LINKING RISK COMMUNICATION AND BIOMEDICAL ETHICS: THE CASE OF PRE- IMPLANTATION GENETIC DIAGNOSIS by Holly Ann Longstaff A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Interdisciplinary Studies) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2009 © Holly Ann Longstaff, 2009 11 Abstract The fundamental objective of this thesis as a collective work is to contribute to the interdisciplinary body of research that seeks to integrate the fields of bioethics and risk analysis. The goal was to move risk communication beyond procedural ethics to focus on substantive ethics or the values that guide such strategies. The research study discussed here includes three phases: (1) interviews with a heterogeneous collection of experts to determine the risks and benefits that pre-implantation genetic diagnosis (PGD) may pose to individuals, groups, and society; (2) mental model interviews with lay citizens to determine shared misconceptions and background knowledge of PGD; (3) and a series of risk communication experimental workshops with lay citizens informed by phase 1 and 2 findings. Phase 1 employed a novel method of eliciting expert judgment by incorporating the views of conventional (e.g., physicians) and unconventional experts (e.g., lived experience of disability) relevant to PGD technology. The approach is based on the mental models method as described by Morgan and colleagues (2002). In total, 8 experts were interviewed individually to protect the diversity of expressed views. All were shown 3 charts based on an extensive literature review. This process allowed for qualitative feedback capable of incorporating a range of discrete skills sets and communication styles. The final products of the interview process were a revised series of charts (3 revised originals plus 2 additional charts) that illustrated technical, personal, and value-oriented messages communicated by those deeply engaged in PGD research. The lay mental model interviews were conducted with 16 citizens. As anticipated, most were not familiar with PGD technology but were familiar with IVF, which must occur in conjunction with PGD. This finding is significant as IVF and PGD share many risks and benefits. Important misconceptions were also identified that were ultimately corrected during the workshops. For example, interview participants had a very difficult time anticipating the stakeholders of PGD and the full multi-scalar outcomes of this technology. Many also assumed that users of PGD technology were infertile. Understanding the multiple scalar nature of PGD is vital to understanding the full array of risks and benefits posed by this technology. It is also important for citizens to understand that users of PGD technology are submitting themselves to the risks of IVF although they are not infertile and could have their own biological children in most cases. The deliberative workshops represented a holistic approach to rick communication. In total, 57 individuals were recruited to participate in 1 of 8 workshops. Randomly assigned participants in half of the workshops (Treatment B) viewed structured decision making (SDM) materials in the form of consequence matrices (informed by phase 1 expert results) while the others did not. Results showed that workshop participants were very successful in learning a wide range of value and technically-oriented risk messages. However, those who received SDM training were better able to make ethical decisions concerning risk communication strategies for PGD technology. This finding underscores the value of including SDM materials in communication strategies for ethically contentious issues. Table of contents iii Table of contents iii List of tables v List of figures vi Acknowledgements vii Dedication viii Co-authorship statement ix Chapter one: Linking risk communication and biomedical ethics: the case of pre-implantation genetic diagnosis 1 Introduction 1 Topic domain: systemic hazards 2 Pre-implantation genetic diagnosis: an example of a systemic hazard 4 Novel risk communication strategies for addressing systemic hazards 7 Overall study design 9 Phase 1: literature review and expert mental model interviews 9 Phase 2: lay mental model interviews 11 Phase 3: experimental deliberative workshop with lay participants 12 Organization of thesis papers and discrete objectives 14 Works cited 17 Chapter two: Integrating bioethics concepts with risk communication theory and applications 20 Introduction: the value of ethical analysis 20 Guidance from substantive ethics 22 From how to why: moving beyond procedural ethics 22 The state of substantive ethics within risk communication: a few examples 25 Guidance from public ethics 30 Employing bioethics guidance to communicate the risks and benefits of systemic health hazards 33 Conclusions 38 Works cited 42 Chapter three: Characterizing expert mental models for systemic hazards: the case of pre-implantation genetic diagnosis 48 Introduction 48 Study context: pre-implantation genetic diagnosis 54 The expert literature review: clarifying risks and benefits for systemic hazards 56 Clarifying risks and benefits at the family scale 59 Clarifying risks and benefits at the group and societal scales 61 The interviews: characterizing expert mental models 64 Experts interviews 64 Final products 69 Conclusions 74 Works cited 77 Chapter four: Learning about systemic hazards through a risk communication experiment: the case of pre- implantation genetic diagnosis 83 Introduction 83 The topic: pre-implantation genetic diagnosis 85 Study overview and samples 87 Mental model interviews 90 Methods 90 Interview results 92 Risk communication experimental workshops 95 Methods 95 Workshop findings 101 Confirmation of mental model interview findings 101 Learning new information 104 Rating the risk communication strategies I l l iv Discussion and conclusions 113 Works cited 119 Chapter five: Using structured decision making to help communicate information about controversial technologies: The case of pre-impanation genetic diagnosis 123 Introduction 123 Experimental design 126 Concepts 126 Research design 129 Risk communication experimental workshop results 133 Reported hopes and concerns for PGD technology 133 Participant opinions regarding PGD technology 138 Rating the workshop 139 Recommendations from participants 140 Evaluating the decision making process 146 Quality of decisions 147 Capacity for future decision making 149 Conclusions 151 Works cited 153 Chapter six: Overview, insights, and future research directions 156 Summary of chapter hypotheses and key insights 156 Study strengths and weaknesses 159 Future research directions 161 Works cited 164 Appendices 166 Appendix A. UBC Research Ethics Board's Certificates of Approval 167 Appendix B. Interview questions for experts 169 Appendix C. Mental model interview questions for lay participants 171 Appendix D. Expert charts (final versions) 172 Appendix E. Example of how expert charts changed over time due to expert participant feedback 177 Appendix F. Risk communication workshop presentation that was used to guide the workshop (this presentation includes the expert based FAQ's and the consequence matrices) 175 Appendix G. Risk communication workshop workbook for lay participants 222 Appendix H. Risk communication workshop scripts for audio recordings (based on Appendix I charts) 248 Appendix I. Risk communication workshop news articles 253 Appendix J. Risk communication workshop links to videos recommended by experts 256 Appendix K. Risk communication workshop database of results 257 V List of tables Chapter one: Linking risk communication and biomedical ethics: the case of pre-implantation genetic diagnosis Tablel.l Consequences that cross domains for stakeholders of PGD at multiple scales 6 Chapter three: Characterizing expert mental models for systemic hazards: the case of pre-implantation genetic diagnosis Table 3.1 Overview of interviewee expertise 66 Chapter four: Learning about systemic hazards through a risk communication experiment: the case of pre-implantation genetic diagnosis Table 4.1 Demographic statistics of interview and workshop participants 89 Table 4.2 Selected introductory mental model interview questions for PGD technology 92 Table 4.3 Workshop format 96 Table 4.4 Unaided question format concerning potential risks and benefits that PGD may pose to a woman 108 Chapter five: Using structured decision making techniques to help communicate information about controversial technologies: The case of pre-impanation genetic diagnosis (PGD) Table 5.1 National Academy of Sciences, 2008 criteria 124 Table 5.2 Workshop research questions 131 Table 5.3 Reported stakeholder concerns: most common response by treatment 137 Chapter six: Overview, insights, and future research directions Table 6.1 Fundamental objectives by thesis chapter 157 VI List of figures Chapter one: Linking risk communication and biomedical ethics: the case of pre-implantation genetic diagnosis Figure 1.1 Conceptual framework for study 10 Chapter two: Blending bioethics concepts with risk communication theory and applications Figure 2.1 Ethical principles guiding risk communication 28 Figure 2.2 Examples of objectives and moral responsibilities by scale 31 Figure 2.3 Bioethics guidance for risk communication strategies regarding systemic hazards 34 Chapter three: Characterizing expert mental models for systemic hazards: the case of pre-implantation genetic diagnosis Figure 3.1 Selected key objectives and the potential influence pathways of decisions at each scale 58 Figure 3.2 Chart indicating social risks and benefits of PGD 72 Chapter four: Learning about systemic hazards through a risk communication experiment: the case of pre-implantation genetic diagnosis Figure 4.1 Conceptual framework for study 88 Figure 4.2 Selected question from pre and post deliberation questionnaire 97 Figure 4.3 Selected rating questions from post-deliberation questionnaire 100 Figure 4.4 Most affected stakeholders of PGD technology 102 Figure 4.5 Individuals and groups who would be most concerned about the use of PGD: coded results of open ended responses from workshop participants 103 Figure 4.6 Understanding the relationship between PGD and IVF 104 Figure 4.7 Correct responses about general background before and after reviewing risk communication materials 105 Figure 4.8 Motivations for pursing PGD 106 Figure 4.9 Correct responses about regulatory framework before and after reviewing risk communication materials.... 107 Figure 4.10 Rating the popularity and trustworthiness of workshop risk communication strategies 112 Figure 4.11 Confidence in knowledge of the risks and benefits of PGD 116 Chapter five: Using structured decision making techniques to help communicate information about controversial technologies: The case of pre-impanation genetic diagnosis Figure 5.1 Consequence tables for PGD at the societal scale: objectives by alternatives 130 Figure 5.2 Workshop format 132 Figure 5.3 Survey 1 results: reported hopes for PGD technology 133 Figure 5.4 Survey 2 results for treatment A participants: reported hopes for PGD technology 134 Figure 5.5 Survey 2 results for treatment B participants: reported hopes for PGD technology 134 Figure 5.6 Survey 1 results: reported concerns for PGD technology 135 Figure 5.7 Survey 2 results for treatment A participants: reported concerns for PGD technology 136 Figure 5.8 Survey 2 results for treatment B participants: reported concerns for PGD technology 136 Figure 5.9 Survey 1 results: Opinions of PGD technology 138 Figure 5.10 Survey 2 results: Opinions of PGD technology 139 Figure 5.11 Survey 2 results: Favourite part of the workshop 140 Figure 5.12 Survey 2 workshop ratings: Least favourite part of the workshop 140 Figure 5.13 Survey 2workshop ratings: Most important factor influencing recommendations 141 Figure 5.14 Survey 2 workshop ratings: Extreme ratings in percentages by treatment 142 Figure 5.15 Recommendations: Who should communicate information about PGD? 143 Figure 5.16 Recommendations: Proposed risk communication strategies for PGD 144 Figure 5.17 Recommendations: Who should comprise the target audience for risk communion strategies about PGD? 145 Figure 5.18 Recommendations: What types of information should comprise risk communion strategies for PGD? 146 vii Acknowledgements I would like to gratefully acknowledge the financial support I received from the Canadian Institutes for Health Research (CIHR) Ethics of Health Research and Policy Training Project and the CIHR Institute of Genetics through The W. Maurice Young Centre for Applied Ethics (CAE) at the University of British Columbia. I also deeply appreciate the financial assistance I received from the Interdisciplinary Studies Graduate Program in the first year of my Doctoral program. Finally, I would like to thank my workshop note takers Sonja Klinsky and Liz Wilcox and all my wonderful colleagues at CAE, especially Peter Danielson. viii Dedication This thesis is dedicated to my thoughtful committee members Tim McDaniels, Michael McDonald, and Susan Cox, my parents Marguerite Longstaff (nee Gibb) and John Longstaff who are both deeply committed educators, and my partner Joey Mikawoz who has always encouraged me to pursue my career with great passion. I also dedicate this work to my informal mentor Michael Burgess. ix Co-authorship statement Each manuscript included in this thesis was authored solely by Holly Longstaff including the identification and design of the research program, performing the research, data analyses, and manuscript preparation. 1 Chapter one: Linking risk communication and biomedical ethics: the case of pre- implantation genetic diagnosis Introduction In March 2009, the US based Fertility Institutes suspended their pre-implantation genetic diagnosis (PGD) eye and hair colour program due to media backlash and public outrage. In a statement appearing on their website the Institute explained that "Though well intended, we remain sensitive to public perception and feel that any benefit the diagnostic studies may offer are far outweighed by the apparent negative societal impacts involved'' (The Fertility Institutes, 2009). Pre-implantation genetic diagnosis is the process of screening fertilized human embryos during in vitro fertilization (IVF). In Canada, the process is typically used to screen embryos for chromosomal abnormalities or specific genetic conditions or select embryos that are tissue matches for sick siblings. Using PGD for non medical purposes such as sex or hair colour selection is frequently viewed as controversial due to obvious eugenic inferences. However, the ethical issues associated with this new emerging reproductive technology are much more complicated and far reaching than they may initially appear. Pre-implantation genetic diagnosis is a systemic hazard comprising a wide range of obscure risks and benefits for individuals, groups, and society at large. Communicating information about this technology therefore poses special challenges for risk communicators. The fundamental objective in this introductory chapter is to explain the unique contributions of this thesis in terms of the topic that is addressed and the methods used to 2 analyze it. It begins by explaining why emerging systemic hazards like PGD deserve our attention in the field of risk communication. Topic domain: systemic hazards Issues falling under the rubric of systemic hazards in the life sciences have the ability to modify living things and therefore foster change at "breathtaking speed...on an unprecedented scale" (OECD, 2003). Analyses of such hazards must therefore stretch beyond conventional probability assessments and cause and effect predictions to involve examinations of "interdependencies and relationships between... risk clusters" (Renn & Klinke, 2004). Many hazards have systemic elements or effects. For example, alcohol consumed during pregnancy may result in a child with health impairments (family scale) who requires long term assistance from publicly funded health care and other social systems (societal scale). However, systemic hazards do not merely involve multiple scales or difficult tradeoffs between individuals and groups. There is the potential for them to have a fundamental effect and even permanently change (Joy, 2000) society's most vital systems (e.g., health, environment) (OECD, 2003). Systemic health hazards such as biotechnology, nanotechnology, or global climate change are unlike other contained hazards in that they cross multiple domains (i.e., economic, regulatory, social, ethical, technological). They are also associated with both normative and interpretative ambiguity and thus often lead to uncertain outcomes for a broad range of stakeholders at multiple scales (Renn, 2006) (see section on PGD below for examples). 3 Norms are temporally bound and may become ambiguous if the application to which they apply evolves (e.g., applying conventional safety standards to substantially equivalent novel products), public attitudes towards that application change, or information about the application decreases or increases (e.g., new risks or benefits are discovered).1 However, it is unclear how novel technologies such as PGD connect with existing social and ethical norms, if at all. Strong normative standards help experts determine socially desirable management options. Yet, with normatively ambiguous problems, experts may have difficulty predicting what is in the best interests of a diverse society (Tansey and Burgess, forthcoming) as the bounded rationality of experts cannot always detect norms embedded within the social rationality of laypeople (Perrow, 1984).2 In some cases, a single interest that all individuals within a society share may not even exist3 while at other times, uncertainties may lead to explanatory complexities. Interpretative ambiguity occurs when it is unclear as to how the consequences of a potential hazard will be experienced by stakeholders at various scales (individuals, groups, society at large). Managing these types of topics as an extension of conventional hazards may foster ethical controversy and public outrage. Emerging health hazards therefore require interdisciplinary approaches to risk communication that blend intellectual contributions from both the bioethics and risk analysis communities. Risk analysis can address many issues surrounding these hazards including predicting probabilities and anticipating 1 Potential hazards can become controversial when generally accepted norms are used to guide the management of novel applications or products through imperfect analogies and norm borrowing. 2 For example, this type of mismatch led to a moratorium of the BC salmon aquaculture industry in 1995. 3 There may be times when it is morally appropriate to treat certain groups differently. For example, in Canada, special guidelines have been developed for conducting research with citizens of First Nation and Inuit descent. For more on this please see Section six of the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans or TCPS (Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada, 1998) and the CIHR Guidelines for Health Research Involving Aboriginal People (CIHR, 2007). Chapter 9 of the Draft 2nd Edition of the TCPS is also dedicated to research involving Aboriginal peoples (Interagency Advisory Panel on Research Ethics, 2008). 4 interdependent relationships between risks and benefits. Risk research can also offer guidance on communicating difficult information about systemic hazards. Bioethics principles and approaches can be used to decipher normative and interpretative ambiguities and illuminate key values the may help to determine the best interests of a diverse society. Pre-implantation genetic diagnosis: an example of a systemic hazard If the term systemic hazard is defined too broadly it quickly becomes meaningless encompassing virtually any hazard with cross scale or multi-system components. Systemic hazards pose special challenges to the field of risk communication and lead to unique outcomes. More specifically, they cross multiple domains and scales and often lead to both normative and interpretive ambiguities for a wide range of stakeholders. The nature of systemic hazards is explored in the passage from Ortwin, 2006 shown below. Systemic risks are at the crossroads between natural events (partially altered and amplified by human action such as the emission of greenhouse gases), economic, social and technological developments and policy- driven actions, both at the domestic and the international level. These new interrelated and interdependent risk fields also require a new form of handling risk, in which data from different risk sources are either geographically or functionally integrated into one analytical perspective. Handling systemic risks requires a holistic approach to hazard identification, risk assessment, concern assessment, toler ability/acceptability judgements and risk management. Pre-implantation genetic diagnosis, the focus of this thesis, fits into this definition well. This controversial reproductive procedure has three main applications that include screening embryos (1) for specific genetic conditions; (2) to detect irregularities in the structure or number of chromosomes; and (3) to determine the sex of an embryo (Health Canada, 2005). Pre-implantation genetic diagnosis was selected as the topic for this study 5 because it a single technology that exhibits all the elements of a systemic hazard. It is therefore an appropriate starting point for analyses of systemic hazards as it is much less complicated than other examples such as global climate change that involve inputs from multiple industries, technologies, and behaviours. As discussed, systemic hazards are special in that they cross multiple domains and scales and exhibit normative and interpretive ambiguities for a wide range of stakeholders including whole societies (Renn, 2006). I will argue that pre-implantation genetic diagnosis can be described as normatively ambiguous because as a novel procedure, it does not directly appeal to any existing moral or social norms. We cannot assume that norms for PGD are identical to those of abortion or IVF even though these procedures share some ethical and social concerns such as rights of the unborn and reproductive rights of citizens. Table 1.1 illustrates how the consequences of PGD cross multiple domains and thus may involve interpretive ambiguities for stakeholders at various scales. For example, screening human embryos could allow families to produce biologically healthy children. However, viewing fertilized human embryos with disabilities as undesirable may normalize or reinforce discriminatory behaviours towards people with disabilities. However, potential outcomes such as medicalizing the pregnancy process or reinforce discriminatory behaviours are best described as unintended consequences of PGD as opposed to intended consequences such as reducing genetic conditions in biological children (DeLeire 1999 and Freeman, 2007). Such consequences are not desired by proponents of PGD but may nonetheless occur with increased acceptance and use of this reproductive technology. As famously argued by Hazlitt (1946), it is important to look "...for all the consequences of a policy instead of merely resting one's gaze on those immediately visible." Thinking through both intended and unintended consequences of systemic hazards helps broaden one's assessment of potential risks and may serve to anticipate and even avoid undesirable outcomes. In addition, screening out groups of people (e.g., those with Down Syndrome) could change the genetic and cultural fabric of society. Genetic materials are special in that they locate individuals within a web of complex relationships4 that include their ancestors, others within their genetic and cultural "group(s)", and future generations. In other words, when PGD alters the genetic landscape of our society, it is altering forever what is passed on to future generations from our genetic history. Table 1.1 Consequences that cross domains for stakeholders of PGD at multiple scales Consequences Family scale Group scale Societal scale Sample domains and examples of potential consequences (bot Health Healthy biological child is born Reducing the number of children with genetic condition X who are born Healthier citizenry overall (in terms of genetic conditions) Regulatory Access to PGD is granted Conditions to be screened are selected (groups of citizens that will be screened out) Locating responsibility for people with disabilities (family or society) Ethical Pressure for "high risk" families to undergo PGD Pressure for all women to eventually participate in screening programs Building an unjust society when discrimination towards people with disabilities is normalized through frequently used screening programs Technological Successfully screening out conditions for families Less assistive technology innovations directed towards citizens with screenable conditions Knowledge gleaned from experiences at the family scale is employed to produce more efficient, safer technologies that lead to widespread use 1 positive and neg Economic High cost of PGD is prohibitive for some families Wealthy groups of citizens have fewer biological children with particular genetic conditions Economic burden of people with certain genetic disabilities is reduced, allowing these health care funds to be allotted to other programs ative) by scale Social Families suffer stigma associated with PGD usage Groups of citizens with screenable conditions are stigmatized as their numbers decrease overall Canadian citizens become less tolerant of people with genetic disabilities as they become increasingly rare I would like to acknowledge Michael McDonald for introducing me to this concept. 7 Each of the sample consequences shown in Table 1.1 involves physical, ethical, and a range of other domains that must be considered by citizens (especially those who will bear the consequences). For example, regulations (some of which are currently in flux in Canada5) will ultimately determine who has access to PGD and who should be responsible for the costs of this procedure (e.g., families, society). There are consequences to individuals who undergo or have PGD performed on them (e.g., unknown long-term health outcomes for children born following PGD and women who undergo PGD), individuals who DO NOT undergo PGD (e.g., pressure to submit to PGD technology if it becomes a standardized medical procedure), groups (e.g., increased intolerance towards people with disabilities and further medicalizing the pregnancy experience for all women) and society (e.g., decreased overall diversity and economic effects within the health care system). Novel risk communication strategies for addressing systemic hazards "Next-generation technologies" like PGD draw on heterogeneous expertise from medical and regulatory experts as well as unconventional experts that include educated stakeholders and activists (e.g., those living with a disability or caring for someone with a disability, those speaking on behalf of people with disabilities). Risk communication strategies for such hazards are therefore particularly difficult to craft as they must 5 Before 2004, there were no regulations, standards, or professional guidelines for the use of PGD in Canada. An agency called Assisted Human Reproduction Canada that reports to the Minister of Health now regulates this technology in Canada. However, the regulatory framework is currently under development and the Province of Quebec has brought a constitutional challenge in regards to certain provisions under the Assisted Human Reproduction Act (Supreme Court of Canada, 2009) 6 Other next-generation technologies include nanotechnology and genetically modified organisms (OECD, 2003), and synthetic biology (Andrianantoandro et al., 2006). 8 communicate a range of technically -oriented (defined here as focusing primarily on technical understandings, the design of alternatives and consequences of PGD including the regulatory landscape) and value -oriented (defined here as focusing primarily on preferences and the emotional or experiential dimensions of PGD such as the lived experience of disability) information to the target audience in order to foster informed decision making and deliberation.7 Leading experts from the field of ethics and risk analysis have all articulated the need for novel strategies and approaches capable of dealing with this new class of hazards (see Renn, 2003 & 2006; Slovic, 2007; & Sherwin, 2008). Such methods call for holistic strategies that draw on input from a variety of sectors, expert groups, and civil society (Renn & Klinke, 2004 and McDonald, 2000). Democratising science policy (McCormick, 2007) through collaborative risk decision making involving experts and lay citizens is now a key objective for many social scientists, risk researchers, and practitioners (Longstaff & Burgess, forthcoming; Webler & Renn, 1995; Renn, 2004; Lidskog, 2008; and Walmsley, 2009). Involving diverse input into risk analysis processes means that one must also be prepared to allow multiple understandings of hazards to inform risk communication strategies in line with Irwin's Citizen Science concept where science is at least in part "developed and enacted" by citizens (Fisher, 1991 and Irwin, 1995). Such strategies will necessarily include both value-oriented and technically-oriented risk messages. However, lay citizens must be communicatively competent to produce meaningful input for policy or risk management decisions concerning hazards, which requires adequate information among other things Many would argue that all values are infused with experiential or technical understanding and all technical understanding has some value-based influences. However, it is useful and important in this case to distinguish among aspects of decision making that rest on emotional and technical understanding as both types of information should inform ethical risk communication strategies for PGD. 9 (Dryzek, 1990). Although it is generally agreed that citizens must be informed during such processes (Sunstein, 2002, Dryzek, 1990, National Academy of Sciences, 2008), specific conditions that would help determine the sufficient amount and type of information are rarely outlined as they will inevitably vary somewhat by hazard. It is also a subject on which lay citizens (who would eventually receive the information) are capable of producing important insights and was therefore one of the deliberative questions posed to participants of this study. The experiment described in this thesis demonstrates one method of crafting a deliberative risk communication strategy that informs participants while incorporating relevant values and norms. It is based on similar efforts conducted by Arvai et al, Renn, and others. For example, Arvai et al sought to enhance decision making through deliberative risk communication experiments that include components of Keeney's Value Focused Thinking approach (Arvai, Gregory, and McDaniels, 2001). Renn introduced the concern assessment to incorporate value based expertise directly into his risk governance framework (Renn, 2006). This novel assessment would complement (not replace) the more conventional technical risk assessment process. Overall study design The purpose of this section is to provide a brief descriptive overview of the entire study described in this thesis. Subsequent chapters provide a more detailed analysis of each individual phase. 