Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Interpreting and making sense of uninformative results of testing for BRCA 1 and BRCA 2 cancer gene mutations Maheu, Christine 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2005-104147.pdf [ 18.59MB ]
JSON: 831-1.0092352.json
JSON-LD: 831-1.0092352-ld.json
RDF/XML (Pretty): 831-1.0092352-rdf.xml
RDF/JSON: 831-1.0092352-rdf.json
Turtle: 831-1.0092352-turtle.txt
N-Triples: 831-1.0092352-rdf-ntriples.txt
Original Record: 831-1.0092352-source.json
Full Text

Full Text

INTERPRETING AND MAKING SENSE OF UNINFORMATIVE RESULTS OF TESTING FOR BRCA1 AND BRCA2 CANCER GENE MUTATIONS by Christine Maheu R.N., John Abbott College, 232, 1987 B.S.N., L'universite de Montreal, 1994 M.S.N., L'universite de Montreal, 1997 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Nursing THE UNIVERSITY OF BRITISH COLUMBIA March 2005 © Christine Maheu, 2005 A B S T R A C T I N T E R P R E T I N G A N D M A K I N G SENSE OF U N I N F O R M A T I V E R E S U L T S OF T E S T I N G F O R BRCA1 A N D BRCA2 C A N C E R G E N E M U T A T I O N S Research suggests that a significant proportion of individuals from families at risk of hereditary breast and ovarian cancer will be found not to have a detectable mutation in their BRCA1 or BRCA2 cancer genes. Although the interpretation of genetic test results is relatively straightforward in families where a mutation has already been identified, little is known about how people who have had breast and/or ovarian cancer in the past as well as a family history of cancer considered at risk for HBOC interpret and make sense of test results concluding that no detectable mutation has been found. This problem is further compounded when they are told that such genetic test results do not completely rule out an inherited mutation because of their strong family history of the disease. While the clinical and research literature refers to these results as uninformative or inconclusive, this study shows that clients' interpretations are much more complex. To date, few studies have focused on affected individuals from families at risk of HBOC who receive uninformative genetic test results. We therefore have little knowledge of how these individuals interpret and make sense of such results and how these results affect their everyday lives, health and illness experiences. This dissertation addresses these lacunae by using an interpretive description approach to examine clients' experiences of genetic testing. Qualitative, in-depth interviews were conducted with 21 affected individuals with a family history of cancer considered at risk for HBOC who received genetic testing and 15 family members. The interview data show that living with a personal and family history of breast and/or ovarian cancer plays an important role in interpreting and then making sense of their genetic test results and in one's perception of probably having an inherited mutation for H B O C . Thirteen generic structures were found to organise beliefs towards the making sense process of interpreting their genetic test results while three types of interpretation of the test results were derived from the participants' accounts. The categories of interpretation are seeing results as: a confirmation of their mutation status, ambiguity regarding their mutation status, and refutation of being a mutation carrier. On the basis of these generic structures and the three types of interpretation, it is possible to see a retrospective narrative of causal reasoning of having a probable inherited mutation that builds upon recognition of a strong family history with breast and/or ovarian cancer. This 7-stage process evolves with changes in people's everyday lives, health and illness experiences. The impact of receiving uninformative test results for BRCA1 and BRCA2 mutations on the lives of affected individuals and their family members requires further examination. We need to understand how such results affect cancer and genetic risk perception and potentially contribute to clients' viewing themselves at chronic risk of cancer and of an inherited mutation. Further investigation is also needed to determine how uncertain genetic risk information is shared among and used by family members. This dissertation offers recommendations to ameliorate the experience of individuals who receive uninformative genetic test results, to improve genetic testing services, and to enhance the genetic knowledge of nurses and their clients. IV T A B L E OF C O N T E N T S Abstract i i Table of Contents iv List of Tables vii List of Figure viii Acknowledgements ix C H A P T E R O N E : Situating the Study 1 Background to the Problem 3 Purpose of the Study 5 The Research Questions 7 Significance of the Research 7 Definitions of Terms 9 Organisation of the Thesis 10 C H A P T E R T W O : Review of the Literature 12 Hereditary Breast and Ovarian Cancer 13 Challenges with Genetic Testing and Counselling 16 Genetic Counselling for Hereditary Breast and Ovarian Cancer 22 Experiencing the process of genetic testing and counselling 27 Interest in Test Information 37 Intended Use of Test Information 40 Psychosocial Issues 44 Psychological impact of having a family history of breast and/or ovarian cancer 44 Benefits and limitations of genetic testing for breast and ovarian cancer susceptibility 45 Perceived risks 50 Nursing implications of genetics 50 Summary 52 C H A P T E R T H R E E : Research Design and Implementation 53 Interpretive Description of Health and Illness Experiences 54 Theoretical framework 54 Research Design • 58 Implementing the Study 59 Recruitment of participants 60 Participant characteristics 63 Sample 66 Data sources 70 Data analysis 73 Standards and credibility measures 77 Ethical considerations 84 Summary 88 C H A P T E R F O U R : The Buildup to Genetic Testing 89 Implications of Living with a Family History of Breast and/or Ovarian Cancer 89 Encountering Genetic Testing 115 V Hearing of the availability of BRCA1 and BRCA2 mutation testing 115 Motivations for having genetic testing 121 C H A P T E R F I V E : Interpreting and Making sense of Genetic Test Results 129 Types of Interpretation Given to Their Test Results 129 The Making Sense Experience 134 Doubt and uncertainty about the outcome of results 138 Too many to be coincidental 149 Age at cancer diagnosis 159 Attributing unique features to their mutation 162 Influence of time 174 Resemblances among individuals with cancer 177 Presence of children 181 Conclusion: Types of Interpretation, Integration of Generic Structures, and the Retrospective Narrative of Causal Reasoning 182 C H A P T E R SIX: What Participants Concluded from their Genetic Testing Experience.. 186 Emotional Reaction to Genetic Testing Results 186 Those who found relief 188 Disappointment and frustration with the results 197 Informative Value of Testing Results for Health-Related Decisions 201 Benefits and Lack of Them 209 Potential Harms of Genetic Testing Foreseen by Participants 219 Risk of discrimination 219 Risk communication issues 225 Views Unique to the Uninformative Group 231 Conclusion 233 C H A P T E R S E V E N : Situating the Study Findings to the Broader Clinical Context 237 Living with Uncertainty 238 Use of the Generic Structures as Guides to Clinical Genetic Testing Practice 242 The Now What Experience of Genetic Testing? 248 Limitations of Study Findings 256 Summary 259 C H A P T E R E I G H T : Summary of Findings and Recommendations for Clinical Practice, Future Research and Policy Development 260 Summary of Findings 260 Recommendations for Clinical Practice, Education, Research and Policy Development 267 Communicating uninformative genetic testing results 267 Providing support with uninformative genetic testing results 270 Directions for education 273 Directions for further research 275 Directions for policy development 278 R E F E R E N C E S 282 Appendix 1: Letter Received by the Study Participants Informing Them of their Genetic Test Results 297 Appendix 2: Preliminary Interview Schedule 300 Appendix 3: Letter of Support From the BCCA Hereditary Cancer Program 305 Appendix 4: Participant Information and Consent Form 307 Appendix 5: Family Genograms of the 21 Primary Participants 311 Appendix 6: Sample: Field Notes 323 Appendix 7: Letter of Invitation to Participate 326 Appendix 8: Sample Initial Analysis of key statements 328 Appendix 9: Sample Ongoing Data Analysis 331 Appendix 10: Demographic Questionnaire 333 Appendix 11: Transcript Segments From a Few Interviews 336 Appendix 12: Database Record with Dummy Data 338 L I S T O F T A B L E S Table 1: Demographic Profile of Primary Participants 64 Table 2: Relationship of Secondary Participants (SP) to Primary Participants (PP) 65 Table 3: Eligibility Criteria for Hereditary Breast and Ovarian Cancer Risk Assessment 91 Table 4: Types of Interpretation Derived from the Participants' Accounts to Uninformative Results 130 vm LIST OF FIGURES gure 1: Pedigree Representing PP21, Sherry's Family History with Breast and/or Ovarian Cancer 92 ix Acknowledgements Bringing to an end my doctoral studies through the completion of my dissertation barely begins to convey the extent of the incredible experience I gained on this journey. Not only was I guided by a renowned researcher, Dr. Sally Thorne, but I also had the good fortune to be invited in to wonderful circles of academics, clinicians, and families, who all contributed immensely to memories of being a doctoral student, that will last forever. First, I would like to express deep gratitude to all of the participants in this study who opened up their homes and their lives so that I could learn about their experiences of genetic testing. Their contribution to the success of this research is without end and, although I cannot name them, I offer my heartfelt thanks for their trust that I could produce knowledge from their shared stories that would serve society in the future. Second, I would like to acknowledge the Fonds de recherche en sante du Quebec, the University of British Columbia Faculty of Graduate Studies, the UBC School of Nursing, the X i Eta Chapter of Sigma Theta Tau, and the Canadian Nurses Foundation/Canadian Association of Nurses in Oncology/AstraZeneca for funding to support this research. Third, I would like to acknowledge the outstanding support and trust I received from my research supervisor Dr. Sally Thorne. She, along with my Dissertation Supervisory Committee, provided me with the intellectual support I needed to complete my dissertation according to my own standards. I thank Dr. Michael Burgess for allowing me to join his group of researchers at the W. Maurice Young Centre for Applied Ethics at the University of British Columbia. 1 thank Dr. Burgess for his guidance. I extend thanks as well to Dr. William McKellin for intellectually challenging me. I am also grateful to Dr. JoAnn Perry and Professor Mary Reidy for their thoughtful comments and questions. Fourth, I also would like to acknowledge the welcome reception I received from the members of the Hereditary Cancer Program at the BC Cancer Agency, in particular Ms. Mary McCullum, Ms. Karen Panabaker, Ms. Pardeep Kaurah, Ms. Jenna Scott, Dr. Doug Horsman, and Dr. Charmaine Kim-Sing. By inviting me to participate in their meetings observe, genetic counselling sessions and the high-risk clinic, and in conversations, they enabled me to witness the real world of genetic testing. The experience was invaluable to my overall understanding. Over the years, this project was also enriched by many conversations I had with friends and colleagues. I wish to thank the scholars working at the W. Maurice Young Centre for Applied Ethics, with whom I attended weekly meetings and discussed the many issues related to qualitative research in the area of genetics: Dr. Bryn William-Jones, Dr. Lori d'Agincourt-Canning, Dr. Sue Cox, Dr. Andre Smith, Dr. Fern Brunger, and Dr. Jessica Easton. I would like to extend special thanks to my compatriots in the Three Musketeers Group - Dr. Louise Racine and Dr. Becky Palmer - for the laughter and worries they shared with me. With my dear friend Becky, I also shared the new experience of motherhood and the challenges it adds to one's academic life. I have been fortunate to have the support, confidence and company of my family to see me through the end of this project; thank you to my parents and siblings for always having faith in me and for your time in looking after my children so that I could peacefully complete the last corrections of my dissertation. Finally, I owe the most gratitude to my husband and children. I carried both of my children, Louca and Delphi, during high peaks in my dissertation - the data collection and writing phases - and both had to grow through stressful times. Yet, today they bless me with X their love and presence. To my husband, Francois Legare, whose partnership with me in all areas of our life meant that I never had to explain why I needed more time and space to accomplish my dissertation work. Many thanks. 1 CHAPTER ONE: Situating the Study Prior to the advent of genetic testing for hereditary breast and ovarian cancer, the combination of breast self-examination, clinical examination and mammography was the only strategy recommended to increase early detection of breast cancer (Delisle, 1997; MSSS, 1998). For families who have a strong history of breast and/or ovarian cancer, genetic testing offers an additional way of assessing the risk of developing these diseases if family members carry inherited mutations of known breast or ovarian cancer genes. Having a personal and family history of breast and/or ovarian cancer increases the risk of carrying an inherited mutation (FitzGerald et al., 1996; Newman et al., 1998; Oddoux et al., 1996). With the discovery of the BRCA1 and BRCA2 breast and ovarian cancer genes (Miki et al., 1994; Wooster et al., 1994), genetic testing for common adult-onset disorders such as breast and ovarian cancer emerged within clinical settings, either as part of research protocols or clinical service. Clinicians, researchers and patients all had high hopes about the efficacy of genetic testing in finding the etiology of breast and ovarian cancers and in identifying individuals at increased risk of developing these diseases as a result of having inherited mutations. Genetic testing for hereditary breast and ovarian cancer is not yet widely available nor offered in all provinces as it is not yet official, integrated clinical practice because of, among other reasons, patent issues. Nonetheless, there has still been considerable uptake of genetic testing for breast and ovarian cancer genes in Canada. For clinicians, one of the valuable aspects of genetic testing for BRCA1 and BRCA2 is the test's ability to identify individuals at risk of hereditary breast and/or ovarian cancer (HBOC) (Lynch et al., 1999). For individuals, genetic testing for BRCA1 and BRCA2 informs them of 2 their actual hereditary risk for breast and/or ovarian cancer and, in part, of their cancers' etiology for those with a past diagnosis because cancer is a multifactorial disease (Hallowell et al., 2002; Lerman, Seay, Balshem, & Audrain, 1995). This test also clarifies the risk of their children and families for hereditary breast and/or ovarian cancer (Armstrong et al., 2000; Lynch et al., 1999). Despite the benefits (such as knowing with certain one's genetic risk status) and risks (such as increased anxiety as a result of this certainty) there are many unresolved limitations that clinicians as well as the general population need to understand. For example, how do you support clinical decision making with the uncertain information resulting from genetic testing (Frost, Venne, Cunningham, & Gerritsen-McKane, 2004)? Uncertain information resulting from genetic testing for BRCA1 and BRCA2 is mainly of two kinds. First, a positive genetic testing result is not a cancer diagnosis but a prediction of risk of developing cancer: Even if individuals are found to carry an inherited BRCA mutation, while they have a high risk of developing breast and/or ovarian cancer, there are no certainties that they will develop the disease and, if they do, no estimate can be given as to when it could appear. With a BRCA1 or BRCA2 cancer mutation, women have a 56% to 87% lifetime risk of breast cancer and a 16% to 40% lifetime risk of ovarian cancer (Di Prospero et al., 2001; Ford, Easton, Bishop, Narod, & Goldgar, 1994). Men found to carry a BRCA1 mutation have a 16% lifetime risk of developing prostate cancer while with a BRCA2 mutation carries a 6% risk of developing breast cancer (Di Prospero et al., 2001). Children of individuals carrying an inherited mutation have a 50% chance of inheriting the mutation. Second, when individuals considered at risk for HBOC are not found to have an inherited mutation, health professionals cannot confirm their risk of cancer or genetic status. The negative test results could either be a true negative, a false negative or - because researchers have not yet identified all conferring inherited mutations of breast and ovarian cancer genes - it 3 is possible that these individuals may carry mutations in genes other than BRCA1 or BRCA2 (Pasacreta, Jacobs, & Cataldo, 2002). Considering that only 5% to 10% of all breast cancer cases are hereditary (Claus, Risch, & Thompson, 1991; McCance & Jorde, 1998) and that there is growing interest in genetic testing for BRCA1 and BRCA2, especially among individuals with a family history of cancer considered at risk for HBOC (Peshkin, DeMarco, Brogan, Lerman, & Isaacs, 2001), we can expect to see a growing pool of individuals who are likely to receive genetic test results of uncertain value such as uninformative genetic testing results. Background to the Problem Before the discovery of the BRCA1 and BRCA2 genes, individuals with a family history indicative of HBOC seeking information about their risk of developing these diseases and that of their family members could only rely on linkage analysis and on models that determine carrier probability for the genes, BRCA1 and BRCA2 such as the BRCAPRO model (Parmigiani, Berry, & Auguilar, 1998).1 Both linkage analysis and models yielded only a probability that one might carry an inherited mutation. Peshkin et al. (2001) note that, even back then, clinicians and scientists struggled with whether to give individuals the results of these analyses, in light of their uncertainty. Now that genetic testing for adult-onset-hereditary-disease has become an important part of clinical genetics practice, we see an increased demand for tests, perhaps as a result of media coverage of genetic discoveries or from referrals by health professionals. In 1996, the BC Cancer Agency and the BC Provincial Medical Genetics Program formed the Hereditary Cancer Program (HCP) to provide genetic education, counselling and testing for BRCA1 and BRCA2 to ' B R C A P R O is a model and software developed by G . Parmigiani, D . A . Berry and O. Agui lar at the Institute of Statistics and Decision Sciences, Duke University, U . S . A . The model uses a Mendel ia approach of autosomal-dominant inheritance to assess the probability of an individual to carry a B R C A 1 and B R C A 2 mutation. The model is based on the individual 's family history of breast and ovarian cancer (BayesMendelLab, 2004). high-risk individuals and families with a strong history of cancer (Williams-Jones, 2002). Although there has been an increase in the uptake of genetic testing for BRCA1 and BRCA2 and we now have an abundance of research since 1994 on the impact of the availability of this test, many unknowns and complexities still remain (Peshkin et al., 2001). The process leading up to genetic testing, receiving the results, and interpreting the results is not simple. The complexities include how to delineate eligibility criteria for testing, how to minimize negative psychosocial reactions, how to support health-care decision making in light of genetic risk values, how to interpret risk values, and how to interpret risk-descriptive statements when risk values are impossible to provide. Even with positive test results indicating the presence of a BRCA1 or BRCA2 mutation, individuals do not receive one precise quantitative cancer risk value but are given a range of the risk to develop cancer. For example, they are told that their risk of cancer will increase between percentage ranges; 56% and 87% for breast cancer (Ford et al., 1994; Struewing et al., 1997) and 22% and 44% for ovarian cancer both by age 70 (Ford et al., 1994; Whittemore, Gong, & Itnyre, 1997). Risk percentages of developing cancer also differ depending on which gene the mutation is found. In addition, these percentages are different for individuals who have already been diagnosed with breast and/or ovarian cancer where they see their risk of developing another primary cancer increased. Research in this area is ongoing. As for individuals within a confirmed hereditary breast and/or ovarian cancer family who receive negative test results for BRCA1 and BRCA2, the test results for these individuals are classified as true-negative. While they see their risk of cancer decrease, their risk is still not at general population's level because of their strong family history with cancer. 2 A primary cancer is the place where the cancer started growing. Recurrent cancer or metastases are escaped cancer cells from the primary site that spread to other parts of the body. Research has found that a high percentage of high-risk families (between 16% and 65%) do not carry detectable mutations of BRCA1 and BRCA2 (Peshkin et al., 2001).3 When not found with a detectable mutation, these families are viewed as part of the "uninformative group". Hence, the availability of genetic testing for these mutations has created a large pool of individuals in a quandary about their perceived and actual risk for breast and ovarian cancer (Iglehart et al., 1998). We are just beginning to see the results of research on this population and the complexities that may result from uninformative genetic test results. Even among those who receive positive test results, many psychosocial consequences are not well understood (Pasacreta et al., 2002). Some known responses to positive BRCA mutation test results are persistent worry, depression, anxiety, confusion, and sleep disturbance (Lynch et al., 1993). While the literature documents known responses to positive and true-negative results, such as survivor's guilt among individuals who test negative within hereditary cancer families (Lerman, Seay et al., 1995), there are still many unknowns. Providing adequate support to those who receive uninformative genetic testing results is even more challenging because of the paucity of studies. While health professionals in cancer genetics have gained knowledge in clinical practice, this knowledge is based mostly on the assumptions, beliefs, and observations of clinicians. Purpose of the Study Few studies were found that focus on individuals and families considered at risk for HBOC who receive genetic testing results other than positive or true-negative. We are just beginning to see some studies that have documented experiential knowledge of individuals' reaction to and understanding of uninformative results of genetic testing for BRCA1 and 3 According to K . Panabacker, genetic counsellor at the H C P , of the 255 individuals tested for B R C A 1 and B R C A 2 from high-risk families, less than 25% received a positive result. 6 BRCA2. Within the literature, where individuals with a strong personal and family history of breast and/or ovarian cancer have received a genetic testing result that indicates that no detectable mutations were found, their results are determined to be either uninformative or inconclusive (Frost et al., 2004; Hallowell et ah, 2002; Hallowell, Foster, Eeles, Ardern-Jones, & Watson, 2004; Peshkin et al., 2001; Schwartz et al., 2002). At the time I wrote my dissertation proposal (the beginning of 2000), very few researchers had studied the experience of women with a past cancer diagnosis who received uninformative genetic testing results for BRCA1 and BRCA2. In light of the lack of knowledge to how individuals interpret and make sense of uninformative genetic testing results for BRCA1 and BRCA2, questions for clinicians become: How do we prepare individuals who rely on us for information when there is little experiential knowledge of how individuals with a past breast and/or ovarian cancer diagnosis interpret and make sense of uninformative genetic testing results? How can we assist individuals as they attempt to make sense of the uncertainty of their results? Inevitably, a critique has been that the arrival of genetic testing in clinics may have preceded the collection of sufficient data to support their use and the clinical knowledge required to interpret all types of results (Bell, 1998; Biesecker & Marteau, 1999; Holtzman & Watson, 1997). This study aims to address this gap by contributing to the experiential knowledge of interpreting and making sense of uninformative, uncertain genetic testing results for BRCA1 and BRCA2 among affected women considered at risk for HBOC who have gone through both genetic testing and the making-sense experience. The study overall purpose is to gain empirical knowledge about individuals' understanding of uninformative BRCA1 and BRCA2 test results but more specifically, to explore, from these individuals' perspectives, how they interpreted such results and how they made sense of them in light of the uncertainty and of their personal and family experiences with cancer. The study also 7 aimed to explore how uninformative genetic testing results affected individuals' everyday lives, health and illness experiences. To best serve this research purpose, a qualitative interpretive description was used (Thorne, Reimer Kirkham, & O'Flynn-Magee, 2004; Thorne, Reimer Kirkham, & MacDonald-Emes, 1997). The Research Questions The research questions for this study are 1. How do affected women considered at risk for HBOC interpret and make sense of their uninformative genetic testing results? 2. How do these results affect their everyday lives, health and illness experiences? The research questions are both descriptive and interpretive. That is, the first part of the first research question focuses on how affected women interpret their genetic testing results. The second part focuses on how they arrive at their interpretation based on how they made sense of them. The second research question explores how the women's interpretation of their results and making-sense experience affected their everyday lives, health and illness experiences. Because I assume that individuals' perspectives differ from the views held by health professionals, I wanted to first explore the women's perceptions. Hence, during the interview process, I did not label genetic testing results as uninformative or inconclusive, just as "your genetic testing results." My goal in using an interpretive description was to produce a rich description of the constructed and contextual nature of individual experience while at the same time allowing for shared realities (Thorne et al., 1997). Significance of the Research As previously said, most studies on the experience of receiving genetic testing results for BRCA1 and BRCA2 focused on individuals who were found to carry a mutation. Furthermore, 8 the studies that have been done are mostly based in the United States and thus, because of the different health care system, bring up issues that are not always relevant in the Canadian context, such as accessibility of testing because of its cost and health and life discrimination risk. One Canadian study, conducted by d'Agincourt-Canning (2003), produced experiential knowledge about some of the complex social and moral issues raised by positive genetic test results for hereditary breast and ovarian cancer. Currently, people who receive genetic test results are assumed to interpret their cancer and genetic risk in the same way that health professionals working in cancer genetics do (Bottorff, Ratner, Johnson, Lovato, & Joab, 1998). However, according to Bottorff, Ratner, Johnson, Lovato and Joab (1998), people tend to conceptualise risk using a constellation of descriptive meanings, while health professionals tend to conceptualise risk as a statistical measure. Further, Richards and Ponder (1996) explain that conflicts between aspects of scientific explanations and lay knowledge of inheritance may impede assimilation of the former. Because postcounselling after testing is not always available to people receiving uninformative genetic test results, there is little opportunity to assess how they interpret and make sense of their results. Yet, the need for this knowledge is urgent, given the growing number of companies, such as Myriad Genetics, who, despite providing supportive materials, rely on family physicians rather than certified clinical geneticist and genetic counsellors to provide counselling. The increased availability and easier access to genetic testing will create a huge pool of individuals and families who receive uninformative genetic test results for BRCA1 and BRCA2. Because nurses (and more specifically genetic nurses) will most likely be the first point of contact for those seeking further information and support before and after genetic testing, nurses need to prepare themselves. This thesis aims to produce knowledge to facilitate this preparation. 9 Definitions of Terms The following are key terms in this study. Affected women: the word 'affected' refers to having previously received a breast and/or ovarian cancer diagnosis. Primary and secondary participants: The primary participants are the affected women who have received genetic testing. Secondary participants represent family members chosen by the primary participants and are addressed as "family members" in this study. Individuals with a family history of risk for hereditary breast and ovarian cancer (HBOC): Individuals assessed to be at increased risk of carrying an inherited mutation predisposing them to breast and/or ovarian cancer, based on having two or more of the following: 1. Cancer in several closely related people, on the same side of the family, in several generations. 2. Cancer at younger than usual ages (e.g., breast cancer in the 30s). 3. More than one diagnosis of cancer in the same person. 4. Specific types of cancer that are linked to specific genes: e.g., breast cancer and ovarian cancer (BCCA, 1999). Having two or more of above factors in one family results in an estimated 20% probability of detecting a gene mutation (dAgincourt-Canning, 2003). Hereditary breast and/or ovarian cancer families: Families in which a mutation has been identified within the BRCA1 and/or BRCA2 genes. BRCA1 and BRCA2: BRCA1 and BRCA2 are the two most common breast and ovarian cancer genes tested for in genetic testing for an inherited susceptibility to the diseases. 'BR' 10 stands for 'breast' while 'CA' stands for 'cancer', while '1' and '2' represent the order in which the genes were discovered. Mutations on either BRCA1 or BRCA2 cancer genes predispose individuals to breast and ovarian cancer. In some families, there is a cluster of only breast cancers, in other families only ovarian cancer and, in still others, both. Although most participants in this study mainly have a family history of breast cancer, I chose to address them all as having a history of both cancers to reduce the risk of inadvertent identification of participants. Uninformative genetic test results for BRCA1 and BRCA2: Within the current study, this term applies only to affected women considered at risk for HBOC who received genetic test results for BRCA1 and BRCA2 indicating that no mutations were identified in the two cancer genes. However, because of their personal and family history of the disease, it is impossible to confirm that they do not carry an inherited mutation or that their cancer diagnosis may have occurred by chance. Lived experiences: Refers to how individuals describe and give meaning to what they have experienced as a result of having come into contact with a clinical context. Organisation of the Thesis This thesis is organised into eight chapters. Following the introduction of the study in chapter 1,1 proceed to a literature review in chapter 2. The literature review focuses on the process of genetic testing and on studies of the implications of breast and ovarian cancer genetic testing. Chapter 3 details my methodology as well as decisions I made throughout the study. Chapters 4 to 6 present the research findings. In chapter 4,1 begin this chapter by describing the implications of living with a family history considered at risk for HBOC presenting the women's interest in genetic testing and their reasons for accepting it. In chapter 5,1 focus on the women's 11 interpretation and making-sense experiences of their test results for BRCA1 and BRCA2. Chapter 6 discusses what participants concluded from their genetic testing experience. In chapter 7,1 discuss how key themes within the study findings can advance the practical science of nursing and inform other health professionals working in the area of cancer genetics. Chapter 8 begins with practical applications of key findings and their clinical applications and concludes with recommendations for clinical practice, research and policy development. 12 C H A P T E R T W O : R e v i e w o f the L i t e r a t u r e According to the Canadian cancer statistics (Canadian Cancer Society, National Cancer Institute of Canada, Statistics Canada, & Health Canada, 2004), breast cancer is still the most frequent cancer diagnosis among women and the second most frequent cause of death for women, following lung cancer. However, the prevalence of breast cancer diagnoses is more than twice that of lung cancer. The Canadian Cancer Society et al. (2004) estimate 21,400 new cases of breast cancer in Canada for 2004 and 5,300 deaths from the disease. Among all breast cancer diagnoses, an estimated 5 to 10% are said to be hereditary. Of these, the predominant cancer genes responsible are BRCA1 and BRCA2 (Lynch et al., 1999). An inherited mutation in either cancer gene increases the risk of developing both breast and ovarian cancer, but the breast cancer is more prevalent. Lynch et al. argue that initial lifetime cancer risk for carriers of BRCA1 and BRCA2 must be interpreted with caution because these estimates are based on cancer prevalence in large families and families at high risk of hereditary breast and/or ovarian cancer due to cancer diagnoses at young ages. Nonetheless, an individual found to carry an inherited mutation of either gene has a 50 to 85% lifetime risk of developing breast cancer (Lynch et al., 1999), compared with an 11% risk in the general population (Canadian Cancer Society et al., 2004). The lifetime risk of developing ovarian cancer is slightly more with a BRCA1 mutation than a BRCA2 mutation: 20 to 60% with a BRCA1 mutation, compared with 10 to 20% with a BRCA2 mutation (Lynch et al., 1999). Canadian Cancer Statistics (2004) estimates the lifetime risk of developing ovarian cancer among women in the general population to be 1.5%. 