alteration techniques) and the most commonly used species used in research (zebrafish and mice), as well the regulatory systems that oversee animal research. This thesis aims to inform animal research policy by contributing to knowledge about societal values. Chapter 2 explores the changing patterns in animal use in order to highlight key areas on which to focus empirical public attitudes research. Chapters 3 and 4 describe two online engagement experiments investigating how acceptance of animal-based research is affected by genetic modification, regulation, invasiveness, and the species used. Chapter 5 develops the public attitudes research further and describes an interview study that examined people’s views toward the creation and use of genetically modified animals in biomedical science. Chapter 6 synthesizes the findings from the four research chapters, and concludes with key policy recommendations: 1) improve scientific reporting, 2) improve data and animal sharing, 3) improve recording of national animal statistics, 4) improve animal welfare assessment, and 5) supplement the Three Rs.  13  2. Worldwide trends in the use of animals in research: The contribution of genetically modified animal modelsφ 2.1. Introduction Russell and Burch introduced the Three Rs – reduction in the number of animals required to gain the same amount of information, replacement of animals with inanimate systems or less sentient species, and refinement of procedures to reduce animal suffering and enhance animal welfare – as guiding principles for humane use of research animals in 1959 (Russell & Burch, 1959), but few data are available to evaluate Three Rs implementation. Indeed, there is no standard method of reporting statistics on animal use from country to country. For example, the UK reports animal procedures when they are started (‘prospective reporting’) (Home Office, 2000). Others, such as the Netherlands, report procedures when they are finished (‘retrospective reporting’) (FELASA, 2007). Yet others, such as Switzerland, report animals ‘in use’ during a given year (FELASA, 2007). However, it is worth noting that retrospective reporting will be required across Europe with the implementation of the new EU Directive (EU Directive, 2010). There is also variation in the types of animals documented; most notably the United States does not include mice, rats, birds, amphibians or reptiles when reporting research animal numbers (USDA, 2006). Countries such as Canada (CCAC, 2005a) and the United Kingdom (Home Office, 2000) have been documenting detailed annual animal statistics from as early as 1975, whereas other countries, such as Australia (Australian Association for Humane Research, 2004) and Norway (Norwegian Animal Research Authority, 2005) were not doing so until recently. Together, this variation makes it impossible to use national records to meaningfully assess worldwide trends in the use of research animals. Bibliometrics is the scientific and quantitative study of existing published literature (Narin et al., 2004) and is used to compare individuals, institutions and countries in terms of scientific production. In this study a novel bibliometric approach was developed to bypass variation in reporting of research animal statistics between countries, and to document detailed aspects of worldwide trends in animal use, such as proportions of animals used, species (e.g. mice, rats, zebrafish etc) and strains (e.g. C57Bl/6 mice, Wistar rats etc). The analysis paid particular  φ  A version of this chapter has been published. Ormandy E. H., Schuppli, C. A. & Weary, D. M. (2009)  Worldwide Trends in the Use of Animals in Research: The Contribution of Genetically Modified Animal Models. Alternatives to Laboratory Animals, 37: 63-68  14  attention to the contribution of genetically modified animal models to worldwide trends, including the type of genetic modification procedure, such as pronuclear microinjection or gene-targeting. 2.2. Methods The journals were selected using the following criteria: a) general, high-impact journals (as opposed to specialist, lower impact ones) (Lewison, 1996) – ISI journal citation reports were used to select journals with impact factors above 20 (ISI Web of Knowledge, 2008); b) journals that publish articles from institutes worldwide; c) journals that publish original research with clear materials and methods; d) journals that publish articles that report the direct use of animals; e) journals that publish articles that report the use of genetically modified animals; f) wellestablished journals that date back to the birth of genetic modification technologies involving animals, around 1980. Science (impact factor: 30.0), Cell (impact factor: 29.2), Nature (impact factor: 26.7) and Nature Biotechnology (impact factor: 22.7). were the only journals that fit all criteria as listed above. Original research articles between 1983-2007 were sampled. This time frame was chosen to reflect the birth of genetic modification techniques: the first transgenic mouse was reported in 1980 (Gordon et al., 1980) From each article: country of the corresponding author, species (both vertebrates and invertebrates) and strain used were recorded. Using existing definitions (FELASA, 1995; CCAC, 1997; Olson & Sandøe, 2004) a genetically modified (GM) animal was classified as one that has been genetically altered – via the integration of foreign genes, or the deletion, modification or altered expression of genes that are already present – to have specific characteristics it would not otherwise have. For those articles involving the direct use of genetically modified animals background strain, genetic modification method (for example, pronuclear microinjection) and resulting genetic modification (for example, gene ‘knock out’) were recorded. The original intention was to collect detailed data on the number of individual animals used per article, however, the majority of articles in the sample that reported the use of smaller species, such as mice and rats, did not describe the numbers of individual animals used. This contrasts with articles that reported the use of larger species, such as pigs and sheep, in which numbers of individual animals used were clearly documented. Due to this lack of consistency the numbers of articles using animals were reported, rather than numbers of animals used.  15  Data were collected from 2 issues per year (April and October). Each issue contained a varying number of original research articles ranging from 6 to 21, and in total data was collected from 2691 articles. 2.3. Results The articles sampled that involved animal use (n=734) had corresponding authors from 24 countries, with the majority of articles from the United States (65.5%). Other high-output countries in our sample were: Japan (6.1%), the United Kingdom (5.3%) and Germany (5.1%). On examining 2691 original research articles published between 1983 and 2007 in the 4 highest impact journals in the biosciences (Nature, Science, Cell, and Nature Biotechnology), a decline in the number of articles reporting use of live non-human vertebrates from 1983 to 1992 was found. The numbers of articles remain relatively steady between 1992 and 2000, followed by a rise in use from 2000 to 2007 (Fig. 1). Mice (49.3%) were the most common species used, followed by rats (15.8%), nonmammals (14.4%), farm animals (7.6%), small mammals – excluding rats and mice (6.9%), primates (4.3%), and companion animals (1.6%). The reported use of invertebrates has remained consistent at an average annual level of 7.6% (minimum 3.0%; maximum 15.8%). The percentage of articles sampled that reported GM animal use increased from 3% to 20% between 1992 and 2007. In contrast, the percentage of articles reporting the use of non-GM animals decreased from 20% to 10% over the same period. Mice (91.4%) were by far the most common species used for genetic modification, followed by rats (3.6%), zebrafish (2.3%), pigs (1.4%), and other species including chickens, sheep, and cows (1.3%). The percentage of articles reporting the use of mice rose from 2.7% in 1992 to 17.3% in 2007 (Fig. 2). The most commonly reported mouse strains used for genetic modification procedures were C57Bl/6 mice (48.1%), followed by 129Sv (11.1%), Balb/c (4.3%), CD1 (2.5%), and FVB (0.3%). Reported use of gene-targeting techniques has increased from 0% in 1992 to 15% in 2007. However, over the same time period the use of random integration techniques has remained consistent at an average annual level of 3.3% (minimum 1.4%; maximum 6.8%) (Fig. 3).  16  Figure 2.1. Trends in research animal use from 1983-2007 As reported in Science, Cell, Nature and Nature Biotechnology. Shaded area shows the number of articles reporting the direct use of live non-human vertebrates as a percentage of total articles (n = 2691). Dotted line shows the percentage of articles that reported the use of non-GM vertebrates. Solid line shows the percentage of articles that reported the use of GM vertebrates. All data are expressed in relation to publication date.  Figure 2.2. Trends in genetically modified (GM) animal species from 1983-2007 As reported in Science, Cell, Nature and Nature Biotechnology. The number of articles reporting GM mice (open circles), rats (open squares), zebrafish (open diamonds) and pigs (open triangles) expressed as a percentages of all articles sampled (n = 2691), in relation to publication date.  17  Figure 2.3. Trends in techniques used to genetically modify animals from 1983-2007 As reported in Science, Cell, Nature and Nature Biotechnology. The number of articles reporting GM animal use (n = 220) that describe random integration (solid line) and gene targeting (dotted line) procedures, in relation to publication date.  18  2.4. Discussion The majority of the articles were published by authors in the United States, where limited animal use statistics are collected, and Japan, where no animal use statistics are collected. The remaining top countries, the United Kingdom (UK) and Germany, are required to record annual animal statistics under EU Directive 2010/63/EU (EU Directive, 2010). Indeed the UK exceeds the requirements of the Directive and collects the most detailed data on research animal use. However, as the Directive does not specify a common format, nor the level of detail for the data collection, animal use statistics from European countries cannot be meaningfully compared. The data we have provided here addresses these challenges and meets the aim of bypassing the variation in the annual reporting of annual animal statistics in order to map trends in research animal use since 1983. The bibliometric method developed here has several limitations. The method can only be used to assess trends in those animals used in published research, and this does not generally include those used in teaching, testing and breeding. The bibliometric method used focused on high-impact journals and so only provides a small subset of all research animal use. Most importantly, the approach shows how the number of studies using animals has changed over time, and does not directly assess the numbers of animals used. Some data are available on the number of animals used, such as those reported by the UK Home Office (Home Office, 2000), and these data show very similar historical patterns to those we have reported here. The UK statistics also separate GM from non-GM animals, and they too have shown an increase in GM animals and a decrease in non-GM animals, with overall animal use being on the increase. The Canadian Council on Animal Care records animal use using categories of invasiveness, with category ‘A’ being the least invasive and category ‘E’ being the most invasive. They record all GM animals as category ‘D’, and have shown that the numbers of animals in category ‘D’ have risen in recent years (CCAC, 2005b) suggesting an increase in the use of GM animals. A recent study (Taylor et al., 2008), using UK annual statistics from 1999 to 2005, shows a strong relationship between number of papers published and the number of animals used per year (R2=0.91). This relationship indicates that bibliometric measures, such as those used in the current study, can provide a reasonable indicator of research animal use. As the data collection described began with articles published from 1983 (this date corresponds with the first available genetically modified animal models), the proportion of studies including GM animals could only increase. More interesting is the rise in overall animal use, and 19  the fact that this increase appears to be due to the use of GM animal models. The effect that GM animal models have on pushing up animal numbers has sparked public concern over the use of GM technology (Macnaghten, 2004). In addition, this rise in animal use clearly runs counter to the principles of Replacement and Reduction, so the increasing use of GM animals in research may also raise additional concerns for the Three Rs. Several factors - the number of animals required to generate new GM lines and both unpredictable and unintended effects on the animal genome – may increase the risk of welfare problems and pose particular challenges to the principle of Reduction (Buehr et al., 2003; Schuppli et al., 2004). Many of the embryos that undergo genetic modification procedures do not survive and of those that do survive only a small proportion - between 1-30% - are genetically modified (Robinson et al., 2003). This inefficiency requires large numbers of ‘founder’ animals in order to produce GM animals that are of scientific value. This in itself thwarts efforts to implement Reduction. Since governing bodies do not typically collect numbers of founder animals or GM animal breeding colonies, current estimates of animal use based on annual statistics are conservative. Furthermore, many animals are exposed to potentially harmful yet routine procedures that are specific to genetic modification methods. For example, during the creation of GM animals, oocyte and blastocyst donor females may be induced to superovulate via intraperitoneal or subcutaneous injection of hormones, genetically modified embryos are surgically implanted to female recipients, males are surgically vasectomized to induce pseudopregnancy in female embryo recipients, and all offspring need to be genotyped, typically from tail biopsies or ear notching (Dennis, 2002; Wells, 2002; Brown & Corbin, 2002). The genetic background of a strain can have a profound effect both on the genetic modification procedure and the resulting phenotype (Yoshiki & Moriwaki, 2006; Barthold, 2002), perhaps explaining the preference for certain strains. Using a consistent background strain helps reduce variability in what can be an unpredictable procedure. However, due to the strong influence of the genetic background, researchers are currently advised to carry out genetic modification on mixed background strains (e.g. C57Bl/6 x 129Sv) (Sanford et al., 2001), which can also play a part in increasing the number of founder animals required in order to generate background strains. GM animals can be generated in a variety of ways. The most common methods are: a) direct pronuclear microinjection of a foreign DNA construct (a transgene) (Gordon et al., 1980), b) transfection with retroviral or episomal vectors (Orwig et al., 2002), and c) gene targeting 20  techniques using homologous recombination in embryonic stem (ES) cells (van der Meer et al., 2001). Genetic modification by any of these techniques does not necessarily result in reduced welfare for the animals involved. However, genetic modification techniques that involve random integration (i.e. pronuclear microinjection) are known to be relatively unpredictable and inefficient, primarily due to limitations in the control of the integration site of foreign DNA (van der Meer et al., 2001). In such cases, several lines of GM animals that differ only in the integration site will be generated (Verbeek, 1997) thus increasing the numbers of animals involved. More precise methods such as gene targeting show promise in both increasing the predictability and efficiency of genetic modification, thereby reducing animal numbers and refining the techniques to minimize potential harm. Gene targeting techniques, such as gene knock-out using the Cre-loxP system, have evolved to allow greater control of DNA integration, however there remain unexpected interactions of the introduced DNA with host genes (Yoshiki & Morikawi, 2006). Interfering with the genome by inserting or removing fragments of DNA may result in alteration of the animal’s normal genetic homeostasis (Costa, 1997), which can affect the behaviour and well being of the animals in unpredictable ways. The data presented here show that although the use of gene targeting techniques is increasing, a substantial minority of studies continue to use random integration methods, that are less efficient and predictable. Transitioning from the use of random integration methods to gene targeting techniques provides an opportunity for refinement. Although complete substitution of research animals with human tissues or computer models is the ultimate goal of Replacement, some people consider that the Replacement principle has expanded to include replacing sentient species with those that are considered less sentient; for example, replacing the use of mice with invertebrates (Gauthier & Griffin, 2005). However, our data reveal that while invertebrate use has remained consistent, use of vertebrates has increased over the past 15 years. 2.5. Conclusion The worldwide trends in research animal use described in this paper highlight the rising use of GM animals. This directly challenges the goals of the Three Rs, particularly reduction. Key recommendations are to encourage policy makers to collect breeding animal numbers so that more accurate animal statistics can be recorded and assessed over time, and to promote a shift towards more targeted and efficient genetic techniques. The intention is that the patterns described here 21  will provide a basis for informed dialogue on the use of animals in research. Using this research as a starting point, the following two chapters explore public attitudes towards the use of animals in research, and in particular, the how genetic modification, regulation, level of invasiveness, and the species of animal used affect people’s acceptance.  22  3. Public attitudes toward animal-based research: Effects of invasiveness, genetic modification, and regulationφ 3.1. Introduction In response to people’s concerns over the use of animals in research, many countries have introduced regulatory requirements, or other forms of oversight, to control how animals are used. For example: Canada uses a quasi-regulatory voluntary compliance system under the guidance of the Canadian Council on Animal Care (CCAC) (CCAC, 2011), the United Kingdom uses a legislative system via the Home Office three-tier licensing program and the Animals (Scientific Procedures) Act 1986 (Walsh & Richmond, 2005), and the United States uses a partly legislated and a partly accreditation-based system via a combination of the Animal Welfare Act, Public Health Service policy and Association for the Assessment and Accreditation of Laboratory Animal Care (AAALAC) (Hamm et al., 1995; Bayne, 2008). Regulatory systems are claimed to operate under a utilitarian framework, whereby some attempt is made to evaluate the costs and benefits of proposed research and a project is only allowed to proceed if the net perceived benefits of the research outweigh the net costs to animal welfare. Commonly stated goals of regulation include maintaining public accountability (Orlans, 2000; Griffin et al., 2007) and public acceptance of research practices involving animals (Nuffield Council on Bioethics, 2005), with regulation aiming to “reconcile the needs of science with the just claims of humanity” (Hampson, 1989, p. 100). It follows that since the public often funds animal-based research (directly or indirectly), and are often listed as potential beneficiaries of the results, understanding people’s attitudes about animal research is important for regulatory systems to maintain public accountability. People’s attitudes can influence their moral reasoning. For example, Knight and Herzog (2009, p. 454) argue that, “ understanding attitudes toward the use of animals for human benefit may relate to wider knowledge of attitudes, action, and emotion, and also moral decision making processes.” People’s attitudes regarding the acceptability of animal-based research vary widely, ranging from complete acceptance to complete opposition (Eurobarometer, 2010; Gallup Poll, 2010; MORI Poll, 2010). In addition, people’s attitudes may change over time - for example the recent MORI Poll (2010, p. 51) indicates that the percentage of people in Great Britain who φ  A version of this chapter has been accepted for publication. Ormandy E. H., Schuppli, C. A. & Weary, D. M. (in press) Public attitudes toward the use of animals in research: Effects of invasiveness, genetic modification, and regulation. Anthrozoös  23  strongly agree that the “The Government should ban all experiments on animals for any sort of research” has declined from 16% in 1999 to 7% in 2009. Attitudes may also change as core beliefs change (Knight et al., 2010). Previous studies on people’s attitudes have identified elements of animal-based research that affect acceptability, including: (a) purpose of the research and its scientific merit (Pifer et al., 1994; Knight & Barnett, 2008); (b) species of animal used (Driscoll, 1992; 1995); (c) level of invasiveness or harm that animals are exposed to during research (Richmond et al., 1990); and (d) the use of genetically modified (GM) animals (Macnaghten, 2004; Schuppli & Weary, 2010). Typically, increased invasiveness results in a decrease in acceptance of animal-based research (Richmond et al., 1990; Hagelin et al., 2003). However, people’s views can be nuanced, with evidence of ‘conditional acceptance’ (Macnaghten, 2001; Knight et al., 2003). For example, people may accept invasive research provided that certain conditions are met, such as the use of analgesia and anaesthesia, the use of humane endpoints, and high standards of animal care (Gallup & Beckstead, 1988; Plous, 1996). Typically, conditions such as these are met by adherence to Russell and Burch’s (1959) Three Rs – reduction, refinement and replacement – which are typically implemented via regulations or other policies and guidelines that oversee animal-based research (Fenwick et al., 2009). Several studies have shown that support for either the creation or use of GM animals may be low (Macnaghten, 2001, 2004; Gaskell et al., 2003a; Schuppli & Weary, 2010). Participants reported that genetic modification of animals is inherently wrong or goes against nature, or alternatively they expressed fear that genetic modification may lead to unexpected and unpredictable, bad consequences (Macnaghten, 2001, 2004; Gaskell et al., 2003a). Concerns have also been raised regarding the numbers of animals required to generate GM animal models (Macnaghten, 2004). In addition, a series of Eurobarometer studies indicated that “trust in [the] government in making regulations on biotechnology” is low (Gaskell et al., 2003a, p. 15). Despite evidence of low levels of acceptance regarding GM animals, the use of GM animal models in research has increased rapidly over the past 15 years (Ormandy et al., 2009). This increase in animal use suggests that common practices in research are falling out of step with public values, and challenges regulatory bodies to ensure the relevance of their regulations. The aim of the current study was to examine how regulation, invasiveness and genetic modification affect people’s acceptance of the use of animals in three specific research protocols. Based on research examining people’s attitudes to invasiveness and genetic modification, three predictions were proposed. First, that support for the use of animals in minimal risk research (non24  invasive, and not using GM animals) will be high, even if research is unregulated. Second, that unregulated animal-based research will be less acceptable if it involves invasive techniques, but acceptance of invasive research will increase if the research is regulated. Third, that unregulated animal-based research will be less acceptable if it involves the use of GM animals, and that participants will continue to oppose the use of GM animals in research even when regulation is in place. This third prediction is based on the moral opposition to genetic modification that has been documented in previous literature on this topic. Such views tend to be less malleable, and less likely to change in response to the addition (or removal) of regulation. To test these predictions we used an online public engagement experiment. Participants were presented with different research scenarios to examine three independent variables: (1) regulation (unregulated or regulated research), (2) invasiveness (feeding trial with or without surgically implanted fistulas), and (3) genetic modification (feeding trials with or without GM animals). Each treatment dichotomizes what in reality is wide range of options. For example, there is a variety of oversight systems in different countries that govern the use of animals in research; the specific examples used in this experiment provide hypothetical situations for participants to consider. According to Christensen and Gomilla (2012), asking people to contemplate hypothetical scenarios can provide valuable insight into the processes that underlie human moral judgments and decision-making. Although most participants live in countries where some form of regulation exists, asking them to think about animal research that is regulated versus unregulated will help us to understand what aspects of regulation do and do not satisfy public trust. 