10 Phase 1: literature review and expert mental model interviews This thesis project is divided into three phases shown below in Figure 1.1. The first phase involved (A) a literature review and (B) mental model interviews with experts to confirm and supplement these findings. The review included position statements and information produced by the National Society of Genetic Counsellors and the American Board of Genetic Counseling, academic papers written within the last five years and included on PubMed, and policy documents regarding PGD. The purpose of the literature review was to clarify the risks and benefits of PGD across domains at multiple scales (family, group, society). These findings were then transformed into a series of 4 charts that recorded major themes and possible outcomes (2 additional charts were added in accordance with advice from experts). Figure 1.1 Conceptual framework for study Phase 1 • Literature review Expert mental model interviews Phase 2 • Lay mental model interviews Phase 3 • Risk communication workshops with lay citizens Expert mental model participants were initially identified from contributions in position statements, academic papers included on PubMed, and policy documents regarding PGD. Additional experts were also added to this group based on recommendations by two or more participants in a snowball sampling method. All who participated in this study are widely acknowledged as experts in this narrow field. The study topic draws on a diverse range of experts that include some lay stakeholders who have specialized knowledge of PGD (e.g., parents of children with disabilities). Risks and benefits at the family scale 11 drew primarily on experts in reproductive medicine, genetics & genetic counselling, and the experience of living with or caring for someone with a disability. Major topics associated with the group and societal scales called for experts in the ethical, legal, economic, and social dimensions of PGD and disability rights advocacy. The expert interview phase of the study was broken into three steps. First, the expert was contacted via email to request their participation in this project. During this first email exchange, the expert was provided with a brief overview of the entire project and their potential role in this phase of the study. If the expert agreed to participate, they were then sent a more detailed explanation of their responsibilities, the charts produced through the literature review, and a consent form. Each individual was given at least a week to review the charts, make comments, and prepare for the interview, which would take place in person, over the phone, or via email depending on their preference and location. In total, eight experts were interviewed with a fairly even gender split. Interviews lasted about one hour on average. Phase 2: lay mental model interviews The second phase includes mental model interviews with lay citizens. The objectives of these interviews were to determine shared misconceptions and background knowledge of PGD. All lay participants were recruited through poster advertisements and were randomly assigned to be either interviewed or participate in one of two workshop treatment groups (phase 3). During the interviews, participants were asked open ended 12 questions that helped them to explore their own mental model of this novel technology. Participants were also asked about more specific risks, benefits, and stakeholders for PGD technology to determine awareness of or ability to anticipate information included in the finalized expert charts from phase 1. In total, sixteen individuals participated in this process, which lasted between 45-60 minutes on average. Findings were then used to complement phase 1 results and inform phase 3 workshop materials. Phase 3: experimental deliberative workshop with lay participants Findings from phase 1 and 2 were combined to inform materials used during a series of experimental deliberative risk communication workshops with lay citizens (phase three). The materials emphasized either value-oriented or technically-oriented information relevant to PGD. In total, fifty-seven individuals were recruited through poster advertisements to participate in one of eight 2.5- hour workshops. Those involved in the workshops (and lay mental model interviews) do not to represent any specific group or population. The purpose of the workshop experiment was to test learning outcomes and differences between treatments (structured and unstructured decision making) during a risk communication experiment. This study includes a within-subject,8 before and after comparison design that analyzes how all participants learn about the ethically contentious and novel risks and benefits associated with PGD. The experiment in this study employs a between-subject design 8 Within -subject is a term used to describe studies in which repeated measurements are taken from the same participants while between -subject is a term used to describe the process of asking 2 groups of participants the same questions but with a different treatment. 13 described as follows. Randomly assigned participants in half of the workshops (Treatment B) viewed structured decision making (SDM) materials in the form of consequence matrices while the others did not. Treatment A participants received no information regarding the objectives and alternatives for PGD in terms of consequence tables, and were therefore considered the control group. The matrices were derived from expert interview findings. The deliberative workshops were not experimenting with the success of various risk communication methods (e.g., video, FAQ's, etc) to communicate particular risk messages about PGD. Instead, participants were presented with a great deal of information about PGD in a range of formats and openly asked to evaluate them. By making this process transparent (as opposed to part of the experiment by showing participants informationally equivalent communications), participants were able to offer advice such as switching the content of one approach with another, or ruling out one method but not its contents. The objective of the workshop experiment was to examine the influence that a series of consequence matrixes had (if any) on participants' decisions about risk communication strategies for PGD. The two areas of analysis that are of primary interest to this study are the (1) within- subject before and after comparison and (2) between subject two treatment or SDM experiment. I analyzed all study data using the individual unit of analysis as opposed to comparing groups for several reasons. First, the before and after and treatment group sizes that comprised this study were slightly small for statistical tests typically used to analyze differences between groups (e.g., T-Test, Matched Pairs T-Test, Analysis of Variance (ANOVA)) (Levin and Fox, 2004). Secondly, two groups (in the case of the before and after comparison) is an insufficient number to confirm significant variations 14 between results. In other words, with only two groups, one cannot determine if differences are random in nature. Examining differences between groups therefore requires one to examine differences between multiple groups. However, financial constraints on this student project made recruiting participants for multiple groups impossible and examining differences at the individual unit of analysis the more practical alternative. Finally, testing group differences between SDM treatments (the between subject experiment) would require group scale analysis such as content analysis of qualitative discussions or group oriented tasks and questions. Individual workbook tasks were used in this study because it is a method of eliciting responses that can be managed easily by a single researcher without the need for expensive transcriptions and extensive qualitative coding of group conversations. Statistical significance at the individual unit for both the within subject and between subject analyses was therefore determined using the McNemar test for the significance of changes at a 95% confidence interval (Sirkin, 1995). This formula measures the number of individuals who respond differently after being exposed to new information or a new condition (Daniel, 1990). Organization of thesis papers and discrete objectives The first paper in this thesis outlines the conceptual framework for the project. I argue that risk and bioethics experts each have valuable contributions to make in the area of systemic health hazards and that lessons from the field of bioethics can fortify risk communication efforts for these difficult problems. Furthermore, lessons from the field of bioethics could potentially shed light on many significant aspects of risk communication for systemic hazards including how to: identify the moral values that guide our efforts; 15 build an ethical communication strategy; choose experts that will ultimately inform that strategy as well as the target audience; and evaluate findings. The second paper focuses specifically on phase one of the thesis project. Characterizing expert judgment is an integral component of all risk communication strategies. This paper attempts to convince the reader that we must broaden our understanding of expertise for systemic hazards like PGD to include non-conventional experts (e.g., those speaking on behalf of or living with disabilities) in order to encourage holistic or "all things considered" ethical judgments (McDonald, 2000). However, broadening the scope of expertise should not involve lowering the bar for inclusion or appealing to extreme outliers. It is important that one maintain good reasons and evidence for including each expert point of view, and balance such perspectives in a fair and justified manner. All experts who were included in this study were identified through a thorough literature review and confirmed by other PGD experts in a snowball sampling method. In addition, including new types of experts presents the need for new methods of gathering and compiling expert judgments that can make sense of value-oriented and technically- oriented risk information. The approach presented in this paper experiments with the Mental Models Method (Morgan et al, 2002) but is also inspired by bioethics concepts and principles. The third paper discusses learning outcomes for phase two (mental model interviews with lay citizens) and three (risk communication experiment with lay citizens) of this study. It demonstrates that the mental models approach can effectively communicate the risks and benefits of novel and unfamiliar hazards. Despite their lack of previous awareness of 16 PGD, participants were in fact successful in acquiring a range of technically -oriented information about this emerging reproductive technology. However, they were concerned about the value-oriented risk and benefit messages contained in certain risk communication methods (e.g., news media articles and videos). Although they agreed that the lived experience of disability and other aspects of these messages were important to convey, they also believed that such messages were biased and intended to manipulate instead of informing viewers' decisions. The final paper discuses the results of study phase three or the experimental risk communication workshops. These findings demonstrated that lay participants, especially those who review even minimal SDM materials, (in this case a series of consequence matrixes that summarize the kinds of information required for responsible decision making (Hammond et al, 1999 and Clemen & Reilly, 2000) derived from expert interview findings) are able to make well informed, inclusive, and therefore meaningful decisions about a complicated novel technology. Workshop findings also fared well against the National Academy of Sciences, 2008 criteria that judged the quality of the workshops and associated outputs. It was concluded that incorporating SDM materials in risk communication efforts is in fact worth the extra time they consume and effort from participants as these materials helped participants to make better, more holistic, and ultimately more ethical decisions. 17 Works cited Andrianantoandro, E., Basu, S. Karig, D., and Weiss, Ron. (2006). Synthetic biology: new engineering rules for an emerging discipline. Molecular Systems Biology 2:0028 Arvai, J., Gregory, R., and McDaniels, T. (2001).Testing a Structured Decision Approach: Value Focused Thinking for Deliberative Risk Communication. Risk Analysis, 21:6, 1065-1076 Clemen, Robert and Reilly, Terence (2000). Making Hard Decisions: An Introduction to Decision Analysis. Duxbury Resource Center CIHR, Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada. (1998). Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (with 2000, 2002 and 2005 amendments). Retrieved on March 16,2009 from http://www.pre.ethics.gc.ca/policy- politique/tcps-eptc/docs/TCPS%20October%202005_E.pdf CIHR. (2007). CIHR Guidelines for Health Research Involving Aboriginal People. Retrieved on March 16, 2009 from http://www.cihr-irsc.gc.ca/e/29134.html Daniel, Wayne .W. (1990) Applied Nonparametric Statistics, 2nd Edition. PWS-Kent Publishing Company, Boston, Massachusetts. Dryzek, John. (1990) Discursive Democracy: Politics, Policy and Political Science. Cambridge: Cambridge University Press. Fisher, Ann. (1991). Risk communication challenges. Risk Analysis. 11,2, pl73-179. Hammond JS, Keeney RL, Raiffa H. (1999). Smart choices. A practical guide to making better decisions. Harvard Business School Press, Boston, MA. Hazlitt, Henry. (1946). Economics in One Lesson Harper and Brothers Publishers New York, pg 215. Health Canada. (2005). Issues related to the regulation of pre-implantation genetic diagnosis under the Assisted Human Reproduction Act. Assisted Human Reproduction Implementation Office. Interagency Advisory Panel on Research Ethics. (2008). Draft 2nd Edition of the Tri- Council Policy Statement: Ethical Conduct for Research Involving Humans. Retrieved on March 16, 2009 from http://www.pre.ethics.gc.ca/policy-politique/docs/TCPS-Draft2- eng.pdf 18 Irwin, A. (1995). Citizen science. A study of people, expertise and sustainable development. London: Routledge Joy, B. (2000). Why the Future Doesn't Need Us. Wired, 8.04, April Levin, Jack and Fox, James Alan. (2004). Elementary Statistics in Social Research: The essentials. Pearson Education Incorporated, USA Lidskog, Rolf. (2008). Scientised citizens and democratised science. Re-assessing the expert-lay divide. Journal of Risk Research. Vol. 11, No. 1-2, pg 69-86. Longstaff, Holly and Burgess, Michael. (Forthcoming). Recruiting for representation in public deliberation on the ethics of biobanks. Public Understanding of Science. McDonald, Michael. 2000. Biotechnology, Ethics, and Government A Synthesis. Prepared for The Canadian Biotechnology Advisory Committee Project Steering Committee On Incorporating Social and Ethical Considerations into Biotechnology. http://cbac-cccb.ca/epic/site/cbac-cccb.nsfyen/ah00438e.html Morgan, Granger M., Fischhoff, Baruch, Bostrom, Ann & Atman, Cynthia J.(2002).Risk Communication: A Mental Models Approach. Cambridge University Press. National Academy of Sciences. (2008). Thomas Dietz and Paul C. Stern, (Editors) Panel on Public Participation in Environmental Assessment and Decision Making, National Research Council. Organisation for Economic Co-operation and Development or OECD. (2003). Emerging Systemic Risks. Final Report to the OECD Futures Project. Paris, France. Perrow, C. (1984). Normal accidents: living with high risk technologies. Basic Books. Renn, Ortwin. (2003). Acrylamide: Lessons for Risk Management and Communication. Journal of Health Communication, 8, 5,435-441. Renn, Ortwin, (2004). The challenge of integrating deliberation and expertise. In Risk Analysis and Society: An interdisciplinary characterization of the field. McDaniels, Tim and Small, Mitchell (eds). Cambridge University Press, p 289-366. Renn, Ortwin (2006). White Paper on Risk Governance: Towards an Integrative Approach. International Risk Governance Council. Retrieved on November 12, 2008 from http ://www. irgc.org/IMG/pdf/IRGC WP No 1 Risk Governance reprinted version . pdf Renn, Ortwin & Klinke, Andreas. (2004).Systemic risks: a new challenge for risk management. Science and Society. October; 5 (Suppl 1): S41-S46. 19 Sabrina McCormick. (2007). Democratizing Science Movements: A New Framework for Contestation. Social Studies of Science . 37, 4, 609-623. Sherwin, Susan. 2008. Whither bioethics? How feminism can help reorient bioethics. The international journal of feminist approaches to bioethics 1:1: 7-27. Sirkin, R Mark. (1995). Statistics for Social Sciences. Sage Publications Inc. Thousand Oaks, California Slovic, Paul. (2007). "If I look at the mass I will never act": Psychic numbing and genocide. Judgment and Decision Making, 2,2, 79-95. Sunstein, Cass. (2002). "The Law of Group Polarization." Journal of Political Philosophy. 10,2. Tansey, James and MM Burgess, (forthcoming). Complexity of public interest in ethical analysis of Genomics. In D. Castle and K Culver (Eds), Genomics and Public Policy: Issues and Perspectives for Canada and the World. UBC Press. The Fertility Institutes. (2009). What's New. Retrieved on March 14, 2009 from http://www.fertility-docs.com/news_events.phtml?ID:=23 Walmsley, Heather. (2009). Mad scientists bend the frame of biobank governance in British Columbia. Journal of Public Deliberation: Vol. 5: No. 1,6. Webler T. & Renn O. (1995). A brief primer on participation: philosophy and practice. In O. Renn T. Webler & P. Wiedemann (Eds.) Fairness and competence in citizen participation. Pg. 17-33. 20 Chapter two: Integrating bioethics concepts with risk communication theory and applications 9 Introduction: the value of ethical analysis Ethics matters to risk communication. Unethical work masquerading as risk communication (e.g., to foster damage control, whitewash controversy or fulfill legislative obligations without real intent (Renn, 2004)) has the potential to erode public trust in the field of risk analysis (Morgan and Lave, 1990) and leave communication strategies incapable of achieving important objectives such as strategic or collaborative decision making with lay citizens (Null, 2002). Effective risk communication strategies therefore require strong ethical foundations to ensure that they are both ethically constructed and serving the public good. Although appeals for collaboration between the fields of applied ethics and risk communication have been made over time, limited progress has occurred to date.10 Some of these appeals point to the need for additional research on the ethical implications of particular communication methods (see the McComas, 2006 review of risk communication research from 1996-2005). Others call for more general advice regarding systemic hazards11 (Johnson, 1999; Sherwin, 2008). Whatever the purpose, encouraging risk analysts to engage with ethics literature should not be understood to imply that past or current risk communication work is necessarily unethical. It should instead be viewed 9 A version of this chapter has been submitted for publication. Longstaff, Holly. Blending bioethics concepts with risk communication theory and applications. 10 A few notable researchers who work in this interdisciplinary area include the following: Sherwin, 2001, 2008; Burgess, 2003, 2004 & 2005; Keeney, 1984, 1992, 1994; Renn, 2004, 2006; Morgan and Lave, 1990; Johnson, 1999; and Sellnow et al. 2008. 11 For more on systemic hazards please see the introductory chapter of this thesis. 21 as a way of enhancing and improving what risk analysts do and how they do it. For example, risk communicators working within the health domain can benefit from engaging with bioethics literature, or the critical examination of advances in the field of health technology and medicine. Those working in the area of bioethics represent a wide range of disciplinary backgrounds including philosophy, applied ethics, medicine, religion, social sciences, and law. These practitioners and academics spend their careers working on topics of great relevance to the field of risk analysis (e.g., protection of human research subjects, the ethical implications and governance of novel technologies) providing research findings and recommendations that should not be ignored. Many who are unfamiliar with the field of bioethics falsely believe that this area of research is inherently and endlessly subjective or thinly disguised political or religious rhetoric about right and wrong actions. In summary, moral judgements are "best described as integrative, holistic, or all things considered judgements requiring the intelligent integration of different types of knowledge and expertise" (McDonald, 2000 and Hurka 1993) and those who engage in bioethics research conduct systematic analyses of value-laden areas in a scientific manner (Danielson, 2007). Bioethicists also acknowledge sets of consensus values commonly shared by risk researchers (see section on the state of substantive ethics). Ethical analysis can be helpful in "determining responsibility, fair distribution of outcomes, and fair procedures", among other things (Sherwin, 2008). I will argue that the field of ethics can provide informative insights on a wide range of issues in risk communication including how to: identify the moral values that guide our efforts; build an ethical communication strategy; choose experts that will 12 In for example the 1979 Belmont Report and Beauchamp and Childress, 2001. 22 ultimately inform that strategy as well as the target audience; and evaluate findings. The fundamental objective of this paper is to demonstrate how bioethics concepts can help fortify risk communication strategies. In particular, it will show how explicitly considering substantive ethics generally and public ethics more specifically will help risk communicators cope with novel systemic hazards within the health domain. Conclusions are provided in the final section. Guidance from substantive ethics From how to why: moving beyond procedural ethics Much attention in the field of risk communication has been focused on procedural ethics or ethically constructed strategies for pursuing assumed (but often not explicitly considered) ends. For example, a great deal of effort had been directed to developing strategic frameworks for risk communication, such as those created by Government bodies in Canada, the United States, Turkey, the UK and others. Such frameworks naturally lead to greater examination of and (ideally) consensus on ethical procedures for risk communication. The processing of technically-oriented risk information has also been a prolific area of study for risk communicators. Studies consistently show that most individuals have difficulty understanding the "dose makes the poison" concept in exposure assessment (especially in regards to carcinogens (McCallum et al, 1990), and 13 Selected examples include: The Public Health Agency and Canada's (PHAC) framework for strategic risk communications; The Interagency Working Group on Import Safety's strategic framework for import safety; The Government of Turkey's strategic communication framework and plan for the prevention and control of avian and pandemic influenza; The Department for Environment, Food and Rural Affairs' strategic framework for the control of bovine tuberculosis. 23 dose response assessments (especially the relationship between high doses in animals and extrapolations to human health risks) (Fisher, 1991). Small probabilities (at or below one in a hundred lifetime risk) are also notoriously difficult to communicate (Fisher, 1991). Tversky and Kahneman (1974) demonstrated that judgment becomes increasingly impaired as probabilities approach 0.01 (Tversky and Kahneman, 1974) and Prelec 1998 showed that individuals are more likely to overestimate small probabilities if they are associated with catastrophic events. Contributions from these areas are a vital step forward in advancing ethical risk communication as strategies must be ethically constructed and understood (procedural ethics) to be effective. However, one must pay close attention to substantive ethics, or the moral values that guide us, when crafting risk communication strategies in addition to procedural ethics as both influence our ability to achieve just outcomes (Sherwin, 2001). The following case helps to illuminate the differences between procedural and substantive ethics and the significance of the latter concept to the field of risk communication. It will demonstrate how ethically constructed strategies created by well intentioned analysts may still lead to unethical outcomes. Imagine for a moment that this case takes place within A CULTURE14 that has major health disparities between its Aboriginal (approximately 5% of the total population) and non-Aboriginal citizens. For example, A CULTURE Aboriginal citizens experience higher rates of Fetal Alcohol Spectrum Disorder, a condition that describes a range of disabilities linked to alcohol exposure in the womb. Suppose that a risk communicator was hired by A CULTURE'S Federal Health Department to communicate the ways in which an individual's risk for this disorder can be minimized. This risk communicator 141 employ the convention A CULTURE to refer to a hypothetical specific culture. 24 truly wants to help the Aboriginal community address this health problem. She takes great efforts to construct a communication strategy that fits within the guidelines set by her region's strategic framework and is very thorough in her methods. She ensures that the materials she creates are effective in helping all interested citizens clearly understand the risk factors for Fetal Alcohol Spectrum Disorder and how to avoid them. Hence the risk communication process meets the test of procedural ethics and is unfolded by A CULTURE'S Federal Health Department. However, the risk communication strategy outrages A CULTURE'S Aboriginal population. They explain that the strategy has stigmatized them as a people even though the disorder is not solely an Aboriginal health problem. They also point out that although they were consulted by the risk communicator after the health topic was picked, they would have suggested a different, and in their opinion, more important health issue if given the opportunity. So despite the fact that the risk communicator was motivated by good intentions and constructed an ethical risk communication strategy, the process does harm15 to the target population by failing to address their health concerns and ultimately stigmatizing them. It therefore violates many of the moral obligations and shared principles that both bioethicists and risk analysts hold dear (see section on the state of substantive ethics) and is thus substantively unethical.1 The concept of harm can also be difficult to characterize. While this term may imply physical bodily harm to some, others may argue that it should also include emotional and social impacts as outlined in the A CULTURE example. 16 The case presented here is intended to remind the reader of the infamous Nu-chal-nuth First Nations incident involving the unauthorized use of blood samples for research purposes (Macintosh, 2005). Similar cases could be made for risk communication strategies focusing on products that caused great suffering in animals or those produced through child labour (see Sellnow at al, 2008 pg 150 for more on these examples). 25 The state of substantive ethics within risk communication: a few examples Risk communication is the process of informing and at times directing human behaviour in ways that are perceived to be in the public interest. This desire to do what is in the best interest of the public is an example of substantive ethics. To determine this "good" risk analysts typically evaluate options according to criteria such as risks or costs and benefits, but some have moved beyond these metrics to examine what may motivate and obligate an analyst to communicate with citizens. Morgan and Lave argue that we can limit harms associated with persuasive efforts by ensuring that the goals of communication strategies are explicit, overt, and altruistic (Morgan and Lave, 1990). In this way, Morgan and Lave begin to move beyond procedural ethics by examining motivation. They state that a communicator's motivations can be divided into two types: caring about how the strategy is used and not caring. The "caring" motivated risk communication messages can be further divided into covert or overt messages, which can be categorized as having either altruistic or selfish goals. These authors suggest that ethical dilemmas are more likely to occur when objectives of the risk communicator and target audience do not match and when the communicator produces covert, selfishly motivated messages. Although the authors do cover new ground in their paper, their arguments are somewhat problematic. First, it is not clear whose interests should be cared for in this framework and whose interests should be a priority. Secondly, Morgan and Lave would have us believe that overt messages are more likely to be ethical, provided that they were motivated by caring. The A CULTURE example above demonstrates how strategies created by caring analysts can nonetheless harm citizens. They are assuming that virtuous communicators are ethical communicators, but this is not always true. 26 Producing virtuous risk analysts who are ethically aware is certainly a good thing. However, such communicators may still create unethical strategies with the best of intentions if they do not understand and take into account the diverse interests and values of their audience, balance those interests in a fair and transparent manner, and determine a morally acceptable course of action. (Sellnow et al, 2008). The broad ranging interests of a diverse public are explored by those who suggest that risk communicators are morally obligated to consult citizens regarding issues that substantially influence their lives (e.g., GMO foods, radon gas, and genetic science applications). Schrader-Frechette refers to this obligation among her four reasons for why managers ought to involve the public in risk analysis. 17 She argues that it is ethical to involve the public when they have a stake in the consequences of an assessment (Schrader-Frechette, 1995). Burgess and Tansey make similar statements in their article on democratic deficits and the ethics of public engagement. They report that a democratic deficit occurs where there is unequal access to shape regulatory decisions (Burgess and Tansey, 2006). Others suggest that risk communicators are also obligated to help direct attention and limited financial resources on actual risks (Fisher, 1991). In other words, communicators must not allow public attention to shift from real to phantom risks. Fisher discusses this concept in her case analysis of tritium clean up after the Three Mile Island disaster. Residents objected to having this toxic water trucked away from the 17 Schrader-Frechette refers specifically to the risk assessment process but this argument can be appropriately applied to risk communication. Other motivations for including the public according to Schrader-Frechette are 2) logical, "uncertainty in risk assessments cause the invocation of value judgments", 3) ontological "assessments affect many things like equity, future generations, civil liberties. They should not just therefore have scientific inputs or bounded scientific rationality", and 4) democratic ideals "if the public cannot participate then it reduces democracy. It is not value neutral" See Schrader- Frechette, 1995. 27 site despite the fact that releases were well below permitted levels. Not fully understanding the risks involved, they instead agreed to have it boiled, which released the tritium into the air. This option was not only more expensive, it was also more hazardous to the residents' health (Rasmussen, 1990). Finally, Johnson creates a framework for identifying ethical issues in risk communication in a paper that responds to the earlier Morgan and Lave piece (see Figure 2.1). He argues 1 St that communicators must consider how each stage in his proposed framework "upholds or undermines" three general obligations (Johnson, 1999). These obligations relate to the commonly held mid-range bioethical principles of autonomy (self governance),19 beneficence (do good), non malfeasance (do no harm), and justice (do what is in the best interests of all) (Beauchamp and Childress, 2001). These principles are described as mid - range because unlike rules or specific moral judgments, they do not dictate a clear course of action (Dworkin, 1975 pg 1060). On the contrary, they are general, often conflicting with each other across contexts and in doing so, foster considered decision making by exposing tradeoffs and conflicting considerations (Keeney, 1984). 18 The stages include: identifying the issue, setting goals, knowing the issue, knowing the audience, knowing the constraints, assessing audiences, identifying messages, identifying methods, (after implementing a strategy) debriefing, and final evaluation. 19 Autonomy is not addressed in Johnson, 1999. 28 Figure 2.1 Ethical principles guiding risk communication 20 Johnson's three ethical principles (Johnson, 1999) Generally accepted mid range bioethics principles (Beauchamp and Childress, 2001) 1. Cause no pain, suffering, or loss 2. Improve the condition of the relevant constituency 3. Treat all groups fairly, tell the truth, and keep promises Nonmaleficence- do no harm "• Beneficence-do good Justice -do what is in the best interest of all It is perhaps useful to return to the A CULTURE example at this point to help think through the ways in which risk communication could be improved by appealing to the shared substantive bioethics and risk communication principles discussed thus far. It would seem that a virtuous risk communicator would recognize the democratic deficit exposed by the A CULTURE situation, where Aboriginal citizens are clearly unable to influence health policy within their community. This virtuous analyst would be motivated 20 Johnson's third principle encompasses a range of issues including truth telling and honesty but is nonetheless similar in nature to Beauchamp and Childress' description of justice. It should be noted that the list of bioethics principles presented here is not intended to be exhaustive. They are coherent with one another and represent a common morality (Marcum, 2008 and Schmidt-Felzmann, 2003). The principles are included here only to demonstrate that the most commonly accepted and cited principles in bioethics are already embedded within the work of most risk analysts. Other noteworthy ethics principles include The Belmont Report's basic ethical principles (The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1979); those discussed in the World Medical Association's (WMA) Declaration of Helsinki (The World Medical Association, 1964); and those highlighted in the World Health Organization's Pandemic Influenza Preparedness and Response document (World Health Organization, 2009). 