13 Before the discovery of these two cancer genes (BRCA1 in 1994 and BRCA2 in 1995) (Lynch et al., 1999; Wooster et al., 1994), estimating an individual and family risk for breast cancer for those with a family history was based solely on mathematical models such as the Claus model (Claus, 2001)4 and the BRCAPRO (Berry et al., 2002), although the BRCAPRO model comes closer to reflecting genetic risk/ Individuals and families with a strong family history of breast and/or ovarian cancer demonstrated high interest in inherited susceptibility testing (Lerman, Seay et al., 1995). The breakthrough in cancer genetics with the identification of the BRCA1 and the BRCA2 genes allowed genetic testing to enter clinical settings within research protocols. Since that time, we have seen a steady buildup of studies focusing on potential and actual impacts of genetic testing on people's lives. This first section of this chapter aims to provide an overview of hereditary breast and ovarian cancer. The second section focuses on the process of genetic counselling and the types of genetic testing that can be addressed in counselling, as well as genetic counselling issues and ethical concerns with genetic testing. The third section presents literature on the interest in genetic testing among women with a family history to breast and/or ovarian cancer and the intended use of test information, while the last section reviews literature about currently known psychosocial issues in genetic testing for BRCA1 and BRCA2. Hereditary Breast and Ovar i an Cancer Genetic susceptibility is an important factor in the development of breast and ovarian cancer, as shown with the discovery of BRCA1 and BRCA2. Other known inherited gene mutations that result in an increase risk to develop breast cancer but are not as penetrant as the 4 The Claus model was developed by Claus et al. and mainly based their estimate on family risk factors such as age at diagnosis for affected family member (Claus et al., 1991). 5 Comparisons of the sensitivity and specificity of the B R C A P R O model to genetic sequencing for B R C A 1 and B R C A 2 mutations demonstrated that B R C A P R O sensitivity in picking up mutations was at least 85% (Berry et al., 2002) and its specificity was 68.2% (Phillips, 2002). 14 BRCA1 and BRCA2 cancer genes are: the ataxia-telangiectasia mutated (ATM) gene, the 1 lOOdelC CHEK2, the phosphatase and tensin (PTEN) tumor suppressor gene, the p53 tumor suppressor gene, the androgen receptor gene (AR), the hereditary nonpolyposis colorectal cancer (HNPCC) gene and the estrogen receptor gene (Bennett, Gattas, & Teh, 1999). Although most breast cancer cases are not hereditary, 5 to 10% of all breast cancers are thought to involve one of the above genes (Claus et al., 1991), although 70% to 90% of all hereditary breast cancers harbour a BRCA1 or BRCA2 mutation (Bennett et al., 1999; Lynch et al., 1999). A family history of breast and ovarian cancer is recognised as one of the most important risk factors for the diseases (Yang & Lippman, 1999). Other nonhereditary factors documented to increase risk for breast cancer include having a personal history of breast cancer, early menarche, late menopause, lack of breastfeeding (a protective factor), use of birth control pill, use of estrogen replacement therapy, childbirth delayed after age 30, and lack of full-term pregnancies (Kelly et al., 2004). Other clinicopathological features of individuals and families suggestive of hereditary breast and ovarian cancer include early disease onset, high incidence of bilateral breast cancer, and high incidence of multiple primary carcinomas in other organs (Anderson, 1992; Nomizu et al., 1997; Ormiston, 1996). A family history of cancer with the following characteristics indicates a probable inherited susceptibility: a woman with two or more first-degree relatives (mother, sister or daughter) with breast and/or ovarian cancer, and breast cancer occurring in a manner consistent with Mendelian transmission of autosomal dominant inheritance (Ormiston, 1996). This means that breast and/or ovarian cancer appears on the same side of the family, among same-blood family relatives. When individuals within families are identified as having an inherited mutation such as in their BRCA1 and BRCA2 genes, their children have a 50% chance of inheriting the mutation. 15 However, unlike single-gene syndromes, such as Huntington's disease that follow the laws of simple Mendelian genetics,6 individuals with an inherited mutation of either the BRCA1 or BRCA2 gene will not necessarily develop the disease but, rather, will be at increased risk of developing it. Ormiston (1996) explains that, unlike single-gene syndromes, for hereditary breast cancer to develop, a person must have inherited one mutated gene and develop a second, somatic mutation in the same region of the gene with the inherited mutation. In sporadic cancers, both somatic mutations are acquired after fertilisation, in the same region of the gene and the mutations occur at a much later time in life. Both scenarios of mutated genes leading to cancer are known as the 'two-hit theory' of carcinogenesis (Ormiston, 1996). Cancer genes, when not mutated, actually protect against cancer by repairing DNA and suppressing tumours. It is now possible, through genetic testing, to determine if one is carrying a mutated gene. Carrier, presymptomatic, and predisposition testing are the three general genetic tests now offered. Carrier testing indicates whether a healthy individual carries one copy of a gene mutation and involves testing for a gene protein product (Ormiston, 1996). This test has implications for the decision to procreate, in that of those identified as gene mutation carriers who have children with a partner who is also a carrier, each of their children faces a 25% chance of inheriting each of their parents mutated gene and, if they do so, becoming affected with the disease. As with recessive inherited conditions, two copies of the inherited mutation are necessary to develop the disease. If only one copy is inherited from one of the parents, the person will be known as a carrier but will not develop the disease. Examples of common recessively inherited conditions include cystic fibrosis, Tay-Sachs disease, and sickle-cell anemia. In presymptomatic testing, healthy individuals learn whether or not they will eventually develop a genetic condition. The most common presymptomatic testing is for Huntington 6 Three modes of Mendelian inheritance are possible: autosomal recessive, autosomal dominant and sex linked. 16 Disease. Only one mutated gene of a pair of genes is needed to determine whether the individual will be affected by the illness. This type of inheritance is autosomal dominant (Ormiston, 1996). With susceptibility or predisposition testing, individuals learn if they are at increased risk of developing a condition but do not know with certainty whether they will develop the disease. The most recent common types of predisposition testing are for breast and ovarian cancer susceptibility such as genetic testing for BRCA1 and BRCA2 (Ormiston, 1996). The Hereditary Cancer Program (HCP), where the study participants were recruited, reported three categories of results for predisposing testing for breast and ovarian cancer susceptibility for individuals who met the HCP eligibility criteria for genetic risk assessment:7 (a) positive - mutation was found in one or both cancer genes tested; (b) negative - known familial inherited mutations were not found; or (c) uninformative - either a change in the sequence of their amino acids has been found but with no known documented clinical relevance yet reported or no mutation was found but because of their strong individual and family history of the disease, the absence of an inherited mutation cannot be confirmed. Challenges with Genetic Testing and Counselling The provision of genetic testing for BRCA1 and BRCA2 has not escaped the political turmoil. In Canada, most clinics provide genetic testing for BRCA1 and BRCA2 within research protocols only because of limitation use imposed by Myriad Genetics' patented BRACAnalysis test. Among the Canadian provinces offering testing are British Columbia, Ontario, Alberta and 7 Eligibility criteria used within the HCP can be viewed in table 3. These criteria tend to vary among provinces, but testing is usually limited to individuals and families with a family history of risk for HBOC. The HCP genetic testing protocol is based on a 20% priori probability of finding a mutation. According to a prior study that recruited its study participants in the same cancer agency as the current study, the 20% cut-off point represents the likelihood of finding a mutation in individuals already diagnosed with cancer compared with finding a mutation in an unaffected individual (less than 10%). However, eligibility criteria are subject to change with new findings indicating specific variations in individual and family histories of the disease that either increase or decrease the likelihood of carrying an inherited mutation, as well as newly identified mutations. In 2000, the HCP called for an ad hoc committee to review the eligibility criteria used by other centres conducting genetic testing for BRCA1 and BRCA2. 17 Quebec (Eggertson, 2002; Williams-Jones, 2002).8 The other provinces mostly provide counselling services but send their samples for analysis to laboratories in Canada; mostly in Ontario (Williams-Jones, 2002). While in 2000 and 2001, the U.S. Patent Office has awarded a series of patents to Myriad Genetics that suspended testing in British Columbia in July 2001 (but still continued to offer genetic counselling), other Canadian provinces such as Ontario, Alberta and Quebec continued to offer testing. The Ontario government is currently challenging the patent. Myriad is demanding that all BRCA1 and BRCA2 testing be conducted in their laboratories for a cost of about $2500US, in comparison to a cost of approximately $1200 -$1500 by the British Columbia Cancer Agency Hereditary Cancer Program (BCCA HCP) laboratories. Since February 2003, the BCCA HC Program has resumed testing in its own laboratory (Y. Ridge, personal communication, February 28, 2005). In 1996, the BC Cancer Agency and the BC Provincial Medical Genetics Program established the Hereditary Cancer Program (HCP) to provide genetic education, counselling and testing for BRCA1 and BRCA2 to individuals and families with a strong history of cancer (Williams-Jones, 2002). Through the creation of an official Cost Centre for the HCP in 1997 by the Ministry of Health, the HCP aimed to offer clinical service while first being established as a research protocol. In the United States, it is currently possible to collect your own DNA sample and to have it sent directly to a laboratory for a substantial fee. Whereas the United Kingdom and Canada finds direct marketing of genetic testing inappropriate without the involvement of proper qualified personal, Myriad Genetics, the U.S. base biopharmaceutical and genomics company that initiated an intensive five months direct-to-consumer advertising campaign for breast cancer 8 A n in-depth review of the current legal issues related to genetic services for B R C A in light of patent regulations is beyond the scope of this study. For a thorough review, see chapter 2 of Bryn Will iams-Jones ' (2002) doctoral dissertation "Genetic testing for sale: Implications of commercial B R C A testing in Canada." 18 genetic testing (Myriad Genetics, 2002) still recommends that all individuals considering genetic testing first consult with their doctor. While laboratories conducting genetic tests are not required to participate in quality programs for genetic testing, both the U.S. Task Force and the U.K. Advisory committee strongly recommend against "over-the-counter" genetic testing for late onset diseases (Jamieson, 2001). In Europe, The Public and Professional Policy Committee (PPPC) of the European Society of Human Genetics (ESHG) organised a workshop in November 1999 in Amsterdam, where 51 experts from 15 European countries derived recommendations for genetic screening programs. Recognizing the possible harms that may arise with genetic testing, the PPPC and the ESHG recommended the establishment of proper organizational criteria similar to those found in Canada such as: having all programs be governed by a governmental or nongovernmental body and having in place guidelines for testing, genetic counselling and delivery of information. As well, they recommended that all individuals who access genetic programs have the possibility to meet face-to-face with a trained professional if needed following the receipt of test information (Godard, ten Kate, Evers-Kiebooms, & Ayme, 2003). While maintaining provision of genetic testing for BRCA1 and BRCA2 within local clinics may be uncertain because of Myriad Genetics patent laws, the HCP amongst other clinics offering testing aims to continue offering genetic services in the most holistic manner — that is, providing pre- and post-counselling services with genetic testing. With genetic testing for adult-onset-hereditary-diseases becoming an important part of clinical genetics practice, we see an increased demand for tests, perhaps as a result of media coverage of genetic discoveries or from referrals by health professionals. While eligibility criteria for testing may vary between testing agencies of the same country and of different 19 countries, the criteria usually include having a strong family history with the disease or early onset occurrence of the disease. The eligibility criteria set by the HCP at the time of the study in 2000 can be viewed in table 3. These criteria represent a 20% chance of finding a mutation in the BRCA lor BRCA2. Because new discoveries and new developments in genetics are ongoing, regulations of eligibility criteria are likely to change as a result of these. Within the context of the current study, participants were recruited from the BC Cancer Agency HCP. Most individuals are referred to the program by their health care practitioners. While testing agencies referral process may differ, the HCP uses a triage system set by eligibility criteria for genetic testing for risk of HBOC so that its limited genetic counselling resources can be made available to those whose family history with cancer is most suggestive of hereditary breast and ovarian cancer (HBOC). Although all individuals referred to the HCP can receive genetic assessment, only those who meet eligibility criteria are offered testing following the receipt of their genetic counselling session. Tested individuals who are found with a mutation in their BRCA1 and/or BRCA2 meet again with a genetic counsellor and a medical geneticist to discuss further results implications for themselves and their family members. For individuals who receive uninformative results, their results are first explained through a telephone call made by a genetic counsellor and the genetic counsellor's explanations are further followed by a written letter. The HCP does reiterate to all individuals who have received genetic testing that provision of face-to-face discussion is always possible. From January 1997 to December 1998, the HCP reported over 250 individuals who received uninformative genetic testing results for BRCA1 and BRCA2. At the time of this study, the lapse of time between obtaining genetic testing and receiving results averaged 18 to 24 20 months for new cases and four to six weeks when a mutation has been identified within a family member.9 Hence, for individuals who received genetic testing between 1997 and 1998, their full results of both BRCA1 and BRCA2 were obtained in 1998 to 2000 (K. Panabacker, personal communication, January 11, 2000). These individuals constitute the population pool for this study. Recent numbers indicate that the total number of referrals to the Cancer Agency between January 1998 to June 2004 is approximately 5,900 (records were not kept previously). Of those, 80 to 90% were for hereditary breast and ovarian cancer risk assessment. Again, of the 5,900, approximately 4,000 were seen for genetic counselling and more than 508 subsequently received uninformative genetic test results, while 345 received positive results, and 192 received true-negative results. The remainder are still awaiting their results (M. McCullum, personal communication, August 12, 2004). Although laboratories offering screening for BRCA1 and BRCA2 mutations may differ in the technology they use, three main types have been documented in studies of BRCA1 and BRCA2 mutation screening: full sequencing (Menkiszak, Brzoskoj et al., 2004), the Protein Truncation Test (PTT) requiring polymerase chain reaction (PCR) amplifications (FitzGerald et al., 1996; Iglehart et al., 1998; Malone et al., 1998), and Single-Strand Conformational Polymorphism (SSCP) (FitzGerald et al., 1996; Iglehart et al., 1998; Malone et al., 1998). A new system of testing, Denaturing High Performance Liquid chromatography (DHPLC) is now being tested. Using DNA from the peripheral blood lymphocytes of an individual already diagnosed with breast and/or ovarian cancer, identification of a possible BRCA mutation is best achieved 9 Reasons given to the extent of time between testing and providing results are the unexpected demand for this test and the limited availability of technicians to process all D N A samples (K . Panabacker, personal communication, January 11, 2000). 21 through full sequencing of the whole gene. However, full sequencing is often not possible largely for economic reasons, thus a modified screening approach is employed. Often, a combination of the Protein Truncation Test (PTT) and single-strand conformational polymorphism (SSCP) are used to detect possible mutations - which can then be characterized by DNA sequencing. Full sequencing of the BRCA 1 and BRCA 2 genes, for example, would offer 99% clinical sensitivity11 - meaning that the chance of failing to detect a change in the sequence of the gene is less than 1%. Changes such as large genomic deletions, however, will not be detected by full sequencing. Sequencing has the same weakness as SSCP in this regard (M. Anderson, personal communication, June 7, 2000).12 The PTT can only determine the length of a specific protein but cannot detect a change in the sequence of the gene. That is, the PTT can only detect mutations that create premature protein termination codons. This results in a shortened protein product that can be detected by gel electrophoresis. PTT will produce false negative results if the gene mutation results in a simple amino acid substitution (one for another) or an "in-reading-frame" addition or deletion (loss or gain of a single amino acid) (M. Anderson, personal communication, June 7, 2000). These types of mutations may or may not have physiological significance in the function of the 1 0 However, as attested by Peshkin et al., (2001), full sequencing is still not 100% sensitive. Large deletions (loss of genetic material) can go undetected, and there are variants that are not detectable by this method. Fu l l sequencing of a gene refers to the process of determining the order of the individual 's four nucleotide bases in a D N A molecule known as: adenine, thymine, guanine, and cytosine (University of Illinois at Chicago, 2001). 1 1 The accuracy of a test is described by its clinical sensitivity and specificity. Cl in ica l sensitivity reflects the probability that a test wi l l be positive when a particular sequence is present (Holtzman & Watson, 1997). The failure to detect a change in a specific sequence reduces a test's clinical sensitivity. Cl in ica l specificity refers to the probability of the test being negative when no change in the gene sequence is found (Holtzman & Watson, 1997). 1 2 M . Anderson is the laboratory technician who conducts all D N A tests at the Hereditary Cancer Program where the study participants were recruited. 1 3 Here is an abbreviated glossary to genetic terms used in this paragraph: Genes are pieces of D N A ; D N A is composed of four bases. Amino acids represent a combination of three of the four bases which are the building blocks of protein. Codons is a three letter word specifying either the start of a protein, end of a protein or one of the amino acids building blocks. A stretch of D N A correspond to a gene that codes for a particular protein that has a particular function. A n d exon is the region of a gene that contains the code to produce a specific protein (National Human Genome Research Institute, 2004). 22 BRCA1 or BRCA2 gene. If, however, the PTT detects a change in the length of a specific protein, then sequencing will be done in that specific area of the gene to determine the specific characteristic of the mutation. Depending on the setting, in instances where the PTT does not detect any mutation in the major exon observed likely to harbour mutations known to a specific hereditary disease, SSCP is employed on the remaining exons. SSCP is used to find gel mobility variations in the exon, in comparison with normal genes. The presence of a variation indicates a mutation. If a variation is found, sequencing is done for that area of the gene to determine the mutation. With SSCP, a risk of producing a false negative result is possible because the test will not detect large deletions on the exon. Genetic Counselling for Hereditary Breast and Ovarian Cancer Genetic counselling for hereditary breast and/or ovarian cancer is based on past practices with Huntington Disease (Wiggins et al., 1992). "Genetic counselling has evolved to mean a communication process by which individuals and families come to learn and understand relevant aspects of genetics, to make informed health decisions, and to receive support in integrating persona] and family genetic information into their daily lives" (Lea, Francomano, & Jenkins, 1998, p. 111). Genetic specialists, including nurses, participate actively in the advancement and services of genetic programs in many areas of the US and Canada. Although Canadian genetic programs and services are not as long established as those in the US, we have seen increased interest in Canada in developing both genetic programs and genetic nursing practice. The Ontario government announced in 2000 their decision to fund genetic testing for breast, ovarian and colon cancer through nine regional centres, for a total of 21 clinical genetic clinics as part of the 23 provincial health insurance (Williams-Jones, 2002).14 Since then, a group of Canadian nursing researchers began studying the expansion of genetic nursing in Canada. The research is funded by CIHR and was entitled Genetic Services and Adult Onset Hereditary Disease: Current and Future Nursing Roles (Bottorff et al., 2004). The increased availability of genetic testing and the shortage of genetic counsellors require nurses to prepare themselves to participate in genetic counselling. Some documented roles of nurses include supporting individuals and their families during the counselling process, helping them to integrate new genetic information into their daily lives, collaborating with genetic counsellors to support individuals and families, and coordinating genetic health care (Lea et al., 1998). Other genetic specialists involved in genetic counselling include clinical geneticists (MDs), medical geneticists (Ph.Ds), clinical molecular geneticists (MDs or Ph.Ds), genetic counsellors (health professionals with Master degrees), advanced-practice genetic nurses (nurses with Master of Science or Doctoral degrees), and genetic nurses (Lea et al., 1998). This team of genetic specialists collaborates in providing support, genetic evaluation, information, and resources to individuals and their families. The team members also obtain and interpret complex family history information, evaluate and diagnose genetic conditions, and interpret and discuss complicated genetic test results (Lea et al., 1998). The person seeking testing and their family members have an active role in decision making throughout the counselling process. Guiding principles of genetic counselling include nondirectiveness, awareness of social and cultural differences, confidentiality, and privacy (Lea et al., 1998). Genetic counselling aims to respect individuals' right to self-determination (Michie & Marteau, 1996). The nondirective 1 4 See also the following link for more information on the Ontario government announcement to funding genetic testing for breast, ovarian and colon cancer : 24 approach also respects decisions that reflect the clients' personal values, beliefs, and interests. It also requires genetic specialists to be aware of their personal values and how they may be reflected in communicating genetic information. Furthermore, nondirectiveness means that genetic specialists must provide all information about the risks, benefits, and limitations of a particular genetic test, and about prognosis for and management of the genetic condition (Lea et al., 1998). Awareness of social and cultural differences calls for genetic specialists to recognise how values and beliefs can affect genetic counselling. Cultural differences are not solely associated with race or ethnicity, as Clarke (1997) points out, but also with a wide range of formative experiences. Individuals from different ethnic groups may experience problems when they are in contact with genetic services. Linguistic and cultural barriers may influence their understanding of and response to genetic counselling and screening (Clarke, 1997). Cultural differences may also affect autonomous decision making about genetic testing. Further empirical studies of beliefs about inheritance and disease are needed to improve professionals' understanding of them. Genetic specialists must also assess how their personal beliefs influence the way they collect and communicate genetic information (Lea et al., 1998). Personal values of clients should not be seen by health professionals as limitations to their understanding of genetic information. In fact, having a better understanding of their clients' values and beliefs should enable genetic specialists to adapt their counselling to understand better their clients. Confidentiality and privacy are important in genetic counselling as well. Genetic test results can affect family members, intentionally or not. Individuals do not always want their family members and friends to know of their genetic status. While not telling family members 25 about the test result could deprive them of information that could affect decisions about their lives, the current state of knowledge about prevention or early detection of breast or ovarian cancer does not justify such disclosure. However, legal interpretations from authors such as Burgess, Laberge, and Knoppers (1998) and Surbone (2001) suggest that this view may change if more effective treatments show in the future that disclosing genetic mutation status may prevent harm. Women with a familial history of breast cancer should usually be advised to take early detection measures whether or not they have BRCA1 or BRCA2 mutations (Burgess, 1999; Koenig et al., 1998). Moreover, the risk of cancer to other family members when an individual within a family is found with an inherited mutation is not immediate and hence, not likely to change substantially as a result of genetic information (Koenig et al., 1998). Generally, genetic specialists should honour any wish for confidentiality of the individual who seeks genetic testing. Because there is a possible risk of insurance discrimination, ensuring and accurately describing the risk related to confidentiality of genetic information becomes even more important (Burgess etal., 1998). Privacy, confidentiality and threat of genetic discrimination are the most cited ethical issues related to genetic testing (Burgess, Knoppers, & al., 1999; Knoppers, 1998; Surbone, 2001). Describing these potential social harms and psychological distress that may arise from genetic testing is a key part of informed consent.15 Informed consent should always be obtained before genetic tests are conducted and opportunity for the individual to participate fully in the decision process needs to be provided (Burgess et al., 1998). Moreover, informed consent should 1 5 There exist detailed theoretical literature on the ethical implications with genetic testing, however, as the current study is not focused on the moral and ethical issues with genetic testing, only an abbreviated review of the content address in documented and observed informed consent to genetic testing is presently address. For a current literature review on the contemporary bioethics literature on genetic testing for hereditary breast/ovarian cancer, see d'Agincourt-Canning (2003). Her study addressed the complex moral and social issues generated by genetic testing. A l s o see Burgess (2001) for a review of the ethical and social issues in genetic testing beyond informed consent. 26 not be limited to the technical and medical aspects of testing but should also include social and ethical implications (Surbone, 2001), as well as alternatives to proposed treatments. While full autonomy in genetic testing can never be attained fully, a standardised protocol that explicitly describes the process of genetic counselling and informed consent provides at least some guidance towards obtaining informed consent (Burgess, 1999; Burgess & d'Agincourt-Canning, 2001; Burgess et al., 1999). Within the issue of privacy, individuals can be told that, although the clinic will keep their test results confidential, there is always a risk that genetic discrimination may occur (Koenig et al., 1998). Health professionals and others have an obligation not to disclose private medical procedures unless authorised by the patient. In addition to discrimination in life insurance, disclosure of genetic information can cause discrimination by health insurance providers and, discrimination in the work place (Burgess et al., 1998; Surbone, 2001). Psychological distress can result when genetic testing is offered without pre- and post-test genetic counselling (Burgess et al., 1998). What genetic testing may reveal should be discussed before the test within the informed consent procedure (for example, paternity).16 Follow-up genetic counselling should assess how individuals have integrated the test results and their implications into their lives (Burgess et al., 1998). Although risks to privacy and confidentiality as well as the risk of discrimination have been discussed by many in the ethics and legal communities with an interest in genetics, few of these risks have been empirically documented in Canada. However, according to Surbone (2001), cases of discrimination following genetic testing for BRCA have been documented in the United States. As Surbone (2001, p. 156) warns, "we are not our genome: yet we culturally tend to see genetic data as more fundamental and more inalterable than any other equation of life." 1 6 Other psychosocial issues that may arise from genetic testing are discussed in the next section of this chapter. 27 D'Agincourt-Canning's (2003) doctoral study, which addresses moral agency in genetic testing for hereditary breast and ovarian cancer, proposes that, in light of the genetic information being familial as well as personal, that perhaps the concept of relational responsibility could be assessed for its potential contribution to the informed consent process. Relational responsibility complements an individual's right to autonomy and reinforces the complex social nature of genetic testing (Burgess & d'Agincourt-Canning, 2001). D'Agincourt-Canning (2003) argues that further empirical study focusing on the experiential impact of genetic testing on people's everyday lives is needed to understand the moral agency of individuals and families who have received results of genetic tests to hereditary breast and ovarian cancer. What has been described and discussed so far are the technical and ethical issues related to genetic testing and counselling. What follows is an overview of the process of experiencing genetic testing and counselling. Experiencing the process of genetic testing and counselling The information presented in this section is derived from the literature, as I do not have first-hand clinical experience in genetic counselling, although I have observed three genetic counselling sessions. The process observed in these sessions is representative of the report by Lea, Francomano and Jenkins (1998) on the genetic counselling process. These authors explain that genetic counselling can occur over an extended period and usually entails more than one visit. The long process allows time for individuals and families to understand genetic information before deciding whether or not to have genetic testing done. Although the process of genetic testing may differ from one country to another, individuals and their families can be asked to first complete a survey that contains a family history questionnaire. The survey is usually provided by the family physician who, based on the family history, recommends genetic testing. 28 Once the completed questionnaire is received by the agency, a genetic nurse evaluates the family history of the individual to decide if the person meets criteria for genetic risk assessment. If the individual does meet the criteria, she/he is put on a waiting list to be seen by a genetic counsellor. Prior to the counselling appointment, the individual receives a telephone call from the genetic nurse explaining what may be expected from the counselling session and who can come for testing. However, the nurse will also explain that she may choose to have the family members make their own appointments and undergo the process separately. During the pre-counselling phase, the nurse will also assess the emotional and social status of the individual and her family (Lea et al., 1998). Usually all individuals seen for genetic counselling are seen again after receiving the results of their genetic test. According to Lea et al. (1998), first phase of genetic counselling is assessment and information gathering. This includes asking individuals and family members why they were referred and what they expect from the session. According to my observations of genetic counselling and the literature, genetic counsellors then review the family history; determine the clients' level of knowledge about hereditary breast and ovarian cancer; communicate genetic information; discuss genetic testing, its accuracy and its pros and cons; and describe existing supports such as networking groups and the follow-up genetic counselling that can be expected. Following each of the sessions, genetic counsellors, with individuals' permission, write a report for the family physician and coordinate the care to be given by other specialists as needed. For example, when family members agree to give blood for genetic testing but live far apart, the counsellor will make the necessary arrangements to have the person tested if genetic testing is offered in their area or arrange to have their blood sample shipped to the testing agency. 