Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

"What brings you here today?" : perspectives of older women on hearing, help-seeking for hearing problems,… Carson, Arlene Jane 2000

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

Item Metadata


831-ubc_2000-565211.pdf [ 17.83MB ]
JSON: 831-1.0089705.json
JSON-LD: 831-1.0089705-ld.json
RDF/XML (Pretty): 831-1.0089705-rdf.xml
RDF/JSON: 831-1.0089705-rdf.json
Turtle: 831-1.0089705-turtle.txt
N-Triples: 831-1.0089705-rdf-ntriples.txt
Original Record: 831-1.0089705-source.json
Full Text

Full Text

"WHAT BRINGS Y O U H E R E TODAY?" PERSPECTIVES OF OLDER W O M E N O N HEARING, HELP-SEEKING FOR HEARING PROBLEMS, A N D THEIR FIRST AUDIOLOGY APPOINTMENT by A R L E N E JANE C A R S O N B.Sc , McGill University, 1977 M . Sc., The University of British Columbia, 1982 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIRMENT S FOR THE D E G R E E OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (School of Audiology & Speech Sciences) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A October 2000 © Arlene Jane Carson, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ftucjiolo^y. J J>pQ&cS\ Sciences The University of British Columbia Vancouver, Canada Date Dd&bT I 3 f ZMrtO DE-6 (2/88) ABSTRACT Persons with hearing problems typically wait many years from the time hearing difficulties are noticed until an appointment is made for an audiologic assessment. Some hard-of-hearing persons never seek help. Many who finally seek and could benefit from rehabilitative help delay taking further action after their hearing is tested. These facts suggest the need for more research into the nature of help-seeking for hearing loss. The present study explored the perspectives of older women on their hearing, hearing problems, and help-seeking for hearing problems around the time of their first audiologic assessment. A major goal of this research was to explore the links between individuals' experiences living with a hearing loss and the process of help-seeking. A related goal was to examine how these experiences and the help-seeking process may influence and be influenced by the interaction of help-seekers with the audiology system. The main participants in this study were older women living in the community who were seeking help for age-related hearing loss (presbycusis). They were interviewed in-depth from the time they booked their first audiology appointment, through the appointment itself, and up to six weeks following the appointment. Data also included interviews with family members and the assessing audiologist, participant observation of the audiology appointments by the researcher, and journal writing by each participant. Data were coded and organized into a theoretical framework, guided by the principles of grounded theory. The findings of this study indicate that participants engaged in a protracted self-assessment of their hearing before, during, and after seeking help for hearing problems. The self-assessing process is influenced by many factors that facilitate or delay help-seeking; some factors may do each at different points in time. The result of the complex combining of these factors is a "push-pull" effect that moves persons iteratively toward and away from help-seeking. This finding is expressed in the core category of the theoretical framework of this study: "the spiral of decision-making in self-assessing Ill hearing". Findings suggest that the audiology assessment itself and, in particular, the different rehabilitative foci of the audiologist and the hard-of-hearing person may be an important influence in self-assessing and decision-making regarding hearing problems. Three themes: contrasting/comparing, cost vs. benefits, and control, were identified as significant in self-assessing. Persons contrast and compare their hearing against many "yardsticks". They evaluate the relative costs and benefits, including the perceived loss of control, of specific actions taken for hearing problems. The exact nature of the relationships among these three themes remains to be determined. The results of this study underscore the need for further research into the nature of the interaction between audiologists and hard-of-hearing persons. As well, the findings of this study may be applicable to the more general study of help-seeking for health conditions related to aging. iv TABLE OF CONTENTS Page Abstract ii List of Tables x List of Figures xi List of Appendices xii Acknowledgements xiii CHAPTER 1 INTRODUCTION 1 The General Problem 1 The Purpose of this Study 7 The Rationale for this Study 7 Summary 9 CHAPTER 2 LITERATURE REVIEW Overview 10 An Ecological Framework 11 Linking Concepts in Social Ecology with Core Assumptions in Health 13 Example of an Ecological Framework: The ICIDH-2 15 Hearing Loss 18 The Psychosocial Impact of Hearing Loss 18 Effects at the Level of the Individual 18 Effects on Family Relationships 21 Attitude of Others: Stigma 23 Communication in Aging: Additional Stigmatizing Effects 24 Women versus Men 27 The Individual's Responses to Hearing Loss 29 The Construct of Coping and Related Theories 29 Issues of Identity and Self-image 34 Summary of the Literature on Hearing Loss 35 Help-seeking 36 Theories and Models of Health Behaviour 37 V Health Belief Model 37 Self-efficacy 37 Locus of Control 38 Theory of Reasoned Action 38 Theory of Planned Behaviour 39 Application of Theories to Help-seeking for Hearing Loss 39 A Model of Health Services Utilization: The Andersen-Newman Model 40 Exploring Underlying Processes in Help-seeking 43 The Construct of Illness Behaviour 44 Common Sense Model 45 Exploring Delay in Help-seeking 46 Gender Differences in Help-seeking 47 Aging Effects in Help-seeking 48 Summary of General Literature on Help-seeking 49 Help-seeking and Oral Health 51 Summary of Help-seeking and Oral Health 53 Help-seeking for Hearing Loss 54 Studies of Help-seeking for Hearing Loss 54 Factors that Delay Help-seeking 61 Factors that Facilitate Help-seeking and Cues to Action 63 Summary of Literature on Help-seeking for Hearing Loss 65 The Role of the Audiologist and Audiology Service Delivery in Help-seeking 66 "Traditional" Audiologic Practice 68 Self-assessment Instruments 69 Ecological Audiology 70 Clinician-Patient Communication 72 General Summary and Review of Research Questions 76 CHAPTER 3 METHODOLOGY 80 Overview 80 Qualitative Methodology 81 Grounded Theory 83 The Study Design 86 The Sample and the Setting 86 Participants 86 vi Family members 87 The Audiologist and the Audiology Setting 88 Audiology Service Delivery in British Columbia 88 Recruiting Participants 89 Nature of the Data 90 Interviews 90 Interviews with participants 91 Interviews with family members 94 Interviews with the audiologist 94 Participant Observation 95 Documents 96 Journal 97 The "Communication Profile for the Hearing Impaired" 97 "Client Intake History" Form 97 Audiologic results 98 Researcher's notes 98 Analytic Method 99 Transcribing Interviews 100 Coding of Interviews and Documents - 102 Data Analysis in the Context of this Study 103 Themes 109 Writing Memos 110 Presenting Findings 112 Ethical Considerations 113 Issues of Confidentiality and Informed Consent 113 The Role of Researcher versus Clinician 114 Criteria for Assessing Quality 114 Auditability 115 Credibility and Fittingness 117 Reflexivity: The Researcher as Instrument Tool 119 Summary 120 CHAPTER 4 THE PARTICIPANTS CHAPTER 5 OUTLINE OF THE THEORETICAL FRAMEWORK The Core Category The Theoretical Framework CHAPTER 6 SELF-ASSESSING Reacting to Hearing Challenges Piecing Together Relying on Others Persevering Avoiding Abandoning Reacting to Consequences Reflecting Feeling Stress Self-contradicting Minimizing Longer-term Consequences Effects on Relationships Reduced Quantity of Interaction Reduced Quality of Interaction Security Issues Influences Relationships Spouse Other Family Friends and Cohort Members Aging Lifestyle External Factors Chronology of Hearing Challenges Retrospective Prospective Worries Summary viii CHAPTER 7 HELP-SEEKING 167 Disclosing 168 Private vs. Public 169 Family Physician or Specialist 171 Booking a Hearing Test 176 Expectations of Being Tested 176 Seeking Information 178 Gaining Closure 180 Prospect of a Hearing Aid 181 Considering a Hearing Aid 183 Negative Aspects 183 Factors Related to the Device 183 Change and Adjustment 186 Changing identity 188 Losing Control 189 Positive Expectations and Hopes 190 Summary 191 CHAPTER 8 THE HEARING TEST 192 The Hearing Test 193 Hearing Test Procedures 193 Test Environment 195 Relating to the Audiologist 196 Interpreting Results 200 Longer-term Impact 211 Summary 216 CHAPTER 9 PRESENTING THEMES AND INTEGRATING FINDINGS 219 Themes 219 Contrasting/comparing 219 Cost vs. Benefit 223 Control 227 Themes, Categories and the Core Category 234 Summary 241 ix CHAPTER 10 IMPLICATIONS 243 Reflections on the Study 243 Strengths 244 Limitations 245 The Challenge of Methodology 246 Implications 248 Implications for the Literature 248 Major Contributions 248 Hearing Loss 249 Psychosocial impact of hearing loss 249 Gender and aging effects 252 The individual's responses to hearing loss: Coping theory and research 253 Help-seeking 254 Help-seeking for hearing loss 254 Factors that delay help-seeking 255 Factors that facilitate help-seeking 255 Cues to action 257 Theories 257 Theories of health behaviour 257 Theories and models of health services utilization 258 The Role of Communication 260 Implications for Policy and Practice in Audiology 261 The Focus on Measuring Impairment 263 The Focus on Time 264 The Focus on the Hearing Aid 265 Implications for Future Research 268 Concluding Remarks 271 REFERENCES 272 X LIST OF TABLES Table 2.1 Personal and Environmental Factors in Health and Illness 14 Table 2.2 Personal and Environmental Factors in Hearing Loss and Help-seeking for Hearing Loss 78 Table 3.1 Rationale for Using a Grounded Theory Approach in the Present Study 86 Table 3.2 Table 5.1 Preliminary Theoretical Framework Illustrating Axial Coding 108 Theoretical Framework Related to Core Category: The Spiral of Decision-making in Self-assessing Hearing 130 xi LIST OF FIGURES Page Figure 2.1 Model of the 1CIDH-2 16 Figure 3.1 Diagram of Set-up for Participant Observation and Recording of Audiologic Assessment 96 Figure 3.2 Example of Open Coding 106 Figure 3.3 Two Examples from Memos on the Theme "Control" 110 Figure 3.4 Example of Method Memo 111 Figure 3.5 Example of Profile Memo 111 Figure 3.6 Example of Relationship Memo 111 Figure 5.1 Iterative Process Relating Self-assessing, Help-seeking, and Outcomes 128 Figure 6.1 Interaction of Properties of Self-assessing 133 Figure 8.1 Factors that Influence Interaction between Audiologist and Patient in Audiologic Counselling 201 Figure 9.1 Theoretical Hypotheses on the Theme "Contrasting/comparing" 224 Figure 9.2 Theoretical Hypotheses on the Theme "Cost vs. Benefit" 227 Figure 9.3 Theoretical Hypotheses on the Theme "Control" 233 Figure 9.4 Theoretical Hypotheses on the Inter-relationship of Themes 236 Figure 9.5 Model Integrating Themes, Categories, and Core Category 237 Figure 9.6 Illustrations of Themes within Self-assessing 239 LIST OF APPENDICES Append ix A: Schedule of Interviews for Participants and Family Members 293 Append ix B: Interview 1: Topic Areas 294 Append x C : Interview 2: Topic Areas 295 Append xD: Interview 3: Topic Areas; Post-audiologic Assessment 296 Append ix E: Interview with Family Member: Topic Areas 297 Append) xF: Interview with Audiologist: Topic Areas 298 Append xG: Instructions for Journal Writing 299 Append) ix H: "Client Intake History" Form 300 Appendi x l : Sample of Audiologic Test Results 301 Append) x J: Description of Audiologic Assessment and Test Environment 303 Append) xK: Explanation of Audiologic Tests and their Interpretation 306 Append) x L: Transcription Key 307 Append) x M : Consent Form for Participants 308 Appendi x N: Consent Form for Family Member 311 Appendi x O: Consent Form for Audiologist 313 Append) xP: Audiologic Results for Each Participant (PI- P7) 315 Xlll ACKNOWLEDGEMENTS I am grateful to many friends and colleagues who have inspired and cheered me on this journey. Thank you... To the women who participated in this research, for numerous cups of tea and for openly sharing your stories and insights about hearing. To the audiologist who agreed to take part in this project, for your enthusiastic support of research and your commitment to your patients and to audiology. To my advisory committee, in particular Kathy Pichora-Fuller, for consistent support and good direction in matters academic, and for challenging me to do ever better. To Keith Fuller and Deron Stewart for technical support deciphering the vagaries of Word. To my parents, Stan and Jenny, for your love and unwavering support of my adventures over the years. And most of all, my deepest love and gratitude to Z. Emily and Fred for your unconditional love and patience, and for every shoulder rub, kiss and hug. You help me to keep it all in proper perspective. 1 CHAPTER 1 INTRODUCTION The General Problem "What brings you here today?" I often initiated a conversation with a patient who was seeing me, an audiologist, for the first time with this question. It was simultaneously helpful and frustrating. Helpful in that, given the short time available in a busy clinic, it was my best chance for "open" discussion of the patient's reasoning behind seeking professional help for hearing problems. It was frustrating in that I felt this question and the answers it elicited just chipped at the tip of an iceberg. Many older individuals with acquired hearing loss related a litany of communication problems that stretched back many years; yet this was their first hearing test. My professional experience is consistent with research that indicates a lag of anywhere from eight to more than twenty years between the time one is first aware of hearing difficulties and when professional hearing help is sought (e.g., Brooks, 1979; Getty & Hetu, 1994; Kyle, Jones, & Wood, 1985; Stephens, Barcham, Corcoran, & Parsons, 1980; Watson & Crowther, 1989). This lag has important implications for the person with hearing problems, those in close relationships with that person, and the hearing health professional who provides audiologic rehabilitation. M y personal experience as a rehabilitative audiologist raised many questions. Why do persons with hearing problems, particularly related to aging, wait so long to seek help from audiologists? Do people seek help from other sources or use other resources of which audiologists are unaware? What are the factors or triggers that finally bring the hard-of-hearing person to book a hearing test? For every person who shows up to be tested, many more never get to the point of making an appointment. For those who do book and keep their appointments and for whom rehabilitative plans are discussed, many delay further action, often for several years. Some never take subsequent rehabilitative steps to alleviate their hearing difficulties. More 2 questions arise: Why do so many who could benefit from rehabilitation reject or not avail themselves of these services and products? What factors differentiate those who book a hearing test, get their hearing tested, and take subsequent rehabilitative action from those who don't seek help or who stop short of rehabilitative action? Is it possible that the nature of audiology services today is a factor in rejection and underutilization of services? It is difficult to reconcile hard-of-hearing persons' poor involvement with audiologic services and products with the data on age-related hearing loss, presbycusis. The findings of independent population-based studies of hearing loss among community-dwelling older adults in the United States (Gates, Cooper, Kannel, & Miller, 1990), Europe (Biering-Sorensen, Christensen, Sorensen, & Parving, 1997; Davis, 1989) and Australia (Ward, Lord, Williams, & Anstey, 1993) are similar. Auditory deficits increase markedly with age, beginning as early as one's thirties (Davis, 1989), and progress over time until there is clinically significant and measurable deterioration in the ability to detect sound (for a review see Willott, 1991). On an audiogram1 this is noted as a predominantly bilateral, sensorineural, high-frequency hearing loss. Hearing loss has been identified as the third most prevalent chronic disability among older adults, surpassed only by arthritis and hypertension (Binnie, 1994; Haber, 1994). Both the incidence and prevalence of hearing loss increase with age. Plomp (1978) reports that at least 25% of the population at age 65 has a hearing loss sufficient to interfere with communication. Up to 50% of community-based adults aged 75-79 years have some degree of audiometrically measured hearing loss (for reviews see Kricos & Lesner, 1995; Willott, 1991). For the population living in long-term care facilities, this percentage is much higher (Hoek, Paccioretti, Pichora-Fuller, McDonald, & Shyng, 1997; Hull & Griffin, 1989; Schow & Nerbonne, 1980; Shultz & Mowry, 1995). The societal impact of presbycusis will increase in the years to come, because older adults are the fastest growing segment of society and life expectancy 1 An audiogram is a graph that plots hearing threshold for pure tones across frequencies (typically 250 Hz through 8000 Hz) for each ear. 3 continues to increase.2 Thus, the significance of presbycusis as an escalating population health issue is well documented. Presbycusis is broadly attributed to peripheral auditory dysfunction, such as hair cell degeneration within the inner ear, and/or central dysfunction, associated with decline in structures above the level of the auditory nerve. Deteriorating auditory function may be the result of biological changes in cells, vascular degeneration, and/or other environmental insults accrued as normal sequelae to the aging process (Willott, 1991). Such changes ultimately affect an older person's ability to process auditory stimuli, most importantly speech. Frequency discrimination is poorer in older listeners, as is the ability to process sound binaurally (i.e., listening with both ears to the same or different stimuli). There is a reduced ability to differentiate the spectral components of a complex sound and to filter out background noise. A l l these effects may be subtle when considered in isolation but substantial when considered together, especially in difficult listening situations (Schneider, 1997; Willott, 1991). Many deleterious changes occur within the aging auditory system, even in the absence of any measurable elevation of pure tone thresholds. Measures of hearing sensitivity, such as the audiogram, provide a poor measure of impairment because there can be substantial histopathological change in the auditory periphery without corresponding elevation of pure-tone thresholds. It is also important to keep in mind that, while presbycusis refers to auditory system dysfunction, deficits in the ability of older persons to process auditory stimuli, especially speech, may result from deficits in cognitive, psycholinguistic, and other central processing systems in addition to auditory system deficits (for a review, see Pichora-Fuller, 1997). The impairment of presbycusis briefly outlined above carries with it wide-ranging adverse effects on the physical, cognitive, emotional, behavioural, and social function of Older adults (Mulrow et al., 1990; Rousey, 1976; Thomas & Gilhome Herbst, 1980a). Moreover, the importance of hearing in effective communication means that the 2 In Canada, 21.8% of the population in 2031 is expected to be aged 65 and over, compared with 10.6% in 1991 (Statistics Canada, 1992a). In the year 2000 in British Columbia, one in eight persons is 65 years of age or older (Government of British Columbia, 1993). 4 detrimental effects of hearing loss are not restricted to the hard-of-hearing person but extend to communication partners (Hetu, Jones, & Getty, 1993). Older women with hearing loss may be particularly disadvantaged due to the centrality of communication in women's social roles and the multiplicity of their roles at home and in the community (Baruch, Biener, & Barnett, 1987). Hearing aids are the cornerstone of audiologic rehabilitation. Few who have a hearing aid, however, find that it overcomes all their hearing problems (Holmes, 1995). More than half who are fitted with aids abandon or greatly limit their use a short time thereafter (Hickson, Hamilton, & Orange, 1986). Nevertheless, much research is devoted to improving hearing instrument technology, under the assumption that if only the amplified signal can be improved sufficiently, hearing instruments will be embraced by currently dissatisfied users as well as by the vast untapped market that could benefit from amplification but have never tried it. Statistics spanning many decades and from many countries, however, consistently indicate that only 10 % to 20 % of those who may benefit from hearing instruments make use of them, despite the deleterious psychosocial effects noted above (Davis, 1989; Gabbard, 1994; Gates et a l , 1990; Jones, Victor, & Vetter, 1984; Kochkin, 1999; Liston, Solomon, & Bannerjee, 1995; Ward et al., 1993). Relatively little research effort has been expended to investigate non-auditory and non-instrumental factors behind the delay in help-seeking and the poor adoption of rehabilitative solutions by hard-of-hearing persons. A strong point is made by Ross (1997) and Erber, Lamb, and Lind, (1996), who state that the future success of rehabilitative audiology lies not only in greater advances in hearing aid technology, but in greater attention to the social, psychological, and environmental factors associated with hearing loss. Attention to these factors forms the basis of an ecological model of practice and research in audiology, the features of which will be described in this and the next chapter. The consequences of the impairment of presbycusis for the activities and participation of older adults in everyday life may be modulated by many factors, including such social, psychological, environmental, as well as physiological factors. These may include the specific nature of the person's hearing-related impairment (e.g., the specific frequencies affected or the degree of central auditory deficit), and impairments in other 5 areas such as vision or cognition, particularly in relation to aging (Lawton & Nehemow, 1973; Schneider & Pichora-Fuller, 2000). Personal factors such as one's health beliefs and behaviours related to health (e.g., Maiman & Becker, 1974) and coping (Folkman & Lazarus, 1980) play a role, as does the social environment, particularly as it influences interpersonal communication (Ryan, Giles, Bartolucci, & Henwood, 1986), and the personal, social, and communication goals of the hard-of-hearing person (Goffinan, 1963; Jaworski & Stephens, 1998; Pichora-Fuller, Johnson, & Roodenburg, 1998). If we are to better understand why help from hearing professionals is not sought or is delayed, we need to explore these diverse factors and their influence on hard-of-hearing persons' perceptions of their hearing problems and their choices of remedial actions. As the above list shows, our exploration must be mterdisciplinary, since we have much to learn from the intersection of audiology with other fields: social psychology, cognitive psychology, gerontology, communication theory, and health promotion, to name a few. A review of the literature on help-seeking for other health conditions may also be useful to discover if hearing loss is unique among health conditions in how help-seeking is delayed or avoided. We also need to examine the role that audiologists and the current culture of audiology and general health service delivery play in delaying or facilitating help-seeking. There is evidence that negative or resigned attitudes of physicians toward hearing loss pose barriers to help-seeking in many cultures (Brink et al., 1996; Gilhome Herbst, Meredith, & Stephens, 1991; Swan & Gatehouse, 1990). There has been little research, however, exploring how the system of audiologic service delivery and the audiologist-patient interaction may facilitate or hinder the rehabilitative process, even after the first help-seeking visit to an audiologist. Are audiologists and the audiology service delivery system implicated in patients' rejection of rehabilitative recommendations, or in the fact that other rehabilitative aids, such as assistive listening devices (ALDs) and communication strategies training have adoption rates even lower than that of hearing aids (Kochkin, 2000; McCormick, Pichora-Fuller, Paccioretti, & Lamb, 1994; Unger, 1994)? Consideration must be given to how a system (as practised in many jurisdictions of the United States and Canada) that ties payment solely to the hearing aid product rather than overall service may influence the rehabilitative process. Research has shown that hearing 6 aid use is significantly increased where a comprehensive rehabilitation program accompanies the hearing aid fitting. For example, Brooks (1979) found in Britain that the introduction of a counselling program increased the regular use of aids from 17% to 33%. Similar positive effects of rehabilitative and ongoing audiological support on the use of hearing aids and assistive listening devices have been noted in other cultures (e.g., in Sweden; Andersson, Melin, Scott, & Lindberg, 1994) and in specific populations, such as residents of long-term care facilities (Lewsen & Cashman, 1997). Few studies have specifically explored help-seeking for hearing loss. Most of these studies have been population based (e.g., all patients attending a large medical clinic) and have focused on auditory factors associated with help-seeking, using questionnaires and audiometric measures of hearing impairment and disability to assess determinants of help-seeking (Brink et a l , 1996; Gilhome Herbst et al., 1991; O'Mahoney, Stephens, & Cadge, 1996; Swan & Gatehouse, 1990). None of these population-based studies explored the role that the audiologist and audiology system may have played in patients' help-seeking or hearing health service utilization. A different research tack was taken by Getty, Gagne, and McDuff (1996), who undertook a qualitative study of the obstacles to seeking help for hearing loss by conducting focus groups with twenty-four older community-based men and women who had already sought help from an audiologist. This study, while exploratory, discovered a wide range of factors from the personal (e.g., resistance to change) to societal (e.g., complicated referral process to get a hearing aid) that delayed or facilitated help-seeking for hearing loss. Their study illustrates the value of qualitative methodology to help elucidate the subtle but important contextual factors involved in social processes. These factors are given opportunity to emerge in the more open-ended and free-ranging exploration that is a feature of qualitative research. This approach was adopted in the research for the present dissertation. 