29 to help direct public attention to this important deficit by crafting persuasive strategies that accurately take into account and communicate the diverse interests and values of all A CULTURE citizens. Through employing the mid range principles in Figure 2.1, the communicator would first notice that the autonomy of A CULTURE Aboriginal citizens was being violated. Although they did participate in the risk communication strategy, they were not consulted early enough in the process. These citizens were therefore not given the opportunity to exercise self governance by directing attention to health topics of the greatest concern to their community. Although the research could benefit many individuals within this society (beneficence), these benefits must also be measured against harm done to the group in the form of stigmatization (non-maleficence). A simple weighing of net risks and benefits in cost-benefit form will not suffice here. The principle of justice requires that risks and benefits be distributed fairly across groups. Ultimately, the ends (reducing Aboriginal citizens' risks for Fetal Alcohol Spectrum Disorder) although important, may not justify the means (violating Aboriginal citizens' right to self governance). After considering all relevant issues discussed above, the analyst may advise A CULTURE'S Federal Health Department to avoid targeted risk communication efforts for controversial topics in the future and instead focus on educating the entire citizenry about the risks of such disorders. She may also suggest additional public engagement activities around health priorities, especially for groups at increased risk. Guidance from public ethics 30 Public ethics, as proposed by Sherwin, is a specific form of substantive ethics that probes moral responsibility at multiple scales of impact in relation to systemic hazards (Sherwin, 2008). This concept of a collective ethics represents a move away from a responsive ethics to a more proactive model. Specifically, she calls on her bioethics colleagues to address broader questions that tend to be overlooked when bioethicists respond to narrow, albeit significant, controversial questions surrounding issues such as rights of the fetus. Topics addressed through public ethics are unlike conventional bioethics topics in that they are unbounded, present novel ethical dimensions (e.g., "new oughts"), and involve cross-scale, ambiguous impacts for a wide range of stakeholders (Sherwin, 2008). For example, personal actions alone (e.g., using public transit, consuming local products) cannot mitigate the risks and benefits of systemic hazards (e.g., the effects of global climate change on a community). It could be argued that all actions (even those of powerful organizations) stem from a collection of individual personal decisions and that all citizens could theoretically unite to foster tremendous social change. However, this line of reasoning assumes that all citizens are equally powerful, able, and equipped to voice dissention within a society that privileges certain genders, physical abilities, worldviews, sexual orientations, and cultural groups, among other things. Alternatively, individual actions such as screening human embryos for disease may lead to unintended consequences for groups of people (e.g., stigmatization of people with disabilities) and alter the genetic makeup of society for all future citizens. Figure 2.2 below shows some of the key value objectives that are likely to be present at each scale for systemic hazards in the health domain. These objectives were derived through interviews with experts (see 31 chapter 3 of this thesis) and are guided by the bioethics principles shown in Figure 1. The figure also outlines a series of key questions that may help sort out moral responsibility at and between scales. Figure 2.2 Examples of objectives and moral responsibilities by scale Key value objectives by scale guided by Figurel bioethics principles Key questions to help determine moral responsibility at and between scales Individuals: to protect and maintain the health of their immediate family (promoting autonomous decision making) Groups: To encourage better lives for members of Group X (fostering the principle of beneficence) and to avoiding discrimination against Group X (fostering the principle of nonmaleficence) Society: To foster overall health & wellbeing of Society X, to foster overall diversity of Society X, to treat all members of Society X equally (fostering the principle of justice) What information types and amounts are needed and who should contribute to these materials? What social and ethical norms are influencing or constraining the ability of the individuals to achieve their objectives? What responsibilities do individuals believe that they should have to groups and society? What tradeoffs are they willing to live with (if any) and why? What actions can groups take that will allow them to achieve their objectives while still respecting the autonomy of individuals or the ability of other groups to achieve their legitimate objectives? Is there a time when a group's objectives must be set aside for the good of society as a whole? How do we identify stakeholders for these novel hazards? Which group's objectives should be prioritized for each systemic health hazard? Are we satisfied with our current norms and systems? Can we change them in ways that allow individuals and groups to do good for others and themselves? * Arrows indicate pathways of influence 32 Sherwin seeks guidance from the work of other feminist scholars in her quest to pursue public ethics. One concept of particular influence is referred to as relational ethics (Sherwin, 1998). This term describes the idea that individuals, groups, and institutions are all connected to one another in our society. As shown in Figure 2.2, the ability to achieve key value objectives at each level are shaped (and constrained) by the moral responsibilities present at other scales of influence. Those studying through the lens of relational ethics examine the quality of decision making at the individual scale within in a wider social context that includes the influence of norms, regulations, responsibilities, culture, and other factors. Relational ethics is thus a political notion promoting autonomous decision making (Burgess and d'Agincourt-Canning, 2001). This is an idea that is also recognized within the field of health risk communication. In their discussion of genetic testing for cancer risks, Croyle and Lerman report that communication strategies should focus on both individual and family risk while discussing management plans (Croyle and Lerman, 1999). Ball, Evans, and Bostrom (1998) recommend encouraging doctor-patient "decision making partnerships" in order to expand the range of available options. Burgess and d'Agincourt-Canning have found that relational responsibility is a significant influence on an individual's decision to pursue genetic testing (Burgess and d'Agincourt-Canning, 2001). In other words, an individual's responsibilities as a caregiver, wife, mother, etc. may well influence and perhaps constrain the health related choices she ultimately makes. 21 Value focused thinking seeks to "clarify, define, and structure" an individual's fundamental values (in the form of objectives hierarchies) to guide decision making. Keeney argues that values are the fundamental component of decision-making and focusing on values as objectives can help one to achieve more desirable situations which he calls "decision opportunities" as apposed to "decision problems" (Keeney, 1992; Keeney, 1994, and McDaniels, 2000). 33 Employing bioethics guidance to communicate the risks and benefits of systemic health hazards Figure 2.3 illustrates how lessons from the field of bioethics can inform and fortify risk communication strategies within the novel regulatory and decision making frameworks discussed below. Public substantive ethics, as described earlier, provides the conceptual framework for identifying moral responsibilities at multiple scales that influence our ability to achieve key value objectives for individuals, groups, and society when dealing with systemic hazards. More general substantive ethical considerations offer moral guidance for risk communicators such as the moral values that guide risk management plans, the target audience, and identifying the range of experts and stakeholders who will inform the strategy. For example, it is common and appropriate for risk communicators to seek the advice of conventional experts such as scientists and regulators who have cultivated a deep understanding of a particular hazard through study and analysis over time. However, in some cases, it may be important to include the views of unconventional stakeholders who also posses relevant as well as morally significant information (such as lived experiences). Such views, much like the views of conventional experts, are formed over many years, are often difficult for non stakeholders to anticipate, and cannot be assumed. Procedural ethics focuses on ways to build and evaluate ethical strategies. 34 Figure 2.3 Bioethics guidance for risk communication strategies regarding systemic hazards Public substantive ethics Conceptual framework for identifying moral responsibilities at multiple scales that influence our ability to achieve fundamental value objectives for individuals, groups, and society in relation to systemic hazards General substantive ethics Guidance for identifying experts and stakeholders, the moral values to guide risk management plans, and the target audience Unconventional stakeholder experts Conventional experts Lay citizens General procedural ethics Guidance on building ethical communication strategies and evaluating findings Developing risk communication strategies that include value-oriented and technically- oriented information Experts from both the fields of risk analysis and bioethics agree that systemic health hazards call for new approaches and methods (discussed below). This paper proposes ethically framed and constructed risk communication as a way of bridging the gap between what the individual can and ought to do and what we think our institutions and others can and ought to do in the face of systemic hazards. Schrader-Frechette (1995) and Burgess (2003), among others, argue that citizens have the right to determine the kind of society in which they want to live. Yet it is impossible for each individual citizen to directly shape the risk analysis processes for a given hazard.22 Risk communication efforts can gather groups of citizens together to examine systemic hazards in order to Involvement here refers to meaningful and direct democratic participation in contrast to other democratic actions such as voting for elected officials. 35 provide public input into or oversight over each risk analysis phase.23 The fundamental objectives of this approach to risk communication are therefore to (1) provide moral guidance for normatively ambiguous problems and (2) consider (and potentially mitigate) consequences associated with interpretative ambiguities that typically lead to multiple scales of impact for a broad range of stakeholders.24 This holistic approach is inspired equally by the work of risk researchers and bioethicists who study similar problems yet rarely have the opportunity to collaborate. While risk analysts call for new regulatory and decision making frameworks capable of incorporating risk communication methods and findings, bioethicists provide advice on how to determine the range of moral responsibilities and values that ought to inform and guide those strategies. Risk researchers recognize the fact that methods capable of addressing ambiguous problems are ineffective without appropriate support from regulatory and decision making frameworks. Marden and co-authors point out that current frameworks may not be structured to cope with particular risk communication outputs (e.g., the ethical dimensions of technologies) (Marden, Longstaff, and Levy, 2006 and forthcoming). For example, in Canada, existing regulations and laws are being stretched to cover ambiguous problems but these efforts have been controversial (e.g., regulating GM, farmed, and wild salmon under the Fisheries Act). McDaniels et al, and Renn therefore present new frameworks capable of considering, directing, and mitigating the consequences of ambiguous hazards while incorporating novel communication strategies. In his recent 23 Typically, the public is brought into the risk analysis process only after hazards are identified and assessed. 24 For example, lay participants can help clarify shared ethical and social norms. Diverse groups of citizens may also be able to draw on personal experiences and worldviews in order to anticipate the outcomes of systemic hazards for other individuals and groups. 36 2006 white paper, Renn proposes a new Risk Governance management framework to cope with ambiguous systemic hazards. Renn's framework includes five core processes, 1) pre-assessment, 2) risk appraisal, 3) tolerability and acceptability judgment, 4) risk management, and 5) risk communication. Significantly, risk communication is located at the centre of the framework influencing and influenced by all other processes. Of particular interest to communicators is the impact that communication is recommended to have on the pre-assessment phase, which includes problem framing, early warning, screening, and determination of scientific conventions. Renn also introduces readers to the concept of a Concern Assessment, which includes gathering information on public risk perceptions, social concerns, and socio-economic impacts. This assessment is placed alongside the conventional risk assessment phase. Other examples of novel regulatory frameworks include the UK's Inter-departmental Liaison Group on Risk Assessment (ILGRA) Risk Communication Sub-Group's 1998 document "Risk Communication: A Guide to Regulatory Practice" (ILGRA, 1998), and the Canadian Government's 2000 Integrated Risk Management Framework (Treasury Board of Canada Secretariat, 2001). Both documents incorporate risk communication into every stage of the analysis process and could easily absorb communication methods and findings (Treasury Board of Canada Secretariat, 2001). McDaniels et al 2005 & 2006 propose a new risk management decision making framework capable of addressing the ethical and normative dimensions of multi-scalar systemic hazards (in this case salmon aquaculture). This framework involves (1) characterizing the regulatory context for a problem at each scale, (2) eliciting objectives from all affected parties at each scale, and (3) characterizing performance 25 This is similar to Schrader-Frechette's probabilistic risk assessment in moral mathematics. See Kristin Schrader-Frechette, 1987. 37 measures for those objectives. McDaniels et al have found that even the most diverse groups often share the same fundamental value objectives for controversial hazards. Establishing these points of convergence can foster communication between groups and allow managers to arrive at decisions all stakeholders can live with. For systemic ambiguous hazards, one must choose communication methods capable of addressing a broad range of individual, group, and societal risks and benefits that incorporate both technically-oriented and value-oriented risk information. Novel methods such as risk ranking and more conventional methods such as focus groups can be used to effectively engage citizens during such risk communication sessions. It is common during risk ranking exercises to ask diverse groups of participants to rate various management options associated with a hazard (EPA, 2005). However, this exercise can also be used to choose hazards that ought to be considered for risk assessments thereby including participant values and interests in the most preliminary stages of the risk analysis process. With systemic hazards such as genetic and genomic science applications, there is often substantial disagreement as to how or even if technologies or activities associated with these topics should be pursued thus taking the discussion beyond procedural ethics into the area of substantive public ethics. Participant recommendations may include additional research, additional regulations, or additional public input from other sub populations. Interactions between diverse participants during focus groups or workshops can be used to expose society's ethics and reveal the norms that should guide risk communication strategies (Burgess and Tansey, 2006, Burgess, 2003). Communication efforts informed through such processes will therefore reflect the values of citizens and risk analysis 38 options will be justified by citizens and not just technical risk experts (Fung, 2003). Such discussions allow for greater representativeness and inclusiveness (Burgess and Tansey, 2005; Longstaff and Burgess, forthcoming) by allowing participants the freedom of expressing under represented or unarticulated opinions (Burgess, 2004). In this way, participants are able to "move beyond established notions of what is relevant" in order to "interrogate assumptions" (Levitt, Weiner, and Goodacre, 2005) and are thus capable of informing an ethical evaluation of the topic at hand (Burgess and Tansey, 2006). The results of these discussions allow citizens to consider the socially significant consequences of a hazard while determining the range of values that ought to shape risk analysis conclusions. Ethical decisions concerning the consequences of ambiguous systemic hazards must be "all things considered" judgments (McDonald, 2000). Methods that foster deliberative engagement encourage citizens to balance and even override self interest (even if this interest is very high) in favour of just policies for all (Ball, Evans, andBostrom, 1998). Conclusions The objective of this paper was to show how lessons from the field of bioethics can help risk analysts cope with systemic hazards in the health domain through risk communication. This paper represents a preliminary step towards integrating a selection of complimentary theoretical concepts from both fields in order to foster communication between the two disciplines. Much work remains to be done and progress in this area 26 An additional step will involve writing a manuscript for the bioethics community from the risk analysis perspective. This paper will outline a series of insights that could be used to fortify bioethics research (e.g., 39 will inevitably involve a number of significant challenges. McDaniels, Renn, and others present innovative ways of incorporating the moral dimensions of risk problems into regulatory and decision making frameworks while recognizing the consequences of ambiguous problems at multiple scales. However, it should be noted that overhauling current frameworks would be both difficult and expensive. For example, Renn states that risk communication under his framework should be performed by an interdisciplinary team of communicators, which would require additional training and perhaps even changes in personnel. In addition, some risk communication methods such as focus groups and workshops are time consuming and expensive. A great deal of research needs to be accumulated to determine if these methods are in fact effective in providing moral guidance for and considering the consequences of systemic health hazards. Only a broad overview was provided here. A range of empirical data must first be collected before one can justify using such methods to inform actual policy. Questions of immediate concern include: approaches for blending value-oriented and technically-oriented risk informative for lay audiences in ways that are both informative and trustworthy; the range of stakeholders and experts that ought to inform risk communication materials for a particular systemic hazard; and the minimum amount of information that should be provided during risk communication sessions for systemic hazards that are invariably complex, to name only a few. Informational sufficiency or the functional aspects of the informed consent process has been a prolific area of research for the field of bioethics despite the fact that the social empirical evidence on the effectiveness of communication methods, relevant risk perception findings, and the benefits of public engagement) 40 dimensions in which the informed consent process is embedded is often ignored (Corrigan and Williams-Jones, 2006) and there is a general "lack of consensus regarding the most appropriate method with which to communicate medical risks" (Ghosh and Ghosh, 2005). Nevertheless, it is now widely recognized that presentation of data has a strong influence on patient decisions (Mazur and Merz, 1993) and that decision theory can and should be used to improve the informed consent process (Holmes-Rovner and Wills, 2002). It appears that any decision aids, "regardless of the format or graphic representation of data" serve to increase patients' knowledge (Holbrook et al., 2007) and that combining both quantitative and qualitative techniques is particularly effective (Ghosh and Ghosh, 2005). However, it is vital to point out that engaging citizens in deliberative risk communication efforts about emerging technologies is no substitute for patient centered risk communication efforts that seek to foster an informed consent relationship between a patient and a genetic counselor or physician. Informed consent directly involves an individual (or group in some cases) by requesting permission for an action that would not otherwise be permitted (e.g., a medical intervention). Deliberative risk communication on the other hand is performed at the social or aggregate level for collective decision making. While the objective of a deliberative exercise with citizens is to develop socially acceptable risk communication strategies, the intent of the informed consent relationship is to help patients make informed choices that they and their families can live with. Lastly, bioethicists, risk communicators, and others need to be provided with opportunities to meet in settings that foster (and respect) an interdisciplinary approach to dealing with systemic hazards. Initiating such collaborations will likely be difficult given 41 the fact that communication between the groups rarely occurs despite the emergence of new interdisciplinary journals, funding opportunities, and degree programs at a variety of major academic institutions. Nonetheless, systemic health hazards must be confronted as they have the potential to significantly impact a broad range of stakeholders and systems. Meeting this new risk communication challenge is one way that communicators and bioethicists can come together to promote the overall health and wellbeing of individuals, groups, and society. 42 Works cited Ball, Leslie K., Evans, Geoffrey, and Bostrom, Ann. (1998). Risky Business: Challenges in Vaccine Risk Communication. Pediatrics, 101, 453-458. Beauchamp, T and Childress, J (2001) Principles of biomedical ethics, fifth edition. Oxford University Press. Beecher, Henry. (1966). Ethics and clinic research. New England Journal of Medicine. 274, pg 1354-60. Burgess, Michael and d'Agincourt-Canning, Lori. (2001). Genetic testing for hereditary disease: Attending to relational responsibility. The Journal of Clinical Ethics, 12,4,361-372. Burgess, Michael M. (2003). Starting on the right foot: public consultation to inform issue definition in genome policy. Electronic Working Papers Series. W. Maurice Young Centre for Applied Ethics, University of British Columbia. Retrieved on September 22, 2008 from www.ethics.ubc.ca. Burgess, Michael. (2004). Public Consultation in Ethics: An Experiment in representative ethics. Journal of Biomedical Inquiry. 1, 1.4-13. Burgess, Michael, and Tansey, James. (2005). Complexity of public interest in ethical analysis of genomics: Ethical reflections on salmon genomics/aquaculture. Electronic Working Papers Series. W. Maurice Young Centre for Applied Ethics, University of British Columbia at www.ethics.ubc.ca. Burgess, Michael and James Tansey. (2006). Democratic deficit and the politics of "informed and inclusive" consultation'. Hindsight and Foresight on Emerging Technologies. Ed. E Einsiedel and R Parker. Vancouver: UBC Press. Burgess, Michael and Williams-Jones, Bryn. (In progress). Prenatal testing (working title). Corrigan, O. and Williamsjones, B. (2006). Pharmacogenetics: the bioethical problem of dna investment banking. Studies in History and Philosophy of Biological and Biomedical Sciences. 37, 3, 550-565. Croyle, Robert T., and Lerman, Caryn. (1999). Risk communication in genetic testing for cancer susceptibility. Journal of the National Cancer Institute Monographs, 25. 59-66. Department for Environment, Food and Rural Affairs. (2005). Bovine TB: Strategic Framework for the sustainable control of bovine tuberculosis (bTB) in Great Britain. 43 Retrieved on September 22, 2008 from http://www.de.fra.gov.uk/aiiimalh/tb/strategv/newstrategy.htm Dworkin, Ronald. (1975). Hard Cases. Harvard Law Review, 88, 6 , pp. 1057-1109 Emanuel, Ezekiel, Robert Crouch, John Arras, Jonathan Moreno and Christine Grady- Editors. (2003). Ethical and Regulatory Aspects of Clinical Research: Readings and Commentary. Johns Hopkins University Press. EPA. (2005). Community based watershed management. Lessons from the national estuary program. Retrieved on September 22, 2008 from http://www.epa.gov/owow/estuaries/nepprimer/documents/srAppendixC.pdf¥search= %22risk%20ranking%20epa%20management%22 EPA. (2006). National Environmental Justice Advisory Council. Retrieved on September 22, 2008 from http://www.epa.gOv/compliaiice/environmentaliustice/nejac/.index.htnil. Fisher, Ann. (1991). Risk communication challenges. Risk Analysis, 11,2,173-179. Fung, Archon. (2003). Survey Article: Recipes for Public Spheres: Eight Institutional Design Choices and their Consequences. The Journal of Political Philosophy, 11, 3, 338-367. Ghosh AK, Ghosh K. (2005). Translating evidence-based information into effective risk communication: Current challenges and opportunities J Lab Clin Med. 145,4, 171-80. Government of Turkey. (2006). Strategic communication framework and plan for the prevention and control of avian and pandemic influenza in Turkey. Retrieved on September 22, 2008 from http://www.unicef.org/turkeY/pdf/he3.pdf Holbrook, Anne, Renee Labiris, Charles H. Goldsmith, Kaede Ota, Sandra Harb, Rolf J. Sebaldt. (2007). Influence of decision aids on patient preferences for anticoagulant therapy: a randomized trial. CMAJ. 22; 176,11,1583-7 Holmes-Rovner, Margaret,Wills, Celia E.(2002). Improving Informed Consent: Insights From Behavioral Decision Research. Medical Care. 40 9, 30-38. Hurka, Thomas. (1993). Ethical Principles in Ethics and Climate Change, Harold Coward and Thomas Hurka (Editors). Wilfrid Laurier University Press, Waterloo, Ont. p. 23 Interagency Working Group on Import Safety. (2007). Protecting American consumers every step of the way: A strategic framework for continual improvement in import safety. Retrieved on September 22, 2008 from http://www.importsafetv.gOv/report:/report.pdf 44 Inter-departmental Liaison Group on Risk Assessment (1998). Risk Communication: A Guide to Regulatory Practice. Retrieved on September 22, 2008 from http://www.hse.gov.uk/aboutus/meetings/ilgra/risk.pdf Jasanoff, Sheila (1989). Differences in national approaches to risk assessment and management. Presentation at symposium on managing the problem of industrial hazards: the international policy issues. National Academy of Sciences. Washington DC. Feb 27. Johnson Branden B, & Slovic P. (1994). "Improving" risk communication and risk management: legislated solutions or legislated disasters? (Editorial). Risk Analysis, 14,6,905-6. Johnson, Branden, B. (1999). Ethical issues in risk communication: continuing the discussion. Risk Analysis, 19, 3, 335-348. Keeney, Ralph. (1984). Ethics, decision analysis, and public risk. Risk Analysis, 4, 2, p 117-129. Keeney, Ralph L. (1992). Value focused thinking: a path to creative decision making. Harvard University Press, 16-22. Keeney, Ralph L. (1994). Creativity in decision making with value focused thinking. Sloan Management Review, 35, 4, 33-41. Levitt, Mairi, Weiner, Kate, and Goodacre, John. (2005). Gene Week: a novel way of consulting the public. Public Understanding Science, 14, 67-79. Longstaff and Burgess. (Forthcoming). Recruiting for representation in public deliberation on the ethics of biobanks. Public Understanding of Science. Longstaff, Holly. (Under review). Characterizing expert mental models for systemic hazards: the case of pre-implantation genetic diagnosis. Journal of Risk Research. Macintosh, Constance. (2005). Indigenous Self-Determination and Research on Human Genetic Material: A Consideration of the Relevance of Debates on Patents and Informed Consent, and the Political Demands on Researchers. Health Law Journal Vol. 13, pg 213-252. Marden, Emily, Longstaff, Holly and Levy, Ed. (2006). The Policy Context and Public Consultation: A Consideration of Transgenic Salmon. Journal of Integrated Assessment, 6, 2, 73-97. Marden, Emily, Longstaf, Holly, and Levy, Ed. (Forthcoming). Transgenic Salmon: Regulatory Oversight of an Anticipated Technology. In Hindsight and Foresight on Emerging Technologies. Ed. E Einsiedel and R Parker. Vancouver: UBC Press. 45 Mazur DJ, Merz JF. (1993). How the manner of presentation of data influences older patients in determining their treatment preferences. J Am Geriatr Soc. 41, 223- 8. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. (April 18, 1979). The Belmont Report: Ethical Principles and Guidelines for the protection of human subjects of research. Retrieved on June 17, 2009 from http://ohsr.od.nih.gov/guidelines/belmont.html Null, Marilyn. ( 2002). A Risk Communication Revolution: Process vs. Event. Presentation at the Baltimore, MD Best Practices Conference. Retrieved on September 22, 2008 from http://ch.ppm- www.apgea.army.mil/risk/PDF/BestPractices Null.pdf Marcum, James (2008) Review - An Introductory Philosophy of Medicine: Humanizing Modern Medicine. Springer McCallum, DB, Hammond SL, Morris, LA, and Covello, VT. (1990). Public knowledge of chemical risk in six communities: analysis of a baseline survey. EPA. McDaniels, Timothy. (2000). Creating and using objectives of ecological risk assessment and management. Environmental Science and Policy, 3, 299-304. McDaniels, Timothy L., Dowlatabadi, Hadi, Stevens, Sara.(2005). Multiple scales and regulatory gaps in environmental change: the case of salmon aquaculture. Global Environmental Change Part A, 15,1. 9-21. McDaniels, Timothy, Longstaff, Holly, and Dowlatabadi, Hadi. (2006). A Value- Based Framework for Risk Management Decisions Involving Multiple Scales: A Salmon Aquaculture Example. Journal of Environmental Science & Policy, 9, 5, 423- 438. McDonald, Michael. (2000). Biotechnology, Ethics, and Government A Synthesis. Prepared for The Canadian Biotechnology Advisory Committee Project Steering Committee On Incorporating Social and Ethical Considerations into Biotechnology. Morgan, Granger, and Lave, Lester (1990). Ethical considerations in risk communication practice and research. Guest editorial. Risk Analysis, 10, 3, 355-358. National Research Council's Committee on Risk Perception and Communication. (1989). Improving risk communication. The National Academies Press. Perrow, C. (1984). Normal accidents: living with high risk technologies. Basic Books. PHAC. (2007). A Framework for Strategic Risk Communications within the Context of Health Canada and the PHAC's Integrated Risk Management. Retrieved on 46 September 22, 2008 from http://www.phac-aspc.gc.ca/publicat/2007/risk-corn/index- eng.php Prelec, D. (1998). The probability weighting function. Econometrica, 66,497-529. Rasmussen. (1990) "Risk analysis: what challenges remain?" SRA conference keynote address cited in Fisher, Ann. (1991) op cit. Renn, Ortwin. (2003). Acrylamide: Lessons for Risk Management and Communication. Journal of Health Communication, 8, 5,435-441. Renn, Ortwin, (2004). The challenge of integrating deliberation and expertise. In Risk Analysis and Society: An interdisciplinary characterization of the field. McDaniels, Tim and Small, Mitchell (eds). Cambridge University Press, 289-366. Renn, Ortwin (2006). White Paper on Risk Governance: Towards an Integrative Approach. International Risk Governance Council. Retrieved on September 22, 2008 from http://www.irgc.org/irgc/proi ects/risk characterisation/ b/contentFiles/IRGC_WP N o_l_Risk_Govemance (reprinted_version').pdf#search=%22%20renn%20risk%20go vernance%20towards%20an%20integrated%20%22 Schmidt-Felzmann, Heike(2003) Pragmatic Principles - Methodological Pragmatism in the Principle-Based Approach to Bioethics. Journal of Medicine and Philosophy, 28:5,581-596 Schrader-Frechette, Kristin. (1987). Parfit and Mistakes in Moral Mathematics. Ethics, 98, 50-60. Schrader-Frechette, Kristin. (1995). Evaluating the Expertise of Experts. Risk, 6, 115- 126. Sellnow, Timothy L., Robert R. Ulmer, Matthew W. Seeger, Robert S. Littlefield. (2008). Ethical Considerations in Risk Communication in Effective Risk Communication: A Message-Centered Approach (Food Microbiology and Food Safety) Springer New York, pg 147-161. Sherwin, S. (1998). A Relational Approach to Autonomy in Health Care, in The Feminist Health Care Ethics Research Network, S. Sherwin (coordinator) The Politics of Women's Health: Exploring Agency and Autonomy, pp. 19-47. Philadelphia, PA: Temple University Press . Sherwin, Susan. (2001). Towards an Adequate Ethical Framework for Setting Biotechnology Policy. The Canadian Biotechnology Advisory Committee Stewardship Standing committee. Retrieved on September 22, 2008 from http://www.cbac-cccb.ca/epic/internet/incbac-cccb.nsf/en/ah00404e.html 47 Sherwin, Susan (2008). Whither Bioethics? How feminism can help reorient bioethics. The international journal of feminist approaches to bioethics, 1,1, 7-27. Slovic, Paul. (2007). "If I look at the mass I will never act": Psychic numbing and genocide. Judgment and Decision Making, 2,2, 79-95. Treasury Board of Canada Secretariat. (2001). Integrated Risk Management Framework. . Retrieved on September 22, 2008 from http://www.tbs- sctgc.ca/pubs pol/dcgpubs/RiskManagemeni/rmf-cgrOl- 1 e.asp#Risk%20Management Tversky, A, and Kahneman, D (1974). Judgment under uncertainty: heuristics and biases. Science, 185,1124-1131. World Health Organization. (2009). Pandemic Influenza Preparedness and Response document. Retrieved on June 17, 2009 from http://www.who.int/csr/disease/influenza/PIPGuidance09.pdf World Medical Association. (1964). World Medical Association Declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects. Retrieved on June 17, 2009 from http://www.wma.net/e/policy/b3.htm Zimmerman, Rae. (1993). Social equity and environmental risk. Risk Analysis, 13, 6, 649-666. 