3.2. Methods 3.2.1 Experimental design Participants (n = 681) were randomly assigned to either an unregulated or regulated scenario (Table 3.1). Our definition of regulation was fashioned from common components of the frameworks used in Canada, the United Kingdom, and the United States. Specifically, the regulated treatment group involved research that: a) is subject to independent ethical review; b) is subject to third party facility inspection; and c) requires accurate reporting of research animal numbers. This definition was presented to each participant when they were allocated to the regulated treatment group. The experiment used a contingency design: how participants answered each question determined the next question they received. This design resulted in a total of 14 possible paths 25  through the various questions (Figure 3.1). The experiment was designed so that questions followed a sequence from minimal risk to higher risk with regard to invasiveness and genetic modification. For example, the first question presented a non-invasive, non-GM research scenario, and asked participants whether they were willing to support the use of 100 pigs, fed two different natural grain diets, in an experiment designed to reduce phosphorus pollution in commercial agriculture. Participants who responded, “Yes” to this first question were then randomly assigned to either a question presenting invasive research or a question presenting the use of genetically modified (GM) animals, and again asked about their support. The invasive question again proposed the use of 100 pigs, but this time pigs were surgically implanted with a fistula to allow researchers to examine the nutritional effects of the experimental diets that were previously proposed; the genetic modification question also proposed the use of 100 pigs, but these animals were genetically modified to better digest dietary phosphorus. These questions were based upon real research scenarios: the invasive question was developed around the research of Guillot et al. (1993) in which pigs were surgically implanted with a duodenal fistula; and the genetic modification question was developed around the EnviropigTM – a GM pig that has been developed to have reduced phosphorous output (Golovan et al., 2001).  26  Table 3.1. Independent variables manipulated in the experiment.  Variable  Regulation  Levels / Treatment groups  Research scenario (text as it appeared to participants)  Unregulated  The research is unregulated. That is to say there are no formal regulations to control or oversee how animals are used in research.  --------  --------  Regulated  Non-Invasive  Invasiveness  --------Invasive  Non-GM  The use of animals in research is regulated. Attributes of this system include: - Regulation that aims to promote ethical use of animals - Formal review of a) research team members, b) the research facility and c) experimental proposals by an independent panel including scientists and members of the general public - Accurate reporting of numbers of animals used - Random and routine inspection of the research facility by a third party The researchers propose feeding two different natural grain diets to 100 pigs. These diets are predicted to reduce phosphorous in pig manure. Excess phosphorous is an important cause of agricultural pollution. -------To determine the effects of these diets on pig digestion and nutrition, the researchers now propose surgically implanting a tube (fistula) into the side of the pig. The tube will provide researchers easy access to the small intestine so that they can collect digestive fluid from the pigs. The researchers propose feeding two different natural grain diets to 100 pigs. These diets are predicted to reduce phosphorous in pig manure. Excess phosphorous is an important cause of agricultural pollution. --------  Genetic Modification  -------GM  In a follow-up study, the researchers have proposed using the identical procedures (i.e. 100 pigs fed two different diets), but this time using pigs that have been genetically modified to reduce the amount of phosphorous in their manure.  27  Figure 3.1. An illustration of pathways through the questions. Participants were first allocated to Unregulated (Panel A) or Regulated (Panel B) scenarios. ‘R’ indicates random assignment of the participant to either the invasive question or the genetically modified (GM) animal question.  A  B  28  Participants starting in the unregulated treatment, who responded “No” to any question, were asked the same question again but this time with regulation in place. Participants who were initially assigned to the regulated treatment, and who responded “Yes” to any question, were asked the same question again, but this time with regulation removed. In addition to “Yes” or “No” answers, participants were asked to provide reasons for their answers. This field was made mandatory, so text input was required to complete each question and move on to the next. Participants who did not want to comment could leave an “x” in the text field, but most respondents (93%) provided substantive comments in addition to their “Yes” and “No” responses. The average comment was approximately 25 words, and ranged from 3 to 60 words in length. Participants were also asked a series of demographic questions (e.g. age, sex identity, diet preferences, political stance, etc.; Table 3.2) that previous studies (e.g. Hagelin et al., 2003) have shown to be relevant in influencing people’s attitudes towards animals, and towards animal-based research. 3.2.2. Pre-testing The experiment was tested several times for clarity of content and technical errors before being published online. Alpha testing (for spelling errors and computer glitches) was done within the research team. Beta testing (for clarity of content and flow of the questions) included the responses of 15 people who were outside the research team and unfamiliar with the experiment prior to testing. These beta testers took part in the experiment as if they were genuine participants, however, the data they generated were not included in the final analysis due to changes that were made after their feedback. The level of readability for the questions was assessed as grade level 8.3 and reading ease 59.8% according to the Flesch-Kincaid Grade Level Score (Friedman & Hoffman-Goetz, 2006). Reading scores are intended to indicate the comprehension difficulty of academic English. These scores mean that the questions should have been understandable by people 13 years and older. 3.2.3. Recruitment Participants (age 19 years or older) were recruited for the experiment using Facebook: an online social networking website. Four Facebook groups were specifically targeted on October 13th 2009: an animal advocacy group (1,289 members), an anti-vivisection group (2,046 members), a pro-research group (4,106 members), and an environmental advocacy group (8,020 members). Assumptions were made about the characteristics of these different stakeholders, so 29  participants were targeted for recruitment based on these assumptions (purposive sampling) (Tashakkori & Teddlie, 2003). In addition, a Facebook event page was created, and people were invited to be part of the event. The event page, containing a link to the experiment, encouraged people to participate and to invite others; a total of 892 people were added to the event guest list. Sampling in this way allowed for the recruitment of a group of ‘lay persons’ that were not specifically targeted for their affiliation with a particular stakeholder group. Since the experiment existed on the web, it was also freely available to anyone with Internet access. The use of Facebook facilitated rapid recruitment of participants (208 participants in 12 hours), but provides a sample that should not be viewed as being representative of any specific population. 3.2.4. Statistical analysis Occasionally participants responded to questions more than once; in these cases only the responses from their initial participation were included, so each participant served as an independent experimental unit in the analysis. Some of the demographic questions allowed for graded responses. For example, the question “How would you rate your familiarity with animal welfare?” had the response options: “not familiar”, “somewhat familiar” and “very familiar”. However, to ease interpretation of odds ratios, these responses were analysed dichotomously: “familiar” (comprising those who had responded either “somewhat familiar” or “very familiar”) versus “unfamiliar”. Spearman’s rank correlation was used to determine whether demographic variables were correlated. If two demographic variables were correlated (rs > 0.8) then logistic regressions were run twice: once with one of the correlated variables and once with the other. The final model included those variables that provided the best fit. This correlation analysis is typically used with continuous predictors, but is also appropriate for dichotomous predictors (Dohoo et al., 2010) and has been reported in previous studies (Pearl et al., 2008; Pearl et al., 2009). Univariable logistic regression was used to test the effect of each demographic variable on support for research described in the non-invasive/non-GM treatment group. The effect of regulation was also included in each logistic model. Demographic effects that were significant (i.e. P<0.05) in the univariable models were then included in the final multivariable logistic regression models. These models also included the effect of regulation, and the 2-way interactions between regulation and each demographic variable, testing participant support for: 1) the noninvasive/non-GM question, 2) the invasive research question and 3) the GM animal question.  30  3.2.5. Comment analysis Qualitative analysis focused on why participants did or did not support the proposed research and why they switched when conditions changed. The first stage of the comment analysis involved the reading and assigning of codes: i.e. “tags or labels for assigning units of meaning to the descriptive or inferential information compiled during a study” (Miles & Huberman, 1994, p. 56). The comments were then read again and codes were checked for consistency, and altered slightly as the comment data were interpreted (following Coffey & Atkinson, 1996, p. 26-53). Initially, all the comments were analysed without focusing on treatment groups, but in a second stage of analysis (to better understand the effects of regulation, genetic modification and invasiveness), the comments and their codes were analyzed in relation to question. Effects of regulation were examined first, followed by a comparison of invasiveness and genetic modification. The quotes used to illustrate the coding (see Results) were selected on the basis of being the most representative of the assigned codes (Table 3.3). 3.2.6. A note on reflexivity As a result of a personal background in animal welfare science, there may have been greater emphasis put on animal welfare related terms (such as harm, pain, suffering, cruelty, reduction and refinement). All attempts were made to remain objective. 3.3. Results A total of 681 participants completed the experiment. Of these, 632 answered all of the demographic questions. Results of the logistic regression include only those participants that answered the questions on research scenarios and all the demographic questions, but the descriptive and qualitative analysis includes all participants. 3.3.1. Demographics The majority of participants (58.2%) were 19-29 years old. Two thirds of participants (66.7%) were female and most participants (62.2%) had college or university level education. Participants came from 26 different countries: the majority were from the United States (37.7%), Canada (33.9%), United Kingdom (14.3%), Australia (3.8%), Denmark (1.8%) and Norway (1.3%). Other countries included Brazil, Sweden, Spain, Cyprus, South Africa, Belgium, France, Germany, Ireland, Bosnia and Herzegovina, China, India, Indonesia, Qatar, Japan, Kenya, Panama and Switzerland. Many participants (44.9%) stated that they had been involved in the animal advocacy/protection movement, with 12.4% expressing “frequent” involvement. 40.5% of 31  participants were “very familiar” with animal welfare and 59.3% of participants stated that they had been involved in the environmental movement (13.9% expressing “frequent” involvement). 19.8% of participants were directly involved in animal research, and 27.1% were “very familiar” with animal research. In terms of diet preference 21.0% of participants indicated that they were vegetarian. 73.4% owned pets, 38.1% were from a rural background, and 62.5% considered themselves to be politically “liberal” or “somewhat liberal.”  32  Table 3.2. Participant demographics (n=632), ‘--‘ indicates the percentage of participants that did not provide an answer to a given demographic question Demographic question Age  Response Options  % Participants  19-29 30-39 40-49 50-59 60-above --  58.2 24.2 6.2 4.0 2.4 5.0  Male Female --  31.5 66.7 1.8  Secondary College/University Masters Doctorate Other --  10.2 62.2 14.0 7.4 1.8 14.4  Yes No --  44.9 53.8 1.3  Minimal Occasional Frequent  52.9 34.6 12.4  How would you rate your familiarity with animal welfare?  Not familiar Somewhat familiar Very familiar --  9.5 48.3 40.5 1.7  Have you ever been a member of, or supported the environmental movement?  Yes No --  59.3 38.9 1.8  Minimal Occasional Frequent  46.1 40.0 13.9  Yes No --  19.8 78.6 1.6  Sex identity  Level of education  Have you ever been a member of, or supported, the animal advocacy/protection movement? If so, please rate your level of involvement:  If so, please rate your level of involvement:  Are you directly involved with some aspect of animal research (i.e. research team member, technician etc)?  33  Demographic Question How would you rate your familiarity with animal research?  Response Options  % Participants  Not familiar Somewhat familiar Very familiar --  20.9 47.5 27.1 4.5  Do you consider yourself to be vegetarian/vegan?  Yes No --  21.0 77.7 1.3  Do you currently own a pet?  Yes No --  73.4 25.3 1.3  Rural Urban --  38.1 60.0 1.9  Liberal Somewhat liberal Neutral Somewhat conservative Conservative --  41.3 21.2 22.6 8.5 4.3 2.1  Do you come from a rural or an urban background?  Politically, how do you consider yourself to be?  34  3.3.2. Participant responses Non-invasive/non-GM question: non-GM pigs fed two different diets In the question proposing non-invasive/non-GM research in an unregulated environment 296 of 363 (81.5%) participants supported the research (part i of Tables 3.4 and 3.5 combined). In the question proposing non-invasive research in a regulated environment 268 of 318 participants (84.3%) supported the research (part ii of Tables 3.4 and 3.5 combined). Comments indicated that their support was based on the following factors: scientific merit due to research validity; benefits to the environment, animals, industry, and society; and no costs to animal welfare. Some participants specified conditions to their support, such as: animal health, animal welfare and humane practice. Example quotes: “Seems like a valid experiment” “Beneficial to the agricultural industry to improve techniques and reduce pollution.” “There is a clear potential benefit from the research, and on first impression the research does not appear likely to cause undue harm to the pigs.” The remaining participants (n=67 in the unregulated environment; n=50 in the regulated environment) were opposed to the non-invasive research (parts i and ii of Tables 3.4 and 3.5). In the unregulated treatment group, 9 of the 15 unsupportive participants (60.0%) were willing to shift their response to one of support if the research was regulated (part iii of Tables 3.4 and 3.5 combined). However, the remainder of participants (n=52) were opposed to animal research even when regulation was in place (part iii of Tables 3.4 and 3.5 combined). Their comments denote that opposition was based on being absolutist against animal research either because animals cannot consent (because it is wrong, it is exploitative, or that we should use humans instead), lack of trust in researchers, and a belief that the research lacks scientific merit. Example quotes: “Absolutely never. Not even if it meant saving my life. I will never agree that it is acceptable to take another creature's life to extend my life or improve the quality of my life. Animals are not and never will be ours to experiment on.” “I do not believe it is ethical to conduct research on non-human animals. They cannot consent, unlike volunteer human participants. Further, I have seen, firsthand, the treatment that research animals can be subjected to, and that treatment is unacceptable.” 35  Invasive question: non-GM pigs undergoing invasive procedure Of the 148 participants who had initially supported unregulated non-invasive research 59 (39.9%) were no longer willing to support the research if it involved an invasive procedure (Table 3.4, part iv). Their comments suggest that this drop in support is due to several factors: the belief that refinement and replacement alternatives are both available, the belief that the research is cruel, too invasive, painful, or that the animals are in discomfort or their health is affected, and belief that the research lacks scientific merit because it is unnecessary or not scientifically valid. Some participants specified conditions under which they would have maintained their support: if fewer animals were used (reduction), if scientific merit was improved, if regulations were in place, and if there was no risk to animals or humans. Example quotes: “I think that they can get the results without having to perform surgery. That is what the faecal matter, urine and saliva swabs containing saliva from the pig are there for.” “100 pigs is too many – fistulate only a sub-group.” “It is too invasive and there are other ways to accurately assess pig digestion and nutrition without doing that and I think it would significantly affect their quality of life beyond that which they would have suffered anyway on a farm.” However, 30 of the 59 unsupportive participants (50.8%) were willing to change their response, and support the invasive research if it was regulated (Table 3.4, part v). Their comments reveal that the primary factor influencing the switch from opposition to support was regulation, particularly the implementation of formal inspection, animal welfare standards, increased public accountability (through greater openness) and ethical review of protocols. Some participants specified conditions to their support: animal welfare assessment, implementation of refinement, and an assurance that the research has scientific merit. Example Quotes: “This regulation implies that they will attempt to make the pigs' lives more comfortable. The third party regulating it would hopefully find a middle ground between people who don't care about the animals and the people who don't want the pigs in captivity.” “With the experimental proposals being reviewed by both scientists and public I think this would ensure the discomfort for the pigs would be kept to a minimum.”  36  “If this regulatory procedure can properly establish whether it is necessary to perform surgery to address the question then the answer is yes.” Of the 132 participants who had supported regulated non-invasive research, 53 (39.4%) were no longer supportive when invasive methods were used, even though regulation was still in place (Table 3.4, part vi). Their comments indicated that this drop in support was due to the belief that the research is too invasive, that refinement and replacement alternatives are available, and that the research is unnecessary, not scientifically valid, or has unworthy beneficiaries. Example Quotes: “This seems like an excessive treatment that would cause the animal discomfort as well as increase its susceptibility to infection.” “Surgery/implants seem excessive. They can get those answers from the naturally processed product.” “Seems unnecessary to me.” Of the 79 participants who were supportive of regulated invasive research, 61 (77.2%) were no longer willing to support the research when regulation was removed (Table 3.4, part vii). Their comments indicated that this drop in support was due to a lack of regulation, particularly a lack of inspection, and reduced accountability as a result, lack of scientific merit, lack of trust in researchers and the belief that refinements are available. Example Quotes: “Being accountable to somebody is the key to research and if there is no regulation there is no accountability.” “I don't trust the human nature of the average scientific researcher enough to let a group of them do something like this on an unregulated basis. Unregulated, I suspect the researchers might start to cut corners with the health and safety of the animals.” “I would worry that unregulated research would produce results that could not be repeated should the research results need to be confirmed.” Eighteen of 79 supportive participants (22.8%) did not switch their response based on the removal of regulation and instead answered “Yes” throughout the experiment (Table 3.4, part vii). Their comments suggest strong trust in researchers and a focus on benefits, especially to society. Example quotes: “It’s a case of innocent until proven guilty. There is nothing suggesting that these researchers will not perform their duties in a humane and responsible manner.” 37  “Anything to help humans live healthier lives.” “I would hope that the researchers would conduct themselves and their research in an ethical manner despite being outside of a regulated area.” GM animal question: GM pigs being fed two different diets Of the 183 participants who had initially supported the unregulated use of non-GM pigs, 64 (43.2%) were no longer supportive when the research involved the use of GM pigs (Table 3.5, part iv). Their comments indicate that the drop in support was due to: complete opposition to genetic modification and costs to animal welfare, the environment and human health. Some participants indicated that they might support the research with adequate regulation and risk assessment. Thirty of the 64 unsupportive participants (46.8%) indicated that they were opposed to genetic modification, but did not specify why. Example Quotes: “Genetically modifying anything is wrong.” “It's unnatural and potentially cruel to the animals.” “Pigs should not be genetically modified. It could be dangerous to the pigs and the people who may choose to eat them” Of the 64 participants that were opposed to unregulated research involving GM pigs, 21 (32.8%) were willing to change their response to one of support when the research was regulated (Table 3.5, part v). Their comments indicated that factors influencing their switch from opposition to support include: regulation, in particular third party ethical review, inspection, and increased public accountability through greater openness. Some participants specified conditions to their support: animal welfare assessment, and a worthy beneficiary when applying research findings. Example Quotes: “I support the more-regulated version of this study because of the formal review. An independent panel is a good idea.” “I support regulated research because it provides the human animal a means to be responsible to its environment and be held accountable for its rapport with aforesaid environment.” “If this system is in place then it would be up to the panel to decide if the merit of the experiment is worthy of being carried out. If they support it then I would support it.” Other participants who received the GM animal question began the experiment with regulation present. Of the 136 participants who had supported the use of non-GM pigs, 50 (36.8%) were no 38  longer supportive when the research involved GM pigs, even though regulation was still in place (Table 3.5, part vi). Their comments indicate that this drop in support was due to opposition to genetic modification, costs to animal or human health, and a belief that the research is unnecessary and lacks scientific validity. Example quotes: “I do not support any type of genetic modification on any substances that are intended for eventual human ingestion.” “I feel that GM animals are often born from cruel situations. A lot of animals are created and suffer to produce one viable animal. I feel that these practices are inhumane.” “I do not support the use of genetic modifications period.” Of the 50 unsupportive participants (in both unregulated and regulated treatment groups), only two participants mentioned concerns about the consequences of consuming GM pigs as a factor in their decision-making. Of the 86 remaining participants who were willing to support the use of GM pigs in regulated research, 59 (68.6%) were no longer willing to give their support when regulation was removed (Table 3.5, part vii). Their comments indicated that this drop in support is due to: lack of regulation, reduced animal welfare, lack of scientific merit, lack of trust in researchers, and costs to human and animal health. Example Quotes: “If this research is unregulated, then I would worry that the pigs’ "best interests" were not kept in mind and that they might be likely to suffer in either bad conditions or due to inappropriate treatment.” “The researchers would have no oversight and nothing would stop them from taking it too far and hurting the animals.” “Regulation is important not only for animal welfare, but also to help validate results. Unregulated research is not as widely accepted by the scientific community.” As with the invasive question, a proportion of participants (27 of 86, 31.4%) responding to the GM animal question continued to answer “Yes” throughout the experiment (Table 3.5, part vii) because of (as comments indicated) trust in researchers, benefits (to the environment, animals, society, industry and science), and being ‘for’ genetic modification (because it is the same as breeding, GM pigs have already been created, it is efficient). 39  Example quotes: “One must have faith in the ethics of fellow scientists when research must be carried out in an area where supervision cannot be exercised. Science cannot be baby-sat at all times.” “Environmental and human concerns outweigh animal rights.” “Pigs have been ‘genetically modified’ by artificial selection already.” It is worth noting that for all questions except the GM animal question, most participants were clear about the framework they used to make their decision. For example, 97 of the 138 (70.2%) participants who answered the invasive question (both regulated and unregulated – Table 3.4, parts v and vii) indicated in some way that they weighed up costs and benefits when deciding whether to support the use of pigs. The remaining 21 participants (15.2%) were less conditional in their responses. In contrast, only 31 of the 150 (20.6%) participants who answered the GM animal question (both regulated and unregulated – Table 3.5, parts v and vii) indicated that they weighed up costs and benefits when making their decision. The majority of the remaining participants (114 of 150, 76.1%) were less conditional in their response. The proportion of participants using cost-benefit decision-making for the invasive is 0.72. The proportion of participants using cost-benefit decision-making for the GM animal question is 0.26. This difference in proportions is significant (χ2=4; df=1; P=0.05). 