7 The Purpose of this Study This study explored the perspectives of older women on their hearing, hearing problems and help-seeking for hearing problems around the time of their first audiology assessment. A major goal of this research was to construct substantive theory, through the use of qualitative grounded theory methodology, linking individuals' experiences living with a hearing loss with the process of help-seeking. A related goal was to examine how these experiences and the help-seeking process may influence and be influenced by the interaction of help-seekers with audiologists. In essence, this study was an in-depth exploration of the question "What brings you here today?" expanded beyond the clinical context and placed in a research context. The results of this exploration may have significance beyond the field of audiology, having application to the more general study of help-seeking for health conditions in aging. Rationale for this Study This research was inspired by a desire to learn more about what prompts persons with hearing problems to seek help from audiologists or inhibits them from doing so. The small number of studies to date on help-seeking for hearing loss have been primarily quantitative, procuring data from large samples using questionnaires that are short, closed ended, and designed from the perspective of the researcher, who may make assumptions about how help-seeking "should" proceed. These studies have explored help-seeking only retrospectively, subject to greater influence of recall bias than a prospective study. Also, most studies of help-seeking for hearing loss have articulated the role that family members may play in influencing help-seeking without collecting data directly from these important family members. Furthermore, I am aware of no studies that have considered the influence of audiologists and the system of hearing health care delivery in facilitating or delaying help-seeking for hearing loss and subsequent rehabilitative action or inaction. The interaction of patient and audiologist is an important dynamic that has not been examined. In short, the research on help-seeking to date has not been centred on the experience and 8 perspective of the person with hearing problems. This research has not given hard-of-hearing persons the space and time to voice their stories about how they perceive hearing problems, the help-seeking process, and the audiology system that provides services and products. The current study attempted to address some of these shortcomings by adopting a qualitative research design whose primary source of data was a series of individual in-depth interviews with women. The research design was unique in three respects. First, data were collected at several points encompassing a period of time both before and after each participant's first audiology appointment. In contrast to previous retrospective studies, the present study afforded a novel prospective perspective into how hearing loss is experienced and the factors that may influence help-seeking around the time when help is sought. Second, data were collected not only from the participants with hearing problems, but also from a close family member and from the audiologist who assessed each participant, resulting in unique combinations of perspectives on some of the study questions. Third, the inclusion of the audiologic appointment as an important component of the data collection provided important information on the clinician-patient interaction in a typical clinical context. These unique aspects of research design highlight how the research questions were investigated within an ecological and patient-centred research paradigm. This approach considered the dynamic nature of hearing loss, how it may evolve over time, and how it may be influenced by many contextual factors, both personal and environmental. This study restricted its sample to older women for four reasons. First, few studies have concentrated on women with hearing loss, especially older (retired) women. Second, those studies that have looked at women (usually with men included in the sample) have Suggested that there are significant differences in the way that the impairment, disability, and handicap of hearing loss affect men and women, and differences in the way that men and women deal with those effects (Garstecki & Erler, 1995). Women even appear to differ in the words they use to describe the effects of hearing loss in their everyday lives (Waridel, 1995), a finding that may be of particular significance in qualitative studies, where the data consist largely of the words of participants. Women were chosen 9 exclusively to avoid the potential confound of gender in these areas. Third, studying presbycusis in older women largely avoids the potentially confounding effects of noise-induced hearing loss, which is present in a significant percentage of older men. Finally, the fact that older women outnumber, and will increasingly outnumber, older men calls for more research to discover how the needs of this expanding cohort may be better met. Summary It is recognized among audiologists that audiologic rehabilitative services are underutilized. Many people who have a hearing loss do not seek help from audiologists. Many who access services wait many years before doing so. Many who finally seek help do not take further rehabilitative action once their hearing is tested. This brings up many questions about the nature of help-seeking for hearing loss. Most studies of help-seeking for hearing loss to date have been population-based, using primarily quantitative measures that have focused on auditory factors associated with help-seeking. In contrast, qualitative studies have shown promise to elucidate important and overlooked psychosocial and environmental determinants of help-seeking. The most productive general area for future research on help-seeking for hearing loss is an ecological approach to explore such psychosocial and environmental factors in-depth, centred on the perspectives of hard-of-hearing individuals. The goal of the current qualitative study was to link the experience of individuals living with presbycusis with help-seeking by focusing on the perspectives of older women regarding their hearing, hearing problems, help-seeking, and interaction with audiology services. The next chapter undertakes a more thorough review of the literature pertinent to this research goal. 10 CHAPTER 2 LITERATURE REVIEW Overview The purpose of this chapter is to review and synthesize the literature important to the research topic by exploring two major substantive areas: hearing loss (particularly presbycusis) and help-seeking. Throughout the discussion in this chapter, the case is made for conceptualizing hearing loss, help-seeking, and practice and research in audiology within an ecological approach or framework. The chapter begins by defining what is meant by an ecological approach and giving an example of an influential ecological health model. The review of the literature on hearing loss provides a description of what is known about hearing loss, particularly in terms of its psychosocial impact on hard-of-hearing individuals and their significant others. This leads to discussion of the range of responses of individuals to the impact of hearing loss, and an examination of coping theory and the role of coping in adjustment to hearing loss. There is discussion of the influence of gender and aging on how hearing loss is experienced and handled. A review of theories and models pertinent to age-related communication completes this section. The discussion then turns to the central research topic of help-seeking, which is conceptualized as one form of coping. The literature on factors and issues that affect help-seeking is first reviewed. This literature falls into two categories: theories of health behaviour derived primarily from social and cognitive psychology, and theories drawing from epidemiological and health education studies of health services utilization. Studies of help-seeking and oral health among older adults are then considered for their ability to 11 shed light on factors that affect help-seeking for another condition that parallels presbycusis in many of its characteristics.1 Having moved from the broad to the specific in the exploration of studies on help-seeking, investigations of help-seeking specific to hearing loss are then considered. The final area of discussion is the role that the system of health service delivery and the interaction between clinician and patient2 may play to promote or delay the help-seeking process. This chapter aims to weave together the strands of knowledge in all these areas to form a composite picture of what is known, and what remains unknown, about help-seeking for hearing loss. With this canvas as a backdrop, the chapter closes with a review/revisiting of the research questions. An Ecological Framework "Ecology" refers to the study of natural systems, emphasizing the interdependence of one element in a system upon every other element (Lawton & Nahemow, 1973). The word ecology derives from the Greek word "oikos", meaning house or habitation. Although people are more familiar with the term ecology in relation to its biological roots in the study of plant and animal systems, the features of ecological systems can be applied as well to complex human systems. Each individual, organism, or element in an ecological system (eco-system), has an optimal or preferred environment where its living conditions are in balance, that is, are most suitable and sustainable.3 Environment encompasses all external forces to which organisms are actually or potentially responsive (Stokols, 1977). The eco-system, and 11 explored much of this literature on theories and models of help-seeking, and on help-seeking for oral health only after the analysis of my research data was complete. Thus, the review of the literature served not as an influence in my analysis, but as a reference base to which I compared the analysis of my findings. This approach is in keeping with grounded theory, which aims for discovery through the data, without too much bias introduced by total immersion in the literature at the outset of data collection. 2 The term 'patient' is preferred over 'client': the word 'patient' derives from the Latin wordpatiens, meaning "to suffer", and from the Latin words paene, meaning "almost", and penuria, "need". People seek help because they have need, are suffering, and wish to move toward wholeness (Miller & Crabtree, 1994). 3 What Borg (1998) refers to as one's "preferendum". 12 individual elements that constitute it, are subjected to disturbances that tend to force it away from this balance. Balance may be regained by either altering the environment to the preferences of the individual, or the individual may adapt or acclimatize to the new environment, or both may change to some degree. Behaviour is characterized as those responses made by individuals that either promote or impair their survival (or the restoration of balance) in the face of environmental fluctuations (Stokols, 1977). Ecological paradigms have developed in disciplines as diverse as sociology, education, human geography, economics, public health, and psychology (Green, Richard, & Potvin, 1996; Stokols, 1992). Within psychology, the analysis of environment-behaviour relationships has evolved from traditions such as behaviourism (e.g., Watson, 1913; Skinner, 1953). Many theories and models have contributed perspectives on the functional significance of the environment and its impact on behaviour, for example, conceptualizing this relationship as being mediated largely by cognitive processes (social learning theory) (e.g., Bandura, 1969), as focusing on the role of personality attributes (Murray, 1938), or, from the interactionist perspective, as emphasizing the importance of the subjective or perceived environment (dependent on both external cues and individualized cognitive strategies) as a determinant of behaviour (field theory) (Lewin, 1935). Ecological psychology (Barker, 1968) has focused on the collective processes by which groups adapt to the environment, while the field of environmental psychology (Proshansky, Ittelson, & Rivlin, 1970) places more emphasis on intrapersonal processes, such as perception, cognition, and learning. Investigations into the effects of such stressors as noise (Glass & Singer, 1974) and crowding (Stokols, Rail, Pinner, & Schopler, 1974) on the behaviour of individuals and groups are examples of topics in environmental psychology. The field of social ecology has subsumed these two fields, conceptualizing behaviour as a joint product of intrapersonal processes (physiological and psychological) and environmental dimensions at the social, physical, institutional, and cultural levels (Stokols, 1977; 1992). 13 Linking Concepts in Social Ecology with Core Assumptions in Health This kernal of the history of ecological concepts in psychology is provided to introduce an ecological framework of health and health promotion that links the concepts of social ecology to core assumptions about the dynamics of human health and strategies to promote health (StokoIs, 1992), both in terms of individual wellness and collective well-being (World Health Organization, 1984). These linkages are outlined below. First, in a social ecological paradigm, the well-being of participants in any situation is assumed to be influenced by multiple facets of both the physical environment (e.g., geography, architecture, and technology) and the social environment (e.g., culture, economics and politics). Moreover, the health status of individuals and groups is influenced by a variety of personal attributes, including genetic heritage, psychological dispositions, and behavioural patterns. Table 2.1, (from Stoko Is, 1992), lists several categories of personal and environmental factors that play either an etiologic or moderating role in human health. Second, analyses of health and health promotion should address the multidimensional and complex nature of human environments and individuals' relationships with them. For example, Stokols (1992) outlines how individuals' physical and emotional well-being.are enhanced by environments that are predictable and controllable by the person while at the same time being novel and challenging. Third, like environments, the participants therein can be studied at varying levels (ranging from individuals, small groups, and organizations, to populations), and with diverse methodologies (e.g., questionnaires, observations, interviews, epidemiologic analyses). Moreover, as Stokols states: "the social ecological perspective assumes that the effectiveness of health-promotion programs can be enhanced significantly through the coordination of individuals and groups acting at different levels, such as family members who make efforts to improve their health practices, corporate managers who shape organizational health policies, and public health officials who supervise community health services." (1992, p.7) 14 Table 2.1 Personal and Environmental Factors in Health and Illness (from Stokols, 1992, p. 13) Biopsychobehavioural factors Sociophysical environmental factors Biogenetic Psychological Behavioural Geographic Architectural / Technological Sociocultural Family history Sense of Dietary Climatic and Injury-resistant Socioeconomic of illness coherence regimens geologic risks architecture status (e.g., floods, Exposure to Psychological' Alcohol earthquakes, Nontoxic Social support vs infectious hardiness consumption hurricanes, construction social isolation pathogens draught) materials in (e.g.,viruses) Self-esteem Smoking buildings Social climate in Ground-water families and Immunologic Creativity Exercise contamination Ergonomic institutions competence patterns design of work Optimism Radon areas and other Modeling and Inoculation Sleep patterns contamination environmental conformity and Pessimistic of soil settings processes medication explanatory Safety history style practices Ultraviolet Environmental Cultural and (e.g., use of radiation aesthetics religious beliefs Congenital Health locus safety belts, and practices disability of control bicycle Atmospheric Indoor and helmets, safe ozone outdoor air Organizational Disabling Interpersonal sexual and depletion pollution (e.g., or political injuries skills prenatal "sick building instability behaviours) Global syndrome") Cardiovascular Extroversion warming Economic reactivity Participation Effective design changes (e..g, Coronary- in health Health of health care job loss) Chronological prone (Type promotion consequences facilities age A) orientation programs of reduced Health media & biodiversity Vehicular and communications Developmental Cancer-prone Compliance passenger safety stage (TypeC) with Restorative Health orientation prescribed potential of Noise pollution promotion Gender medical wilderness and programs in Depression/ regimens other natural Electromagnetic organizations anxiety environments radiation and communities Use of Hostility/ community Water quality Health-suspiciousness health and treatment promotive services systems legislation Health- Solid waste Environmentally relevant treatment and protective decisions and sanitation regulations actions made systems on behalf of Availability of others health insurance and community health services 15 One illustration of an ecological framework whose development was spearheaded by an influential coordinating body is the World Health Organization's (WHO) ICIDH-2, the International Classification of Functioning and Disability (ICIDH-2, 1999). Example of an Ecological Framework: The ICIDH-2 The WHO's International Classification of Functioning and Disability (ICIDH-2, 1999) has heightened recognition of the importance of an ecological approach to the study of health conditions and determinants. In 1980, the WHO introduced the International Classification of Impairments, Disabilities and Handicaps (ICIDH). This classification system defined "impairment" as a physical, physiological, or anatomical loss or abnormality of function, "disability" as the loss or reduction of normal ability resulting from impairment, and "handicap" as the detrimental effect that disability has on an individual's life, especially on the activities and roles he or she normally performs. The ICIDH-2 has recently recast this trio of concepts to encompass the range of function, not just dys-function, associated with health conditions. The impairment dimension has been expanded to include body functions, the physiological or psychological functions of body systems and structures, and the anatomical parts of the body, in addition to impairment. The disability dimension has been renamed the activity/activity limitations dimension. Activity is defined as the performance of a task or action by an individual; activity limitations are difficulties an individual may have in the performance of activities. Note that this dimension is defined at the level of the individual and is based on the actual performance of activities (rather than what a person could, or might, do). The handicap dimension has been recast as the participation/participation restrictions dimension. Participation refers to an individual's involvement in life situations in relation to her/his health conditions, body functions and structures, activities, and contextual factors. Participation restrictions are problems an individual may have in the manner or extent of involvement in life situations. These three dimensions are thus organized at the body, individual, and societal level. 16 In contrast to the ICIDH (1980) model, in which impairment resulted in disability, which in turn resulted in handicap, the ICIDH-2 model conceives of functioning and disability as a dynamic interaction between health conditions and contextual factors. A problem may exist at one level without necessarily involving the other levels; for example, an individual may have participation problems due to a stigmatizing attitude by society that persists even after the individual has recovered from the impairment or activity limitations associated with mental illness. Contextual factors represent the complete background of an individual's life and living, and consist of both personal and environmental factors. Personal factors include gender, age, other health conditions, coping styles, social background, education, and other conditions that influence how an impairment is experienced by an individual. Environmental factors range from physical factors such as climate and terrain to social attitudes, institutions, and laws. Contextual factors may pose barriers or act as facilitators to activities and participation. The dimensions of ICIDH-2 and their interaction are schematized in Fig.2.1. Health condition (disorder or disease) Activity & Structure Environmental Factors Personal Factors Figure 2.1 Model of the ICIDH-2 (ICIDH-2,1999. section 5.1) The ICIDH-2 reflects the WHO Constitution's broad conception of health as "a state of complete physical, mental, and social well-being and not merely the absence of disease or irmrmity" (ICIDH-2, 1999, section 4.3.8). The ICIDH-2 represents a biopsychosocial model that attempts to integrate the medical model of disability with the social model of disability. The medical model views disability as a personal problem requiring treatment that is aimed at a cure or the individual's adjustment and behaviour change. The social model of disability, on the other hand, views disability as a socially-created problem rather than an attribute of the individual, with the solution hinging on the collective responsibility of society at large to make the environmental modifications necessary for the full participation of people with disabilities in all areas of social life. In summary, the ICIDH-2 acknowledges that functioning happens at the level of the "whole person" and is the product of an interaction between one's physical or mental condition and the social and physical environment. It also emphasizes the crucial role that context plays in influencing an individual's activities and participation in everyday life. In summary, at the hub of an ecological approach to the study and promotion of health and healthy social systems is an appreciation of the dynamic interdependence between organism and environment. This fundamental principle demands that health and health conditions not be compartmentalized: the health of an individual (or community) must be considered in the context of the environment in which it lives and with which it interacts. With the foregoing discussion in mind, let us now consider the health condition of hearing loss within an ecological framework that moves beyond impairment to consider psychosocial impact on activities and participation in everyday life. 18 Hearing Loss The Psychosocial Impact of Hearing Loss Chapter 1 outlined the characteristics and consequences of the impairment of presbycusis and highlighted the growing importance of presbycusis as a population health concern. The study of the impairment of presbycusis is more straightforward than the study of its psychosocial impact. This is because the multidimensional effects of hearing loss span many facets of everyday living and may be modulated by many factors unique to each individual. Let us now turn our attention to the non-auditory consequences of hearing loss in daily life, the psychological and social manifestations of reduced hearing ability. In ICIDH-2 terms, we will now consider the impact of the impairment of hearing loss on the activities and participation of the hard-of-hearing person at the level of the individual. Following this will be an exploration of the effects of hearing loss on family relationships, stigma and the attitudes of others toward hard-of-hearing persons, effects related to communication styles, and the effect of gender on how hearing loss is experienced. The section concludes by reviewing how the individual responds in terms of coping and identity. Effects at the Level of the Individual Forty years ago, Ramsdell (1960) made an important contribution to the psychology of hearing and hearing loss when he proposed that sound has at least three levels of meaning. First, there is a social level at which sound is utilized in verbal language to convey meaning and relate to others. Second, there is a signal or warning level, at which an outside stimulus such as a siren relays the need for action or response. The third level of meaning is the most fundamental level of awareness, a so-called "primitive" level at which sound endows the human being with a sense of the "aliveness" of the world around him/her. Today, Ramsdell's three levels of sound meaning are encompassed in the term "soundscape". A soundscape is "an environment of sound with emphasis on the way it is 19 perceived and understood by the individual, or by society. It thus depends on the relationship between the individual and any such environment." (Truax, 1978, p. 126). To some degree, the extent to which Ramsdell's levels of meaning, or one's soundscape as described by Truax, is disrupted by hearing loss depends on the severity of one's impairment. A person with a profound hearing loss, for example, can be expected to experience disruption at all three levels of sound meaning, or to have an extremely limited soundscape. Beyond such gross distinctions in the function of sound, however, an individual's audiogram, as one measure of impairment, cannot predict the psychosocial effects of hearing loss (e.g., Erdman & Demorest, 1998b). The impact of hearing loss on the meanings of sound depends to a great extent on the individual's life context.4 The psychosocial effects associated with presbycusis are well documented, but their pattern of expression is highly individualistic. Four reasons why their effects may be highly individualistic are elaborated as follows. First, as mentioned in Chapter 1, presbycusis is characterized by insidious onset and progression of hearing loss in both ears. Ongoing adaptation to tiny increments in hearing loss may prevent recognition of the problem, especially as there is usually no comparative loss between ears to notice. As a result, presbycusis is a greater challenge to study than acute health problems (such as a heart attack) which are characterized by better-defined symptoms of more dramatic onset (such as severe pain). Presbycusis parallels other chronic conditions associated with aging, such as arthritis, in that symptoms may be easily incorporated into some general framework explained away by the fact that one is "just getting old" (e.g., the aches and pains of arthritis) (Zola, 1986). Second, because presbycusis by definition is tied to the aging process, the psychosocial effects of presbycusis are influenced by a myriad of factors associated with aging and life experience. These include physiological changes, such as those incurred as a result of cumulative exposure to hanriful noise levels, and psychological changes, such as may accompany altered occupational and family roles. 4 This notion of the soundscape introduces the concept of acoustic, or auditory, ecology, which refers to the interaction of listeners with their sound environment. Acoustic ecology includes the range and type of acoustic environments with which the listener/communicator interacts, with an emphasis on relationships one has with sound, and on the basis of sound, including the meaning and importance of soundscape components to the listener (Truax, 1978). 20 Third, because, presbycusis, like other forms of hearing loss, exerts its main effects on social interaction through communication, the effects of hearing loss are not restricted to the person with the impairment but affect and are affected by communication partners. Finally, related to the above three points, it is important to remember that the processes involved in aging and presbycusis are dynamic processes that evolve over time. In summary, one's experience of presbycusis is shaped by many influences over many years, leading to a complex but subtle and highly variable interplay of factors that determine the overall psychosocial impact of hearing loss. Hearing loss leads to adverse effects on the physical, cognitive, emotional, behavioural, and social function of older adults (see Mulrow et al., 1990, for a comprehensive list of studies). Hearing loss has been shown to lead to increased stress levels, anxiety, and social withdrawal. Altered self-concepts manifested in lowered self-esteem, feelings of inferiority, insecurity, and loss of autonomy may ensue (Rousey, 1976). In a study by Thomas and Gilhome Herbst (1980a), several factors were identified that significantly differentiated hearing-impaired from normal-hearing older adults. Those with hearing loss rated their health to be poorer, were less likely to be able to get out without help, ventured out less far, were less satisfied with their extent of mobility, reported having fewer friends than in the past, reported less enjoyment from life than previously, and were more likely to be rated as depressed. UMrnann, Larsen, Rees, Koepsell, and Duckert (1989) discovered a significant association between hearing loss and cognitive dysfunction in patients with Alzheimer's disease. Careful not to presume a cause and effect relationship between hearing loss and Alzheimer's, Uhlmann and colleagues concluded that hearing impairment may be an important risk factor for presenting Alzheimer-like conditions in terms of cognitive dysfunction. Indeed, some apparent cognitive deficits have now been shown to arise from perceptual deficits in vision and hearing (Schneider & Pichora-Fuller, 2000). Such findings highlight the potential wide-ranging secondary effects of hearing loss in terms of functional health status, in addition to the more intangible effects of hearing loss on quality of life. 21 Effects on Family Relationships Regardless of the age of the hard-of-hearing person, the interactive nature of communication means that the detrimental effects of hearing loss are not restricted to the hard-of-hearing person but extend to her/his communication partners. Hetu, Jones, and Getty (1993) provide a comprehensive discussion of this topic, by reviewing the limited number of studies of the impact of acquired hearing loss on the relationships of middle-aged and older adults. Intimate relationships, such as those with a spouse or other family member within the same household, are very vulnerable to the effects of hearing impairment. An unimpaired partner may share parallel handicaps with a hard-of-hearing kin in terms of stress, isolation, negative self-image, and difficulties in family interaction. Within the home, family members of hard-of-hearing persons report problems associated with communication, such as irritation at having to repeat comments or having to tolerate the loud volume of the television. Extra effort is involved where the uriimpaired family member must assume added responsibilities such as always answering the telephone. Feelings of isolation and negative self-image arise from reduced communication, especially intimate conversation, among all family members. Interactions between family members are also strained outside the home. There is the effort of and irritation at having to act as interpreter for the hard-of hearing person. A spouse may harbour resentment arising from restrictions imposed on a couple's social life where the hard-of-hearing spouse avoids social encounters such as parties or leaves early due to fatigue. A normal-hearing family member's own self-image may be undermined by a spouse or parent's inappropriate communication behaviours, such as speaking in too loud a voice or interrupting conversation to talk on an unrelated topic. Hetu et al. (1993) point out that with progressive acquired hearing loss, coping with and adjusting to hearing difficulties is a process that family members are engaged in with the hearing impaired spouse or parent even before a clear awareness of the impairment and its consequences emerges. However, while the problem of hearing loss is shared among family members, usually the potential solutions are not. The adjustments that family members make are commonly not openly discussed, negotiated or 22 acknowledged (Hetu, Lalonde, & Getty, 1987). As a result, efforts to improve communication often do not solve problems but may actually lead to greater feelings of misunderstanding and isolation within the family, manifested in part by both reduced frequency and reduced content of communicative interaction. For example, the hearing-impaired person may raise the television volume, and the irritated partner may tolerate it or ask repeatedly for it to be turned down. One or the other person may end up abandoning the listening activity because a satisfactory solution for both parties is never found.5 This example highlights how coping styles and strategies are dependent on the effects due to the nature of the interpersonal interactions among communication partners, due to factors related to their social roles, and due to their problems concerning shared space. Concepts from symbolic interaction theory (discussed in more detail in the next chapter) are useful in understanding the complex relationships among factors related to the coping styles and strategies of hard-of-hearing people. According to this theory, persons with hearing loss, like everyone else, are actors who take on different roles in different situations. From interaction to interaction, each person carries a definition of self that is shaped by social and personal experience, as well as the responses of others. Each actor in a social setting creates a "script" in conjunction with other actors, aiming to selectively reveal what each believes is most desireable from one's definition of self (Goflman, 1959). If this includes concealment of one's hearing loss in work and social settings, home may be viewed as the "backstage" setting where one may unwind and suspend pretense. It is possible that a controlling, managing role may be adopted with significant others at home (hence, turning the T V volume up), whereas a passive, avoidance role is assumed in the "outside world". These coping strategies impose changing roles for significant others as well. Outside the home, for example, a family member may be intermediary, protector or interpreter, whereas inside the home he or she may choose to withdraw from interaction with the hard-of-hearing relative. Hallberg and Barrenas (1993) defined four different television is cited as the most common problem area in several community-based surveys concerning hearing (e.g., Barcham & Stephens, 1980; Stephens, Lewis, Charny, Farrow & Francis, 1990b ), including studies of other cultures (e.g., in Poland; Golabek, Nowakowska, Siwiec, & Stephens, 1989). 