29 While much of the information needed to make a genetic risk assessment is gathered by the genetic nurse during the precounselling phase, this information is then reviewed by the genetic counsellor with the individual during the session. Following this review, the genetic counsellor gathers and discusses with the individual her or his family history with cancer. This phase consists of filling out a family pedigree based on information shared by the individual during the precounselling phase and information shared during the session. The pedigree represents the family's medical history and helps the genetic counsellor determine if a strong possibility exists for an inherited susceptibility to breast and/or ovarian cancer (Lea et al., 1998). The genetic counsellor collects information about the number of affected and unaffected individuals with breast cancer and other forms of cancer from the maternal and paternal sides of the family and inquires about the ages at which diagnoses were made. Based on my observation of genetic counselling sessions, the pedigree is used by the genetic counsellor to explain to individuals the differences between sporadic and hereditary breast cancer. The discussion focuses on the inheritance of cancer if the family history suggests a strong probability of this. 1 7 The genes associated with breast and ovarian cancer are discussed and presented graphically and the mutation process that can lead to cancer is explained. This part of the session is usually presented by both a medical geneticist and a genetic counsellor and also includes a discussion of the risks, benefits and limitations of testing, and available treatments, if an inherited susceptibility to the disease is found (Lea et al., 1998). However, presenting and explaining notion of inheritance is not as straightforward an activity as presented in the Richards and Ponder (1996) study. 1 7 This part of counselling can be sensitive as, to some individuals, viewing their family history in a pedigree may bring their first awareness of a probable inherited susceptibility to breast and/or ovarian cancer. 30 Richards and Ponder (1996) conclude through hypothesis testing that people tend to derive their understanding of inheritance from concepts of social relationship of kinship, which Richards and Ponder (1996) explain are sustained by everyday social activities. The authors posit that closeness in genetic terms is determined by the closeness of social relationship and social obligations. For example, the authors note that breast cancer and ovarian cancer were described by their study participants as a "proneness" to cancer passed by mothers to daughters. Conversely, women from families with ovarian cancer may not see themselves at risk because they "take after their father's side of the family" (Richards, 1996). Overall, people recognise an equal contribution by both parents to inheritance but are less aware of the genetic connection between an individual and more distant relatives. Richards (1996) notes that some inheritance principles understood by lay people are that family resemblance to an affected family member makes that individual more at risk to develop the familial disease and believe that the disease will appear at the same age as the affected relative. There is also the idea of inheritance as being able to "ski" a generation. Cox and McKellin's (1999) research demonstrates that social components, rather than biological components, shape the everyday understanding and experience of hereditary risk. Their conclusion concurs with that of Biesecker (1997b), whose secondary analysis of studies and book chapters that had looked at the psychological aspects of cancer genetics and susceptibility testing shows that risk perceptions are also shaped by the complexities of experience in a cancer family and by factors influencing the understanding of risks such as personality traits, illness perception, family relations, mood, and anxiety. Assessing how individuals interpret inheritance and how they draw upon these interpretations when understanding hereditary risks is all part of the counselling process. Whereas most sessions take up to an hour or more depending on the agency and the individual's 31 need, a second face-to-face counselling session is, at times, requested by individuals before making their decision for testing. Follow-up education will often be done by the genetic nurse. Concomitantly, at the end of the session, the individual and family members are informed that it takes time to understand genetic information and that a networking group is ready to help answer 18 their questions and concerns. While follow-up genetic counselling when giving out genetic testing results is recommended standard practice, it is not always feasible. When genetic testing is done through a private company such as Myriad Genetics, the onus is on the provider or on the individual to seek pre- and post-test counselling through their personal health care provider. However, in cases where genetic testing was done within a research protocol, postcounselling is usually included in the process. But, at times, because of cost and time constraints, only those who receive positive test results may be offered a one-to-one session, while individuals who receive uninformative results are given support over the telephone and sent a letter explaining the meanings and implications of their results. They are usually offered a one-to-one session only if they feel the need. Again, procedures for informing individuals and families of their test results depend on the agency (Lea et al., 1998). Providing genetic test results is an important part of the genetic team's role. However, communicating risk information to the lay public can be challenging. One of the challenges to communicating cancer risk information is in moving beyond the individual's assessment of risk to assessing their risk within their familial context. That is, in that family history of cancer is an important risk factor in establishing personal risk, situating cancer risk information within the 1 8 Because most genetic testing is offered through research programs, not all settings that offer this service also coordinate the networking groups. The Hereditary Cancer Program of the B C Cancer Agency offers such a service for those who have received genetic counselling, regardless of whether or not they have cancer or received their genetic test results. 32 familial context becomes important. Other challenges include how risk information is communicated, as the way the message is framed can influence how the content is understood and acted upon (Sarfati, Howden-Chapman, Woodward, & Salmond, 1998). For example, following an empirical literature review of thirteen studies, Edwards and al. (2003) found that detailed personal risk communication in the form of numerical calculations of risk was associated with a smaller uptake of screening while risk communication stated as categories (high, medium and, low) or simply listed as personal risk factors presented higher uptake of screening programmes. Most screening programmes addressed in the studies reviewed were in relation to mammography screening, breast cancer risk and genetic testing. Bottorff, Ratner, Johnson, and McCullum (1999) also comment on how physicians' struggle with conveying limitations of cancer risk knowledge for fear that this information could be misinterpreted. The literature documents that health professionals and lay people differ in the ways they understand risk information (Peshkin et al., 2001). That is, risk figures can be given in genetic counselling as odds or as percentages (Harper, 1993). Pitfalls exist for both methods, however. With odds, some individuals may not clearly grasp that odds ratios refer to the future and not to the past. As Harper (1993) explains, just because an individual has two affected children in succession does not make it less likely that the next child will be affected since chance has no memory. Also, the possibility exists that individuals reverse or misinterpret odds ratios. Nevertheless, when risk values are framed in positive ways, for example the risk of not developing breast cancer for carriers of BRCA1 and BRCA2 gene mutations is approximately 30% as opposed to the risk of developing breast cancer for carriers of BRCA1 and BRCA2 is approximately 70%, the experience can be less threatening and may lead to different decisions 33 (van Zuuren, van Schie, & van Baaren, 1997).19 Consideration should be given to how risk estimates and figures are interpreted by individuals who seek and receive genetic testing. These interpretations need to be explored during counselling sessions, since studies have shown that lay people do not always understand principles of probability (Macdonald, Doan, Kelner, & Taylor, 1996). Communicating cancer risk also poses challenges for people who do not always understand the meaning of lifetime risk for cancer and how such risk can be modified by individual factors (Rimer, 1999). Lipkus and Hollands (1999) suggest the use of graphic visual display to enhance the understanding of numerical risk. While diverse approaches to providing risk information may help individuals understand their theoretical risk to cancer, none can eliminate the ambiguity resulting from the inherent uncertainty associated with risk information. Future studies are still needed to assess the best approach in presenting uncertainties with risk estimates. Studies that have looked at communication of risk information note that some individuals may feel more "prone to cancer and in having the familial known inherited mutation to cancer" when they see physical or behavioural resemblances between themselves and the individuals in the family who have developed cancer or who have been found to have an inherited mutation increasing their susceptibility to cancer (Richards, 1996; Richards & Ponder, 1996). While some individuals within one family may see themselves more prone to cancer and to an inherited mutation, others may see themselves less at risk. For example, within the context of hereditary breast cancer, men have an equal risk to women in inheriting a BRCA1 or BRCA2 mutation and passing it down to their children but men may still perceive themselves less at risk for inheriting a mutation associated with one of these cancer genes because of seeing breast cancer as less 1 9 The topic of decision making with uncertain results is discussed later within this chapter under the section "interest and intended use of test information". 34 relevant to them and being a women's disease. However, even less understood than how perceptions of risk are formed and how risk is communicated are the conceptual structures individuals use to interpret and make sense of genetic test results. There is a considerable absence of studies that have empirically researched this process. As Lynch et al. (1999) attest, because identification of the BRCA1 and BRCA2 mutations is relatively recent, knowledge of how best to translate genetic information in a clinical setting is still ongoing. Even though providing genetic test results for BRCA1 and BRCA2 may seem relatively straightforward, in many instances the process becomes complex, not only because of the cases encountered within clinical settings but because of the many complex issues in genetic counselling (Peshkin et al., 2001). Even with positive genetic test results, the process is complex, as the risk of developing cancer cannot be precisely quantified and management options are still uncertain. However, recent studies indicate that prophylactic surgeries protect against recurrence of the cancer or new primary cancer among those with an identified BRCA mutation (Narod, 2002). Affected individuals found to carry an inherited mutation have an approximately 40% risk of developing contralateral breast cancer over the next decade (Narod, 2002). 20 The experience of true-negative results can also be compounded by confusing interpretations because an individual may interpret the results as no longer being at risk of ever developing breast and/or ovarian cancer and experience their good fortunate with survivor's guilt over family members found with the familial mutation (Peshkin et al., 2001). Or, as is the focus of this study, individuals with a family history of risk for hereditary breast and/or ovarian cancer may receive results indicating no mutation was found without a straightforward explanation from genetic health professionals on how to interpret and make sense of their results. Thus, genetic 2 0 Recall that a true negative genetic test results for B R C A land BRCA2 can only be received by individuals from families with an identified inherited mutation of either cancer gene. 35 counsellors and other health professionals providing support and education face many challenges - especially with individuals receiving uninformative genetic test results. In fact, they will likely face many such individuals, as studies have demonstrated that 16% to 66% of all high-risk families do not carry detectable mutations in their BRCA1 and/or BRCA2 cancer genes (Peshkin et al., 2001). Within the health literature, guidance can be found for interpreting uninformative BRCA1 and BRCA2 genetic test results; however, as said earlier, most of this information is from clinical literature as opposed to empirical literature. Authors such as Lynch et al. (1999), Peshkin et al. (2001) and Iglehart et al. (1998) who have conducted studies on the complexities of interpreting uninformative genetic testing results to BRCA cancer gene mutations maintain that counselling individuals from families considered at risk for HBOC can be a huge challenge. Such as, Lynch et al. point out, "there is currently no test that is sensitive enough to pick up all possible mutations" (p. 97). Hence, individuals receiving uninformative genetic test results must be told that there is still a possibility that they carry a mutation undetected by current tests. Iglehart et al. recommend that individuals know about and understand these limitations to genetic screening before deciding to have genetic testing done. Interpretation of negative results from studies conducted on other inherited diseases such as for familial adenomatous polyposis (FAP) also show that some individuals fail to be reassured of their decreased risk to general population risk as a result of having received a negative result (Michie, Smith, Senior, & Marteau, 2003). Although representing a small scale study of nine participants, explanations given by Michie et al. provide some understanding on the reasons for the lack of reassurance experienced despite receiving negative results. Among the nine participants, four continued to perceive self at increase risk for FAP because they doubted the efficacy of a blood test to determine one's risk 36 for a disease that occurs in the bowel. The participants also perceived DNA testing as too simple compared to an invasive testing of colonoscopy that can actually see inside the bowels. Other findings from their study also show that participants expressed doubts about the sophistication of DNA testing technology to detect all mutations as well as viewing their results as transitory. That is, the testing could only test for known mutations to date. To illustrate the complexities that may arise with genetic counselling setting of individuals who receive uninformative results, I will summarise a case study presented by Peshkin et al. (2001). This case was drawn from their clinical research with high-risk individuals who received genetic counselling and testing. It is a case of an affected woman with a previous breast cancer diagnosis at the age of 41. Her family's history of cancer comes from her paternal side and presents only with breast cancer. There are six cases of breast cancer in three generations with four considered early-onset cases (i.e., diagnosed before the age of 50), one male breast cancer and one case of bilateral breast cancer. The index case (this affected woman) underwent full genetic sequencing for BRCA1 and BRCA2 and was not found to have a mutation or any variants. The authors conclude that, in light of her strong family history that gave her an overall 90% risk of carrying an inherited mutation, hereditary breast cancer cannot be ruled out. Possible interpretations of her situation were multiple and complex. First, she was told that current tests are not 100% sensitive in detecting all mutations and, therefore, it is still possible that she may carry a mutation. As well, she was told that her suspected mutation may be in an as yet unidentified breast cancer gene. Finally, she was told that there is a possibility that her breast cancer may be a sporadic case within a hereditary breast cancer family. The authors felt, however, that this explanation is unlikely because she was a first-degree relative of a male breast cancer case and she was the youngest in the family to receive a breast cancer diagnosis. 37 Hence, she was advised to have other affected family members tested and that, if their tests were negative, it would be highly unlikely that all the tested women in this family would develop sporadic breast cancer considering the high occurrence of cancer in their family in comparison to what is expected within general population. Upon this, she would be recommended to enter additional research studies, such as those offering linking analysis. The above case clearly presents some of the complexities that were faced by the genetic team at the Hereditary Cancer Program when the participants of the current study received their "uninformative" genetic test results. While Peshkin et al. (2001) recognise that the time-consuming nature of providing comprehensive services may make it difficult for one genetic professional to provide all the care, they recommend referral to an oncology nurse or a nurse specialising in cancer genetics, because they have ongoing contact with affected individuals through regularly scheduled follow-up appointments. To conclude this section on challenges to genetic testing and genetic counselling, one area that further complicates the making-sense process of individuals receiving uninformative results is how to use such results in their medical management. The following section reviews interests in genetic testing and intended use of test information. Interest in Test Information Research involving individuals considered at risk for HBOC suggests that substantial proportions are interested in genetic testing for breast and ovarian cancer susceptibility. Comparison between two American studies on this subject showed significant differences in the level of interest between individuals recruited from the general population and individuals recruited from a breast cancer centre who had at least one first-degree relative with a past breast and/or ovarian cancer diagnosis. The latter study, by Lerman et al. (1995), showed a greater 38 interest among its study participants with a family history of cancer (91% or 96 out of 105 participants) compared with Tambor, Rimer and Strigo's (1997) study of the general population in which 69% of participants showed interest in genetic testing for breast and ovarian cancer susceptibility. Other studies, such as those by Lipkus, Iden, Terrenoire, and Feaganes (1999) and a Canadian study by Bottorff et al. (2002) produced comparable results. After controlling for differences in age, education, personal history of breast cancer, and knowledge of genetics, Bottorff et al. found that individuals with at least one first-degree relative with breast cancer were two times more likely to be interested in genetic testing for breast and ovarian cancer susceptibility than those with no family history. However, their research showed lower overall rates of interest than reported in both Canadian and American studies. Bottorff et al. reported an interest rate of 28.5% among the 761 individuals surveyed who had no personal history of breast cancer and a rate of 30% among the 260 individuals surveyed who had a personal history of breast cancer. However, as other studies on interest in genetic testing have indicated, lower interest may suggest that, despite media coverage, individuals from high-risk families are not sufficiently informed about the availability of testing to decide to use this technology (Andersen, Bowen, Yasui, & McTiernan, 2003; Bottorff et al., 2002). Finally, Donovan and Tucker (2000) conclude from their study that no relationship exists between interest in genetic testing and individuals with a family history of breast cancer. According to Tambor et al. (1997), sociodemographic factors common among members of the general population interested in genetic testing are being younger than 60 years and being Caucasian rather than African American. Tambor et al. hypothesise that younger women may be more interested in knowing their probability of carrying an inherited mutation in order to decide whether to have children, while older women may be more concerned about their risk of cancer 39 because of their advanced age. Bottorff et al. (2002) found that sociodemographic factors influencing interest in genetic testing were being less than 50 years old, having a positive family history of breast cancer, and having more years of education. Psychosocial factors reported to influence interest in genetic testing for breast and ovarian cancer susceptibility include having a sense of vulnerability to breast cancer (Press, Yasui, Reynolds, Durfy, & Burke, 2001), cancer worry, and beliefs that having testing will result in increased access to cancer screening (Durfy, Bowen, McTiernan, Sporleder, & Burke, 1999; Gwyn, Vernon, & Conoley, 2003). Beliefs related to breast care influence interest in genetic testing. Those who believe that mammograms will benefit them and their family, and that they will feel more in control over their health were found to be more likely to be interested in testing (Tambor et al., 1997). As with the results of work by Lerman et al. (1995), interest in testing was associated with acquiring information useful for other family members such as daughters. The authors conclude that interest in testing differs among individuals from different age groups, races, levels of schooling and use of health care. In light of what is known about sociodemographic factors influencing interest in genetic testing for breast and ovarian cancer susceptibility, both Lerman et al. (1995) and Bottorff et al. (2002) recommend that these factors be used in targeting community-based education about genetic testing. They also recommend that research be done to test different approaches to education and counselling based on specific characteristics of target populations. Although it can be hypothesised that when planning for cancer genetic services, interest in genetic testing will always be higher than actual uptake (Cappelli et al., 1999) I found very few Canadian studies that actually compared interest and uptake in the general population. Finally, Marteau and Croyle (1998) suggest that uptake rates for genetic testing are higher when there are effective ways of treating or preventing the condition. They report that an average of 50% of 40 those interested in genetic testing for breast and ovarian cancer susceptibility proceed with testing compared to 10% for Huntington's disease, for which there is no treatment and 80% for familial adenomatous polyposis, for which there is effective treatment. Intended Use of Test Information I found very few studies that looked specifically at the use of genetic testing for BRCA1 and BRCA2 with the intent to use test information to support health management decisions to reduce risk of breast and ovarian cancer among individuals considered at risk for HBOC. What is even more crucial is gaining empirical understanding of the use of genetic results to inform health-related decisions when test results are uninformative or are uncertain. Both of these contexts will be discussed below. One study on this topic by Meiser et al. (2000) looks at intent to pursue prophylactic bilateral mastectomy among women at risk of HBOC. The study found that 19% of the 333 women surveyed said that they would consider prophylactic mastectomy if the test revealed that they carry an inherited mutation in their BRCA genes. This was in contrast with the majority (47%) who said that they would not consider a mastectomy while 34% said they were unsure and 1% had already undergone a prophylactic mastectomy. Statistical measures from the Meiser and al. study showed a correlation between breast cancer anxiety and consideration of mastectomy. As well, the highest proportion of women considering prophylactic mastectomy were within the 30 to 39 age group (Meiser et al., 2000). Perceived risk as opposed to actual risk was also found to correlate with consideration of prophylactic surgery. Although these results concern hypothetical situations whereby the individuals are only viewed as having a theoretical risk to an inherited mutation compared to an actual risk, they nonetheless provide an overview to the context of having to make health-related decisions with inconclusive or uninformative genetic 41 testing results. Uninformative test results are of limited use in decisions about prophylactic mastectomy because they do not provide a definitive measure of genetic mutation status and of risk of cancer. Although some recent empirical evidences suggests that prophylactic bilateral mastectomy reduces morbidity and mortality among individuals at risk of HBOC (Meijers-Heijboer et al., 2001; Narod, 2002), further studies establishing this surgery's effectiveness are still needed (Burke et al., 1997; Meiser et al., 2000). Bouchard et al. (2004) also suggest that there are cultural differences that influence recommendations of prophylactic surgery. Their study showed that cancer geneticists in Montreal (Quebec, Canada) and in Manchester (Great Britain) were more likely to recommend prophylactic mastectomies than cancer geneticists from Marseilles (France). One study that provides prospective evidence of the usefulness of prophylactic surgery for unaffected individuals identified with a BRCA1 or BRCA2 gene mutation is by Meijers-Heijboer et al. (2001). Their results show that 3 years after surgery, among the 76 out of the 139 women identified with a BRCA1 or BRCA2 mutation who chose to have bilateral prophylactic mastectomies, none developed invasive breast cancer. This compares with the 8 women out of 63 who chose regular surveillance instead of surgery that did develop breast cancer. These eight women developed invasive breast cancer. While more studies are needed to determine the long-term protection of prophylactic mastectomy for women identified with a BRCA mutation, this next study does point in favour of the surgery to confer protection against the appearance of cancer. Among the 22 carriers of BRCA mutation diagnosed with a previous breast cancer before the age of 42, at 12 years of follow up, half had developed a second primary breast cancer (Haffty et al., 2002). The majority of cancers were classified as second primary tumours. 42 While not all genetic scientists would agree, Narod (2002) views it prudent to recommend bilateral mastectomy for women with a past breast cancer diagnosis known to carry an inherited mutation of BRCA1 or BRCA2. He also suggests the use of tamoxifen (a cancer preventing drug) to reduce the incidence of second primary cancers. Generally, health professionals presume that those who learn of a genetic risk will be more motivated to strictly follow breast cancer surveillance measures (Bredart, Autier, Audisio, & Geragthy, 1998; Lerman, Seay et al., 1995). Studies, however, have not conclusively demonstrated this assumption to be correct in all cases. Reactions to positive genetic test results may be more complex than was initially presumed. As demonstrated in the 1992 study by Kash, Holland, Halper, and Miller, high cancer anxiety led to poor adherence to a regime of breast examinations by health professionals and monthly breast self-exams among women with a family history of breast cancer. Although researchers hypothesize that women who carry BRCA1 and/or BRCA2 cancer gene mutations are able to reduce their risk of developing and dying from breast cancer by using preventive measures such as prophylactic mastectomy (Lerman, Hughes et al., 1998; Narod, 2002; D. L. Newman et al., 2004; Rebbeck et al., 2004), controversy exists about the efficacy of strategies for preventing cancer in carriers (including prophylactic surgery) and strategies for managing surveillance of cancer (Bredart et al., 1998). Such controversies are doubts in the efficacy of breast cancer mammography in women under the ages of 40 as well as the lack of prospective studies showing a protective use of tamoxifen, prophylactic mastectomy and prophylactic oophorectomy in reducing the risk of breast and ovarian cancer. Other associated controversies with prophylactic surgeries are complications of premature menopause, and the few studies that have addressed the quality-of-life issues in women who have opted for 43 prophylactic surgeries (Eisen, Rebbeck, Wood, & Weber, 2000). Fasouliotis and Schenker (2000) believe that more studies are needed to assess the efficacy of screening strategies in reducing morbidity and mortality for breast and ovarian cancer among high-risk individuals. Certainly more studies are needed about women's interest in using genetic testing to support and inform health-related decisions. As put forth by the Bottorff et al. (2002) study, the most frequently cited reason for genetic testing among the 260 women with a past breast cancer diagnosis surveyed was for simple curiosity. Bottorff et al. posit that this finding might reflect individuals' expectation that genetic testing will tell them categorically whether they will get the disease or not rather than providing a probability of cancer risk based on genetic status. Another reason for testing was to warn family members and to take preventive action, although specific actions were not mentioned by survey participants. The utility of uninformative genetic test results in assisting individuals with risk management decisions is undocumented. However, one recent study suggests that such test results may lead to decisions based on incorrect assumptions in individuals' personal interpretation of their results (Frost et al., 2004). This study by Frost et al. took place in the High Risk Breast Cancer Clinic at Huntsman Cancer Institute, University of Utah. The clinic is both a research facility and a clinic for individuals with a family history of breast and/or ovarian cancer. Frost et al. conducted a qualitative pilot study using focus groups and individual interviews with 15 women enrolled in their breast-cancer surveillance program. The study examined how women made health-related decisions when provided with uncertain genetic test results for BRCA1 and BRCA2. Based on the type of uncertain results, women were enrolled in three different groups: Group 1 (n=6)—affected women with breast and or ovarian cancer who received results indicating a variant of uncertain significance; 44 Group 2 (n=4)—unaffected women found to carry an inherited mutation; and Group 3 (n=5)—affected women considered at risk for HBOC but for whom tests did not identify an inherited mutation. The authors define uncertainty in group 3 as an absence of explanation for their personal and family history of cancer, their future risk of cancer and that of others in their family members. The authors report that women who received their results face-to-face understood the difference between uncertainty in test results and uncertainty in their risk of developing cancer. Thus, understanding such a difference in their risks as well as having the opportunity to discuss the implications of uncertain results in light of having to make health-related decisions facilitates the making sense process. While the authors did not report what were the health-related decisions to be made, they nonetheless point to the notion that when inconclusive results are unexpected and misunderstood, individuals are not able to use genetic information for health care decision making. Clearly, more empirical studies are needed to assess how individuals receiving uninformative genetic testing results are able to use such information to assist them in making risk management decisions. Psychosocial Issues Psychological impact of having a family history of breast and/or ovarian cancer Learning that one has a genetic predisposition to breast and ovarian cancer is stressful and there is potential for adverse psychological outcomes (Esplen et al., 1998). However, Kelly et al. (2004) note from their study that once the initial shock of learning one's genetic status has passed, this news may be accompanied by a decrease in an overall general stress, although most studies point to the contrary (Friedman et al., 1999; Lynch et al., 1993). One type of stress, however, that does not dissipate with time is living with a family history of cancer considered at 45 risk for HBOC. This stress tends to recur in each new generation with each new cancer diagnosis. Pasacreta, Jacobs and Cataldo (2002) point out that anticipation of breast cancer development among unaffected individuals from known hereditary cancer families may be more psychologically distressing than actually having or having had breast cancer. Individuals from families with histories of cancer are known to experience symptoms of general distress, have frequent intrusive thoughts and can at times deny their risk of cancer (Zakowski, Valdimarsdottir, & Bovbjerg, 2001; Zakowski et al., 1997). Bish et al. (2002a) document that distress associated with increased risk perception to cancer can have detrimental effects, such as failing to interpret appropriately genetic test results (Lerman, Lustbader et al., 1995), reduced participation in screening and surveillance programs (Kash et al., 1992) and excessive breast self-examination. Consequently, when individuals enter a clinic for genetic counselling and genetic testing, they may be already experiencing distress related to their own cancer and genetic risks, and that of others in their family. As for Lerman, Seay et al. (1995), they support that affected and unaffected individuals with a family history of cancer indicative of HBOC do live with an overestimation of their risk to develop breast cancer. Overall, women, unaffected and affected, with a HBOC experience multiple psychological stresses ranging from increased perception of risk for cancer to increased perception of carrying mutations that will increase the cancer risk of family members (Bish et al., 2002a). Benefits and limitations of genetic testing for breast and ovarian cancer susceptibility The introduction of genetic testing for breast and ovarian cancer susceptibility was accompanied