3.3.3. Statistical results None of the demographic responses were highly correlated  (rs< 0.8 in all cases),  indicating that multicollinearity was not a concern in the analysis. Univariable logistic regressions were carried out for the non-invasive/non-GM question (i.e. non-GM pigs being fed two different diets). The results showed that vegetarianism, familiarity with animal welfare, involvement in animal advocacy, familiarity with animal research and involvement in environmental advocacy were associated with decreased support for the use of research animals. Age (being older) was associated with increased support (Table 3.6. a). These demographic variables, along with the effect of regulation, were included in multivariable logistic regression models testing responses to the non-invasive/non-GM question, the invasive research question, and the GM animal question (Table 3.6 b). A multivariable logistic regression model was carried out for the non-invasive/non-GM question (i.e. non-GM pigs fed 40  two different diets). Non-vegetarians were 7.7 times more likely than vegetarians to support the use of pigs in research that involved feeding two different diets (P<0.0001). Older participants were 2.0 times more likely to support the research than younger participants (P=0.097). Regulation was not a significant predictor of support. There was no interaction between regulation and any of the demographic variables in this model or any of the models reported below. A second multivariable logistic regression model was carried out for the question involving surgical implantation of a fistula into non-GM pigs (invasive question). Nonvegetarians were 2.5 times more likely than vegetarians to support this use of pigs (P=0.0003), and participants were 1.8 times more likely to support the invasive research if it was regulated (P=0.0021). A third multivariable logistic regression was carried out for the question involving the feeding of two different diets to GM pigs (GM animal question). Participants who stated that they were familiar with animal research were 2.3 times more likely to support the use of GM pigs in research than those that claimed to be unfamiliar with animal research (P=0.0010). Regulation was not a significant predictor of support.  41  Table 3.3. An overview of codes that emerged from all stages of the qualitative data analysis. During comment analysis comments were separated according to the regulation treatment and invasiveness and GM animal questions in order to better understand why people responded the way they did in accordance with the manipulation of the three independent variables. Code Name  Description of Code  Decision Making Framework Consequentialist/conditional  Those who indicated that they weighed up the costs and benefits in making their decision about animal-based research  Non-consequentialist/non-conditional  Those whose comments indicated that their decision was not based on consequences or conditions  Costs  Those who commented on only the costs of the research  Animal Welfare  Those who indicated that costs to animals, such as harm, suffering, invasiveness, cruelty or abuse are important  Humane Practice  Those who indicated that costs can be minimized through efforts to reduce animal numbers, refine experimental techniques and replace animals in research  Human Health  Those who identified concerns about costs to human health  Unknown consequences  Those who identified concerns about unknown consequences  Benefits  Those who commented on only the benefits of the research  To environment  Those who indicated that the research has benefits to the environment  To animals  Those who indicated that the research has benefits to animals  To science  Those who indicated that the research has benefits to science  To industry  Those who indicated that the research has benefits to industry  Costs versus benefits  Those who indicated that they weighed up the costs and benefits in making their decision about animal-based research  Regulation  Trust Scientific merit  Those who expreseds belief that regulation ensures that benefits outweigh costs by in turn ensuring: a) ethical review, b) inspection, c) third party involvement, d) accountability, e) openness and f) animal welfare standards Those who indicated that trust in researchers, regulators or experts to ensure that benefits outweigh costs is important. Those who indicated that if the research is necessary, valuable and valid, then the benefits outweigh the costs  42  Code Name  Description of Code  Against animal research Wrong  Those who expressed the belief that animal-based research is wrong  Use humans  Those who stated that humans should be used instead, or that research should only be acceptable if it also acceptable for humans Those who expressed the belief that animal-based research interferes with the nature of an animal  Nature Those who expressed the belief that animal-based research exploits animals Exploits animals Those who indicated that animals cannot consent to being involved in research No consent  For genetic modification Same as breeding  Those who expressed the belief that genetic modification is no different from breeding practices  GM pigs already created  Those who indicated that the pigs have been created already and so it is fine to use them in research  Efficient  Those who expressed the belief that genetic modification is an efficient practice  Against genetic modification Wrong  Those who expressed belief that genetic modification is wrong  Unnatural  Those who expressed the belief that genetic modification is an unnatural practice  43  Table 3.4. Results for the invasive question. Number of participants per question responding “Yes/No.” A vote of “Yes” indicates support for the proposed research. Text in bold at the top of the table columns represents the changes in treatments. Underlined values in the table represents the number of participants that changed their “Yes/No” response based on the addition/removal of invasiveness or regulation. Roman numerals (i-vii) mark where specific results are discussed in the main text of the paper.  A. Invasive Regulated  (n = 6) Yes = 3 No = 3  Non-invasive Regulated  (iii) (n = 32) [Yes = 6 No = 26  Non-invasive Unregulated (i) (n = 180) Yes = 148] [No = 32  Invasive Unregulated (iv) (n = 148) Yes = 89 No = 59]  Invasive Regulated  (v) (n = 59) Yes = 30 No= 29  B. Non-invasive Regulated (ii) (n = 157) Yes = 132] No = 25  Invasive Regulated (vi) (n = 132) Yes = 79] No = 53  Invasive Unregulated  (vii) (n = 79) Yes = 18 No = 61  44  Table 3.5. Results for the genetically modified (GM) animal question. Number of participants per question responding “Yes/No.” A vote of “Yes” indicates support for the proposed research. Text in bold at the top of the table columns represents the changes in treatments. Underlined values in the table represents the number of participants that changed their “Yes/No” response based on the addition/removal of invasiveness or regulation. Roman numerals (i-vii) mark where specific results are discussed in the main text of the paper.  A. GM Regulated  (n = 9) Yes = 6 No = 3  Non-GM Regulated  (iii) (n = 35) [Yes = 9 No = 26  Non-GM Unregulated (i) (n = 183) Yes = 148] [No = 35  GM Unregulated (iv) (n = 148) Yes = 84 No = 64]  GM Regulated  (v) (n = 64) Yes = 21 No= 43  B. Non-GM Regulated (ii) (n = 161) Yes = 136] No = 25  GM Regulated (vi) (n = 136) Yes = 86] No = 50  GM Unregulated  (vii) (n = 86) Yes = 27 No = 59  45  Table 3.6. (a) Demographic variables and their effect on participant’s willingness to support animal-based research. Individual logistic regressions were carried out for all demographic variables (as reported in Table 3.2). Only those that were significant predictors of support (p < 0.05) are included here. Regulation was factored into the model. CL = confidence limits. Variable  Odds ratio  CL (95%)  R2  P value  Vegetarianism  0.10  0.06 to 0.16  0.24  <0.0001  Animal advocacy  0.29  0.18 to 0.45  0.08  <0.0001  Familiarity with animal welfare  0.34  0.22 to 0.56  0.06  <0.0001  Age  2.1  1.3 to 3.4  0.03  0.0018  Environmental advocacy  0.56  0.34 to 0.91  0.02  0.0181  Familiarity with animal research  0.56  0.34 to 0.83  0.02  0.0101  Table 3.6. (b) Results from the multivariable logistic regressions for the invasive question and the GM animal question. Question  Variable  Odds ratio  CL (95%)  R2  P value  Interpretation  Vegetarianism  0.13  0.07 to 0.23  0.3  <0.0001  Vegetarian participants were 87% less likely to support the noninvasive/non-GM research  Age  2.0  1.2 to 3.5  0.3  0.0097  Older participants (>30) were 2.0 times more likely to support the non-invasive/non-GM research  Regulation  1.8  1.2 to 2.6  0.07  0.0021  Participants were 1.8 times more likely to support the invasive research if it was regulated  Vegetarianism  0.4  0.25 to 0.67  0.07  0.0003  Vegetarian participants were 60% less likely to support the invasive research  Familiarity with animal research  2.3  1.4 to 3.8  0.04  0.0010  Participants familiar with animal research were 2.3 times more likely to support research involving GM animals  Non invasive/ non-GM  Invasive  GM animal  46  3.4. Discussion 3.4.1. Relationship between demographic variables and support for animal research Vegetarianism was a significant predictor of decreased support across all treatment groups. This fits with previous findings (Furnham & Pinder, 1990; Broida et al., 1993; Hagelin et al., 1999; Schuppli & Weary, 2010); however, care needs to be taken in assuming that vegetarianism is a predictor of attitudes. Being vegetarian can be an action that results from a certain attitude towards or beliefs about animals. It also relates directly to opinions regarding the use of animals in food production. As the research scenarios in the study involved farmed pigs, vegetarian participants may have assumed that these animals would go into food production after the proposed research was completed. Thus the negative views of some vegetarian participants may have been associated with farming pigs for food production, rather than being against the research per se. Typical rates of vegetarianism in the UK, US and Canada are roughly 3-4% (American Dietetic Association, 2003; Harris Interactive, 2008; Davies, 2009). The overrepresentation of vegetarians among our participants likely allowed a stronger test of the effects of this variable. Other variables were predictors for specific scenarios. For example, professed familiarity with animal welfare was associated with decreased levels of support for the non-invasive/non-GM question, perhaps because people familiar with animal welfare are likely to have animal interests at heart when making these decisions. This finding echoes the work of Knight et al. (2004), who found a positive correlation between belief in animal mind and opposition to animal use. Belief in animal mind has been broadly categorized as belief in an animals’ ability to make decisions or solve problems (animal cognition) (Knight et al., 2009), and an animals’ ability to experience pain and suffer (animal sentience) (Herzog & Galvin, 1997). Age was also an influential variable in the non-invasive/non-GM treatment group, with older participants being more supportive. This is in line with previous findings showing that older people are more supportive of the use of animals in research (e.g. Hagelin et al., 2003). Professed familiarity with animal research was a predictor of support for the use of GM pigs in research. This is perhaps because individuals claiming to be familiar with animal research may also be familiar with genetic modification, as this is a commonly used procedure in laboratory science (Ormandy et al., 2009). Due to the complex nature of the science involved in genetic modification, those involved in scientific practices such as animal research may have a better understanding of the process and thus feel better able to judge the risks.  47  Other demographic variables were not predictors of support for the specific questions, but may have influenced the overall levels of support for animal use. For example, our sample of participants likely over-represented politically liberal views compared to the general population. Research has shown that people who are politically more left leaning are less supportive of animal experimentation (Buss et al., 1986; Eurobarometer, 2001; Herzog et al., 2001). This may also be explained by differences in people’s ‘world view’ or ethical ideology because attitudes towards animals are closely related to attitudes towards other political and social matters (Furnham & Pinder, 1990). In previous work, sex identity has been found to relate to attitudes towards the treatment of animals (including research animals), with virtually all studies finding that a lower proportion of females accept the use of animals in research compared to males (Hagelin et al., 2003). We did not find an effect of sex identity on responses to specific questions, but the over representation of females in our sample may have affected the overall levels of support. 3.4.2. Influence of experimental treatments on participant support Support for non-invasive research or research using non-genetically-modified animals was high in both the unregulated and regulated treatment groups. Animal-based research became less acceptable if it involved invasive techniques, but the acceptance of the invasive research increased if regulation was in place. These findings fit with the predictions made at the outset of the experiment. There are certain conditions that, for some people, make animal-based research more acceptable (e.g. Hagelin et al., 2003). Such conditions often involve minimising costs to the animals to ensure that benefits outweigh the costs. It follows that regulation should encourage participants to switch from opposition to support for the animal use. Indeed, the multivariable logistic regression showed that regulation was a predictor of support in the invasive question. Comments by participants also identified regulation as a factor in influencing their decision to support the proposed research. As predicted, regulation had little effect on the (already low) level of support for research involving GM animals, although 21 of 64 of participants changed their response from opposition to support when regulation was added. The comments from participants who switched from opposition to support when regulation was added focused on issues such as third-party ethical review, inspection, increased public accountability through greater openness, and the value of the research and its scientific validity (which some participants believe is gained through having regulations in place). Participants who did not support the use of GM pigs often invoked a less 48  conditional approach to genetic modification, stating that genetic modification is “wrong” or “unnatural.” These findings fit with other studies that have examined people’s attitudes towards the genetic modification of animals (MORI, 1999; Macnaghten, 2001; 2004). Eurobarometer surveys (Gaskell et al., 2003a) also found low levels of support for the use of GM animals, especially for animals used in agriculture, and “greater opposition to GM food than to GM crops suggest[s] that the public may be more concerned about food safety than the environmental impacts of agri-food biotechnologies” (Gaskell et al., 2003a, p. 11). 3.4.3. Decision-making frameworks underlying participant responses A combined analysis of the “yes/no” results and participant comments suggest that the invasive question promoted a utilitarian decision-making framework. That regulation was not a significant predictor of support for the GM animal question may be due to the use of a less conditional and more ‘rule-based’ framework than a consequentialist one. This interpretation is supported by the qualitative data, with a higher percentage of less conditional views emerging in comments from the GM animal question than in the invasive question. This result highlights a potential difference in the decision-making frameworks that participants use when they are deciding whether or not to support the use of GM animals in science, compared with decisions about whether to support animal research in general. The difference in decision-making may be due to different attitudes about the types of research the animals were involved in. Some participants who were asked about invasive research were clear that the use of pigs was acceptable as long as harm was minimized, but research was unacceptable if there was a belief that alternative non-invasive methods could be used. This finding fits with previous research indicating that underlying beliefs about the availability of alternative methods shape people’s attitudes about animal research (Knight et al., 2003; 2010). The weighing up of costs and benefits, as seen in the invasive treatment, may also imply a more orthodox rational approach, where decision-making is mediated by conscious appraisal of available information related to the events prior to making the judgment about whether the research should proceed. This perhaps fits with more traditional theories of moral reasoning, such as Kohlberg’s rationalist framework which “assumes that moral judgments are based on a priori evaluation of relevant and available information” (Knight & Herzog, 2009, p. 454), or the theory of planned behaviour which assumes that behaviour is governed by beliefs, norms and intentions to act (Ajzen, 1991). Indeed, most participants in the invasive treatment were clear about their concerns, and outlined conditions regarding what they were willing to accept and why. For example, some participants commented that they accepted invasive research provided that the 49  animals did not experience any unnecessary suffering or harm, that the benefits outweighed the harms, that the research was regulated and that the use of animals was reduced, refined and replaced wherever possible. Some participants gave their own recommendations for how to minimize pain and distress, by suggesting anaesthesia, analgesia, and humane endpoints for the study. Such recommendations are already implemented via regulation and implementation of the Three Rs, indicating that (for the most part) current regulatory systems are likely aligned with societal concerns regarding invasive animal-based science. In contrast, when making moral judgments about newer, emerging technologies such as genetic modification, a more intuitionist approach (as proposed by Haidt, 2001) may take precedence, and underlying beliefs and norms might still need to be formed in order to make a priori evaluations. Many participants that were asked about research involving GM pigs indicated that they think genetic modification is ‘wrong’. This seems to support the social intuitionist theory, which posits that, “moral judgments are like aesthetic judgments…gut feelings happen to us quickly, automatically, and convincingly…we know immediately that the act in question is right or wrong. Then, if somebody asks us to explain our judgment, we search for reasons why our judgment is correct” (Haidt, 2002, p. 54). It is worth noting that a high proportion of participants (46.8%) stated their opposition to genetic modification without elaborating on the nature of their opposition. This result suggests that responses are being made intuitively, without clearly articulated reasons for the opposition. If this is the case it may prove valuable to track people’s decision making over time, as more becomes known about genetic modification technology. The less conditional approach that participants used when deciding to support or oppose the use of GM pigs may also stem from emotional responses to the unknown. In support of this Slovic (2006, p. 3) argues that, “responses to uncertain situations appear to have an all-or-none characteristic that is sensitive to the possibility rather than the probability of strong positive or negative consequences, causing very small probabilities to carry great weight.” These results indicate that, if scientists continue to create and use GM animals, they need to better inform the public about the nature of the genetic modification, potential risks, and how these have been addressed. Our findings also call into question whether the Three Rs framework, which aims to minimize harms to animals in science, is an effective tool for addressing seemingly rule-based concerns regarding GM animals.  50  3.4.4. Implications for animal policy and animal welfare Comments indicated that certain regulatory components affect participants’ acceptance of animal-based research. In particular, ethical review, inspection or assessment, third party involvement (for both ethical review and inspections), and greater accountability to the public through openness (which in turn generates trust). These findings highlight key areas for policy makers and institutions to improve and develop. Care must be taken in making more general inferences about the opinions of the ‘public’ based on the results of this study. There are some drawbacks to recruiting participants using the Internet; for example, relatively little may be known about participant characteristics, and what is known may be questionable (Dillman, 2000; Stanton, 1998). In addition, people with stronger views are more likely to participate, and they may not represent the majority of views in wider society. For these reasons the results are not intended to provide a basis for inferences about general levels of support for animal use. However, because participants were randomly allocated to different treatments, the current sample is appropriate for drawing inferences about the effects of the treatments that were tested within this experiment. Moreover, the comments provided by participants highlight some of the conditions under which participants find invasive (but not GM) animal research more acceptable, including reduction in animal use, refinement of scientific practices so that less harm is caused, and replacement of animals where possible – the pillars of the Three Rs. Regulatory oversight systems are typically responsible for implementing the Three Rs. A more novel finding is the reduced effect of regulation in affecting responses to use of GM animals. This result suggests that policy makers will need to take other approaches to address concerns over the creation and use of GM animals. 3.5. Conclusion Current regulatory systems typically use a utilitarian framework that weighs perceived costs and benefits. The present study showed that regulation is a predictor of support for invasive animal-based research, but has less effect on support for research involving GM animals. When making decisions about the use of GM animals in research participants seemed to be more intuitive and less conditional in their decision-making and evaluation, rather than using utilitarianbased reasoning. The use of different decision-making frameworks creates a challenge for regulators that have until now relied almost exclusively on a utilitarian approach to decisions about the use of animals in research.  51  The study presented in Chapter Two showed that mice and zebrafish are the most commonly used species in research. The following chapter uses the same online public engagement tool discussed in the current chapter to explore people’s willingness to accept the use of zebrafish or mice in biomedical research.  52  4. Factors affecting people’s acceptance of the use of zebrafish and mice in researchφ 4.1 Introduction As discussed in Chapter One, a wide variety of factors are known to influence people's attitudes towards animal research ranging from beliefs about animal sentience to concerns about scientific merit. This study focused on three factors in particular: species of animal, genetic modification techniques, and regulatory framework. The most commonly used species are mice, zebrafish, and rats (Ormandy et al., 2009), and the number of animals used is now greater than ever due to an increase in the use of genetically modified (GM). People’s attitudes towards research animal use tend to differ depending on the species involved (Driscoll, 1992; 1995; Serpell, 2004). For example, members of the public tend to be less willing to accept the use of companion animals (e.g. dogs or cats), in part because these animals are viewed as having higher mental abilities compared to many other species used in research (Eddy et al., 1993). Similarly, Animal Care Committee members have been shown to be less comfortable with research using companion animals and non-human primates (Schuppli, 2011) because of beliefs about the sentience of these species. It has been shown that animals that are closer relatives of humans tend to evoke more positive affect (Eddy et al., 1993) and those animals perceived as ‘cute’ tend to be preferred (Gunnthorsdottir, 2001; Herzog & Burghardt, 1988; Lawrence, 1989). Species such as fish and invertebrates are typically rated below mammals, and as such are considered an appropriate relative replacement for mammals in research (CCAC, 2005a; DeTolla et al., 1995; Fabacher & Little, 2000). However, the welfare of laboratory fish has received little academic interest despite the mounting evidence that fish are sentient and have the capacity to feel pain (Chandroo et al., 2004). New scientific technologies, such as genetic modification, can also affect attitudes towards animal research. Several studies have shown that support for either the creation or use of genetically modified (GM) animals is low (e.g. Macnaghten, 2004; Gaskell et al., 2003a; Schuppli & Weary, 2010). People have expressed fundamental moral opposition to genetic modification (Ormandy et al., in press a), and concern regarding its ‘unnaturalness’ and potential to lead to unknown consequences (e.g. Macnaghten, 2001; Eurobarometer, 2001; Birke et al., 2007). φ  A version of this chapter has been accepted for publication: Ormandy, E. H., Schuppli, C. A. & Weary D.  M. (in press) Factors affecting people’s acceptance of the use of zebrafish and mice in research. Alternatives to Laboratory Animals.  