48 Chapter three: Characterizing expert mental models for systemic hazards: the case of pre-implantation genetic diagnosis27 Introduction Characterizing how experts think about problems is an integral component of risk analysis for virtually any hazard. Individuals cultivate expert knowledge by working within a particular discipline or accruing specific skills or experiences over time. Risk analysts harness this expertise in several areas: probability elicitation, structuring risk assessments, developing modeling platforms, providing input on risk management decisions, and a range of other activities. The intention of this paper is to present a novel method of characterizing expert judgments about systemic hazards (as defined in the Introduction of this thesis) derived from mental model interviews in order to inform a participatory risk communication experiment. Numerous approaches exist for eliciting expert information about potential hazards. The intention of most is to obtain judgments in formats relevant for policy-makers (De Bruin, 2006). Some methods can be described as consensus oriented (De Bruin et al., 2006; and Cooke and Goossens, 2004), which seek to clarify and obtain an integrated view of the facts of the matter or the breadth of issues relevant to the topic at hand. Such approaches work best for problems that draw on cohesive groups of experts with similar skill sets. However, new emerging hazards, referred to here as systemic hazards, involve uncertain outcomes at multiple scales, cross economic, social, ethical, technological and other A version of this chapter has been submitted for publication. Longstaff, Holly. Characterizing expert mental models for systemic hazards: the case of pre-implantation genetic diagnosis 49 domains, and are not bound by national borders (Renn, 2006).28 These novel hazards require cooperation across sectors to support robust governance, risk management, and risk communication strategies (see McDaniels, Longstaff, and Dowlatabadi, 2006 and Renn, 2006 for examples ). Communication strategies for these hazards often draw on heterogeneous expertise including from, at times, somewhat unconventional experts. In addition, it can be difficult for risk communicators to collate information across these diverse expert groups as there is often no consensus amongst heterogeneous experts about the nature, extent, and even the reality of systemic problems. It is asserted here that these new problems therefore require novel methods of eliciting and characterizing expert judgments capable of clarifying uncertainties and the full range of likely outcomes to relevant stakeholders. The need to identify new methods, theoretical perspectives, findings, and goals in the field of risk communication that are capable of coping with novel hazards is a new challenge for risk analysts. This challenge has been underscored in a variety of locations including a recent special edition of the Journal of Risk Research that presented a selection of papers from the "New perspectives on risk communication: uncertainty in a complex society" 2006 research conference that took place in Goteborg, Sweden (Boholm, 2008). Systemic problems are unique in that they often involve both normative and interpretative ambiguity (Renn, 2006). Ambiguous risk problems differ from other complex hazards in 28 Examples of systemic health hazards include: global climate change, genetically modified foods, biobanking and other genetic and genomic science technologies. 29 In his recent 2006 white paper, Renn proposes a new "Risk Governance" management framework capable of handling ambiguous problems. Renn's framework includes 5 core processes, 1) pre-assessment, 2) risk appraisal, 3) tolerability and acceptability judgment, 4) risk management, and 5) risk communication. Significantly, risk communication is central to the framework influencing and being influenced by all other processes. Of particular interest to communicators is the influence that communication is recommended to have on the pre-assessment phase, which includes problem framing, early warning, screening, and determination of scientific conventions. 50 that there is typically substantial disagreement as to how or if technologies or activities associated with these problems should be pursued at all. Ambiguous risk problems require cooperation and engagement across sectors, expert groups, and civil society to determine an appropriate course of action. As such, these problems call for new participatory risk communication methods and approaches that reflect a range of social values (Renn and Klinke, 2004). The fact that a topic is normative (e.g., walking on the right side of the sidewalk) does not justify pursuing participatory risk communication. Normative ambiguity occurs in the "absence of a strong normative standard" (e.g., where there is no clear right side to walk on) (Burgess and Williams-Jones, in progress). Interpretative ambiguity occurs when it is unclear as to how the consequences of a potential risk will be experienced by individuals and groups at multiple scales and across domains. Participatory risk communication can be used to promote democratic public oversight or input for every phase of risk analysis (Renn, 2006 and Powell & Leiss, 1997).31 Systemic hazards32 are ripe with interpretative ambiguity. For example, health biotechnologies like pre-implantation genetic diagnosis (PGD) may lead to a range of multi scalar physical, emotional, ethical, economic, and social impacts that should considered by decision makers. Pre-implantation genetic diagnosis may pose risks and benefits to individuals who undergo or have PGD performed on them (e.g., unknown long-term health outcomes for children born following the procedure and women who 30 Take for example the ambiguous risk problem of genetically modified foods and activities associated with this topic such as mono-culture and mega-farming. 31 Unlike other approaches in which the public is brought into the risk analysis process only after hazards are identified and assessed. 32 Pre-implantation genetic diagnosis is used to analyze and screen genetic material from fertilized human embryos during in-vitro fertilization (see study context section.) 51 undergo PGD), individuals who do not undergo PGD (e.g., denial of insurance coverage to people who neglect testing and thus "choose" to bear children with disabilities), groups (e.g., by further medicalization of the pregnancy experience ), and society (e.g., through economic impacts on the Canadian health care system). Hoffmann et al report that most approaches for aggregating diverse expert judgements can be characterized as either behavioural (e.g., Delphi ) or mathematical (e.g., probabilistic decision modeling) (Hoffmann, et al., 2006). These manipulate analytic languages derived through quantifying expert judgments (Fischhoff, 1989). Characterizing the risks and benefits of systemic hazards, such as PGD, call for a systematic method of eliciting expert information that can also account for distinct communication patterns and skill sets of conventional (e.g., genetic counsellor, physician) and unconventional experts (e.g., those living with disability, caregiver perspective). It is for this reason that the methods described here were based on a behavioural approach known as the mental models method as described by Morgan et al. (2002).34 33 Delphi method is an expert decision making model where participants revise answers in response to group feedback in rounds (if necessary). The views of earlier rounds inform each successive round until participants come to agreement (Dalkey, 1969). Morgan and co-authors explain that an individual's mental model is informed by that person's life experiences or worldview and help the individual to navigate through a very complex world. The mental models method recognizes that the minds of laypeople are not homogenous. Instead, this approach makes allowances for a person's individuality while disseminating information in a systematic and ethical way. Each individual will use a set of ideas or mental models to predict the world around them and incorporate these ideas into their decision making processes. The dissemination process is designed to allow participants to examine a wide range of risks and benefits of a topic thereby allowing that participants to determine which options align with their values and beliefs (Morgan et al, 2002). 52 A mental model is a collection of thoughts and beliefs that a person holds concerning the functioning of the world (Morgan et al., 2002).35 During the first stage of the mental models approach for risk communication, expert information is organized into a format that speaks to these models by addressing major uncertainties or points of common interest. The approach calls for a broad range of experts to be interviewed to minimize interviewer bias. One potential drawback associated with techniques typically used for characterizing expert judgments is that they often rely on quantitative methods, such as questionnaires, that cannot make allowances for diverse communication patterns of participants (See for example Bruine de Bruin et al., 2006 and Goossens, et al., 1998). In addition, experts run the risk of operating at one level or in one domain and can only report on their side of the problem. For example, a physician who conducts PGD develops a deep understanding of the technical and perhaps ethical dimensions of this technology and how successful this procedure can be in avoiding particular disabilities in the biological children of their patients. However, this expertise cannot be substituted for the lived experience of someone who cares for children with screenable disabilities (who we are attempting to avoid) or who has lived with that disability all their life. On the other hand, it is inappropriate for a risk communicator to allow the lived experience of disability to replace expertise concerning the technical and other dimensions of PGD when crafting their communication strategies. For the case of systemic hazards that involve multiple scales of impact across multiple domains, it is important for risk communicators to promote a holistic view amongst experts in order to characterize the full range of risks and benefits posed by the hazard in question (Renn and Klinke, 2004). 35 It should be noted that the concept of a mental model can be traced back even earlier to the field of cognitive psychology (Holland et al, 1986; Johnson-Laird, 1983; and Gentner & Stevens, 1983 ) 53 The mental models method employs charts that show deterministic decision pathways including uncertain outcomes (Morgan et al., 2002). While other studies (e.g., Willems et al, 2005) have employed charts to elicit judgment from homogenous expert groups, using charts to collate the judgments of a heterogeneous group of experts for systemic hazards is a novel use of these decision-making tools (Willems et al., 2005). In the experiment discussed here, literature findings regarding the risks and benefits of PGD were captured in a series of charts that were used to explore expert mental models throughout a series of individual interviews. Using charts allows the interviewer and participant to systematically explore the full range of risks and benefits of PGD at each scale and domain but does not constrain the responses or communication styles of the expert. This paper focuses specifically on the ways in which the author characterized expert mental models of PGD and is one component of a larger research project. 3? The following section includes background information on PGD, the topic of interest for this study. The next section presents the results of an extensive literature review that sought to clarify risks and benefits at the family, group, and societal scales according to literature produced by experts relevant to PGD. The author's approach to characterizing expert mental models is described in the following section. This was accomplished through a series of expert interviews that were intended to confirm and supplement the literature review findings. Key insights and conclusions are presented in the final section. The book by Morgan et al uses the term influence diagram to refer to these figures. However other writers see influence diagrams as more formalized and with somewhat different conventions than seen in the Morgan work (e.g., Clemen, 1997). I use the term charts to avoid confusion. 37 Phase one of this project was approved by the University of British Columbia's Behavioral Research Ethics Board (ID#H07-02574). 54 Study context: pre-implantation genetic diagnosis Pre-implantation genetic diagnosis is used to analyze and screen genetic material from i n fertilized human embryos during in-vitro fertilization (IVF). During PGD, cells (called blastomeres) or the polar body (structure extruded from the egg or embryo after fertilization that contains a copy of the maternal genetic material) are extracted from the fertilized embryo at day 3 when the embryo has 8-10 cells (Chen, 2005). DNA from these embryos is then multiplied thousands of times to identify specific genes. Embryos that do not carry genetic markers for a specific condition are selected and implanted into the uterus of the female patient. Pre-implantation genetic diagnosis has a very high barrier for entry in Canada and only affects a very small proportion of live births. In Canada, approximately 12% of the population identifies with a disability and only about 3% of disabilities are solved through tests like PGD. Currently, fewer than 20 children have been born following PGD in Canada but over 1000 have been in the world (Health Canada, 2005). Under Canadian law, embryos that are not selected can be destroyed or donated to stem cell research, provided that the donors have given their free and informed consent and that no part of the embryo or the embryo as a whole was obtained through commercial transactions (CIHR, 2005). There are three main applications for PGD, all of which are controversial for various reasons. The first application involves using the procedure to diagnose genetic conditions or chromosomal abnormalities through two different types of tests. The first detects 38 It is important to note that although PGD must occur in conjunction with IVF, individuals pursuing IVF treatment are not required to undergo PGD. 39 It is also possible to select FOR specific conditions such as deafness (Robertson, 2003). 55 irregularities in specific genes or DNA segments to avoid conditions such as cystic fibrosis or Tay-Sachs syndrome. The second test detects irregularities in the structure or number of chromosomes in order to avoid conditions such as Down Syndrome. The condition of having too many or too few chromosomes is known as aneuploidy. The test for aneuploidy is also called genetic screening because one is not testing for any specific marker. Some physicians hope that using PGD to detect aneuploidy in an embryo during IVF will also help to avoid first trimester pregnancy losses in women who may have otherwise experienced miscarriages due to chromosomal abnormalities or problems associated with age related infertility (Chen, 2005). The second application for PGD is to determine the sex of an embryo. Sex selection can be used to avoid male embryos and instead, select female embryos that are less likely to be affected by sex linked disorders such as Duchenne muscular dystrophy or hemophilia. Individuals may also choose the sex of their embryo for non-medical reasons including "family balancing". However, sex selection for this purpose is banned in Canada under Section 5 (1) (e) of the Assisted Human Reproduction Act (AHR Act) (Department of Justice Canada, 2004). The final application for PGD is tissue typing or human leukocyte antigens (HLA) matching of tissue. During this process, PGD is used to select a saviour sibling or an embryo that is a tissue match for an existing sick child. The chance of rejecting donated tissue (i.e., bone marrow, umbilical cord blood) is significantly reduced if the donor's antigens are similar to those of the older sibling (Devolder, 2005; and Wilkinson, 2004). 56 The expert literature review: clarifying risks and benefits for systemic hazards The Mental Models Approach recommends seeking out a diverse range of experts when characterizing the known risks and benefits of a hazard. However, identifying relevant experts for the topic of PGD is not a straightforward task. Conventional approaches to stakeholder mapping are typically used to characterize stakeholders in order to predict their ability to influence the daily functions and key objectives of an organization, thereby addressing Freeman's (1994) "Principle of Who or What Really Counts." Although legitimacy is a key concern in such processes, some stakeholders groups that have morally legitimate claims may be inadvertently overlooked. Stakeholder mapping approaches are extremely effective in helping businesses to identify the most disruptive groups by mapping stakeholders according to criteria like power and interest (Gardner et al., 1986) or power to influence, legitimacy of the relationship, and urgency of the claim (Mitchell, Agle, and Wood, 1997). Determining "power positions" through the "systematic classification of entities of interest" (Sharma, 2003) help organizations to manage stakeholders (Newcombe, 2003). However, many unconventional stakeholders of PGD technology (e.g., people with disabilities, caretakers of children with disabilities) are neither powerful nor influential. In addition, their small numbers and limited influence is likely to only decrease if PGD becomes socially acceptable and used with increased frequency. Although PGD is a medical procedure performed by physicians, a thorough literature review of this technology revealed that the risks and benefits of PGD could be experienced at multiple scales, or levels of impact, by families, groups, and society. In other words, the risks and benefits of PGD stretch beyond the potential for PGD technology to achieve successful outcomes and the physical risks or benefits that 57 this procedure may pose to the women who use it and the children who are born following it. To identify or map stakeholders and experts groups that include both the relatively powerful (e.g., physicians who perform PGD and control access to it) and the relatively powerless (e.g., caretakers of children with screenable conditions), it is therefore necessary to employ different means of characterizing expertise. Some potential decision objectives for PGD at various scales are shown in Figure 3.1 below. For the case of PGD, constraints set at broader levels (e.g., regulations) are intended to shape behaviour at narrower scales. However, the ideal of autonomous decision making promoted within the Canadian health care system (outlined in the section on risks and benefits at the group and societal scales) and the objectives of special interest groups (e.g., people with disabilities) also constrain the decision making system. It is necessary to find a way to integrate all of these ethically fraught decisions about a single technology into one systematic risk communication strategy. This goal was achieved here by conducting a thorough literature review of documents produced by experts relevant to PGD technology and confirming and supplementing these findings during expert mental model interviews. 58 Figure 3.1 Selected key objectives and the potential influence pathways of decisions at each scale. Families 1. To avoid condition X 2. To select sex of child 3. To create saviour sibling Groups (For example: people with disabilities) 1. To encourage better lives for those affected by disability 2. To reduce discrimination against people with ^ * Society 1. To foster overall health & wellbeing of Canadians 2. To roster overall diversity of Canadian society 3. To treat all Canadians equally 4. To foster individual autonomy and self development The following section discusses sources for the information in Figure 3.2. It clarifies that risks and benefits posed by PGD technology at the family scale, and the group and societal scales. The information presented here is based on a thorough literature review of position statements and information produced by the National Society of Genetic Counsellors and the American Board of Genetic Counseling, academic papers written within the last five years and included on PubMed, and policy documents regarding PGD. 59 Clarifying risks and benefits at the family scale The literature review began by identifying the risks and benefits of PGD at the family scale or the risks and benefits that the technology may pose to a woman's body and to the embryo, fetus, and child. Many of the issues discussed in this section would be similar to those addressed by a genetic counsellor40 and can be characterized as primarily physical, technical, or psychological. Physical risks associated with PGD are often related to IVF, which, as previously mentioned, must be conducted in conjunction with PGD. For example, hormone injections that stimulate ovulation and the egg retrieval process during IVF are both physically invasive and will pose some risk to a woman's body. Other short- term risks that families must consider include Ovarian Hyperstimulation Syndrome (OHSS) and infections. There are also benefits associated with PGD and IVF (other than producing a biological child). For example, IVF lowers the risk of ectopic pregnancies or pregnancies that occur outside the uterus, which can endanger a woman life, cause severe bleeding and pain, and can lead to the death of a fetus. That said, the background risk of ectopic pregnancy is small (less than 2% in Canada).41 One significant physical risk associated with IVF, and therefore PGD, is multiple births, which is linked to the age of mother, the number of embryos implanted, and the quality of those embryos. Risks that multiple births pose to women include premature labour, difficult deliveries, physical and emotional stress. Risks to fetuses and children who are twins, triplets, etc. include increased risk of prematurity, neonatal mortality, SIDS, birth 40 All those pursuing PGD must (1) be diagnosed with or be a carrier for a serious genetic condition and (2) consult with a genetic counsellor. 41 Statements derived from expert interview conducted during phase 1 of this project. Statistics Canada, Health Statistics Division reports that approximately 1.2% of pregnancies in Canada were ectopic in nature in 1994. For more on ectopic pregnancy rates in Canada please see Public Health Agency of Canada, 2002. 60 defects, developmental disabilities, and child abuse (Multiple Births Canada, 1998). Conducting single embryo transfers (often referred to as SET) during IVF for best candidates42 helps to eliminate this risk but also reduces the chance of becoming pregnant because fewer embryos are implanted overall. It is also important to mention that PGD cannot guarantee a "normal" embryo. There are a series of technical challenges inherent to PGD that "can lead to a misdiagnosis of the embryo" and result in an abnormal fetus and child (Hudson, 2004). Testing occurs under strict time constraints and in addition, only small amounts of DNA are examined. Also, there is the potential for the child to carry or develop conditions that were not screened out through PGD. Still other risks and benefits associated with PGD are psychological in nature. The psychological benefits of avoiding a dreaded condition must be weighed against a number of additional factors. In some cases, the biological father and not the woman who undergoes PGD will actually be the carrier of the genetic disorder being avoided. Perceived or actual resentment and guilt associated with this realty could increase stress levels experienced by family members. In addition, families may be asked to submit to additional tests throughout a pregnancy after PGD has occurred. Tests such as amniocentesis or CVS are often used to detect misdiagnosis (stemming from some of the technical challenges discussed earlier) and health of the fetus over time. A recent study determined that 96% of surveyed American IVF clinics either recommended or required 42 Younger woman under 35 who have good quality embryos among other things are typically defined as the best candidates for SET (Kissin et al., 2005). 61 such additional tests (Baruch et al., 2008). Continued or repeated testing may induce a sense of prolonged uncertainty and increase stress levels for family members. Clarifying risks and benefits at the group and societal scales Most would likely agree that there can be great social and biological pressure to reproduce. Yet families may suffer with infertility, have difficulty conceiving, or experience miscarriages and infant deaths (Franklin and Roberts, 2006). Those who are fertile may have an inheritable genetic condition that makes it almost impossible to have their own biological healthy children. Families in this situation may feel pressure to pursue technological solutions such as PGD to solve their fertility dilemmas and may fail to consider other alternatives such as adoption.43 Some view entitlement over reproduction as an extension of one's entitlement to control his or her body. In addition, it may be asserted that parents should be entitled to a healthy child through whatever means possible (Leader, 1999). These views likely stem (in part) from the emphasis that Canadians and our Canadian health care system places on the bioethical principle of autonomy.44 As explained by Sherwin (2000), patient autonomy is a "central value within virtually all health care ethics feminist or otherwise." But is adoption really a practical alternative for Canadians? Public adoption, or adoption through a government agency, for an infant with no "special needs" may take up to 8 years. Private adoption may take less time but can cost anywhere from $10,000-$ 18,000. Costs for international adoption may also take less time but are also costly varying from $15,000 in Haiti, to $30,000 in Guatemala, and as high as $50,000 in the US (Canada Adopts, 2001). 44 Autonomy is defined as the ability to govern one's own actions without interference from others (Beachamp and Childress, 2001). 62 The prevailing norm within the health care system is to view competent patients as autonomous individuals who purchase health care services and actively participate in (or even dictate) the management of their own health care. This concept of the autonomous decision maker can be found in the National Society of Genetic Counsellor's (NSGC) Code of Ethics and Position Statements and literature produced by the American Board of Genetic Counselling (ABGC) as well as in the ethical codes of many other health practitioners (i.e., Canadian Nurses Association Code of Ethics for Registered Nurses (Canadian Nurses Association, 2002), Canadian Medical Association Code of Ethics (Canadian Medical Association, 2004)), and policies and legislation of various Canadian provinces (i.e., The Ontario Consent to Treatment Act, the Substitute Decisions Act, and the Advocacy Act, The BC Health Care (Consent) and Care Facility (Admission) Act, the Yukon Territory Care Consent Act, The Quebec Civil Code and the Act Respecting Health Services and Social Services (Manitoba Law reform Commission, 2004)). While respecting an individual's autonomy is desirable in a health care setting, we must also consider how actions at the family scale may potentially harm citizens at broader levels by thinking through other generally accepted ethical principles such as distributive justice.45 PGD technology may affect groups and society as well as individual families. For example, decisions made about this technology at the family scale may have an effect on groups such as all those who chose to give birth and people with disabilities. At the present time, PGD is rarely used in Canada; but suppose it were to become a standard part of prenatal care. When new technologies that screen and perform diagnostic tests on 45 Distributive justice can be understood as the fair distribution of risks and benefits across all groups (Beachamp and Childress, 2001). 63 fetuses are made widely available and used with increased frequency, these procedures may become standard practice and therefore difficult for all families to refuse (Shakespeare, 2005). Couples may not understand that they have the option of refusing standardized tests or screens and the range of potential consequences associated with these procedures may be masked.46 A good example of this phenomenon is ultrasound technology. Ultrasound is now standard practice for pregnancies in Canada and is typically viewed as an opportunity to see and bond with the fetus (Cook, 2003; Mitchell, 2001; Rothman, 1989; and Taylor, 1998) rather than as a medical screening procedure to detect abnormalities. Williams et al, 2005 report that many women are not prepared for negative results associated with this routine procedure and do not adequately consider the consequences of potential negative risk estimates in their decision to pursue ultrasound. This research team also found that many individuals falsely believed that negative results could be addressed by medical experts through processes like in vivo surgery (Williams et al., 2005). In reality, additional information cannot necessarily mitigate negative outcomes. Lastly, some women were found to experience long term anxiety from false positive problematic screens (Tymstra, 1986; and Weinans et al., 2000). Another potential outcome of using PGD is that there will be fewer people born who have genetic disabilities. Some may argue that this will result in a net economic benefit for the Canadian health care system. Some are concerned that if PGD is viewed 46 However, others are quick to point out that currently, most people are not eligible for PGD (as it has a high barrier for entry) and additional tests are not an option for those who oppose abortion. In other words, the reason some couples choose PGD in the first place is because it happens before pregnancy occurs. They do not want to use prenatal tests such as CVS or amniocentesis and then abort affected fetuses. 64 favourably in this way, then perhaps Canadians will begin to experience social pressure to submit to such technologies and remain consumers of genetic tests throughout their pregnancies. Pursuing technologies that screen for disability may also influence how we view all people with disabilities. Perhaps PGD will reduce our society's overall compassion for people with disabilities when they are viewed as undesirable. Society may even become less tolerant, or "try less" for adults with disabilities who already exist.47 This outcome not only affects people with disabilities, it helps to shape the social and ethical norms of society. In short, it is in everyone's best interest to create a society that fosters diversity and compassionate behaviour towards all citizens.48 The interviews: characterizing expert mental models This section discusses the process of characterizing expert mental models for PGD. The first section describes the interview process while the second presents the final products of those interviews. Experts interviews The first step in phase one of this project was to complete a thorough literature review regarding the risks and benefits of PGD across domains at multiple scales (family, group, society) (see expert literature review section). These findings were then transformed into a series of charts that recorded major themes and possible outcomes. The charts were 47 Statement derived from expert interview conducted during phase 1 of this project. 48 For more on disability rights critiques of prenatal testing and screening please see Shakespeare, 1998 & Parens and Asch, 2000 65 organized around key themes regarding the technological and social risks and benefits of PGD. There were 6 charts it total, two of which (regulatory considerations and conditions screened for by PGD) were added in accordance with expert advice (see Appendix D to view charts and Appendix B for expert interview questions). The next step was to identify and recruit experts to evaluate these charts. Experts were selected to represent the major themes that emerged from the literature review. Additional experts were also added to this group based on recommendations by two or more participants. This review, which also served as one of the comprehensive exams for my doctoral program, was quite extensive in nature. It covered a vast array of information including position statements and information produced by the National Society of Genetic Counsellors and the American Board of Genetic Counseling, academic papers written within the last five years and included on PubMed, and policy documents regarding PGD. It was also informed by disability studies research and an extensive qualitative ethnographic study of couples who have undergone PGD (Franklin and Roberts, 2006). All who participated in this study are widely acknowledged as experts in this narrow field. Maintaining confidentiality of interview findings was therefore a particularly important and challenging experience. The study topic draws on a diverse range of experts that include some lay stakeholders who have specialized knowledge of PGD (e.g., parents of children with disabilities). Even characterizing risks and benefits at the family scale involved a somewhat diverse group of medical experts who could help families work through the physical, psychological, and technological issues surrounding PGD. Table 3.1 illustrates the wide 66 range of expertise that was represented by the expert group. In total, eight experts were interviewed with a fairly even gender split. TABLE 3.1 Overview of interviewee expertise Experts 1. MrA 2. MrB 3. MsC 4. MsD 5. MsE 6. MsF 7. MsG 8. MrH Ethical, economic, legal. & social dimensions V cr Q- -_ ID o o o •S Q- a. (D T3 c r 01 §. 