23 approaches of wives toward spouses with occupationally induced hearing loss: pretending there is no problem (co-acting), playing down the problem (minimizing), controlling, steering, or advising the hard-of-hearing person (mediating), and separating themselves from the spouse (distancing). One cannot easily generalize about the specific effects of hearing loss on intimate relationships; every relationship is unique and dynamic and the effects that emerge are a function of many variables. In addition to the severity of the impairment, it seems likely that specific effects of hearing loss would depend on the length of time family members have shared the hearing problem and their relationship. Other factors that may play a role include the person's ability to adjust, as well as his/her age and lifestyle. The review by Hetu et al. (1993) encompassed studies that looked at families of workers with noise-induced hearing loss as well as retired couples, where, as mentioned earlier, additional factors related to aging seemed to combine with the effects of occupational hearing loss. These factors and their effects are highly individualistic; nevertheless their importance is evident. Attitudes of Others: Stigma One's broader social sphere of interaction includes contact with friends, acquaintances, and strangers in a work or community setting. The attitudes of these people and of members of society in general toward hard-of-hearing persons strongly contribute to the psychosocial disadvantages of hearing loss. Much of the work in this area has been undertaken by looking at the attitudes of coworkers of males with occupationally induced hearing loss (Hetu, Getty, Beaudry, & Philibert, 1994; Hetu, Getty, & Waridel, 1994). By far, the strongest response of those with unimpaired hearing is the stigmatization of the hard-of-hearing person (see Hetu, 1996 for a comprehensive discussion of stigma and hearing loss). Stigma refers to the "discrediting" of an individual as a result of her/his inability or failure to meet the expectations associated with a role in a specified social setting (Goffman, 1963). The spoiled sense of identity that accompanies 24 stigma is acutely felt by those with disabilities because inappropriate stereotyped responses from others are commonplace (McKellin, 1994; Pettygrove, 1985). Anticipating such stereotypes (e.g., equating hearing loss with impaired cognitive skills), leads to a range of responses by hard-of-hearing persons in their struggle to preserve and present an unstigmatized definition of self. One response may be to conceal one's hearing loss, which has the unfortunate effect of feeding the stigmatization process when coworkers, for example, make disparaging remarks or jokes about hearing loss in the presence of the hard-of-hearing person (Hetu et al., 1994). Communication in Aging: Additional Stigmatizing Effects Older adults, whether or not they have significant hearing loss, may also be stigmatized by the communication style others use with them. Research on communication, especially in inter-group encounters, has shown that people accommodate their manner of talking to each other in a variety of ways. This is the essence of Communication Accommodation Theory (Giles, 1984), which has grown into an mterdisciplinary model of the processes that determine how relationships and identity influence and are influenced by communicative interaction. On the one hand, to the extent that a person wishes to gain approval and identify with an interlocutor, he or she will converge his or her speech style to the other person, adjusting it so that it is more similar. On the other hand, speech divergence can communicate in-group identification, dissociation, or power differential. Convergence is the typical pattern. However, in talk with elderly persons, convergence may not be toward the older person's actual competence but toward a stereotypical societal view of old age as a time of dependence and sharply declining abilities. Health care providers, for example, especially physicians, often use denigrating or patronizing speech styles in talking to elderly patients (Ryan & Butler, 1996). Intergenerational communication studies also reveal negative stereotypes attached to older persons by their younger interlocutors and how these stereotypes alter the way that younger conversation partners interact with older adults (Coupland, Coupland & Giles, 1991). 25 Ryan, Giles, Bartolucci and Henwood (1986) have coined the term "communication predicament of aging", which refers to situations in which undesirable discrepancies between the actual communicative competence of an elderly person and the negative perception of his or her competence result in inappropriate communication accommodation. Inappropriate accommodation is noted particularly in the language that may be used by caregivers or family members. Inappropriate communication accommodation is epitomized by simplistic "baby talk" characterized by such features as high pitch, exaggerated intonation, and reduced grammatical complexity. Such inappropriate accommodation reflects lowered expectations of communicative competence and conveys a fundamental lack of respect for the older adult, such as is exhibited by patronizing or controlling talk that leads to constrained opportunities for the elder to communicate. An extreme example is the situation in which an older person's problems are discussed with a third party as if the elder were not present. In response, an older person's self-perception may erode to match these negative expectations and judgments, rendering them self-folfilling. Further support for the role of communication in the social construction of self (another example of symbolic interactionism) comes from a social learning framework provided by Baltes and Wahl (1996). They observed and coded behavioural interaction patterns between older people and their social partners in everyday activities, in both long-term care institutions and private dwellings. Baltes's model of learned dependency (1988, 1995) sparked interest in exploring the possible consequences of these exchanges on how older people maintain and develop dependent and independent behaviours. Findings showed that the dominant interaction pattern was one in which the dependent behaviours of older people were immediately attended to and given positive reinforcement, whereas independent behaviours were ignored or discouraged, a pattern that Baltes and Wahl labeled the "dependency-support script" coupled with an "independence-ignore script". This pattern was particularly evident when the older person was engaged in personal care. For example, where a resident tried to put on his shirt (an independent act), a staff member said, "I told you not to dress yourself; you always get it wrong" (a dependence-supportive act). The pattern was apparently replicable in males and females. In response, 26 older persons generally complied with the expectations of their social partners, reinforcing the pattern of interaction. Similar to Ryan's "communication predicament of aging" model, a downward spiral may ensue, with the older person becoming increasingly less interactive and more dependent upon caregivers. This reduction in communicative interaction threatens the health of the older adult by threatening his/her autonomy. As well, emotional health, which is important to overall health status, is a function of one's social support as fostered by good communication (Nussbaum, Thompson & Robinson, 1989). The role of communication is not restricted to sharing information or conversational content between parties. This ideational, prepositional, or transactional role of communication is important, but communication also serves an equally important relational role (Brown & Yule, 1983). The relational or interactional role is the social role of communication that enables us to establish and maintain relationships with others and that contributes to our sense of identity. Compared to younger adults in the workforce, older persons in their retirement years may attach relatively more importance to the interactional, relational, or social function of communication than to the transactional, ideational information exchange function (e.g., Johnson & Pichora-Fuller, 1994). With increasing age, as one becomes more acutely aware that one's time spent with others is limited, one's emotional needs in conversation take greater precedence (Carstenson, 1996). Communication, therefore, serves an important role in promoting a sense of autonomy and connectedness well into old age. In addition to the contribution that factors associated with aging may make to the psychosocial impact of hearing loss, there is evidence of differential effects of hearing loss on women compared with men. As the current study involves older women exclusively, let us now consider the literature on gender and hearing loss. 27 Women versus Men Research shows that hearing impairment6 and its effects on activities and participation impact women and men differently. Waridel (1995) studied working women who had occupationally-induced hearing loss. These women reported problems on the job related to their hearing, such as excessive fatigue, stigma reinforced by peers and boss, and a feeling of having to prove one's ability to a greater extent to make up for having a hearing loss. In addition, these women reported feeling very disadvantaged on the home front. At the social 'hub' of the family, nurturing children (and husband) and often acting as spokesperson to the outside world, these women felt they had a more demanding social role than their spouses. Their communication obligations meant that they could not rest from the stress of communicating after a day's work, as their spouses could. A difference in social roles between men and women results in different psychosocial disadvantages associated with hearing loss. Hetu et al. (1993) report that men's traditional role as "breadwinner" leads to concealment of their hearing impairment on the job out of fear of giving an image of being "diminished" or "less manly". Women on the other hand report greater feelings of inadequacy at home than on the job because of the importance they place on their role as 'communicator' to and for their family. Women's different social roles also encompass multiple roles in society. Much more than men, women simultaneously play the role of parent, spouse, caregiver to a relative or relatives, community member, and employee. Studies of women with multiple roles suggest that they gain both socially and emotionally from complex role responsibility (Baruch, Biener, & Barneff, 1987). To my knowledge, however, no studies have examined how such responsibility may be affected by the stress of communication difficulty arising from hearing loss. 6 In terms of impairment, women, on average, have poorer low-frequency thresholds (up to 1000 Hz), but better high-frequency thresholds than men (Pearson et al., 1995). This "gender reversal" phenomenon is noted across the life span in studies with older participants (age 50 to 80+) (Jerger, Chmiel, Stach, & Spretnjak, 1993) and in studies with young to old participants (Erdman & Demorest, 1998b). Jerger et al. (1993) suggest that gender differences may be due to variance in noise exposure, the possibility of greater atrophy of the stria vascularis in women, and the possible relationship between cardiovascular disease and hearing loss, particularly among older women. 28 The two genders also seem to have different coping mechanisms to deal with the effects of hearing loss. There is evidence from population data that women self-rate hearing disabilities as slightly less of a problem than men (e.g., Davis, 1995). Another study indicated that working women with noise-induced hearing loss are more likely than men to try to pass as "normal" hearing (Hallberg & Jansson, 1996). Men, much more than women, impose their needs strongly on the family and come to expect family members to accommodate to their hearing loss (Jones, Kyle & Wood, 1987). Hence, although interpersonal communication is of great importance to women (Garstecki & Erler, 1995; Erdman & Demorest, 1998b), women receive (and may expect?) less social support and understanding on the homefront for their communicative difficulties compared with men (Jones et al., 1987; Waridel, 1995). Women and men also appear to differ in the words they use to describe the effects of hearing loss in their everyday lives. For example, in Waridel's (1995) study, women spoke of the "pleasure lost" in activities once enjoyed; men in similar qualitative studies never used such a phrase (Getty & Hetu, 1994). These researchers expressed the need for further exploration of gender differences in the psychosocial effects of hearing loss, and a need to relate findings to the broader literature on gender studies and gender and disabilities (Hetu et al., 1993; Garstecki & Erler, 1995; Erdman & Demorest, 1998b). In summary, the literature on the psychosocial effects of hearing loss provides evidence that the impact of hearing loss is wide-ranging, affecting not only the hard-of-hearing person, but others with whom she/he interacts on both an intimate and casual basis. At the societal level, negative attitudes and behaviours of others (whether they be coworkers, family members or care-givers) toward hard-of-hearing persons may diminish self-esteem and negatively influence self-perception. Such effects are exacerbated for older adults with presbycusis, due to stereotypic negative attitudes toward the aged. Social partners play a critical role in mediating between the experience of hearing loss and the behavioural coping responses of hard-of-hearing persons. It is to these responses that we now turn our attention. 29 The Individual's Responses to Hearing Loss It is well acknowledged that psychological factors mediate between the hearing impairment and its effects on activity and participation. It is therefore important to explore the psychological construct of coping and its role in the individual's responses to hearing loss. The above discussion has already made reference to the coping behaviours that hard-of-hearing persons and significant others engage in when responding to the impact of hearing loss in their lives. The construct of coping is discussed further below. Lazarus and Folkman's well-known theory of stress and coping is introduced, and two theories that relate the concept of control with stress and coping (the theory of attentional overload, and the theory of learned helplessness) are also mentioned. This section concludes with a discussion of the role of coping in the preservation of identity and self-image. The Construct of Coping and Related Theories Coping theory and research posits that coping processes intervene between stressful events and psychological adaptation or adjustment (Folkman, Lazarus, Gruen, & DeLongis, 1986). Coping has been defined as "cognitive and behavioural efforts to manage (reduce, minimize, master, or tolerate) demands (external or internal) that are appraised as taxing or exceeding the resources of the person" (Lazarus & Folkman, 1984). It is a dynamic process of purposeful efforts focused on the resolution of difficulties that place demands on the organism for adjustment (Zautra & Wrabetz, 1991). Coping has been conceptualized as being either a "trait" that refers to stable properties of a person, or as a "state" that deals with transient reactions that change with circumstances. One may then differentiate between "coping styles" and "coping strategies" to mark this trait-state distinction. A well-known, ecological7 model of stress and coping based on the cognitive theory developed by Lazarus and Folkman (1984) views the person and the environment as being in a dynamic, mutually reciprocal, bidirectional relationship. "Stress" is 7 The term "ecological" is not explicitly used by Lazarus and Folkman (1984). 30 conceptualized as a relationship between the person and the environment that requires constant appraisal. Transactions that are appraised as stressful because they are perceived to pose a threat, challenge or harm to the individual require a coping response. Coping is seen to serve one of two functions: problem-focused coping works to deal with the problem causing distress; emotion-focused coping works to regulate the distress. Problem-focused coping is goal-directed and includes strategies for gathering information, decision-making, plarining, and resolving conflicts in order to solve or manage problems that impede or block goals and create distress (Lazarus & Folkman, 1984). Emotion-focused forms of coping include distancing and escape-avoidance responses (Folkman et al., 1986). Studies have shown that people use both forms of coping in virtually every type of stressful encounter (Folkman et al., 1986). Moreover, the coping behaviours used are determined by characteristics of both person and environment (Felton, Revenson, & Hinrichsen, 1984). Coping processes lead to an event outcome that either is a favourable resolution, an unfavourable resolution, or no resolution. Particular emotions are generated throughout the processes of appraisal, coping, and the evaluation of event outcomes. A chronic disability imposes repeated, sometimes unremitting, demands for coping (Lazarus & Folkman, 1984). Distress continues wherever an unfavourable or unresolved event outcome follows a coping response. Heim (1995) points out that, since an individual's principal aim in coping is to adapt to the new situation resulting from disease (illness), one cannot not cope with illness demands. In Heim's (1995) opinion, "good" coping involves taking an active stand, including tackling issues, seeking information, solving problems, and seeking social support with optimism and confidence; "bad" coping, however, implies a passive ruminating, withdrawn attitude, often with self-accusation or resignation. Heim does point out, however, that the distinction between "good" and "bad" coping depends on the perspective of the observer: what may seem inappropriate to a family member or health professional may seem appropriate to the "coper". Folkman et al. (1986) caution that it is important not to value a particular form of coping without reference to the context in which it is used. Pettygrove (1985), in relating her personal experience with glossectomy, states that "what appears to be noncompliance or rebellion against treatment may in fact 31 be an attempt to assert more self-control or to avoid some task that would lower self-esteem" (p. 109). In addition, because coping is a process, one's approach to coping is likely to change over time as the demands of the illness and one's reaction to those demands change. To explore the strategies utilized by hard-of-hearing persons in their adjustment to hearing loss, audiological research has been directed predominantly to the development of scales or self-assessment instruments (discussed more in a later section) (e.g., Andersson, Melin, Lindberg, & Scott, 1995; Demorest & Erdman, 1987). With a few exceptions, however, (e.g., recent research by Stephens, Jaworski, Lewis & Asian, 1999 on specific communication strategies adopted in specific listening environments) these pencil-and-paper tools have not considered the role that either contextual or underlying psychological factors may play in the individual's choice of particular coping strategies. In contrast, qualitative research by Hallberg & Carlsson (1991) has looked specifically at the psychological construct of coping in relation to hearing loss. They interviewed middle-aged employed men and women with long-standing hearing loss, some of whom wore hearing aids. They found two major themes related to coping among these participants; persons with acquired hearing loss either opted to "control the social scene" or "avoid the social scene". They reported that each participant used a variety of strategies, but showed a preference for one of these two management patterns. "Controlling the social scene" is a pattern characterized by the hard-of-hearing individual actively managing situations and the listening environment and taking responsibility for outcomes. Strategies may include verbal and nonverbal instructions to others to enhance understanding. The need to control the social scene is viewed by Hallberg and Carlsson as an activity comparable with Folkman and Lazarus's (1984) problem-focused coping. "Avoiding the social scene", a pattern adopted less often by women than men in their study, includes the strategy of minimizing the hearing loss, for example, by joking about difficulties or by making positive comparisons between the self and others having worse hearing. Nonverbal communication strategies, such as lip-reading and positioning oneself near the talker, may also be utilized to minimize the attention drawn to one's hearing loss. Hallberg and Carlsson (1991) view avoidance of the social scene as partly comparable to 32 the emotion-focused coping of Folkman and Lazarus (1980). They concluded that both management patterns have a mix of positive and negative personal and psychosocial outcomes. A negative psychosocial outcome may result, for example, when one finds oneself in an uncontrollable situation. As Folkman et al. (1986) state, "people who are repeatedly in uncontrollable situations experience helplessness, become increasingly passive in their coping efforts, and ultimately experience demoralization and depression" (p. 571). The relationship between perceived control and response to stressors has been the focus of several theories. Two such theories are the theory of attentional overload and the theory of learned helplessness. The theory of attentional overload (Cohen, 1978) posits that uncontrollable or unpredictable stimuli require more extensive monitoring (because of their novelty or complexity) than controllable events. Uncontrollable stimuli are, therefore, more likely to deplete attentional resources and to result in impaired task performance and interpersonal relations. The quotation above (from Folkman et a l , 1986, p.571) relates to Seligman's (1975) theory of learned helplessness that states that, through repeated exposure to uncontrollable events, individuals reduce their attempts to influence the environment because they come to believe that outcomes are independent of their behaviour. This theory has ties to the construct of locus of control (discussed in a later section). Nevertheless, researchers have paid increasing attention to the link between our well-being and our control over our environment (Stokols, 1985).8 Stokols (1985) emphasizes the importance of considering the sociophysical context of human stress. As he outlines, the degree of "fit" between person and environment (also termed congruence) in a particular setting depends on perceived controllability and environmental salience. Controllability concerns one's perceived ability to modify or maintain the environment, or regulate one's exposure to it, in accordance with personal preference and well-being. Quality of life indicators in aging (Raphael, Renwick, & Brown, 2000) include the element of environmental choice, as well as control, in the definition of a "quality environment". Choice and control within specific environments contribute to quality of life by enhancing a person's perceptions concerning her/his decision-making abilities and opportunities (see Pichora-Fuller & Robertson, 1994 for an example of this among hard-of-hearing residents in a long-term care facility). 33 Environmental salience comprises both motivational salience, which concerns the extent to which one associates settings with psychologically important needs, and perceptual salience, which refers to the degree to which features of the sociophysical environment support or hinder achieving one's goals in that setting. Stress, therefore, and one's response to it, is multidimensional and changes from situation to situation and across time. For example, a hard-of-hearing person's choice of a particular coping strategy to deal with misunderstood conversation at a party is influenced by her recollection of the degree of success or failure experienced previously when that strategy was used in previous comparable situations. The compensatory coping strategies utilized by older hard-of-hearing persons in conversation change as environmental context changes, and as personal resources change, such as with the decline of hearing and cognitive resources in aging. The meaning of a message in conversation results from the overlap between the content dimension and the relational dimension of communication. In addition to linguistic cues, the relational dimension relies on nonverbal paralinguistic and prosodic cues. These cues include such features as tone of voice, pitch, stress, rhythm, volume, intonation, and rate of speech. These features support and supplement the content dimension of conversation, by relaying information such as the emotional state of a speaker and relaying cues that regulate turn-taking. Villaume, Brown and Darling (1994) suggest that presbycusis involves auditory losses along these two dimensions, content and relational, and suggest that these two dimensions of presbycusis exhibit different patterns of aging. Young-old adults may have fairly severe losses on the content dimension of listening (e.g., reduced ability to identify phonemes), but little loss on the relational dimension (e.g., good ability to use prosody). In contrast, old-old participants (late 70's and 80's) exhibit major losses on both the content and relational dimensions. In terms of coping strategies, young-old adults pay greater attention to paralinguistic cues in conversation to compensate for the content cues they miss. However, as their ability to use paralinguistic cues deteriorates with age, adjusting to a communication partner may become an uncertain process. Such increasingly uncontrollable situations may lead to a shift in conversational style that becomes exaggerated as either more passive or more dornineering. Indeed, old-old adults, when 34 compared to young-old adults in Villaume et al.'s study, exhibited a much more restricted style of conversation. Co-occurring health conditions, such as visual problems (Erber, 1996), exacerbate these effects. Studies such as these highlight how the availability or lack of personal resources in many realms (both physical and psychological) may determine the choice of coping strategies utilized at any one point in time, as well as their evolution over the lifespan. Given the importance of communication in older adulthood and the disadvantageous results of maladaptive behaviours, it is important that older adults utilize effective coping styles and strategies to deal with hearing loss in ways that are consistent with their goals, and that enhance their health and quality of life. Issues of Identity and Self-image There is a common theme that runs through the literature on coping: the preservation or continuity of identity and self-image. With regard to the goal of adapting or adjusting to illness or disability, coping cushions the emotional impact of illness or disability and preserves identity. This point has been raised in terms of family relationships and the stigma attached to hearing loss in work or social settings. Hallberg & Jansson (1996) note that: denial (of hearing loss) or self-deception is interpreted as an attempt to protect the normal identity and to avoid being defined as a deviant in social interactions... Denial, or self-deception, could also be seen as a coping strategy, aimed at concealing the hearing disability in order to preserve a positive self image, (pp. 340-341) It has been seen how hearing loss may threaten one's self-image and identity at every level of interaction, from the personal to the public to the societal. Given the enormous energy many hard-of-hearing persons devote to protecting their "normal" identity, it is curious that issues of identity and self-image have not assumed greater importance in the literature on acquired hearing loss (for exceptions see Getty & Hetu, 1994; Laszlo, 1994; McKellin, 1994). Fortunately, a review of more recent research suggests that this is changing (e.g., Stephens et al., 1999). Unlike the Deaf, with their 35 strong sense of group identity or culture,9 there is no recognized homogeneity of experience among hard-of-hearing persons to unite them. Their social experience is diverse, and, although their impairments may be similar, there is not necessarily any outward sign of hearing loss to signal group membership. The notion of shared meaning that is a defining feature of culture is largely lacking, except as it emerges in rehabilitation.10 We will return to this important point in the discussion of help-seeking. Summary of the Literature on Hearing Loss The literature reviewed above has illustrated the wide-ranging and highly individualistic impact of hearing loss on the hard-of-hearing person and her/his family members. The stigmatizing attitudes of others toward the hard-of-hearing person may compound the psychosocial disadvantages of hearing loss, particularly as these attitudes are expressed through communication, especially with older adults. There is evidence that hearing loss affects women and men in different ways, and the two genders appear to respond in different ways to these effects. The impact of hearing loss is mediated by one's responses to hearing loss as influenced by psychological factors, and the construct of coping and related theories can be used to explain the response of the individual to the stress of hearing loss. We have seen that coping really encompasses all the behaviours an organism engages in consciously or subconsciously in response to stress, with the goal of adjusting to that stress. We move now to consider help-seeking, which may be considered a subset of coping behaviours. 