53  Although the bulk of public-attitudes research has focused on consumer attitudes toward GM food production animals, there is evidence that biomedical applications of GM animals may be more acceptable than food related applications (Schuppli & Weary, 2010). However, some research has shown that there is public concern about the methods of developing GM animals, including concerns about increased numbers of research animals used due to the relatively inefficient nature of some genetic modification techniques (Macnaghten, 2004; Ormandy et al., in press a). Attitudes towards other genetic technologies, such as ethyl-N-nitrosourea (ENU) mutagenesis, have yet to be evaluated. ENU mutagenesis is a common practice: a filtered search of the Mouse Genome Informatics (MGI) database for mutant mice that have been created using ENU yielded thousands of results, and a PubMed search using the terms ‘mouse’ and ‘ENU’ pulled up 836 articles and 75 reviews. ENU mutagenesis is also often used on zebrafish (de Bruijn et al., 2010). The presence of an appropriate regulatory framework also affects people’s willingness to accept animal-based research. For example, people’s support for research involving pigs increased when the research was conducted within a regulated environment (Ormandy et al., in press a); regulation had less effect when GM pigs were used, reflecting overriding concerns about genetic modifications (Ormandy et al., in press a). Based on the literature regarding how public attitudes towards animal research are affected by species, genetic modification, and regulation, we predicted that: 1) support for the use of zebrafish will be higher than support for the use of mice, even if research is unregulated; 2) support for the use of GM zebrafish will be low (compared to the use of zebrafish in ENU mutagenesis), even if the research is regulated; and 3) support for research involving zebrafish or mice will increase if the research is regulated. To test these predictions we designed an online experiment in which participants were presented with different research scenarios, all focused on creating animal models for biomedical research on skin cancer. The study was designed to explore where and why people draw the line in terms of what they are willing to accept, with a view to informing animal welfare policy.  54  4.2. Methods 4.2.1. Experimental design Participants (n = 467) were randomly assigned to either an unregulated or regulated treatment. Participants in the regulated treatment were told that the proposed research was subject to independent ethical review, subject to third party facility inspection, and that researchers were required to accurately report research animal numbers (Table 4.1). The experiment used a contingency design: how participants answered each question determined the next question they received. This design resulted in a total of 14 possible paths through the questions (Figure 1). These different paths allowed for the manipulation of the variables of interest as well as providing the research team with a means to examine where people draw the line in terms of what they are willing to accept. The first question for all participants described a research proposal in which zebrafish were to be exposed to chemicals that cause mutations in germ line cells (Table 4.1). The research scenario presented to participants involved ENU mutagenesis, a common technique used to create mutant animal models (de Angelis et al., 2000). It was made clear that the intention of using ENU mutagenesis was to create animal models for the study and treatment of skin cancer. Participants were first asked whether they were willing to support the use of 100 zebrafish in this research. Those who responded, “yes” were then randomly assigned to either the species treatment (involving ENU mutagenesis of mice) or the genetically modified (GM) animal treatment (involving the use of GM zebrafish), and again asked about their support. The species treatment proposed the use of 100 mice exposed to ENU mutagenesis; the GM animal treatment proposed the use of 100 GM zebrafish. Both treatments proposed the use of animals to create models of skin cancer and were based upon real research (Fujii et al., 1976; Amatruda et al., 2002; Mizgireuv & Revskoy, 2006). Participants starting in the unregulated treatment and who expressed opposition at any point by responding “no”, were asked the same question again but this time with regulation in place (Figure 4.1A). Participants who were initially assigned to the regulated treatment, and who supported either the use of mice (species treatment) or the use of GM zebrafish (GM animal treatment), were asked the same question again, but this time with regulation removed (Figure 4.1B). In addition to “yes” or “no” answers, participants were asked to provide reasons for their answers. Participants who did not want to comment could leave an “x” in the text field, but most respondents (93%) commented on their “yes” or “no” response. 55  Figure 4.1. An illustration of pathways through the questions. Participants were first allocated to Unregulated (Panel A) or Regulated (Panel B) scenarios. ‘R’ indicates random assignment of the participant to either the species treatment group or the genetically modified (GM) animal treatment group. A.  B.  56  Table 4.1. Independent variables manipulated in the experiment. Initially, all respondents were randomly assigned to either the unregulated or regulated treatment and then asked if they would approve research that proposes “the use of 100 zebrafish to study skin cancer”. Participants who answered “Yes” to this research scenario were then randomly assigned to either the species or genetically modified (GM) animal treatment group. Variable  Regulation  Levels / Treatment groups  Research scenario (text as it appeared to participants)  Unregulated  These researchers are unregulated. That is to say there are no formal regulations to control or oversee how animals are used in research.  --------  --------  Regulated  Zebrafish  Species  --------Mice  Genetic Modification  The use of animals in research is regulated. Attributes of this system include: - Regulation that aims to promote ethical use of animals - Formal review of a) research team members, b) the research facility and c) experimental proposals by an independent panel including scientists and members of the general public - Accurate reporting of numbers of animals used - Random and routine inspection of the research facility by a third party The researchers propose the use of 100 zebrafish to study skin cancer. Zebrafish are a species of small, tropical freshwater fish, commonly used in scientific research. To study skin cancer the zebrafish will be immersed in a chemical solution that causes tumours. -------To further their understanding of skin cancer the researchers now propose the use of 100 mice. To study skin cancer the mice will be injected with a chemical that causes tumours to grow at the site of the injection.  Non-GM  The researchers propose the use of 100 zebrafish to study skin cancer. Zebrafish are a species of small, tropical freshwater fish, commonly used in scientific research. To study skin cancer the zebrafish will be immersed in a chemical solution that causes tumours.  --------  --------  GM  To further their understanding of skin cancer the researchers now propose the use of 100 genetically modified zebrafish. To study skin cancer, the zebrafish will be genetically altered so that they carry a gene that activates tumour growth.  57  4.2.2. Recruitment Participants were recruited for the experiment using Facebook. Four types of Facebook stakeholder groups were targeted: an animal advocacy group (1,049 members), an anti-vivisection group (2,055 members), a pro-research group (4,095 members), and an environmental advocacy group (8,021 members). Assumptions were made about the characteristics of these different stakeholders, so participants were targeted for recruitment based on these assumptions (purposive sampling) (Tashakkori & Teddlie, 2003). In addition, the survey was available online for public access. To further characterize participants, a series of demographic questions was included (Table 4.2). This allowed the influence of various demographic factors on participant support to be tested. In addition, a series of additional questions were asked regarding species sentience, genetic modification, and regulation of animal-based research (Table 4.3). These questions allowed us to better understand participant views towards the three variables being manipulated during the experiment. 4.2.3. Statistical analysis Univariable logistic regression was used to test the effect of each demographic variable on support for ENU mutagenesis in zebrafish. Regulation was also included in each logistic model. Demographic effects that were significant (i.e. P<0.05) in the univariable models were then included in a final multivariable logistic regression model. The multivariable model also included regulation, and the 2-way interactions between regulation and each demographic variable. 4.2.4. Comment analysis Qualitative analysis focused on trying to understand why participants did or did not support the proposed research and why they switched when conditions changed. The first stage of the comment analysis involved the reading and assigning of codes: i.e. “tags or labels for assigning units of meaning to the descriptive or inferential information compiled during a study” (Miles & Huberman, 1994). The comments were then read again and codes were checked for consistency, and altered slightly as the comment data were interpreted (Coffey & Atkinson, 1996). Initially, all the comments were analysed without focusing on treatment groups, but in a second stage of analysis (to better understand the effects of regulation, genetic modification, and species), the comments and their codes were analyzed in relation to question.  58  4.3. Quantitative results 4.3.1. Demographics Many of the participants (36.9%) were relatively young (19-29 years old), and the majority (74.8%) were female and had at least some (56.1%) post-secondary education. Participants were from Canada (69.6%), the United States (18.2%), the United Kingdom (7.7%), and Australia (1.0%) and other countries (all less than 1%) including Afghanistan, Columbia, France, Germany, India, Italy, New Zealand, Portugal, Singapore, Turkey and the United Arab Emirates. Many (44.3%) of the participants stated that they had been involved in the animal advocacy/protection movement, with 37.6% expressing “frequent” involvement. Similarly, 54.7% stated that they had been involved in the environmental movement, with 37.8% expressing “frequent” involvement. The majority of participants (54.9%) considered themselves to be “very familiar” with animal welfare, 33.3% were directly involved in animal research, and 42.3% were “very familiar” with animal research. Approximately one quarter (23.7%) of participants indicated that they were vegetarian, 78.0% owned pets, 31.9% were from a rural background, and 60.4% considered themselves to be politically “liberal” or “somewhat liberal.” 4.3.2. Additional closed-ended questions When asked to indicate their level of agreement with the statement “genetic modification of animals is an acceptable practice,” 63.0% of participants totally agreed (Table 4.3). When asked about whether animals can experience pain, suffering, happiness and pleasure, 92.8% totally or mostly agreed that dogs are capable of experiencing these states, and 45.4% totally or mostly agreed that mice and fish are capable of experiencing these states. When asked to indicate their level of agreement with the statement “public authorities (e.g. governments) can be trusted to regulate the use of animals in research,” 67.9% totally or mostly agreed. 4.3.3. Statistical analysis A total of 467 participants completed the experiment. Of these, 415 answered all of the demographic questions. Logistic regression analysis included only those participants who provided their demographics, but the descriptive and qualitative analysis describing participant responses includes all participants. Results from the univariable logistic regressions showed that vegetarians, females, and those involved in animal advocacy or environmental advocacy were less likely to support the use of non-GM zebrafish in ENU mutagenesis research (Table 4.4. a). In contrast, older participants, 59  and participants who were themselves involved in animal research, considered themselves familiar with animal research, and a higher level of education were more likely to give support. The multivariable model shows that participants who are animal advocates were 2.4 times less likely to support ENU mutagenesis in zebrafish (Table 4.4. b). Younger participants, and vegetarians were both 1.7 times less likely to support this procedure. In contrast, those participants who claimed to be familiar with animal research were 2.1 times more likely to support the ENU mutagenesis in zebrafish. There was no interaction between regulation and any of the demographic variables in this model.  60  Table 4.2. Participant demographics (n=415), ‘--‘ indicates the percentage of participants that did not provide an answer to a given demographic question Demographic question  Response Options  % Participants  Age 19-29 30-39 40-49 50-59 60-above --  36.9 20.0 16.9 17.4 6.6 2.2  Male Female --  23.8 74.8 1.4  Secondary College/University Masters Doctorate Other --  7.6 56.1 14.8 19.1 1.4 1.0  Yes No --  44.3 54.7 1.0  Minimal Occasional Frequent  26.4 36.0 37.6  How would you rate your familiarity with animal welfare?  Not familiar Somewhat familiar Very familiar --  6.3 36.6 54.9 2.2  Have you ever been a member of, or supported the environmental movement?  Yes No --  54.7 43.4 1.9  Minimal Occasional Frequent  15.1 47.1 37.8  Yes No --  33.3 64.5 2.2  Sex identity  Level of education  Have you ever been a member of, or supported, the animal advocacy/protection movement? If so, please rate your level of involvement:  If so, please rate your level of involvement:  Are you directly involved with some aspect of animal research (i.e. research team member, technician etc)?  61  Demographic Question How would you rate your familiarity with animal research?  Response Options  % Participants  Not familiar Somewhat familiar Very familiar --  14.0 41.1 42.3 2.6  Do you consider yourself to be vegetarian/vegan?  Yes No --  23.7 73.9 2.4  Do you currently own a pet?  Yes No --  78.0 20.3 1.7  Rural Urban --  31.9 64.4 3.7  Liberal Somewhat liberal Neutral Somewhat conservative Conservative --  36.5 23.9 25.1 8.2 2.9 3.4  Do you come from a rural or an urban background?  Politically, how do you consider yourself to be?  62  Table 4.3. Additional questions given to participants (n=415) to establish attitudes towards species sentience, genetic modification and regulation of animalbased research Question  Response  % Participants  Dogs can experience pain, suffering, happiness and pleasure:  Totally agree Mostly agree Mostly disagree Totally disagree Undecided --  85.1 7.7 0.7 3.9 0.7 1.9  Mice can experience pain, suffering, happiness and pleasure:  Totally agree Mostly agree Mostly disagree Totally disagree Undecided --  15.9 29.5 14.0 27.1 11.3 2.2  Fish can experience pain, suffering, happiness and pleasure:  Totally agree Mostly agree Mostly disagree Totally disagree Undecided --  10.1 35.3 20.8 24.4 8.0 1.4  Genetic modification of animals is an acceptable practice:  Totally agree Mostly agree Mostly disagree Totally disagree Undecided --  63.0 26.6 2.2 3.1 2.7 2.4  Public authorities (e.g. governments) can be trusted to regulate the use of animals in research:  Totally agree Mostly agree Mostly disagree Totally disagree Undecided --  31.9 36.0 13.4 2.4 13.0 3.1  63  4.3.4. Participant responses Level of opposition to ENU mutagenesis of zebrafish was high in both regulated (61.5%) and unregulated (69.3%) treatment groups (parts i and ii of Tables 4.5 and 4.6 combined). Of the 158 participants who were initially opposed to the unregulated ENU mutagenesis research involving zebrafish, 124 (78.4%) maintained their opposition even after regulation was added (part iii of Tables 4.5 and 4.6 combined). Due to low levels of support in the opening questions, sample sizes for subsequent questions randomly allocating participants to either the species or GM treatments were smaller than expected: n=37 for unregulated species treatment, n=46 for regulated species treatment, n=33 for unregulated GM treatment, and n=46 for regulated GM treatment. Those participants that had been supportive when asked about research involving non-GM zebrafish tended to maintain their support when either species was switched to mice or the GM treatment was added, regardless of whether regulation was in place or not. Specifically, the percentage of initially supportive participants who continued to support animal use despite the change in species or procedure was 95.6% (n=35/37) for unregulated research on mice, 93.5% (n=43/46) for regulated research on mice, 90.9% (n=30/33) for unregulated research on GM zebrafish, and 95.6% (n=44/46) for regulated research on GM zebrafish.  64  Table 4.4. (a) Demographic variables and their effect on participant’s willingness to support ENU mutagenesis in zebrafish Individual logistic regressions were carried out for all demographic variables (as reported in Table 4.2). Only those that were significant predictors of support (p < 0.05) are included here. Regulation was factored into the model. CL = confidence limits. Demographic variable  Odds ratio  CL (95%)  R2  P value  Animal advocacy  0.28  0.20 to 0.40  0.12  <0.0001  Involvement in animal research  3.4  2.3 to 4.9  0.1  <0.0010  Vegetarianism  0.38  0.26 to 0.59  0.05  <0.0001  Familiarity with animal research  2.3  1.7 to 3.3  0.06  <0.0001  Sex identity  2.0  1.3 to 3.0  0.03  0.0009  Environmental advocacy  0.59  0.42 to 0.83  0.03  0.0017  Level of education  1.6  1.2 to 2.1  0.03  0.0030  Age  0.67  0.48 to 0.70  0.02  0.0200  Table 4.4 (b) Results from the multivariable logistic regression for the ENU mutagenesis in zebrafish question. Pseudo R2 = 0.2 Variable  Odds Ratio  CL (95%)  P value  Interpretation  Animal advocacy  0.42  0.26 to 0.67  0.0003  Participants involved in animal advocacy were 58% less likely to support the proposed research  Familiarity with animal research  2.1  1.2 to 3.9  0.0049  Participants familiar with animal research were 2.1 times more likely to support the proposed research  Vegetarianism  0.59  0.35 to 1.0  0.0360  Vegetarian participants were 41% less likely to support the proposed research  Age  1.7  1.1 to 2.5  0.0132  Older participants (>30) were 1.7 times more likely to support the proposed research  65  Table 4.5. Quantitative results for the species treatment group. Number of participants per question responding “Yes/No”. A vote of “Yes” indicates support for the proposed research. Text in bold at the top of the table columns represents the changes in treatments. Underlined values in the table represents the number of participants that changed their “Yes/No” response based changes in species or regulation. ENU = ethyl-N-nitrosourea, induced chemical mutagenesis procedure. Roman numerals (i-vii) mark where specific results are discussed in the main text of the paper.  A. ENU Mice Regulated  (viii) (n = 14) Yes = 9 No = 5  ENU Zebrafish Regulated  (iii) (n = 79) [Yes = 14 No = 65  ENU Zebrafish Unregulated (i) (n=116) Yes = 37] [No = 79  ENU Mice Unregulated  (iv) (n = 37) Yes = 35 No = 2]  ENU Mice Regulated  (v) (n = 2) Yes = 1 No= 1  B. ENU Zebrafish Regulated (ii) (n = 120) Yes = 46] No = 74  ENU Mice Regulated (vi) (n = 46) Yes = 43] No = 3  ENU Mice Unregulated  (vii) (n = 43) Yes = 12 No = 31  66  Table 4.6. Quantitative results for the genetically modified (GM) animal treatment group. Number of participants per question responding “Yes/No”. A vote of “Yes” indicates support for the proposed research. Text in bold at the top of the table columns represents the changes in treatments. Underlined values in the table represents the percentage of participants that changed their “Yes/No” response based on changes in procedure (ENU mutagenesis or genetic modification) or regulation. ENU = ethyl-N-nitrosourea, induced chemical mutagenesis procedure. GM = genetically modified. Roman numerals (i-vii) mark where specific results are discussed in the main text of the paper.  A. GM Zebrafish Regulated  (viii) (n = 20) Yes = 19 No = 1  ENU Zebrafish Regulated  (iii) (n = 79) [Yes = 20 No = 59  ENU Zebrafish Unregulated (i) (n=112) Yes = 33] [No = 79  GM Zebrafish Unregulated  (iv) (n = 33) Yes = 30 No = 3]  GM Zebrafish Regulated  (v) (n = 3) Yes = 3 No = 0  B. ENU Zebrafish Regulated (ii) (n = 119) Yes = 46] No = 73  GM Zebrafish Regulated (vi) (n = 46) Yes = 44] No = 2  GM Zebrafish Unregulated  (vii) (n = 44) Yes = 11 No = 33  67  4.4. Qualitative results and discussion 4.4.1. Costs and benefits of ENU mutagenesis Opponents to ENU mutagenesis (for both zebrafish and mice), tended to focus on the harms to the animals. A proportion (34%) of the participants commenting on welfare costs expressed the belief that ENU mutagenesis is painful. For example, one participant commented that the research “clearly will cause harm to the fish”, while another stated their opposition to ENU mutagenesis even if it was regulated because “it is still hurting the fish because they are still put in the chemicals that cause tumours.” Concern for pain was related to sentience by one participant: “there is no difference between sentient creatures being immersed in a chemical solution unregulated or sentient creatures being immersed in a chemical solution regulated. The issue is the ability to feel pain.” The belief that zebrafish or mice can feel pain is one aspect of ‘belief in animal mind’ (or animal sentience), which has been shown to affect attitudes towards research animal use in some (but not all) cases (Knight et al., 2004; Knight et al., 2009). For some participants (15%), concerns about animal suffering were framed in terms of the Three Rs – replacement, reduction and refinement (Russell & Burch, 1959) - although the terms were not necessarily used. For example, one participant referred to reduction “I think the research is using too many mice, fewer should be used” and another referred to refinement, “the pain should be alleviated with analgesia and anaesthesia.”  Some of the opponents to ENU  mutagenesis (11%) raised questions about whether non-animal alternatives are available. For example, one participant wrote, “It is still unethical to use any animal in this way - are there no alternatives?” While someone else commented, “I would prefer to see grant money used to find alternatives to research using animals.” Other opponents to ENU mutagenesis (5%) expressed the opinion that human tissue should be used instead. For example, one participant commented that, “it is not relevant to get data from fish. The scientists should use human data with toxicogenomics and cell culture.” Another opponent stated that, “Better results would be obtained by using human skin cells rather than inflicting cruelty to animals.” These comments suggest that opposition to ENU mutagenesis is rooted in two beliefs. First, that the method causes pain to the animals involved, and second, that the use of animals in the proposed research is not necessary, and that a more relevant way to conduct the research is through the use of human tissue. Some supporters of ENU mutagenesis (27%) tended to focus on the benefits of research to human and animal health. For example, one participant indicated that their support was rooted in the belief that this research should proceed because of the “potential value of the project to 68  human/animal health.” Another supporter expressed a similar view by stating, “I believe in medical research that will save and improve the quality of life for all humans.” However, many supporters of ENU mutagenesis (31%) also stipulated conditions to their support, primarily related to concerns about animal pain, suffering, and the Three Rs. For example, one supporter stated several conditions: “I don't like the idea of animal testing but feel it is necessary to do so at times, so yes, [but] ONLY if there is no pain caused to the mice AND if there is no alternative to using live animals AND if they are provided with living conditions and nutrition that are good for them and cause them no stress AND if they are to be euthanized, that it this done in a humane manner” (emphasis added by participant). Such results provide support for decision making processes proposed by Serpell (2004), Stafleu et al. (1999), and Ideland (2009), who posit that when making decisions about animal use, a common strategy is to trade off the utility and necessity of using animals (in this case benefit to human health and condition that there are no alternatives), against concern for animal welfare (in this case minimizing pain and providing humane care). 4.4.2. Species sentience Overall, the results showed little effect of species: 94% of participants that supported ENU mutagenesis in zebrafish also supported ENU mutagenesis in mice (parts iv and vi of Table 4.5 combined). However, 11 of the 70 participants who were supportive of using zebrafish in ENU mutagenesis mentioned in their comments that species sentience, or the location of zebrafish on the phylogenetic scale was an important factor. For example, one participant commented, “It is necessary to use an animal model to study cancer, and zebrafish are lower on the evolutionary chain than other larger species, so if they can be used instead it is preferred.” While others said, “I've heard fish do not have pain receptors, I figure the fish are unaware” and “Again, medical research is important. I'm glad that more intelligent animals such as monkeys are not being used.” Ten participants did switch from support to opposition when species changed from zebrafish to mouse. One participant explained that, “Mice are very intelligent so I consider this far more serious than using fish” illustrating that for some people, the perceived difference in sentience between species is important. The validation questions about animal sentience (Table 4.3) also support this interpretation; more participants totally agreed that mice are sentient (i.e. they can experience pain, suffering, happiness and pleasure) (15.9%) compared to zebrafish (10.1%); in contrast, 92.8% considered that dogs are sentient.  