by much enthusiasm among many families at risk of HBOC. They could finally learn whether they had a heightened risk of cancer and whether they would transmit mutated genes to their children (Lerman, 1997). Other perceived benefits were learning one's genetic risk 46 status for breast and ovarian cancer susceptibility, planning for appropriate treatments if found with a mutation, providing information to relatives who want to know of their risk for an inherited mutation, helping to decide about life and disability insurance, and assistance in making lifestyle changes to prevent cancer (Cappelli et al., 1999). While there are benefits to genetic testing such as assisting at risk families make medical decisions and promote cancer risk-reducing behaviours (Lerman, 1997) there are, as well, unfavourable implications to learning one's genetic status. That is, while learning more precisely about one's potential lifetime risk of developing breast cancer by knowing if one is carrying an inherited mutation has been perceived as a benefit (Biesecker, 1997a; Sagi, Kaduri, Zlotogora, & Peretz, 1998; Schwartz et al., 2002), this benefit can also be attenuated when individuals experience estrangement from family members with differing results, experience guilt over the possibility of having transmitted a mutation to their children, lose hope and, experience anxiety and depression (Biesecker, 1997b; Lynch et al., 1997). For noncarriers within hereditary breast cancer families, a hypothesise benefit is feeling relief in discovering that their risk of developing breast cancer is the same as that of the general population (Lerman, Hughes et al., 1998). Nevertheless, because genetic testing is in its early years and most of the research in this area has focused on those who discover that they have a genetic mutation, the lack of empirical studies of individuals who tested negative to breast and ovarian cancer susceptibility does not permit us to draw conclusions of possible similarities from this group to the uninformative group about the degree to which relief from anxiety and uncertainty can be found. However, one recent study assessed the difference in cancer-specific distress and perceived risk for breast and ovarian cancer between individuals from known hereditary breast cancer families who tested positive and those who tested negative (Schwartz et 47 al., 2002) as well for individuals who received uninformative results. The authors had hypothesised that, since an uninformative result was determined as inconclusive to an individual's genetic status to an inherited mutation, they did not expect to see significant reductions in distress or perceived risk. Both of these variables were assessed through a structured telephone interview which was then followed by a face-to-face counselling session. Among their participants, 78 tested positive, 58 tested negative and, 143 received uninformative results. No significant difference was found from baseline to 6 months following disclosure between the positive group and the uninformative group on the mean scores of distress outcomes. This study, however, did find that cancer-specific distress and perceived risk for breast and ovarian cancer had significantly decreased for individuals found not to carry an inherited breast and ovarian cancer gene mutation. Hence, the uninformative group followed similar path as the positive group and did not exhibit a decrease in distress as reported among relatives who received a negative test results from known HBOC families. Another benefit foreseen by individuals going for testing is learning more precisely about the risks for their children or future children to inherit a mutation, and deciding about whether or not to have children. The desire for reassurance can be powerful but can also be counteracted if test results are not as expected (Claes et al., 2004). Hence, during genetic counselling, emphasis should be placed on the fact that, if an individual has a mutated gene associated with breast cancer, then their risk of transmitting it to their children would be 50%. Also, the possibility exists that individuals will learn that they did, in fact, transmit a mutated gene to their children. Reactions to such situations include self-blame for passing the genetic mutation and feeling guilty for failing to protect one's children (Biesecker, 1997b; Claes et al., 2004; Lynch et al., 1999). Although the precise effect of testing for breast cancer 48 susceptibility on family relationships is not yet fully understood (Macdonald et al., 1996), some effects have been documented among individuals such as daughters, siblings, parents, and first-or second-degree relatives. Individuals may unwillingly learn more about their own susceptibility following the revelation of the genetic status of a family member (Bredart et al., 1998; Croyle, Achilles, & Lerman, 1997). Family members may pressure each other to go ahead or not to go ahead with genetic testing (Lynch et al., 1999). Adverse effects on an individual's self-concept is another potential risk of genetic testing (Biesecker, 1997b; d'Agincourt-Canning, 2003). How mutation status disclosure may influence the quality of life of an otherwise healthy person found to be at increased risk of developing cancer in the future is largely unknown. In addition, disclosing genetic information may cause people to worry about when the disease will develop, as it has been observed that some individuals equate having an inherited mutation of their BRCA cancer genes with receiving a cancer diagnosis (Welch & Burke, 1998). While early disclosure of one's genetic status may prompt some to be more regular with their cancer screening, early disclosure can also interfere with an individual's present life and goals for the future (van Zuuren et al., 1997). Further, Croyle et al. (1997) comment on that offering information about the probability of developing cancer can complicate the task of maintaining a healthy concept of self. Individuals who live with the perception to being at risk of cancer because of a known mutation in their family and who learn that they do not carry this mutation may have difficulties incorporating into their self-identify that they are not at extremely high hereditary cancer risk (Lynch et al., 1999). Noncarriers may also no longer see themselves at risk or feel that their risk is below average. According to Lerman, Daly, Masny, and Balshem (1994), 72% of the women they studied who are at risk for hereditary breast cancer expected to go on worrying even if they received a result 49 indicating that no mutation had been found. Hypothetically, individuals from known hereditary breast cancer families who decline genetic testing may be doing so to protect their sense of wellbeing or because they feel that proven strategies for preventing cancer are still lacking. While there needs to be an understanding that not all health problems are the result of heredity, there may be a societal tendency towards genetic reductionism - providing simplistic medical explanations to multifactorial, social problems such as in the case for breast cancer when breast cancer is a multifactorial disease. Nelkin and Lindee (1995) labelled this trend genetic essentialism - reducing the self to a molecular entity and discounting the environmental context in which our genes exist. The risk is that genetic essentialism may create new forms of social eugenics and lead to new social classes based on genetic inheritance (Macdonald et al., 1996). Macdonald et al. (1996) believe that holding such a restrictive view of genetic essentialism may result in genetic testing one day becoming mandatory, so that people may lose their right to make their own decisions if they want testing or not. Genetic essentialism could also lead to discrimination, as it has in the past for certain social groups (for example, Black Americans at risk of sickle-cell anemia) (Macdonald et al., 1996). Other limitations of genetic testing for BRCA genes include the degree to which test results are uncertain (Frost et al., 2004; van Zuuren et al., 1997). From their study evaluating uncertainty in genetic counselling process, van Zuuren et al. (1997) describe the distressing situation in which individuals and families found themselves when they were looking for reassurance from genetic testing but instead received uncertain results, van Zuuren et al. note that this inconsistency where individuals consult experts to get certainties but receive uncertainties with their genetic consultation creates a bad match. This incongruence is further exacerbated when the intent of testing is to support health decision making. 50 Perceived risks There are many studies showing that despite receiving genetic counselling and testing to correct individuals' actual risk to an inherited mutation and risk to cancer that their perceived risks remained unchanged. A study at North York General Hospital and Toronto-Sunnybrook Regional Cancer Centre in Ontario evaluating the impact of positive test results for BRCA1 and BRCA2 mutations found that perceived risk of acquiring a second cancer among 18 affected individuals increased from 46% to 57% after receiving test results. As for 6 unaffected individuals, their perceived risk for cancer increased from 27.5% to 47.5%. These findings indicate affected and unaffected people who received positive results experienced an increased perception of cancer risk while Schwartz et al. (2002) study findings indicate that people within HBOC families who received negative results experienced a reduction of perceived risk. As well, those who perceive they are at increased risk of cancer because of their strong family history may experience similar stress as those who received confirmation of their mutation (Baum, Friedman, & Zakowski, 1997; Burgess & d'Agincourt-Canning, 2001). Perceived risk can also increase not only after receiving positive results but also following genetic counselling. Bish et al. (2002a) found that after genetic counselling, perception of risk for ovarian cancer increased among affected women from high-risk families. In addition, some women's perceived risk for cancer increased while they were waiting for their results of testing. Changes in perceived risk for breast and ovarian cancer and to risk of carrying an inherited mutation have yet to be documented for individuals who received uninformative genetic test results. Nursing implications of genetics Genetic discoveries made within the last two decades have slowly begun to affect Canadian genetic nursing. Although still relatively small and not yet fully recognised, genetic 51 nursing is a growing health care specialty within Canadian nursing (Bottorff et al., 2004; Green, 2004). Following their interviews with 22 nurses in genetic services for adult-onset hereditary diseases, Bottorff et al. (2004) report that challenges for genetic nursing in Canada include role ambiguity, lack of recognition for nursing expertise, limited availability of genetics education, isolation, and the instability of nursing positions. Because the first point of entry into the health care system for the general population is often through nurses, there is a need to provide educational support to nurses so that they can better assist individuals as they make sense of their experiences with genetic testing. While there is currently an international effort to develop education in genetics, there are as well many issues for leadership in education in genetics. Some of these issues are the increased need to incorporate genetic knowledge by all nurses; to convince administrators and faculty of the need to add more genetics content in nursing programs; and to support nurse leaders in genetics to disseminate the importance of genetics as an essential content for health care throughout the world (Feetham, 2001). At best, what is needed from nurses working in all settings is a basic understanding of genetics to know when referral to cancer genetics agencies may profit their patients. Lashley (2000) says, "Minimally you want to develop professionals who can think genetically when approaching a clinical situation or problem that on the surface may not appear to be genetic" (p. 797-798). The International Society of Nurses in Genetics advocates that nurses recognise the integral component of genomics within their role of promoting health and wellbeing (International Society of Nurses in Genetics, 2003). Responses from 975 Canadian nurses to a survey regarding their educational needs in the areas of genetic testing and adult-onset hereditary disease indicated that they perceived themselves to be inadequately prepared to provide genetic care (Bottorff et al., 2004). Educational needs include not only the psychosocial needs and 52 services related to genetic testing but also how to provide support in interpreting results, as well as the ethical, legal, and social issues associated with genetic testing (Pasacreta et al., 2002). Summary This chapter reviewed various implications of genetic testing for BRCA1 and BRCA2 mutations. Empirical studies about the impacts of genetic testing on people's everyday lives are scarce, although more are beginning to appear. A better understanding of the impact of receiving uninformative genetic test results for BRCA1 and BRCA2 mutations is needed to counsel these individuals appropriately. In the chapters that follow, I endeavour to contribute to this lack of empirical knowledge. 53 C H A P T E R T H R E E : Research Design and Implementation My interest in nursing practice and the type of knowledge that is useful in guiding our practice led me to examine methodologies suited to the objectives of my research. Because my intent was not to look at how individuals' discourses can be constrained or regulated by social processes beyond individual's consciousness, I believed that traditional qualitative methodology such as that of critical social theory and grounded theory limited the exploratory nature needed to answer the research questions. I needed an approach that acknowledged human subjectivity and interpretation in people's health and illness experiences. Moreover, the complexity of clinical nursing requires that knowledge generated in nursing research be not only theoretical but also inform nursing practice. I, therefore, needed an approach that provided guidance towards producing practical nursing knowledge. Because of the interdisciplinary nature of nursing, nurses are often called upon to work in innovative, evolving clinical settings where little nursing knowledge exists to guide their practice. Genetics is one such domain, where the nurse's role is becoming more visible and expanding (Thorne, 2001). Therefore, nursing research that informs practice in this area is greatly needed, especially knowledge grounded in context and in human subjectivity that has the potential to bring understanding to lived experiences. The challenge of describing and interpreting shared realities such as health and illness experiences from the perspective of those who live them requires a flexible research process that does not constrain participants' liberty to express their views, without being a priori categorised. A qualitative methodology that offers such a flexible process to study shared realities without 54 losing sight of the individual is interpretive description - a noncategorical qualitative alternative for developing nursing knowledge (Thorne et al., 1997). The present qualitative inquiry, then, is guided by interpretive description. The concepts most commonly associated with interpretive description are inductive research, subjectivity of experience within common experiences of health and illness, and contextualising particular cases in an everyday-life perspective (Lowenberg, 1993; Thorne et al., 2004; Thorne et al., 1997). Interpretive description has been used as an analytic framework by researchers and scholars who find they can use it to create practical nursing knowledge out of shared human experiences and unique individual experiences (Buissink-Smith & Mcintosh, 1999; Irwin, Thorne, & Varcoe, 2002; Ramfelt, Severinsson, & Lutzen, 2002; Stajduhar et al., 2002). In this chapter, I begin by presenting the analytical framework of interpretive description, explaining how this approach is distinctively useful to nursing inquiries. I then describe the general principles guiding my study. I begin by reporting on recruitment strategies and participants selection criteria. I go onto describe the data sources, commenting on interviewing and on field notes. I then outline the data analysis procedure used to arrive at my research findings. I conclude with a discussion of the standards and credibility measures I employed to establish the scientific integrity of the research process and its ethical conduct. Interpretive Description of Health and Illness Experiences Theoretical framework Thorne et al. (1997) argue that interpretive description is a valid form of qualitative inquiry for understanding how people experience their health and illness, so that nursing practice can be adapted to better meet the needs of these patients. Just as traditional empirical methods such as quantitative studies prove limited when used in inquiries of human subjectivity, so do 55 qualitative approaches developed in other disciplines other than nursing.21 A noncategorical qualitative approach called interpretive description as been proposed as an alternative (Stajduhar et al., 2002; Thorne et al., 2004; Thorne et al., 1997). Like the constructivist approach, the interpretive description approach recognises the contextual and constructed nature of health and illness experiences of individuals who come into contact with clinical settings and relates how clinical context can influence an individual's subjective interpretation to their experience. The goal of interpretive description is not to produce new truths, but to identify common clinical phenomena that, within individuals' contexts, bring about new understandings and new meanings (Thorne et al., 1997). The philosophy of interpretive description shares perspectives with traditions of interpretive and naturalistic inquiries (Denzin, 1989; Lowenberg, 1993; Schwandt, 1994). Epistemological foundations underpinning interpretive description include the view that, because reality is multiple and constructed, it can only be studied holistically (Thorne et al., 2004). That is, in-depth understanding comes from studying the phenomena within multiple realities, not from viewing the phenomenon extracted from the reality in which it occurred. The knower and the would-be knower are inseparable; they interact to influence one another. Because no a priori 2 1 Whi le many qualitative approaches developed from different disciplines do not seem entirely different from other qualitative research methodologies, there are main differences within the choice of phenomenon studied. Grounded theory was created by Glaser and Strauss (Glaser & Strauss, 1967) and originated in sociology for the study of social processes. Interpretive description was created by Thorne, et al. (Thorne et al., 1997) and originated in nursing for the study of clinical phenomena. Whereas grounded theory has been used to study clinical phenomena in the past, its purpose differs from interpretive description. Grounded theory aims to generate a theory by creating a schema of the phenomenon through the development of concepts. Interpretive description aims to produce a synthesis of the main themes and patterns of a phenomenon studied. Whi le interpretive description may still contribute to generating theories whereby analysts may test concepts derived from such studies to further bring refinement of plausible relationship between them, its purpose is not focused on theory development per se (Stewart, 1998). I chose interpretive description to guide my study as I believed that too little was known of the current study population "the uninformative group" to be able to pre-determine the value of generating a theory of the underlying social processes related to the experience of making sense of interpreting genetic testing results. 56 theory can encompass all the possible realities of a studied phenomenon, interpretive description encourages theory grounded in the data. Researchers who choose interpretive description believe that no research is value-free; rather research is socially constructed, historically located and shaped by both the researcher's and the participant's beliefs and emotions (Denzin, 1996). Emotions and beliefs permeate all aspects of interpretive research: personal lived experiences, the home and family, health and illness experiences, and in power relations. Power is viewed as the ability to define what is legitimate. The aim of interpretive description is not to determine how these emotions, beliefs and power correlate with the phenomenon studied, but to acknowledge their influences on how an individual will experience a clinical procedure. Interpretive description focuses on how the studied clinical phenomenon occurs, how individuals' interpret their clinical experiences, and how patterns are formed and altered throughout the making-sense experience. In interpretive description, the individuals studied are not perceived as objects of research but as collaborators in the research process. Consequently, unlike traditional post-positivist approaches to the building of knowledge, researchers working within an interpretive description approache engage in the subjectivity of participants to understand and express more effectively the emergent patterns (Denzin, 1989). This reflexive, engaging, dialectic approach to research is said to reduce this objectification of study participants by paying attention to power inequalities between knower and would-be knower (Guba & Lincoln, 1994). Engaging also refers to the reflective process that I engaged in within the study. Being aware that I could not remove all power inequities between myself and the participants, I made explicit to participants at the outset of interviews as well as in our first contact by telephone that this research was part of my doctorate. I shared with them that I do not have a family history of 57 breast and/or ovarian cancer, nor have I shared this experience through someone close to me. I did not pretend to comprehend fully where they are coming from in interpreting and making sense of their genetic test results. I did share, however, my many years of reading on the subject, observing genetic counselling sessions, participating in clinical reviews of individual genetic testing cases, and participating in weekly discussions with a group of researchers conducting inquiries in the ethical, social, and legal implications of genetic testing. I tried to decrease power inequities by emphasising the importance of their experiences to my learning and, more importantly, to the improvement of genetic services for those coming after them. Concomitantly, I was still aware that complete sharing of power within the research process was impossible, as I defined the orientation of the interviews by making participants aware of the focus of my study and I made the final decisions of what would be considered valuable knowledge. Research strategies in interpretive description draw upon work in grounded theory, naturalistic inquiry and ethnography, and on phenomenological approaches for data collection (Thorne et al., 2004). Such strategies include purposive and theoretical sampling that reflects an awareness of possible structures that may bring variations within the phenomenon studied. Data analysis begins during data collection and researchers locate emerging patterns within the data and compare them to existing knowledge on an ongoing basis. The end product of interpretive description is a synthesis of the main patterns and processes in the phenomenon studied. Individual stories are used to illustrate differences and similarities, thereby rendering individuals visible within identified patterns. This form of clinical knowledge development appeals to those working directly with patients because it provides insights on how to apply aggregate knowledge to individual cases (Thorne et al., 2004; Thorne et al., 1997). 58 The write-up of individual cases and common patterns is thick description and interpretations derived from individuals' lived experiences (Denzin, 2001).22 Thick description, as opposed to thin description that simply states facts, provides the context of the phenomenon explored; uncovers meanings that organise experiences; and traces the phenomenon's evolution. Thick description presents meanings and feelings as a text that can be interpreted (Denzin, 1989). The interpretive description approach recognises that lived experiences reflect the social world in which the individual lives. Interpretation aims to illuminate the meanings and concepts that organise a person's experiences of health and illness (Denzin, 1989). This aim requires that the researcher be an informed reader of the phenomenon studied. That is, unlike phenomenology, interpretive description researchers locate their findings within existing knowledge, so that the new knowledge can be linked, and contrasted, to the work of others in the field (Thorne et al., 1997) and to the researcher's prior understanding of the phenomenon (Denzin, 1989).23 In addition, Thorne et al. (1997) argue that an analytic framework constructed out of a sound critical analysis of existing knowledge builds a stable platform from which to launch a qualitative inquiry. Research Design A l l the above features of knowledge construction created the theoretical foundation for my methodological approach to studying how people interpret and make sense of their uninformative genetic testing results and how these results are lived in their everyday lives. This approach directed me to: 1. Design a preliminary interview schedule based on existing knowledge but leaving space to reformulate the initial interview schedule. 2 2 See appendix 8 for a sample of initial analysis of key statements with thick description and interpretation. 2 3 M y personal assumptions about the phenomenon studied at the outset of the study and my evolving understanding are injected throughout the report of research results in chapters 4 to 6. 59 2. Recognise the importance of thematic outliers as evolving selection criteria for study participants. 3. View participants' experiences as permutated by past and present lived experiences, including family relations and interactions, and by subjectivity reflecting complex cognitive and emotional interactions. 4. Engage in the subjectivity and interpretation of participants' narrative accounts so as to clarify my understanding of the lay theory developed by individuals to render their experience meaningful. 5. Construct knowledge in collaboration with participants by sharing with them patterns and themes in the data. 6. Seek out ways in which aggregate processes can further enhance understanding of individual cases. 7. Consider how clinical knowledge derived from the study can be applied in the development of nursing practice. Implementing the Study Having set out the methodological approach that guided my research design, I now turn to the methodology itself. I will also detail the interpretive lens that guided the research and analysis process as well as document my efforts to fully engage in inductive reasoning by reflecting on the research analysis and processes. In this study, the primary participants are individuals who had genetic testing for BRCA1 and BRCA2, while the secondary participants are family members chosen by the primary participants who did not have genetic testing (except for one). In accordance with interpretive description, in-depth interviews, field notes and clinical documents were my primary data sources. Transcription of interviews and field notes was 60 ongoing during the research process, to facilitate location of emergent themes within the findings. Recruitment of participants Primary participants were recruited for this study with the assistance of genetic counsellors and the educational nurse at the Hereditary Cancer Program of the BC Cancer Agency. Based on initial sampling criteria presented below, the counsellors and nurse made the initial contact with potential participants,24 who had already undergone genetic testing for BRCA1 and BRCA2 mutations, had been informed of their results through a telephone call from one of the health professionals from the Cancer Agency informing them that no mutation had been identified in their BRCA1 and BRCA2 cancer genes, and had received a letter informing them of these results. In this initial contact, the counsellors and nurse explained my study and sought their participation. If the person was interested to hear more about the study, they were asked for permission to release to me their names and telephone numbers. If they agreed, I telephoned them to describe further the goals of my study, what their involvement would be and seek their participation. If they voiced interest, I encouraged them to take time to consider participating. In the meantime, I sent them an information letter and an informed consent form that further outline the aims of the study. I told them that I would contact them a week following the receipt of the letter to determine their interest in taking part in the study. I also explained that they could refuse to participate or withdraw from the study at any time. If they agreed to participate, verbal consent was obtained one week following the receipt of the letters and written See Appendix 3: Letter of support for recruitment from the B C C A Hereditary Cancer Program. 2 5 See appendix 4 for a copy of the participation information and consent form. The information letter and consent form was sent to all participants via express post to maintain respectable timeline between voicing an interest in participating and in conducting the actual interview. 61 consent at the day of the interview. Al l interviews were conducted within 2 weeks of receiving verbal consent. Potential recruitment of family members for the purpose of answering more in depth Research Question 2 was discussed with primary participants after they had received the letter of information and informed consent form. They were also told that they were not obliged to identify potential family members in order to participate in the study. I explained that a significant family member could be an individual who they felt supported them through their genetic testing experience or someone interested in their experience. If they could identify anyone, I asked them to make the initial contact, and seek permission for me to contact the person. Only at the interview did I obtain names and telephone numbers of potential secondary participants. Among primary participants who had identified their partner as their significant family member (n=3), all partners chose to be interviewed on the same day as the primary participant. As well, one primary participant brought her sister on the day of the interview, as she also wanted to participate. Written informed consent was obtained from the sister when I conducted a separate interview with her. Of 23 potential participants contacted, only one declined while another asked to postpone her interview because she was receiving treatment for a recurrent breast cancer at the time. She asked that we keep in contact every week or so to see if she had enough energy to set a tentative interview date as she insisted on participating. I kept in contact with her for over 6 months. The last time I called, her husband shared with me that she had passed away from recurrent breast cancer but had been very sorry she had not been able to participate in my study. I continued recruiting participants until I heard consistent repetition of themes in the interviews. As well, as I neared 20 interviews, I heard few new insights in participants' stories. 62 That is not to say that their stories were completely similar, but that I was gaining very little new knowledge; hence, I felt that I had achieved thematic comprehensiveness. In interpretive description, small samples are considered ideal for inquiries of clinical phenomena where the purpose is to capture themes and patterns within participants' subjective perceptions (Thorne et al., 2004). However, the sample must be large enough to portray participants' experience well enough to inform clinical practice. As for recruitment of secondary participants, I had intended to recruit an average of five family members per primary participant. In the end, I had an average of one secondary participant per primary participant and the maximum was three. The low number of family members per primary participants reflects an analytical decision I made during the course of data collection and analysis. Early in the data collection phase I realised that the family members' accounts of interpreting and making sense of test results were quite different from the accounts of those who had actually lived the experience as the former could only anticipate how they might have lived such an experience. Hence, I no longer emphasised the need for primary participants to identify significant others unless they really felt that someone made a difference in their experience. I had also realised that the content of the secondary participants' interviews contributed only minimally to the research and mostly to answer Research Question 2, assuming that their everyday lives of primary participants included interaction with significant others who helped them through their experience. Hence, I subsequently focused on the primary participant data and included secondary participants' data in the analysis only when it added depth to the primary participants' accounts. 63 Participant characteristics The final research sample comprised 21 primary participants and 15 secondary participants.26 Table 1 gives baseline demographics of the 21 participants and table 2 summarises the relationships between family members interviewed and the study participants. The education levels of primary participants varied: 7 1 % had at least a college education, while 29% had completed high school only. Of the former, 57% had completed at least one year of university level or obtained a full degree. All participants were women. Since the main focus of analysis is the primary participants' accounts of their experience of interpreting and making sense of their genetic test results, participants, refers only to primary participants. When input from secondary participants input is included, I specifically say so and address them as the family members interviewed. 64 Table 1: Demographic Profile of Primary Participants Sex Birth place Female 21 Canadian-born 18 Male 0 European-born 3 Age No. of offspring 41-50 10 None 4 51-60 8 One 1 61-70 2 Two 8 Older than 70 1 Three 8 Age at breast/ovarian cancer3 Marital status 20-30 2 Single 2 31-40 7 Married/common law 15 41-50 9 Separated/divorced 4 51-60 2 Level of education 61-70 0 High school diploma 6 Older than 70 1 College diploma 3 Employment Status At least one year of university 12 Full time 10 Residence Part-time 4 Greater Vancouver13 14 Retired 4 Fraser Valley 0 7 Unemployed 1 On leave from job 2 Note. Total number of participants (N) = 21. 