9Among many Deaf, a sense of group identity has been developed from an early age through the use of sign language, as well as attendance at residential schools. 1 0 The term rehabilitation is used in the broadest sense to include not only the services of health professionals but also other sources of help. As such, an important component of the rehabilitative process may involve participation in a self-help group or consumer association such as the Canadian Hard of Hearing Association (CHHA). 36 Help-seeking The ecological paradigm introduced at the outset of this chapter identified the link between behaviour and underlying personal psychological factors (see again Table 2.1). This link was underscored in our discussion of coping and theories related to coping and stress. In attempting to better understand the process that leads to the culminating point of booking a hearing test, it is helpful to extend this discussion by exarnining major psychological theories and conceptual frameworks developed to attempt to explain health behaviour as it might apply to help-seeking. A number of these theories and frameworks will be introduced in this section; however, with a few exceptions (Brink et a l , 1996; Hyde & Riko, 1994; McCormick et al., 1994; Noh, Gagne & Kaspar, 1994,) they have not been considered in the audiologic literature in general or in audiologic research on help-seeking in particular. We begin the discussion with a description of five major social-psychological and cognitive-psychological ideas drawn from the health education and health promotion literature. These are: the Health Belief Model, the concepts of self-efficacy and locus of control, the theory of reasoned action, and the theory of planned behaviour. The Health Belief Model was one of the first models of health behaviour to be developed; the other constructs outlined here have built upon the theories underlying this model. Following this description of personal psychological constructs, an outline of the Andersen-Newman model is presented. This important structural model of health service utilization incorporates personal and environmental factors that are posited to affect the use of health services. As will be highlighted, however, this model has been criticized for its focus on structure and its failure to consider aspects of process that underlie help-seeking and service utilization. Two models that focus on aspects of process in help-seeking: the construct of illness behaviour, and the Common Sense Model, are then presented. One final model presented explores the issue of delay in help-seeking (exploring stages of delay). The review of the theoretical literature on help-seeking concludes with a discussion of the effects of gender and aging on help-seeking. 37 Theories and Models of Health Behaviour Health Belief Model The Health Belief Model attempts to predict health-related behaviour in terms of certain belief patterns. It was developed in the 1950's to explain and predict why people engage or don't engage in specific preventive behaviours, such as participating in tuberculosis screening programs. The model stated that, in order for an individual to take action to avoid a disease, he/she would need to believe that 1) he/she was personally susceptible to that disease (perceived susceptibility), 2) the occurrence of the disease would have at least a moderately severe effect on some component of his/her life (perceived seriousness), and 3) taking a particular action would in fact be beneficial by reducing susceptibility to or the severity of the effects of the condition, and that it would not entail overcoming important social or psychological barriers (perceived benefits of and barriers to taking action, otherwise termed cost-benefit analysis). Furthermore, cues to action were stated to link action to perception. Hence, a "stimulus", either internal (e.g., perception of bodily states), or external (e.g., interpersonal interactions, mass media communications), was assumed to be necessary to trigger the specific health behaviour (Maiman & Becker, 1974). Self-efficacy Self-efficacy is a concept from social learning theory (Bandura, 1977) that has been incorporated into the Health Belief Model (Rosenstock, Strecher, & Becker, 1988). It refers to one's beliefs in one's abilities to mobilize resources both internally and externally to meet situational demands. Self-efficacy incorporates the concept of reciprocal determinism (as opposed to operant conditioning) in that it acknowledges a person's ability to act upon her/his environment as well as to be acted upon by the environment. It is a particularly salient construct because issues of perceptions of self-confidence and control are important in long-term modifications of behaviour, such as behaviours related 38 to communication and hearing. The kinds of coping strategies one adopts to compensate for hearing-related communication difficulties, for example, will depend in part on one's perceptions of self-efEcacy, including one's ability to control environmental factors contributing to communication difficulties. The degree to which professional recommendations regarding rehabilitation are followed may also be influenced by one's perception of one's ability to fulfill such recommendations. Locus of Control The concept of control is reflected in psychology's "health locus of control" construct, which refers to the extent to which a person believes her/his health is determined by forces internal or external to the self. Those who believe health is determined by forces such as chance, fate, luck, or powerful others are said to have an "external" locus of control, whereas those who feel that one's own efforts and abilities determine health are said to have an "internal" locus of control (Goldsteen, Counte, & Goldsteen, 1994). Locus of control is closely related to another construct, "mastery", which some view as central among the personal resources which people draw upon in the face of threats posed by events and objects in their environment (Turner, Frankish, & Phillips, 1991). Mastery is seen to be conditioned largely by one's past history of successes and failures in meeting such challenges. Theory of Reasoned Action This theory incorporates the concept of behavioural intention, which is felt to be the link between beliefs and behaviours. The theory of reasoned action holds that before an actual behaviour occurs, one must have the intention to carry out the behaviour (Fishbein & Ajzen, 1975). It also specifies that the strength of an individual's intention to carry out a specific behaviour is a function of personal attitudes as well as subjective norms. These attitudes, in turn, are influenced by beliefs concerning the efficacy of action in achieving expected outcomes and by the attitude toward those outcomes. 39 Theory of Planned Behaviour The theory of planned behaviour is an extension of the theory of reasoned action that emphasizes the importance of volitional control for predicting behaviour (Ajzen, 1985). It is recognized that actual control is an intangible measure; rather the perception of control is the available measure. Hence, perceived control over a behaviour should be high when an individual perceives few obstacles, internal and external, to performing the behaviour. Perceived control may be based on personal experience or observation of the experience of close other. Application of Theories to Help-seeking for Hearing Loss This short synopsis of theories of health behaviour introduces important psychological constructs that have found their way into models of health services utilization. They resonate with some of the findings of research on help-seeking, and in particular, as we shall see, behaviour related to help-seeking for hearing problems. For example, we have noted how presbycusis and its consequences may develop gradually. If an individual is not aware of these subtle changes, her perceptions of susceptibility and severity are likely to be low and help-seeking action is unlikely. If she is frequently in uncontrollable listening situations where her coping strategies are unsuccessful, she may give up trying to understand conversation and withdraw from social interaction (learned helplessness). Additionally, if she has an external locus of control, she may be less likely to seek help because she rationalizes that hearing loss and its effects are outside her control. Moreover, if the costs of taking action are perceived to outweigh the benefits of taking action, action will likely be delayed or avoided. As previously mentioned and discussed further in this chapter, research findings indicate that many costs accompany getting one's hearing tested. There are costs in terms of the stigma associated with revealing one's hearing loss and adopting "hard-of-hearing" as part of one's identity. There is the cost of wearing a hearing aid or using an assistive listening device, not only financial cost, but social cost in terms of the stigma attached to it and the altered "body image" it creates 40 (Helman, 1995). The knowledge and skills necessary for the successful use of such devices may seem overwhelming, especially to the elderly, thereby reducing their self-efficacy. The perceived minimal benefits of hearing aids are often based on personal attitudes developed in response to the poor modeling provided by unsuccessful users. Very often, cost exceeds benefit and no action is taken. One can see how these psychological theories resonate with the attitudes, values, belief patterns, and behaviours associated with presbycusis, in term of actions adopted on a day-to-day basis to deal with hearing challenges, and in terms of seeking help for longer-term solutions, such as considering a hearing aid. Now we turn to consider one model that has been invoked in numerous studies of help-seeking and health services utilization: the Andersen-Newman Model. A Model of Health Services Utilization: The Andersen-Newman Model The psychological constructs described above have found their way into models of health services utilization, underscoring the importance of these constructs in help-seeking and health services utilization. The Andersen-Newman model (Andersen & Newman, 1973) was developed over a number of years based on a series of national surveys in the United States of that population's use of and expenditures for medical services. This model is helpful in exploring the factors that determine help-seeking behaviour and the use of health services in general. The Andersen-Newman model views the utilization of health services as a type of individual behaviour, but it also delineates the influences of societal determinants and the health services system itself on individual determinants of utilization. The model characterizes health services utilization into type (e.g., use of hospital, physician, medication), purpose (primary, secondary, tertiary or custodial care), and unit of analysis (which distinguishes between, for example, initial contact with a physician versus number of services received in a given time period). The authors stress the importance of how health service utilization is measured because the components of the model and their relative influence vary considerably, depending on the service under analysis. For example, 41 factors that influence preventative (primary) care (e.g., annual checkups and immunizations) differ from those related to diagnosis and treatment. In terms of individual determinants, the model assumes that a sequence of conditions determines whether or not people use services and the volume of services they consume. Use is posited to depend upon: 1) the propensity of an individual to use services as predicted by various predisposing variables, 2) his/her ability to secure services (enabling variables), and 3) his/her illness level. Predisposing variables are individual characteristics that exist prior to illness. These characteristics include demographic factors such as age, sex, marital status, and past illness; social structural (or status) factors such as education, occupation, and ethnicity; and, beliefs about illness and health care, which encompass knowledge about disease, attitudes toward health services, and values concerning health and'illness. Enabling variables are conditions that permit one to act on a value or satisfy a need regarding a health service. These conditions may reside at the individual or family level in terms of financial resources, whether or not the individual has a regular source of care, and the nature and accessibility of that care. Enabling characteristics of the community that may affect service use include the ratio of health personnel and facilities to the population, the price of health services, and aspects related to geographic region and the rural-urban character of a community. The third level of determinants affecting an individual's use of health services is state of illness as self-perceived and also as evaluated by the practitioners who treat the individual. Measures of self-perceived illness include number of disability days experienced during which the individual is unable to do what he/she usually does (analogous to "activity limitation" in the ICIDH-2), and self-report of symptoms. The state of illness evaluated by practitioners is included because, once an individual seeks care by accessing the health care system, the nature and extent of the care received is in large part based on professional evaluation (assessment of symptoms and diagnoses) of the individual. In addition to the individual determinants of health service use, the model posits that there are two categories of societal determinants of health service utilization: 42 technology and norms.11 Societal norms reflect legislative components, such as how medical care is financed (which Andersen and Newman postulate is the norm which has the greatest effect on health service utilization), and societal beliefs and homogeneity of values. The health services system structures the provision of formal health care goods and services in society, and consists of two major dimensions: resources and organization. The resource dimension includes the total volume of labour and capital relative to the population served and the distribution of these resources within a country (or more regionally, within a state or province). Organization refers to access and structure within the health care system. Access specifies the requirements that must be met and the barriers that must be overcome before medical care is received and/or continued; for example, such factors as health care budgets and waiting time affect access. Structure refers to characteristics of the system that determine what happens to the patient following entry into the system; for example, the nature of medical practices of the primary practitioners who first see the patient.12 Of the variables cited in the Andersen-Newman model, research by many investigators has revealed that variables related to needs, estimated by indices of health status, are the most consistent and powerful predictors of utilization (Anderson & Newman, 1973; Wolinsky, 1978; Mutran & Ferraro, 1988; Strain, 1991) These indices of health status vary from study to study but usually consist of some combination of subjective (self-reported) and objective measures, including activity limitations, number of chronic conditions, presence of pain, degree of worry (perceived seriousness), and perceived health status (e.g., compared to one's peers). The Andersen-Newman model has been very influential in laying out determinants of help-seeking and health services utilization at both individual and environmental levels. 1 1 Andersen & Newman (1973, p. 100) define technology and norms as follows: "Technology" is defined as " a set of principles and techniques useful to bring about change toward desired ends." "Norms" represent "the spectrum of modes whereby social systems induce or insure normal compliance on the part of members." 1 2 Anderson & Newman point out that all these components are inter-related. They also acknowledge that many other factors may directly influence an individual's use of a health service besides the health services system. 43 For example, this model was influential in the development of the Precede-Proceed model of health promotion planning (Green & Kreuter, 1991).13 For all its influence, however, the Andersen-Newman model has come under criticism for equating self-assessed health status with need and for its focus on structure (e.g., laying out the components that influence utilization) without consideration of the processes underlying structure. The next section elaborates upon these concerns. Exploring Underlying Processes in Help-seeking Several researchers (e.g., Mechanic, 1979; Hansell, Sherman & Mechanic, 1991; Strain, 1991) have argued that measures of need for medical care are not "pure" measures of illness, and that answers to questions on health, for example in health surveys, reflect, in addition to morbidity, a complex pattern of illness perception and behaviour (Mechanic, 1979). The Andersen-Newman model has been viewed as a fixed, sequentially-ordered look at utilization that fails to tap the interactive, contextual, and developmental processes underlying the use of health services (Mechanic, 1979). Moreover, a focus only on volume of services utilized, as in the Andersen-Newman model, neglects all those who cope with health issues in other ways, through the use of informal means (e.g., self-care remedies), as well as formal services (Penning & Chappell, 1990; Strain, 1991). 13Within the field of health promotion, the Precede-Proceed Model of health promotion planning (Green & Kreuter, 1991) has emerged as one of the most comprehensive and adaptable ecological models for designing and implementing health promotion programs. "Precede" stands for "Predisposing, Reinforcing and Enabling Constructs in Educational Diagnosis and Evaluation." "Proceed" stands for "Policy, Regulatory, and Organizational Constructs in Educational and Environmental Development." As the name indicates, this model, like the Andersen-Newman model, places great emphasis on the constructs of predisposing and enabling factors, and, additionally, on reinforcing factors because, as the authors elucidate, whether we are trying to explain, predict, or change a behaviour, either at the individual or collective level, we must attend to the factors underlying that behaviour, which fall mostly into the psychological domain. Green & Kreuter (1991) define predisposing factors as those antecedents to behaviour that provide the rationale or motivation for the behaviour; as with the Andersen-Newman model, these consist of knowledge, attitudes, beliefs, values, and perceived abilities. Enabling factors are characteristics of the environment that enable a motivation to be realized. Reinforcing factors (not considered in the Andersen-Newman model) occur subsequent to a behaviour to provide the continuing reward or incentive for the behaviour. Green and Kreuter (1991) emphasize that no single behaviour or action is caused by just one factor and any plan to influence behaviour must consider all three sets of factors. 44 Two conceptual frameworks from medical sociology that attempt to address the elements of process in help-seeking are the construct of illness behaviour (Mechanic & Volkart, 1961), and the Common Sense Model (Prohaska, Leventhal, Leventhal, & Keller, 1985), which incorporates the construct of illness representation. The Construct of Illness Behaviour The study of illness behaviour was introduced (Mechanic & Volkart, 1961) to "depict the large variability in reactions to symptoms and illness and to identify the various sociocultural, environmental and psychological factors that affected such reactions." (Mechanic, 1995, p. 1208). This developed into the construct of illness behaviour, referring to: "the varying ways individuals respond to bodily indications, how they monitor internal states, define and interpret symptoms, make attributions, take remedial actions and utilize various sources of formal and informal care."(Mechanic, 1995, p. 1208). As Mechanic stated (1995, p. 1208), this construct therefore views illness "not only (as) a state of a person, but (as) a mode of coping." Work in pain research and chronic illness (which may include hearing loss), for example, has long noted striking variability among individual responses to seemingly comparable body states. A central component in the construct of illness behaviour is the importance attached to self-attention and body awareness. The concept of illness behaviour has directed researchers to look at the myriad of psychosocial variables that influence the process of help-seeking. The way one responds to internal states (e.g., perception that symptoms are worsening vs. abating), the impact of that response on one's social roles, the resources available to successfully cope with health-related problems, and the overall effect of one's approach to illness on physical and psychological well-being all contribute to one's self-assessment of health status. It is therefore not surprising that lower self-assessed health would be correlated with greater help-seeking and health service utilization. 45 Common Sense Model Leventhal and colleagues extended this consideration of process in their Common-Sense Model, which posits that people develop a representation of illness threats that guides their health-directed behaviours, with attributes ascribed to these threats such as symptoms and label, causes, course (i.e., the interval from symptom onset through illness duration) and consequences for health (Leventhal, Meyer, & Nerenz, 1980). This cognitive model was expanded and transformed to the self-regulatory model of illness behaviour (Cameron, Leventhal, & Leventhal, 1993), which hypothesizes that emotional responses add to the cognitive illness representation to direct coping responses. Findings on help-seeking based on this model were obtained in a longitudinal study of middle-aged and older adults matched on age, gender and health status (Cameron et al., 1993). In the study of Cameron and colleagues (1993), symptoms were identified as necessary, but insufficient, cues to help-seeking. Compared with their matched controls, help-seekers rated their symptoms as more severe and as having greater long-term consequences (as measured by perceived limitations on activities and perceived lack of control over the problem). Symptoms, according to these authors, are the targets for coping responses and serve as points of reference for appraising coping outcomes. Those who seek help engage in more active coping strategies but rate their coping efforts as less effective than matched controls. Social communication about the problem was identified as an important coping procedure and significant others advised those with problems to seek care in fully 50% of help-seeking cases, highlighting the importance of social support and "social permission" in help-seeking (Zola, 1973), Cameron et al. concluded that help-seeking is based on cognitive representations of symptoms that are more elaborate and well-defined in those who seek help than in those who don't seek help. They also concluded that seeking help serves to relieve the load of emotional distress created by symptoms. In another study, older adults were interviewed and their medical records for the year before and after the interview were reviewed (Hansell et a l , 1991). Results suggested a significant, albeit weak, correlation between increased volume of initial visits to physicians and higher levels of body awareness. Their 46 results also suggested that greater body awareness, both physiological and psychological, is associated significantly with poorer self-assessments of health over time, (the authors acknowledge that more objective, external assessments of health status would have been preferable to self-report measures). Turner et al. (1991), in a study of service utilization by physically disabled persons, concluded that physical limitations (as defined using measures of activity limitation, pain, self-report of health, and psychological distress) play a significant causal role in service utilization. It was hypothesized that this finding may be related in part to an increased "preoccupation" (p. 294) with health on the part of physically limited individuals, (which could be interpreted as greater attention to bodily states). One final model will now be presented that specifically explores delay in help-seeking. Exploring Delay in Help-seeking Many studies have investigated delay in seeking medical help, conceptualizing delay as the total time elapsing from when the person first noticed a symptom up to the time he or she was seen by a doctor. Safer, Tharps, Jackson, and Leventhal (1979) divided total delay into three components: appraisal delay, illness delay, and utilization delay. They studied how different processes may work at each of these stages and predict utilization. Appraisal delay was defined as the number of days that elapsed from the day the patient first noticed his/her symptom up to the day he/she concluded he/she was ill. The authors posited that perception/sensation make one aware of symptoms, and that appraisal and coping strategies were the most probable predictors of appraisal delay. Symptoms that provide well-defined and strong sensations were felt to be more likely to lead to appraisals that something specific is wrong and to shorten the appraisal period, whereas vague complaints or weak and varying symptoms are hard to define and interpret, consequently lengthening the appraisal period. Illness delay is defined as the number of days from the end of the appraisal delay up to the day when the patient decided to seek professional help. The final stage of delay, J 47 utilization delay, is the period between the point at which the patient decided to seek help to when the patient was actually seen at the clinic. In their study, patients who had short total delays were persons who did not have a competing personal problem (e.g., divorce) and who had a painful symptom. Patients with old symptoms and those who imagined possible, severe consequences of their illness had long illness delays. The best predictors of long utilization delays were concern about the monetary cost of treatment, the absence of pain, and the belief that one's symptoms could not be cured. The results supported the hypothesis that different factors mediate delay in each of the three stages, while recognizing that the length of delay in one stage could affect the length of delay in another stage. The analytic framework provided by Safer et al. (1979) is useful conceptually, although its value in predicting help-seeking, as with other studies in the literature, may be limited by study design. Specifically, these studies are retrospective and subject to the recall biases associated with retrospection. Also, studies revolve around illness episodes that last days or weeks before help is sought, unlike presbycusis, the symptoms of which may drag on for years before help is sought. Gender Differences in Help-seeking Many studies have reported that females have higher rates of health service utilization than males (Tessler, Mechanic, & Dimond, 1976; Verbrugge, 1979). Compared with men, women have been found to engage more frequently in health-directed behaviours and seek health-related information; this greater interest in and concern with health is attributed to social role differentation (Verbrugge, 1985). These gender differences seem to disappear for older adults (Liang et a l , 1999; Mutran & Ferraro, 1988). There is evidence that older women are more likely than men to suffer from chronic ailments such as arthritis and musculoskeletal problems (versus the much higher incidence of heart attack and strokes among older men) (Verbrugge, 1984). Mutran and Ferraro (1988) found that although older women report more symptoms (also reported by Hibbard & Pope, 1983), perhaps related to the greater prevalence of chronic problems in women, this did not translate into a greater number of visits to physicians. Gender 48 differences in discretionary use of physician services disappeared after self-assessed need for medical care was considered; that is, men who assessed their health as poor were more likely to see a physician than were women who assessed their health as poor. Also, given equal levels of disability and overall health status among older adults, it is men who are more likely to be hospitalized, which suggests differences in social factors that are determined primarily by physician preferences rather than by biological differences. It seems difficult to extract a main effect of gender on help-seeking; moreover, it seems of little utility to even attempt to extract one, given the complexity of psychosocial factors that influence help-seeking. As an example, family dynamics, most notably between spouses, may be such that one member may exert a significant influence upon the other with regard to help-seeking. As a second example, consider the fact that a greater percentage of older women than men live in poverty (Townson, 2000), and that a significant link has been established between poverty and reduced health status (Wilkinson, 1992). Therefore, the finding that more older women than men seek help related to health status may reveal more about help-seeking factors related to socio-economics rather than about gender per se as a factor in help-seeking. Aging Effects in Help-seeking There is consensus that growing older leads to a greater average need for health care services. Attitudes about the health care system appear to affect usage among older adults: medical skepticism reduces usage while a stronger belief in the value of preventive health care appears to promote the use of services (Penning & Chappell, 1990; Strain, 1991). Family and spousal support appears more important than peer influences for older groups (Umberson, 1992); however, advice from friends might be influential among older persons Jiving alone who are unsure how to interpret a physical complaint (Edwardson, Dean & Brauner, 1995). These findings demonstrate how influences may change and/or be cumulative across the lifespan and with life circumstances, highlighting the need to consider help-seeking within the social context of the individual (Haug, Musil, Warner, & Morris, 1998). 