69  4.4.3. Supportive of genetic modification Participants who were initially supportive of using zebrafish for ENU mutagenesis often continued to be supportive when the GM animal question was proposed (96% continued support parts iv, vi, and viii of Table 4.6 combined). These participants commented that GM animal models are more accurate, reduce animal numbers, and are preferable to ENU mutagenesis. For example, one participant commented, “I agree with the use of genetically modified research species as it increases the probability of successful research and also allows for a reduction in the numbers used as you can use less fish and have a greater chance of all (or almost all) generating tumor growth.” Another participant stated that they preferred genetic modification to ENU mutagenesis because it may be less painful, “The genetically modified fish may grow tumours without the pain of being dipped in chemicals.” Support for genetic modification in our experiment concurred with results from the validation questions asked at the end of the survey where a high proportion of participants totally agreed that genetic modification is an acceptable practice (63.0%) – Table 4.3). This result may be due, in part, to the question order and a reflection of views about specific types of animal use. The question asking about general acceptability of genetic modification as a practice came at the end of the experiment, after participants had given their responses to the specific research scenarios, with many of the participants judging the GM animal scenario to be less painful, more accurate and more efficient than ENU mutagenesis. This may have framed the issue in a way that made participants more accepting of genetic modification for this particular experiment. In addition, if participants’ answers were specific to the biomedical research applications given in this experiment, rather than more general application of GM animals, this result may well be an indication of different views towards the genetic modification of laboratory animals compared to the genetic modification of food animals. This difference in attitudes based on the end use for GM animals (i.e. food versus biomedicine) and the resulting benefits has been found in other studies (Ormandy et al., in press a; Gaskell et al., 2000; Lassen et al., 2006). 4.4.4. Against genetic modification A few of the participants who were initially supportive of ENU mutagenesis using zebrafish switched to opposition when the GM animal question was proposed (6% - parts iv, vi, and viii of Table 4.6 combined). These participants explained that they were fearful about unknown consequences. For example, one opponent to genetic modification commented, “I can't put it to words very well, but the idea of genetic mutations that we create scares me. Maybe a  70  latent, perhaps irrational fear that the mutated genes will have unexpected results that aggressively invade the species or other species or somehow upset a delicate balance…” 4.4.5. Regulation Regulation was mentioned by 88 of the 105 (84%) participants who changed their answer in response to the shift in regulation. For example, one participant who switched from support to opposition when regulation was removed commented, “Regulations are important to ensure humane use of animals (for example humane endpoints), adequate project review, inspections, etc. These things are important not only for humane reasons but also to help ensure sound scientific results.” Another participant who switched from opposition to support when regulations were put in place stated, “Before, it was not regulated and the health and welfare of the animals was not the first priority…Also, having a third party inspect the facility is a great method for maintaining a professional and properly run facility. This method will decrease the chance of animals being mistreated.” This indicates both that participants were paying attention to the shift in variables, and that regulation is an important component affecting some people’s willingness to support animal-based research. Specific components of regulation that participants identified as being valuable include ethical review, inspection of facilities, animal welfare monitoring programs, reporting of animal numbers and openness (in order to maintain public accountability). 4.4.6. Influence of demographics on participant support Vegetarianism was a predictor of decreased support, which confirms previous findings (Broida et al., 1993; Hagelin et al., 2003). Typical rates of vegetarianism in the UK, US and Canada are roughly 3-4% (American Dietetic Association, 2003; Harris Interactive, 2008, Davies, 2009). The over-representation of vegetarians among our participants (24%) likely allowed a stronger test of the effects of this variable, and may have influenced the significance of the effects of vegetarianism in this study. Animal advocacy was also associated with decreased levels of support. This too fits with previous findings (e.g. Hagelin et al., 2003), suggesting that people’s underlying beliefs about animals (i.e. their willingness to advocate for protecting them) influences their willingness to support this type of animal-based research. Our participant population had a high percentage of animal advocates (44%), which may have skewed the results, leading to the high levels of opposition to ENU mutagenesis in zebrafish. In previous public-attitudes research, age has been shown to be an influential factor, with younger people being less supportive of animal research in many cases (e.g Hagelin et al., 2003). Our results showed a significant effect of age on level of support. Over representation of younger 71  people in our participant population likely allowed for a stronger test of the effects of this variable, and may have influenced the significance of effects of age on support. Familiarity with animal research was associated with higher levels of support; some other studies have also shown a positive association between familiarity (or knowledge) of a particular topic and positive attitudes (Gaskell et al., 2003a; Qin & Brown, 2007). However, this pattern is not consistent across studies. In a previous experiment that asked participants about their willingness to support the use of pigs in research (Ormandy et al., in press a), familiarity with animal research was associated with lower levels of support, similar to some previous findings (Broida et al., 1993; Knight et al., 2003; Pifer et al., 1994). One explanation for the difference in results between the agricultural model used in the previous experiment, and the biomedical model used here, may be the difference in research settings. We suggest that general preconceptions about the day-to-day life of animals are different for laboratory and farm animals, whether or not they are being used for research. The general image of life in the laboratory setting may often be worse than the reality (explaining why knowledgeable individuals are more willing to support this use), while the reverse may be true for pigs and other farm animals. Sex identity and level of education also influenced participant support, with females being less likely to support animal-based research than males, and those with higher levels of education being more likely to show support. The effect of sex identity and education level in attitudes towards animals has been well documented, and our findings match those of other studies (e.g. Ormandy et al., in press a, Gallup & Beckstead, 1988; Matthews & Hezog, 1997; Kendall et al., 2006). One consistent theme in the public-attitudes literature is that participants often draw a distinction between different types of animal research (Birke et al., 2007). Using animals for medical experiments that will benefit human health tend to be more positively regarded than those used for cosmetics testing (Aldhous et al., 1999; Kane, Parsons & Associates, Inc, 1989). On this basis we had expected the majority of participants (i.e. more than 50% of participants) to support research intended to aid the study of skin cancer. The lower than expected levels of support may have related to the perception that ENU mutagenesis is a painful procedure; research that is perceived as being painful or invasive is often less supported (Hagelin et al., 2003, Richmond et al., 1990). For example, our previous work showed that non-invasive research on pigs was more highly supported, but this support declined when the research required an invasive procedure (Ormandy et al., in press a). However, in this study only 10.1% of participants totally agreed (35.3% mostly agreed) that fish can experience pain, suffering, happiness, and pleasure indicating 72  that there may be something else about the ENU mutagenesis procedure that participants found objectionable enough to oppose, despite the suggested health benefits. 4.5. Implications for policy In Canada, animals that have undergone ENU mutagenesis (i.e. chemically induced mutation) are classified as “genetically engineered” and are considered alongside other animals that are classified in the same way, but that have been produced using different genetic alteration techniques. The results presented here show low levels of support for ENU mutagenesis procedures, and a preference for other genetic modification techniques where animals are born with a predisposition for tumour development. This illustrates the importance of asking more nuanced questions about current procedures in animal research when evaluating people’s opinion, rather than asking more general questions about animal research or genetic modification. 4.6. Conclusions Many of the participants in this study (65%) said that they were opposed to ENU mutagenesis in zebrafish. The primary reasons for this opposition were 1) concerns about animal welfare (i.e. pain and suffering), 2) concerns about reduction and refinement implementation (i.e. too many animals being used, or lack of refinement), and c) concerns that the use of animals in this particular type of research is unnecessary (due to the belief that using human tissue would provide a better means of researching skin cancer). When asked about genetic modification, most participants were supportive. The results suggest that biomedical applications using GM animals are acceptable provided that animal welfare concerns are taken into consideration and that there is sufficient regulation. However, many participants perceived ENU mutagenesis to be painful, suggesting that research that exposes animals to this procedure is not well supported, regardless of whether mice or zebrafish are used. This result calls into question the assumption that invasive research will be considered more acceptable if performed on “lower” species, such as fish, that sometimes serve as replacements for mammals in animal research. It also suggests that more research is required on the welfare effects of ENU mutagenesis and how these might be mitigated. Greater efforts may also be required to inform the public about scientific practice, and allow feedback via public engagement, to reduce opportunities for disconnects between common scientific practice and societal values.  73  5. The use of genetically modified animals in biomedical science: An exploration of stakeholder views 5.1 Introduction Public-attitudes research to date has shown high levels of concern regarding the genetic modification of animals, but less concern for the creation and use of genetically modified (GM) animals in biomedical research (e.g. Schuppli & Weary, 2010). Reasons for not supporting genetic modification of animals include pragmatic concerns about the environment and human health, as well as concerns about ‘playing God’, or feelings of repulsion or unease about the practice (Macnaghten, 2001; 2004). The majority of public-attitudes research relating to GM animals has focused on attitudes to GM food animals (e.g. Costa-Font, 2008), or public attitudes towards specific health-related efforts, such as the use of GM pigs for xenotransplantation (Einsiedel & Ross, 2002), but little work to date has addressed creation of GM animals for use in biomedical research. This is a pertinent topic since, in the past decade, the use of GM animals in research has more than doubled (Ormandy et al., 2009). As in many other jurisdictions, the current mechanism in Canada for taking into account societal values in decisions about how animals are used in research are: 1) the requirement that a community representative is present on animal care committees (ACCs) and assessment panels (CCAC, 2006), and 2) the representation of an animal protection organization (the Canadian Federation of Humane Societies - CFHS) on the national council. However, these individuals (both the community representatives, and the CFHS representatives) may have difficulty representing societal values. ACC community representatives may feel isolated or inadequate because of a lack of technical understanding of the science; they may have been recruited for convenience rather than with a mandate to represent community values; and the chairperson may not make specific efforts to empower them to provide a community viewpoint (Schuppli & Fraser, 2007). Moreover, although the CFHS representatives are well positioned to advocate for the interests of animals, they may have difficulty representing the views of the wider public. When considering attitudes towards the creation and use of GM animals, it is also important to evaluate the perspectives of research technicians and animal care staff: individuals who are responsible for the day-to-day care of animals in laboratories. Research technicians are minority members of ACCs, and in the current terms of reference the presence of research technicians or animal care staff is recommended, but not required, as part of the minimum quorum (CCAC, 2006). Moreover, policy development at the national level rarely allows formal input from research technicians or animal care staff. Therefore, an important stakeholder perspective is 74  perhaps being underplayed when decisions are made regarding national and institutional policy, and when making decisions about whether to approve projects involving animals. In addition to gaining perspectives from members of the public and research technicians, it is also valuable to involve experts in the field and ask them about their views. Scientists working within the field of genetic modification can provide valuable insight into daily practice, and any challenges that they face. However, not all researchers working with animals are using genetic modification techniques or GM animals. Hence, including the perspectives of animal researchers who do not create or use GM animals in their work may add another dimension to the views of the scientific community in relation to genetic modification as a practice. Gaining appreciation for a diversity of values and concerns is an important step in being able to understand what issues should be addressed in policy that regulates the use of animals in research. To begin to address the current lack of information in this area, an interview-based study was carried out to examine the perspectives of researchers, research technicians and animal care staff, and members of the public. The aim was to explore participants’ attitudes, values and concerns about the creation of genetically modified animals for their use in biomedical science. 5.2 Methods 5.2.1. Sampling Purposive sampling methods (Tongco, 2007) were used to recruit 20 Canadian participants representing three different stakeholder groups: 8 animal researchers (6 of whom used GM animals in their research), 5 animal care staff or research technicians, and 7 members of the public (4 community representatives on Animal Care Committees, 3 with no direct involvement with animal research). Animal researchers were recruited from a list of individuals that had been involved in reviewing draft national guidelines. Animal care staff and research technicians were recruited at the 2010 Canadian Association of Laboratory Animal Science (CALAS) symposium – a symposium specifically targeting research technicians and animal care staff. Four members of the public were initially recruited from a list of community representatives who sit on Animal Care Committees. Snowball sampling was then used to recruit participants who had no direct links to animal research. Two additional participants were recruited from a previous study (Ormandy et al., in press a) in which they expressed the view that “genetic modification is wrong.” One of these participants was a wildlife researcher and the other was member of the public – bringing the total numbers to 8 researchers, 7 members of the public and 5 research technicians. 75  5.2.2. Participant demographics Of the eight researchers who participated in the interviews, six had been actively involved in the creation and use of GM animals as part of their research. The remaining two researchers had not been involved in the creation and use of GM animals, but had used animals as part of their research: they were both wildlife biologists, and had done field work. All five research technicians and animal care staff were directly involved in caring for laboratory animals. Two were facility managers for universities, while the remaining three were research technicians that either worked in animal care services (e.g. laboratory animal breeding), or were assigned to specific research projects. Participants were from 5 different provinces of Canada (British Columbia, Alberta, Quebec, Ontario and New Brunswick). Participants included 11 males and 9 females ranging from 30-60 years old. The majority (13 of 20) had not been members of animal advocacy organizations, were not vegetarian (16 of 20), were pet owners (13 of 20), came from an urban background (13 of 20), and considered themselves politically liberal or somewhat liberal (16 of 20). All participants claimed to be familiar or somewhat familiar with animal welfare. All participants totally or mostly agreed that dogs and mice have the capacity to experience pain, suffering, happiness, and pleasure. Most participants agreed that fish have the same capacities, with 4 participants stating that they were undecided. 5.2.3. Interview protocol The semi-structured, open-ended interviews lasted 45 to 90 minutes. Interviews were conducted either face-to-face (n=8) or via the telephone (n=12), and were recorded using a digital audio recorder. The day before the interview, each participant was given two statements and asked to indicate their level of agreement: 1) “Genetic modification of animals is an acceptable practice”, and 2) “Public authorities (e.g. governments) can be trusted to regulate the use of animals in research.” Response options were “totally agree”, “mostly agree”, mostly disagree”, totally disagree”, and “undecided.” These responses were used as starting points for conversations during the interview. Each interview was split into questions about four main themes. Participants were asked: 1) “how do you feel about genetic modification of animals as a practice?” 2) “do you think that the creation and use of GM animals should be regulated differently from other forms of animal research?” 3) “in your experience, are the Three Rs being implemented with regard to the creation 76  and use of GM animals in research?” and 4) “what do you think that the public thinks about the use of GM animals in research?” Each of these questions led to further probes, for example, where relevant the question about the Three Rs was followed up with, “what do you see as challenges to implementing the Three Rs?” and the question about how participants felt about genetic modification of animals as a practice was followed up with, “do you think that limits should be placed on genetic modification?” In many cases these probes led to further questions to try to understand participants’ views. New issues were included in the interview protocol as they arose, to be used in subsequent interviews. 5.2.4. Data analysis Each participant was assigned an alphanumeric code. The letter signified whether the participant was a researcher (R), a research technician (T), or a member of the public (P). The number signified the chronological number of the interview (1-20). “GM” or “no GM” was added to the researcher codes to indicate whether these individuals used GM animals in their research or not. “ACC” was added to the codes for members of the public that were community representatives on Animal Care Committees. These codes accompany the quotes presented in the results below. All participants are represented in the results. Analysis of the interviews was carried out following the process described by Coffey and Atkinson (1996) and Knight et al. (2003). Interview recordings were transcribed verbatim. Each transcript was read several times, then “open coding” was carried out: each word, line, and paragraph was examined to divide text into smaller chunks. When all the text had been divided in this way, descriptive codes were allocated. These codes were grouped into categories, and similarities, differences, relationships and patterns were noted as they emerged. This required reading and re-reading of transcripts, and memos of issues and ideas were recorded throughout the process. 5.3. Results 5.3.1. Responses to closed-ended questions When asked to indicate their level of agreement with the statement “genetic modification of animals is an acceptable practice,” four participants (one researcher, three research technicians) totally agreed, seven mostly agreed (five researchers, one research technician, one member of the public), two mostly disagreed (both members of the public), four totally disagreed (three members of the public, one researcher), and three were undecided (one researcher, one research technician, one member of the public). 77  When asked to indicate their level of agreement with the statement “public authorities (e.g. governments) can be trusted to regulate the use of animals in research,” five participants totally agreed (one researcher, four research technicians, one member of the public), seven mostly agreed (four researchers, three members of the public), three mostly disagreed (one researcher, one research technician, one member of the public), two totally disagreed (one researcher, one member of the public), and three were undecided (one researcher, two members of the public). 5.3.2. Justification for and against genetic modification In general, participants recognized that genetic modification has become a prevalent technology, with the use of GM animals in research “growing exponentially” (T12). When asked about their feelings toward genetic modification, participants tended to have mixed feelings. Support for the creation and use of GM animals in research tended to be stronger for all participants if there were clear benefits to human health. When asked about the genetic modification of animals destined for human consumption (referred to hereafter as GM food animals) there was less support, and stronger concerns were expressed. For example, one research technician stated that, “when we’re talking agriculture, I’m not really sure how much I agree with [genetic modification]. When it comes to [biomedical] research I agree with it” (T17), while one member of the public commented, “for farm animals I’m definitely against [genetic modification]. Research, I don’t know enough about it, but it depends what the research is for. If it will contribute to knowledge that we really do need I guess I’m more in favour of it” (P18). This illustrates that support may depend on the type of benefits gained from GM animals, and that there is a difference in support for GM animals used in biomedicine and GM food animals, with GM animals in biomedicine deemed more acceptable. Participants raised several different justifications both for and against the genetic modification of animals. In some cases the comments were specific to the creation and use of research animals, and in some cases the comments were about genetic modification in general. Arguments for genetic modification Some participants focused heavily on the potential benefits provided by GM research animals, such as improving human health -“[genetic modification of animals] just opens the field completely in terms of what we can look at in terms of human disease and treatments” (T10) or perhaps improving farm animal welfare – “if we can use genetic modification to reduce the suffering of animals in production units, we should talk about that” (R6 – GM). Participants also highlighted the benefits of GM research animals to scientific research, claiming that genetic modification facilitates science through, a) increasing “understanding of biological processes” 78  (P9 - ACC), b) reducing variability between animals, therefore allowing more focused research and reduction of animal numbers in a given experiment – “I think GM animals might even decrease the numbers of animals we have to use because it’s so much more specific” (T17), and c) granting the ability to get research results more quickly – “they’re doing genetic modification to get a faster result” (P15 - ACC). A few participants also mentioned that genetic modification provides more control than selective breeding – “genetic engineering is much more controlled than what people have done with selective breeding in the past” (R4 – GM). Generally, those who were supportive of the genetic modification of animals, and had hands-on experience, were of the opinion that the procedures used to create GM animals are no worse than other lab procedures in terms of animal welfare. Some supportive participants rationalized genetic modification by saying that it is a logical extension of the long term practice of selective breeding, and that genetic modification utilizes naturally occurring processes (e.g. random mutations) – “humans in the last hundreds and thousands of years have bred animals for their purposes, be it for food or as draught animals or pets. They use them in many different ways, and genetic modification is another extension of that” (R3 – GM). Arguments against genetic modification Some concerns were related to the procedures for creating GM animals, with several participants highlighting concerns about: a) the inefficiency of genetic modification and the need for high numbers of breeding animals, as well as the production of surplus animals that do not carry the genetic alteration of interest – “the one thing about working with genetic modification is that there’s often a huge amount of animals wasted to get to the final model that has the proper genetics. We often have to sacrifice a lot of animals along the way that haven’t picked up the mutation” (T10), b) the relevance of GM animal models to human disease, and the translatability of results from GM animal experiments to human conditions – “they’re researching something that isn’t present in nature, so maybe the findings might not be as accurate as if they were using an actual real animal that’s produced in nature”(P19), and c) human errors and the potential for mistakes to be made in laboratories – “People are not naturally conscientious so it’s very easy to make a mistake” (P1 – ACC). Other concerns were less specific to the use of GM animals in biomedical research, and related to general feelings of unease about genetic modification as a practice, with some participants stating that they think genetic modification is unnatural (with a fear that “nature will bite back,” P9 – ACC), or violates the ‘integrity’ of animals. One member of the public expressed concerns that motives of corporate greed are driving an increase in the creation of GM animals – 79  “the only reasons I’ve ever known of why people genetically modify animals is for corporate reasons…I think it’s done out of greed” (P19). Other more general comments about genetic modification as a practice included, a) concerns over the “power to design life” (R20 – no GM), b) concerns over equity of GM animals and their products, especially in relation to the global farming industry – “there could be issues where, and we’re getting onto a more difficult topic here, where we’re talking about equity between developing world farmers and the developed world biotechnology companies and such” (R6), and c) concerns about ethical practice – “in terms of ethics I don’t think we have the right to alter animals’ genetics for our own purposes. I don’t think that’s a good thing to do” (P18). Some concerns were related to negative consequences that could result from genetic modification. For example, a) concerns about the containment of GM animals and negative effects on wild populations and the environment if they escape – “I think we have to be very careful about their inadvertent release into the wild. I think that’s a big concern” (R2 – GM), b) unknown consequences to laboratory animals specifically – “we don’t know what’s going to happen when we knock down one gene in a mouse, we have absolutely no idea, so there’s a risk involved” (T10) – or general unknown consequences – “I’m not sure we understand the long term effects of what it can do…I don’t know if scientists know that, but I imagine they don’t” (P18), and c) animal welfare concerns because of the unpredictability of genetic modification – “the phenotypes arising can sometimes be surprising and the impact on the animal unforeseen for protracted periods of time” (T12). 