65 a3 participants had more than one primary breast cancer. bGreater Vancouver areas include participants from Vancouver, North Vancouver, West Vancouver, Coquitlam, and Langley. cFraser Val ley area include participants from Surrey, Port Coquitlam, White Rock, Richmond, Maple Ridge, and Delta. Table 2: Relationship of Secondary Participants (SP) to Primary Participants (PP) Relationship Specifics Total Identification of PP with SP Spouse/Partner Sibling Husband Sister Child Daughter Extended family member Female Cousin Niece Total number of secondary participants (family members interviewed) SP1/PP15 SP1/PP16 SP1/PP20 SP1/PP1 SP1/PP2 SP1/PP10 SP1/PP5 SP2/PP5 SP1/PP18 SP1/PP21 SP2/PP3 SP1/PP3 SP2/PP5 SP1/PP22 SP2/PP20 15 66 In order to provide a clearer portrayal of those who participated in the study, I built a pedigree for each of the participants from information gathered during the interviews (see Appendix 5) as well as from the demographic data collected at the end of each interview.271 refer to these pedigrees throughout the results chapter as a visual means to orient readers to the study findings. Permission to build pedigrees was sought from the participants. I told them that I would omit familial characteristics that might disclose their identity. Participants were quite enthusiastic about the idea and even offered their medical files to help build their pedigrees. However, wanting to be clear about my position, I explained to them that my role was one of researcher, not nurse, and that this did not give me permission to access their medical files. Some participants still associated me with the health professional from the BC Cancer Agency who had first contacted them on my behalf. Some commented that "I was the first they had heard from the Cancer Agency since they had been handed their genetic test results". Even after when I clarified my position, many participants said how good it was for them to express how they had felt about their genetic test results. Selection criteria In this research, the first selection criterion consisted of participants who had undergone genetic testing for BRCA1 and BRCA2 mutations and received results considered uninformative or inconclusive by clinicians. Although the clinicians and health professionals working in the BC Cancer Agency call these results uninformative, they are also called inconclusive in some of the literature on genetic testing. One of my dissertation committee members made the comment that I should not assume that participants share the view that their results are uninformative. If my aim was to seek how they interpreted their results and made sense of them, I had to avoid introducing these results as the Cancer Agency defined them. Very few participants actually 2 7 See Appendix 10 for the demographic data collection form used to help build the pedigree. 67 referred to their results as uninformative or inconclusive, although many spoke about the uncertainty of their test results. Because I viewed the results as inconclusive in terms of their outcome, I struggled to stay true to the reflective process and power issues, such as not thinking that my perception was the most accurate when participants asked for my view of their results. The line between engaging with knower and would-be knower is vague. That is, I was always wary that participants might adopt my interpretation of their results and how I made sense of them when my intent was to learn how they interpreted and made sense of their results. I do share the view that groups from different professional and social backgrounds tend to interpret testing differently. Whenever participants asked what my position was I answered that I was most interested in hearing their view. My aim in selecting potential participants was to obtain as heterogeneous a group as possible and to include what may seem at first outliers, as seen in earlier times with older use of qualitative methods by nurse researcher trying to adhere to the dominant rules of normal sciences approaches that aim for their sample to be homogeneous (Thorne et al., 1997). Thorne et al. explain that excluding odd cases from an inquiry results in producing identical, quantifiable data that is devoid of almost all human subjectivity. Hence, the general principles of theoretical sampling are relevant to the design of research with an interpretive description approach when the purpose is building nursing knowledge. The variables to purposive and theoretical sampling consisted of those identified within the genetic testing literature as having potential to influence interest in uptake and interpretation of genetic testing for breast and ovarian cancer susceptibility. In the course of the interview process, characteristics other than those identified a priori came to light, such as differences among individuals who came from large families compared with small families. Hence, toward 68 the end of the middle of the data collection, a few additional individuals from large families were approached to participate in the study. I used the same procedure in establishing first contact as outlined above in the Recruitment section. Writing field notes soon after each interview facilitated my recognition of another variable influencing the studied phenomenon: having a mother with a past diagnosis of breast and/or ovarian cancer. Although I had already identified this variable in the literature review and selected it as a sampling variable, I had recruited only two such women, among the potential participants. Hence, I selected a few more participants meeting this criterion from the available candidate pool, numbering about 255. Further variables that potentially influence participants' interpretation and making sense experience of their results were identified through discussion with the genetic nurse at the Cancer Agency, as well as the genetic counsellors. They recommended certain participants because they had observed differences among their experiences of receiving and interpreting their genetic test results. In our discussions, I also shared with them potential influential variables I had observed in the genetic counselling sessions at the Cancer Agency (about 24 hours' worth). I also drew potential influencing variables from the monthly clinical review meetings I attended in the Hereditary Cancer Program between 1999 and mid-2002, when I then moved to a different province. These meetings provided a forum for discussion of difficult and unusual cases. Attending these meetings were medical geneticists, oncologists, pathologists, genetic counsellors, nurse educators, a medical anthropologist, a bioethicist, and graduate students interested in genetic testing. These meetings were chaired either by the director or the acting director of the Hereditary Cancer Program. All of these professionals contributed to the case management by making recommendations about treatment and follow-up care. These meetings 69 also allowed staff to share and discuss new knowledge on standards of genetics, clinical practice and ethical issues. The final variables for sample selection were: Age: Below and above 50 (Reason: Interest in and use of testing can differ between younger and older women); Gender: A l l women (Reason: Only one man in the available pool of candidate had received uninformative results); Children: With and without children (Reason: Common incentive for interest in genetic testing); Family history: Mother diagnosed with breast cancer and died as a result of the disease vs. mother still alive or died of cause other than breast cancer (Reason: Known to influence perception of risk for an inherited mutation); Size of family: Families of two or fewer siblings in two generations vs. more than two siblings in two generations, both from same side of the family where breast and/or ovarian cancer occurred (Reason: During dialectic data collection and analysis, it became apparent that differences occurred between the two, while their similarities brought about further distinctions to the phenomenon studied); Education: University vs. college vs. high school (Reason: Education level thought to influence individuals' interpretation of testing by assuming that college to university people might have been introduced to some more complex concepts of genetics compared to those learned in high school); Country of origin: Canadian-born vs. not Canadian-born (Reason: Differences in ethnicity might bring about different beliefs and experiences of testing); 70 Residence: Greater Vancouver vs. Fraser Valley (Reason: Proximity to testing site seen as possibly influencing testing experience); Marital status: Live-in spouse vs. single, divorced, or widowed (Reason: Influence of support system on genetic testing experience); and • Year genetic testing was done: 1997, 1998, 1999 or 2000 (Reason: See how events unfold over time until receipt of test results in 2000. Note: Receipt of genetic testing results for BRCA1 began in 1997 while results of BRCA2 testing began in 1998. Data collection took place in the year 2000). My aim in using sampling variables was to produce a guide for the genetic counsellors and the genetic nurse assisting me in recruitment. I aimed to have at least two individuals identified for each variable. Because the sample was being drawn from a large population (255), this goal was possible and attained. Data sources The main data source for this research was in-depth, open but focused and exploratory, interviews. Following each interview, I wrote field notes.28 The field notes were a valuable source of data to begin identifying key themes that serve to guide ongoing data collection, data analysis. My secondary sources of data were, as identified above, genetic counselling session observations, monthly meetings of the Hereditary Cancer Program group, and weekly meetings with interdisciplinary scholars working in the area of genetic testing. I conducted all of the interviews with the participants face-to-face, in participants' homes or in other places of their choosing. Nearly all interviews with family members were also conducted in their homes. The exception was telephone interviews, when either the individual lived far from Vancouver or in the United States. Most participants mostly lived in locales 2 8 See appendix 6 for a sample of field notes mad following interviews. 71 around Vancouver such as Langley, North Vancouver, West Vancouver, and Coquitlam. Others lived in Lower Mainland suburbs such as Surrey, White Rock, Delta, Abbotsford and Maple Ridge. The interviews with the participants ranged from an hour and a half to two hours while those with the family members ranged from half an hour to an hour. All were audiotaped (with 29 30 consent) and transcribed. ' All transcribed interviews were transferred into QSR N5 software for qualitative analysis for easier management and retrieval of data. As well, a database was used to keep participants' record of contact, field notes, and easy access to personal journal.31 A l l participants were asked to choose a code name to safeguard their anonymity. Many participants told me either on the first or second telephone contact that they were not sure if they had something valuable to share. Hence, at the interviews, I began by sharing the aim of the study to establish the direction of the interview. I then invited the participants to start by telling me about their breast cancer experience. With the family members, I asked that they start with how they felt about their family member's experience of genetic testing. Beeson (1997) indicates that storytelling by individuals who have lived a meaningful experience helps to create order, make sense of their lives, and construct action. Therefore, I welcomed first-person accounts as valuable data. The knowledge gained from the participants' storytelling added richness to my understanding of their experience. I transcribed half of the interviews, while the other half were transcribed by two self-employed transcriptionists. I chose to hire professionals as I felt that they probably had an ethical protocol for working with personal information from tapes. The issue of confidentiality was discussed with both transcriptionists. 3 0 See Appendix 11 for a reading of interview segments from a few participants. 3 1 See Appendix 12 for template of database recording. 3 2 See Appendix 2 showing preliminary interview schedule for both primary and secondary participants. The bolded content represents ongoing additions to the interview schedule as data collection progressed. Although I took these preliminary interview schedules to all interviews, I did not always look at them during the interview because I knew the content by heart and because I did not want to interrupt the flow of the participants' narrative. The very few Individuals who expected to be asked questions in a formal, directive manner felt more comfortable when I gave them possible topics that could be discussed in the interview. 72 A few participants expected direct questions and did not seem comfortable starting with the storytelling. Hence, with them, I would start by asking that they tell me about their genetic testing experience and then how they interpreted their results. With the exception of a few participants, most felt more comfortable talking first about their breast cancer experience, then later about their experience with genetic testing. To the interviews, I brought my list of topics to cover and the prompts to use if necessary, although the way I used prompts - if at all - varied 33 from one interview with another. ' Most participants needed little prompting. Since data collection and data analysis were done iteratively, data collection took over a year. I spent time immersed in the data interspersed with time in the field. This iteration is a principle behind many qualitative approaches. The inquiry was constantly refined through this iteration, testing and challenging emerging conceptualisations of the studied phenomenon (Thorne et al., 1997). Through writing my field notes following each interview, I reflected on key and recurring points mentioned by the participants, as well as new questions I asked that brought new perspectives. This activity led me to question some of my initial constructions and explore new constructions in subsequent interviews. I found myself asking questions such as: What led this participant to respond in this way? What am I hearing and what am I not hearing? What is different and similar within the interviews conducted thus far? I marked these insights as outstanding questions in my field notes, to be either tested or negated by the next participants I met. This reflection allowed further clarification of emerging key themes: I asked participants to reflect on what these emerging themes meant within their lived, health and illness experiences. This strategy further contributed to the validity of the findings. See appendix 7 for the list of prompts sent to the participants as part of the letter of invitation that accompanied the information sheet and informed consent form. 73 Data analysis An interpretative description approach requires an inductive analytical interpretation to data analysis. While I believe that experienced qualitative researchers would not see the need for more guidance to data analysis in interpretive description, than that put forth by Thorne et al. (2004), I felt that my limited experience in qualitative data analysis required me to find more detailed guidelines for qualitative data analysis. Reviewing methods used in other qualitative approaches that share similar philosophical principles with interpretive description (ethnography, grounded theory and phenomenology, among others), I believed that interpretive interactionism (Denzin, 1989, 2001) provided further guidance in my search to find themes and build patterns and commonalities without losing sight of individual variations within them. I was advised by experienced qualitative researchers to be cautious of any method that claimed to offer a recipe for data analysis because there is no such thing - qualitative data analysis tends to be reshaped by ongoing inductive reasoning. I also understood that qualitative data analysis can only be learned through experience and practice (Thorne et al., 2004). As such, although Denzin's interpretive interactionism (Denzin, 1989; 2001) provided me with some guidance to organise my initial data analysis, it was only when I began to actually analyse data that I took full notice of the intellectual juggling between data collection, data analysis, and generating valuable, plausible and credible findings. Hence, the techniques I used to arrive at an inductive, analytical interpretation, were borrowed among those recommended by Denzin for interpretive interactionism (1989)34 and from Thorne et al. for interpretive description (2004). In interpretive interactionism, as in interpretive description, the end goal is to produce a meaningful ' Some of the techniques to data analysis borrowed from Interpretive Interactionism (Denzin, 1989) are discussed in the next pages. 74 account of experiential knowledge. Thus, what I present below is my reflective, learned data analysis process to knowledge development of my studied phenomenon. The purpose of thick description and interpretation in data analysis as put forth by Denzin (1989; 2001) and by Thorne et al. (2004) is to tease out the meanings and feelings present in participants' experience. To begin this process, I first bracketed main themes within the participants' stories that spoke directly to how they interpreted their results and the structures that informed their belief towards their interpretation of results. Key themes were mainly recurring points within the interviews or ideas participants emphasised while telling their stories during the interviews. These themes ranged from two to as many as 10 lines or more of transcribed-verbatim accounts. I deliberately did not select one or two words to represent evolving themes, to avoid constraining the analytical process within specific categories. Because the study had as its main focus how affected women interpreted and made sense of their genetic test results, relevant segments of the transcripts were coded under such a category. However, this was not a means to limit my view of their experiences but for initial data management. The main analysis came later when all of the segments coded under this category were analysed for possible themes and structures that could help illuminate the meanings people gave to their experience. These emergent key statements served my initial attempt to describe and interpret the meanings found in participants' accounts. I asked myself, what are the intentions, feelings, beliefs of the participants within the key statements? To synthesise the participants' accounts of their making-sense process, I looked for lay theories participants might have used to understand their experiences. Denzin (1989) says that lay theories are often found in participants' stories. By looking for and analysing some of these lay theories, I was able to identify more concepts that 3 5 See Appendix 8 for a few examples of key statement segment. 75 structured their interpretation and making-sense experiences. One example of such a lay theory is how one participant explained to me that, within each passing generation, breast cancer diagnoses seemed to be occurring 10 years earlier in family members with each passing generations. She felt that this pattern was too prominent to be just coincidental or the result of chance and that there were too many similarities between the ages at which individuals were diagnosed with breast cancer to also be the result of chance. This lay theory grew out of her family history with the disease and her belief that she may be carrying an inherited mutation despite being told that no mutation was found. The dialectic approach facilitated my seeing initial differences and similarities in the participants' experiences. For example, I noticed that some differences depended on personal familial characteristics. Participants who came from large families and/or also who had a mother who had been diagnosed with breast and/or ovarian cancer tended to attach different meanings to some of the emergent themes influencing how they interpreted and made sense of their genetic test results. Once I located themes and structures within each participant's story, I formulated an initial description and interpretation of their meaning based on the context of each individual. 3 6 I then looked at the essential themes as a whole to begin understanding possible processes and patterns in participants' experiences and what was meaningful to them. Thus, I came to see a chronology to the participants' experience. That is, participants made it clear to me that they had not just "arrived at the decision to have genetic testing overnight," but that there are factors that led them to become open to this new science. One of the major factors is the implications to living with a family history of breast and/or ovarian cancer. Then, before they consider genetic See Appendix 8 for examples of initial analysis of key statements and the formulation of their description and interpretation. 76 counselling and obtaining genetic testing, there is a search for answers to the question "Why would I want such a test?" This first phase is in chapter 4. When they finish this first phase by deciding to get tested, they enter into the experience of genetic testing. This second chronological phase is represented in chapter 5. Perhaps, having had the opportunity to talk about how they felt with their genetic testing experience and how they made sense of their results, all the participants reflected on the "so what?" question: What was this all worth to me? Even without prompting from me, their stories moved to this issue, which happened to be my second research question: How do their results affect their everyday lives, health and illness experiences? This third phase is represented in chapter 6. When I began to look at the participants' experience as a whole through the lens of main themes and structures identified, I focused on attempting to classify and order them as they occurred, regardless of which participant's experience the structures belong to. This process is known in interpretive interactionism as the construction phase (Denzin, 1989). In the construction stage of data analysis, I often asked questions such as "What am I seeing here? Is there any order to all of these structures? Is there a beginning, middle, and end? What do these themes mean in participants' lives? And why am I not seeing something I expected to see? When confronted with two opposing views among participants, I asked "How does this occur? What situation brings people to have such an opposing view? How is it that I see such a pattern among these individuals but not among others? With the answers to these questions, I was able to derive how key themes and structures affected each other and show how they cohere to form patterns and processes.37 When I observed 3 7 See appendix 9 for a sample of ongoing data analysis used throughout the study to assist in constructing patterns and processes. 77 that there was absence of some of the key themes and structures in a personal participant's story, I contrasted the participant's family and personal lived experiences with other participants' stories who presented with these key themes and structures, thereby showing how lived experiences can shape and alter patterns, processes, and structures within the aggregate experience of interpreting and making sense of uninformative genetic test results. This method also brought to light what is particular to the lived experience of an individual person (Thorne et al., 1997). This final contextualising stage puts the phenomenon studied back into the social world where it occurred and where the structures identified take full meaning (Denzin, 1989, 2001). Within interpretive description approach, the dialect activity of comparing and contrasting individual cases and patterns with the aggregate experiences of all participants forms the basis of its strength: "a respect for knowledge about aggregates in a manner that does not render the individual case invisible" (Thorne et al., 1997, p. 171). Contrasting stories and discovering how lived experiences shape patterns and processes produces valuable clinical knowledge. Nurses can then see how aggregate knowledge can be applied to individual cases. This is a form of knowledge nurses have long demanded, one that facilitates their everyday professional practice (Thorne et al., 1997). Standards and credibili ty measures As understood by Crotty (1998), establishing the credibility of our research requires that we report our research process. This permits the reader to judge the trustworthiness of our findings and decide what she or he can retain from them. Further, Thorne et al. (2004) state that what is most important in establishing the credibility measures of our findings is not so much presenting "a litany of attributes such as trustworthiness, transferability or making claims about one's integrity (such as reporting an 'audit trail')" (p. 15) but rather presenting our analytic 78 decisions and showing how these are contextualised within the final study. I have described my data analysis procedures above, and I have given examples of it. More examples are presented in chapters 4 to 6. As with other qualitative research, rigour is a critical component of interpretive description, as research designs often evolve with ongoing data collection and analysis. Another component critical to interpretive description is explicit accounting of researcher's biases that may influence our study. Although completely setting aside our biases is naive (Thorne et al., 1997), as we can never remove ourselves completely from the research process, I will share one personal bias that I struggled with during the conduct of this research. One of my biases is a belief that we, as health professionals, often interpret clinical tests and results differently than our patients do. Consequently, although I interpret the genetic test results that participants of this study received as inconclusive (meaning that no conclusion can be drawn from their results), in the interviews I referred to their results just as "your genetic testing results." I tried not to influence their interpretations of the results by not showing approval or disapproval of their interpretations, even if at times the participants' interpretation seemed unrealistic. Validity of findings in interpretive description is defined in their potential to create mental heuristics that confirm hunches of expert clinicians from the field studied where they would see new understandings of their reflective practice observations (Thorne et al., 2004; Thorne et al., 1997). Hence, I sought feedback from expert clinicians within genetics throughout my data analysis in peer-reviewed scientific presentations and nursing rounds at two different universities. The feedback I received encouraged me to refine my data analysis, as some of my interpreted structures were either challenged for their content or questioned for their clarity by audiences at my presentations. 79 Additionally, the dialectic approach in interpretive description builds in a credibility check that is more productive in ensuring validity than taking data excerpts back to study participants. Thorne et al. (1997) recommend taking key statements in the form of interpreted data back to participants where such an approach can produce important insights when individuals reflect on why such key statements do or do not quite represent their view. To give an example within the current study, a few participants had all said something I found rather unusual: "Those who test positive for an inherited mutation are the lucky ones." This contradicted some of the literature, in which receiving a positive test result for a mutation of BRCA1 and/or BRCA2 is perceived as stressful (Frost et al., 2004; Tercyak et al., 2001). However, this statement could also refer to viewing positive testing as eliminating uncertainty about one's genetic status (Lerman et al., 1996). So, I brought this key statement to subsequent interviews for the participants' critical consideration. Some said "Well it never occurred to me to think of them as the lucky ones, but I certainly do understand where this might come from, as ..." and then apply their reflection to their own experience with genetic testing. Field notes and journalling also facilitated the evolution of my reflection during data analysis and led me to question my early assumptions about participants' experiences. In the field notes, I recorded situations, people, and places being observed, as a well as my general demeanour on interview days. I also would write postinterview comments and observations in my journal, which I used to reflect on the research process. The first few journal entries focussed on my comfort with conducting open interviews. I also enjoyed having a place where I could reveal my inner thoughts on the research process. The last journal entries focussed more on struggles with my analytic framework and their resolution. Although I viewed this journal as a private, safe haven as I reflected on my integrity within the research process, I also realised that 80 this journal was another means by which others can judge the credibility of my findings. Hence, I reluctantly share two entries in my journal: one made following my third interview, the second following the completion of all the 36 interviews. Today I interviewed my third participant. We were just the two of us. Again, I was surprisingly amazed at my qualities in conducting open unstructured interviews, as this is how I truly see them. However, as I was congratulating myself, I wondered what my committee would think. Would they think differently if they'd been in the interview with me? Would they think differently if they heard my responses and questions on the tape? This led me to wonder if I should do the interview having in mind my committee. In the first instance, I thought it would be good idea, as I do have to be credible to myself and to them at the same time. So, I should be comfortable with the fact that they would be listening to my every word. Well, that is just the problem. I would become so conscious that I am being evaluated and heard on every word I said that I would lose track of the interview process and probably lose the easy, comfortable feeling I have in interviewing the participants. The participants would probably feel my tension and perhaps start becoming conscious of the tape recorder and the fact that we are not really having a discussion but more, as Mishler (1986) would say, a traditional stimulus-response interview. That is totally what I am trying to avoid. Therefore, I will keep doing my interviews with the mindset that I have been doing them. That is, as if we are having a conversation, with them sharing with me a part of their lives. (Research journal entry, December 18,2000) The focus of the second research journal entry is on the analysis process. I am questioning the orientation analysis of my secondary participants: 81 I have done a fair amount of writing of my findings from key statements, although much more interpretive description needs to be brought out for result chapters. I have prepared a presentation for next Monday for the Sigma Theta Tau conference, and it has helped me to synthesise further my results and see more clearly growing patterns and processes, such as the chronology of their experience and the concept of time. Today I wanted to take a break from the participants' analysis and go to the family members' analysis. Somehow I suddenly felt blocked about how to code and analyse the transcripts. I have read what I wrote previously using the reflection framework I followed with the primary participants, but I am not sure how to proceed with the family members/secondary participants. Their experience seems both to differ from the participants' and not to follow the focus of the study. With the family members and the primary participants I used the same research questions, because I assumed that this experience is lived in relation with others, not in isolation so I wanted to have the family members' perspectives. But, in the end, I realise now that this is not the case and that the aim of this study is not to compare the two populations. Or should I be thinking about that? (Research journal entry, October 31, 2002) Following this entry, I went back to my dissertation committee to seek their advice. I explained that, although the family members' stories added to the second research question (where the focus is on exploring how the participants' experience of genetic testing affected their everyday lives, health and illness experiences), because the family members had not actually lived the experience of genetic testing, it became difficult to integrate experiential knowledge of this experience. My committee and I decided that, from this point on, the family members' accounts would only be added when it contributed to the participants' accounts in answering the 82 second research question. As theorised by Thorne et al. (2004), "If the findings look too similar to the analytic framework with which one entered the study, they may reflect the mind's capacity to 'fit' data rather than to ask good questions and generate useful conceptualizations" (p. 10). This experience made me realize more fully that qualitative research must remain open to change through an iterative, reflective, critical and analytical process (Morse & Field, 1995). Throughout the data collection and analysis processes, I continually examined my preliminary beliefs and assumptions about the studied phenomenon in relation to the interpretations and decisions I made throughout the research process. This ongoing reflection forced me to reflect on my own position in relation to the research (Hall & Stevens, 1991) - an added component towards achieving scientific credibility (Lather, 1991). I have already presented examples of how I modified the analytic process, but now I would like to reflect on my assumptions when I entered the study. Although largely influenced by my reading on genetic testing for breast and ovarian cancer susceptibility, I originally viewed genetic testing as a tool in breast and ovarian cancer screening for individuals identified at high-risk of developing the disease. However, in the course of the study, I came to realise the slippery slope in this assumption: providing individuals with their genetic mutation status, that these individuals were being offered the same treatments as high risk individuals already diagnosed with the disease (d'Agincourt-Canning, 2003). Further, I had not realised that the availability of this testing might also create a new pool of individuals categorised as at risk for cancer or the "at-risk health status "(Bottorff et al., 1999) and bring about the possible psychological harm that may result with these labels. I also assumed that individuals from high-risk cancer families would be automatically interested in genetic testing and that they would see its availability as an opportunity to be 83 screened more precisely on their own cancer risk and that of others in their family. Perhaps my position as a nurse influenced the latter assumption, in that I have relied on clinical tests to assist me at arriving at clinical care decisions, either in relation to my own care, my family's care or, most often, the care I provided to the many patients I nursed over the years. I have come to trust and rely heavily on clinical tests and their findings and just assumed that, if a new test came along that could convey additional information about one's disease risk, that people would be interested in obtaining this test. Although this assumption was rarely challenged in the interviews, what the participants did say was, "How can you offer a test when you can't offer definitive result?" In addition, I assumed that individuals who received a genetic test would automatically want to tell as many family members as possible they were having this test. It became impossible for me to maintain this assumption without sharing it with the participants, in light of what I was hearing from them. The majority had decided not to tell their family they were having the test before obtaining their results because, as they explained, "Why worry others when we don't know the outcome of the results?" When I questioned this view, I came to learn from the participants that genetic testing can play a very small part in their everyday lives compared to more stressful events such as their breast cancer experience. I realised then that I could not discuss how they interpreted and made sense of their results without exploring what it meant to live with a personal and family history of breast and/or ovarian cancer. Thus, the interviews helped me to redefine my initial assumptions and guided me toward new areas of inquiry that I had not previously intended to explore, such as looking at how the individual made sense of their genetic testing results in light of the implications of living with a family history of breast and/or ovarian cancer. The interviews also gave me an appreciation of 84 the complexities of receiving uninformative genetic testing results and the complex meanings generated by living with a family history of cancer, both of which I have not personally experienced. Now, I would like to briefly reflect on the choices I made in the lines of inquiry I pursued with the participants which, in the end, affected the knowledge I produced (Whittaker in d'Agincourt-Canning, 2003). Seeing that there is a paucity of studies on the population studied in this research, I felt that the immediate clinical knowledge need was for understanding how these individuals interpret and make sense of their specific type of genetic test results, as opposed to focusing on the communication of risk. Although I do see the latter as important within the context of genetic testing because of its inherent complexity, I felt that such a topic would 38 require the full focus of a study. Thus, while communicating risk and interpreting risk values are important aspects to study in genetic testing, I do not feel that I can do justice to their inherent complexities even though some issues relating to these two topics are interwoven within the findings of my research. Ethical considerations In the previous section, I discussed the standards and credibility measures applied in this study that were critical to the integrity of the research. In this section, I address responsibilities of the researcher to protect study participants - also known as responsible research conduct (Steneck, 2002). While protection of study participants cannot rest solely on the integrity of the researcher, clearance to conduct this study was obtained by the Behavioural Research Ethics i n Board at the University of British Columbia, and from the BC Cancer Agency. 