49 It is important to distinguish between one's proclivity to seek help and the actual use of health services. As Barer, Evans, Hertzman, & Johri (1998) elucidate, health care utilization among older adults is largely a function of the way that health services are delivered to this population segment. In the last few decades, they point out, greater health care investigation and intervention among the elderly who seek help has led to a greater proportion of hospital admissions and longer hospital stays among the older population segment. Greater health care utilization, therefore, is not just a function of individuals' perceived needs, but is heavily influenced by the system of health care delivery, especially once help-seeking has been initiated.14 Many studies of older adults and their utilization of health care only examine the number of physician visits, often without even making a distinction between patient-initiated and physician-initiated visits. In terms of patient-initiated visits, older adults are actually quite judicious in their use of services. Elderly adults attribute many symptoms, especially mild chronic ones, to the aging process (Kart, 1981). Older adults are less likely to see chronic mild symptoms as illness warnings (Prohaska et al., 1985), and symptoms that are less likely to be associated with illness by older respondents are ones that might be expected (correctly or incorrectly) to increase with age (Kart, 1981). Furthermore, those who view variables associated with stress as predictors of health service utilization acknowledge that stress may be less relevant in old age, with evidence that the elderly attempt to minimize the experience of stress as a coping strategy (Leventhal, Leventhal, Schaefer, & Easterling, 1993). Summary of General Literature on Help-seeking The preceding review has presented only a subset of the voluminous literature on health behaviour, help-seeking and health services utilization, with the intent to provide an overview of some of the major theories and models, and research studies related to the 1 4 The complexity of health problems and associated medications in the elderly, paired with an increasingly fractionated system of diagnosis and treatment among numerous medical and para-medical specialties, also leads to more frequent and/or longer hospital stays. 50 topic of help-seeking. This review has spanned many disciplinary perspectives but nevertheless reveals a convergence of themes regarding the factors that contribute to help-seeking, health services utilization and the decision-making process behind such utilization. In general, symptoms appear to be a necessary but insufficient stimulus to help-seeking. More to the point, it is the way symptoms are interpreted and responded to that affects decisions around help-seeking (i.e., do they become labeled as illness representations or develop into illness behaviour). There is evidence that those who are more aware of symptoms (have greater body awareness), who interpret symptoms as severe, who perceive that they are not coping effectively with these symptoms, and who believe in the efficacy of treatment are more likely to seek out services. There appears to be consensus that a wide range of social, psychological and physiological factors interacts in the dynamic process of help-seeking. The relative contribution of each category of factors is not yet known, and indeed may be too variable across individuals and situations to be determined. Consider, for example, the additional layer of complexity introduced by the effects of culture and ethnicity, which have not even been touched upon in this discussion. Moreover, assumptions based on simplistic analyses must be challenged. While the process leading up to a decision to seek help for a health condition may be highly individualistic, service utilization encompasses factors outside the decision-making realm of individuals that are mediated by service delivery models, especially physician preferences, and socio-economics. This points to the importance of considering the context (e.g., social, economic, and political) within which wide-sweeping generalizations are made about sub-populations, in particular the aging population. Just as important social determinants of health across the lifespan have been established (Hertzman, Frank, & Evans, 1994; Hertzman, 1999), the social determinants of help-seeking and health care utilization must be carefully considered. An over-riding methodological shortcoming of many studies of help-seeking and health services utilization is their reliance upon self-report of health as a measure of health status that in turn defines need. In this research design context, it should come as no surprise that subjects who rate their health poorer seek help more than others whose view 51 of their health is more positive. Research is needed to address what mediates between psychosocial and physiological variables and one's self-appraisal of health and particular health conditions (e.g., Haug et al., 1998). The studies reviewed suggest that factors underlying one's choice of coping styles and strategies, and their perceived effectiveness may play a key mediating role. The literature reviewed also underscores the need for better research paradigms to investigate help-seeking. While the literature described above provides much information on help-seeking and the utilization of health services, one challenge in trying to relate this literature to help-seeking for hearing loss is that population-based studies often cover a wide range of health conditions. Given evidence that symptoms may be an important factor in help-seeking, one question that arises is: How does help-seeking for acute, possibly life threatening, episodes of illness differ from help-seeking by older adults for chronic conditions more closely related to presbycusis? To look at this issue more closely, it is worthwhile to explore studies of help-seeking for a specific health condition or issue somewhat similar to presbycusis. Some well-designed research on oral health in older adults provides a good example. Help-seeking and Oral Health My years of clinical observation, as well as comments of the participants in the current research project, suggest that older persons with hearing problems often spontaneously compare their hearing with other bodily systems that are in decline with age, most commonly eyes and teeth. Patients compare dental and hearing problems in terms of how well their teeth and ears work (if they have their own teeth) and the limited usefulness of available prostheses (if they have or are considering dentures or a hearing aid). There are several perceived similarities between hearing aids and dentures, such as their ability to improve function and their limited ability to restore function without irritation or inconvenience. Thus, an exploration of help-seeking for dental problems may provide, along some dimensions, an interesting parallel to help-seeking for hearing problems. 52 Oral health in older adults has been a research focus of MacEntee and colleagues. Their work is distinguished by a blend of quantitative and qualitative methodology and a refreshing interdisciplinary approach. In one study that looked at the influence of age and gender on oral health and the use of services (MacEntee, Stolar, & Glick, 1993), over 500 independently-living adults over 70 years of age were interviewed about their health and social support and a subset of255 (49%) were examined by a dentist. No significant differences in age, gender, self-reported oral health or use of services were noted between the group who agreed to an examination and those who declined. The self-assessment of oral health was much more optimistic than the clinical data revealed: 78% rated their teeth or mouth in at least good condition, compared with a rating of good in only 45% of mouths examined. This optimistic self-assessment of dental status matched self-assessment of general health status. Age and gender were not significantly associated with oral health and dental complaints, but both age and gender did influence the use of dental services, with greater use noted among women, younger participants, and those who were not concerned about the cost of service. A majority claimed to have visited a dentist "since age 65", but dental use declined as age increased, even though visits to primary care physicians increased. Compared with younger participants, a greater proportion of older participants reported that they only went to dentists when they had pain. A major finding was that the use of dental services for older adults appeared to match the pattern of use established in their younger years. This finding is consistent with the findings of the health services utilization studies cited above, insofar as patterns of previous utilization of health services are among the best predictors of subsequent utilization, after correcting for health status (Eve, 1988; Turner et al., 1991). About one-third of all subjects complained of a mouth problem, and about half of those who complained said that the problem had been present for at least a year. Complaints were most frequently related to denture discomfort and dental pain. Over one-third of those with a long-standing problem would not or could not identify why they failed to seek professional help. The remainder gave a variety of reasons why help was not sought, in the following order of frequency: the problems were unimportant, cost was a concern, getting to a dentist was difficult, previous treatment had been unhelpful, and 53 sickness preempted action. Almost one third complained that they could not get dental treatment.15 In an effort to overcome limitations of the. structured interviews used in then-studies, MacEntee and colleagues engaged a smaller number (24) of independently-living elders in open-ended interviews prompted by the question: What is the significance of oral health in the lives of older adults? (MacEntee, Hole, & Stolar, 1997). Findings were quite different in this qualitative paradigm. Here older adults described oral health not from a focus on dysfunction, but rather within the context of three interacting themes: comfort, hygiene, and health. These three themes illustrated that oral health was important for social interactions at least as much as for personal comfort and general health. The narratives of these elders on the topic of oral health also emphasized the positive role of adaptation and development as an integral part of successful aging. On the basis of these quantitative and qualitative studies, MacEntee (1996) recommended: that future investigations of behaviour and beliefs relating to oral health should be sensitive to this balance of development and decline that seems to regulate our lives as we get older, and I suggest that quantitative methods alone fail to penetrate much below the surface of this complicated psychological environment. (p.80) Summary of Help-seeking and Oral Health MacEntee et al 's (1993) findings show both similarities and differences with the previous studies on help-seeking that have been reviewed. Two points of similarity are that delay occurred before help was sought for dental problems and that use Of dental service in younger years was a good predictor of its use in old age. Factors pertaining to service delivery were noted to be important: older adults expressed concern about the cost and availability of dental services. One noteworthy difference compared with other help-seeking studies reviewed was the finding that the use of dental services declined with age in MacEntee et al 's study, despite an increase in the use of physician services. This difference may reflect elders' lower priority attached to dental care compared with other 1 5 The factors underlying this complaint were not elucidated by MacEntee et al. (1993). 54 health conditions, a skepticism about the efficacy of available solutions, and/or the lack of available services. An important revelation provided by MacEntee et al.'s (1997) qualitative approach was the balance that older adults attempt to reach between seeking help for symptoms and adjusting to those symptoms. For example, many older adults with ill-fitting dentures chose to accommodate to the problem rather than seek help. This pattern matches the general help-seeking literature and studies of aging that suggest that, compared with younger persons, older adults may tolerate symptoms that are of greater severity or for a longer time before they seek help. Against the backdrop of perspectives gleaned from the above studies, help-seeking for hearing loss will now be considered in detail to elucidate factors that are likely to promote and hinder help-seeking. Help-seeking for Hearing Loss There have been few studies that specifically examined help-seeking related to hearing loss, despite statistics cited in Chapter 1 that indicate that only a small percentage of persons who may benefit from audiologic rehabilitation services and products seek them out and that an even smaller percentage utilize them. Hearing aids, assistive listening devices, and communication strategies training are not ideal solutions, but they have proven to be effective in reducing the disabling effects of hearing loss and enhancing quality of life (e.g., Mulrow et al., 1990; Pichora-Fuller & Carson, 2000; Weinstein, 1991). Why do people not avail themselves of these solutions? The following studies provide clues. Studies of Help-seeking for Hearing Loss Gilhome Herbst, Meredith and Stephens (1991) looked at the social implications of hearing impairment as they affect help-seeking activity in two different populations of elderly people. The goal of their study was to explore social and cultural differences 55 between the two populations. The samples consisted of all persons aged 70 years and over registered at two particular general medicine practices, one in an inner London borough (n= 253) and one in a Welsh village (n= 99). The London population was studied between 1978 and 1980 and reported on earlier by Humphrey, Gilhome Herbst, and Faruqi (1981). The Welsh study took place in 1989. It should be noted that at least the Welsh sample included a large number of people who had hearing loss due to significant occupational noise exposure. In both samples, approximately 25% of hearmg-impaired participants (with hearing impairment roughly defined as a mean loss of 35 dB H L or greater across the "speech frequencies" in the better ear) replied negatively when asked if they had ever noticed any difficulty with their own hearing. Of those who did admit to difficulty, 25% of the Welsh group said they had never consulted their doctor16 about it. Forty four percent (17 out of 39) of the Welsh group who did consult with a physician obtained a hearing aid; the paper does not discuss how much time may have elapsed between the visit to the physician and obtaining the aid, or how many times the patient may have discussed his/her hearing at appointments with the doctor before a referral was made. Results indicate that 56% of hard-of-hearing persons who did consult with their doctor about hearing problems did not obtain an aid because the physician did not refer them for one. Details are not provided regarding how this question about consultation was worded, or whether the response format was closed- or open-ended, although such details may have influenced how participants responded. Thus it is unclear to what extent the physician may have been responsible for the lack of follow-through (especially given that, in the opinion of the authors, the physician involved in this study was sympathetic to the concerns of hard-of-hearing persons). Other factors that may be implicated, such as a patient's attitude toward hearing aids and issues surrounding service delivery (e.g., long waiting lists or convoluted referral and follow-up procedures) are not reported.17 The authors conclude that, for the London sample, the two major determinants of help-seeking 1 6 Please note that all references to "doctor" or "physician" in describing British studies refer to a general medical practitioner. In the British hearing health care system there are specialists called "audiological physician" and "otolaryngologist" (ENT). 1 7 The authors state that "the process of procuring an aid was found to be equally problematic in each sample" (p. 2 1 2 ) , but these problems are not explained. 56 were the severity of impairment and the number of years hearing loss was noted before reaching retirement age. Swan and Gatehouse (1990) sought to identify factors that lead individuals to seek management for their hearing loss. They compared 269 consecutive new adult referrals to one audiology clinic (consulters) with 286 others identified in a British National Study of * 18 Hearing as having never sought medical advice as an adult about their ears or hearing (non-consulters). Their aim was to compare the level of disability and handicap in these two groups after controlling for impairment, age, gender, and socioeconomic status. Impairment was measured as the better-ear and worse-ear pure-tone averages at .5, 1,2 and 4 kHz. Disability was measured using a monaural speech discrimination test in quiet, with the non-test ear masked. Reported disability and handicap were assessed in two ways. First, subjects were asked to give their own estimate of their hearing in each ear as a percentage of normal hearing (100%). Second, all non-consulters and 60% of consulters completed the Institute of Hearing Research Hearing Questionnaire (Swan & Gatehouse, 1990, p. 160). Compared with non-consulters, results showed that consulters had greater measured disability, and consulters whose better average threshold was less than 40 dB H L had greater reported handicap and disability (for everyday speech, speech in quiet, and sound localization) when measured disability and impairment were controlled.19 Consulters were also found to have more asymmetrical hearing than non-consulters. Swan and Gatehouse (1990) conclude that the main factors that influence help-seeking are disability and handicap, which do not necessarily correspond to one's hearing impairment. It is noteworthy that a greater percentage of consulters were women than men. The authors acknowledge the limitations of using one measure of disability in this study (monaural speech discrimination in quiet, the choice of which was determined by the standard test battery in use at the audiology clinic from which all consulters were drawn), and from the test options chosen for the National Study of Hearing. They conclude their article by stating that a valid and practical test of disability to measure the efficacy of Participants were a stratified population sample from the electoral register. 1 9 More severely impaired consulters and non-consulters reported similar severe disability and handicap, a result that may reflect a floor effect in the questionnaire scale. 57 rehabilitation is not yet in use. Such a measure is required before there can be a shift from the current practice of managing patients on the basis of pure-tone thresholds (impairment). The imperative for this shift is suggested by the fact that patients consult because of disability and handicap (albeit defined by limited available measures), issues that do not necessarily correspond directly to their impairment. A study by Brink, Wit, Kempen, and Heuvelen (1996) examined help-seeking for hearing loss and compared groups showing dissimilar help-seeking behaviours in terms of their attitudes toward hearing loss and hearing aids. Their community-based study was part of a Dutch longitudinal study on aging. The mean age of the 624 participants was 68.9 years (range 57-92), 56% of whom were female. Hearing was assessed for air-conducted tones at .5, 1, 2 and 4 kHz. Participants with a mean loss of 35 dB H L or more in both ears were considered hearing impaired; 144 (23.1%) of the sample met this criterion. Help-seeking for hearing loss was assessed by a questionnaire, which inquired about discussion of hearing problems with a physician, whether a hearing aid had been tried, and possession and use of a hearing aid. Each hearing-impaired participant completed a twenty-item questionnaire on perceptions of disability and handicap. In addition, the investigators developed and administered an attitude questionnaire based on factors in the revised Health Belief Model, that probed beliefs about the severity of one's hearing problems, benefits of using a hearing aid, associated costs (or barriers), and interpretations of the opinions of significant others on hearing loss and hearing aids; In this study, 27% of hard-of-hearing participants had not discussed hearing loss with their doctor, a further 26% had discussed it but decided not to try a hearing aid. Three percent did not keep the aid after a trial period. Forty-one percent of participants reportedly used a hearing aid; a further three percent reportedly had an aid but did not use it. Help-seeking was not found to be related to sex or age (when level of hearing impairment was controlled), but participants who took greater help-seeking action were generally more impaired. Two distinct patterns of differences in attitude among the help-seeking groups were found. First, hearing aid users perceived their hearing problems to be more serious, reported greater social pressure to seek help, and saw greater benefits of aid 58 utilization. In contrast, those who did not consult a doctor reported the least severity in terms of perceived hearing problem, social pressure, and expected benefits from a hearing aid. Second, those who consulted a doctor but did not try an aid perceived greater stigma associated with hearing aid use. They also reported that significant others saw more disadvantages than benefits to an aid and had advised thenragainst obtaining one. However, the researchers did not solicit the views of significant others directly, and they suggest that caution be exercised in interpreting the views of others filtered through hard-of hearing respondents. Brink et al. (1996) raise the important point that the fact that attitudinal factors cor vary with help-seeking behaviour, as outlined in the above findings, does not prove that attitude causes behaviour; it is equally reasonable to conclude that behaviour causes attitude. They cite Festinger's theory of cognitive dissonance (Festinger, 1957), which suggests that individuals adjust their beliefs to their behaviours, and other psychologists (e.g., Bern, 1967), who argue that this adjustment may be explained simply by the fact that people observe their own behaviour and infer their attitudes and beliefs from it. The authors conclude that only prospective studies, not yet undertaken, that demonstrate the existence of beliefs prior to the behaviour could establish that attitude is a determinant of behaviour. A common conclusion of the three studies described above is that greater help-seeking behaviour is correlated with greater perceived seriousness (disability and handicap) of a hearing problem. This finding is consistent with results from the general literature on help-seeking and health services utilization that show that variables pertaining to needs are the best predictors of service use. Nevertheless it remains a challenge to extract the true factors influencing hearing aid use and help-seeking. For example, i f one asks participants whether the following statement (taken from Brink et al., 1996) is true, partly true, or not true for themselves (as a way to tap severity of loss as a factor in assessing attitude): "I can still manage quite well without a hearing aid", an affirmative answer that correlates strongly with a lack of help-seeking is interesting but reveals little about the factors underlying this perception. Questionnaires are limited in the insights they provide into the factors that influence the responder. A forced-choice response format, in 59 particular, constrains respondents to answers that may reflect assumptions inherent in the questionnaire. In short, such questions don't adequately tap into why help is not sought. The findings of Brink et al. (1996), like those of Gilhome Herbst et al. (1991), raise important questions about the roles that physicians and the system of hearing health care delivery play in help-seeking. There may be many reasons why one quarter of participants did not try an aid after consulting their doctor; Brink et al. (1996) discuss this issue in terms of patients' attitudes toward hearing loss and hearing aids. Their inaugural study did not make explicit a number of key points, such as the nature of the steps involved, or the potential barriers to getting from the physician's office (assuming this step is required) to the hearing aid and other rehabilitative solutions, whether the doctors who were consulted all recommended a hearing aid trial and why they did or did not do so, what kind of waiting, list may be a deterrent, what financial considerations may have been at play, and so on. To date, the importance of such contextual factors has not been adequately considered in population-based studies of help-seeking for hearing loss. Future studies of help-seeking for hearing loss should investigate such factors. Brink et al., (1996) suggest that their results show a higher incidence of help-seeking for hearing impairment compared with the British study by Gilhome Herbst and colleagues (1991), after correcting for age differences and criteria differences for defining hearing impairment between samples. A greater percentage of Dutch than British hard-of-hearing participants sought help from their doctor (73% vs. 57%), and obtained a hearing aid (44 % vs. 25%). These differences may reflect true cultural differences in help-seeking that result from more positive attitudes toward help-seeking, and/or more streamlined and comprehensive hearing care services in the Netherlands. Such a conclusion is speculative, however, without knowing the service delivery model in place for each study, and, in the absence of details about the wording of the questions asked, the reasons for the particular responses. For example, more patients may seek help and obtain hearing aids when the state or another third party pays for hearing aids, and/or the participant may be reluctant: to disclose dissatisfaction Or non-use in such cases. In summary, the studies described above document that many adults with hearing impairment don't seek help by consulting their doctor, and many who do consult do not 60 take, or are not referred for, further rehabilitation. They also suggest that help-seeking is positively correlated with the perceived seriousness of one's hearing problem, although we have seen how this finding may be a reflection of how disability and handicap are measured. The emphasis of these quantitative population-based studies has been on the role of auditory factors in help-seeking. With the exception of Brink et al 's attention to attitudes and health beliefs, or predisposing factors (Brink et a l , 1996), the specific psychosocial factors that may promote or hinder help-seeking for hearing loss have not been addressed. The role of co-occurring health conditions, for example, has not been looked at, nor, as mentioned, have the effects of culture or service delivery models on help-seeking been examined. In summary, what is needed in research on help-seeking for hearing loss is an approach that is more ecological, that is, one that more thoroughly examines the contributions of the hard-of-hearing person, significant others, and the environment. In contrast to the quantitative focus of the above studies, Getty, Gagne, and McDuff (1996) qualitatively explored the obstacles to seeking help for hearing loss. They conducted a series of focus group interviews with 24 women and men over 70 years of age who lived in an urban Quebec community and who had seen an audiologist and obtained a hearing aid or aids (information was not obtained on how many used their aids). A follow-up study was carried out with a similar population that public health nurses identified as having a hearing problem but that never consulted a professional about hearing problems. While the quantitative studies cited above focused on auditory factors related to the hard-of-hearing individual that may prompt help-seeking, Getty et al 's (1996) study revealed a wide range of factors from the personal (micro) to societal (macro) that work to prevent or delay help-seeking for hearing loss. These factors are included in the following section that summarizes the literature on how such factors seem to delay help-seeking. 61 Factors that Delay Help-seeking The insidious advance of hearing loss with advancing age is such that often many years may pass before one is aware that one's hearing has deteriorated. An important point is made by Gagne Hetu, Getty, and McDuff (1995), who remind us that having a limited awareness of a problem is not the same as denying that a problem exists. Pichora-Fuller and Kirson (1994) report that not only hard-of-hearing persons but also young and old normal-hearing persons tend to attribute difficulty understanding spoken language to non-auditory factors. Many of these attributions may be valid. Getty et al. (1996) reported that elderly persons sometimes confuse the symptoms of hearing loss with those of senility. So we see that it is frequently difficult to link cause and effect when it comes to hearing impairment and its consequences. Presbycusis is also passively accepted by seniors as a "normal part of aging" (Brink et al., 1996). Maurer and Rupp (1979) dubbed this "geriapathy," and we have seen this phenomenon in other studies of help-seeking among the aged. Physicians and other professionals, and indeed society at large, are also of the opinion that hearing loss comes with age and can't be helped or is not consequential enough to warrant hearing aids (e.g., Humphrey et al., 1981). Such attitudes may be a reason why many who do consult their doctor about hearing problems are not referred for further help (Gilhome Herbst et al.,1991; Lichtenstein, Bess, & Logan, 1988). Hearing loss is a chronic condition, but because it is not painful (like arthritis20) or life-threatening (like cancer or heart disease) it may be deemed low priority for help-seeking (Getty et al., 1996). Furthermore, seniors in Getty et al's study (1996) spoke of their resistance to change "at my age" and of difficulty learning the new skills involved in hearing aid use. Studies indicate that older listeners tend to report less handicap than younger listeners with the same hearing loss (e.g., Garstecki & Erler, 1996; Gatehouse, 1991; Lutman, 1991). This is despite the greater deleterious effects of noise and reverberation on aged listeners due to a decline in peripheral and/or central processing abilities. This under-It is noteworthy that significant help-seeking delay is noted as well for arthritis, even for those with significant pain (reported in Zola, 1986). 