5.3.3. Animal welfare and the Three Rs When asked about their concerns regarding the creation and use of GM animals in research, participants tended to comment that animal welfare concerns depend on the specific genetic modification under investigation. One researcher raised the issue that there are perhaps greater welfare concerns with non-GM animals because there is less scrutiny over the animal use protocols and the welfare assessment of the animals. However, all participants recognized the importance of reducing harms to all animals in laboratories, including GM animals. When directly asked about the Three Rs, some participants (researchers and research technicians in particular) raised issues that could be viewed as obstacles to improving animal welfare and implementing the Three Rs, while others raised issues that could be seen as opportunities for animal welfare and Three Rs improvements.  80  Obstacles to improving the Three Rs One research technician was of the opinion that Three Rs are not being implemented with regard to GM animals –“I just don’t think [the Three Rs] are being followed” (T13). This particular research technician went on to comment that there is a lack of ability to share animals, and a lack of communication so Three Rs implementation is being restricted by politics – “It’s really hard to give another scientist your mice after you’re done with them, for political reasons…I think people’s intentions are sometimes really good, but I don’t think that politics allows for these intentions to be followed through…there is definitely a failure to communicate between different places” (T13). Two researchers also expressed the belief that there are no nonanimal alternatives to genetic modification and that it “cannot be mimicked by anything else” (R4 – GM). These issues challenge the principles of replacement and reduction. Several participants raised issues related to the high number of breeding animals used, and the high number of surplus animals generated during genetic modification procedures – “[genetic modification is] so inefficient right now…I definitely think it needs to get more efficient than it is” (T10). One researcher recognized the increasing use of zebrafish in particular – e.g. “There’s so much being done now in basic research with transgenic zebrafish that we’re going to see quite an increase in the use of zebrafish” (R7 – GM), while one of the research technicians reflected on the need to maintain heterozygote animals, which may increase animal numbers – “say a gene knockout could lead to infertility, we need to breed animals that are not homozygous for that knock-out, so in that situation it’s going to increase the number of animals that you need because you’re going to generate animals that aren’t complete knock-outs” (T17). These points identify challenges to the principle of reduction. Some participants recognized that genetic modification is unpredictable in nature, and unanticipated welfare concerns may arise. As a result, it may be challenging to implement humane endpoints at the planning stages of creating new GM animal lines: “what happens when you start to genetically modify animals?…are there different endpoints?…are there different levels of stress that we’re not aware of?” (P9 –ACC), and, “you’ve got a whole class of genes where you don’t know what you’re going to get when you disrupt it. So, because of that it’s not easy to anticipate the kind of suffering that the animal is going to go under” (R2 – GM). One ACC community representative also mentioned that some laboratories require that GM mice are kept in ventilated caging systems – “…they’re in a whole section by themselves, in cages that are ventilated. They do have some type of enrichment but they certainly don’t have the type of living conditions that the regular [laboratory animals] have…I often wonder what it’s like to live in a 81  ventilated rack” (P9 – ACC). Ventilated caging, and the unpredictable effects of genetic modification on the animal’s welfare pose challenges to the principle of refinement. There was also concern that animal welfare is not broadly appreciated in the scientific community. One research technician felt that little attention was paid to suffering due to genetic modification if the phenotype was deemed ‘normal’ for that particular line of GM animals. Opportunities for improving the Three Rs GM animal use in research may be beneficial if it reduces the number of animals needed for a given experiment through reducing the variation between animals – “when I think of animal numbers in relation to the use of GM animals, I think it decreases the numbers of animals because it decreases the variability” (T14). One researcher also saw it as an advantage that genetic modification allows the use of mice rather than larger animals – “mice are used as the predominant model organism, very often because you can do genetic modification, and you can use mice instead of using larger animals, which is always an advantage” (R8 – GM). This participant went on to elaborate that using mice is advantageous because they are less expensive to house, and they are less contentious to use in research compared to primates or companion species. Most participants saw the importance of increased welfare assessment and monitoring for GM animals. Nearly all participants identified the need for a mechanism to act quickly if welfare concerns arise, and some of those working in laboratories indicated that many of their concerns about animal welfare had been addressed now that post-approval monitoring has been implemented via revised CCAC requirements – “There has been a lot of positivity about postapproval monitoring because of the unanticipated welfare concerns that might arise with genetically modified animals when you’re creating a new line” (T17). Some participants also saw the opportunity for improving animal user training specific for GM animals and procedures, for example, handling and genotyping. For example, one researcher stated, “with genetics there’s a specific approach to genotyping that animal, and phenotyping them, and a lot of that isn’t taught in our animal facility” (R2 – GM). 5.3.4. Regulation Participants were divided on whether separate regulations were required for GM animals in research. Those in favour of current regulation agreed that regulation is necessary and effective, and that there is more scrutiny for GM animal protocols. Specifically, participants mentioned the importance of third-party involvement, ethical review, and post-approval monitoring. Those who were more critical of regulations raised concerns over a perceived increased regulatory burden 82  related to GM animal creation and use in research. These participants (mainly researchers) claimed that current guidelines require a lot of resources (that may be unavailable) and that the paperwork involved may actually distract from animal welfare –“In the university often the scientists are ending up filling out all those forms, so how many mice, and which treatment, and that’s become cumbersome…scientists are often very displeased with all the paperwork involved” (R4 – GM). One researcher also stated that increasing the regulatory burden might result in researchers taking their work to countries with less regulatory oversight. One researcher pointed out that genetic models are not always well described on animal care protocols, potentially making ACCs less able to make decisions about whether to approve protocols involving either the creation or use of GM animals. In addition, one research technician mentioned that while ACCs may be generally effective, they may be naive about the actual numbers of animals used to create GM animal models – “I think a lot of the members of our animal care committee may not have a full grasp of what goes on because the animal numbers that they are approving are the actual models, and I don’t think they really know how much it takes to get to that point. All they really see is the number of animals that are going to be ordered and number of animals that are going to be used, but that might be seven animals and seven animals, but in the time between the seven and the seven it goes up to two hundred and fifty at some point because they’re breeding, and breeding, and intercrossing” (T10). This may also be seen as a challenge to the principle of Reduction. Two participants (both community representatives on ACCs) said that they did not trust the oversight system in place in Canada, saying that it is too easy for researchers to be in noncompliance and that there needs to be better feedback to the ACC regarding the outcomes of approved studies so that ACCs can make better decisions about similar studies in the future. This was deemed especially important for newly created GM animal models which have unanticipated outcomes. One specific concern raised regarding regulation was the issue of public reporting of animal numbers. Two participants (one ACC community representative and one research technician) expressed unease about breeding animal numbers, as well as the lack of publicly available data reporting the numbers of animals culled – “I think the numbers are significantly higher than what’s being reported” (P5 – ACC), and “Euthanasia records would probably be the best way to capture the [animal] use” (T10).  83  Limits to genetic modification When asked about what limits can or should be placed on genetic modification, all participants responded by saying that there should be limits placed on the pain and distress (or suffering) caused by genetic modification – e.g. “I think that the seeking out of knowledge does not necessarily give us permission to impart such a derangement to an animal that it would encounter excruciating pain or distress” (T12), and “we shouldn’t be creating pain just for the sake of accomplishing something that’s dubious at best.”(P15 – ACC). However, some researchers and research technicians recognized that this is challenging due to the unpredictable nature of genetic modification. Three participants also agreed that there should be limits to how much of an animal’s nature is changed by genetic modification, but their comments around that point were not clearly articulated. For example, one member of the public commented, “are we genetically modifying animals to be less than what they are?” (P16), while a researcher stated, “I think the limits are….it’s hard for me to put into words because I would want to look at that being as something with it’s own life to live…I don’t know to what extent that mouse [expressing green fluorescent protein] would be aware of the light that is coming out of him, I don’t know how it would affect his sense of security for instance” (R20 – no GM). Additional concerns were raised in response to being asked about what limits should be placed on genetic modification. One researcher expressed concern regarding the extent to which animals are ‘humanized’ – “I guess it would be kind of revolting to think that you might be able to put together a set of characteristics that would define something as recognizable as a human being” (R2 – GM). While a member of the public wanted there to be limits on the types of GM animal that are created – “the benefits have to be very explicitly detailed:… what are the reasons for doing this, and any dangers that might be posed by this, are they outweighed by the benefits?”(P16). 5.3.5. Public awareness and engagement All participants were of the opinion that the public is uninformed about genetic modification practices, especially in laboratories. Many participants were of the opinion that the scientific community should improve communication with the public, and that there is a need for greater openness and use of lay language regarding animal research in general, and genetic modification specifically. In addition, some participants recognized the need for openness, and better education about animal research and genetic modification at universities and colleges – “I do think we need to be a little more transparent and, you know, educate people about how animals are used for research…I think high schools are a good place to start” (T17). 84  Tied to this, many participants expressed concern about the general lack of information available regarding the creation and use of GM animals in research. Participants used words such as ‘secret’, ‘confidential’, and ‘private’ to describe animal research. When asked about public accountability, all participants agreed that there should be full disclosure of animal research, including the creation and use of GM animal models, and specifically animal numbers. When asked what the public thinks about genetic modification, participants were of the opinion that there are varying levels of acceptance among the general public. They linked low levels of public acceptance to fear, religion, broader worldviews about ‘nature’, or an emotional response; for example one researcher commented, “I think the public has got the idea that if you genetically modify something you’ve created a monster” (R11 – no GM). In contrast, they linked higher levels of acceptance with the belief that there are no alternatives to using GM animals to create certain disease models, the utility of products, and personal experience of the benefits that can be gained from animal research, specifically health issues that may be better treated by developing drugs in GM animal models. Overall participants expressed a belief that the public does not know enough to be fully accepting of genetic modification. However participants seemed to agree that higher levels of acceptance could result from better public education and engagement on genetic modification. In particular, participants felt that public engagement should focus on the creation and use of GM animals in biomedical research since their use is legitimized by having extra benefits (i.e. more accurate disease models, which enable the developments of more targeted therapeutics); e.g.“ the first thing that will change their perception is direct benefit” (R4 – GM). Nearly all participants expressed the need for public engagement and open discussions about the genetic modification of animals and their future applications. One participant noted “a balanced discussion, that includes benefits, is important if industry and the scientific community is trying to have the public come along with them on this journey” (P16) and “we need to make decisions as a country.” 5.4. Discussion Participants were more supportive of GM animal creation and use for biomedical research (due to the perceived human health benefits) than for food production. This mirrors the findings of other studies relating to both animal use in general (Knight et al., 2003), and GM animals specifically (Gaskell et al., 1998, 2000; Lassen et al., 2006; Royal Society of Canada, 2001; Schuppli & Weary, 2010). Models explaining this difference in support for the different applications of GM animals suggest that when people (particularly non-experts) consider GM 85  food animals they may develop what is known as a lexicographic process, where an important attribute (i.e. risk or no-risk) dominates the decision (Gaskell et al., 2004). In contrast, when people consider GM animals used in biomedicine, many of the commonly stated risk factors are removed (e.g. there are limited environmental concerns because the animals are housed in biosecure facilities, and the animals will never be consumed), enabling the use of the ‘expected utility method’, which involves consideration of all the benefits as well as the costs (Gaskell et al., 2004; Costa-Font, 2008). Participants perceived several key challenges for implementation of the Three Rs. One important issue raised was the lack of communication between facilities and lack of sharing of animals and data. Animal welfare organizations like the RSPCA in the UK also recognize these challenges and actively encourage data and GM animal sharing between different departments and institutions (e.g. Osborne et al., 2009a). Data and animal sharing will facilitate implementation of reduction, since different laboratories and institutions will not be recreating the same lines of GM animals. Policy makers could intervene, and more specifically push for sharing of animals and data. From participant comments it seems that there is a tension between reducing animal numbers in a given experiment through the use of GM animals (because there is reduced variability between the animals under study), versus the high numbers of animals required to create those GM animal models in the first place. In a previous study it was shown that pronuclear microinjection, which is a relatively inefficient technique for creating new GM animal lines, is the most commonly used technique (Ormandy et al., 2009). This highlights a potential target for policy makers to push for the use of more efficient genetic modification techniques (e.g. gene targeting). Participants highlighted the importance of public accountability, with some participants noting that, in Canada, breeding animal numbers are not required to be reported. This becomes a concern with GM animals because the inefficiency of genetic modification procedures results in high numbers of breeding and surplus animals (Robinson et al., 2003). To date, no country provides the public with numbers of breeding animals used in laboratories: numbers tend to focus on the numbers of animals used in experiments. For example, the UK reports the number of animals used in scientific procedures, and since they consider genetic modification a scientific procedure, they count numbers of GM animals. However, the numbers reported do not capture the total numbers of animals needed to create and maintain the GM animals of interest, or the surplus animals that are created that do not carry the genetic alteration required. 86  Participants also noted that humane endpoints might be difficult to establish in advance due to the unpredictable nature of genetic modification. This highlights the need for robust welfare assessment strategies that allow for flexibility and fast responses to welfare concerns that may arise. Welfare assessment strategies for genetically modified animals have been refined in the UK (Wells et al., 2006), with the use of electronic databases to record welfare information for individual animals (J. Bussell and M. Gardiner, personal communication), and the development of a standard vernacular to describe the physical characteristics (phenotype) of new GM animal lines, particularly mice (www.mousewelfareterms.org). Participants, particularly researchers, saw opportunities to improve animal user training that is specific to GM animal models and procedures such as genotyping. Such training may help to implement the principle of refinement, as well as improving the efficiency of genetic modification (therefore reducing the number animals required to produce the GM animals of interest). These findings also echo discussion from a recent workshop that was held to engage Canadian ACC members on the creation and use of GM animals in research (Ormandy, 2010). Participants also discussed the secrecy surrounding animal research, including the use of GM animals. Most participants felt that more effort should be put into public engagement on issues relating to animal research and the creation and use of GM animals. In terms of public engagement, participants were in agreement that more effort is needed to promote dialogue with the public on issues related animal research. Decisions about the creation of new GM animals are made with little engagement from the public. For GM food animals in particular, it is clear from many different studies that there is a lack of public support (Frewer et al., 1997; Gaskell et al., 2000; Priest, 2000; Lassen et al., 2006; Schuppli & Weary, 2010). If more effort had been made to engage to public on the end application of GM food animals, and their benefits, perhaps the resources that were devoted into creating animals like the Enviropig™ (which was recently refused for commercial agriculture on the grounds that there was not enough consumer acceptance, and the risks were still unknown – Schmidt, 2012) could be put into technologies that are more likely to resonate with public values. This highlights the need for policy makers to develop ways of engaging the public on these issues. 5.5. Conclusions This study shows that there is support (from researchers, research technicians and members of the public) for the use of GM animals in biomedical research as long as there are tangible human health benefits and the Three Rs are implemented. This highlights the need for 87  policy makers to push for better implementation of the Three Rs, particularly as GM animals are often used in research. Several key efforts could be made: a) encourage researchers to share data and animals more freely, b) encourage improved animal user training, specifically with regard to genetic modification techniques, c) collect and publish breeding animal numbers as well as animals used in experiments, and d) develop robust engagement strategies that allow input from researchers, research technicians, and members of the public and promote a balanced discussion of the issues that inform policy development.  88  6. General discussion and recommendations 6.1 Introduction This thesis opened with a critical review of existing literature on public attitudes toward the human use of animals, and in particular, animal-based research. Gaps in the current literature were identified, and these gaps were addressed in the subsequent research chapters. The research study presented in Chapter Two mapped worldwide trends in animal-based research, and showed that the use of animals in research is increasing, in part, due to the increasing use of genetically modified (GM) animal models (especially mice and zebrafish). Given this shift in the use of animals, the following research chapters focused on public attitudes towards the creation and use of GM animals in research. Chapters Three and Four described two online engagement experiments that explored attitudes towards the use of GM animals in agricultural and biomedical research (with additional focus on regulation, invasiveness, and species used in research). Together, the findings indicate that participants found the creation and use of GM animals in biomedical research more acceptable than the creation of GM food-production animals. In addition, regulation was a significant predictor of support for the invasive research, but not for the use of GM animals. These differences may be due to different decision-making processes used by participants. Surprisingly, no difference in support was found between the use of zebrafish and mice in biomedical research. Chapter Five further explored attitudes towards the use of GM animals in research through a series of semi-structured interviews with researchers, research technicians, and members of the public. The findings again illustrate greater acceptance of GM animals used for biomedical purposes versus for food. The study also highlighted key issues in the creation and use of GM animals in research (e.g. data and animal sharing, animal user training, etc.) that are important for policy makers to address. A recent paper on animal research states “Let us recall the need for constantly monitoring societal ethical concerns, and staying abreast of them in our actions” (Rollin, 2010, p. 10). In keeping with this statement, this thesis has: a) given an overview of the public-attitudes literature relating to animal research, b) mapped worldwide trends in animal research to identify the contribution of genetically modified animal models to those trends, c) presented three separate studies that explore societal attitudes and ethical concerns related to the creation and use of GM 89  animals in research, and d) developed new methods for online research that can be used to get greater public participation in policy debates. 6.2. Limitations and successes of the thesis research 6.2.1. Reporting of animal numbers While conducting the bibliometric study described in Chapter Two, the data collection strategy needed to be altered. The initial aim was to document the numbers of animals reported in each journal article; however, for many of the articles, the reporting of animal numbers was inadequate. As a result, the numbers of articles reporting animal use were recorded, rather than the numbers of animals per article. This limitation to the bibliometric study highlights a need for more accurate reporting of animal-based research, which is discussed in more detail in section 6.3.1. 