3 8 For references to studies looking at risk communication in the context of genetic testing for B R C A ! and B R C A 2 mutations, see Bottorff et al., 1998; Bottorff, Ratner, Johnson, & M c C u l l u m , 1999; Croyle & Lerman, 1999; Edwards et al., 2003; Lipkus & Hollands, 1999; Lynch et al., 1997; and Rimer, 1999. 3 9 See appendix 3 for a copy of the letter of support received from the B C Cancer Agency. 85 The potential exists that people's lives may be reshaped as a result of the researcher entering the lives of individuals by asking them to share their stories. Because researchers enter into these relationships with specific aims and goals tat are independent of participant benefit, the primary responsibility of researchers should be avoiding harm to participants (Clandinin & Connelly, 1994). Responsibilities I took to heart were: ensuring informed consent throughout the research process, maintaining privacy (protecting the identities of participants), and protection from harm (physical, emotional, or any other kind) (Fontana & Frey, 1994). Within this study, ways in which I tried to make the participants' experience positive included respecting as much as possible the time limit for the interviews as written on their consent form. As well, I made it a point to only discuss the topics outlined in the cover letter sent with their participation information sheet and consent form.40 Therefore, I did not question them about other areas of their lives unrelated to the focus of the study. Finally, as a token of my appreciation for their time and their participation in the study, at the end of each interview, I offered all participants both a small scented candle and a three-piece Lindor chocolate box. As well, the day following the interview I sent all participants a thank you card. I also sent them all a Christmas card in December 2002 updating them on my progress, since I had told them that I would send them a 10 - page written summary of my findings once I defended the dissertation. I explained in the card that my writing phase took longer than expected because of maternity leave and family obligations, but that I had not forgotten my commitment to them. Subsequently, I received a Christmas card from a secondary participant who wished me well and provided me with an update on her family's newest diagnosis of cancer. See appendix 7 for a copy of the accompanying cover letter. Its content represents a synthesis of the interview schedule I brought to each interview. 86 With the aim of minimising any possible harm of the participants having engaged in my research, before asking a question I constantly asked myself, would this question cause pain to the participant? If so, does their protection from harm preclude asking this question? I always emphasised with participants that their comfort was always more important than my need to obtain information. I wanted them to feel free not to discuss topics that they were not ready to discuss with a stranger. As an example, during one interview, discussion moved toward the diagnosis of a participant's mother with breast cancer. Tears started rolling down her face. I immediately stopped the tape and told her that if it was too uncomfortable a topic that I did not need to hear about this experience to understand how she made sense of her results. The participant agreed to change the topic and shared how she had not yet discussed with anyone the death of her mother following her cancer diagnosis. Although I remember thinking at the time that hearing about this experience would have benefited my overall view of the emerging process within my phenomenon studied, I did not want to put the dissertation needs ahead of the participant's comfort. Maintaining confidentiality was another important ethical concern I tried to maintain throughout the study. Because of the nature of the research framework, whereby I interviewed different people within the same family, I found myself at times being questioned about what some family members had said, knowing that the other person's response was not what the interviewee expected. For instance, one participant said to me, "I am sure my sister must have told you about our mom?", yet the sister's view of their mother's experience with breast cancer was not the same. This event propelled me to be even more cautious about keeping all participants' interview data anonymous. Some strategies employed were fictionalising identifiable familial characteristics (Fontana & Frey, 1994) such as asking participants to choose 87 a code name that only they could recognise, hence some of the code names are unusual. Throughout the entire project, I remained highly aware of the ethical issues that may arise as a result of having entered into a researcher-participant relationship. Consent for me to contact potential participants was given during a telephone call from either a genetic counsellor or the nurse at the Hereditary Cancer Program of the BC Cancer Agency. For those who expressed an interest in participating in the study, a letter of information and a consent form were sent prior to the interviews but following our first initial telephone contact.41 The information and consent form specified that the student researcher would be contacting them to find out if they were still interested in participating; it specified the estimated length of the interview, and asked that that the interview take place in the participant's home or in an alternate location if desired. The letter also asked that the participant pass a copy of the letter of information to other family members who might be interested in participating in the study. The consent form included a statement regarding risks such as becoming upset as a result of having discussed some topics. Participants were told that they were always free to contact the genetic counsellors and the genetic nurse at the Cancer Agency to discuss ongoing issues in relation to their experience with genetic testing. Participants were also told at the beginning of the interview that they could refuse to answer any question they judged to be too upsetting and that they could ask me to erase any part of the tape recording if they felt that disclosing any information made them uncomfortable. None of the participants requested this. Participants were also told that their participation was voluntary, that they were free to stop the interview at any time, and that they could choose to withdraw from the study without jeopardising their health care or access to genetic services. 4 1 See appendix 4 for a copy of the participant's information and consent form. While I obtained written consent at the beginning of the interviews, I also recognised that consent is ongoing throughout the interview as well as throughout the research process. The right of participants to privacy was further protected by presenting only selected data to my dissertation committee members that would not compromise participants' confidentiality. Interview transcripts were kept in secure computer files protected by a password and all identifying information was stored separately from the transcripts. Paper files of participants were stored without using their names and instead used numbers such as primary participant 1 (PP1) and secondary participant 3 (SP3) of primary participant 1 (PP1) that became SP3/PP1. Biographical information was modified as necessary to conceal possible identifying characteristics of the participants and wi l l continued to be so in future published and unpublished work. 4 2 The data were not destroyed once the transcripts were completed as I obtained consent from the participants to retain the data to permit future secondary analysis. Summary In this chapter, I presented the rationale for use of interpretive description by nursing researchers studying health and illness experiences. I also described some of the reconceptualisation that occurred in the research process as a result of its interactive nature. I outlined the research methodology and the methods employed to gather and analyse data. I addressed standards and measures used to establish credibility of research findings and described ways in which I attended to ethics. I now turn to chapters 4, 5 and 6, in which the findings of the research are presented. Hence my use of breast and/or ovarian cancer when speaking of all participants' history with cancer as opposed to saying that a particular individual only had breast cancer or ovarian cancer in their family history. Without this condition, I believe that maintaining anonymity becomes perilous. 89 CHAPTER FOUR: The Buildup to Genetic Testing When I began my study seeking how individuals and family members interpret and make sense of uninformative genetic test results, I asked individuals to describe their personal experiences with breast and/or ovarian cancer and the experience of others in their families. Not until I did a few interviews did I take full notice of the implications of such a starting point in understanding what would lead individuals to be receptive to genetic testing. Just as Hallowell et al. (2004) found when they looked at the responses to uninformative genetic test results for BRCA1 and BRCA2 mutations among affected individuals with a family history to hereditary breast and/or ovarian cancer (HBOC), it is impossible to discuss individuals' interpretations of such genetic test results without first providing the context of these responses. Thus, I begin this chapter by describing how having a family history of risk for breast and/or ovarian cancer brings about certain discourses that you would not normally see in other families. In the first section, I present implications of these found in the participants' accounts that either facilitated or brought contention to their interpretation and the process of making sense of their uninformative genetic test results. In the second section, I present how genetic testing entered participants' lives and the reasons that led them to undergo genetic testing. In the final section, I build a foundation for exploring how participants interpreted and made sense of their genetic test results, the subject of chapter 5. Implications of Living with a Family History of Breast and/or Ovarian Cancer Many of the participants explained how they looked to genetic testing to explain their breast cancer diagnosis and their past family history of the disease. Before they went on to explain how they made sense of their genetic test results, all participants except one started by 90 sharing with me what it had been like for them to live with a diagnosis of breast and/or ovarian cancer and, more specifically, what it had been like to live with a family history of cancer. Their journey of learning to live with the disease led many to seek genetic testing as a way of finally answering why breast and/or ovarian cancer was so present in their family. Living with a family history of a disease deepened their struggles to make sense of uninformative43 genetic test results. Before presenting the implications of living with a family history of breast and/or ovarian cancer as described by the participants of this study, I first present a pedigree of what constitutes having a family history considered at risk for HBOC and look at how one study participant met eligibility criteria for genetic testing. Although eligibility criteria vary among cancer programs, hereditary breast cancer susceptibility is usually characterized by early onset, cancer occurring in both breasts, multiple primary cancers; and in conjunction with other cancers (Ormiston, 1996). First, I review the eligibility criteria for testing presented in Table 3 that study participants met, and then follow with an example of a pedigree, built from a participant's account of her family history. " Although very few participants described their results as uninformative or inconclusive, mostly all participants except four viewed their test results as uninformative or inconclusive as defined in chapter 1. That is, because of their personal and family history with cancer, that it was impossible to confirm that they do not carry an inherited mutation or to conclude that their cancer may have occurred by chance. 91 T a b l e 3: E l i g i b i l i t y C r i t e r i a f o r H e r e d i t a r y B r e a s t a n d O v a r i a n C a n c e r R i s k A s s e s s m e n t G e n e t i c r i s k is assessed a p p r o p r i a t e i f : A f a m i l y h i s t o r y meets one o f the f o l l o w i n j c r i t e r i a : • A woman with breast cancer diagnosed at age 35 or younger OR • A woman with ovarian cancer diagnosed at age 50 or younger OR • An Ashkenazi Jewish woman with breast or ovarian cancer OR • Have a blood relative with a confirmed BRCA1 or BRCA2 mutations A f a m i l y h i s t o r y that i n c l u d e s at least t w o o f the f o l l o w i n g on the same side of the family: • Cancer in 2 or more closely related family members (i.e., parents, siblings, children, grandparents, aunts, uncles) *> Cancer diagnoses at younger ages than expected in the general population (i.e. premenopausal breast cancer) OR •S» A woman with ovarian cancer diagnosed at age 50 or younger OR • Multiple primary cancers in one individual OR • A male with a breast cancer diagnosis O R Note. The eligibility criteria presented here are from the Hereditary Cancer Program, B C Cancer Agency (BCCA) and represents those used within the year the study was conducted. Reprinted with permission from the B C C A . 92 PP2I: Sherry SP1 : Susan - Sister I N D K X : P P and proband P R O B A N D : Niece Female ^Dx 5J lAgc inf |Wa Dc early after >5i Dx 4n Dx .14 rO 1 1" 50; 2 n d 63 Male Deceased Do not know what type ol" cancer Hreast cancer Ovarian cancer 0 0 OQ O I-ympho Brain cancer Attended genetic counselling session Dx |#| = Diagnosed at age # 13c 1#J = Deceased at age # 17 16 Note: INDEX: Had genetic testing. PROBAND: Initiated genetic testing proce Figure 1: Pedigree Representing Sherry's (PP21) Family History of Breast and Ovarian Cancer Starting from Table 3 and looking at Sherry's44 family pedigree, Sherry had to meet at least one criterion from the left column or at least two criteria from the right column. Sherry met the first criterion: in the left column. Sherry was diagnosed with breast cancer at the age of 35. Her family history also met other criteria from the left column. Her niece was diagnosed with ovarian cancer at the age of 34. As for the last two criteria, Sherry was not an Ashkenazi Jew nor does she have in her family an individual with a confirmed BRCA1 or BRCA2 gene mutations. Nonetheless, she meets the criteria for testing under this category by having received a breast cancer diagnosis at the age of 35. Looking at the right column in Table 3, Sherry's family history met three of the four criteria - some of them more than once - while only two are necessary for eligibility. For the criterion breast and/or ovarian cancer in at least two closely related women on one side of the family, Sherry's family history meets this criterion in many ways. That is, both All first names presented in the dissertation represent code name chosen by participants. 93 she and her sister had a diagnosis of breast cancer, as did her mother. Sherry's two nieces who are siblings were each diagnosed with cancer, one with breast and the other with ovarian. As for the second family history criterion, cancer diagnoses at younger ages than expected in the general population (i.e. premenopausal breast cancer), Sherry's family history again meets this criterion more than once. Sherry was diagnosed with breast cancer when she was premenopausal, as were her two nieces. While Sherry's sister was diagnosed at the age of 50, Sherry said during the interview that her sister was premenopausal at the time she was diagnosed. The third criterion is having at least one individual with multiple primary cancers. Sherry's sister received two primary breast cancers, one cancer occurring at the age of 50 and the other at the age of 63. As well, Sherry's other sister, a family member interviewed in this study identified as SP on the pedigree, sent me a letter in January 2003 saying that Sherry had been diagnosed with lung cancer, believed to be another primary cancer. I remembered Sherry saying in her interview that she had always been a healthy person, jogged regularly and had never smoked a cigarette in her life. The fourth criterion is having a case of male breast cancer in the family. Sherry's family has none. Sherry was told that no mutation had been found in the two breast and ovarian cancer genes tested for but because of her personal and strong family history alone, the clinic could not conclude with certainty that there was no inherited mutation and that the family should continue screening as previously recommended. These conflicting results have been a challenge to make sense of for Sherry, in light of what she has come to live with in her everyday life. In the next section, having illustrated what a family history considered at risk for HBOC can look like, I now discuss the themes derived from participants' interviews about the implications of living with a family history of breast and/or ovarian cancer. 94 Expecting breast cancer to appear at some point. Individuals with a family history of breast and/or ovarian cancer and who have been in close contact with family members who have developed the disease often speak of constant thoughts that the disease must be inherited. A study by Iglehart et al. (1998) showed that affected women with a family history tend to overestimate their risk of carrying hereditary factors for breast and/or ovarian cancer. However, the authors explain that such feelings may come from prolonged care for someone significant who had breast cancer, such as a mother. As a result, individuals tend to internalise their risk to cancer from embodied experiences of others to cancer. For many of the participants, this has been their reality. One participant, whose mother developed breast cancer while still a preteen, described the experience as follows: "I basically prepared myself to have cancer all those years. It was like I knew that probably at some point I would have cancer." As described by Chalmers, Thomson and Degner (1996), perception of risk for cancer can be intensified by the breast cancer experience of others within the family. This experience has been termed empathetic (dAgincourt-Canning, 2001; 2003), referring to the subjective knowledge derived from personal, close experience with others in the family who experience cancer. Increased awareness of risk for breast and/or ovarian cancer in self and others. This increased awareness of risk to breast cancer in self and others brings about different ways of being. For example, one participant spoke of her need to always be on the "lookout" for breast cancer, while another participant spoke of being "attuned" to verbal and written information about breast cancer. She explained that she did this for her two sisters and herself. As a result of her breast cancer diagnosis, she found fear of a second primary cancer or a recurrence difficult to let go of. 95 For many of the participants with multiple cases of breast and/or ovarian cancer in their family, dealing with some family members being diagnosed with more than one cancer, or a recurrence, was not unusual. This brought about additional concerns: Perhaps they as well would have to live through a recurrence or another cancer. Just as Vickberg (2003) concluded, it was the younger participants in this study who often feared a recurrence. Many of the participants looked to their family history to assess their possible risk for cancer and that of others in the family. This has been suggested by Richards (1996) as a way lay people explain inheritance. Other lay theories include physical characteristics and personality traits. In other words, if you resemble a family member with cancer, you may inherit the disease. I found in this study that participants not only use notions of physical characteristics and personality traits to establish their risk for another cancer or a recurrence but also use logical concepts of pathophysiology to explain their perceived risk for cancer. Participants whose mothers' breast cancer recurred definitely thought that "perhaps cancer was still lurking inside of them."45 However, according to the Jobsen, Meerwaldt and van der Palen (2000) prospective cohort study, a family history of cancer is not predictive in cancer recurrence among women older than 40. Nonetheless, participants have learned to live with a perceived risk of recurrence. With the disease being part of their family history, participants with children (especially daughters) commented that they also had to live with the thought that perhaps one day their daughters too would be diagnosed with breast cancer, just like them. One participant, Donna, even said that her daughter does not fear cancer because she has come to accept that it will happen to her. Here is Donna sharing a conversation she had with her daughter about the daughter's view of her risk of developing breast cancer: 4 5 Recall that all 21 participants have been affected by breast or ovarian cancer. 96 My daughter said, T know what happened to my grandmother. I know what happened to you. I expect it to happen to me but I don't really want to think about that now.' My daughter has kind of accepted that this is something that's going to happen. She's just waiting for the other shoe to drop. She's a very strong young woman, but she's also a very pragmatic young woman. And so I think, for her, and I think I can speak for her cousins too, that...I mean they've seen this in three very close family members and they just expect it. It's going to happen. (Donna, PPM, field conversation, September 7, 2001 speaking of her 20 year old daughter) A second participant also contemplated what it must be like for her daughter to live with the notion that her mother, her grandmother and her great grandmother were all diagnosed with breast cancer: My daughter is 20 years old now. I just can't imagine what it must be like for her now to be fourth generation, and knowing that all women in the past three generation have had it. I grew up with that so I can be kind of cavalier about it, of course I am going get cancer, right. But in my heart I could hope that I wasn't. But by now not having found the mutation, I hope that the cycle will stop with me. (Stephanie, PP12, field conversation, July 18, 2001) Both of the above participants held fatalistic views that, because their mothers had breast cancer and they had breast cancer, it is almost inevitable that their daughters too will one day be diagnosed with it. With many of the individuals within the study there exists a burden of anticipation that seems internalised by the family's history of cancer, which then influences the perceived risk of cancer for themselves and others. Burden of anticipation based on the family legacy of breast cancer was also documented in d'Agincourt-Canning's study (2003), where she also found an anticipation of cancer among individuals at high-risk for breast and/or ovarian cancer who underwent genetic testing. However, fear of recurrence has also been documented among individuals with or without a strong family history of breast and/or ovarian cancer as a result of a cancer diagnosis (Gil et al., 2004; LaTour, 2002; Vickberg, 2003). These studies found that illness uncertainty continued long after cancer diagnosis and treatment. The most frequent trigger of this fear was hearing about someone's else cancer (Gil et al., 2004). As explained by some participants in this study, "Hearing of someone else's recent cancer diagnosis 97 within the family is not an unusual event" and therefore, participants have come to internalise that no one is immune to cancer, even i f they have already been diagnosed with one. Whether or not cancer appearing in the participants' families is a true reflection of them having an inherited mutation or the result of sporadic cancers, this is the reality that participants are l iving with on a daily basis. In his book on illness narratives, Frank (1995) named the "chaos narrative" to reflect this continued perception of vulnerability that some individuals experienced following an illness or life threatening health condition. Becoming comfortable with the word cancer. Some participants explained that, as a result of having come to terms with their increased risk for breast and/or ovarian cancer because of their family history, they have also learned to be comfortable with the word "cancer," as it became an expected part of their future. A s Stephanie said, "What else was I to assume?" Before she finally received her breast cancer diagnosis at the age of 50, she had already gone through two previous cancer diagnoses, the first at 20 years old when she was diagnosed with precancerous cells in her cervix and the second at 35 when she was diagnosed with uterine cancer. Just like her mother, who had received two cancer diagnoses, one at the age of 32 of breast cancer and the other at 66 of lung cancer, Stephanie was not surprised to have more than one cancer. D'Agincourt-Canning (2003) documented similar findings in her study. She found that unaffected individuals drew on other family members' cancer experiences to construct how cancer could eventually happen to them. Family patterns of breast cancer have certainly been used to inform the participants' perception of their risk of future cancers. Within the literature, becoming comfortable with the word cancer has also been seen as a coping strategy among individuals from high-risk families. Appleton, Fry, Rees, Rush and Cul l 98 (2001), who studied the psychosocial effects of living with an increased risk of breast cancer among 25 unaffected women, reported that women use a variety of cognitive strategies to cope with this increased risk. They focus on the present, avoid potentially worrying breast cancer cues, and think positively. Thus, they adopt an optimistic attitude about the future. The strategies I found are similar to those within Appleton et al.'s study (2001); however, they also found that the participants in their study had to learn to live with this threat, whereas the participants in my study said they had become comfortable with the word cancer as a result of seeing breast cancer as part of their families - as a family artifact. Breast cancer was omnipresent within their lives, they explained. Within their homes were pictures that remind themselves and others of those who had received a breast and/or ovarian cancer diagnosis. The scars were not only psychological but physical: the pictures showed that some of the women in the family are breastless. Although all participants experienced stress upon their cancer diagnosis, they also celebrated their cancers' departure. One participant described how, when it came time for her to have her breast removed as part of her breast cancer treatment, she and her sister had a huge celebration the night before. They took pictures of her naked breasts and ones with lingerie to celebrate the departure of one breast, scheduled for a mastectomy the next day. Becoming comfortable with the word cancer also seemed to come as a result of witnessing a close family member's experience with cancer. The idea of having cancer was not foreign, even before they receive their own diagnosis. One participant, Juniper, explained how she felt she had already gone through the cancer experience even before receiving her own cancer diagnosis because of the way her mother involved Juniper in her emotional and physical experience with cancer. Here is my reinterpretation of Juniper's story of what it was like for her to grow up with a family history of breast cancer: When Juniper's mother was diagnosed with breast cancer at the age of 45, she pulled Juniper and her sister, both still in their preteens, out of school so that they could look after their younger brother. The mother felt she was unable to do so anymore as a result of her breast cancer. Juniper said that her mother went into a depression and showed the many facets of her cancer experience to her children. The family's support system was limited to the immediate members because they had migrated from another country and had no relatives in Canada. With the diagnosis, her mother, having no outside support system, further isolated herself. Her young children became her only support. Her mother showed her two daughters all her scars - physical ones and the mental ones. Juniper mentioned how her mother's mental pain was more difficult to deal with than her physical pain, even when her mother showed them the gruesome scars from her radical mastectomy. And they did look gruesome back then, Juniper added. There began Juniper's story of feeling as if she had already been through the experience of breast cancer, even before receiving her own diagnosis. Juniper also shared how her mother later had her other breast removed out of fear. However, 5 years after her mother's breast cancer diagnosis, Juniper explained that her mother viewed her experience with cancer as an epiphany. Her mother realised she wanted to live life to the fullest and thereafter embarked on many projects that were meaningful to her and, at times, to her family as well. Her mother's epiphany was a learning experience for Juniper. She knew she could now live through cancer, as her mother did. However, she promised herself that, contrary to her mother, she would not fear the cancer experience and that she would not share all of her intimate emotional experiences with her children. Like her mother, Juniper was diagnosed with breast cancer at the age of 45. When Juniper made the decision to have genetic testing, she did not share this news with her mother but did tell her children that no mutations had been found. 