62 reporting of hearing difficulties may be due in part to a lack of awareness of changes in hearing abilities as noted above, but may also reflect some combination of change in communication expectations, needs, or environments (Cheesman, 1997). Furthermore, one's communication goals may have a strong influence on the degree of handicap experienced in conversation (e.g., Johnson & Pichora-Fuller, 1994; Pichora-Fuller, Johnson, & Roodenburg, 1998). This relates to the earlier discussion of communication in aging where it was noted that older persons may attach more importance to the interactional, relational or social function of language than to the transactional, ideational component. Other factors that delay help-seeking are associated with knowledge of and attitudes toward the available "solutions" to hearing loss. Hearing aids have a "bad" reputation and are seen as devices that do not "cure" hearing loss, but instead are "ineffective crutches" that make the impairment more visible (e.g., Getty et al., 1996; Humphrey et a l , 1981; McCormick, Pichora-Fuller, Paccioretti, & Lamb, 1994). Hallberg and Carlsson (1991) suggest that by increasing the visibility of a hearing loss a hearing aid may actually contribute to increasing one's handicap by increasing stigma (see also Goffman, 1963). In addition, there is little knowledge about other rehabilitative solutions to handicapping situations, such as use of assistive listening devices and communication strategies (e.g., Getty et al., 1996). Stephens (1996), for example, reports that "Hearing Concern," the British association of hard-of-hearing persons, identified the need for information on assistive devices as one of its top priorities. Another group of factors are ones that may restrict accessibility to services and products. For example, some people may be impeded in help-seeking for hearing problems because of impaired mobility or physical disability; many other individuals are limited by financial barriers (e.g., Franks & Beckmann, 1985). Getty et al., (1996) cite additional organizational factors associated with the distribution of audiological services. Often there is limited knowledge by the public of the services available. In residential care settings across Canada, for example, there is a lack of knowledge as to where to obtain hearing devices for residents and training in the use and care of hearing instruments is often 63 lacking (Carson, 1997). Sirnilarly, in the province of Quebec, Getty et al. (1996) noted that the consultation and referral process in the community is long and complicated. Finally, with regard to the issue of identity, it has been suggested that while there is no shared identity or culture linking hard-of-hearing people, all those who have their hearing tested share the common experience of "coming out" and revealing the possibility of being hard of hearing to at least a selected portion of the public world. In avoiding the possible diagnosis of "hearing impaired," the individual avoids the serious social consequences of a publicly recognized hearing disability. Going public with one's hearing loss entails a reconstruction of one's sense of self to incorporate hearing loss into one's identity, at both the individual and group level. It may also entail a redirection of one's life expectations. Such personal evaluations and shifts require the deployment of significant cognitive and emotional resources (e.g., McCormick et a l , 1994; McKellin, 1994). Factors that Facilitate Help-seeking and Cues to Action Studies indicate that there are two main factors that trigger help-seeking by a hard-of-hearing person. As previously discussed, one factor is the severity of the disability and handicap related to the hearing loss (as opposed to the hearing impairment per se). The second is social pressure from significant others to take action. Let us consider this latter deterrninant in more detail. Social pressure is exerted by significant others for the hard-of-hearing person to "do something" about her/his hearing problem (Getty et a l , 1996; Hetu, Riverin, Lalande, & St. Cyr, 1988; Hickson, Hamilton, & Orange, 1986; O'Mahoney, Stephens, & Cadge, 1996). O'Mahoney et al. (1996) report on results of a questionnaire they administered to 95 consecutive new patients seen at two audiology centres in Britain (London and Cardiff) over a three-month period. Patients were asked to identify whom they perceived as being the main motivating influence in their first seeking help for hearing loss. In analyzing the data, these patients were subdivided into two groups: those who worked or were seeking work outside the home (Group A), and those whose main occupation was within the home (e.g., homemakers) or were retired (Group B). Not surprisingly, these two groups differed in mean age (A: 47.6 years; B : 70.3 years) and mean hearing threshold in the better ear 64 (A: 28.8 dB HL; B : 45.3 dB HL). There were significant differences between the two groups in who was reported to be the prime motivator for help-seeking, with four categories identified: self, general practitioner (GP) or other health professional, family member, and other. Prompting by a family member was reported almost twice as often in Group B (55%) compared with Group A. The researchers muse whether this may be an indication that often it is the family who feels more handicapped than the patient. Issues related to lifestyle and lifespan may account for the group differences: Group B members may have greater hearing impairment, but their retired lifestyle may also afford them more contact with family and therefore more opportunities for family members to prompt them to get their hearing checked, regardless of whether they live alone or not. It was also found that Group B reported self as motivator half as much as Group A and "GP or health professional" as motivator three times as much as Group A . The study, however, did not explore whether it may have been the patient, rather than the professional, who initiated the discussion of hearing loss at an appointment. If the patient raised the topic, then the category of "self-motivated" is underrepresented in their findings. If the patient raised the topic at the insistence of a family member, then the category of family member is underrepresented. Furthermore, the authors acknowledge that the study does not address the issue of how often patients may have raised the issue of hearing with a physician before a referral was finally made. The answer to the question " who do you perceive to be the main motivating influence behind your first seeking help for your hearing loss" does not describe how subjects may interpret their first "motivating influence" in help-seeking: is it the hospital referral (where the questionnaire was administered) or the visit to the physician (as the authors imply). If it is the visit to the doctor, how does the GP act as the main influence? The authors stress that the findings of this pilot study are tentative and answers to their overall research question concerning the relationship between the outcomes of rehabilitation and self-motivation to seek help await further studies now being undertaken. O'Mahoney et al 's (1996) study reveals once again that it is challenging to find answers to questions about help-seeking within a research paradigm that relies on questionnaires. Given the complex nature of help-seeking, related to the interweaving of 65 psychosocial factors (family dynamics, service delivery issues), and methodological interpretation problems for self-report measures, researchers must seek alternative paradigms to more adequately understand this topic. In their group interviews with seniors, Getty et al. (1996) identified several other triggers to help-seeking for hearing loss. In addition to pressure from family members, a significant change in lifestyle often prompted action, such as the death of a spouse, a divorce, or a marriage. This is consistent with findings of research on other chronic conditions. Meillier, Lund, and Kok (1997), for example, undertook a qualitative study of cues to action in change of lifestyle to prevent coronary heart disease in males. Changes in life circumstances which, in the authors' words, constituted a "breach with existing life patterns" (p.45), included moving to a new place, starting a new job, and having a new partner, and were notable factors in lifestyle changes for seven of their ten interviewees. Other triggers to help-seeking reported by participants in Getty et al's (1996) interviews were changes in visual capacity (which may constitute a change in lifestyle if visual changes became severe and/or support systems or compensatory resources became inadequate), and meeting someone who is satisfied with an audiologic service. Summary of Literature on Help-seeking for Hearing Loss In summary then, the research literature on help-seeking for hearing loss to date highlights four important and related points.21 First, hard-of-hearing persons seek help related to their handicap (and, to a lesser extent, their disability) rather than their impairment. Second, there does not exist a strictly linear relationship among handicap, disability, and impairment. Two individuals with the same audiometric profile (impairment) may differ greatly in handicap. A discussion of handicap is valid only as it prioritizes the perspective of the hard-of-hearing person and contextualizes it to that person's life experiences. The proposed changes to the WHO's ICIDH reflect a growing realization of these qualifications. Third, handicap is not a fixed entity for any individual. A change in 21Expressed here with reference to ICIDH (1980) terminology, the context in which the research referred to was carried out. 66 lifestyle or change in communication demands, whether abrupt or gradual, may lead to widely varying degrees of handicap at different times. Even without such dramatic change, one's perception of one's handicap evolves over time. Fourth, many agents extrinsic to the hearing loss (i.e., environment) influence the hard-of-hearing person's decision to seek help. They include the influence of significant others, the social roles and communication needs of the hard-of-hearing person, the level of hearing accessibility in society, and the system of service delivery. As we have seen, this last agent, the service delivery system, has been afforded little attention in studies of help-seeking in general, or in studies specifically concerning help-seeking for hearing loss. We turn now to consider this influence, and in particular the influence of the audiology system on help-seeking, focusing on the relationship between the audiologist and the patient. The Role of the Audiologist and Audiology Service Delivery in Help-seeking This section reflects upon the role that the system of health care delivery, particularly the system of hearing health care, may play in help-seeking for hearing loss. The section begins with a review of traditional audiologic practice, including a discussion of self-assessment tools in rehabilitation. In keeping with a major theme of this review, the case is then made for the need to expand upon an ecological service delivery model in audiology. The section concludes with a discussion of clinician-patient communication, its importance within an ecological paradigm, and a review of the few existing studies of audiology-patient interaction. At the organizational level, the "system" of audiology may exert considerable influence on the hard-of-hearing person. As McKellin (1994) states: "audiologists... are important players in the cultural definition and social construction of hearing loss" (p.219). There is, however, a poor realization within traditional audiologic practice that the culture of audiology and the nature of its relationship with its patients render it part of the environment of the hard-of-hearing person. A more ecological approach would require that this issue be explicitly considered in audiological practice and research. 67 The discussion in this chapter has touched upon how health service delivery may affect health service utilization. At the "macro" level of policy and program planning, the Andersen-Newman model (1973) posits that access to a service is influenced in part by the financial and personnel resources allocated to the health care system, which in turn affect what services are covered publicly and/or privately, and the length of waiting lists. As with the studies of health care in general, studies of help-seeking for hearing loss have also implicated service distribution, financial barriers, and hard-of-hearing persons' lack of knowledge about services as factors that delay or prohibit help-seeking (Getty, 1996). Analogous studies of oral health in the elderly reveal service delivery issues, especially among institutionalized elders (MacEntee et a l , 1993). Thus, even if one is highly predisposed to seek help, there may be many factors extrinsic to the individual that intervene to restrict or delay access to care. Depending on how hearing health care service delivery is structured, the physician may play a pivotal role in help-seeking for hearing loss. As gatekeeper, he or she may determine whether or not a patient is referred to an audiologist for hearing help. Family physicians in Denmark (and likely in other cultures) frequently refer elderly hard-of-hearing persons to an ENT, rather than directly to an audiologist, which further delays the rehabilitative process (Parving, Christensen, & Sorenson, 1996).22 Recall that the Andersen-Newman model posited that the evaluation of a patient's illness state by the professional largely determines the nature and extent of patient care. Even where a physician's referral is not required for an audiologic assessment, the physician's opinion may still be sought, given the central role that a primary-care physician plays in many patients' lives, especially among older adults. The above points are important factors that work to help or hinder a patient from making an appointment with an audiologist. Nevertheless, once a patient walks into the audiologist's office, it is the interaction between the audiologist and the patient that may Parving et al. (1996) note that GPs' practice of referring older patients to an ENT is prevalent, despite the fact that these patients would always be examined by an audiological physician in the audiology department prior to audiometry. 68 determine the success or failure of rehabilitation. For this reason, it is important to consider the nature of this interaction. "Traditional" Audiologic Practice The way that audiology has been traditionally practised has much in common with standard medical practice, both having derived from the medical or curative model of health. Among the features of this model is a philosophy that views "disease" as an attribute of the individual, and the physician as the agent of healing through identification and treatment of pathology. This "find it and fix it" approach (Keller & Carroll 1994) sees the role of patients as one of "complying with" or "adhering to" medical recommendations, with little input into the decision-making process (Miller & Crabtree, 1994). In the medical model, the audiologist moves from impairment to disability to handicap, focusing first on quantifying the hearing impairment in order to "treat" the disability to reduce handicap. The bulk of time in an initial assessment is devoted to test procedures to measure impairment. A rehabilitation program is designed with the audiologist often as primary decision-maker and the hearing aid as the primary rehabilitative tool. Patients are expected to "comply" with "recommendations" to try a hearing aid, with little attention paid to their perceptions of their hearing-related problems compared to the attention given to the assessment of impairment. In his critique, Janzen (1995) suggests that audiologists' aim in rehabilitation is to instill certain beliefs and expectations in hard-of-hearing people so that they come to understand themselves and their problems from the specialist's perspective. Like other health practitioners, the audiologist typically dominates the communication process, setting the agenda and controlling both the subject and the style of discussion (see Giles, Coupland, & Wiemann, 1990, for studies of the discourse styles of physician-patient and nursing home staff-resident interactions). The area of discourse analysis in audiologist-patient interaction is to date essentially uncharted territory (one exception is Coupland & Jaworski, 1997, discussed later in this chapter). 69 In the medical model, both the patient and her/his health are usually assumed to be very static over time. There has been a lack of appreciation of how the needs of patients, their perceptions of their hearing, and their definition of self may evolve over time (Gagne et a l , 1995; Luborsky, 1995). Inherent in the design of clinic schedules, and even the design of outcome measures, is an assumption that the process of rehabilitation should be accomplished rapidly (Heaton, 1992; McCormick et a l , 1994). Another component of "traditional" audiologic practice are self-assessment instruments. Self-assessment instruments In recognition of the need to elicit patient perspectives on hearing loss, the past three decades have seen a proliferation of self-assessment instruments in audiology, developed by practitioners and researchers (for a comprehensive review of self-assessment instruments in hearing and related functions, and discussion of relevant issues in self-assessment, see Noble, 1998). Such tools, however, depending on the purpose for which they are used, often fall short of their goal because the questions posed therein are the product of authors' assumptions of how disability and handicap are expected to be manifested (Gagne et a l , 1995; Noble, 1978; 1983; 1998; for an opposing view, see Demorest, & Erdman, 1986). Also, as we have seen in research on help-seeking, several conditions pose serious challenges to the interpretation of the results, e.g., a participant's lack of awareness of a health problem, or a perception of it as relatively unimportant, or a hidden agenda that attempts to conceal or exaggerate a condition, particularly if the researcher or practitioner infers health status from self-report measures. These issues amount to questions concerning validity. Questions that may have no relevance in an individual's experience are suspect in terms of content validity. Construct validity (referring to the theorized nature of the trait or ability under investigation) is at risk when questions attempt to tap communication abilities that may not have great relevance to actual communication. For example, a general question about speech understanding ignores the role of the relationships between 70 interlocutors and their communication goals that may have a significant impact on then-degree of success in understanding. Self-assessment tools decontextualize the experience of hearing loss and any elaborations (e.g., story-telling) or departures the client makes from the structured question-answer-question sequence are often discarded as irrelevant or confounding to the survey plot (Jylha, 1994; Mechanic, 1995; Mishler, 1986). This nomothetic approach (Noble, 1998), whereby individual variation is treated as a source of "noisiness" affecting the assessment task, is a feature of what Mishler (1984) refers to as the "voice of medicine" versus the "voice of the lifeworld". In contrast, an ideographic approach (Noble, 1998) attempts to tap the uniqueness of the individual case by some degree of open-ended inquiry. Ideographic measures may pair specific questionnaire items with particular contexts identified by the.respondent (e.g., Dillon, James, & Ginis, 1997). Such measures are an attempt to reconcile the need for individualized input from patients with the time available at the appointment. These approaches, however, cannot replace the kind of meaningful assessment obtained through more open-ended qualitative approaches. Ecological Audiology The overarching goal of an ecological approach to audiology (Noble, 1978; 1983; Noble & Hetu, 1994) is to find solutions to hearing problems by contextualizing hearing loss within a person's real life situations. The environmental psychology literature (Stokols, 1992) underscores the importance of adopting, as Gagne and colleagues advocate, a problem-solving approach to hearing loss (Gagn6 et a l , 1995; Gagne, McDuff & Getty, 1999) that focuses on specific person-environment situations of handicap, instead of focusing on the person as hearing-handicapped. Ecological audiology has been practised in various forms for various purposes, including clinical conversation-based therapy (Erber, 1988), group aural rehabilitation programs (Israelite & Jennings, 1995), rehabilitation programs in special settings such as homes for the aged (Hoek et a l , 1997; Jennings & Head, 1994; Pichora-Fuller & Robertson, 1994), evaluation of an outreach program to seniors provided by seniors 71 (Carson, 1997), community programs promoting communication and hearing health (Worrall, Hickson, Barnett & Yiu, 1998) and in occupational health research (Getty & Hetu, 1991; Hetu & Getty, 1991). It is noteworthy that these examples have used a community participatory approach often initiated by community members, or intensely involving them in the planning, implementation, and evaluation of services. Nevertheless, we have yet to reconcile a philosophy that is truly centered on the hard-of-hearing person with the resource and time constraints of the typical audiology setting, whether that be hospital health unit, or private practice. A n update of one of the best-known models in audiology highlights the growing appreciation of the ecological paradigm in audiology. Almost twenty years ago, Goldstein and Stephens (1981) put forward a management model to guide clinical practice in audiologic rehabilitation. While comprehensive for its era, this model of evaluation and remediation was decision-analytic in its approach (similar to other models influenced by the medical model such as one presented by Turner, 1991), and miinmized the patient's involvement in this decision-analytic process around remediation. Stephens (1996) has revisited this model, acknowledging the failure of the initial version to consider psychosocial variables, especially the influence of significant others. To correct this, the updated model incorporates a "disability and handicap" component (based on WHO ICIDH, 1980) as its new starting point, and adds an intermediary component of "integration and decision-making" between evaluation and remediation, which Stephens describes as "a vital which key decisions are made jointly between the professionals and the hearing impaired people together with an input from Significant Others" (1996, p. 61). An important new component of an ecological approach in audiology is to examine the relationship between patient and audiologist in the clinical interaction. Such an examination should include consideration of the patient's knowledge of audiology services, the patient's expectations of those services and their efficacy, how well (or poorly) those expectations are met, and how the degree to which expectations are met may influence future rehabilitative action. There is, for example, evidence that patient satisfaction with his/her physician has been shown to improve patient "adherence" to 72 therapeutic recommendations (Greene, Adelman, Friedman, & Charon, 1994). Examining the clinical interaction should also encompass exaniining the audiologic assessment: from the test procedures and physical environment to the discourse style of patient and clinician. Clinician-Patient Communication It is encouraging that the field of health communication is flourishing and that the topic of clinician-patient communication and its improvement is Of increasing interest (e.g., Northouse & Northouse, 1998). To date, most studies have been centered on the communication between physicians and patients, with relatively little work directed to the interaction between other health professionals and patients. Keller and Carroll (1994) describe three perspectives that inform the discussion of the behaviours of physicians toward patients, most notably verbal behaviours. The first perspective addresses core beliefs. Core beliefs are shaped by the physician's central philosophy regarding the processes of disease and hearing, usually conceived of within the reductionistic medical model outlined above. The "voice of medicine" versus the "voice of the life world" of which Mishler speaks, for example, refers to the expression of these core beliefs, the very different orientations from which physicians and patients talk and think about health (Mishler, 1984). The second perspective views the physician-patient relationship as a consequence of the roles that each party enacts towards one another. How and what one says in a dialogue is, in part, role determined (reminiscent again of symbolic interactiohism). The nature of the relationship and the resulting communication behaviour will change depending upon the role position that both physician and patient take, although it is the physician who usually has more control given the function of the setting and the power differential between the two parties. The third perspective views the physician-patient relationship as a consequence of how physician and patient behave toward one another at a verbal level. Here, the psychological and communication skills of the physician are critical. Success in interaction is more likely if the physician can accept and utilize the knowledge that the thought and 73 articulation processes of the physician and patient are essentially different, and vary across individuals for both groups. The physician and patient have, in essence, different world views of health and illness: the patient comes from a personal experience of illness which includes lifestyle consequences, fears, and altered roles that can only be appreciated through the telling of her/his personal story; the physician comes from a background of schooling that has ingrained unique vocabulary (resulting in particular discourse patterns) and problem-solving techniques (e.g., a decision-tree approach to diagnostics). These perspectives are informed by theories in psychology and sociology and have relevance for analyzing the interaction between audiologist and patient. A number of valuable studies have approached health communication from a primarily sociolinguistic perspective in a wide variety of care-giving contexts (e.g., Coupland, Robinson & Coupland, 1994; Nussbaum, 1990; Street, 1991) and, as discussed earlier, in various sub-populations of patients such as the elderly (e.g., the work of Ryan et a l , 1986). To date this multi-disciplinary health communication perspective has not been applied to the study of the audiologist-patient interaction, except for a recent study by Coupland and Jaworski (1997). Coupland and Jaworski (1997) recorded an interaction between an audio logical physician and a patient and analyzed it to illustrate various theoretical perspectives on the social dimension of communication. The audiological physician used a highly ritualized interview format of question-answer-question to establish the nature, causes and solutions to the patient's hearing problems, but it was noted that this transactional mode frequently gave way to a more relational, interpersonal style of commimication. As predicted by communication accommodation theory and relevance theory (Sperber & Wilson, 1986), examples from this study showed how both audiologist and patient accommodate their speech to each other and attempt to make comments or questions as mutually relevant or interpretable as possible. In addition, the frequent shift of the audiologist away from the medical frame (analogous to the concept of role described above) to the relational one illustrated what Coupland et al. (1994) describe as "complex processes of frame negotiation, where doctors and patients blend their socio-relational and bio-medical priorities" (p. 94). Thus, a serious medical (transactional) frame may shift to a joking 74 personal (relational) frame as a way to build rapport and trust in a context where painful disclosures or examinations may be necessary. Coupland and Jaworski (1997) emphasize the importance of the social dimension of communication in the medical or audiological context, and that illustrate that theories of communication may help to elucidate the relevant cognitive processes that support the goal of sharing meaning between clinician and patient. Only a small number of other studies of audiologist-patient interaction have been reported in the literature. Doyle (1994) examined aspects of patient-clinician interaction in audiology by documenting activity as it naturally occurs in the clinic. Initial consultations of fifty adult patients (mean age 73.8 years) with ten audiologists in Australian government hearing clinics were audiotaped in a study of the behaviour of audiologists and their patients who were in the initial stages of rehabilitation. Recordings were analyzed to determine how consultation time was used, and what questions patient and clinician asked, and to characterize interpersonal aspects of the consultations. A profile of clinical interaction emerged in which patients appeared largely passive. On average, audiologists asked four times as many questions as patients, with "direct requests for information" comprising just under half the total number. Doyle (1994) states: Question asking by audiologists and clients indicated the apparently different agendas for the two participants. Audiologists' requests for information reflected their orientation to the task of documenting the client's hearing loss and moving to the yes or no hearing aid decision. Clients' requests for information reflected a need for knowledge about the use, benefit, and appearance of hearing aids. Consistent with their generally passive role, clients often directly solicited audiologist opinion in decision-making, (p. 224) Results indicated that all ten audiologists tended to have a routine method of presenting information (e.g., six out of ten used the patient's audiogram as the focus of their informational counselling), a finding consistent with survey data obtained by Martin, Adabie and Descouzis (1989a) who reported that audiologists in both the United States and Australia "appear primarily to proceed in standard and preset ways in conveying what is considered to be important information" (p.50). Audiologists also displayed a conversational style that appeared to "tempt" or convince the patient to accept 75 amplification (e.g., "You wouldn't have to wear it all the time", p. 221). Despite these similarities, an interesting finding of the study by Doyle (1994) was the variation in consultation characteristics across the ten audiologists along four dimensions: total consultation time, proportion of consultation time in which talk occurred, proportion of consultation time devoted to hearing assessment, and the number of patients' direct request for information. The author posits that "the individual clinician may have a powerful effect on hearing aid rehabilitation processes and, perhaps, outcomes"(p. 224), and she suggests that such potential effects should be the focus of further study. A second study (Kargas & Doyle, 1996) looked at patient satisfaction with the use of time in initial audiological consultations (assessments) with patients of all ages at two urban Australian audiology clinics (for patients under 15 years of age, the accompanying parent was the subject). The basic research question was: "Does patient satisfaction with the experience of participating in an audiological consultation vary with: 1) total consultation time; 2) time spent on various forms of information transfer within consultations (e.g., history taking, informational counseling, or other talk associated with test instructions, greetings and rapport building); or 3) time kept waiting immediately pre-consultation?" Analyses of 60 audiotaped interactions that categorized and measured the time spent in each of the above activities were compared with patient responses to a satisfaction questionnaire (four-point scale) that was completed immediately following consultation. Results indicated that satisfaction ratings were generally high and that total consultation time and waiting time pre-consultation in relation to total consultation time contributed most to variation in patient satisfaction ratings (accounting for 42% of the variance). Only three patients out of 60 expressed dissatisfaction; these three had significantly longer waiting times (an average of 28 minutes) and shorter consultation times than satisfied patients (with waiting time of roughly nine minutes). Regardless of the length of the consultation, most time within a consultation in this study was spent on "other talk" (i.e., not history-taking or informational counselling), a finding that resonates with the comments of Coupland & Jaworski (1997) above regarding the importance of the social dimension of communication. Both Australian studies described above suggest the 76 need for further study of audiologist behaviours and the relationship of that behaviour to rehabilitation outcomes including client satisfaction and degree of hearing aid use. General Summary and Review of Research Questions The literature reviewed in this chapter has documented presbycusis as a growing health concern that has significant and wide-ranging psychosocial effects not only on the hard-of-hearing individual, but also on the family and friends of the individual. The coping responses persons deploy have been shown to mediate between the impairment of presbycusis and its psychosocial consequences. The literature suggests that coping styles and strategies (on the part of hard-of-hearing individuals as well as significant others), and their perceived effectiveness, also play a key role in one's self-assessment or perception of health status, which in turn may signal the need for help-seeking. The literature on help-seeking, both with regard to hearing loss and more generally, highlights many factors that may help or hinder help-seeking. The way these personal and environmental factors combine is complex and varied. Generally, it is not the presence of symptoms per se that determine help-seeking, but how one interprets and responds to symptoms. There may be large variability in these responses, perhaps as a result of one's degree of attention to bodily states, and one's beliefs, values and attitudes about hearing and health. Social and psychological constructs inform empirical research by elucidating individual characteristics and personal attributes (e.g., beliefs and attitudes) that are posited to influence coping and help-seeking. Such predisposing factors are explored in theories and models of health behaviour, help-seeking and health services utilization, which also consider the role that enabling and reinforcing factors, predominantly environmental, play in help-seeking. Theories and models of communication contribute to explaining how social and communicative roles, intents and expectations may influence coping and help-seeking, and how these factors may differ by gender and with aging. Such models and theories are just beginning to be applied to the special situation of the audiologist-patient interaction to determine this influence on help-seeking and rehabilitative outcomes. Broader issues 77 surrounding the nature of service delivery and its impact on help-seeking have not yet been considered in audiology. As well, there is a need to explore further gender and aging issues related to how these factors influence the help-seeking process. The diverse literature on help-seeking and health behaviour can inform audiology by corroborating help-seeking factors already documented in studies of hearing loss, and offering novel perspectives that may inform audiologic research, including the present study. In summary, the path from impairment, through the range of particular coping responses, to the culminating help-seeking step of booking an audiologic appointment is influenced by a complex array of predisposing, enabling, and reinforcing factors from the micro (personal) to the macro (societal) level. Table 2.1 at the opening of this chapter listed some of the personal and environmental factors that play either an etiologic or moderating role in health and illness. Table 2.2, modeled after Table 2.1, focuses on the personal and environmental factors that the literature review in this chapter has revealed as playing a possible etiologic or moderating role in hearing loss, and factors that may act as determinants in help-seeking for hearing loss. The present review of the literature has highlighted the methodological shortcomings of utilizing research designs that do not tap the intricacies and important subtleties of the dynamic, interactive, and long-term nature of the help-seeking process. The qualitative research approaches reviewed (in studies of oral health and hearing loss) revealed aspects of the help-seeking process that quantitative studies failed to uncover. The findings of this literature review underscore the importance of an ecological approach to audiology, to mirror the shift toward an ecological view of health that is occurring in health research and practice more generally. An ecological approach to the study of presbycusis and help-seeking is one that contextualizes hearing loss within the real world situations of each hard-of-hearing individual. Moreover, within this individualized approach, the focus is on solving specific situations of handicap identified as problematic and salient by the individual, in conjunction with significant others. 78 Table 2.2 Personal and Environmental Factors in Hearing Loss and Help-seeking for Hearing Loss (Note: HL= hearing loss; HA= hearing aid; HOH= hard-of-hearing person) Biopsychobehavioural factors Biogenetic Psychological Behavioural Sociophysical environmental factors Sociocultural/physical Impairment: length of time rate of change symmetrical amount of loss type of loss Other health conditions Mobility Vision Cognitive deficits Gender Age Self-concept: e.g.: self-esteem autonomy sense of identity self-efficacy perceived control Beliefs: e.g.,: perceived severity of HL perceived efficacy ofHA's perceived efficacy of coping beh. Attitudes: toward HA's health services Attribution: e.g. to aging Communication: needs expectations, goals, importance Knowledge:HL,HA other solutions, service delivery Coping responses: problem-based emotion-based Talked to doctor re: hearing Lifestyle Soundscape: range and salience Lifestyle Communication: acoustic factors of speech linguistic and paralinguistic factors style e.g., accommodating social factors e.g., roles, goals,salience environmental factors e.g. bkgnd noise gender and age effects Family members: pressure on HOH person to seek help social support e.g., in communication coping responses Attitudes: negative: stigmatizing HL and aging positive: satisfied w HA, audiology Social roles: women older adults medical: physicians, audiologists Socioeconomic status; financial resources Technology e.g., ease of use ofHA's Hearing health care system: resources organization (medical model) physicians: attitudes, referral and intervention style; follow-up access issues: service availability waiting list, referral process, location financial structure Culture, ethnicity 79 The purpose of this research study is to qualitatively explore, within an ecological framework, four perspectives identified in this literature review as warranting study or more in-depth study. Specifically, these perspectives are: 1) the perspective of hard-of-hearing older women on their hearing, hearing loss, coping responses, and help-seeking, elicited at the point in time when a help-seeking step is taken; 2) the perspective of family members and their role in how presbycusis is experienced within the family and in help-seeking; 3) the perspective of hard-of-hearing individuals on the audiology system, including their expectations of their first interaction with an audiologist, their impressions and interpretations of the interaction and its outcomes, and an exploration of how the system of audiology service delivery may influence further rehabilitative action for presbycusis; and, 4) the perspective of the assessing audiologist and the nature of her interaction with these women and the service delivery system within which she plays a crucial role. The next chapter will discuss methodological issues related to these broad areas of inquiry. 80 C H A P T E R 3 M E T H O D O L O G Y Overview This chapter outlines the methodological strategies and procedures that were used in the present study. The chapter moves from the general to the specific. It begins with a discussion of qualitative methodology and the rationale for adopting a grounded theory approach. Details of the study design follow, including information about the sample and the setting, the recruitment of participants, and the nature of the data. The techniques and process of data analysis are then described. The chapter closes with a discussion of ethical issues and the criteria for assessing the quality of the study. The main purpose of this research was to explore the perspectives of older women on their hearing, their hearing problems, and help-seeking related to their hearing. These perspectives were examined retrospectively and prospectively in the weeks leading up to and following each woman's first audiologic appointment. While the focus was on these women as the core participants, a secondary purpose of the research was to explore the perspectives of a close family member and the audiologist who assessed the core participants. In particular, the audiologic appointment was a focal point at which the interaction between each woman as a patient and the audiologist who evaluated her hearing for the first time was explored. As highlighted in the previous chapter, several aspects of the study design are unique. Very little information on the psychosocial aspects of hearing loss has been gained prospectively. The views of significant others are seldom solicited in studies. There are few studies that focus on the perspectives of women, and none of which I'm aware that have looked at help-seeking for hearing problems by older women. Moreover, research on the clinician-patient interaction has been almost non-existent in audiology. For these reasons, it was anticipated that this study would be important for the potentially significant 81 contribution it could make to the knowledge base. The following section presents the rationale for choosing a qualitative approach, and in particular grounded theory methodology. Qualitative Methodology The landscape of health research is changing, reflecting fundamental shifts in philosophy on the part of health practitioners, researchers, and policy-makers. Two such shifts that have occurred are: 1) in how health is defined and measured, and 2) in how the relationship between researcher and research participants is viewed. Perhaps the greatest barometer of change in how health is defined and measured is the development of the International Classification of Functioning and Disability, the ICIDH-2, by the WHO, as was described in the previous chapter (ICIDH-2, 1999). The ICIDH-2 represents a fundamental paradigm shift away from an earlier emphasis on "disease" (WHO, 1980) and toward a bio-psycho-social model of health that recognizes the importance of contextual and environmental determinants of health and the dynamic relationship between the individual and her/his social and physical environment. The ICIDH-2's focus on health rather than disease, function rather than dysfunction, and the person-in-context rather than the disease in isolation reflects both a new social understanding of disability and a response to changes in the provision of health care. It may be hypothesized that there is a parallel relationship between these changes and the increasing use of qualitative methods in research. The second shift listed above is the recent recognition of the voice of the patient as a health research participant. Research in the social sciences has for some time employed qualitative methods to listen to the voice of those most familiar with the subject matter at hand.1 In contrast, the influence of biomedicine, 1 For example, "action research" in its various forms has been implemented for many decades in the social sciences, often employing qualitative methodology (Reason, 1994). Action research is defined as social research where researchers and stakeholders define the research question and cogenerate relevant knowledge toward implementing social change (Greenwood & Levin, 1998). with its emphasis on positivism, technology, and treatment intervention has prevailed in health research. Only recently has the perspective of the research participant, the recipient of health care services, been given voice, assisted in part by qualitative research methods (Miller & Crabtree, 1994). Qualitative research is also termed "naturalistic inquiry" (Denzin & Lincoln, 1994, p. 4). It is well suited to uncover relationships between individuals and their ecological contexts because it accounts for the influence of environment rather than controlling for its effect in study design, data collection or analysis, as would be the case in quantitative research. Qualitative research takes many forms, most of which have the following features in common (as outlined by Miles & Huberman, 1994). The research involves intense and/or prolonged contact with a "field" or life situation, typically an "everyday" life situation of individuals, groups, societies or organizations. Using an inductive rather than a deductive approach, the researcher attempts to capture the perspective of local actors "from the inside" and explicate the ways people in particular settings come to understand, account or, take action and otherwise manage their day-to-day situations. Little standardized instrumentation is used to gather data, with the researcher being the main "measurement device" in the study. Finally, much of the data and the product of most analysis consist of words, rather than the graphs and numbers common in quantitative research. As the literature review in Chapter 2 noted, very few qualitative studies have been undertaken in audiology. Not coincidentally, such studies have been conducted by researchers and practitioners who advocate an ecological approach to the practice and study of audiology. Such an approach considers the person-in-context, where contextual factors may include interpersonal, organizational, and societal influences. Examples of audiology research groups engaged in qualitative study have been mentioned in the review of the literature: Hetu, Getty, Gagne, and colleagues in Montreal (Hetu, Getty & Waridel, 1994; Hetu, Jones & Getty, 1993; Hetu, Riverin, Lalande, & St.Cyr, 1988) and Hallberg and associates in Sweden (Hallberg & Carlsson, 1991; Hallberg & Jansson, 1996). The methodology of the Montreal group has concentrated on in-depth interviews and focus groups using primarily a phenomenological approach to describe the meaning of hearing loss in the lives of participants. Hallberg and colleagues have adopted a grounded theory approach and have combined qualitative with quantitative methods. The choice of research strategy among the various forms of qualitative approaches is determined by the purpose of the study, the nature of the research questions, and the resources available to the researcher (Morse, 1994). For example, phenomenology would be the method of choice if the research question concerns the meaning of a phenomenon, or the essence of the experience. If the question concerns describing the nature of a phenomenon, then answers are best obtained using ethnography. If the questions concern an experience and the phenomenon in question is a process, then the method of choice is grounded theory (Morse, 1994). The next section explains the rationale for the choice of grounded theory in the present study. Grounded Theory Grounded theory is a qualitative methodology that uses systematic and well-defined procedures to develop an inductively derived theory about a phenomenon (Strauss & Corbin, 1990). In grounded theory methodology, data collection, analysis and eventual theory stand in close relationship to one another (Strauss & Corbin, 1998). A theory is a set of interrelated concepts, definitions, and propositions that present a systematic view of events or situations by specifying relations among variables, in order to explain and predict the events or situations (Glanz & Rimer, 1995). With a grounded theory approach, the researcher does not begin a study with a preconceived theory in mind, but rather allows theory to emerge from the data. Grounding concepts in data is a main feature of this method; the theory emerges from the data and the data is used to illustrate, or ground, the theory (Strauss & Corbin, 1998). In the development of such theory, there is a continuous interplay between the data collected and its 84 analysis; thus, grounded theory methodology is often referred to as the "constant comparative method" (Strauss & Corbin, 1998). The development of grounded theory was influenced by the theoretical traditions of sociology and social psychology, in particular the symbolic interactionist tradition of the Chicago School of Sociology, which itself arose out of the philosophical perspective of pragmatism (Mead, 1934). The theory of symbolic interactionism has three basic premises. First, humans act toward objects (objects may be concrete things or abstract ideas) and other humans on the basis of the meanings that these objects hold for them.2 Second, these meanings arise out of social interaction. Third, meanings are not fixed but may be modified by humans as they interact with and interpret objects. This perspective also includes points raised in the previous chapter: 1) persons are actors who take an active role in responding to situations; 2) the various "roles" persons play, or "careers" they move through, differ in different social settings; 3) "careers" have two sides to them, the public, accessible side and the private, internal side that is linked to images of self and identity. As Strauss (1969) asserts, the central object to be negotiated in interaction is identity of self. In summary, symbolic interactionism focuses on interaction; interaction is a process; a process is by definition dynamic, not static; and central to the evolving meaning in interaction is the concept of identity. As the literature on presbycusis reveals, the psychosocial effects of presbycusis and individuals' responses to these effects hinge on issues that also involve interaction, process, and identity. Grounded theory, with its roots in symbolic interactionism, is well suited to the present study of hearing loss over time as it relates to initial help-seeking in older women. The quotation below (from Fife, 1994) is about symbolic interactionism in the conceptualization of meaning in illness. By substituting "hearing loss" for "life-threatening illness" it may be seen how this approach and 2 One definition of "meaning" is an individual's perception of the potential significance of an event for the self and one's plan of action (Fife, 1994). grounded theory methodology may yield comparable benefits in the study of presbycusis: Symbolic interactionism.. .links the perception of self, the social context, cognition and behavior so that the role of meaning in persons' responses to the disruption of life that results from the occurrence of a life threatening illness can be more clearly understood. It provides the framework for understanding the reciprocity between the individual's internal response to the crisis of a life-threatening disease and the ways in which that response is modified by interaction with the social world. Subsequently, based on this theoretical perspective there is the possibility that the meaning persons associate with their illness and its treatment could serve as a clinical marker, and provide some indication of the quality of adaptation persons will be able to achieve, (p. 316) Grounded theory is a way of thinking about and conceptualizing data that is particularly suited to experience-oriented and process-oriented research questions (Morse, 1994). A grounded theory approach is often the method of choice in studies of chronic health conditions. In fact, it was through Glaser and Strauss's substantive research of chronic illness and dying that many of the tenets of grounded theory were developed and applied by its authors (Glaser & Strauss, 1975). Table 3.1 provides a synopsis of the rationale for using grounded theory in this investigation. Grounded theory emerged as the methodology best suited to this study as a result of the following factors: the nature of the research area, the research questions, and the researcher's interest not only in describing phenomena but in developing substantive theory to explain those phenomena. 86 Table 3.1 Rationale for Using a Grounded Theory Approach in the Present Study Conditions Consistent with Grounded Theory The Present Study "Process" questions: experience over time or change Explores the experience of presbycusis and help-seeking over time Interest in patterns of action and interaction between and among various types of social units (i.e., "actors") Study of presbycusis involves study of patterns of interaction; perspective of family member and audiologist are included Data sources include interviews, participant observation, journals, fieldnotes, and memos Data sources are interviews, participant observation, fieldnotes and memos Emphasis on substantive theory development arising from substantive research topic Interest in building theory from substantive data to explain research phenomena for population studied The Study Design The following discussion of study design covers three main topics: the sample and setting, how participants were recruited, and the nature of the data. The Sample and the Setting The total sample in this study included older women (the participants), one close family member for each woman, and the assessing audiologist, who was the same for all women. Participants The core sample comprised women from an urban centre on Vancouver Island, British Columbia. Anticipating that each participant would have a unique story to tell 87 about her hearing, I aimed to reduce the effect of confounding variables on my results by choosing as homogenous a sample as possible. My sampling strategy included setting specific criteria for participation in this study. These inclusion criteria were that each participant 1) be age 55 years or older; 2) have no prior hearing aid experience; 3) have had no hearing test (conducted by an audiologist) within at least five years; 4) have English as her first language and 5) live independently in the community, preferably with at least one other person. Each woman selected 6) was to have no serious medical condition or cognitive impairment that would affect her participation in the study. The final criterion was that, 7) according to either the referring physician or my judgment based on screening questions, the reason for booking the appointment with the audiologist was for assessment of presbycusis uncomplicated by other significant symptoms such as severe tinnitus, dizziness, or middle-ear problems. In total, seven women participated in this study. These seven women ranged in age from 72 to 82 years. Three of the women lived with their husbands, and four lived alone. None of the women had ever had her hearing tested by an audiologist, except one who had had an audiologic assessment seven years previously and a recent hearing screening test by her ENT. One other woman had also had her hearing screened by an ENT two to three years previously. Family Members Six of the seven participants gave their consent for me to interview a family member. Thus, in addition to the core sample of seven older women, I interviewed three husbands and three daughters. Although I broached the topic of interviewing a family member over several interviews with one participant, she continually stressed that her family who lived nearby (son and granddaughters) were "too busy" to be available. 88 The Audiologist and the Audiology Setting I also collected data from the audiologist who assessed the seven participants. Specifically, I had a fifteen to twenty minute interview with the audiologist no later than 24 hours after her assessment of each participant. There were three criteria that guided my search for an appropriate audiology setting for this study. The first was dictated by research design: in order to have the opportunity to contact and interview participants twice before their audiology appointment, I needed to work with a setting having at least a two-week waiting period for assessments. Second, I required a setting having a sufficient caseload of patients with presbycusis to permit data collection within a reasonable time (six to eight months). Third, I wished to work with an audiologist who provides high quality service and who has had many years of experience working with persons who have presbycusis. The audiologist chosen for this study has had her own sole-charge private practice at the same location for over a decade. She is a clinical faculty member at the University of British Columbia (UBC) School of Audiology and Speech Sciences for which she supervises audiology student interns. As such she is considered on the basis of observation by U B C faculty to provide exemplary audiologic service. I chose to work with only one audiologist to maintain consistency of testing across participants. Audiology Service Delivery in British Columbia Audiologic services to adults in the province of British Columbia, Canada, are provided through public and private facilities. However, because the priorities of public facilities are testing children, difficult-to-test populations, and advanced diagnostic testing, the vast majority of persons with presbycusis are seen by private-practice audiologists. It is not necessary to have a referral from a physician to make an appointment. There is no public funding of hearing aids for adults (except for those on social assistance). Audiologic assessments are provided at no charge if conducted at a public facility; there is no public funding for assessments conducted at a private facility. The audiologist who participated in the present study charged $50 for a one-hour assessment (a fee similar to that charged 89 by other private-practice faculties). If the assessment resulted in a hearing aid fitting, this charge was applied as a credit against the purchase price of the aid. Recruiting Participants I worked with the booking secretary at the audiology setting to recruit participants. I discovered that this secretary obtains few details of the patient's background information when she books appointments. She usually asks only the patient's name, phone number, and whether the appointment is a first visit. I was in telephone contact with the secretary every few days to check on appointments that had just been booked to determine if there were new potentially eligible participants. The secretary telephoned prospective recruits, and asked for their permission for me to contact them. In my subsequent telephone conversation with the women who agreed to allow me to phone them, I explained the purpose and value of my study and asked screening questions to determine if they met the study criteria.3 In total, 15 women were telephoned. Five of these women did not meet one or more study criteria (i.e., two were already hearing aid users, one was too young, one appeared to have significant tinnitus, and one was in poor health). Three women declined to participate due to lack of interest (1), concerns about confidentiality (1), and because a spouse was quite ill (1). I arranged a first interview with each of the seven participants who took part in this study after determining that each of these women met the criteria for inclusion in the study, expressed genuine interest in participating, and had, I felt, a clear and realistic understanding of what would be expected of her as a participant in the study. A l l seven participants completed the study. Only two of the seven participants accepted the $125 honorarium that was offered at the end of the last interview. 