6.2.2. Addressing the criticisms of online public engagement Chapters Three and Four describe online public engagement experiments. Some of the criticisms of online public engagement tools or surveys include sampling bias (due to selection bias), participant drop out, and the need for participants to have good computer skills. The first criticism, sampling bias, refers to the tendency of a sample to over- or underestimate a population parameter. This can occur through different types of selection bias, for example, inadequately representing some members of the population in the sample (undercoverage error), non-response bias, or voluntary selection bias. The online engagement experiments described in Chapters Three and Four used non-probability-based sampling, meaning that sampling bias is a potential limitation. Participants were not randomly selected from a defined total population, and did not have a known, non-zero chance of being selected. This may lead to undercoverage bias (underrepresentation of certain members of the total population) and voluntary selection bias. Chapters Three and Four did not the aim to make inferences about the general public or a defined total population. Rather, the aim was to examine the differences between two or more experimental treatment groups. However, voluntary selection bias may have resulted in an overrepresentation of individuals with strong opinions. A second challenge to online public engagement is participant drop-out within the experiment. As the online engagement instrument required participants to write a comment before they could proceed to the next question, there was a risk that participants would quit before the end of the experiment. However, the drop-out rate in both online engagement studies presented in 90  Chapters Three and Four was relatively low (2.4%). In addition, the comments that people gave were useful, with 93% of participants’ comments being substantial (average comment length = 25 words; range = 3-60 words). Interestingly, those participants that indicated either strong opposition or support tended to give shorter comments; in contrast, those participants with more nuanced views tended to give longer comments. Finally, online engagement tools and surveys have also been criticized for requiring participants to have good computer skills. Researchers cannot assume that all members of their target population will have the skills necessary to navigate an online engagement experiment. Thus, as part of the demographic questions in Chapters Three and Four participants were asked about their level of computer literacy: 73.5% of participants self-identified as “very good” when asked to rate their skill at using the Internet. However, a lack of computer skills is still a limitation in terms of representing societal demographics, and may in part explain the over-representation of younger participants in the Chapters 3 and 4. This may also contribute to undercoverage bias. An additional limitation of the online public engagement studies described in Chapters Three and Four is that although quantitative and qualitative data were collected together, there remains room for misinterpretation during analysis because it is not possible to follow-up with individual participants to their responses. For example, if an unsupportive participant writes that, “genetic modification is wrong,” it is difficult to understand why they think genetic modification is wrong. For this reason, the follow-up study using interview-based methods was conducted which allowed a deeper exploration of people’s attitudes through face-to-face interviews. 6.2.3. Limitations to interview study Qualitative research provides rich, in-depth data that are often missed in quantitative studies. However, there are several limitations to the study described in Chapter Five, including the use of telephone interviews, the use of only one researcher to carry out all the interviews, data analysis and interpretation, and the small sample sizes for the various stakeholder groups represented. The first limitation to the interview study described in Chapter Five was that some interviews were conducted face-to-face and others were conducted over the telephone. Face-toface interviews are, arguably, richer experiences, since the interviewer can see how the participant is responding in terms of their body language as well as through their verbal comments (Opdenakker, 2006). Face-to-face interviews also allow a better connection between the interviewer and participant through eye contact. However, the ability to conduct interviews over the telephone allowed participants to be recruited and interviewed across five different provinces 91  without travel expenses. Telephone interviews may also overcome barriers that participants may face in terms of participating in face-to-face interviews. The second limitation is that the interviews were all conducted, transcribed, analysed and interpreted by one person. This lack of triangulation with other researchers may have affected the validity and reliability of the analysis and interpretation of the interview data, and as such, may be seen as a limitation. However, having only one person carry out all stages of the interview research process allowed for continual reflection on the interview findings: data analysis and interpretation started at the very first interview, and was an iterative process throughout the study. Finally, the aim of the interview study was to represent a diversity of views, and the different stakeholders represented were researchers, research technicians, and members of the public. Since only twenty interviews were conducted, this meant that a minimum of five, and a maximum of eight participants were recruited from each stakeholder group. This limits any claims that can be made about differences between the views raised by the different stakeholders. 6.3. Recommendations 6.3.1. Improve scientific reporting The reporting of animal-based science has recently been criticized by scientists who are interested in how the Three Rs are applied. For example, a recent study of editorial policy by the RSPCA in the UK (Osborne et al., 2009b) assessed 12 criteria that they felt should be required by journals. One of the criteria was to report the number of animals and the species used. The maximum possible score using the criteria was 12, but the highest score achieved was 9 (achieved by only one of 236 journals included in the study). The average score, taking all journals into account, was 1.51. In relation to this issue, the reporting of statistical methods used in animal-based science has been criticized. A study by Kilkenny et al. (2009) unearthed several key problems with the scientific reporting of animal research. In a strategic review of 271 different articles describing animal use data (with 48 of these articles examined in greater detail), Kilkenny et al. (2009) identified many articles that omitted basic details about the strain, sex, age, and weight of the animals used. They also found that in 6% of the articles studied the number of animals used in the main experiment could not be determined. Reporting animal numbers is essential so that the biological and statistical significance of the experimental results can be assessed or the data re-analyzed, and is also necessary if the experimental methods are to be repeatable. The findings of Osborne et al. (2009b) and Kilkenny 92  et al. (2009) highlight the need for more stringent scientific reporting. Recent efforts have been made to encourage this. First, the Animal Research: Reporting In Vivo Experiments (ARRIVE) guidelines issued by the National Centre for the 3Rs (NC3Rs) in the UK (Kilkenny et al., 2010), the ILAR Guidance for the Description of Animal Research in Scientific Publications (ILAR, 2011), and the gold standard publication checklist (GSPC) (Hooijmans et al., 2010) outline key information to be included in scientific articles that report animal use. The aim of the guidelines is to improve the reporting of research using animals. To date, 85 scientific journals and 6 UK funding bodies endorse the use of the ARRIVE guidelines. Second, the Eighth World Congress on Alternatives and Animal Use in the Life Sciences held in Montréal in 2011, asked participants to agree to an international declaration on the synthesis of evidence to advance the Three Rs principles in science. The Declaration calls for improvements in the planning, executing, reporting, reviewing and translating animal research (Montréal Declaration, 2011; Leenaars et al., 2012) and, in part, seeks the same goals as the ARRIVE guidelines, in that it requires improved reporting of animal research. Based on the unforeseen limitations to the study presented in Chapter Two, and the findings from Chapters Three, Four, and Five, in which participants indicated that they value public accountability, the international adoption of the principles outlined in the ARRIVE guidelines, ILAR, GSPC, and the Montréal declaration are recommended. 6.3.2. Improve data and animal sharing It has been argued by Canadian animal care committee members (Ormandy, 2010) and elsewhere (Osborne et al., 2009a) that if data and animals are not shared freely within the scientific community, there is a risk of unnecessary repetition of experiments, and unnecessary use of animals. However, researchers are often in competition for grants, or for publication in high impact journals. As a result, a culture of confidentiality surrounds certain research practices. This is especially evident with the creation of new GM animals, where researchers may have a desire to patent their ideas, techniques, and even the animals they have created. The use of GM animals has already been identified as challenging the Three Rs principle of reduction, since the use of GM animals has been shown to contribute to a reversal of the downward trend in overall reported animal research (Ormandy et al., 2009). The lack of data and animal sharing reinforces this challenge to reduction. As such, a key recommendation is for policy makers to encourage data and animal sharing between researchers. In a recent UK document, it was recognized that “a significant barrier to sharing is the dissemination of knowledge on what is available to share” (Osborne et al., 2009a, p. 11). However, several GM animal databases exist for 93  researchers to ensure that their GM animal lines are quickly and easily accessible, for example, the International Mutant Strain Resource (IMSR - http://www.findmice.org//index.jsp). In the UK there is also a Mouse Locator Network (MLN), which is an e-mail network through which requests for GM animals are posted and disseminated. Policy makers outside the UK should consider encouraging similar efforts to advance data and animal sharing. 6.3.3. Improve recording and reporting of national research animal statistics In Chapter Five some participants raised concerns about how research animal numbers are collected and reported out in the publicly available annual animal statistics, in particular the recording of breeding animal numbers. This issue becomes particularly pertinent with the creation and use of GM animals because of the high numbers of breeding animals required to create the GM animal of interest. Currently, the animal numbers that are reported to the public only cover the number of animals used in scientific procedures. With the increasing use of GM animals, there has been a corresponding increase in the number of breeding animals kept in animal facilities. Given that the efficiency of genetic modification techniques has been estimated at between 1-30% (Robinson et al., 2003), current animal numbers reported are likely to be conservative. This creates problems for public accountability, as many more animals are being housed in laboratories than are being reported to the public. A key recommendation (for Canada, and internationally) is to start to collect and report breeding animal numbers alongside the current animal use statistics. 6.3.4. Improve animal welfare assessment In Chapter Four participants raised welfare concerns with induced mutagenesis techniques, in particular the use of ethyl-N-nitrosourea (ENU) to induce mutations in zebrafish (Ormandy et al, in press b). The perception was that this procedure is painful, but no research date has explored the potential pain that might result from the procedure. Participants in Chapters Three, Four and Five also identified the importance of having robust welfare assessment strategies in place. This was deemed particularly important for newly created GM animals, since there is an element of unpredictability to genetic modification techniques and researchers cannot always anticipate the effects. Welfare assessment is an important aspect of refinement and Three Rs implementation, and a set of key recommendations on the welfare assessment of GM mice have been made by the National Centre for the Three Rs (NC3Rs) in the UK (Wells et al., 2006). In particular, the NC3Rs encourages a) structured welfare assessment for new GM animal lines (and GM animals that are newly introduced to a given establishment), and b) the use of a GM animal “passport” to 94  detail the welfare assessment history of each individual GM animal or line, so that when GM animals are transported to different locations within or outside the institution in which they are created, their welfare assessment information travels with them. This allows animal care staff to be able to provide for the needs of any GM animals that have impaired welfare as a result of their genetic alteration. A key recommendation is the refinement and adoption of these practices in Canada. In particular, special attention should be paid to welfare assessment of fish in laboratories. 6.3.5. Supplement the Three Rs The Three Rs are widely seen as the underpinnings of humane animal research, and aim to reduce the harms experienced by animals in research. However, the creation and use of GM animals challenges the Three Rs principles. This calls into question whether the Three Rs are fully applicable to the creation and use of GM animals in research. Such questions have been raised elsewhere: notably, Schuppli and Fraser (2004) explored how we might supplement the utilitarian basis of the Three Rs with principles based on deontology and relational ethics. The Three Rs are an important component of ethical animal experimentation, but should not be thought of as the only approach to carrying out ethical research. For example, an alternative theoretical framework, the ‘five freedoms’, is used in farm animal welfare. The five freedoms outline five key criteria that should be met if good animal welfare is to be achieved: 1) freedom from hunger and thirst, 2) freedom from discomfort, 3) freedom from pain, injury and disease, 4) freedom to express normal behaviour, and 5) freedom from fear and distress (Brambell Report, 1965). In New Zealand, the five freedoms are used as part of the ethical framework for humane animal research alongside the Three Rs: animals are classified according the ‘domains of welfare compromise’ that result from being research subjects (Mellor & Reid, 1994). Given that the Three Rs are currently being challenged by the creation and use of GM animals, it would be beneficial to consider domains of welfare compromise (or the ‘welfare status’ of animals) as a supplement to the Three Rs in other countries as well. In addition, it may be useful to think in terms of positive rights for animals (such as the right to be taken into consideration when making decisions) and our relationship with them. Kymlicka and Donaldson (2012) challenge us to see animals as fellow citizens (or “denizens”) in society. Given that our fellow humans are also used as research subjects, some of the key principles (e.g. justice) that we apply to human research subjects may also be relevant if we are to view animals as fellow citizens.  95  One of the key aspects of using humans as research subjects is the voluntary provision of informed and ongoing consent. However, there are instances where human subjects are not able to provide such consent for themselves. In these cases a next of kin or someone with a close relationship to the person being asked for their consent is able to provide third-party consent. Animal subjects are in a similar position in that they cannot provide their consent. However it may be possible to draw on relational ethics and obtain third-party consent from the individuals that have the closest, day-to-day relationship with the animals under study; namely, the research technicians and animal care staff. As mentioned in Chapter Five, research technicians and animal care staff are underrepresented when making decisions about animal research, and when developing policy. It follows that a key recommendation would be to make research technicians part of the minimum quorum for animal care committees, and to invite them to be part of the committees that draft national guidelines on the use of animals in research. 6.4. Conclusions Improvements in animal-related regulation and public policy often arise from changes in societal attitudes and opinions (Kirkwood & Hubrecht, 2001). However, current mechanisms for including public opinion in animal research policy are lacking. The research presented in this thesis has provided a means of exploring people’s attitudes about the creation and use of different GM animal models in research, with a view to informing animal policy. By improving public engagement on issues related to animal research, and during decision-making processes, democratic decisions can be made that better reflect citizen values. In time, this should increase democratic legitimacy in a controversial area of research. 6.5. Next Steps A valuable avenue for future research would be to establish how citizens in a democratic society want to be engaged in decisions about animal research and animal policy. This could be achieved through focus groups or deliberative engagement events, which aim to encourage a diversity of views to be represented and may provide a more educative experience for all participants than online engagement or interviews.  96  Bibliography Adams, C. J. (1994). Bringing peace home: A feminist philosophical perspective on the abuse of women, children and pet animals. Hypatia, 9: 63-84. Ajzen, I. (1991). The theory of planned behaviour. Organizational behaviour and human decision processes, 50: 179-211. Ajzen, I., & Fishbein, M. (2000). Attitudes and the attitude-behaviour relation: reasoned and automatic processes. European Review of Social Psychology, 11(1): 1-33. Aldhous P., Coghlan A. & Copely, J. (1999). Let the people speak. New Scientist, 2187: 26. Allport, G. W. (1954). The Nature of Prejudice. Beacon Press: Cambridge, MA. Amatruda, J. F., Shepard, J. L., Stern, H. M. & Zon L. I. (2002). Zebrafish as a cancer model system. Cancer Cell, 1: 229-231. American Dietetic Association. (2003). Position of the American Dietetic Association and Dieticians of Canada: vegetarian diets. ADA Reports, 103 (6): 748-765. http://www.vrg.org/nutrition/2003_ADA_position_paper.pdf Accessed on September 15, 2011. Arluke, A. B. (1988). Sacrificial symbolism in animal experimentation: Object or pet? Anthrozoös, 2(2): 98-117. Australian Association for Humane Research (2004). Statistics – Animal use in research and testing, Australia. http://www.aahr.asn.au/statistics.html Accessed Oct 15, 2008. Baldwin, E. (1993). The case for animal research in psychology. Journal of Social Issues, 49: 121-131. Barthold S. W. (2002). “Muromics”: Genomics from the perspective of the laboratory mouse. Comparative Medicine, 52(3): 206-223. Batt, S. (2009). Human attitudes towards animals in relation to species similarity to humans: a multivariate approach. Bioscience Horizons, 2(2): 180-190. Bayne, K. (2008). Standards for the rearing environment of laboratory animals in the United States. Alternatives to Animal Experiments, 14 (special issue): 47-50. Birke, L., Arluke, A. & Michael, M. (2007). The Sacrifice: How Scientific Experiments Transform Animals and People. Purdue University Press: West Lafayette, Indiana. Blackshaw, J. & Blackshaw, A. W. (1993). Student perceptions of attitudes to the human animal bond. Anthrozoös, 6: 190-198. 97  Boogaard, B. K., Oosting, S. J. & Bock, B. B. (2006). Elements of societal perception of farm animal welfare: a quantitative study in The Netherlands. Livestock Science, 104: 13-22. Bowd, A. D. & Bowd, A. C. (1989). Attitudes toward the treatment of animals: A study of Christian groups in Australia. Anthrozoös, 3: 20-24. Braithwaite, V. A. & Huntingford, F. A (2004). Fish and welfare: do fish have the capacity for pain perception and suffering? Animal Welfare, 13: S87-S92. Brambell Report (1965) Report of the Technical Committee to enquire into the welfare of animals kept under intensive livestock husbandry systems. HMSO: London, UK. Broida J., Tingley L., Kimball R., & Miele, J. (1993). Personality differences between pro- and anti-vivisectionists. Society and Animals, 1: 129-144. Brown, G. A. J. & Corbin T. J. (2002). Oocyte injection in the mouse. In Transgenesis Techniques: Principles and Protocols (ed. A. R. Clarke), pp. 39-70. Humana Press: Totowa, NJ. Buehr, M., Hjorth, P. J., Hansen, A. K. & Sandøe, P. (2003). Genetically modified laboratory animals – what welfare problems do they face? Journal of Applied Animal Welfare Science, 6(4): 319-338. Burgess, M. M. (2012) Governance of nonhuman animals at the University in a democratic society. University of British Columbia Green College Lecture Series: Bringing the collective together: nonhumans, humans and practice at the University. Vancouver, Canada, March 29, 2011. Burgess, M. M. and Tansey, J. (2008). Democratic deficit and the politics of "informed and inclusive" consultation'. In E. Einseidel and R. Parker (Eds) Hindsight and Foresight on Emerging Technologies, pp. 275-288. UBC Press: Vancouver. Buss, D., Craik, K. & Dake, K. (1986). Contemporary worldviews on technology and resources: Their political attitudinal content. Journal of Applied Social Psychology, 15. CCAC (2011). About the CCAC. http://www.ccac.ca/en/About_CCAC/About_CCAC_Intro.htm Accessed on September 15, 2011. CCAC (2006). Terms of reference for Animal Care Committees. Available at: http://www.ccac.ca/Documents/Standards/Policies/Terms_of_reference_for_ACC.pdf CCAC (2005a). Guidelines on the care and use of fish in research, teaching and testing. Available at: http://www.ccac.ca/Documents/Standards/Guidelines/Fish.pdf CCAC (2005b). Facts and Figures: Trends in Animal Use. http://www.ccac.ca/en/Publications/New_Facts_Figures/trends/trends_intro.htm Accessed March 21, 2008.  98  CCAC (1997). Guidelines on: Transgenic Animals. http://www.ccac.ca/en/CCAC_Programs/Guidelines_Policies/GDLINES/TRANSGE1.HTM Accessed June 25, 2007. Chandroo, K. P., Duncan, I. J. H. & Moccia, R. D. (2004). Can fish suffer?: perspectives on sentience, pain, fear and stress. Applied Animal Behaviour Science, 86: 225-250. Christensen, J. F. & Gomilla, A. (2012). Moral dilemmas in cognitive neuroscience of moral decision-making. A principled review. Neuroscience and Biobehavioural Reviews, 36(4): 12491264. Coffey, A. & Atkinson, P. (1996). Concepts and coding. In Making Sense of Qualitative Data: Complementary Research Strategies, 26-53, ed. A. Coffey and P. Atkinson. Sage: Thousand Oaks, California. Costa P. (1997). Use of transgenic animals and welfare implications. In Welfare Aspects of Transgenic Animals – Proceedings EC Workshop (eds. L. F. M. Van Zutphen & M. Van Der Meer), pp 68-77. Springer-Verlag: Berlin. Costa-Font, M. (2008). Consumer acceptance, valuation of and attitudes towards genetically modified food: Review and implications for food policy. Food Policy, 33: 99-111. Crettaz von Roten, F. (2012). Public perceptions of animal experimentation across Europe. Public Understanding of Science. Available online at: http://pus.sagepub.com/content/early/2012/02/01/0963662511428045 Cummins, R.A. & Gullone, E. (2000). Why we should not use 5-point Likert scales: The case for subjective quality of life measurement. Proceedings, Second International Conference on Quality of Life in Cities (pp.74-93). Singapore: National University of Singapore. Daly, B. & Morton, L.L. (2006). An investigation of human-animal interactions and empathy as related to pet preference, ownership, attachment, and attitudes in children. Anthrozoös, 19: 113127. Davies, L. (2009). Public attitudes to food issues. GfK Social Research: Report for the Food Standards Agency. London, UK. http://www.food.gov.uk/multimedia/pdfs/publicattitudestofood.pdf Accessed on September 15, 2011. de Angelis, M. H., Flaswinkel, H., Fuchs, H., Rathkolb, B., Soewarto, D., Maschall, S., Heefner, S., Pargent, W., Wuensch, K., Jung, M., Reis, A., Richter, T., Alessandrini, F., Jakob, T., Fuchs, E., Kolb, H., Kremmer, E., Schaeble, K., Rollinski, B., Roscher, A., Peters, C., Meltinger, T., Strom, T., Steckler, T., Holsboer, F., Klopstock, T., Gekeler, F., Schindewolf, C., Jung, T., Avraham, K., Behrendt, H., Ring, J., Zimmer, A., Schughart, K., Pfeffer, K., Wolf, E. & Balling, R.. (2000). Genome-wide, large-scale production of mutant mice by ENU mutagenesis. Nature Genetics, 25: 444-447. de Bruijn, E., Cuppen, E. & Feitsma, H. (2010). Highly efficient ENU mutagenesis in Zebrafish. Methods in Molecular Biology, 546: 3-12. 99  Dennis M.B. (2002). Welfare issues of genetically modified animals. Institute for Laboratory Animal Research, 43: 100-109. DeTolla, L. J., Srinivas, S., Whitaker, B. R., Andrews, C., Hecker, B., Kane, A. S. & Reimschuessel, R. (1995). Guidelines for the care and use of fish in research. ILAR Journal 37, 159-173. Dietrich, H. & Schibeci, R. (2003) Beyond public perceptions of gene technology: community participation in public policy in Australia. Public Understanding of Science, 12: 381-401 Dietz, T., Frisch, A. S., Kalof, L., Stern P. C. & Guagnano, G. A. (1995). Values and vegetarianism: An exploratory analysis. Rural Sociology, 60: 533-542. Dillman, D. A. (2000). Mail and Internet Surveys: The Tailored Design Method. John Wiley & Sons: New York. Driscoll, J. W. (1992). Attitudes towards animal use. Anthrozoös, 5: 32-39. Driscoll J. W. (1995). Attitudes towards animals: Species ratings. Society and Animals, 3(2): 139150. Dohoo, I., Martin, W. & Stryn, H. (2010). Model-Building Strategies. In Veterinary Epidemiologic Research 2nd Edition, 317-334, ed. I. Dohoo, W. Martin & H. Stryn. AVC Inc: Prince Edward Island, Canada. Ecomonist (2012) The roar of the crowd: Crowdsourcing is transforming the science of psychology. May 26, 2012. Available at: http://www.economist.com/node/21555876 Eddy, T. J., Gallup, G. G. & Povinelli, D. J. (1993). Attribution of cognitive states to animals: Anthropomorphism in comparative perspective. Journal of Social Issues, 49 (1): 87-101. Einseidel, E. F. & Ross, H. (2002) Animal spare parts? A Canadian public ocnsultation on xenotransplantation. Science and Engineering Ethics, 8: 579-591 Elam, M. & Bertilsson, M. (2003). Consuming, engaging and confronting science: the emerging dimensions of scientific citizenship. European Journal of Social Theory, 6(2): 233-251. EU Directive (2010) Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010 on the protection of animals used for scientific purposes. Available at: http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2010:276:0033:0079:En:PDF Eurobarometer (2010). Science and Technology Report. European Union. http://ec.europa.eu/public_opinion/archives/ebs/ebs_340_en.pdf Accessed on September 30th, 2011.  