3 g: o O CD O § O > § JE. S ^ zi w 3 8 Eugenic applications/fall into the w rong hands or used for bad Don't know/No opinion purpose • Treatment A (survey 1) • Treatment B (survey 1) In the post communication questionnaire, the most common concern for treatment A participants was still eugenic applications. Some also remained concerned about PGD being used for some general bad purpose or to exploit an unidentified group. However these participants were also now concerned about unnecessary applications for PGD such as sex selection and for other non-fatal conditions, as well as inequitable access to the technology (see Figure 5.7). 136 14 12 10 8 6 4 2 0 Figure 5.7 Survey 2 results for treatment A participants: reported concerns for PGD technology 8 8 I , 1 5 4 Eugenics For non-fatal conditions Inequitable access For bad purpose/to exploit a Treatment A (post) Treatment B participants also maintained eugenic concerns but the leading concern after reviewing communication and SDM materials was now the potential for PGD to lead to segregation and discrimination, unnecessary applications (e.g., screening for non- fatal conditions and sex selection), and inequitable access (see Figure 5.8). They were also concerned about the technology becoming imposed on individuals and the ways in which the PGD may narrow diversity for Canadian society. Figure 5.8 Survey 2 results for treatment B participants: reported concerns for PGD technology 14 12 10 SegregatiorVdiscrimination For non-fatal conditions Inequitable access Eugenics Imposed Narrow diversity a Treatment B(post) 137 Differences also emerged between treatment A and B participants for questions regarding anticipated stakeholder concerns. When asked about stakeholders and PGD technology before receiving risk and benefit information in the first questionnaire, participants from both treatments most commonly reported that stakeholders would likely be primarily concerned about PGD because of religious convictions, general ethical beliefs, because they perceived PGD as "playing God", etc (see Table 5.3). These were still the most commonly reported stakeholder concerns after receiving risk and benefit information for treatment A participants (those who did NOT review SDM materials). However, in the second survey, or after reviewing risk and benefit information and SDM materials, the most commonly reported anticipated stakeholder concerns for treatment B participants were that PGD may ostracize some, create class differences, and discriminate against people with disabilities or infringe on their rights. Table 5.3 Reported stakeholder concerns: most common response by treatment Survey 1: most popular response for treatment A participants Because it goes against how they want others to live /goes against their convictions/playing god /murdering embryo (n=7) Survey 2: most popular response for treatment A participants Ethical /religious reasons/ eugenics/playing god/different values (n=16) Survey 1: most popular response for treatment B participants Erosion of faith in god /ethics/religious reasons/playing god (n=12) Survey 2: most popular response for treatment B participants Ostracization /creating class differences/discrimination against people with disabilities/infringe on their rights (n=12) 138 Participant opinions regarding PGD technology Most participants in both treatments reported that they had no opinion regarding PGD in the first questionnaire (see Figure 5.9). This is unsurprising given that most were unfamiliar with this technology before the workshop. Nearly all (93%) participants stated in the first questionnaire that they had never heard of the term pre-implantation genetic diagnosis or PGD. Figure 5.9 Survey 1 results: Opinions of PGD technology (shown in percentages) 16 14 12 10 8 6 4 2 0 9 , - = 5 • r J • | 0 0 14 14 I Strongly supportive Somewhat Somewhat opposed Strongly opposed supportive • Treatment A (pre) • Treatment B (pre) No opinion Don't care about PGD Overall opinions did not vary significantly by treatment by the end of the workshop (see Figure 5.10). In the second questionnaire, only two participants reported that they still had no opinion of PGD and most reported that they were now either somewhat or strongly supportive of this technology. Statistical significance was determined using the McNemar test for the significance of changes at a 95% confidence interval (Sirkin, 1995). This formula measures the number of individuals who respond differently after being exposed to new information or a new condition (Daniel, 1990). Figure 5.10 Survey 2 results: Opinions of PGD technology 139 Strongly supportive Somew hat Somew hat opposed Strongly opposed supportive • Treatment A (pre) • Treatment B(pre) No opinion Don't care about PGD Rating the workshop Differences between treatments emerged when participants were asked to rate their favourite part of the workshop (see Figure 5.11). Participants from each treatment favoured the questionnaires, general background information presented in FAQ's, and information on the technical risks and benefits to women, embryos, fetuses, and children equally. However, participants from treatment B (those who DID review SDM materials) were somewhat more likely to prefer the private voting process and the group discussion and significantly more likely to prefer strategies that focussed on the social risks and benefits of PGD and the conditions screened for by PGD. 140 Figure 5.11 Survey 2 results: Favourite part of the workshop IR - 14. - 13 - 10 - ft - fi - 9 . 0 - , 5 4 ^ ^ ^ Q m- uestionnaires 8 4 • II - 1 Private voting 1 2 11 _ 1 1 — 1= • • 7 | z . 1 H •^ FAQ presentation Info on technical risks/benefits to women/E,F,&C I 1 Info on social R/B 14 8 H H t Info on conditions screened for by PGD 15 11 Group discussior • Treatment A (post) • Treatment B (post) When asked to identify their least favourite part of the workshop most participants in both treatments stated that the information presented on the technical risks and benefits of PGD to women, embryos, fetuses, and children was their least favourite part (see Figure 5.12). Treatment B participants were also significantly more likely to pick the O f questionnaires as their least favourite activity. Figure 5.12 Survey 2 workshop ratings: Least favourite part of the workshop Questionnaires Private voting FAQ presentation Info on technical Info on social R/B Info on conditions Group discussion risks/benefits to screened for by women/E,F,&C PGD I Treatment A (post) • Treatment B (post) It is possible that treatment B participants rated the questionnaires as their least favorite part of the workshop because they preferred the group discussion portion. They might have believed that if the workbook exercises were reduced, they would have been able to engage in a longer discussion. These were also the participants who viewed the SDM materials so therefore had to wait much longer before they were allowed to discuss the issues with other participants (when compared to treatment A participants). 141 Participants in both treatments reported that the most important factor informing their private vote, or their recommendations for risk communication strategies, was the general background information about PGD addressed in the FAQ power point presentation (see Figure 5.13). Nineteen participants from treatment B and 16 from treatment A stated that this information source influenced their decision the most. The second most important source of information for participants was the news articles that described aspects of PGD from the viewpoint of those personally affected by disability. Responses concerning the least important factors influencing participants' final vote also showed very few significant differences between treatments. Figure 5.13 Survey 2workshop ratings: Most important factor influencing recommendations FAQ- Audio-impacts Chart-impacts Audio-impacts Chart-impacts Audio-social Charts-social Videos- News articles- Discussion Opinions Other background women women embryo, fetus, embryo, fetus, impacts impacts conditions conditions groups formed before child child event B Treatment A • Treatment B Finally, participants were asked how stressful, difficult, enjoyable, worthwhile, and interesting they found the workshop on a four point scale. Figure 5.14 shows those who offered extreme responses of 1 or 4 out of four. Those who did NOT receive SDM materials in treatment A had a higher percentage of participations who rated the 142 workshop as extremely stressful, enjoyable, and worthwhile. Ratings for extreme difficulty were similar for both treatments as were the percentage of those who reported that the workshop was extremely interesting. It is worth noting that extreme percentage ratings for treatment B participants were lower overall for this set of questions. Figure 5.14 Survey 2 workshop ratings: Extreme ratings by treatment (shown in percentages) 100- Qn - an - 70 - 50- 4(1 - ?n . 90 . •tn . n . -IA 64 o l 48 • • • • II Stressful 4 6 41 1 Difficult 25 | 14 mm II 1 I Enjoyable • Treatment A • Treatment B 48 ^m ^ ^ B ^ ^ H ^—M Ĥ Worthw hile ' ' b» 1 Interesting Recommendations from participants After reviewing all risk communication information (and SDM materials for the case of treatment B participants), participants engaged in a facilitated group discussion that asked otr them to provide risk communication recommendations. Participants were asked to consider: who should communicate information about PGD; who should comprise the target audience; preferred risk communication strategies for PGD; and what information should be included in those strategies. After the group discussion on these issues was A brief overview of risk communication practices and objectives was provided before each group discussion. 143 complete, participants were then asked to fill out the second questionnaire, which included a voting section where they could privately write down their recommendations. They were explicitly told that their private vote need not agree with any recommendations reached during the group discussion. Recommendation responses revealed no significant differences between treatments. Participants from both treatments agreed that risks communicators for PGD should be both medical professionals and a federal department such as Health Canada. Group discussions revealed that participants had come to view PGD as a multi-scalar hazard that posed risks and benefits to individuals (e.g., families who want to have children), groups (e.g., those who can afford PGD), and society (e.g., all Canadians). They believed it was therefore appropriate to use different communicators (see Figure 5.15) and strategies (see Figure 5.16) to reach each of these target audiences (see Figure 5.17). Figure 5.15 Recommendations: Who should communicate information about PGD? nr-^ ~2 T Medical Trained Health Academics Unidentified The media Other professional counselor Canada/Gov experts • Treatment A • Treatment B 144 The most popular proposed risk communication methods for all participants were a face to face discussion with a medical expert and a website, which correlates well with the recommended risk communicators. Pamphlets were also a popular choice. The telephone survey was the only method to receive no votes by participants from either treatment. Figure 5.16 Recommendations: Proposed risk communication strategies for PGD Public service Pamphlet Website F2F discussion Phone survey Interviews or announcement w ith medical discussion expert groups I Treatment A (post) • Treatment B (post) Other Recommended target audiences were also similar for participants in both treatments. The most popular choice was all interested Canadians although treatment B participants were slightly more likely to choose this audience than their treatment A counterparts. The second most popular choice for all participants was married couples who want to have children followed by people considering IVF or PGD and who can afford it. 145 Figure 5.17 Recommendations: Who should comprise the target audience for risk communion strategies about PGD? ?n - 1 * 5 - 10 5 n - 19 *n *n - • • M ar rie d co up les /co up les w ho w an t ch ild re n 14 I ~ ~ 1 10 9 1 • 5 V, CO CD § & T3 .£ CO « < 5 1 Pe op le co ns ide rin g I VF or PG D/ tho se w ho ca n af fo rd it 6 o — AB Th os e w ho w an t to ad dr es s a pr ob lem 5 1 Te en s/a ll s ch oo l ag ed ch ild re n • Treatment A • Treatment B •> 2 0 1 1 c CD E 1 < i 2 0 1 1 M ed ica l st ud en ts/ ex pe rts 1 0 M os t C an ad ian s Participants from both treatments also agreed on the types of information that should comprise risk communication strategies for PGD (see Figure 5.18). They agreed that just about all information types presented during the workshop should be included although a significant number of treatment B participants reported that information about PGD regulations were more important than the others. In addition, only a few treatment B participants believed that the consequence tables (SDM materials) were important to include.87 87 Participants were asked to consider difficult tradeoff during the workshop. They understood that they were supposed to recommend key items (e.g., components, information types) that ought to be included in all risk communication strategies about PGD. They were aware that including certain items meant that others would have to be left out as it is unrealistic to retain the large amount of detail included in the workshop. It is the author's assumption that participants left the matrices out because they likely viewed the matrices (which took up to 20 minutes to discuss) as a simple way of reducing the amount of information for a future strategy. Also the value of the matrices is not obvious to those involved in the experiment. 146 Figure 5.18 Recommendations: What types of information should comprise risk communion strategies for PGD? T »-2S 27 2f—= „ - • | y _ ?i-23 23_23 IBIBli General info Regs Conditions R/B to women R/BtoE,F, &C Social R/B Decision Other matrixes • Treatment A • Treatment B Voting confidence levels were also very similar for participants of each treatment. In total, 86% reported that they were either somewhat (61%) or very confident (25%) in their vote. Evaluating the decision making process This section employs the 2008 Academy of Sciences criteria to evaluate the quality of decision processes for participants of the deliberative risk communication workshops about PGD. The focus is specifically on process related quality or the effect (if any) that SDM applications had on how participants managed complex information about a novel, controversial topic. The first key finding to introduce is that differences between the two treatments could be easily overlooked if only major workshop outcomes were examined. For example, participant final recommendations for risk communication strategies about PGD revealed very few differences between treatments with most agreeing on who should communicate this information, preferred risk communication strategies and the ^ 147 types of information that ought to be included in those strategies, and preferred target audiences. Overall, opinions of PGD did not vary significantly by treatment by the end of the workshop and confidence levels in participant generated recommendations were also very high and consistent across treatments. Differences emerged only when examining the hopes and concerns and perspectives that informed key workshop outcomes. Quality of decisions Information was added and considered in the process At the beginning of the workshop in the first questionnaire, almost one third of participants (30%) failed to identify any particular hopes or concerns for PGD technology. However, treatment B participants reported on a more complex and wider network of hopes and concerns by the second questionnaire. While shared post questionnaire responses from treatment A participants centered mostly on disease prevention and decision making, shared responses from treatment B participants also covered topics such as public education and funding arrangements. Eugenic applications were one of the leading concerns for treatment A participants both before and after reviewing communication materials. However, over the course of the workshop, the leading concern for those who had viewed SDM materials (treatment B) became the potential for PGD to lead to segregation and discrimination. Common concerns for treatment B participants also included additional topics such as PGD technology becoming imposed on individuals and the ways in which the technology may narrow diversity for Canadian society. 148 Outputs reflected a broad view of the issues that participants believed were important Participants agreed that the issues presented to them throughout the workshop were relevant and thus important to include in any future risk communication strategies with the exception of the SDM materials. Suggestions for additional information to include in forthcoming risk communication strategies mostly called for supplementary details on topics already covered during the workshop as opposed to new items (e.g., information on other conditions screened for by PGD, more stories from the perspective of people with disabilities). Most agreed that it was sensible for risk communicators to rely on a collaborative process like this one when determining risk communication strategies for PGD.88 Innovative ideas were generated for solving problems Participants were given the opportunity to describe their vision of a risk communication strategy for PGD in the voting section of the second questionnaire. Many innovative and extremely detailed strategies were proposed and a few even included intricate diagrams and figures. Some of the more notable suggestions included a 1-800 hotline number for risk and benefit information about PGD; a documentary film about PGD produced for CBC television and schools; videos to be aired on YouTube; webcasts; referendums on PGD policy; and chat rooms for those going through the PGD process. In total, 46 participants agreed that it was sensible, 7 said they did not know, and 3 reported that it was not. 149 Capacity for future decision making Participants became better informed about relevant issues Mental model interviews with lay citizens that took place prior to the workshops revealed that participants had very little background knowledge of PGD and had a difficult time anticipating particular stakeholders of this technology. All participants were successful in learning a great deal about PGD during the workshop (see Chapter 4 of this thesis). However, participants who reviewed SDM materials (treatment B) had an easier time identifying with the concerns of people who may be discriminated against through this technology (e.g., PGD may ostracize some, create class differences, discriminate against people with disabilities, or infringe on their rights) in addition to other stakeholders. People with disabilities are often identified as a major stakeholder group for PGD technology but were overlooked by most mental model interviewees during phase 2 of this study. Interestingly, those who reviewed SDM materials were more likely to prefer information on the social risks and benefits of PGD and on the conditions screened for by PGD when compared to their treatment A counterparts. Perhaps treatment B participants favoured this information because they were able to identify with a wider range of stakeholder groups. Scientists gained an understanding of public concerns The workshops described here have no official mandate. However the recommendations produced by participants could be easily communicated to scientists, policy analysts, and regulators working in this area. The final step in this project is to share anonymized 150 findings on a password protected website accessible to all experts and lay citizens who participated in various study phases. Participants gained skills Two of the new skills gained by all workshop participants were learning how to participate in a deliberative group discussion and design risk communication strategies. Most reported that the workshop had been a valuable learning experience. A significant majority (84%) also stated that they would be "very" or "definitely likely" to attend a workshop like this again in the future. Man y treatment B participants were able to recognize complex tradeoffs by reviewing SDM materials. Learning about alternatives through the lens of value objectives at each scale encouraged participants to consider important tradeoffs inherent to each choice. Treatment B participants were therefore better able to articulate tradeoffs that occur between scales for PGD. For example, they understood that pursuing this technological solution could lead to great perceived benefits for families and children while potentially discriminating against people with disabilities and reducing overall diversity for society. Many ultimately viewed this as a justified tradeoff as indicated by the overall support for the technology by the end of the workshop. The ability to identify and cope with tradeoffs is applicable to a range of other problems and is a skill that participants could easily employ during actual decision making opportunities that regularly emerge in their daily lives. 151 Conclusions Findings from the risk communication workshops described here contribute empirical evidence to the body of knowledge surrounding public involvement efforts. The results demonstrate that lay participants, especially those who review even the most minimal SDM materials, are able to make well informed, robust, inclusive, and therefore meaningful decisions about a complicated novel technology. Workshop findings also fared well against the National Academy of Sciences, 2008 criteria that judged the quality of the workshops and associated outputs. It was found that SDM materials fostered holistic or "all things considered" judgements (McDonald, 2000). Treatment B participants were more likely to understand and articulate the range of stakeholder hopes and concerns for PGD technology (research question #1). The materials also helped these participants conceptualize complex financial and ethical tradeoffs between families, groups, and society (research question #2). Pre-implantation genetic diagnosis is a technology that is caught up in the culture wars over abortion. Results indicate that participants who received SDM materials are more likely to acknowledge, but get past, the most obvious and politically charged ethical concerns (e.g., PGD is playing God, concerns about killing human embryos, PGD is a slippery slope to eugenics) in order to analyze murky long-term ethical dilemmas that could be easily overlooked without careful analysis as illustrated in the quotation shown below (research question #3). It's okay that science discover[s] more and more things. But I hope that we are all prepared to use these findings for the benefit of all human kind, not for exclusion of some people or [for] destruction. -Amy 152 In closing, it seems that including SDM materials in risk communication workshop are in fact worth the extra time they consume and effort from participants (research question #4). The counterfactual experiment described here demonstrates that SDM materials helped participants to make better, more holistic, and ultimately more ethical decisions. In addition, participants did not feel overwhelmed by the inclusion of additional SDM materials. On the contrary, those who did NOT receive SDM materials in treatment A were more likely to rate the workshop as extremely stressful even though participant ratings for extreme difficulty were similar for both treatments. 153 Works cited Arvai, J., Gregory, R, and McDaniels, T. (2001).Testing a Structured Decision Approach: Value Focused Thinking for Deliberative Risk Communication. Risk Analysis, 21:6, 1065-1076. Beierle, Thomas C , and Jerry Cayford. (2002). Democracy in Practice: Public Participation in Environmental Decisions. Resources for the Future Bohman, J F. (1998) The coming of age of deliberative democracy. Journal of Political Philosophy 6 (4), pg. 399-423. *• Bohneblust, H., & Slovic, P. (1998). Integrating technical analysis and public values in risk-based decision making. Reliability Engineering and System Safety, 59,151-159. Burgess, MM. (2004). Public consultation in ethics: an experiment in representative ethics. Journal of Bioethical Inquiry. 1.1, pg 4 -13. Clemen, Robert and Reilly, Terence (2000). Making Hard Decisions: An Introduction to Decision Analysis. Duxbury Resource Center Daniel, Wayne .W. (1990) Applied Nonparametric Statistics, 2nd Edition. PWS-Kent Publishing Company, Boston, Massachusetts. Dietrich, Heather and Schibeci, Renato. (2003). Beyond Public Perceptions of Gene Technology: Community Participation in Public Policy in Australia. Public Understanding of Science. 12. pg 381 - 4 0 1 . Dryzek, J. (1990) Discursive Democracy: Politics, Policy and Political Science. Cambridge: Cambridge University Press Fukuyama, F., and F. Furger (2007) Beyond Bioethics: A Proposal for Modernizing the Regulation of Human Biotechnologies. Paul H. Nitze School of Advanced International Studies. Gregory, R, Lichtenstein, S., & Slovic, P. (1993). Valuing environmental resources: A constructive approach. Journal of Risk and Uncertainty, 7, 177-197. Gregory, R., & Keeney, R. (1994). Creating policy alternatives using stakeholder values. Management Science, 40, 1035-1048. Gregory, R., McDaniels, T., Fields, D. (2001) "Decision aiding, Not Dispute Resolution: A New Perspective for Environmental Negotiation". Journal of Policy Analysis and Management 20 (3), pg. 415-432. • 154 Gutmann, Amy, and Thompson, Dennis. (1996). Democracy and Disagreement. Cambridge: Belknap Press. Hammond JS, Keeney RL, Raiffa H. (1999). Smart choices. A practical guide to making better decisions. Harvard Business School Press, Boston, MA. Juanillo NK, Scherer CW (1995). Attaining a State of Informed Judgments: Toward a Dialectical Discourse on Risk in Communication Yearbook. Vol 18, Burleson B (ed) pp 278-299. Thousand Oaks, CA, USA: Sage Publications. Keeney, Ralph. (1982). Decision analysis: an overview. Operations Research. 30, 5, p 803-838. Keeney, Ralph L. (1992). Value focused thinking: a path to creative decision making. Harvard University Press, pg 16-22. Keeney, Ralph, Detlof von Winterfeldt and Thomas Eppel. (1990). Eliciting Public Values for Complex Policy Decisions. Management Science. 36, 9, pp. 1011-1030 Longstaff, Holly and Burgess, Michael. (Forthcoming). Recruiting for representation in public deliberation on the ethics of biobanks. Public Understanding of Science. Longstaff, Holly (Under review). Learning about systemic hazards through holistic risk communication Longstaff, Holly. (Under review) Characterizing expert mental models for systemic hazards: the case of pre-implantation genetic diagnosis. McComas, K. (2004). Strategic risk communication for new and emerging risks: When even the 'best-laid' plans go wrong. EMBO Reports, 5 (Special Issue), S61-S65 McDaniels, Timothy L. (1996). The Structured Value Referendum: Eliciting Preferences for Environmental Policy Alternatives. Journal of Policy Analysis and Management. 15, 2, pp. 227-251. McDaniels, T., Gregory, R. (2004) Learning as an objective within structured decision processes for managing environmental risks. Environmental Science and Technology 38(7), pg. 1921-1926 McDaniels, T, (2000) Creating and using objectives for ecological risk assessment and management. Environmental Science and Policy 3, pg. 299-304 McDonald, Michael. (2000). Biotechnology, Ethics, and Government A Synthesis. Prepared for The Canadian Biotechnology Advisory Committee Project Steering Committee On Incorporating Social and Ethical Considerations into Biotechnology. 155 National Research Council. (1996). Paul C. Stern and Harvey V. Fineberg, Editors; Committee on Risk Characterization. Understanding Risk: Informing Decisions in a Democratic Society National Academy of Sciences. (2008). Thomas Dietz and Paul C. Stern, (Editors) Panel on Public Participation in Environmental Assessment and Decision Making, National Research Council. Public Participation in Environmental Assessment and Decision Making. O'Doherty, K and Burgess, M. (2008) Engaging the Public on Biobanks: Outcomes of the BC Biobank Deliberation. Public health genomics Payne, J. W., Bettman, J. R., & Johnson, E. J. (1992). Behavioral decision research: A constructive processing perspective. Annual Review of Psychology. 43, pp 87-132. Rowe, G and Frewer, Lynn. (2000). Public Participation Methods: A Framework for Evaluation. Science, Technology & Human Values, Vol. 25, No. 1, 3-29. Rowe, G. and Frewer, L.J. (2005). A Typology of Public Engagement Mechanisms. Science, Technology, & Human Values. 30 (2), pg. 251-90. Sirkin, R Mark. (1995). Statistics for Social Sciences. Sage Publications Inc. Thousand Oaks, California Slovic, P. (1995). The construction of preference. American Psychologist. 50, pp 364- 371. Sunstein, C. (2002). The Law of Group Polarization. Journal of Political Philosophy 10(2): 175-95. Sunstein, Cass (2002). Designing Democracy, What Constitutions Do. Oxford: Oxford University Press. Walmsley, Heather (2009) Mad scientists bend the 'frame' of biobanking governance in BC - how ethnographic attention to stories can help deliberative events 'listen'. Journal of Public Deliberation. Vol. 5: No.l, Article 6. Zeckhauser, R., and Viscusi, K.W. (1996). The risk management dilemma. In H. Kunreuther and P. Slovic (Eds.), Annals of the American Academy of Political and Social Science, Special Issue, Challenges in Risk Assessment and Risk Management (pp. 144-155). Thousand Oaks: Sage. 156 Chapter six: Overview, insights, and future research directions Summary of chapter hypotheses and key insights The fundamental objective of this thesis was to contribute to the interdisciplinary body of research that seeks to integrate the fields of ethics and risk analysis. It was shown how guidance from bioethics can fortify health risk communication strategies and ultimately foster holistic decision making for a virtually unknown systemic hazard. The chapters included in this thesis have attempted to explain the unique contributions of this study in terms of the topic addressed and the methods used to analyze it. The first introductory chapter explained why systemic health hazards like pre-implantation genetic diagnosis (PGD) are an important area of study for the field of risk communication and how novel methods can be used to elicit risk communication recommendations from lay citizens. Some have argued that democratizing science policy though deliberative efforts may be used as an excuse to persuade citizens or "dilute accountability" of policy makers (Lidskog, 2008). This objection seems particularly persuasive when public engagement efforts fail to show evidence of meaningful outcomes. Research presented here demonstrates that lay citizens can in fact produce thoughtful and well informed recommendations about extremely complex and unfamiliar health topics in line with their own values. The chapters that comprise this thesis introduce novel ways of approaching systemic hazards that include intellectual contributions from the field of bioethics and experiments with deliberative risk communication. Table 6.1 shows the fundamental objectives for each chapter. 157 Table 6.1 Fundamental objectives by thesis chapter Chapter Fundamental objectives Two: Blending bioethics concepts with risk communication theory and applications Three: Characterizing expert mental models for systemic hazards: the case of pre- implantation genetic diagnosis Four: Learning about systemic hazards through a risk communication experiment: the case of pre-implantation genetic diagnosis Five: Using structured decision making techniques to help communicate information about controversial technologies: the case of pre-impanation genetic diagnosis • to demonstrate how bioethics concepts can help fortify risk communication strategies. • to share a novel method of eliciting expert judgment • to gather empirical evidence that shows lay citizens can in fact learn a great deal about a range of value-oriented and technically-oriented risk/benefit messages • to evaluate the quality of lay participant decision processes associated with a series of deliberative risk communication experimental workshops Chapter two outlined the various ways in which bioethics theory can be used to fortify risk communication strategies for systemic health hazards. It showed how considering substantive and public ethics can help risk communicators identify the moral values that should inform their efforts; build an ethical communication strategy; choose experts that will ultimately inform that strategy as well as the target audience; and evaluate findings. Emerging systemic hazards require more than new theoretical frameworks, they also call for new approaches. The main objective of chapter three was to introduce a novel method of eliciting expert judgment. It demonstrated how the mental models approach (Morgan, 2002), supplemented with bioethics theory discussed in chapter two, can be used to elicit and characterize expert judgments capable of clarifying uncertainties and outcomes for a wide range of heterogeneous stakeholders for PGD technology. 