100 Becoming comfortable with the word cancer applies mostly to participants in whose families cancer occurred in the previous generation. Those whose families' cancer experiences appeared only within their generation did not talk about becoming comfortable with talking about cancer - illustrating an evolving level of comfort with the word cancer. While some participants became comfortable with the word cancer and comfortable with their increase risk for cancer, a few were not comfortable at the thought of losing their breast to cancer. Although images of breastless women were familiar to some of the participants, this did not mean that they were at ease with the notion. The following quote speaks to this issue: Having most of my adult life seen my mother with no breasts, I knew what it was going to look like. It wasn't as if I didn't know what it was going to look like. You get used to the word cancer and are almost comfortable with it even when one gets a diagnosis. But what has been harder is losing the breast. It's like, there's two different issues happening. I had a harder time with losing my breasts as opposed to having cancer. (Juniper, PP2, field conversation, December 13, 2000) In light of what is known about individuals living with a high-risk of disease, we can hypothesize that time allows acceptance of one's risk of cancer, while having to accept a mastectomy as a treatment for breast cancer is an immediate decision that does not allow much time for reflection. However, according to two studies that explored how well women accepted loss of a breast following a mastectomy, only 7% in Kiernan et al. (1981) study reported negative problems resulting from their surgery - work, social or sexual problems - whereas among this 7% (20 women), sexual problems were cited most often. From their study, Metcalfe et al. (2004) also concluded that 97% of the women surveyed were satisfied with their decision but that younger women below the age of 50 were less likely to report satisfaction than older women. The women in the Kiernan et al. study who expressed greater concern with losing their breast were also younger than most of the other women. Just as in the Metcalfe et al. study, for women in my study under the age of 50 who underwent a mastectomy to treat their breast cancer or 101 prophylactic mastectomy, having reconstructive surgery immediately after facilitated their acceptance of losing their breasts. A study by Maguire (1989) demonstrated that psychiatric morbidity was lower among women who underwent reconstruction immediately following mastectomy while Rowland et al. (2000) study shows that women from the lumpectomy group (n=l 119) compared to women from the mastectomy with reconstruction (n=327) or the mastectomy-alone group (n=511) reported fewer problems with their body image and feelings of sexual attractiveness. Cancer in our family is part of our upbringing - it is the norm. Individuals with a strong family history of breast cancer in my study explained how the topic of breast cancer came to be a common subject of conversation within the family and at family gatherings. Reminders of breast cancer around the house became keepsakes of the family's history. Although the following quote is from only one participant, many others expressed in similar terms this view that cancer was part of their upbringing: My children are aware of everybody's cancer in the family and have always been aware that I have had breast cancer. They see the old pictures of me and they are like reminders of my past history with breast cancer. I had a mastectomy and although I am wearing a prosthetic bra in these pictures, they all feel they could tell the difference. (Sherry, PP21, field conversation, September 18, 2001) Sherry later described how she came to view breast and/or ovarian cancer as an expected event by making a parallel with how society has come to view car accidents as an expected part of having so many cars on the road. She has come to view cancer as the norm in her family. Before her own breast cancer diagnosis, there were three confirmed breast cancer diagnoses in Sherry's family. After Sherry's breast cancer diagnosis, there were two more relatives who received cancer diagnoses — her two nieces who are siblings. One developed breast cancer while the other developed ovarian cancer. 102 It used to be that it became common to hear about car crashes and people being killed. There just seemed to be so many people, it just became normal that some individual would be killed in a car crash. Or to hear every couple of months of somebody you knew being dead. That was sort of the norm. And so this is what it is like for me with breast cancer. Breast cancer in my family is the norm. (Sherry, PP21, field conversation, September 18, 2001) As a result of their perceived increased risk for cancer from seeing so many cancers in their past and present family, some participants shared that "There is always the thought that, someone else in the family will be diagnosed." The participants of my study have all had breast cancer and have a strong family history of the disease. By virtue of their family's history, they feel at risk to develop another cancer. For those who share these latter thoughts, the idea that their family members are likely to develop cancer is often omnipresent. Hearing of another breast cancer within the family is an almost inescapable reality that they learned to live with. Role of family guardian. Many study participants adopted the role of family guardian. They make it their responsibility to look out for their children, siblings and sometimes extended family members. As family guardian, they tried to reduce the family's risk of cancer, and to provide them with information to increase their screening behaviour and to make them more aware of their increased risk for cancer. The concept of family guardian as related to the ill relative is not new within the literature. Past studies have described how the il l relative within high-risk families becomes the manager of cancer-related information for others in the family unaffected by breast cancer (Chalmers et al., 1996). This guardian position is also seen as providing a role model that one can survive their cancer and for the need to be proactive in decreasing their risk of cancer. Seeing herself somewhat as the family guardian - being the eldest sibling, Ginger explained that she felt she needed to gather as much information as possible about her disease on 103 behalf of her siblings. She felt that, the more information one has, the more power one has over the disease. She did, however, specify that, although she was going through genetic testing as a means of obtaining more information about her cancer risk, she had not decided what she would do with the information. While the participants may have enacted the family guardian role unconsciously, some of the family members interviewed also spoke of viewing their family member who underwent genetic testing as role model — a role model of living through a cancer experience. Like I mean, the whole idea of saying you know, the survivor of breast cancer, I mean it seems like a term that's being used so much and it really kind of bugs me but I mean, you have somebody who's lived through it. And so I have sort of a living role model and the fact that if you developed breast cancer, they are means of ways of getting over it. Not to say that it is not a killer, I mean it is. People do die. It's just depends on how it gets into the system right. (Jill, SP2/PP3, Joyce's daughter, field conversation, December 18, 2000) Although Jill has not lived through a cancer diagnosis, she described her experience of living with a family history of cancer using terms similar to those used by affected individuals. Having become aware of her increased risk for cancer because of her family history with the disease, Jill has learned to live with this risk. She does not think that one survives breast cancer but that one can learn to live with having a diagnosis. Other individuals in her family who have lived through their breast cancer diagnosis are role models for her. They show that a cancer diagnosis is not a death sentence. Need to be proactive. Because of the cancer risk in their family, the participants spoke of their need to proactively prevent another cancer in themselves and to prevent cancer in their family. As said previously, seeing oneself and others in the family at higher risk of developing cancer when one has experienced the disease is not unique to high-risk families, but is often seen among cancer 104 survivors (Macleod, 2000; Mast, 1998; Mishel, 1988, 1990). Being proactive may also be an adaptive mechanism to help them deal with concerns they frequently mentioned such as, "Why did I get breast cancer?" and "Was it something I did?" According to Macleod's (2000) study, illness uncertainty is positively linked to fear of recurrence and symptom distress. Hence, although unaffected individuals such as the family members interviewed who do not live with cancer and are therefore most likely not experiencing the same probability of illness uncertainty,46 they could nonetheless be experiencing what is known as living with chronic risk (Kenen, Ardern-Jones, & Eeles, 2003b). Just as most of the participants viewed their genetic test results as uncertain, they also viewed the origin of their cancer as uncertain. In attempting to understand why they got breast and/or ovarian cancer, they compared themselves with others in the family, in terms of who they believed was more at risk of developing the disease. Even participants who strongly believed they had an inherited mutation and that this mutation was the cause of their cancer still struggled with the idea that current genetic testing technology could not confirm their genetic mutation status. Hence, the participants also learned to live with uncertainties about the origin of their cancer and with uncertainties about their genetic mutation status. This double uncertainty lends weight to the need to explore further the Kenen et al.'s (2003b) concept of living with chronic risk. A couple of participants explained that if they could not figure out why they developed breast cancer in the first place, how could they know they are being diligent enough now to prevent another cancer? Not having an answer as to why they developed breast and/or ovarian cancer left them constantly in search of, "What did I do wrong?" Bottorff, Ratner et al. (1998) explain that uncertainty exists when the probability of outcome is unknown, while risk is associated with the probability that each outcome is known. Yet , the authors acknowledge that even known probabilities do carry some form of uncertainty. 105 When I asked participants how they felt about individuals who were told that a mutation had been discovered in their breast and ovarian cancer genes, three participants answered in the following ways: Barney: They are the lucky ones, for at least now they know why they have developed breast cancer. (PP7, field conversation, December 4, 2001) Stephanie: I would not say they are lucky but I would say that I would envy them. At least they have something to hang their hats on. (PP12, field conversation, June 27, 2001) Erika: I was looking for an explanation for my breast cancer, but I would not have liked to be told that I had the gene. (PP10, field conversation, September 7, 2001) The above examples expose the complexity of the participants' capacity to tolerate uncertainty. The participants were forced to struggle with the uncertainty of the etiology of their cancers and the uncertain outcome of their genetic test results without finding relief. Messecar and Kendall (1998), who studied the meanings of uncertainty and the adjustment of wives whose husbands were called to active duty during the Persian Gulf War, formulated the concept of struggling without relief as one of the three main ways wives came to terms with the separation from their husbands when their return was uncertain. Continuous uncertainty about the origin of the participants' cancers may explain their need to seek ways to be more diligent. One participant, Erika, reported becoming more aware of everything that could be considered carcinogenic. She mentioned how she had, at one point, almost stopped eating for fear that all food contained carcinogenic agents. Another participant, 106 Victoria, decided to grow an organic garden in the summer and, the rest of the year, refused to eat any vegetable or fruit that was not organic. Participants explained that, when you have been diagnosed with breast cancer and have had many experiences with breast cancer among your family and friends, these proximity experiences41 remind you of your risk of developing another cancer and the need to be constantly vigilant. As Erika said, "Yours ears just perk up as soon as you hear the words breast cancer, and respite is very seldom." However, Erika explained that this overabundant information caused her to "switch off at one point," because she was unable to process the information anymore. Her metaphor of switching off parallels turning a light on and off. A study by Appleton et al.(2001) found similar results; some women's heightened sensitivity to breast cancer cues prompted them either to increase their vigilance or consciously avoid the cues. The findings of my study suggest that the role of family guardian and being a living role model to others in the family can be overwhelming. The responsibility of keeping abreast of all "anti-cancer behaviours," as a participant described it, may fulfill their need to stay diligent. This responsibility, however, should perhaps not come without support. That is, Stephanie mentioned how it is "just too difficult to leave it up to us to find our cancers and to help others around us." Participants felt they needed specialised support to enact their roles effectively. Hence, Stephanie requested on going assessment by a high-risk breast cancer clinic because she believes such a clinic has the latest evidence-based knowledge of breast and/or ovarian cancer. Three participants held the opposite view. They viewed the suggestion from their high-risk clinic that they could now be monitored by their general practitioner as "being released" from the high-cancer-risk population. They perceived their cancer risk to be lowered. However, 4 7 Proximity experiences of family's and friends' breast cancer have also been defined as embodied experience - as opposed to bodied experiences. I find the distinction interesting, as both terms convey anxiety-provoking feelings. For a fuller description of the meanings attached to embodied experiences, see d'Agincourt-Canning (2003). most participants did not share this view. As some said, "A genetic test, no matter what the results are, does not erase our family history of the disease." Here is Stephanie expressing her need to be screened in a high-risk clinic. I must add that Stephanie is convinced that cancer runs in her family with a "100% certainty" and, therefore, firmly believes that she carries an inherited mutation: "You can't find this disease by yourself, 'breast or ovarian cancer' and so that is why I want to stay in this specialised clinic." (Stephanie, PP12, field conversation, June 18, 2001) Reassurance from being assessed in a high-risk breast cancer clinic has been documented in another study in which the psychosocial effects of living with an increased risk of breast cancer were examined (Appleton et al., 2001). The authors described how many women not only felt privileged to be assessed in specialised clinics but also felt confident about both the expertise of the staff and the equipment. Increased exposure to breast and ovarian screening and testing. It can only be speculated that, perhaps as a result of their family history with breast and/or ovarian cancer, the participants have gone through much more than the usual amount of screening and testing. Other studies indicate that women with a family history of breast cancer used screening procedures such as mammography more often than those without such a family history (Costanza, Stoddard, Gaw, & Zapka, 1992; McCaul, Branstetter, Schroeder, & Glasgow, 1996). Participants in the Mahon and Casperson (1995) study commented that they participated in more screening activities as a result of the reminders sent by their cancer screening centres. Costanza et al. (1992) specify that it is only when women are aware that having a family history of cancer is a risk factor for breast cancer that they increase their use of mammography. However, other studies focusing on high-risk women have documented that rates of mammography do not differ between high-risk women and women of average risk (Andersen et 108 al., 2003; Drossaert, Boer, & Seydel, 1996). Constanza et al. (1992) comment that, based on their study findings, further education is needed among family physicians to increase mammography utilisation among high-risk women who do not know that having a family history of cancer increases your risk of breast and/or ovarian cancer. The increased use of follow-up screening within my study could reflect participants already being clients of a cancer agency, identifying themselves as high-risk, and being identified as such by their oncologists. Some participants remarked that they understood that having a family history of breast cancer increased their risk for the disease. For example, participants were very aware that, when they found a lump, it needed to be taken seriously. As Juniper explained to me during her interview, "I understand that my lumpy breasts are not the same as your lumpy breasts" -knowing that I do not have a family history of breast cancer. Participants commented on how, for example, a suspicious lumpy breast or a fibrous cyst was investigated more thoroughly. Hence, before some of the participants received their first diagnosis of breast cancer, they had already experienced many screening and testing procedures for cancer. The following account looks at one participant's experience. By the time I was barely over 30 years old, I had already experienced my first mammogram. By the time I was 40,1 had about five or six done. And by the time I was 41,1 had one breast aspirated from a fibrous cyst. Yes, I had been thinking about cancer for a long time. Then, you also have those worrisome mammograms that tend to lead to other mammograms soon after. (Juniper, PP2, field conversation, December 13, 2000) A new breast cancer diagnosis in the family is often a pivotal moment reminding others that perhaps they should have follow-up screening and mammography done, regardless of whether or not they are due for their regular checkup. This again increases their exposure to screening procedures. One participant commented on how, when she was found to have breast cancer, her three sisters had their mammograms redone. As a result of this retesting, two sisters 109 were told their mammograms were normal while one was found to have a suspicious lump, which led to more tests revealing a cancerous tumour. The increase in screening and testing procedures results not only from participants' requests following a family member's recent breast and/or ovarian cancer diagnosis, but also from recommendations by their doctors. Stephanie describes how her doctor recommended that she consider prophylactic surgery in light of her family history of breast cancer: When the Cancer Agency found out I still had my ovaries, they panicked and said you shouldn't have those because of your breast cancer history and your family history with the disease...He said, 'You still got your ovaries?' ands I said yes. And he said, 'You have breast cancer?' I said yes. He said, 'There is a huge connection with family history. Whether or not it's a gene it doesn't matter, families that have a history with breast cancer often have ovarian cancer.' So he said, 'Well it's got to come out. (Stephanie, PP12, field conversation, June 27, 2001) When family physicians are aware of their patients' strong family history of breast and ovarian cancer, these individuals typically receive more intense screening. And, as described in past studies, physician recommendation for screening is a strong predictor of acceptance and adherence to regular screening (Lerman, Rimer, Trock, Balshem, & Engstrom, 1990; Stefanek & Wilcox, 1991). Having to go through more than their usual share of screening and testing is not without consequences. Not all participants found a sense of security in being screened so intensely. Gilligan, echoing others, expressed how she came to dread her yearly mammograms. Gilligan was diagnosed with breast cancer at the age of 26 and had bilateral prophylactic mastectomy in her 40s: Prior to my decision to have prophylactic surgery, going through my yearly mammograms was just a big fear. Usually what would happen is that I would go in for my mammogram, and it would come back questionable. So I would have to go in for a second mammogram and again they were having questions. And so you know I'm in there for a couple of hours, which feels to me like a very long time. I then would have to go through many visits, repeated mammograms, and then to ultrasound - to finally be 110 told that the mass turned out to be just a cyst. Just going through that every year, I was beside myself and so it confirmed in my mind that a mastectomy was the way to go. There was a huge fear there that you keep under control a lot of the time but yeah, at one point, you have enough. (Gilligan, PP16, field conversation, August 14, 2001) "Having enough" meant to Gilligan that she did not want to live with this constant fear that perhaps one day she might be found to have another breast cancer, or a recurrence like her mother. Following this response, Gilligan described how she felt a huge sense of relief from her decision to have a bilateral mastectomy followed by reconstructive surgery. She expounded on how she had come to view her breasts as a major source of stress and anxiety and, by having had her breasts removed, she felt she had removed a major risk in her personal life and was now more ready to take on new risks in other parts of her life, such as in her professional life. Her husband, who was interviewed as part of this study, spoke of the impact of her surgery on their lives. Here are the field notes taken from his interview: His wife's surgery was lived as a major turning point in his life. He describe that he felt his wife became more relax and hence, he became relaxed. He also emphasised many times how this change had loosened the tension in their relationship and had made it so much better because she no longer feared for cancer. (Pierre, SP1/PP16, Gilligan's husband, field notes, August 16, 2001) Such an experience from affected women with a family history of risk for hereditary breast and ovarian cancer (HBOC) who received uninformative results to their BRCA1 and BRCA2 genetic testing has also been documented in Hallowell et al.'s (2004) study, in which a participant described that having her breasts removed followed by breast reconstruction was like being born again, in the sense that she could now move on with her life. Within participants' lives, there was a sense that having so many screening and testing procedures was the norm. Just as some participants came to view breast and/or ovarian cancer as the norm in their family, so were multiple screening and testing procedures. Gladys, along with I l l two sisters, developed breast cancer, spoke casually about the number of mammograms one of her sisters had: My sister chose not to have a double mastectomy. She chose to have a mastectomy - only on one side. And she has mammogram four times a year, I think. So they watch her very closely and nothing else has occurred. I mean her surgeon indicated that there's a very good chance that she will get cancer again, but nothing so far has happened and she is being carefully followed. (Gladys, PP17, field conversation, August 30, 2001) Familiarity with clinical trials. Having a strong family history of breast and/or ovarian cancer meant that many of the participants and their family members met many of the criteria required to enter clinical trials related to breast and/or ovarian cancer. A few participants shared how they were at times solicited and encouraged to participate in clinical trials because of having such a strong family history. These participants made the analogy that, just as they easily met many criteria to enter different clinical trials, they also easily met the eligibility criteria for genetic testing for BRCA1 and BRCA2 mutation analysis. Meeting these criteria for different trials was interpreted by some as a confirmation of their high-risk status and their high likelihood of having an inherited mutation. Ginger said, "Genetic testing is just one more tool to manage my life, to manage my chronic illness." In this statement, Ginger acknowledged that she viewed her past breast cancer experience as a chronic illness that she needs to manage. As expressed earlier in this section, many participants spoke of their concern that breast cancer may reoccur or that they were more at risk of a recurrence or just generally more at risk as a result of having a diagnosis of cancer. In the quote above, Ginger also makes reference to how breast cancer is now part of her reality, part of her life. Breast cancer was not a state outside of her in the sense that once she received treatment, the cancer was no longer with her. On the contrary, like other participants, Ginger 112 emphasised that breast cancer was her "chronic illness to manage" and that it was now part of her life. Ginger's view that breast cancer is a lifelong disease has been reinforced by outside experiences, such as reports of recurrences in her support group: More and more, it seems women are getting breast cancer like 7 years later or 9 years later or 12 years later.. .In this support group where I used to go, I met many wonderful older women who are all survivors. But many of them have had, like my mother, two cancers. (PP11, field conversation, May 7, 2001. Ginger's mother died following her second breast cancer diagnosis) The chronic illness perspective has also been documented by Kenen et al. (2003b). In their study, women with a family history of breast and/or ovarian cancer used various coping strategies to "get on with their lives" and not be dominated by thoughts of long-term risk for cancer. These included adopting a health lifestyle, reducing stress and participating in annual screening. The authors found that women in their study faced issues concerning adaptation and an uncertain future similar to those faced by individuals living with chronic illness. Hence, they describe this process for individuals with a family history of breast/ovarian cancer indicative of HBOC as living with chronic risk. Having to deal with broken plans. Having to deal with broken plans reflect some of the participants' reality whereby as a result of many diagnosis of cancer in the family or having to deal with mortality as a result of some of these cancer, some participants were not be able to follow through with plans they had made. When making short - and long - term plans, most individuals do not have to consider, what if someone gets a diagnosis of cancer? How will this affect our plan? The following family in this study had to address these issues after the fact. Gladys, a participant, relates how her sister's plans and hers sadly changed as a result of her sister's second breast cancer diagnosis. 113 Their other sister had been diagnosed with breast cancer while still dealing with her rectal cancer. But they said, 'Well, you know it looks all right now. We don't see any gaps in the bowel cavity and, you know, maybe try a little bit of liquids and so on. And she started to eat and started to get better. And so after 2 weeks my sister G and I said, 'Well, you know we can't just stay on here. In a month or more, we have to go home to our lives.' So we left and G said to me, 'Look, I have been over three times and you also have been over three times, but I'm retired and my husband's retired so I am more flexible.' So she said I didn't have to go again, if...when N dies - because we knew it would be within the next month or two - that she would handle it. She was going to go to the funeral. So I said, 'that's fine.' So, she was all pumped up to organize it and, when N died, G was diagnosed with a second lump and was told she had to have a mastectomy right away. So, she couldn't go to the funeral. So 1 had to go to the funeral. And that was so upsetting to G that I think that's probably why she's had all this trouble and all this pain because she was...she was absolutely destroyed at not being at the funeral. She and N had been very close in age; they were about 13 months apart. (Gladys, PP17, field conversation, August 14,2001) Here is a reinterpreted biographical account of Emma's experience of broken plans resulting from her two sisters dying from breast cancer as shared during her interview: It only had been recently that Emma could speak about the death of her sisters. Emma had made plans to grow old with her two sisters. Instead, she saw them die of breast cancer as she nursed them both to their deaths. One of her sisters was like a mother to her. Upon this sister's diagnosis, Emma and her husband made the decision to sell their home and move closer to her dying sister. Soon after their move, Emma took her sister into their home and cared for her till the end. Following the death of both of her sisters from breast cancer, Emma was then diagnosed with breast cancer as well, at the age of 43. Emma said that, for a long time and still now, she suffers from survivor guilt. (Emma, PP20, field notes, October 27, 2001) Perhaps this section could also be called adaptation to living with a family history of breast cancer. However, adaptation differs between participants who had children and those who do not. Those who had young children often said that, "Dying is not an option." Evelyn, who did not have children, spoke of her risk of dying from her disease as something to expect. Evelyn was first diagnosed with breast cancer at the age of 46 and again at the age of 50. She said she was still premenopausal when she received her two breast cancer diagnoses. Evelyn explained 114 what she meant with seeing death as something to expect because of her lived experience with cancer: Like living an uncertainty is what one of my psychologists said, you know, and you do but you don't dwell on it, you know. And I'm learning to be more comfortable with the thought of dying and watching, you know, I've watched a lot of friends die and see how gracefully and graciously they do it but it does, I mean it's, it's why I'm retiring in two years. I mean my odds of living to 65 are not great, so why spend all that time working? (Evelyn, PP5, field conversation, March 23, 2001) Evelyn goes on to explain: The head librarian, I saw him just after Christmas and he asked how my holidays were and I told him that my dad had just died and it was unexpected. And he said, um, oh, that must be really hard, it's a long time ago that my parents died but it really causes you to face your own mortality. And I looked at him and I said, 'David, been there, done that twice already.' (PP5, field conversation, March 23, 2001) In the above quote, Evelyn refers to her poor prognosis which had been given to her on both of her cancer diagnosis. In conclusion, after hearing how participants coped with their strong family history of cancer, I now understand more fully their description of genetic testing as "nothing in comparison to their experience with cancer and their family's history of it." Certainly, having understood these individuals' experiences with cancer through the nine themes presented above has given me a better sense of why they came to be interested in having genetic testing. Through these themes I have also come to understand how living with such a history would make individuals more receptive to genetic testing because they have had much time to adapt to living with their perception to an increased risk of a subsequent cancer and risk of cancer for their family members, and, to their belief that perhaps all the cancer cases in their family results from an inherited mutation. In the next section, I look at how genetic testing entered the participants' lives and the reasons that led them to accept genetic testing. Many of the reasons operate within the 115 participants' experience of living with a family history of cancer and, more specifically, within the themes just discussed. Encountering Genetic Testing As I began to analyse the reasons that brought the participants to pursue genetic testing, it became apparent that a process took place before the participants arrived at in genetic counselling. In this section, I begin by describing how participants became aware of the availability of genetic testing for breast and ovarian cancer susceptibility. Then, I present the different reasons for having genetic testing described by the participants in accepting to have genetic testing. Hearing of the availability of B R C A 1 and B R C A 2 mutation testing Very few participants had inquired about the availability of a test that could assess whether their cancers were inherited. Of all the 21 participants, only 5 self-referred for genetic testing for breast and ovarian cancer susceptibility after becoming aware of its existence either through the media or through breast cancer support groups. Five other participants had been contacted by a family member to see if they would have testing done on their behalf and 11 other participants had been made aware of the availability of this test by one of their health care providers. Self-referral. As discussed in the previous section, Implications of Living with a Family History, a newly diagnosed case of cancer in the family often acted as a catalyst for change, by reminding family members of their increased risk to cancer and to perhaps leading them to have a follow-up screening test. This applies to four of the five participants who self-referred. As also discussed in the previous section, each new case of cancer further influences their perception of risk and a 116 need to proactively reduce the risk. These two events led many participants to become attuned once more to all that is said about breast and/or ovarian cancer. Hence, three of the participants described how during this period they came to hear of the availability of genetic testing either through the media or through a breast cancer support group. One participant, Marcy, asked about genetic testing after the death of a cousin from kidney cancer. As Marcy explains, Bernadette - that's my mother's sister's daughter - she had breast cancer when she was 60. That's my first cousin. Her brother had just died of kidney cancer, so we've got quite a few in the family. When you start thinking about it and writing them all down, you start to see how they add up. So that's the reason we wanted to be - asked to be - genetically tested. (PP18, Marcy, field conversation, September 13, 2001. Marcy's two sisters, two female cousins and herself were diagnosed with breast cancer. One female cousin was diagnosed with colon cancer. Al l are on the maternal side of the family) The fact that they all lived in a province where genetic testing for BRCA1 and BRCA2 mutations became available led them to be more aware that such a test was actually possible for them. Upon a family member's request. Similar to those who had self-referred following a reminder of their increased cancer risk, family members (5) who approached primary participants about having genetic testing on their behalf had the same catalysts. Joyce, who had been approached by her cousin Stacey to have genetic testing, explained how her cousin had recently lost her sister to breast cancer. I had the opportunity to interview Stacey. The following quote tells how Stacey came to hear of genetic testing: Every few years I'd hear of another cousin. Right now, out of the three generations, there are 13 women that have had breast cancer in our family. My sister was the last one. So, once my sister developed breast cancer at the age of 46, it got quite personal then. And, um, just watching her go through what she was going through, um, and thinking of her children, (she has three daughters) it just made me want to do as much as I can to find out what's going on with our family. Then several years ago I saw an ad in the paper about this breast cancer research thing. So I asked my doctor if he would refer me, which he did. And that is how the whole thing started for our family. (Stacey, SP1/PP3, Joyce's cousin, field conversation, June 4, 2001) 117 Joyce offers the following interpretation of her cousin's interest in genetic testing: My cousin had asked me to go for genetic testing. She has since lost her sister to breast cancer. Last April she lost her. So she's the one that asked me. She phoned and asked if I would mind and I didn't. It didn't bother me to go down and to see. (Joyce, PP3, field conversation, December 12, 2000) Joyce explained that, before her cousin called, she did not know of the availability of this test but, after discussing it with her cousin, felt that it would be a good idea. Joyce then made the first call to the Hereditary Cancer Program, the clinic her cousin told her offered the test. Joyce understood that she was being asked to have genetic testing, because she was the only living individual with a breast cancer diagnosis who was locally available. Four other participants where also contacted by a relative to have genetic testing on their behalf following a recent diagnosis of breast cancer in one of their close relatives. All of the five family members who contacted a primary participant in this study to seek their willingness to have genetic testing on their behalf all lived close to the Hereditary Cancer Program and were aware of its existence. Hence, it was after these relatives had consulted a health professional from the Cancer Agency that they contacted the primary participants. Of the five participants who were approached by an extended family member, four said that they did not object to having been contacted. They said that they understood their relatives' suddenly increased perception of risk for breast cancer because of having a close family member diagnosed with the disease. Besides, they themselves could now see value to this test for themselves and their own immediate family members. The one participant who had objected to being contacted by an extended family member to have genetic testing on her behalf, spoke to the issue of timing in terms of when she felt it was appropriate to approach a family member about genetic testing. 