100  Eurobarometer (2001). Europeans, science and technology. http://europa.eu.int/comm/research/press/2001/pr0612en_report.pdf Accessed May 25, 2009 Fabacher, D. L., & Little, E. E. (2000). Introduction. In: The Laboratory Fish (ed. G. K. Ostrander), pp. 1-9. San Diego CA: Academic Press. FELASA (Federation of European Laboratory Animal Science Association) (2007). Statement on a proposal from the Laboratory Animal Science Association (LASA) for the European Community to adopt retrospective reporting of scientific procedures on animals. http://www.felasa.eu/Documents/Policy_statements/FELASA%20statement%20on%20retrospecti ve%20reporting_20071113.doc Accessed March 24, 2008. FELASA (Federation of European Laboratory Animal Science Associations) (1995). Transgenic Animals – derivation, welfare, use and protection, pp 3-6. September 1992, revised February. Fenwick, N., Griffin, G. & Gauthier, C. (2009). The welfare of animals used in science: How the “Three Rs” ethics guides improvements. Canadian Veterinary Journal, 50: 523-530. Fishbein, M. (1967). Attitude and the prediction of behaviour. In M. Fishbein (Ed.) Readings in Attitude Theory and Measurement (pp. 477-92). New York: Wiley. Fishbein, M. (1963). An investigation of the relationship between beliefs about an object and the attitude toward that object. Human Relations, 16: 233-240. Frewer, L. J., Howard, C. & Shepherd, R (1997) Public concerns in the United Kingdom about general and specific applications of genetic engineering: Risk, benefits, and ethics. Science, Technology ad Human Values, 22: 98-124 Friedman, D. B. & Hoffman-Goetz, L. (2006). A systematic review of readability and comprehension instruments used for print and web-based cancer information. Health Education and Behaviour, 33: 352-373. Fujii, K., Kurata, H., Odashima, S. & Hatsuda, Y. (1976). Tumor induction by a single subcutaneous injection of sterimatocystin in newborn mice. Cancer Research, 36: 1615-1618. Furnham A. & Heyes C. (1993). Psychology students’ belief about animals and animal experimentation. Personality and Individual Differences, 15:1. Furnham, A., McManus, C. & Scott, D. (2003). Personality, empathy and attitudes to animal welfare. Anthrozoös, 16: 135-146. Furnham, A. & Pinder, A. (1990). Young people’s attitudes to experimentation on animals. The Psychologist, October: 444-448. Gallup G. G. & Beckstead J. W. (1988). Attitudes towards animal research. American Psychologist, 43(6): 474-476. 101  Gallup Poll. (2010). Four moral issues sharply divide Americans. http://www.gallup.com/poll/137357/Four-Moral-Issues-Sharply-Divide-Americans.aspx Accessed on September 15, 2011. Gaskell, G., Allum, N., Bauer, M., Durant, J., Allansdottir, A., Bonfadelli, H., Boy, D., de Cheveigné, S., Fjaestad, B., Gutteling, J. M., Hampel, J., Jelsøe, E., Correia Jesuino, J., Kohring, M., Kronberger, N., Midden, C., Nielsen, T. H., Przestalski, A., Rusanen, T., Sakellaris, G., Torgersen, H., Twardowski, T. & Wagner, W. (2000). Biotechnology and the European Public. Nature Biotechnology, 18: 935-938. Gaskell, G., N. Allum, M. Bauer, J. Jackson, S. Howard, & N. Lindsey. (2003b) Ambivalent GM nation? Public attitudes to biotechnology in the UK, 1991-2002. London School of Economics and Political Science: Life Sciences in European Society report. Gaskell, G., Allum, N. & Stares, S. (2003a). Europeans and Biotechnology in 2002: Eurobarometer 58.0. European Union: Directorate Press and Communication, Public Opinion Analysis Unit. Gaskell, G., Allum, N., Wagner, W., Kronberger, N., Torgersen, H., Hampel, J. & Bardes, J. (2004). GM foods and the misperception of risk perception. Risk Analysis, 24: 185-194. Gaskell, G., Bauer, M., & Durant, J. (1998). Public perceptions of biotechnology in 1996: Eurobarometer 46.1. In J. Durent, M. Bauer & G. Gaskell (eds) Biotechnology in the Public Sphere: A European Sourcebook, pp. 189-214. Cromwell Press: London. Gauthier, C. & Griffin, G. (2005). Using animals in research, testing and teaching. Revue Scientifique et Technique - Office International Épizooties, 24(2): 735-745. Golovan, S. P., Meidinger, R. G., Ajakaiye, A., Cottrill, M., Wiederkehr, M. Z., Barney, D. J., Plante, C., Pollard, J. W., Fan, M. Z., Hayes, M. A., Laursen, J., Hjorth, J. P., Hacker, R. R., Phillips, J. P. & Forsberg, C. P. (2001). Pigs expressing salivary phytase produce low-phosphorus manure. Nature Biotechnology, 19 (8): 741–745. Gordon, J. W., Scangos, G. A., Plotkin, D. J., Barbosa, J. A. & Ruddle, F. H. (1980). Genetic transformation of mouse embryos by microinjection of purified DNA. Proceedings of the National Academy of Science - USA, 77(12): 7380-7384. Griffin, G., Dansereau, M. & Gauthier, C. (2007). Categories of invasiveness – A precautionary approach. Alternatives to Animal Testing and Experimentation, 14: 715-720. Guillot, E., Vaugelade, P., Lemarchali, P. & Re Rat, A. (1993). Intestinal absorption and liver uptake of medium-chain fatty acids in non-anaesthetized pigs. British Journal of Nutrition, 69: 431-442. Gunnthorsdottir, A. (2001). Physical attractiveness of a species as a decision factor for its preservation. Anthrozoös, 14: 204-215.  102  Hagelin J., Carlsson H-E. & Hau J. (2003). An overview of surveys on how people view animal experimentation: Some factors that may influence the outcome. Public Understanding of Science, 12: 67-81. Hagelin, J., Hau, J. & Carlsson, H. E. (1999). Undergraduate university students’ views on the use of animals in biomedical research. Academic Medicine, 74: 1135-1137. Hagelin, J., Hau, J. & Carlsson, H. E. (2000). Attitude of Swedish veterinary and medical students to animal experimentation. Veterinary Record, 146: 757-760. Hagelin, J., Johansson, B., Hau, J. & Carlosson, H-E. (2002). Influence of pet ownership on opinions toward the use of animals in biomedical research. Anthrozoös, 15: 251-257. Haidt, J. (2002). “Dialogue between my head and my heart”: Affective influences on moral judgment. Psychological Inquiry, 13 (1): 54-56. Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review, 108 (4): 814-834. Hamm, T. E., Dell, R. B. & Van Sluyters, R. C. (1995). Laboratory animal care policies and regulations: United States. Institute of Laboratory Animal Research Journal, 37 (4): 75-78. Hampson, J. (1989). Animal experimentation – practical dilemmas and solutions. In The Status of Animals: Ethics, Education and Welfare, 100-110, ed. D. Paterson & M. Palmer. CAB International Press: UK. Harris Interactive. (2008). Vegetarianism in America. Vegetarian Times. http://www.vegetariantimes.com/features/archive_of_editorial/667 Accessed on September 15, 2011. Henry, B. & Pulcino, R. (2009). Individual difference and study-specific characteristics influencing attitudes about the use of animals in medical research. Society and Animals, 17: 305324. Herzog, H. A. (1993). The movement is my life: The psychology of animal rights activism. Journal of Social Issues, 49: 103-119. Herzog, H. A. & Burghardt, G. M. (1988). Attitudes toward animals: Origins and diversity. Anthrozoös, 1: 214-222. Herzog, H. A. & Dorr, L. B. (2000). Electronically available surveys of attitudes toward animals. Society and Animals, 8: 2-8. Herzog, H. A. & Galvin, S. (1997). Common sense and the mental lives of animals: An empirical approach. In R. W. Mitchell (Ed.) Anthropomorphism, Anecdotes and Animals. State University of New York Press: Albany. Herzog, H. A. & Golden, L. L. (2009). Moral emotions and social activism: the case of animal rights. Journal of Social Issues, 65(3): 485-498. 103  Herzog, H. A., Betchart, N. S. & Pittman, R. B. (1991). Gender, sex role orientation and attitudes towards animals. Anthrozoös, 4: 184-191. Herzog, H. A., Rowan, A. N. and Kossow, D. (2001). Social attitudes and animals. In The State of the Animals, 55-70, ed. D. J. Salem and A. N. Rowan. Humane Society Press: Washington, DC. Hills, A. M. (1995). Empathy and belief in the mental experience of animals. Reviews and research reports. Anthrozoös, 8: 132-142. Hooijmans, C. R., Leenaars, M. & Ritskes-Hoitinga, M. (2010) A gold standard checklist to improve the quality of animal studies, to fully integrate the Three Rs, and to make systematic reviews more feasible. Alternatives to Laboratory Animals, 38:162-182 Holmberg, T. & Ideland, M. (2010). Secrets and lies: “selective openness” in the apparatus of animal experimentation. Public Understanding of Science, http://pus.sagepub.com/content/early/2010/08/23/0963662510372584. Accessed Feb 10, 2012. Home Office (2000). Statistics of scientific procedures on living animals. HMSO: London, UK. Ideland, M. (2009). Different view on ethics: how animal ethics is situated in committee culture. Journal of Medical Ethics, 35: 258-261. ILAR (2011) Guidance for the Description of Animal Research in Scientific Publications. National Academies Press: Washington, DC. Irwin, A. (2001). Constructing the scientific citizen: Science and democracy in the biosciences. Public Understanding of Science, 10: 1-18. ISI Web of Knowledge: Journal Citation Reports. http://admin-apps.isiknowledge.com/JCR/JCR?SID=3ACmeBEaKDlebKjilAA Accessed March 25, 2008. Jasper, J., & Nelkin, D (1992). The Animal Rights Crusade. The Free Press: New York. Kalof, L., Dietz, T., Stern, P. C. & Guagnano, G. A. (1999). Social psychological and structural influences on vegetarian beliefs. Rural Sociology, 64: 500-511. Kane, Parsons & Associates, Inc. (1989). Survey of American adults: Report of findings (Unpublished report). Kane, Parsons and Associates, Inc: New York. Kellert, S. R. (1996). The value of life: Biological diversity and human society. Washington, DC: Island Press. Kellert, S. R. (1980a). American attitudes towards and knowledge of animals: An update. International Journal for the Study of Animal Problems, 1: 87-119. Kellert, S. R (1980b). Public attitudes toward critical wildlife and natural habitat issues. National Technical Information Service PB80-138332: Springfield, VA. 104  Kellert, S. R. & Berry, J. K. (1981). Knowledge, affection and basic attitudes toward animals in American society. National Technical Information Services, PB-81-173106: Springfield, VA. Kendall, H. A., Lobao, L. M. & Sharp, J. (2006). Public concern with animal well-being: Place, social structural location, and individual experience. Rural Sociology, 71(3): 399-428. Kilkenny, C., Browne, W. J., Cuthill, I. C., Emerson, M. & Altman, D. G. (2010) Improving bioscience research reporting: The ARRIVE guidelines for reporting animal research. PLoS Biology 8(6): e1000412. Available at: http://www.nc3rs.org.uk/downloaddoc.asp?id=1206&page=1357&skin=0 Kilkenny, C., Parsons, N., Kadyszewski, E., Festing, M. F. W., Cuthill, I., Fry, D., Hutton, J. & Altman, D. G. (2009) Survey of the quality of experimental design, statistical analysis and reporting of research using animals. PLoS ONE, 4(11): 1-10. Kirkwood, J. K. & Hubrecht, R. (2001) Animal consciousness, cognition and welfare. Animal Welfare, 10: S5-S17. Knight, S., Bard, K., Vrij, A. & Brandon, D. (2010). Human rights, animal wrongs? Exploring attitudes toward animal use and possibilities for change. Society and Animals, 18: 251-272. Knight, S. & Barnett, L. (2008). Justifying attitudes towards animal use: a qualitative study of people’s views and beliefs. Anthrozoös, 21(1): 31-42. Knight, S., Cherryman, J. & Nunkoosing, K. (2004). Attitudes towards animal use and animal mind. Anthrozoös, 17: 43-62. Knight, S. & Herzog, H. (2009). All creatures great and small: New perspectives on psychology and human-animal interactions. Journal of Social Issues, 65 (3): 451-461. Knight, S., Nunkoosing K., Vrig A. & Cherryman J. (2003). Using grounded theory to examine people’s attitudes towards how animals are used. Society and Animals, 11(4): 179-198. Knight, S., Vrij, A., Bard, K. & Brandon, D. (2009). Science versus human welfare? Understanding attitudes towards animal use. Journal of Social Issues, 65(3): 463-483. Kymlicka, W. & Donaldson, S. (2011) Zoopolis: A political theory of animal rights. Oxford University Press Inc.: New York. Lassen, J., Gjerris, M., & Sandøe, P. (2006). After Dolly: ethical limits to the use of biotechnology on farm animals. Theriogenology 65, 992-1004. Lawrence, E. A. (1989). Neoteny in American perceptions of animals. In: Hoage, R. J. (ed) Perceptions of Animals in American Culture, pp 57-76. Smithsonian Institution Press: Washington, DC, USA. Leenaars, M., Ritske-Hoitinga, M., Griffin, G. & Ormandy, E. H. (2011) Background to the Montréal Declaration on the synthesis of evidence to advance the 3Rs principles in science, as 105  adopted by the 8th World Congress on Alternatives and Animal Use in the Life Sciences, Montréal, Canada. ALTEX Proceedings of the 8th World Congress on Alternatives and Animal Use in the Life Sciences, Special Issue: 35-38. Lewison, G. (1996). The definition of biomedical research subfields with title keywords and application to the analysis of research outputs. Research Evaluation, 6: 25-36. Lyons, D. (2011). Protecting animals versus the pursuit of knowledge: Evolution of the British animal research policy process. Society and Animals, 19: 356-367. Macnaghten, P. (2001). Animal futures: Public attitudes and sensibilities towards animals and biotechnology in contemporary Britain. Lancaster: IEPPP. http://csec.lancs.ac.uk/docs/pdf/macnaghten_animal_futures.pdf Accessed on September 15, 2011. Macnaghten P. (2004). Animals in their nature: A case study on public attitudes to animals, genetic modification and ‘nature’. Sociology, 38(3): 533-551. Matthews, S. & Herzog, H. A. (1997). Personality and attitudes toward the treatment of animals. Society and Animals, 5: 169-175. Mayer, S. (2003) Science out of step with the public: the need for public accountability of science in the UK. Science and Public Policy, 30: 177-181. Mellor, D. J. & Reid, C. S. W. (1994) Concepts of animal well-being and prediciting the impact of procedures on experimental animals. In: Jenkin, G., Mellor, D. J. & Baker, R. M. (eds) Improving the Well-being of Animals in the Research Environment, pp 3-18. Glen Osmond: SA, Australia. Miles, M. B. & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.) Sage: Thousand Oaks, California. Mizgireuv, I. & Revskoy, Y. (2006). Transplantable tumor lines generated in clonal zebrafish. Cancer Research, 66(6): 3120-3125. Montréal Declaration (2011) Available at: http://www.wc8.ccac.ca/files/WC8_Declaration_of_Montreal_FINAL.pdf MORI. (1999). Animals in medicine and science: General public research conducted for Medical Research Council. MRC: London. MORI. (2010). Views on Animal Experimentation. London: MRC. http://www.ipsosmori.com/DownloadPublication/1343_sri-views-on-animal-experimentation-2010.pdf Accessed on September 15, 2011. Narin, F., Olivastro, D. & Stevens, K. A. (1994). Bibliometrics/Theory, Practice and Problems. Evaluation Review, 18: 65-76.  106  Navaro, J., Maldonado, E., Pedraza, C. & Cavas, M. (2001). Attitudes among animal research among psychology students in Spain. Psychological Reports, 89: 227-236. Norwegian Animal Research Authority (NARA) (2005). http://oslovet.veths.no/dokument.aspx?dokument=99 Accessed Oct 17th, 2008. Nuffield Council on Bioethics. (2005). The ethics of research involving animals. London, UK: Nuffield Council on Bioethics. Ohlendorf, G. W., Jenkins, Q. A. L., & Tomazic, T. J. (2002). Who cares about farm animal welfare? In: Wimberley, R. C., Harris, C. K., Molnar, J. J. and Tomazic, T. J. (Eds.) The Social Risks of Agriculture: Americans Speak Out On Food, Farming, and the Environment. Pp. 87-101. Praeger: Westport, CT. Olson I. A. S. & Sandøe, P. (2004). Ethical decisions concerning animal biotechnology: what is the role of animal welfare science? Animal Welfare, 13: S139-144. Opdenakker, R (2006) Advantages and disadvantages of four interview techniques in qualitative research. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 7 (4). Retrieved from http://www.qualitative-research.net/index.php/fqs/article/view/175/391 Orlans F.B. (2000). Public policies on assessing and reporting degrees of animal harm; international perspectives. In Progress in the Reduction, Refinement and Replacement of Animal Experimentation, Pp. 1075-1082, ed. Balls M., van Zeller A. M. and Halder M.E. New York: Elsevier Science Pub. Co. Ormandy, E.H. (2010). The Use of Genetically Engineered Animals in Science: Genome British Columbia Knowledge Translation Workshop III. May 2010. Available at: http://www.genomebc.ca/portfolio/genomics-and-society/events/knowledge-translationworkshops/genetically-engineered-animals/ Ormandy, E. H., Schuppli, C. A. & Weary, D. M. (in press, a) Public attitudes toward the use of animals in research: Effects of invasiveness, genetic modification and regulation. Anthrozoös. Ormandy, E. H., Schuppli, C. A. & Weary, D. M. (in press, b) Factors affecting people’s acceptance of the use of zebrafish and mice in research. Alternatives to Laboratory Animals. Ormandy, E. H., Schuppli, C. A. & Weary, D. M. (2009). Worldwide trends in the use of animals in research: The contribution of genetically modified animal models. Alternatives to Laboratory Animals, 37: 63-68. Orwig, K. E., Avarbock, M. R. & Brinster, R. L. (2002). Retrovirus-mediated modification of male germ-line stem cells in rats. Biology of Reproduction: 67: 874-879. Osborne N, Jackson I, Cox D, et al. (2009a). Sharing and archiving of genetically altered mice: Opportunities for reduction and refinement. A report of the RSPCA Resource Sharing Working Group (RSWG), Available at: http://www.nc3rs.org.uk/downloaddoc.asp?id=857 107  Osborne NJ, Payne D. and Newman ML. Journal editorial policies, animal welfare and the 3Rs. American Journal of Bioethics 9(12) (2009b): 55-59 Paul, E. S. (1995). Us and them: scientists’ and animal rights campaigners’ views of the animal experimentation debate. Society and Animals, 3: 1-21. Paul, E. S. & Serpell, J. A. (1993). Childhood pet keeping and humane attitudes in young adulthood. Animal Welfare, 2(4): 321-337. Pearl, D. L., Louie, M., Chui, L., Doré, K., Grimsrud, K. M., Martin, S. W., Michel, P., Svenson, L. W. & McEwen, S. A. (2009). A multi-level approach for investigating socio-economic and agricultural risk factors associated with rates of reported cases of Escherichia coli O157 in humans in Alberta, Canada. Zoonoses and Public Health, 56 (8): 455-464. Pearl, D. L., Louie, M., Chui, L., Doré, K., Grimrud, K. N., Martin, S. W., Michel, P., Svenson, L. W. & McEwen, S. A. (2008). Epidemiological characteristics of reported sporadic and outbreak cases of E.coli o157 in people from Alberta, Canada (2000-2002): methodological challenges of comparing clustered to unclustered data. Epidemiology and Infection, 136 (4): 483-491. Peek, C. W., Dunham, C. C. & Dietz, B. E. (1997). Gender, relational role orientation, and affinity for animal rights. Sex Roles, 37: 905-920. Pifer, L. (1996). Exploring the gender gap in young adults’ attitudes about animal research. Society and Animals, 4(1): 37-52. Pifer, L., Shimizu, K. & Pifer, R. (1994). Public attitudes toward public research: Some international comparisons. Society and Animals, 2: 95-113. Plous S. (1996). Attitudes towards the use of animals in psychological research and education: Results from a national survey of psychology majors. Psychological Science, 7(6): 352-358. Plous, S. (1991). An attitude survey of animal rights activists. Psychological Science, 2: 194-196. Priest, S. H. (2000) US public opinion divided over biotechnology? Nature Biotechnology, 18: 939-942. Qin, W. & Brown, J. L. (2007). Public reactions to information about genetically engineered foods: Effects of information formats and male/female differences. Public Understanding of Science, 16: 471-88. Rajecki, D. W., Rasmussen, J. L. & Craft, H. D. (1993). Labels and the treatment of animals: Archival and experimental cases. Society and Animals, 1: 45-60. Richmond, G., Engelmann, M. & Krupka, L. R. (1990). The animal research controversy. The American Biology Teacher, 52(8): 467-471. Robinson, V., Morton, D. B., Anderson, D., Carver, J. F. A., Francis, R. J., Hubrecht, R., Jenkins, E., Mathers, K. E., Raymond, R., Rosewell, I., Wallace, J. & Wells, D. J. (2003). Refinement and reduction in production of genetically modified mice. Sixth report of the 108  BVAAWF/FRAME/RSPCA/UFAW Joint Working Group on Refinement. Laboratory Animals, 37: S1-S51. Rollin, B. (2010) Animal research, animal welfare, and the Three R’s. Journal of Philosophy, Science and Law, 10: 1- 11. Rowe, G. & Frewer, L. J. (2005) A typology of public engagement mechanisms. Science, Technology, and Human Values, 30(2): 251-290 Royal Society of Canada (2001). Elements of Precaution: Recommendations for the Regulation of Food Biotechnology in Canada. Royal Society of Canada: Ottawa. Russell, W. M. S. & Burch R. L. (1959). The principles of humane experimental technique, 238pp Meuthen: London, UK. Sanford, L. P., Kallapur, S., Ormsby, I. & Doetschman, T. (2001). Influence of genetic background on knockout mouse phenotypes. In Methods in Molecular Biology, Vol. 158: Gene Knock-out Protocols (eds. M. J. Tymms & I. Kola), pp 217-226. Humana Press: Totowa, NJ. Schiele, B. (2008). On and about the deficit model in an age of free flow. In D. Cheng et al (Eds) Communicating Science in Social Contexts: New Models, New Practices, pp. 93-118. Springer Sciences: Berlin. Schmidt, S. (2012) Genetically engineered pigs killed after funding ends. Postmedia News, June 22, 2012. Available at: http://www.canada.com/technology/science/Genetically+engineered+pigs+killed+after+funding+ ends/6819844/story.html Schuppli, C. A. (2011). Decisions about the use of animals in research: Ethical reflection by animal ethics committee members. Anthrozoös, 24(4): 409-425. Schuppli, C. A. & Fraser, D. (2007). Factors influencing the effectiveness of research ethics committees. Journal of Medical Ethics, 33: 294-301. Schuppli C.A., Fraser D. & McDonald M. (2004). Expanding the 3Rs to meet new challenges in humane animal experimentation. Alternatives To Laboratory Animals, 32: 525-532. Schuppli, C. A. & Weary, D. M. 2010. Attitudes towards the use of genetically modified animals in research. Public Understanding of Science 19: 686-697. Serpell, J. A. (2004). Factors influencing human attitudes to animals and their welfare. Animal Welfare, 13 (1): S145-S151. Serpell, J. A. (1996). In the company of animals: A study of human-animal relationships. Cambridge: Cambridge University Press. Sherwin, S. 2001. Toward setting an adequate ethical framework for evaluating biotechnology policy. Ottawa: Canadian Biotechnology Advisory Committee. 109  Sieber, J. E. (1986). Students’ and scientists’ attitudes on animal research. American Biology Teacher, 48: 85-51. Slovic, P. (2006). Affect, reason, risk and rationality. European Working Group “Multicriteria Aid for Decisions” Series, 3 (3): 1-5. Stafleu, F. R., Tramper, R., Vorstenbosch, J. & Joles, J. A. (1999). The ethical acceptability of animal experiments: a proposal for a system to support decision-making. Laboratory Animals, 33: 295-303. Stanistreet, M. & Spofforth, N. (1993). Attitudes of undergraduate students to the uses of animals. Studies in Higher Education, 18: 177-196. Stanton, J. M. (1998). An empirical assessment of data collection using the Internet. Personnel Psychology 51 (3): 709-725. Swami, V., Furnham, A. & Christopher, A. (2008). Free the animals? Investigating attitudes toward animal testing in Britain and the United States. Scandinavian Journal of Psychology, 49: 269-276. Tashakkori, A. & Teddlie, C. (2003). Handbook of mixed methods in social and behavioural research. Thousand Oaks, California: Sage. Taylor, K., Gordon, N., Langley, G. & Higgins, W. (2008). Estimates for Worldwide Laboratory Animal Use in 2005. Alternatives to Laboratory Animals, 36: 327-342. Tongco, M. D. C (2007). Purposive sampling as a tool for informant selection. Ethnobotany Research and Applications, 5: 147-158. USDA (United States Department of Agriculture) (2006). AWA Inspections. http://www.hsus.org/web-files/PDF/ARI/awreport2006-1.pdf Accessed March 24, 2008. van der Meer, M., Rolls, A., Baumans, V., Olivier, B. & van Zutphen L. F. M. (2001). Use of score sheets for welfare assessment of transgenic mice. Laboratory Animals, 35: 379-389. Verbeek, J. S. (1997). Introduction of defined mutations into the mouse germline. In Welfare Aspects of Transgenic Animals – Proceedings EC Workshop (eds. L. F. M. Van Zutphen & M. Van Der Meer), pp. 1-17. Berlin: Spinger-Verlag. Walsh, M. & Richmond, J. (2005). Regulation of animal experimentation in the United Kingdom. School Science Review, 87: 85-89. Wells D. (2002). Production of chimeras derived from murine embryonic stem cells. In Transgenesis Techniques: Principals and Protocols (ed. A. R. Clarke), pp. 127-149. Totowa, NJ: Humana Press Humana Press. Wells D. L. & Hepper P. G. (1997) Pet ownership and adults’ views on animal use. Society and Animals, 5: 45-63. 110  Wells, D. J., Playle, L. C., Enser, W. E., Flecknell, P. A., Gardiner, M. A., Holland, J., Howard, B., Hubrecht, R., Humphreys, K. R., Jackson, I. J., Lane, N., Maconochie, M., Mason, G., Morton, D. B., Raymond, R., Robinson, V., Smith, J. A. & Watt, N. (2006) Assessing the welfare of genetically altered mice. Laboratory Animals, 44: 88-103 Wuensch, K. & Poteat, G. M. (1998). Evaluating the morality of animal research: Effects of ethical ideology, gender, and purpose. Journal of Social Behaviour and Personality, 13: 139-150. Yoshiki, A. & Moriwaki, K. (2006). Mouse phenome research: Implications of genetic background. Institute for Laboratory Animal Research, 47(2): 94-102.  111  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073306/manifest

Comment

Related Items