158 Chapter four discussed the results of mental model interviews and risk communication experimental workshops with lay citizens about PGD. It showed that although most participants were initially unfamiliar with this technology, they could overcome many shared misconceptions and learn a range of technically -oriented and value-oriented risk/benefit messages. Furthermore, these messages did not undermine their own views. Unlike other communication efforts (e.g., propaganda) that attempt to actively manipulate thinking, the messages presented in this study enhanced opinions of PGD thereby respecting the autonomy of participants. For example, although participants became more concerned about the risks that PGD may pose to people with disabilities (or other groups), many anticipated risks of this technology were maintained throughout the workshop (e.g., eugenic concerns and the potential for designer babies). Reported benefits were also more complex by the end of the risk communication workshop although many participants maintained the view that an overall healthier population was a significant benefit of PGD technology. Chapter five examined the quality of workshop participant decision processes according to criteria presented in the National Academy of Sciences, 2008 report on public participation. Experiment results fared well against the 2008 criteria (National Academy of Sciences, 2008). The workshop included a counterfactual experiment with structured decision making (SDM) components in the form of consequence tables. It was determined that the SDM materials helped participants to arrive at more holistic, inclusive, and therefore ethical decisions by helping them to conceptualize the complex financial and ethical tradeoffs between families, groups, and society (McDonald, 2000). 159 In summary, this thesis has produced a number of significant insights for existing theory. First, collaboration between applied bioethics and risk analysis is not simply possible, it is productive. I would even argue that such interdisciplinary work is required to effectively cope with systemic hazards in the health domain. There are many interesting overlaps between these two fields of study in terms of shared principles and interests. Ethically informed risk communication efforts, like the experiment described here, help lay participants to deal with complicated tradeoffs and ultimately produce meaningful recommendations in line with their own values. This study also confirmed the value of established risk analysis methods such as the mental models approach and SDM techniques and how these tools can be employed to foster ethical decisions. Study strengths and weaknesses The main contribution of this thesis as a collective work is to demonstrate that lay citizens can in fact make holistic and well informed decisions about systemic hazards through deliberative risk communication efforts. Empirical results presented here are generalizable to other kinds of systemic health hazards and can be used to inform standards and methods of risk communication for government bodies and others in the research community. Nevertheless, a perceived weakness of this study could be that the experimental process had no official mandate and therefore no direct influence on public policy making about PGD technology in Canada (as would be the case for federal program evaluations for example). Some may argue that it is both inappropriate and a waste of citizens' valuable time and energy to produce recommendations in a vacuum. However, the study was intended to examine and test a novel approach to deliberative 160 risk communication, not to statistically represent the views of a particular group for policy guidance. The absence of an official mandate is in fact an overall strength for this study. Official public engagement efforts must produce recommendations in a timely fashion using conservative (and dependable) methods and approaches. The fundamental objective of such efforts is to produce a final product, not to theorize or experiment with novel conceptual frameworks. Academic researchers have the freedom and I would also argue, obligation to pursue research that cannot be conducted elsewhere in private industry or by government officials in order to advance their chosen field of study. Other weaknesses of this study stem from typical financial constraints placed on any student project that made recruiting lay participants for extended workshops held at expensive facilities impossible. This project used space graciously donated by The W. Maurice Young Centre for Applied Ethics at the University of British Columbia and was only able to provide small honorariums (10$) for lay participants who could therefore not be expected to donate large amounts of time to this study. Workshop participants had a maximum of one hour to consider the complex issues surrounding PGD during the large group discussion period. Additional deliberation time would likely lead to more substantial and robust recommendations. In the future, with a grant funded research project, it would be preferable to allow participants more discussion time and perhaps even spread the event over an entire weekend (or two) to allow for meaningful reflection and adequate rest periods. Recent deliberative democracy experiments conducted in the topic areas of biobanking and salmon genomics at The W. Maurice Young Centre for Applied Ethics have demonstrated the importance of such considerations (Longstaff and Burgess, forthcoming). It would also be appropriate to experiment with recruitment filters 161 in order to test the risk communication experiment with different types of lay citizen samples (e.g., family members of people with disabilities, couples, those who have used reproductive technologies). Future research directions Communicating the risks and benefits of systemic hazards is a new challenge for risk analysts. However, notable progress has already been made by researchers including Sherwin, Slovic, Renn, and McDaniels et al. Sherwin's public ethics probes moral responsibility at multiple scales of impact in relation to systemic hazards (Sherwin, 2008). Her framework examines new oughts posed by systemic hazards through employing existing feminist bioethics concepts such as relational ethics (Burgess and d'Agincourt-Canning, 2001). Slovic examines the ways in which citizens cope with or make sense of new systemic topics such as genocide. His recent work explains some of the cognitive failures associated, for example, with processing "losses of life as they become larger" (Slovic, 2007). He argues, like Sherwin, that dealing effectively with systemic hazards requires a multi-system, cross-scale approach that includes a commitment to morally responsible actions amongst citizens and institutions but also system changes to international regulatory frameworks, among other things. Both Renn and McDaniels et al propose new risk management frameworks capable of incorporating the value dimensions of systemic hazards into the risk analysis process. Renn's concern assessment complements the conventional risk assessment model produced by technical experts while McDaniels et al introduce decision making frameworks guided by shared 162 fundamental values of relevant stakeholders (Renn, 2006; McDaniels et al, 2005 & 2006). Future research studies should continue to experiment with deliberative risk communication experiments that incorporate bioethics theory in the human health domain in order to confirm some of the findings presented here. Additional research is required to determine best practices for communicating value-oriented and technically-oriented risk/benefit information that are effective and trustworthy. Future studies should also test the methods used in this study as well as novel approaches (Cox et al, 2009) with different audiences and experiment with more complex systemic topics such as global climate change, stem cell research, and global poverty (UNICEF, 2008). The findings presented here demonstrate that lay citizens can come to understand and make meaningful recommendations about extremely complicated, unknown technologies like PGD. There is therefore good reason to believe that lay citizens could provide meaningful input for a range of other systemic hazards, which will ultimately help analysts develop more inclusive and ethical risk communication strategies. Finally, it is important for the dialogue to continue between bioethics and risk researchers as a great deal of work remains to be done in this new interdisciplinary field. The natural next step in this relationship would be to explore the contributions that risk analysis can make to the field of bioethics. It is important for those conducting work in highly complex areas, like systemic hazards, to understand a wide range of perspectives including those of conventional experts and lay citizens. Ethics researchers working in the health domain could benefit from learning more about how to engage the lay public 163 as well as how to communicate across disciplinary boundaries, topics that have been researched extensively within the field of risk analysis. Works cited Burgess, Michael and d'Agincourt-Canning, Lori. (2001). Genetic testing for hereditary disease: Attending to relational responsibility. The Journal of Clinical Ethics, 12,4, 361- 372. Cox, Susan, Magdalena Kazubowski-Houston, and Jeff Nisker. (2009). Genetics on stage: Public engagement in health policy development on preimplantation genetic diagnosis. Social Science and Medicine. 68,1472-1480. Lidskog, Rolf. (2008). Scientised citizens and democratised science. Re-assessing the expert-lay divide. Journal of Risk Research. Vol. 11, No. 1-2, pg 69-86. Longstaff, Holly and Burgess, Michael. (Forthcoming). Recruiting for representation in public deliberation on the ethics of biobanks. Public Understanding of Science. McDaniels, Timothy L., Dowlatabadi, Hadi, Stevens, Sara.(2005). Multiple scales and regulatory gaps in environmental change: the case of salmon aquaculture. Global Environmental Change Part A, 15,1. 9-21. McDaniels, Timothy, Longstaff, Holly, and Dowlatabadi, Hadi. (2006). A Value-Based Framework for Risk Management Decisions Involving Multiple Scales: A Salmon Aquaculture Example. Journal of Environmental Science & Policy, 9, 5, 423-438. McDonald, Michael. (2000). Biotechnology, Ethics, and Government A Synthesis. Prepared for The Canadian Biotechnology Advisory Committee Project Steering Committee On Incorporating Social and Ethical Considerations into Biotechnology. Morgan, Granger M., Fischhoff, Baruch, Bostrom, Ann & Atman, Cynthis J.(2002).Risk Communication: A Mental Models Approach. Cambridge University Press. National Academy of Sciences. (2008). Thomas Dietz and Paul C. Stern, (Editors) Panel on Public Participation in Environmental Assessment and Decision Making, National Research Council. Public Participation in Environmental Assessment and Decision Making. Renn, Ortwin (2006). White Paper on Risk Governance: Towards an Integrative Approach. International Risk Governance Council. Retrieved on September 22, 2008 from http://wwvv.irgc.org/1rgc/projects/risk_characterisation/_b/coirtentFiles/TRGC_WP_No ovemance (reprinted_version).pdf#search~%22%2()renn%20risk%20governance%20towards%2 0aD.%20i.ntegrated%20%22 Sherwin, Susan (2008). Whither Bioethics? How feminism can help reorient bioethics. The international journal of feminist approaches to bioethics, 1,1, 7-27. 165 Slovic, Paul. (2007). "If I look at the mass I will never act": Psychic numbing and genocide. Judgment and Decision Making, 2, 2, 79-95. Unicef (2008) Unicef Canada annual report 2007-2008. Retrieved on February 21, 2009 from http://www.unicef.ca/portal/Secure/Community/502AVCM/PRESS/publications/2007_2008_Uni cef_AR_en.pdf 166 Appendices Appendix A. UBC Research Ethics Board's Certificates of Approval Appendix B. Interview questions for experts Appendix C. Mental model interview questions for lay participants Appendix D. Expert charts developed after an extensive literature review and then confirm through interviews. Included are charts that discuss the (1) technical risks of PGD to the embryo, fetus, and child; (2) technical risks of PGD to a woman; (3) social risks of PGD, (4) considerations for a Canadian regulatory framework in flux, and (5) and spectrum of conditions screened for by PGD. Charts 4 and 5 were not initially included. They instead were added in response to expert participant feedback. Appendix E. Example of how expert charts changed over time due to expert participant feedback Appendix F. Risk communication workshop presentation that was used to guide the workshop (this presentation includes the expert based FAQ's and the consequence matrices) Appendix G. Risk communication workshop workbook for lay participants Appendix H. Risk communication workshop scripts for audio recordings (based on Appendix D 1-4 charts) Appendix I. Risk communication workshop news articles Appendix J. Risk communication workshop links to videos recommended by experts. Appendix K. Risk communication workshop database of results 167 Appendix A. UBC Research Ethics Board's Certificates of Approval UBC W The University of British Columbia Office of Research Services Behavioural Research Ethics Board Suite 102, 6190 Agronomy Road, Vancouver, B.C. V6T 1Z3 CERTIFICATE OF APPROVAL - FULL BOARD PRINCIPAL INVESTIGATOR: Michael F. McDonald INSTITUTION / DEPARTMENT: UBC/Collegefor Interdisciplinary Studies/Applied Ethics UBC BREB NUMBER: H08-00700 INSTITUTION(S) WHERE RESEARCH WILL BE CARRIED OUT: j Institution | Site UBC Vancouver (excludes UBC Hospital) Other locations where the research will be conducted: The workshops will take place at locations that are most convenient for my participants. For example, if I am successful in recruiting parents from a dance studio, I will hold the workshops at that location. If I am successful in recruiting a group of office workers to participate in a workshop, then I will hold the workshop at their office. The default location for all workshops will be UBC Vancouver campus. All of the workshops will take place in a private room so that outsiders will not be able to observe. CO-INVESTIGATOR(S): N/A SPONSORING AGENCIES: N/A PROJECT TITLE: Linking Risk Communication and Biomedical Ethics: The Case of Pre-implantation Genetic Diagnosis (PGP) REB MEETING DATE: CERTIFICATE EXPIRY DATE: June 26, 2008 bung 26, 2009 DOCUMENTS INCLUDED IN THIS APPROVAL: DATE APPROVED: [June 27, 2008 [Document Name | Version | Date Protocol: A. Research proposal Final May 7,2008 Consent Forms: B. Consent form Final May 7,2008 Advertisements: C. Recruit poster interview Final May 7, 2008 F. Recruit poster workshop Final May 7, 2008 Questionnaire. Questionnaire Cover Letter. Tests: I. Workshop workbook Final May 7,2008 H. Workshop audio scripts Final May 7, 2008 E. Workshop Final May 7,2008 G. Workshop news stories Final May 7, 2008 D. Mental model interview questions Final May 7, 2008 Other: Findings will be posted on this password protected site that currently has no content. http://www.riskcommunication.info/home.html 168 The application for ethical review and the document(s) listed above have been reviewed and the procedures were found to be acceptable on ethical grounds for research involving human subjects. Approval is issued on behalf of the Behavioural Research Ethics Board and signed electronically by one of the following: Dr. M. Judith Lynam, Chair Dr. Ken Craig, Chair Dr. Jim Rupert, Associate Chair Dr. Laurie Ford, Associate Chair Dr. Daniel Salhani, Associate Chair Dr. Anita Ho, Associate Chair 169 Appendix B. Interview questions for experts Linking Risk Communication and Biomedical Ethics: The Case of Pre-implantation Genetic Diagnosis (PGD) Questions for experts: running time 1 hour Introduction Thank you very much for agreeing to participate in this research study. I have asked you to participate in this project because you have been identified as an expert in some aspect of PGD technology. In this hour long interview, I will ask you to review a series of charts that will eventually be used to create risk communication strategies for lay potential users of PGD technology. These charts represent the findings of a literature review I conducted on the social, technical, ethical, and economic dimensions of PGD technology. Please do not be concerned if you are unable to comment on particular issues or charts. I will be interviewing a wide range of experts until all aspects of the charts are covered. The purpose of this interview is to evaluate the accuracy and completeness of these charts and also to identify any phantom risks that may be presented in these charts. This project is being funded through the Research Allowance portions of my Canadian Institutes for Health Research (CIHR) Ethics of Health Research and Policy Training Project and the CIHR Institute of Genetics Doctoral Research Awards. Part 1: Social dimensions of PGD Couples who pursue PGD do so in order to avoid having a child with a specific genetic condition. However, there are many social risks associated with PGD that must also be considered. I would like to start our interview be asking you to comment on the validity of the chart that outlines the social risks of PGD. Ql) Do you think the chart accurately represents the social risks of PGD? Why/why not? Q2) Is there anything missing from the chart? Q3) Is there anything that should be removed from the chart? Now I would like you to comment on the chart that outlines some of the societal factors that may inform a person's decision to pursue PGD. 170 Q4) Do you think the chart accurately represents the societal factors that may inform a person's decision to pursue PGD? Why/why not? Q5) Is there anything missing from the chart? Q6) Is there anything that should be removed from the chart? Part 2: Technical dimensions of PGD Now let's move on to some of the other risks of PGD technology. There are many technical risks associated with PGD that must be considered by potential users. For example, PGD poses certain risks to an embryo, foetus, or child. Q7) Do you think the chart accurately represents these risks? Why/why not? Q8) Is there anything missing from the chart? Q9) Is there anything that should be removed from the chart? Many of the risks posed to an embryo, foetus, or child also pose some risk to the women who undergo PGD. Q10) Do you think the chart accurately represents these risks? Why/why not? Ql 1) Is there anything missing from the chart? Q12) Is there anything that should be removed from the chart? Part 3: Wrap up Lastly, do you have any additional comments that you would like to make specifically about these charts or generally about this project? Thank you so much for your time! 171 Appendix C. Mental model interview questions for lay participants 1. Have you ever heard of pre-implantation genetic diagnosis or PGD? 2. Have you ever heard of a saviour sibling? 3. Have you ever heard of genetic testing? 4. Have you every heard of in vitro fertilization or IVF? 5. How about prenatal testing? (e.g., Amniocentesis) 6. What sort of tests can be performed on a woman while she is pregnant? If any. 7. How did you come to learn this (above) information? 8. DEFINITION: IVF is a technology used by people who have difficulties achieving pregnancy. During IVF, egg cells are fertilised by sperm outside of the body. 9. DEFINITION: PGD is used to analyse genetic material from fertilized human embryos during in vitro fertilization. Screened embryos are selected and put back into the uterus of the female patient and hopefully grown to term. There are 3 applications for PGD. One is to screen embryos for abnormalities or identify markers for specific genetic conditions. The second is for sex selection, which is illegal in Canada. The third is to produce a saviour sibling or a sibling that is a tissue match for an existing sick child. Under Canadian law, embryos that are not selected can be destroyed or donated to stem cell research. 10. Who do you think might be concerned about PGD? 11. Who might be excited about PGD? 12. What kind of risks do you think PGD may pose? 13. What kind of benefits do you think PGD may pose? 14. Are there any risks or benefits that PGD may pose specifically to a woman? 15. How about the embryo, foetus, and child? 16. What kind of social risks do you think PGD may pose if any? 17. Where would you go to find info about PGD? 172 172 Appendix D. Expert charts (final versions) Expert information charts (to be disseminated in handout form before each audio recording is played and collected before the vote/second questionnaire) Rana and Ted's decision: Technical risks of PGD to a woman (chart 1) 173 Decision to pursue PGD Getting pregnant Short term risks of PGD are likely the same as IVF. For example, hormone injections that stimulate ovulation and the egg retrieval process during IVF are both physically invasive and pose some risk to women's bodies. Other short term impacts include Ovarian Myperstimulation Syndrome (OHSS) and infections. OHSS can range from mild to moderate forms in up to 1 in 5 women and in rare extreme cases can lead to death. All physical risks imposed on female partner even if they are not the one affected by the disorder being screened for (e.g., Huntington's Disease) A woman or couple may experience stress associated with undergoing IVF when they are not actually infertile. Age-Women 40 and over are much less likely to have a successful pregnancy through IVF. Pregnancy Additional tests-After PGD has occurred and the embryo has been implanted into the uterus, the woman may be advised to have additional tests such as amniocentesis or CVS to detect misdiagnosis and health of the fetus over time. Some women however avoid these tests and their associated risks to the pregnancy and decide to trust PGD. However, receiving a relief diagnosis can help parents acquire assistance from the system and prepare for the future. Birth and beyond Some argue that there B a link between IVF and breast and ovarian cancer but this appears to be a very soft link that disappears in studies that include large samples. Some also point out that additional long term studies are required before we can statethat IVF or PGD have negligible long term impacts on patients. PGD cannot guarantee a "normal" embryo, fetus, child. There is a potential for misdiagnosis AND a potential for the child to acquirecondrtions that were NOT screened out through PGD. IVF lowers the risk of ectopic pregnancies orpregnancies that occur outside the uterus, which can endanger a woman life, and cause severe bleeding and pain. However the background risk of ectopic pregnancy in natural pregnancies is less than 2%. i n A significant psychological risk of IVF is that the woman may not become pregnant. Psychological risks of IVF or PGD must also be compared to those of abortiore after diagnosis. Risk of multiple births. Typically three to six embryos are used during IVF and as a result, about one trird of IVF live births are multiple in nature. However, multiple birth are even more likely for women who hare ingested fertility drugs instead of undergoing IVF. Risks multiple births pose to women include premature labour, difficutt deliveries, physical and emotional stress. That said, the trend is now to conduct one embryo transfers for best canddates, which eliminates the risk of multiple birth, but also reducesthe chance of becoming pregnant 1 ' ~~i » Rana and Ted's decision: Technical risks of PGD to the embryo, foetus, and child (chart 2) 174 Decision to pursue PGD Physical "risks" to embryo Physical "risks" to fetus IVF lowers the risk of ectopic pregnancies or pregnancies that occur outside the uterus which may lead to the death of the fetus However the background risk of ectopic pregnancy in natural pregnancies is less than 2%. IVF poses less risk than amniocentesis or CVS for the embryo x I Experts state that the risks associated with technologies such as CVS, IVF, and PGD are smal. That said, there is an additional minor increase associated with them. For IVF, PGD, and CVS, the additional risk for the pregnancy is 1 %. However with CVS, there is also the risk that the pregnancy may fail. Amniocentesis has an additional risk of .5%. Lastly, these risks are for each fetus.so in a multiple birth situation, there would bean increased risk of 1% plus 1%. Physical "risks" to child Risk of being bom prematurely. Some will do well, others will not. if bom prematurely, the baby may experience short term risks such as difficulty breathing, undeveloped organs, and death or long term risks such as gastrointestinal issues, cerebral palsy, developmental delays, or chronic lung disease. Risk of multiple births. Risk of multiple birth is linked to the age of mother, number of embryos implanted, and the quality of those embryos. On average, 3-six embryos are used for IVF and as a result, about one third of IVF live births are multiple in nature. These risks can be lowered by implanting fewer embryos. However, multiple birth are even more likely for women who have ingested fertility drugs instead of undergoing IVF. Risks to fetuses and children include increased risks of neonatal mortality, SIDS, birth defects, developmental disabilities, and child abuse according to Multiple Birth Canada. I I i r PGD cannot guarantee a "normal" embryo. The Genetics and Public Policy Centre's 2004 report outlines a series of technical challenges inherent to PGD that "can lead to a misdiagnosis of the embryo" and result in an "abnormal" fetus and child. For example, testing occurs under strict time constraints and in addition, only small amounts of DNA are examined. Also, there is the potential for the child to acquire condtions that were NOT screened outthrotgh PGD. i i i—r Technical medium during IVF can accidentally "turn on" or "turn off" genes while culturing cells. This may increase risks of disorders such as Angelman Syndrome (AS) or Prader-Willi syndrome (PWS). The long term physical health impacts of PGD are iikey similar to iVF, which are known. Children bom through IVF have twice as high a risk of major birth defects (cardiovascular, chromosomal) { 3 or 4-8.5%) and are more likely to have multiple major defects when compared to naturally conceived infants (2 or 3-4%). That said, the probability that these problems will occur is low (as shown by percentages), although they may be significant. Children bom from women who undergo PGD may also be subjected to psychological risks or benefits, for example, if they were bom to create compatible ceils for an existing child (HLA matching). i i ' i i History of paternalistic and even unethical health care practices and abuses (such as those outlined in Beecher, 1966), among other things 4 175 Discrimination against people with disabilities Social and biological pressure to have biological children Shift in health care ethics from paternalism (beneficence) to autonomy People want to manage their risks. We don't want to take it as it comes. We talk abort preferences as if they are risks. 1 Unwillingness to foster change of social norms Difficulty conceiving (miscarriages, infant deaths) although those who use PGD are not necessarily infertile Normalization of discrimination over time Responsibility for disability is located at the individual scale Pressure to pursue technological solutions and failure to consider other alternatives such as adoption Cost not covered by health care system (Regulatory framework is currently in flux) PGD technology to avoid known conditions. Limited by current genetic knowledge, which may expand overtime Will PGD reduce society's overall compassion for people with disabilities? Public adoption (through government agency) in Canada for an infant with no "special needs" may take up to 8 years. Private adoption may take less time but cost anywhere from $10,000- $18,000. Costs for international adoption may also take less time but are also cost!/ varying from $15,000 (Haiti), to $30,000 (Guatemala) and as high as 50000 (US), (source: Canada Adopts!) Societal and economic factors that may inform a decision to pursue PGD (chart 3) Social pressure to submit to such technologies after they become a standard part of prenatd care and to remain consumers throughout a pregnancy Influences the decision process for PGD We all have the "right" to control our own bodies Parents have the "right" to a healthy child 176 What happens when diagnostic tests on fetuses become "standard practice"? But very few are eligible for PGD and would not ha/e additional tests anyway. Questions concerning moral status of fetus can be avoided and it is less harmful to a woman. But PGDposes more risks to an embryo than prenatal diagnoses More pregnancies are treated as potentially problematic, -fc The pregnancy experience is viewed as a medical procedure instead of a natural process. Is PGD a slippery slope to eugenics? Individuals may be surprised by test results (is an ultrasound a baby photo or a screen?) or may falsely believe that problems can be addressed. Women may experience long or short term anxiety associated with the additional knowledge. IE Individual decision making is reframed as available alternatives are reduced (e.g., ability to refuse tests or screens) and the full range consequences are not considered. Concerns may lead others to judge those who use PGD. Also, some may resent those who get to have "normal" children. Is PGD "playing God" with human life? Does an embryo have a "right to life"? Should all life be consid ered "a gitt"? Do users of PGD technology (mostly women) decide who/what types of people should enter society? But genetics are only one aspect of a person. There is a very small risk that genetic tests may reveal information about the biological family without their consent (e.g., biological father). Will PGD impact how we view people with disabilities? Will it make us less tolerant? But who will volunteer to raise disabled children so thatsociety can be diverse? Decision contexts regarding the role of the parent and tie meaning of a pregnancy are altered. Additional burden on users of PGD technology. Social risks of PGD (chart 4) Appendix E. Example of how expert charts changed over time due to expert participant feedback Technical risks of pre-implantation genetic diagnosis (PGD) to the embryo, foetus, and child (pre-expert interview status) Decision to pursue PGD Physical risks to embryo Physical risks to foetus Risk of ectopic pregnancies or pregnancies that occur outside the uterus which may lead to the death of the Physical risks to child Risk of multiple births. On average, six embryos are required for IVF and as a result, about one third of IVF live births are multiple in nature. Risks to fetuses and children include increased risks of neonatal mortality, SIDS, birth defects, develoDmental disabilities and child abuse PGD cannot guarantee a "normal" embryo. The Genetics and Public Policy Centre's 2004 report outlines a series of technical challenges inherent to PGD that "can lead to a misdiagnosis of the embryo" and result in an "abnormal" foetus and child. For example, testing occurs under strict time constraints and in addition, only small amounts of DNA are examined. i i i i Unknown long term health impacts. 178 Technical risks of pre-implantation genetic diagnosis (PGD) to the embryo, foetus, and child (post-expert interview status) Decision to pursue PGD Physical "risks" to embryo Physical "risks" to fetus IVF lowers the risk of ectopic pregnancies or pregnancies that occur outside the uterus which may lead to the death of the fetus. However the background risk of ectopic pregnancy in natural pregnancies is less than 2%. IVF poses less risk than amniocente sis or CVS for the embryo Experts state that the risks associated with technologies such as CVS, IVF, and PGD are small. That said, there is an additional minor increase associated with them. For IVF, PGD, and CVS, the additional risk for the pregnancy is 1%. However with CVS, there is also the risk that the pregnancy may fail. Amniocentesis has an additional risk of .5%. Lastly, these risks are for each fetus, so in a multiple birth situation, there would bean increased risk of 1 % plus 1 %. T T Physical "risks" to child Risk of being born prematurely. Some will do well, others will not. If born prematurely, the baby may experience short term risks such as difficulty breathing, undeveloped organs, and death or long term risks such as gastrointestinal issues, cerebral palsy, developmental delays, or chronic lung disease. Risk of multiple births. Risk of multiple birth is linked to the age of mother, number of embryos implanted, and the quality of those embryos. On average, 3-six embryos are used for IVF and as a result, about one third of IVF live births are multiple in nature. These risks can be lowered by implanting fewer embryos. However, multiple birth are even more likely for women who have ingested fertility drugs instead of undergoing IVF. Risks to fetuses and children include increased risks of neonatal mortality, SIDS, birth defects, developmental Hisflhilitips anri r.hilri ahiisp armrHinn tn Multinlp Rirth Canada I o PGD cannot guarantee a "nornial" embryo. The Genetics and Public Policy Centre's 2004 report outlines a series of technical challenges inherent to PGD that "can lead to a misdiagnosis of the embryo" and result in an "abnormal" fetus and child. For example, testing occurs under strict time constraints and in addition, only small amounts of DNA are examined. Also, there is the potential for the child to acquire conditions that were NOT screened out through PGD. Technical medium during IVF can accidentally "turn on" or "turn off' genes while culturing cells. This may increase risks of disorders such as Angelman Syndrome (AS) or Prader-Willi syndrome (PWS). The long term physical health impacts of PGD are likely similar to IVF, which are known. Children born through IVF have twice as high a risk of major birth defects (cardiovascular, chromosomal) (3 or 4-8.5%) and are more likely to have multiple major defects when compared to naturally conceived infants (2 or 3-4%). That said, the probability that these problems will occur is low (as shown by percentages), although they may be significant. Children born from women who undergo PGD may also be subjected to psychological risks or benefits, for example, if they were born to create compatible cells for an existing child (HLA matching). I I I 1 Appendix F. Risk communication workshop presentation that was used to guide the workshop (this presentation includes the expert based FAQ's and the consequence matrices) Medical risk communication workshop August, 2008 Facilitator: Holly Longstaff Welcome!! • A little about me and my PhD project • Why is this important? • How will I use this information? • Introductions-a little about you 1. What's your name? 2. Have you ever done a focus group or telephone survey before? 