4 8 R e c a l l that, in o rde r to be e l i g i b l e f o r genet ic test ing f o r BRCA1 and BRCA2 mutat ions , a b l o o d s a m p l e f r o m an a f fected i n d i v i d u a l w i t h breast a n d / o r o v a r i a n c a n c e r is needed . 118 It was a cousin that phoned me, and she'd heard about it, and she asked me if I'd go in and give some of my blood. I was just going through chemo and radiation at the time. And I, well I hadn't talked to her in years. Well, phoning me up to ask me to do that after not talking to me in years! Well, that kind of threw me for a loop. But I said, sure, no problem on my part. And that's where she phoned the genetic counsellor, and then the genetic counsellor phoned me. And I said, yeah, I will come in and let them take my blood and stuff like that. (Barney, PP7, field conversation, December 4, 2001) Barney later explained that she was upset at her cousin "calling her out of the blue like that" after not having talked to her for years. Perhaps Barney felt, although she was receptive to genetic testing, it was not the best time to have it done because she was in the midst of her own breast cancer treatment at the time. Certainly, the timing of the approach of when to contact a third party to have genetic testing for someone else's behalf needs further inquiry. Although it was not the purpose of my study to assess and analyse family dynamics and family interactions as a result of genetic testing, this is an area that would certainly be interesting to pursue in future studies, in light of Barney's response. Research questions could include: What are the best clinical guidelines for deciding who should inform third parties of another individual's interest in genetic testing and when is the best time to approach them? As Barney later specified in her interview, she would have preferred to be approached by a genetic counsellor or a genetic nurse rather than being asked by a cousin whom she had not spoken to in years. There is still little empirical knowledge about the best way to approach family members about genetic testing while creating as little harm as possible, when the request is made on behalf of a third party. Another participant, Victoria, explained that her cousin had not sought her own sister's interest in being tested, despite the sister being recently diagnosed with breast cancer but, instead, called upon her. Victoria's cousin felt that her sister was having a hard time adjusting to her breast cancer diagnosis while concurrently receiving chemotherapy and would find genetic 119 testing a burden 4 9 Victoria's diagnosis of breast cancer dated back 11 years. Victoria explained that she had heard of genetic testing and had always been interested in it but had never pursued this interest. Through this call she received from her cousin's and her mother's genetic counsellor, she found out that both of them had independently pursued genetic testing and that her mother had actually had gone ahead with it. She passed away from breast cancer before obtaining her results. Victoria's mother had not shared with her that she had pursued genetic testing. Victoria's cousin, who wanted to have genetic testing done following her sister's diagnosis, inquired about the test but found out that one of the eligibility criteria for testing was a past diagnosis of breast or ovarian cancer in their own body. As noted in the clinical trial study by Loader, Shields, and Rowley (2004), the individual who initially requests testing appears to be the one who undergoes the most stress when asking for genetic testing. This stress associated comprises distress from cancer risk, genetic counselling, distress about general health and emotional health, and breast cancer worry. Recommendation of a health care provider. Eleven of the 21 participants had heard about genetic testing from one of their health professionals while receiving follow-up care or follow-up screening at the Cancer Agency. Contrary to Barney, who felt upset being called "out of the blue" by an extended family member, none of these 11 participants mentioned feeling any anxiety over their health professionals' recommendation that they have genetic testing done. Recommendation from a health care provider has been reported in other studies as the most frequent method of referral for individuals with hereditary cancer syndrome (Campbell et 1 did not have the opportunity to interview Victoria's cousin. Hence, I do not know how she came to hear about the availability of genetic testing and why she felt that it was improper to ask her sister. 120 al., 2003; Mahon & Casperson, 1995; van Zuuren et al., 1997), especially when physicians know that genetic testing can be obtained locally (Wideroff et al., 2003). While these studies indicate that health professionals do know how to identify and know when to refer women with a family history of breast cancer for risk evaluation, not all studies point in that area. A recent review by Hutson (2003) showed that most women who attended these programs tended to be self-referred and had underlying psychological distress. Based on this finding, Hutson recommended that more education be given to health care providers to help them identify individuals at increased risk of hereditary cancer who could benefit from genetic counselling and screening. This recommendation is also supported by recent studies indicating that lack of genetic knowledge and lack of local availability for genetic testing is a barrier to physician referral (Wideroff et al., 2003). When I asked all the participants how genetic testing entered their lives, many first said they were uncertain but that it most likely would have been through one of their health care providers. Given that all of these individuals were being followed in a high-risk-breast-cancer clinic, I thought that most would have been referred by health professionals. Of the 11 participants referred by a health care provider, only 3 actually said that their health professionals had requested a referral for genetic testing to help them make immediate decisions about their breast cancer treatment. Obtaining their first genetic test result took (for some) almost a year; they did not in the end use the test results to decide whether to have a bilateral mastectomy and prophylactic oophorectomy. Instead, following their family physicians' and oncologists' recommendations, they opted to follow through with a bilateral mastectomy. In summary, the process of genetic testing did not always begin with a personal interest. Rather, most individuals were either advised to have such a test by one of their health care providers or received unexpected requests by a relative to get tested, because the relative was 121 ineligible. Among the 21 participants who had genetic testing done, in 16 cases, the genetic testing process was initiated by others. Nonetheless, all 21 could find good reasons for getting tested. These reasons represent the last phase of the process to the built up of genetic testing. Motivations for having genetic testing Just as Hutson's (2003) found in her study, in my analysis of the motivations given by participants for having genetic testing, the main motivators given were not polar opposites but more scaled from having genetic testing for themselves to having it for others as well. However, within these scales, priority distinctions were found. The distinctions reflected aspects of self in relation to others. The same view has also been documented in recent research by d'Agincourt-Canning (2003) who also studied the experience of high-risk women and families with genetic testing for inherited susceptibility to breast and ovarian cancer. Four categories of motivations for having genetic testing were constructed within my study: 1. For others only. 2. For self and others without priority. 3. For others first, and then for self. 4. For self first, and then for others. Within these categories, some priorities were assigned by the primary participants. I derived these four categories from my analysis of the motivations that led the participants to decide to have the genetic test done. The conditions for each one of the categories will be discussed below, but, to begin, here is a brief description of the categories. The first category of motivators given for having genetic testing was for others only. The second and third, although containing the same words, for self and for others, differed in the 122 priority to each word given by the participants. That is, having genetic testing was either done for oneself first, then for others; or for others first, then for oneself as well. This third category is represented as for others first, and then for self. The fourth category is for self first, and then for others. The difference between the fourth and the second and third categories is that, initially, when participants decided to get tested, their motivations for having the test done was for themselves only - until they received genetic counselling and realised that their test results could be important to other family members. Then their motivations shifted categories: from for self only to for self first, and then for others. The latter (fourth) category shows the aspect of self in relation to others evident when individuals make decisions about medical testing in the context of genetics. The four categories, although I present frequency rates for them among the 21 participants, are not mutually exclusive. Rather, I use these categories to illustrate the complexity of decisions involved in genetic testing that the participants underwent. When participants said that they were having the test done for others, they meant others to include immediate family members as well as extended family members. The one participant for who other represented only her children, expressed how she did not feel that it was her responsibility to let her extended family know that she had gone for genetic testing.50 For others only. Of the eight participants whose reasons for having genetic testing were for others only, four had been asked by a third party to have genetic testing done: two by a niece and the other two by a female cousin. Participants explained that they felt genetic testing was too late to tell 5 0 The issue of responsibility for communicating genetic information is a delicate one, because of the ethical implications involved. These include, whose duty is it to inform? What are the harms and benefits of communicating, or not, that one has received genetic information? Although this study's focus was not to assess the ethical issues of genetic testing for individuals and their families, certainly some of the participants brought up some valuable ethical issues. These issues are discussed later in this dissertation. For reviews of some of the ethical implications of genetic testing, see d'Agincourt-Canning's (2003) doctoral dissertation and Burgess (2001). 123 them if they were at increased risk of developing breast and/or ovarian cancer, since they had already developed the disease. Hence, they viewed the test as only useful for family members still unaffected by the disease. Just like in d'Agincourt-Canning (2003) study, very few participants had considered genetic testing just for themselves only. However, as other studies have reported, many individuals have genetic testing out of a sense of responsibility to others, such as for their children (d'Agincourt-Canning, 2003; Lerman & al., 1996; Lerman & Croyle, 1995; Lerman, Daly et al., 1994; Lerman et al., 1996; Lynch et al., 1999). Participants believed that, by having the test done, at least family members would have the option of being genetically tested to see whether they themselves carry an inherited mutation.51 The participants also felt that, if they were found to carry a mutation, they could at least use this information to convince their family members to be more proactive about their health and about cancer screening. This links with the view of some participants that they are family guardians and are always looking for ways to make their family members aware of their risk for cancer. Having genetic testing for others only has also been documented by Bottorff (2002) and Hallowell (2002), who describe how their participants wanted to have genetic testing in order to warn family members of their possible increased risk of an inherited mutation. The following situation illustrates the category for others only as reflected by one participant within this study: I just thought it was something I could do that, if I tested positive, then they would be more aware and they would have more information. But I really think that it's more important for the people that have not tested positive or have not had cancer. You see, genetic testing was nothing to me. It's more for other people, members of the family. I know I am more at risk to cancer. I have already been diagnosed with breast cancer and was told that I was terminal 5 years ago. I mean, what if I had the gene? It made no difference. (Macy, PP1, field conversation, December 11, 2000) 5 1 In the Hereditary Cancer Program, one of the criteria for unaffected individuals to have genetic testing is that an individual within the family must have an identified BRCA1 and/or BRCA2 cancer gene mutation. Not all cancer genetics programs have this criterion. 124 Like Macy, the other seven participants who gave reasons in this category also felt that genetic testing was "too late for them." For self and others without priority. Having genetic testing done for self first, and then for others is the second category of motivations. Here eight participants could see a benefit for themselves, as well as valuable reasons to have the test on behalf of their family members.52 They wanted to find out why there were so many cases of breast and/or ovarian cancer in their families and if all of these cancers resulted from an inherited mutation. Among those eight, three had the test done to assess their risk for ovarian cancer and to help them make informed decisions about having prophylactic mastectomy and/or oophorectomy. These three participants felt that their risk to breast and ovarian cancer was "high enough," with or without the mutation, and went ahead with their surgeries without waiting for their test results. One participant had a prophylactic oophorectomy, while the other two had bilateral prophylactic mastectomy. One of these two had received a lumpectomy as part of her breast cancer treatment but still felt at risk for a subsequent breast cancer. As described in the first section of this chapter, this participant, like others, constantly felt at risk for a subsequent breast cancer and for a recurrence especially since her mother had died from recurrent breast cancer. The participants hoped that genetic testing would provide them with information that could at least tell them if their cancer was hereditary. Juniper explained: I wanted to be more informed. I wanted to know that if I had a genetic type of cancer or a hereditary type, or whatever you call it. I felt that not only did I owe it to myself to know 5 2 When the participants spoke of having the testing done on "beha l f of their family members and extended family, I do not known i f the participants had understood before having genetic counselling that their genetic information would also have implications for others. The genetic counselling process was not studied within this research, but the evolution of one's understanding of the implications of genetic information for others would be an interesting area to study. 125 that so that I could make decisions regarding my other breast and decisions regarding my whole body - my health, you know health wise, that it was also imperative having a sister. I owed it to my sister to avail myself to anything that is being offered to me that could help her out and ease her anxiety about not only having a mother at 45 with breast cancer but having now a sister with breast cancer. I owed it to what I originally thought were only my daughter and my two nieces. But I didn't realize that I also owed it to my nephews, because I didn't realize that it can carry along that way. I felt that I owed it to my whole family; that, if this information was there and I could avail myself to that information, that it was really important for me to get it because it would help us be able to make decisions no matter which way it went. It would help us make the right decisions for our bodies. (Juniper, PP2, field conversation, December 13, 2000) Juniper added further that, while she already had one breast removed as part of her breast cancer treatment, she had seriously considered having her other breast removed if genetic testing showed that she carried an inherited mutation. Although many other studies have documented similar reasons for having genetic testing that go beyond the individual's need (Bottorff et al., 2002; Burgess & d'Agincourt-Canning, 2001; Hallowell et al., 2002; Lynch et al., 1999), Juniper's response shows the aspect of the self in relation to others and how it is difficult to dissociate the self from the rest of the family members in the context of genetic testing. The reasons reported by the above studies are similar to those expressed by the current study participants: to use genetic information to increase utilisation of cancer screening among family members, to adopt more proactive behaviours aimed at reducing general risk to cancer, and to know the etiology of their cancer. Ginger, whose reasons fall under this category, explained that, although uncertain about what she would do with the genetic information, she believed that information was power and that, if the information was available, she should try to obtain it. Although Ginger had no intention of telling her daughter about her results, since her daughter was only 9 years old at the time of her genetic testing, "At least, if one day she wanted to know, then the information would be available for her." Just as the participants of d'Agincourt-Canning's (2003) study that viewed 126 knowledge as power, participants in my study also viewed gaining information as a form of control over their disease and felt that it is better to know if one has an inherited mutation than not to know. For others first, and then for self. Having genetic testing done for others first, and then for self is the third category of reasons. Here, participants' reasons to have genetic testing were for others first then for themselves. Three participants fell under this category. Here are two quotes that distinguishing between the second and third category of reasons for having genetic testing: For self and others I was having it for my children and also for myself. (Gilligan, PP16, field conversation, August 14, 2001) For others to self I was first doing it (genetic testing) for the kids then for my self. (Mimi, PP15, field conversation, August 21, 2001) Gilligan's quote does not give priority to one reason, whereas Mimi's quote does. Evelyn, one of the three participants who said that they were first having the test done for others, then for themselves, explained that she was first having the test done for her siblings and extended family members but that she was also having it done for herself. She said that she did not need this test to tell if she was at increased risk for breast cancer as she already had two breast cancers but did want to know if she was at increased risk to ovarian cancer. For self first, and then for others. The final and fourth category of reasons is for self first, and then for others. Although, as in the third category, only two participants are within this category, its particularity is worth 127 discussing. The then for others reflects how, at first, these two individuals' acceptance of genetic testing was for self only and then became directed toward usefulness to others following the genetic counselling session. Both had entered genetic testing by a recommendation from their physician. The two participants, Becky and Louise, explained that it was during their genetic counselling session that they first became aware of the value of this test for other family members. Louise explained that her initial reason for testing was to confirm that her ovarian cancer diagnosis was "not her fault" - that it was not the result of anything she did in the past. As for Becky, she explained that she "had just gone along for the ride, like one would when recommended to be followed up by a nutritionist." The following quote describes the feelings Becky experienced when she realised the information her test could provide to her family: So I go down there and I was nervous, I was surprised. I didn't know why I was so nervous, but then Dr. X did a very good job at putting me at ease. Then Dr. X starts explaining me that my family history with breast cancer is quite interesting cause - you know - no ovarian cancer but so many late onset breast cancers. And Dr. X said, 'For example, your grandmother and her three sisters were in their 80s, 90s when diagnosed with breast cancer. Now, we'd be interested in knowing what happened like what protected them, you know - if there was a gene that protected them. But, basically, she said there are too many breast cancer cases in my family to be due to chance and she said, you know, we can maybe help you and you can help us.' So I'm starting to think maybe I can contribute something to science. But somehow I was still thinking this is just me. This is just my decision, okay, and I decided now why am I doing this anyway? I mean what good is it going to do and then I decided well, no, I think I want to do it because, what if something changes in the future and there is something they can do about it? Then they will already know that, yes, I am a candidate or no, I'm not a candidate. I mean you never know. And I just feel that information is a good thing...Then Dr. X said, 'You know, we'd like to test you.' I said yeah. She said, 'We like to test the people who had cancer. So we'd like to test you. We'd like to test your aunt Katherine and we'd like to get tissues that maybe still exist from your two aunts that died and your grandmother.' All of a sudden I'm going, 'you mean I have to tell all those people that?' And it didn't sort of hit me, uh, till then. I thought they were just going to test me.... (Becky, PP8, field conversation, April 17, 2001) 128 No participants in this study were simply self-interested in testing. This fact may speak to the notion that genetic information has implications not only for the self but for other family members as well and to what Hallowell (2002), a sociologist, attributes as a sense of 'genetic responsibility" towards others. In an empirical study, Hallowell (2002) noted that women who participated in genetic counselling and testing for breast and ovarian cancer did so out of sense of responsibility to their family (past, present as well as future) to determine their risk to an inherited mutation to the disease. Hence, the current finding further supports the need to assess individual's decision to having genetic testing as to determine the potential impact and lack of if they receive uninformative genetic testing results. In this chapter, I described the process that led individuals to be interested in genetic testing and the reasons that led them to have it. I suggested that there is a buildup before individuals become receptive to genetic technology, which includes living with a family history of breast and/or ovarian cancer and the implications of living with such a history. I offered a list of four categories of motivations for having genetic testing with their respected reasons as offered by the participants. What these categories provide to clinics is a condensed list that encompasses the many reasons already documented in other studies by distinguishes how the self is situated in the reasons people bring to having genetic testing. Further, the four categories also represent incentives for having genetic testing done. These incentives also serve to locate the participants' satisfaction or dissatisfaction with their genetic testing experience, after receiving their results. This is further discussed in chapter 6. 129 C H A P T E R F I V E : Interpreting and Making Sense of Genetic Test Results This chapter seeks to extend the understanding of participants' interpretation of their genetic test results by exploring the meanings attached to their interpretations. Although the participants were told that their results from the test for BRCA1 and BRCA2 mutations were negative (meaning that no mutations were found), the participants had different views on how such results ought to be interpreted. My goal here is to demonstrate how patterns of interpretations are shaped and altered throughout the making-sense experience that is grounded in individuals' belief systems. These belief systems influence how they interpret their test results. First, I present the types of interpretation participants gave to their test results. Second, I lay the groundwork for the structures informing their beliefs to how they interpreted their test results. Then, I move to contrasting individuals' stories to identify differences and similarities that render individual cases visible within the common patterns identified (Thorne et al., 1997). Types of Interpretation Given to Their Test Results Table 4 presents the three types of interpretation that I derived from the participants' accounts. When I asked the participants how they interpreted their test results, they would start off by saying, "It is negative for BRCA1 and 2," then go on to say "but," "or," or "and" They would follow this with a sentence that resembled the ones in the meaning row of Table 4. 130 Table 4: Types of Interpretation Derived from the Participants' Accounts to Uninformative Results Interpretation Confirmation of mutation carrier status Ambiguity regarding mutation carrier status Refutation of mutation carrier status But we (our family) do Maybe we have or And we do not have the have a mutation; either maybe we don't have a genetic type of cancer. Meaning one that they have not mutation and that the yet identified or one that results are incomplete or they have missed. inconclusive The largest number of participants (11 out of 21) interpreted their results as ambiguity regarding mutation carrier status. The other two types are opposite, but not extreme opposites. That is, participants who interpreted their results as a confirmation to their mutation carrier status (six participants) understood that no mutation was found in either gene, but they still believed with certainty that they carry an inherited mutation. There was no doubt in their minds that they are carrying an inherited mutation while the participants who interpreted their results as a refutation to their mutation carrier status (four participants) did not doubt the outcome of their results and believed that they were not carrying an inherited mutation. As for the participants who interpreted their results as ambiguity regarding their mutation carrier status (11 participants), this groups leaves room for uncertainty. Although three types of interpretation were found in the participants' account, some of the participants' interpretations do not fit exclusively into one category. That is, in some parts of their interviews, participants would at times say, "But I know I have an inherited mutation, they just have not found it." Later in the interview, they would add, "My daughters were glad to hear 131 that I do not have an inherited mutation." As well, four participants' interpretations moved at times from saying they were certain they had an inherited mutation to saying that "Maybe we do or perhaps we do not have an inherited mutation and vice versa." As for the four participants who believed they did not have an inherited type of cancer, one would at times comment on how she understood that there is always a possibility she might carry some other yet-unidentified gene mutation, or perhaps the current technology missed her mutation but that for the time being, she is interpreting her results as a refutation to her mutation carrier status. Nonetheless, when I asked her specifically how she viewed her test results, she answered that she was found "negative -meaning that no mutations had been found in her two breast and ovarian cancer genes analysed. How participants arrived at their interpretations of genetic testing results is discussed by presenting generic structures found to inform their beliefs about whether or not they have an inherited mutation. The generic structures represent key statements located in the participants' experiences. The term "generic" in this context portrays how there is not one participant experience that captures all the structures described nor does one structure alone represent an individual's experience. In general, a participant experience will present with some of the 13 structures derived from the synthesis of all the participants' experiences. Hence, the generic structures represent a beginning point which individuals might have used to interpret and make sense of their result. They are generic as well in the sense that they do not imply any linear or causal model but provide rich, detailed description and interpretation to the phenomenon studied. Some people's interpretations are informed by generic structures usually found within one type of interpretation such as to explain why they believed they did not have an inherited mutation, while others used the same structures to explain why they believed they had an inherited mutation. 132 The participants' interpretations and making-sense experiences were extensively coloured by the four possible interpretations in the letter they received from the Cancer Agency (provided below). Recall that the purpose of this study was neither to assess the content of the genetic counselling process nor to evaluate how test results were communicated. Hence, although I highlight part of the letter participants received, I only do so in the spirit of providing the context in which the participants' interpretations developed when they tried to make sense of the test results contained in the letter. The four interpretations offered to the participants were: 1. A mutation may exist in the region of the BRCA1 or BRCA2 gene that our lab looked at, but it has not been detected by our current testing method OR 2. A mutation may be present in the as-yet-untested portion of the BRCA2 gene OR 3. The responsible mutation may be in another, as yet unidentified, hereditary cancer gene OR 4. You do not have an inherited breast/ovarian cancer gene mutation, which means that your cancer diagnosis may have occurred by chance. The Cancer Agency's letter continues with, "We must highlight that this result does NOT mean that we have completely ruled out an inherited breast/ovarian cancer gene mutation in your family. Based on your family history of cancer, we recommend that cancer screening continue as previously recommended." It ends by stating that, at this time, they do not have the necessary resources to undertake additional testing such as full sequencing of both genes.54 The clinical interpretations above have also been reported in the literature (Bish et al., 2002a; Claes et al., 2004; Frost et al., 2004; Hallowell et al., 2002; Hallowell et al., 2004; Iglehart et al., 1998; Lynch et al., 1999; Peshkin et al., 2001; Wong et al., 2001). Peshkin et al. 5 3 Only 65% of the B R C A 2 cancer gene was screened. The Agency explains that this region represents the surface area of the B R C A 2 gene where most clinically significant mutations have been found to date. 5 4 For a complete copy of this letter, see appendix 1. 133 (2001) conclude that, when no mutation is found in an affected individual with a family history of risk for hereditary breast and ovarian cancer (HBOC), that several explanations are possible: (a) a mutation might still exist in the BRCA cancer genes tested but was not detected by the methods used, or (b) the individual might have a rare or as-yet-undiscovered mutation, or (c) that the individual and the clinic should consider that perhaps the person tested represents a sporadic case within a possible hereditary cancer family. Their clini