3. Why did you decide to participate today? Agenda (2.5 hours in total) m^t Please hand in your signed consent forms! Part 1: Pre-deliberation questionnaire (20 min) Part 2: Presentation of background information (30 min) *Short break for snacks! (10 min)* Part 3: Deliberation and private vote (1 hour) Part 4: Post-deliberation questionnaire (30 min) Please pick up your honorarium before you leave! 181 Part 1-30 min Please make sure you hand in your signed consent form Then, complete your pre-deliberation questionnaire 182 Part 2-30 min Task: Learning as much as we can about pre-implantation genetic diagnosis or PGD before we deliberate about it. 183 184 These FAQ's will help you to answer the workbook questions. 184 Pre-implantation genetic diagnosis: background information What is PGD? • PGD is used to analyze genetic material from fertilized human embryos during in vitro fertilization or IVF. Screened embryos are selected and put back into the uterus of the female patient and hopefully grown to term. 185 186 PGD background What is in vitro fertilization or IVF? • IVF is a technology used by people who have difficulties achieving pregnancy. During IVF, egg cells are fertilised by sperm outside of the body. What is the connection between IVF and PGD? • PGD is performed on a patient while she is undergoing IVF (extra step). BUT-People who use PGD do not necessarily have problems achieving pregnancy unlike users of IVF. Now here comes the tricky part! • Although you must go through IVF to use PGD, you do not have to get PGD if you undergo IVF. 186 PGD background Why would someone want to use PGD? There are 3 applications for PGD (1) to screen embryos for abnormalities or identify markers for specific genetic conditions. (2) for sex selection, which is illegal in Canada. (3) to produce a saviour sibling or a sibling that is a tissue match for an existing sick child. Under Canadian law, embryos that are not selected can be kept for future use, destroyed, or donated to stem cell research. What is PGD used for? PGD is used to screen for genetic conditions or chromosomal abnormalities. (TRUE) PGD is used to help infertile couples achieve pregnancy. (FALSE) • PGD is used to determine the sex of an embryo. (TRUE) • PGD is used to diagnose abnormalities during an ultrasound. (FALSE) • PGD is used to match tissue to create compatible cells for an existing child. (TRUE) 187 PGD background How do people get PGD? • It is actually quite difficult to qualify for PGD right now in Canada. You can't just walk into a fertility centre and ask for it. You must be at risk for a known serious genetic disorder. If you are eligible, a trained genetic counselor must explain the risks and benefits of PGD to you before you make your decision. 188 189 Useful definitions (alphabetical order) Amniocentesis: a prenatal test that detects chromosomal abnormalities and other genetic conditions while the fetus is in the mother's whomb. It tests a sample of amniotic fluid, which is the fluid that surrounds the fetus. Chorionic villus sampling or CVS : a prenatal test that detects chromosomal abnormalities and other genetic conditions while the fetus is in the mother's whomb. It takes cells from the placenta, which surrounds the fetus, and it can be performed earlier on than amniocentesis. Chromosomal abnormalities: an abnormal number or structure of the chromosome Chromosome: a chromosome is a bundle of DNA. Every human has 46 chromosomes. Disability types - Medical model of Disability: People disabled by their body-structure related functioning. - Social model of disability: People disabled by the attitudinal and environmental barriers they experience due to their non normative body-structure related functioning that hinder their full and effective participation in society on an equal basis with others. DNA: contains our genetic information. Embryo: stage in prenatal development from fertilization to the 8th week of a pregnancy. Fetus: stage in prenatal development from 9,t' week onward. Gene: a gene is a distinct section of DNA In vitro fertilization or IVF: technology used by people who have difficulties achieving pregnancy. During IVF, egg cells are fertilised by sperm outside of the body. Pre-implantation genetic diagnosis or PGD: technology used to analyze genetic material from fertilized human embryos during IVF. Pre-natal: before birth. 189 Recap! The process of PGD (Remember there are 3 main applications) Application 1: to diagnose genetic conditions or chromosomal abnormalities 1. A couple want to have a biological child. 2. At least one person in a couple is identified as being at risk for a serious genetic disorder. 3. They decide to have PGD to avoid passing on the condition. 4. A trained genetic counselor explains the risks and benefits of PGD to them. 5. The woman undergoes IVF. 6. During IVF, the egg cells are fertilised by sperm outside of the body to make fertilized embryos. 7. The genetic material from those embryos are tested to screen for the specific genetic conditions or chromosomal abnormalities. 8. Embryos that DO NOT carry genetic markers for the specific condition or contain chromosomal abnormalities are selected and put back into the uterus of the woman. 190 The process of PGD Application 2: to determine the sex of an embryo. 1. A couple want to have a biological child. 2. At least one person in a couple is identified as being at risk for a serious genetic disorder that is (a) linked to the sex of the child (called X linked) or (b) the couple want to use PGD for "family balancing" (which is illegal in Canada). 3. They decide to have PGD to have a child of a particular sex. 4. A trained genetic counselor explains the risks and benefits of PGD to them. 5. The woman undergoes IVF. 6. During IVF, the egg cells are fertilised by sperm outside of the body to make fertilized embryos. 7. The embryos are screened and only female (in the case of X- linked disorders) or ? in the case of family balancing are selected and put back into the uterus of the woman. 191 The process of PGD Application 3: to match tissue to create compatible cells for an existing child 1. A couple have a child that is very sick and needs a special procedure such as a bone marrow transplant. 2. They have searched for a donor but there are no matches. 3. They decide to have PGD to produce a baby that is a tissue match for their sick child. 4. A trained genetic counselor explains the risks and benefits of PGD to them. 5. The woman undergoes IVF. 6. During IVF, the egg cells are fertilised by sperm outside of the body to make fertilized embryos. 7. The genetic material from those embryos are tested to see if any match the sick child. 8. Embryos that match are selected and put back into the uterus of the woman. 9. After birth, stem cells are harvested from the umbilical cord of the new baby and used to help the sick sibling (no guarantee). 192 PGD used to detect a spectrum of conditions Some point out that it is important not to trivialize or conflate all genetic conditions. What do you think? In each case, are we selecting for a feature? Are we selecting against pathology? But where does a feature end and pathology begin? Non-fatal conditions Fatal conditions From birth: Down's Syndrome Caused by an extra chromosome Symptoms vary widely Likelihood of having a child with Down's increases with age of mother but can occur at any age in any culture Children with Down's may face more or less health problems than other children depending on the condition • Hyperlink to video from Down's Syndrome Society Late/adult onset: Cancer BRCA1 is a gene that increases risk of breast and ovarian cancer This test detects mutations in BRCA genes BBC News story (see handout) Early onset: Tay-Sachs Disease Genetic disorder Progressive destruction of central nervous system Symptoms begin at around 6 months and may include eventual blindness, mental retardation, inability to swallow, breathing difficulties Most die before 5th birthday Hyperlink to public service announcement video from National Tay-Sachs & Allied Diseases Association htip://www.ntsad.org/ Late/adult onset: Huntington's Disease Most common type of adult-onset condition • Do not develop symptoms (motor disturbance, dementia) until later in life. • It is progressive and degenerative and fatal Washington Post story (see handout) 193 Regulatory framework FAQ's Who regulates PGD in Canada? An agency called Assisted Human Reproduction Canada that reports to the Minister of Health. However, the regulatory framework is currently under development. Has PGD always been regulated in Canada? • No. Before 2004, there were no regulations, standards, or professional guidelines for the use of PGD in Canada. Is PGD covered by the Canadian health care system? No, so PGD is quite expensive and can cost up to $20,000 +. Although some Canadians (such as Ontario citizens) may be partially covered through insurance. How common is PGD in Canada? • Fewer than 20 children have been born following PGD in Canada but over 1000 have been in the world. Can PGD be used for "family balancing" in Canada? No, this use of PGD is illegal here. 194 Considerations for a regulatory system in flux Who will protect the rights of those who are screened for? We accept laws against PGD for sex selection. Why not laws against PGD for certain disabilities? Need some way of ensuring that everyone's rights are not violated. Can informed consent address all of our concerns? Role of the state in influencing practice or regulatory backdrop for PGD (coverage of 1VF, sex selection illegal under AHR Act, legal abortions) Regulations only work if they have teeth. How can we ensure that relevant parties will follow regulations? Preventing certain conditions through technologies like PGD could reduce overall disability related costs for the hearth care system but costs are not currently covered in Canada. There is a limited amount of money available to the health care system. If we pay for people to undergo PGD, then we must reduce funds elsewhere. Is this a tradeoff we can live with? Should cost effective choices for society be imposed on the minority? Should society benefit for selecting against all types of disability? 195 Risks/benefits of PGD (Holly will hand out charts that go with each audio recording) Rana and Teds decision • This decision is about the technical risks/benefits PGD poses to a woman. • Please listen to this audio story and feel free to examine chart 1 (handout). Rana and Ted s decision • This decision is about the technical risks/benefits PGD poses to the embryo, fetus, and child • Please listen to this audio story and feel free to examine chart 2 (handout). Social risks of PGD • We are now moving on to the social risks/benefits of PGD. • Please listen to this audio story and feel free to examine chart 3 and 4 (handout). 200 NOTE: the following section was only viewed by treatment B participants 200 Consequence Matrix for PGD at the individual scale: objectives by alternatives Alternatives Natural: status quo PGD: testing fertilized embryos before they are implanted in the female patient Prenatal diagnosis: includes a variety of tests used to determine the health and status of an unborn fetus (e.g., ultrasound Amniocentesis, chorionic villus sampling) Adoption: private, public, or international Fundamental objectives of families 1) To avoid condition X Chance of avoiding an undesirable condition depends on the condition type (e.g., chance of chromosomal abnormalities increase with age of mother, with autosomal dominant disorders, there is typically one affected parent, with autosomal recessive conditions if both parents are unaffected but both carriers, they have a 25% chance of having an affected child) Embryos can be screened to avoid certain conditions One cannot avoid an undesirable condition, but can determine if the fetus is affected before birth. Some may choose to abort an affected fetus (a difficult option both physically and psychologically). However, others may view this as an unacceptable option for a variety of reasons. Can avoid conditions but it takes a long time, it is expensive, and the couple cannot have a biological child. 2) To select sex of child 50-50 chance of having the desired sex Illegal in Canada Illegal in Canada Can select the sex of the child 3) To create saviour sibling Naturally creating a child that is a perfect match for existing sick child is unlikely but theoretically possible. Embryos can be screened for a perfect match Naturally creating a child that is a perfect match for existing sick child is unlikely but theoretically possible. Highly unlikely Consequence Matrix for PGD at the group scale: objectives by alternatives Alternatives Natural: status quo PGD: testing fertilized embryos before they are implanted in the female patient Prenatal diagnosis; a variety of tests that are used to determine the health and status of an unborn fetus (e.g..ultrasound, amniocentesis, chorionic villus sampling) Adoption: private, public, or international Fundamental objectives of groups (people with disabilities example) 1) To encourage better lives for those affected by disability More people having natural pregnancies will lead to more people born with disabilities and increased diversity for society. If there are more people born with disabilities, society may be more willing to support them (e.g., modifying infrastructures, assistance and support from government programs, wider social networks) PGD is used to avoid people with disabilities, which will reduce the overall number of people with disabilities in our society. This may reduce our overall compassion and available resources for people with disabilities and their families. These techniques may help families prepare for life with a child affected by disability. However, it may also encourage some to abort affected fetuses. Adoption may increase the likeness of placing children who have "special needs" who would otherwise grow up without a family. 2) To reduce discrimination against people with disabilities More people having natural pregnancies will lead to more people born with disabilities and increased diversity for society. If there are more people born with disabilities, our society may begin to show more tolerance for these folks. PGD is used to avoid people with disabilities, which will reduce the overall number of people with disabilities in our society. This may lead some to believe that discriminating against them is acceptable since we are trying to reduce their overall numbers. These techniques may help families prepare for life with a child affected by disability. However, it may also encourage some to abort affected fetuses. This may lead some to believe that discriminating against people with disabilities is acceptable since we are trying to reduce their overall numbers. If adoption becomes socially desirable and popular, it may become more acceptable to adopt children with special needs and less acceptable to «{\o discriminate against them. Consequence Matrix for PGD at the societal scale: objectives by alternatives Alternatives Natural: status quo PGD; testing fertilized embryos before they are implanted in the female patient Prenatal diagnosis: a variety of tests that are used to determine the health and condition of an unborn fetus (e.g., ultrasound, amniocentesis.chorionic villus sampling) Adoption: private, public, or international Fundamental objectives of society 1) To foster overall health & wellbeing of Canadians Some natural pregnancies will result in children with disabilities who may or may not experience short lived or painful lives. PGD will reducing the overall number of those born with specific conditions and will therefore increase the overall number of healthy babies. Prenatal diagnosis can help families prepare for life with children affected by disabilities or avoid affected fetuses who would experience short lived or painful lives. All types of adoption should be encouraged so that all of us have the opportunity to be raised by loving families. 2) To foster overall diversity of Canadian society Some natural pregnancies will result in children with disabilities who contribute to the diversity of our society. Reducing the overall number of people with disabilities will reduce the genetic and cultural diversity of our Canadian society. Prenatal diagnosis can help families prepare for life with children affected by disabilities. However, it may also lead some to abort affected fetuses, which may reduce the overall genetic and cultural diversity of our Canadian society. If adoption becomes socially desirable and popular, it may become more acceptable to adopt children with "special needs", which will increase the overall genetic and cultural diversity of our Canadian society. 3) To treat all Canadians equally Supporting natural pregnancies and therefore not supporting techniques that assist in reproduction means that some Canadians will not be able to have their own biological "normal" children. Part of treating Canadians equally means that there should be equitable access to government services. However PGD is expensive and currently only available to some. If PGD is made available to all, other services will have to be reduced. Part of treating Canadians equally means that there should be equitable access to government services. However, prenatal diagnosis techniques are expensive. If they are made available to all, other services will have to be reduced. All types of adoption should be encouraged so that all of us have the opportunity to be raised by loving families. 203 204 • SDM section complete 204 Short group discussion • A lot of information has been presented in the last 30 minutes, do you have any comments or questions at this point that you would like to share with the group? 206 Break time!! • Please take 10 minutes to grab a snack or take a bathroom break. _̂ _ : - : - : 206 Part 3-1 hour Small group discussion (Feel free to refer to handouts and ask questions!) Question: how should we communicate the risks/benefits of PGD? 207 Rules of discussion • There are no right or wrong opinions here • Its okay to disagree! • Always remember to be respectful - Let others have their say - Try not to judge - Don't cut people off while speaking - No side discussions please - Really listen to what is being said Role of the facilitator • Each question will be discussed in turn by the group with guidance from the facilitator (Holly) • Holly will ... - Go back to FAQ's or definitions upon request - Keep the discussion on track and respectful • Sonjawill... - Keep track of key ideas - Record points of consensus and persistent disagreements Risk communication background • What is risk communication? • Who are risk communicators? • Why do we communicate risk/benefit information? • What role do experts play? • How do we communicate risk? When picking a risk communication strategy.... Consider... • Is this the kind of risk information you need to/can only communicate once? Or will the information change over time? (think tsunami warning vs. heart health information) • Should the method be interactive or one way? (think focus groups vs. public service announcements) • How much can we spend on this method? (think $$-who is funding it?) • How quickly does this information need to get out there? (think food recalls vs. health risks of fatty foods) • Do we have all the information we need or do we need to talk to others? (think exposure to toxins vs. public policy about risks of salmon farming) 211 A few popular risk communication strategies Public service announcement (PSA) • To provide a lot of one way information to people quickly. This is a one shot, expensive format. Pamphlets • To provide one way information to people in a format they can use over time. This is a one shot, cheap format. Website • To provide information to people that can modified over time and can be interactive. This method is only good for those comfortable with and able to access the net. Face to face discussion with medical expert To provide information from one expert. This may or may not be a one shot format and will most likely be interactive depending on the medical expert. Telephone survey • To find out what people think and give them information. This is a one shot format that is both quick and cheap. Interviews or discussion groups To learn about how people think about a topic, determine misconceptions, gather advice, give people new information, and develop risk communication strategies. This is an expensive, time consuming, and interactive format. Something else??????? 212 Discussion questions Questions to consider: 1. How can our risk communication strategies help to achieve the fundamental objectives of families, group, and society? 2. What are the pros and cons of each strategy? 3. What tradeoffs are we willing to live with? 213 214 The following slide was only shown to treatment B participants 214 Recall the relevant fundamental (or most important) objectives of families, groups, and society for PGD Families 1. To avoid condition X 2. To select sex of child 3. To create saviour sibling Groups (For example: people with disabilities) 1. To encourage better lives for those affected by disability 2. To reduce discrimination against people with disabilities Society 1. To foster overall health & wellbeing of Canadians 2. To foster overall diversity of Canadian society 3. To treat all Canadians equally Discussion questions continued Potential risk communicator • Who should communicate this information? (For example: physicians, genetic counselors, Health Canada, academics, someone else?) Target audience • Who should we targeting with these strategies? (For example: all Canadians, people considering PGD, people considering IVF, all women who are pregnant, someone else?) 216 Discussion questions continued Developing a risk communication strategy • What are the best strategies to use for PGD? Public service announcement (PSA), pamphlets, website, face to face discussion with medical expert, telephone survey, interviews or discussion groups, something else? • What types of information should we consider for our strategies? Think about the information provided to you today, is it all relevant? • Should all applications of PGD be treated in a similar manner? In other words, do we use ail types of information for all uses of PGD? • Should certain types of information be optional at times? • Who should have a say when developing these materials? In other words, who should comprise our expert panel? • When should our target audience receive risk information about PGD? Please hand in all background materials and receive voting cards/second questionnaire Private workbook vote Recommended risk communication strategies Feel free to use the strategies developed by the group or develop your own. 1. Who should communicate information about PGD? (For example, physicians, genetic counselors, Health Canada, academics, someone else?) 2. Who is the target audience? (For example, all Canadians, people considering PGD, people considering IVF, all women who are pregnant, someone else?) 3. Check off the types of information that you include in your strategy. Think about the range of information that was presented today. 4. Check off the strategies you like best from the list. Notice there is a space to add a new strategy. 5. Please describe how your strategy would work. 219 220 Part 4-30 min Post-deliberation questionnaire * Remember to pick up your honorarium before you leave! * Do you want to know what I do with all this information? All combined and anonymized results and votes will be posted on a password protected website. 220 221 C om plete 221 222 Appendix G. Risk communication workshop workbook for lay participants Pre-deliberation questionnaire Hi, and thank you for participating in this workshop. Before we begin, I would like to take a moment to explain a few things. First, THIS IS NOT A TEST!!!! You will have the opportunity to revise each of the following questions at the end of the workshop. I know some of these questions may be difficult to answer but please do your best to answer each one. It is perfectly okay to guess! But if you don't want to, you can just write in "don't know". I don't expect you to know all the answers. I am using this questionnaire to determine how much you leam and how your opinions change over the course of the workshop. Please fill in the following information before you begin Name: Date: , 2008 223 222 Please take your time to respond to the following questions. 2) Have you ever heard the term pre-implantation genetic diagnosis otherwise known as PGD? (Please circle one) YES NO 3) Please provide a description of PGD. Remember, it's okay to guess! You will have the opportunity to revise all your responses at the end of the workshop. 4) Where have you heard about PGD? (Feel free to circle more than one response) a) The television news b) The newspaper c) The Internet d) In a magazine e) In a classroom from a teacher/professor f) From a physician/nurse/genetic counsellor g) From my friends/family h) I have never heard of PGD before today i) Other (please describe) 5) As far as you know, what are the applications of PGD? (Feel free to circle more than one response) a) PGD is used to diagnose genetic conditions or chromosomal abnormalities. b) PGD is used to help infertile couples achieve pregnancy. c) PGD is used to determine the sex of an embryo. d) PGD is used to diagnose abnormalities during pregnancy. e) PGD is used to match tissue to create compatible cells for an existing child. f) Don't know 6) Who regulates PGD in Canada? (Feel free to circle more than one response) a) An agency called Assisted Human Reproduction Canada that reports to the Minister of Health. b) No agency takes responsibility for PGD in Canada. c) Each province is responsible for creating its own agency that must report to the Minister of Health. d) Before 2004, there were no regulations, standards, or professional guidelines for the use of PGD in Canada. e) There have been regulations, standards, & professional guidelines for the use of PGD in Canada since the birth of the first test-tube baby, Louise Brown in 1978. f) Don't know 7) Is PGD covered by the Canadian health care system? (Please circle one) YES NO Don't know 8) Have you ever heard the term in vitro fertilization otherwise known as IVF? (Please circle one) YES NO 9) Please provide a description of IVF. Remember, it's okay to guess! You will have the opportunity to revise all your responses at the end of the workshop. 10) Where have you heard about IVF? (Feel free to circle more than one response) a) The television news b) The newspaper c) The Internet d) In a magazine e) In a classroom from a teacher/professor f) From a physician/nurse/genetic counsellor g) From my friends/family h) I have never heard of IVF before today i) Other (please describe) 11) Is there a relationship between PGD and IVF? (Please circle one) YES NO Don't know 12) If a woman wants to use PGD, must she also undergo IVF? (Please circle one) YES NO Don't know 226 13) If a woman undergoes IVF, must she also undergo PGD? (Please circle one) YES NO Don't know 14) What risks and benefits might PGD pose to a woman? Please list as many as you can. Point form is okay. Risks Benefits 15) What risks and benefits might PGD pose to an embryo, foetus, or child? Please list as many as you can. Point form is okay. Risks Benefits 16) What social risks and benefits might PGD pose? In other words, what risks and benefits might PGD pose to society, groups, or individuals or those other than families, doctors, etc? Please list as many as you can. Point form is okay. Risks Benefits 227 17) What are some of the reasons for why people may use PGD? (Feel free to circle more than one response) a) To help infertile people achieve pregnancy. b) To have a baby that could provide tissue for a pre-existing sick child. c) To have a baby of a particular sex. d) To have a baby that is NOT affected by a particular condition/disability. e) To avoid aborting an "abnormal" fetus. f) To have their own biological child. g) Adopting a child is too expensive or takes too long, h) To detect abnormalities during pregnancy. 18) In Canada, is it legal to use PGD to select a baby of a particular sex for "family balancing"? (Please circle one) YES NO Don't know 19) In Canada, can anyone get PGD? (Please circle one) YES NO Don't know 20) Is PGD common in Canada? (Please circle one) YES NO Don't know 21) Who (what individuals or groups) might be concerned about the use of PGD? (Please list below) 228 22) Why might the individuals or groups you mentioned in question 19 be concerned about PGD? 23) How confident are you in your knowledge of the risks and benefits of PGD? (Please circle one) 1 2 3 4 Not at all A little Somewhat Very 24) At this time, what would you say are your biggest hopes for PGD (if any)? For example, I hope that PGD will 25) At this time, what would you say are your biggest concerns regarding PGD (if any)? For example, I fear that PGD will 26) Please circle the response that best describes your opinion about PGD at this time. (Please circle one) a) I am strongly supportive of PGD b) I am somewhat supportive of PGD c) I am somewhat opposed to PGD d) I am strongly opposed to PGD e) I have no opinion about PGD f) I don't care about PGD Please use the space below to explain your above response. 27) Who would likely be the most impacted by the use of PGD? (Feel free to circle more than one response) a) Families b) Women c) Children d) People with disabilities e) Disability activists f) Families affected by disability g) Religious groups h) All Canadians i) Others (please name) j) Don't know 230 231 Private vote and post-deliberation questionnaire Please fill in the following information before you begin Name: Date: , 2008 232 Private vote: Post-deliberation A number of risk communication strategies for PGD were considered during your small group discussion. Please use this space to explain your preferred strategy. 1) Who should communicate information about PGD? (For example, physicians, genetic counselors, Health Canada, academics, someone else?) 2) Who is the target audience? (For example, all Canadians, people considering PGD, people considering IVF, all women who are pregnant, someone else?) 3) What types of information would you include in your strategy? Think about the range of information that was presented today and check off the topics you would include. Information type Check off V 1. General background about PGD (that I presented in FAQ's at the beginning) 2. Information about the PGD regulations in Canada (that I presented in FAQ's at the beginning) 3. Information on the spectrum of conditions that PGD can test for (I showed this in video and news article formats) 4. Technical information on the risks PGD poses to a woman (shown today in an audio recording and a chart-Rana and Ted's decision) 5. Technical information on the risks PGD poses to the embryo, fetus, and child (shown today in an audio recording and a chart- Rana and Ted's decision) 6. Information on the social risks of PGD (shown today in an audio recording and in charts) 7. Other information (please describe) 8. * only for B groups* Charts showing objectives by alternatives for individuals, groups and society (Decision Matrixes) 233 4) Please check off the strategies you like best for communicating the risks of PGD Notice that there is space to add in a new strategy. Strategy Check off V 1. Public service announcement (PSA) 2. Pamphlet 3. Website 4. Face to face discussion with medical expert 5. Telephone survey 6. Interviews or discussion groups 7. Other? 5) Please describe how your strategy would work. Feel free to include drawings, diagrams, anything you want! You are almost done! Please move on to the post deliberation questionnaire. 234 Post-deliberation questionnaire Hi, and thank you for participating in this workshop. Before you leave for the day, I would like you to answer a few final questions. Remember, like the first questionnaire, THIS IS NOT A TEST!!!! This is your opportunity to revise each of your responses from the beginning of the workshop. I know some of these questions may still be difficult to answer but please do your best to answer each one. Once again, it is perfectly okay to guess! I don't expect you to know all the answers. I am using this questionnaire to determine how much you have learned and how your opinions have changed over the course of the workshop. 235 Please take your time to respond to the following questions. These are the same questions that I asked before the workshop. 1) Have you ever heard the term pre-implantation genetic diagnosis otherwise known as PGD? (Please circle one) YES NO 2) Please provide a description of PGD. Remember, it's okay to guess! You will have the opportunity to revise all your responses at the end of the workshop. 3) Where have you heard about PGD? (Feel free to circle more than one response) a) The television news b) The newspaper c) The Internet d) In a magazine e) In a classroom from a teacher/professor f) From a physician/nurse/genetic counsellor g) From my friends/family h) I have never heard of PGD before today i) Other (please describe) 4) As far as you know, what are the applications of PGD? (Feel free to circle more than one response) a) PGD is used to diagnose genetic conditions or chromosomal abnormalities. b) PGD is used to help infertile couples achieve pregnancy. c) PGD is used to determine the sex of an embryo. d) PGD is used to diagnose abnormalities during pregnancy. e) PGD is used to match tissue to create compatible cells for an existing child. f) Don't know 5) Who regulates PGD in Canada? (Feel free to circle more than one response) a) An agency called Assisted Human Reproduction Canada that reports to the Minister of Health. b) No agency takes responsibility for PGD in Canada. c) Each province is responsible for creating its own agency that must report to the Minister of Health. d) Before 2004, there were no regulations, standards, or professional guidelines for the use of PGD in Canada. e) There have been regulations, standards, & professional guidelines for the use of PGD in Canada since the birth of the first test-tube baby, Louise Brown in 1978. f) Don't know 6) Is PGD covered by the Canadian health care system? (Please circle one) YES NO Don't know 7) Have you ever heard the term in vitro fertilization otherwise known as IVF? (Please circle one) YES NO 237 8) Please provide a description of IVF. Remember, it's okay to guess! You will have the opportunity to revise all your responses at the end of the workshop. 9) Where have you heard about IVF? (Feel free to circle more than one response) a) The television news b) The newspaper c) The Internet d) In a magazine e) In a classroom from a teacher/professor f) From a physician/nurse/genetic counsellor g) From my friends/family h) I have never heard of IVF before today i) Other (please describe) 10) Is there a relationship between PGD and IVF? (Please circle one) YES NO Don't know 11) If a woman wants to use PGD, must she also undergo IVF? (Please circle one) YES NO Don't know 238 12) If a woman undergoes IVF, must she also undergo PGD? (Please circle one) YES NO Don't know 13) What risks and benefits might PGD pose to a woman? Please list as many as you can. Point form is okay. Risks Benefits 14) What risks and benefits might PGD pose to an embryo, foetus, or child? Please list as many as you can. Point form is okay. Risks Benefits 15) What social risks and benefits might PGD pose? In other words, what risks and benefits might PGD pose to society, groups, or individuals or those other than families, doctors, etc? Please list as many as you can. Point form is okay. Risks Benefits 16) What are some of the reasons for why people may use PGD? (Feel free to circle more than one response) a) To help infertile people achieve pregnancy. b) To have a baby that could provide tissue for a pre-existing sick child. c) To have a baby of a particular sex. d) To have a baby that is NOT affected by a particular condition/disability. e) To avoid aborting an "abnormal" fetus. f) To have their own biological child. g) Adopting a child is too expensive or takes too long, h) To detect abnormalities during pregnancy. 17) In Canada, is it legal to use PGD to select a baby of a particular sex for "family balancing"? (Please circle one) YES NO Don't know 18) In Canada, can anyone get PGD? (Please circle one) YES NO Don't know 19) Is PGD common in Canada? (Please circle one) YES NO Don't know 20) Who (what individuals or groups) might be concerned about the use of PGD? (Please list below) 240 21) Why might the individuals or groups you mentioned in question 19 be concerned about PGD? 22) How confident are you in your knowledge of the risks and benefits of PGD? (Please circle one) 1 2 3 4 Not at all A little Somewhat Very 23) At this time, what would you say are your biggest hopes for PGD (if any)? For example, I hope that PGD will 24) At this time, what would you say are your biggest concerns regarding PGD (if any)? For example, I fear that PGD will 25) Please circle the response that best describes your opinion about PGD at this time. (Please circle one) a) I am strongly supportive of PGD b) I am somewhat supportive of PGD c) I am somewhat opposed to PGD d) I am strongly opposed to PGD e) I have no opinion about PGD f) I don't care about PGD Please use the space below to explain your above response. 26) Who would likely be the most impacted by the use of PGD? (Feel free to circle more than one response) a) Families b) Women c) Children d) People with disabilities e) Disability activists f) Families affected by disability g) Religious groups h) All Canadians i) Others (please name) j) Don't know 242 Now I would like to ask you a few general questions about the risk communication strategies I used during this workshop. For the following questions, please check off V only one response. Parts of the workshop a. The questionnaires. b. The private voting. c. The presentation of FAQ' s about PGD at the beginning. d. The audio recordings and charts that discussed the technical risks of PGD to a woman and to the embryo, fetus, and child. e. The audio recording and charts that discussed the social risks of PGD. f. The videos and news articles that talked about different types of conditions. g. The group discussion. 1) What was your favourite part of the workshop? 2) What was your least favourite part of the workshop 243 For the following questions, please check off V only one response Of all the information provided to you today, which was the... Information types m. The presentation of FAQ's about PGD n. The audio recordings that discussed the technical risks of PGD to a woman-Rana and Ted's decision. o. The charts that discussed the technical risks of PGD to a woman-Rana and Ted's decision. p. The audio recording that discussed the technical risks of PGD to the embryo, fetus, and child-Rana and Ted's decision. q. The charts that discussed the technical risks of PGD to the embryo, fetus, and child-Rana and Ted's decision. r. The audio recording that discussed the social risks of PGD. s. The charts that discussed the social risks of PGD. t. The videos that talked about different types of conditions. u. The news articles that talked about different types of conditions v. The small group discussion w. The opinions that I held before the workshop. x. Other 3) most important when considering your private vote? 4) least important when considering your private vote? 5) How confidence do you feel about your private vote today? (Please circle one) 1 2 3 4 Not at all A little Somewhat Very 244 For the following questions, please check off V only one response. 4) Of all the risk communication strategies I used today, which was ... Strategy a. The presentation of FAQ's about PGD b. The audio recordings that discussed the technical risks of PGD to a woman or the embryo, fetus, and child -Rana and Ted's decision c. The charts that discussed the technical risks of PGD to a woman or the embryo, fetus, and child -Rana and Ted's decision. d. The audio recording that discussed the social risks of PGD e. The charts that discussed the social risks of PGD. f. The videos that talked about different types of conditions. g. The news articles that talked about different types of conditions. 6) your favourite? 7) your least favourite? 8) the most trustworthy? 9) the least trustworthy? 245 Please answer the following questions on a scale of 1-4, where 4 is very and 1 is not at all. Check off V only one response. Questions 10) How stressful would you say that the workshop was to you? 11) How enjoyable would you say that the workshop was to you? 12) How difficult would you say that the workshop was for you? 13) How worthwhile was the workshop? 14) How interesting was the workshop? 1 not at all 2 a little 3 somewhat 4 very 15) On a scale of 1-4 where 4 is definitely likely and 1 is not at all likely, how likely is it that you would attend a workshop like this again in the future? (Please circle one) 1 2 3 4 Not at all Somewhat Very Definitely 16) In your opinion, is it sensible for risk communicators to rely on a process like this one when determining risk communication strategies for PGD? (Please circle one) YES NO Don't know 17) Please use the space below to describe any suggestions you may have for ways to improve this workshop or any final thoughts you may have about this workshop. Last but not least, please take a few moments to answer the following questions. Your responses will be used for statistical purposes only. Age (Please circle one) 19-21 22-30 31-40 41-50 51-60 61-70 71-80 Cultural background (please describe) Gender Occupation Marital status (please circle one) Married In a relationship Single Divorced Widow Highest level of education (Please circle one) 1. Some high school 2. High school 3. On the j ob training 4. College or some college 5. Trade school/apprenticing or some 6. University degree or some university 7. Graduate school or some Do you have children? (Please circle one) YES NO If YES, how many? You average household income (Please circle one) 1. Under $20,000 2. $21-30,000 3. $31-40,000 4. $41-50,000 5. $51-60,000 6. $61-70,000 7. $71-80,000 247 8. Over $80,000 M,„****,„*,„****„**M******«**********«**********************END 248 Appendix H. Risk communication workshop scripts for audio recordings (based on Appendix I charts) 1) Audio recording of Rana and Ted's decision: Technical risks of pre-implantation genetic diagnosis (PGD) to a woman (compliment to chart 1) Rana and Ted are a couple who really want to have their own biological child. Although they could have a baby naturally they really want to use PGD. You see, Ted is at risk for Huntington's Disease. Ted was profoundly affected by watching his father suffer with the condition and they do not want to pass this disorder on to their child. They have just found out that they are eligible for PGD and have been referred to a fertility centre to receive information about the risks and benefits of PGD from a genetic counsellor. Judy, the counsellor, tells them that PGD poses a number of risks to a woman's body while she is trying to get pregnant. She explains that they can only use PGD if they also undergo IVF. She notes that Rana and/or Ted may experience stress associated with undergoing IVF when they are not actually infertile. She explains that the short term risks of PGD are likely the same as IVF. For example, hormone injections that stimulate ovulation and the egg retrieval process during IVF are both physically invasive and will pose some risk to Rana's body. There are also other short term risks that Rana and Ted must consider such as Ovarian Hyperstimulation Syndrome (OHSS) and infections. OHSS can range from mild to moderate forms in up to 1 in 5 women and in rare extreme cases can lead to death. Judy points out that Rana is over 40 and she is therefore much less likely to have a successful pregnancy through IVF. Judy also notes that all of these risks are to Rana's body even though she is not the one affected by Huntington's Disease. She asks if this will lead to problems for Rana or Ted relationship if Rana feels resentful or Ted feels guilty. Judy says that it is something they must think about before they make their decision. She also states that some of the risks and benefits of PGD are psychological in nature. A significant psychological risk of IVF for some is that the woman may not become pregnant. Psychological risks of PGD must also be compared to those of abortions after diagnosis. Judy explains that some couples choose PGD because affected embryos can be avoided before pregnancy occurs. For religious or other reasons, they do not want to have a baby naturally, find out that it is affected, and then have an abortion before it is born. Judy says that, she understands that Rana and Ted want to use PGD to avoid passing on Huntington's Disease to their child. However, she also adds that PGD cannot guarantee a "normal" embryo, fetus, or child. There is a potential for misdiagnosis AND a potential for the child to acquire conditions that were NOT screened out through PGD. Judy can see that Rana and Ted are looking a little discouraged so she says that there are also some benefits associated with PGD and IVF. She explains that IVF lowers the risk of ectopic pregnancies or pregnancies that occur outside the uterus, which can endanger a woman life, and cause severe bleeding and pain. However the risk of ectopic pregnancy in natural pregnancies is really not a significant risk and is less than 2%. 249 Judy now moves on to risks that may occur during a pregnancy. Judy explains that one significant risks of PGD and IVF is multiple births. Typically three to six embryos are used during IVF and as a result, about one third of IVF live births are multiple in nature. However, she is quick to point out that multiple birth are even more likely for women who have ingested fertility drugs instead of undergoing IVF. Risks that multiple births pose to women include premature labour, difficult deliveries, physical and emotional stress. However Judy says the trend is now to conduct one embryo transfers during IVF for best candidates. Although this eliminates the risk of multiple births it also reduces the chance of becoming pregnant because fewer embryos are implanted. Judy states that if they do in fact achieve pregnancy, Rana and Ted will also have to consider if they want to have additional tests throughout that pregnancy. She explains that after PGD has occurred and the embryo has been implanted into the uterus, they may be advised to have additional tests such as amniocentesis or CVS to detect misdiagnosis and health of the fetus over time. Judy explains that some couples avoid these tests and their associated risks to the pregnancy that come with them and decide to trust PGD. However, she adds that receiving a relief diagnosis could help them acquire assistance from the system and prepare for the future. Lastly, Judy explains that PGD may pose some long term risks to Rana's body. Some argue that there is a link between IVF and breast and ovarian cancer but Judy assures the couple that this appears to be a very soft link that disappears in studies that include large samples. Some researchers also point out that additional long term studies are required before we can state that IVF or PGD have a small long term impact on patients. Rana and Ted tell Judy that they understand that there are many risks to consider. Although they are still pretty sure that they would like to use PGD at this clinic, they will definitely need some more time to think about it. 2) Audio recording of Rana and Ted's decision: Technical risks of pre-implantation genetic diagnosis (PGD) to PGD poses to the embryo, foetus, and child (compliment to chart 2) After discussing the risks that PGD poses to a woman, Judy states that Rana and Ted should also consider the risks that PGD may pose to an embryo, foetus, and child. Judy begins by mentioning again that IVF lowers the risk of ectopic pregnancies or pregnancies that occur outside the uterus. She explains that avoiding an ectopic pregnancy is good for Rana but it is also good for the fetus because ectopic pregnancies can lead to the death of the fetus. However she notes again that the background risk of ectopic pregnancy in natural pregnancies is less than 2%. Judy goes on to say that IVF poses less risk to an embryo than prenatal tests like amniocentesis or CVS because those prenatal tests can lead to miscarriages. She adds that experts in her field state that the risks associated with technologies such as CVS, IVF, and PGD are all quite small. That said, there is an additional minor increase associated with them. For IVF, PGD, and CVS, the additional risk for the pregnancy is 1%. However with CVS remember there is also the risk that the pregnancy may fail. Amniocentesis has an additional risk of .5%. Lastly, Judy explains that these risks are for each fetus, so in a multiple birth situation, there would bean increased risk of 1% plus 1% or 1% for each fetus. At this point Judy returns to the subject of risk associated with multiple births. She tells them that the risk of multiple births during IVF is linked to the age of mother, the number of embryos implanted, and the quality of those embryos. However, multiple births are even more likely for 250 women who have ingested fertility drugs instead of undergoing IVF. She goes on to say that on average, 3-six embryos are used for IVF and as a result, about one third of IVF live births are multiple in nature. Judy notes that these risks can be lowered by implanting fewer embryos. Risks to fetuses and children who are twins, triplets, etc. include increased risks of neonatal mortality, SIDS, birth defects, developmental disabilities, and child abuse. Judy also says that children of multiple births are at risk of being born prematurely. Some will do well, others will not. If born prematurely, the baby may experience short term risks such as difficulty breathing, undeveloped organs, and death or long term risks such as gastrointestinal issues, cerebral palsy, developmental delays, or chronic lung disease. Judy explains once again that PGD cannot guarantee a "normal" embryo. There are a series of technical challenges inherent to PGD that "can lead to a misdiagnosis of the embryo" and result in an "abnormal" fetus and child. For example, testing occurs under strict time constraints and in addition, only small amounts of DNA are examined. Also, there is the potential for the child to acquire conditions that were NOT screened out through PGD. In addition, the techniques used during IVF can accidentally "turn on" or "turn off' genes while culturing cells. Judy explains that this may increase risks of disorders such as Angelman Syndrome (AS) or Prader-Willi syndrome (PWS). Judy says that the long term physical health impacts of PGD are likely similar to IVF, which are known. Children born through IVF have twice as high a risk of major birth defects (cardiovascular, chromosomal ranging from 3 or 4—8.5%) and are more likely to have multiple major defects when compared to naturally conceived infants (ranging from 2 or 3-4%). That said, the probability that these problems will occur is low (as shown by percentages), although they may be significant. Children born from women who undergo PGD may also be subjected to psychological risks or benefits, for example, if they were born as a savior sibling to create compatible cells for an existing child. Rana and Ted tell Judy that they understand that there are many risks to consider. Although they are still pretty sure that they would like to use PGD at this clinic, they will definitely need some more time to think about it. 3) Audio recording of societal and economic factors that may inform a decision to pursue pre-implantation genetic diagnosis (PGD) (compliment to charts 3 and 4) Sadly, there have been many documented instances of unethical health care practices and abuses throughout the world. It is quite easy to locate cases that discuss experiments performed on incarcerated or institutionalized adults and even children without their consent. For example, during World War II Jewish people and others were subjected to horrible experiments against their will. But these cases also occur right here in North America. In the 1930's, Black men included in the "Tuskegee Study" were left to suffer with untreated Syphilis for decades. These, and other similar events, have lead to a shift in how people are treated within the health care system. Polices such as the Nuremberg Code (established after the Nazi atrocities) have been created to protect and inform research subjects. Where we once trusted our doctors to make medical decisions on our behalf, we now demand to be involved in these decisions. Most of us our interested in managing our risks and are not prepared to "take it as it comes". We may even sometimes think of preferences as if they are risks. 251 We also live in a country that sometimes discriminates against people with disabilities. Although we make our world more liveable for people with disabilities by modifying infrastructures (for example, building ramps), or by opening up employment opportunities, we could do more. We could even criminalize discrimination if we wanted to. Some argue that our relaxed attitude towards this type of discrimination has led us to view discrimination as a normal everyday thing. Perhaps our desire to control health related decisions coupled with our decision to ignore certain types of discrimination is what has led us to place the responsibility for disability mostly on families instead of the community or society. But what does this have to do with PGD? Most of us would likely agree that there can be great social and biological pressure to have babies. But not all of us can have our own biological healthy children. Some of us may be infertile, have difficulty conceiving, experience miscarriages, or even infant deaths. And some of us who are fertile, may have a genetic condition running in our family that makes it almost impossible to have our own biological healthy children. People in this situation may feel pressure to use technological solutions like PGD to solve their fertility dilemmas and may fail to consider other alternatives such as adoption. But is adoption really a practical alternative for Canadians? Public adoption, or adoption through a government agency, for an infant with no "special needs" may take up to 8 years. Private adoption may take less time but can cost anywhere from $10,000-$ 18,000. Costs for international adoption may also take less time but are also costly varying from $15,000 in Haiti, to $30,000 in Guatemala, and as high as $50,000 in the US. So families who are at risk for certain conditions and want to have a biological child may decide to use PGD. However PGD technology can only be used to avoid known conditions. Our ability to detect conditions is limited by our current genetic knowledge, although this is likely to expand over time. In addition, costs for PGD are not covered by the Canadian health care system. In fact, the regulatory system for PGD in Canada is currently in flux so we don't really know what it is going to look like in the future. Yet some will argue that despite all these complications, each of us is entitled to control our own reproduction because we are entitled to control our own bodies. In addition, some also believe that parents have the "right" to a healthy child. However, there are other issues to consider. Risks associated with PGD don't just affect the mother, father, and child, they can also affect others citizens. Some insist that as new technologies that screen and perform diagnostic tests on fetuses are made widely available and used with increased frequency, these procedures may become "standard practice" and therefore difficult for families to refuse. Couples may not understand that they have the option of refusing tests or screens and may not understand the range of consequences associated with having these procedures. However, others are quick to point out that currently, most people are not eligible for PGD (as it has a high barrier for entry) and additional tests are not an option for those who oppose abortion. In other words, the reason some couples choose PGD in the first place is because it happens before pregnancy occurs. They do not want to use prenatal tests like CVS or amniocentesis and then abort affected fetuses. But let's suppose for a moment that technologies like PGD were to become a part of standard medical care. When medical technologies become a standard part of care, we sometimes forget what their real purpose is and the consequences of the procedures can be masked. Ask yourself, is an ultrasound a baby photo or a genetic screen? Why do people hang them on the walls of their office? Individuals who have ultrasounds and forget that they are actually genetic screens may be 252 surprised by positive test results or may falsely believe that problems discovered about their fetus can be addressed. In addition, if PGD becomes a standard part of care during pregnancy, will this mean that more pregnancies will be treated as potentially problematic? That the pregnancy experience will be viewed as a medical procedure instead of as a natural process? And would this be a bad thing or not? Also, what happens to the people who receive all this additional information about their embryos and fetus? Some women may experience long or short term anxiety associated with the additional knowledge and choices that screens and tests provide. On the other hand, some people are excited about PGD because with it, unlike abortions, one does not have to be concerned about the moral status of the fetus. PGD occurs before a woman is pregnant so it pre-empts moral dilemmas associated with having an abortion after a positive prenatal diagnosis. Also, when compared to abortion, some view it as less harmful to the woman's body, and less psychologically and emotionally stressful. When compared to prenatal diagnosis however, the risks to the embryo are greater because with PGD, affected embryos are destroyed or donated to stem cell research. In other words, embryos are more likely to live after a prenatal test than during PGD. Others are concerned about PGD because they believe it is a form of eugenics, which is defined as selectively breeding humans or getting rid of types of people like the Nazi's did during World War II. Some are concerned that PGD is "playing God" with human life. Some are asking about the moral status of embryo. Does an embryo have a "right to life"? Should all life be considered "a gift" no matter what? Ethical and other concerns surrounding PGD may lead others to judge those who use this technology. Also, those affected by a condition may resent others who get to have "normal" children. Some argue that users of PGD technology (mostly women) may have too much power because they are deciding who and what types of people should enter society. However, others are quick to point out that genetics are only one aspect of a person. In other words, defining types of people purely by their genetics is misleading. But we should be careful to note that genetic tests are unlike other types of medical procedures. Genetic tests are unique in that they will sometimes reveal information about the biological family, potentially without their consent (such as a biological father). However, there is a very small risk that this will occur with PGD. Right now PGD is rarely used in Canada but suppose it were to become a standard part of prenatal care and used frequently. Will Canadians experience social pressure to submit to such technologies and remain consumers of genetic tests throughout their pregnancies? Will PGD reduce our society's overall compassion for people with disabilities? Will PGD impact how we view people with disabilities? Will we become less tolerant or try less for adults with disabilities who are already here? That said, who will volunteer to raise children with disabilities so that society can be diverse? In any case, PGD is likely going to alter a couples' decision context regarding the role of the parent and the meaning of a pregnancy and may place additional burdens on users of PGD technology. 253 Appendix I. Risk communication workshop news articles The Vow (this is only a portion of the article) When Dave Kendall promised to love Diana 'in sickness and in health,' he meant it The Washington Post By Liza Mundy, Sunday, March 9, 2008; W10 SHE MIGHT HAVE SEEN A SHIMMER ON THE WATER, OR SOMETHING ELSE COULD HAVE TRIGGERED HER REACTION. All Dave Kendall knew was that one minute he was sitting in the stern of his fishing boat, steering, during an excursion on the Occoquan Reservoir about a decade ago. The next minute Diana, his wife of more than 20 years, had fallen from her seat near the bow. She was on the floor of the boat, limbs flailing in what looked like a violent seizure. Diana had had fainting spells before, but doctors had not determined the cause. They didn't this time, either. Yet there were other things that now seemed wrong with Diana: She was having bizarre reactions to music, which made her head hurt so acutely that she would sometimes bolt out of church. One day, she came home from work at Fort Belvoir saying she no longer remembered how to use Excel. Dave, giving her a refresher session, saw that she was losing her focus. And she was gaining weight. Diana, who had always been trim and active, began eating junk food and sitting for days, crying. Dave took her to more doctors, including a neurologist who said he wanted to run a test for something called Huntington's disease. He didn't tell them much about it — just that they had better hope the test came back negative. Dave felt reasonably optimistic while they were waiting for the results. Doing some Internet research, he learned that Huntington's is a hereditary disorder, and they didn't know of anyone in Diana's family who had it. The life the Kendalls had built together ~ decades of hard work that had allowed them to leave behind impoverished childhoods in rural Virginia for the middle-class comforts of suburban Washington — felt secure. But when they returned to the neurologist on August 6,2000, they learned that the future wasn't secure at all. The test had come back positive. "This is really bad news, isn't it?" asked Diana, turning to Dave. She was strikingly calm and dry-eyed. It was the beginning of a transformation in her; deterioration, yes, but also acceptance and what can only be described as grace. It was also the moment ~ Dave realizes in retrospect — when she entrusted herself to his care. It was the end of married life as they had known it, the beginning of marriage in an entirely new form. Same contract, radically altered circumstances. "She put her life in my hands," he says. Diana was 48; Dave was 51. By then, they had been living with a fatal neurodegenerative disease for several years. Parts of Diana's brain were dying, an inexorable process caused 254 by a single abnormal gene, a glitch that affects everything: thought, mood, movement. After the diagnosis, Dave began keeping a journal to record Diana's rapid decline. "She was unable to walk in grocery store to complete shopping ~ sat on bench," he wrote in December 2001, adding, "I start peeling apples, potatoes, etc., to avoid accidents with knives." In February 2002: "Walks like a drunk — falls against wall." In August 2002: "Cannot find hymn number in book. Leans while sitting." In October 2003: "Cannot put her shoes on" and "could not get out of chair." In April 2005: "Could not subtract 6 from 1977." In November 2005: "This is the first Thanksgiving in which I had to spoon feed her." In May 2007: "Di has no ability to read anything or create a thought. I do it and ask her if she likes the thought." More than seven years after Dave began keeping his journal, Diana has lost her balance and her ability to stand unaided. She can count forward from one to 10, but not backward from 10 to one. She is immune to boredom and has no sense of the passage of time. Yet she is still capable of teasing Dave, as well as expressing a deep appreciation of what her disease has cost her husband. Three years ago, he asked her what her New Year's resolutions were. "To read my Bible more," she said, "and to be less of a burden to you." 255 BBC News Story 'I wouldn't have to fear for my child' A report from the Human Fertilisation and Embryology Authority is to recommend extending embryo checks to look for cancer genes, which make people highly susceptible to certain forms of the disease. Karin Cohn, from Middlesex, found out last year that she carries the BRCAl gene - which greatly increases her risk of developing breast or ovarian cancer. Both her sister and mother have had breast cancer, and there is a long family history of both conditions. "It gives people an option not to have to go through the stress and the anxiety that I've been through." -Karin Cohn Karin, who is in her 30s, has already undergone a preventative double mastectomy to remove her breasts, and she also plans to have her ovaries removed to take away her risk of cancer. She said she fully supported the proposal to use pre-implantation genetic diagnosis to select embryos which did not carry the cancer gene. Karin told the BBC: "It's fantastic. It gives people an option not to have to go through the stress and the anxiety that I've been through, making my decision. "If I had had the option, I would have done it. And I would continue to do it until I got a clear embryo. "It would mean I wouldn't have to worry about my child in the future." 'I feel guilty' She said she worried for her four-year-old daughter, Sophie, who has a 50/50 chance of also carrying the BRCAl gene. "We've been advised to wait until later to have her tested. But it is a worry." Karin's mother Pat Gibor said she feels responsible for passing the BRCAl gene on to her daughters. "I feel guilty. Rationally I know I shouldn't, but emotionally I do." And she said she too backed the idea of using PGD to select embryos which did not have the cancer susceptibility gene. "They do step in in other instances, so if it can be done if embryos can be tested for cystic fibrosis or Huntington's disease, I don't see why they can't be tested for the BRCAl gene." 256 Appendix J. Risk communication workshop links to videos recommended by experts 1) Hyperlink to "Belong" Video from Down's Syndrome Society http://www.onetmemedia.com/otm_site/view_shared?p=50a08d8c72399d9fd791.c7& skin id=601&utm source=otm&utm medium=email 2) Hyperlink to public service announcement video from National Tay-Sachs & Allied Diseases Association http://www.ntsad.org/ 257 Appendix K. Risk communication workshop database of results