3 After briefly explaining the study and my background, I asked each potential participant questions in the following order to determine whether study criteria were met: 1) are you 55 years or older? 2) have you ever tried a hearing aid? 3) have you ever had your hearing tested by an audiologist? (if yes, I tried to determine whether this was more than 5 years ago) 4) do other family members live with you ? (I phrased this question about living situation thus to be as non-threatening as possible re: security); 5) how is your health? (to determine if there were other medical conditions that would preclude participating). 6) My last question was to confirm that English was the first language. 90 Nature of the Data The data collected were of several forms: semi-structured interviews with the participants, their family members and the participating audiologist, participant observation of each audiology assessment, and documents. The documents themselves took several forms: a written journal and questionnaire completed by each participant, the clinic "client intake history" form completed by each participant at the audiologist's office prior to the hearing assessment, the audiologic test results provided by the audiology facility, and my notes and memos related to the research project. A description of each of these forms of data collection follows. Interviews Much effort in the research planning phase was devoted to designing the schedule and structure of interviews. Interviews were the main form of data collection, thus I was careful that they be conducted in a fashion that would facilitate information gathering while allowing me to establish and maintain rapport with each participant. Interviews also offered me the opportunity to informally assess a participant's or family member's hearing and her/his communication style and strategies. The first two participants in this study, Cheryl and Phyllis4 were at first considered pilot participants with whom the study design was tested. The only change made to the design related to the issue of validity, which will be described later in this section. Cheryl and Phyllis were therefore retained as participants. Appendix A outlines the schedule of interviews for each participant, family member and the audiologist. Interviews 1 and 2 were held before the participant's audiology appointment. Interview 3 was the "post- audio" interview, conducted for most participants immediately after the audiology assessment. I continued interviewing a participant until I felt that the data collection process was "saturated", that is, the same 4 To maintain confidentiality in writing about this study, I assigned each participant a pseudonym: PI is Cheryl, P2 is Phyllis, P3 is Janet, P4 is Marjorie, P5 is Frances, P6 is Louise, and P7 is Betty. 91 information was being repeated frequently and no new information was forthcoming on topics initiated by the participant or me. This was corifirmed in coding and analyzing the data, in that at a certain point no new properties, dimensions or relationships emerged (Glaser, 1978, p. 53; Strauss & Corbin, 1998, p. 136). For most of the women, this point was reached by the end of interview 4; for two women, a fifth interview was held. In total, 29 interviews were held with the core sample of seven women, six interviews with family members, and seven interviews with the participating audiologist. The interviews spanned the time from the first meeting with the participant which took place between six and eighteen days before the audiology appointment, to the final interview which took place between twenty and forty-two days after the audiology appointment. Interviews with participants Interviews with participants were held in their homes (in either the dining area or the living room) and lasted between forty-five and ninety minutes. The exception was interview 3, held right after the audiology appointment, which took place in a small restaurant next to the audiology clinic and lasted about thirty minutes. A l l interviews were semi-structured. An interview guide for each interview was prepared (discussed further below). A l l interviews were audiotaped and I subsequently transcribed them in detail. Topics within and across interviews ranged from general to more specific and from less personal to more intimate. Beyond the first interview, the questions drew upon data from prior interviews whenever I wished to expand upon a participant's earlier comment or when I felt that gaps remained in my understanding of a participant's previous responses to the same or similar questions. I revisited many questions from interview to interview to confirm or add to participants' original responses. To enable data analysis to accompany data collection, I arranged as wide a time as possible between interviews so that I could thoughtfully listen to each audiotape, and in most cases analyze a first-run transcription of it before the next interview. 92 The interview guides are provided in Appendices B to F. Interview 1 (Appendix B) focused on the woman's personal background, lifestyle and activities, her description of her health, her hearing and hearing problems, her coping style and strategies, and her experience with other health professionals. One important question raised was how the participant's appointment came to be booked at this point in time. Interview 2 (Appendix C) explored this question further, and also looked at the participant's expectations and hopes for the audiologic assessment, her knowledge of audiology and audiologists, and her knowledge of hearing aids. Interview 3 addressed questions related to the audiology assessment: reactions to the test procedures and the test results, understanding of test results, how well expectations were met, and overall impressions of the test (see Appendix D). This was the most "structured" of the interviews, in that my goal in this interview was to explore specific issues related to the just-completed assessment. I felt it was important to talk with the women as closely as possible to the time of the "event", yet I did not wish to tire participants with a lengthy interview on the heels of an hour-long assessment. In two cases, a participant's husband accompanied her to the test and sat in on the audiologist's post-test consultation as well as my post-test interview, so that I was able to direct the occasional off-the-cuff question to him. Interviews 4 (and 5 when held) were held roughly from one to six weeks "post-test". In general, in these interviews participants were asked once again to give their interpretation of their test results and their feelings about those results. We also discussed what action, if any, they intended to take from that point on and the reasoning behind those intentions. Beyond these questions, the topics covered in interviews 4 and 5 were more varied across participants than in other interviews because, by this point, much ground had been covered. Participants by this point gravitated toward topics they wanted to expand upon, and I revisited certain questions that "nagged" at me for elaboration or clarification, because the answers previously obtained were either unclear, incomplete or contradictory. There were two features that the last interview had in common for all participants. First, I made a point to ask each woman about the experience of taking part in this study. Second, in the final forty minutes or so of the final interview, I asked each 93 woman to provide feedback to me on a list of specific points that reflected, to that date, my analysis arid interpretation of their responses to my questions. This synopsis was organized into four sections to facilitate discussion: participants' thoughts on their hearing, booking the hearing test, the hearing test results, and hearing aids. The request for feedback to the synopsis was the only change from the original pilot design (mentioned above). The original plan had been for each participant to be given the opportunity to review the transcript of each of her interviews to verify their accuracy. The pilot participants, Cheryl and Phyllis, were offered this opportunity but declined, stating that they felt that some of their answers to questions may have changed over the course of the study and that they didn't feel the need to go over the interviews in written form. Therefore, the decision was made to review a synopsis of the analysis and interpretation of earlier responses in the final interview rather than asking participants to review transcripts. This modified procedure was in fact used for Cheryl and Phyllis as for the subsequent participants. There were several advantages to conducting interviews at participants' homes. Speaking to these women on their own "turf seemed likely to put them more at ease, and placed them in a familiar and preferential "acoustic" setting for talking to an unfamiliar communication partner. For my part, I welcomed the opportunity to observe where these women lived and to thereby gain insights into their lifestyle and daily acoustic environments. It also afforded me the chance to observe the interaction between that participant and other family members, who were not present for interviews with the participant, but who frequently passed through the room where we were talking. As soon as possible following each interview, I wrote field notes on my observations (described under "documents" below). I felt there was a sense of closure with each participant at our final interview as my questioning and our interaction drew to a close. I requested the opportunity to keep in contact with each woman informally following the study, in order to follow up on any questions I might still have arising out of my analysis of the data. They encouraged me to keep in touch. I subsequently contacted two of the women by telephone for a brief 94 conversation regarding their follow up action toward getting a hearing aid, and I documented the content of these calls in my field notes. Interviews with family members Interviews with family members were also semi-structured and audiotaped and lasted between twenty-five and forty minutes. My goal for these interviews was to gain the perspective of an important "other" on several key questions I had already posed to participants, while also offering them the opportunity to reflect openly on any aspect of their experience with their wife's or mother's hearing (see Appendix E). I was most interested in their description of their mother's or wife's hearing, how long they had noticed problems, what kinds of problems they noted, how they reported their wife or mother to be coping with hearing challenges, and their impression of how the audiology appointment came to be booked. Interviews with family members took place at participants' homes (with one exception where I interviewed a daughter in her own home). I did try to make it clear that I wished to speak alone with family members; nevertheless, participants were present for a portion of these interviews, providing refreshments or socializing. This time together was an opportunity for me to observe the interaction between a participant and her spouse or daughter. Notwithstanding this time together as a "threesome", I felt overall that I had a sufficient period of time alone with family members to ask them pertinent questions with participants out of "earshot". Moreover, I did not feel that family members censored their answers to my questions because their wife or mother was present. Interviews with the audiologist The interview I had with the participating audiologist following her assessment of each participant lasted fifteen to twenty minutes and focused on her perspective of the patient (participant) she had just tested and the assessment process. It explored her perceptions of the patient's expectations for the test and how well she believed they were 95 met, her own expectations, and other impressions (see Appendix F). These interviews were held in the audiologist's office, except for one interview conducted by telephone the day after the assessment because scheduling difficulties on the test day did not allow the audiologist to take time for the interview any sooner. This telephone interview was the only interview not audio-taped; instead, detailed notes of the audiologist's comments were taken and later typed and subjected to analysis. Participant Observation Each participant gave her consent for me to attend, observe, and audiotape her audiology appointment from start to finish. The audiologist also agreed to this. Each audiologic assessment took on average one hour to complete. To be as unobtrusive as possible, I arrived before the appointment to set up the tape recorder on the test side of the sound booth next to the monitor speaker on the audiometer so that both the participant's and the audiologist's speech could be recorded. (For my first participant, I placed the tape recorder on the participant's side of the booth, which had the disadvantage of recording only the participant's voice during the portion of the testing conducted under earphones). I sat as inconspicuously as I could on the audiologist's side of the booth in the corner nearest the door on a short stool, without compromising my view of the participant during her assessment. Figure 3.1 illustrates the set-up for the audiologic assessment, participant observation and recording. I talked with the audiologist only when she initiated conversation with me as she conducted the testing. I remained as observer on the audiologist's side of the booth when the audiologist moved into the booth to be with the patient for the post-test counselling session. Throughout the assessment, I jotted down notes of observations regarding aspects of the test procedures, environment, participant behaviour, or the interaction between audiologist and participant/patient that would not be captured on the audiotaped recording. These were later reviewed and coded as data. Some of these observations also generated questions that I posed to a participant after her assessment; for example, after 96 6 ft-Door 12 ft 6 in Window Door • © AudiometeJ ® Figure 3.1 Diagram of Set-up for Participant Observation and Recording of Audiologic Assessment P = Participant/patient A = Audiologist R = Researcher T = Tape recorder Dimensions of sound-attenuating booth are approximate. observing that Louise whispered her responses to the audiologist's questions during her test, I made a research note of this and asked her about this particular behaviour in our next interview. Documents Documents in this study were of five forms: a journal with regular entries by each participant, a questionnaire completed by each participant, a "client intake history" form completed by each participant, audiologic results on each participant, and the notes of the researcher. 97 Journal As part of the study, each of the seven core participants was asked to keep a daily journal of thoughts related to her hearing. A small, hardcover, lined journal was provided to each participant, along with an instruction sheet (Appendix G) that I reviewed with each woman at the end of the first interview. Each time I saw a participant, I always asked her how her journal writing was progressing; occasionally this triggered the participant to read excerpts or make comments based on what had been written. The journals were returned to me at the last interview and subsequently analyzed. The "Communication Profile for the Hearing Impaired" Each participant was also asked to complete the Communication Profile for the Hearing-Impaired (CPHI) at her convenience. The CPHI is a lengthy self-assessment questionnaire (Demorest & Erdman, 1987) with 145 items in three parts. The goal of the CPHI is to describe the communication performance, communication environment, communication strategies, and personal adjustment of hard-of-hearing adults, as well as the hard-of-hearing person's views of the attitudes and behavior of others toward him or her. My purpose in having the women complete this questionnaire was to compare one quantitative measure of the psychosocial issues being explored with the qualitative findings. The CPHI was chosen from a roster of dozens of self-assessment instruments that are available because of its relative popularity and purported psychometric refinement. The CPHI was given to participants at the end of the first interview and returned to me in person by the last interview. "Client Intake History" form The participating audiologist utilized this one page form (see Appendix H) to obtain information about the patient's history in terms of general medical and specific ear-related problems, and to learn a bit about the patient's hearing difficulties. Question 98 #6: "Whose idea was it to have an audiological evaluation?" was of particular relevance to this study. Each patient (participant) completed this form in the audiologist's waiting room before her appointment began. The audiologist used the form as a starting point in her discussion with the patient in the sound-booth, before starting the audiologic assessment. She made notes on this form of any relevant comments made by the patient relating to the questions on the form. The audiologist provided a copy of the completed form to me following each appointment. I was interested in seeing how well the written answers of participants on the form matched their comments to me in interviews or in their journals. Audiologic results The participating audiologist also made available to me copies of the audiologic results for each participant whose assessment I attended. These included the results of pure tone testing by air and bone conduction, speech recognition (reception) testing, word recognition testing in quiet or with background noise (with incorrect responses to each test word noted on the test form), and the results of a test to determine the patient's most comfortable and uncomfortable loudness level (see Appendix I for a sample of test results). The results of a test of middle ear function (tympanogram) were also included for two participants whose history suggested possible middle-ear dysfunction. The audiologist wrote comments about the patient's or her own subjective impressions during testing, e.g. "left ear sharper". A description of the hearing testing procedures is provided in Appendix J. A glossary of audiologic terms and information on how to read an audiogram is provided in Appendix K . Researcher's notes From the time I was notified by the audiology facility of my first potential participant, I kept a journal in which I recorded my thoughts, observations, and ideas related to this study. Many entries in the journal were related to matters relative to the interviews with participants, such as a description of the physical setting of the interview, 99 my observations of a participant's personality, how I felt the interview and our interaction went, and other points. These entries were considered to be data. Frequently, a participant made very interesting comments between the time the tape recorder was turned off and when I was at the door to leave. I tried to capture these comments in my journal as soon as possible (usually in my car before I drove away). I also made notes in my journal regarding aspects of or incidents in my own life, for example my mood or my baby daughter's teething that kept me up all night, which may have affected the data collection process. A second important set of "notes as data" were the notes I made on interview transcripts to mark communication breakdowns in conversation with a participant. Wherever I noted a breakdown (a significant proportion of which usually occurred just after I asked a question), I made a note in the margin of the transcript. These were not formally analyzed or tallied. Although I did not undertake a discourse analysis of these breakdowns, I considered them important because their occurrence or absence gave me a taste of a participant's ease or difficulty comprehending in a favourable Ustening environment. It was interesting to compare these breakdowns across participants (e.g., variation in number of breakdowns) and also to note how the observed breakdowns in conversation were sometimes at odds with a participant's report of how well she heard one-on-one. This completes the description of the various data sources in this study. The next section describes how the data were analyzed. Analytic Method This section describes the data analysis procedures specific to grounded theory and how they were applied to the data collected in this study. The discussion begins with a description of the process of interview data transcription. Following this, an explanation of the coding process is provided using examples from the data of this study to illustrate the process. A discussion of memo-writing and how the findings are presented completes this section. 100 Data analysis consists of three forms of activity: data reduction,5 data display, and conclusion drawing or verification (Miles & Huberman, 1994, p. 10). While the discussion below will describe each of these three activities sequentially, it should be kept in mind that, except at the very start of data analysis, these activities are iterative and interactive, not isolated and sequential.6 The discussion begins with what I consider to be an important component of the activity of data reduction: transcribing the raw data. Transcribing Interviews The interview data collected comprised twenty-nine interviews with the seven participants of this study, resulting in approximately thirty-five hours of audiotaped recordings. Six interviews with family members of participants resulted in over three hours of audiotaped recordings, and seven interviews with the audiologist rendered another three hours of recording. In addition, each participant's audiologic assessment was audiotaped, resulting in seven more hours of taped recordings. In total, about forty-eight hours of audiotaped data were obtained. I personally transcribed all of the interview data and the audiologic assessment data, except for portions of two final interviews that were transcribed by an experienced transcriber (and which I subsequently checked). Transcribing is a tremendously time-consuming task, taking upwards of eight hours per hour of interview. This process, however, was personally very rewarding. As a student of speech and language sciences 5 Data reduction refers to the process of selecting, focusing, simplifying, abstracting, and transforming the data that appear in written-up field notes or transcriptions (Miles & Huberman, 1994, p. 10). 6 In the course of my doctoral studies, I researched the merits and drawbacks of utilizing a software program (such as NUD*IST) to analyze data. I decided not to employ qualitative software to assist in data analysis for two main reasons. First, the feedback I received from other researchers convinced me that there was a significant learning curve in becoming familiar with such software programs, and I decided that for my first major qualitative undertaking I would rather spend those months immersed in the data of my project rather than in learning about the software. Second, it was the opinion of the seasoned qualitative researchers with whom 1 spoke that each software program deals with data in a slightly different way, and I felt uncomfortable at the prospect of introducing an additional lens to "view", and possibly influence, the interpretation of the data. Instead, I worked through the data using only the computer programs available within my word processor (Word 2000), such as the search and retrieve command for easy access to particular codes in my analysis. 101 with a special interest in discourse analysis, I strove to preserve as much detail as possible of the verbal and non-verbal interaction in the interviews. At the same time, I was acutely aware of what may be lost in the process of transducing auditory data to a written form. Doing a detailed transcription of an interview allowed me to get a feel for the discourse style of a participant in conversation, and for characteristics of the breakdowns in conversation that resulted, at least in part, from the participant's hearing difficulty. It also provided some information on the reaction of, and strategies used by, a participant in response to conversational breakdowns. Transcriptions of the audiologic assessments revealed interesting discourse patterns in the interaction between the audiologist and the patient (participant).7 I adapted a notational schema for transcribing interviews that achieved the degree of detail desired, maintained quality and consistency, and resisted changes in text formatting. My notational system was extracted primarily from the work of Tannen (1984), and is described in Appendix L . The slow line-by-line consideration involved in detailed transcribing stimulated many questions and insights about the data. This process constituted my initial analysis of the data. I made frequent memos to myself on these thoughts, which helped to frame questions and talking points for subsequent interviews with the same or another participant. Undertaking the transcribing myself shortly after the interview truly immersed me in the data so that I felt I achieved a more intimate connection with the data of each interview than if I had received a final printed product transcribed by someone else. In qualitative inquiry, data analysis informs data collection (Strauss & Corbin, 1998). A first draft transcription was carried out as soon as possible following each interview. The focus of this first pass was on interview content that guided the next interview. This was especially important in looking at data from the first interview, because there were a number of questions about the hearing test that I wanted to pose or follow up on with a participant before her audiology appointment, and the second interview was the only opportunity to do so. As well, transcribing the interview data while 7 It was beyond the scope of this study to conduct a detailed study of the discourse between the audiologist and patient (participant). In the future, I hope to undertake a formal discourse analysis of the audiologic assessments, particularly the pre-test and post-test portions, to study the relatively uncharted area of clinician-patient discourse. 102 the experience and flavour of collecting it were still fresh in my mind may have led to theoretical insights that may not have emerged otherwise (Patton, 1987, p. 140). Coding of Interviews and Documents In grounded theory, there is an interplay between induction and deduction in data analysis and theory building. Induction is utilized in working with the data, the words of study participants, to generate labels, or what are referred to as code names for all ideas present in the data. These code names, or labels, will eventually, through the process of constant comparison become the "concepts": the building blocks of theory. In the words of Strauss and Corbin (1998, p.103), a concept is "an abstract representation of an event, object, or action/interaction that a researcher identifies as being significant in the data." The process by which concepts are identified is called "open coding"(Glaser, 1978, p. 56), and the tags assigned to concepts are called "codes". The concepts or codes identified in data analysis are abstract to varying degrees. Grounded theory organizes coding into three levels of abstraction. "Categories" represent the most abstract level of coding, the most central ideas that emerge from the data. Categories answer the big question "What is going on here?" The next level is less abstract, and involves coding data according to "properties". Properties are the general or specific characteristics or attributes of categories. The third and least abstract level of coding is by "types, circumstances, and conditions" otherwise termed "dimensions". Dimensions are characteristics of a property, specifying the range along which a property may vary. As coding proceeds, the grounded theorist makes constant comparisons and asks questions in order to classify concepts as similar or different and to relate concepts to each other. The process by which concepts or codes are organized into different levels of abstraction is termed "axial coding" (Strauss & Corbin, 1998, p. 123). As concepts accumulate and are compared and related to each other, the analyst continues to work with the raw data to see how it fits with the emerging interpretation or classification of concepts. As concepts are named, sorted and classified a "theoretical (conceptual) 103 framework" begins to take shape; over time these concepts (whose names may evolve with the analysis) are grouped into categories, properties and dimensions. This may be viewed as a deductive component of data analysis, in that the theory that is evolving is guided by the ongoing analysis of the raw data, as well as by the researcher's experience with other research and theories. A third type of coding, termed "selective coding" (Strauss & Corbin, 1998, p. 143), involves the process of integrating and refining the theory. Open and axial coding predominate in the early stages of analysis; later in the analysis, selective coding helps to integrate theory by relating the categories to each other and by identifying one central or core idea of the research, termed the "core category" (Strauss & Corbin, 1998, p. 146). As mentioned earlier, data analysis is an iterative and interactive process involving all three types of coding. The analysis continues right into the final stages of writing up the theory. As Glaser (1978, p.7) states, "Grounded theory assumes that part of the method, itself, is the writing of the theory. The way data is coded, ideas are memoed, and memos are sorted are all partly focused on designing and facilitating the writing of the theory." To summarize then, open coding labels categories, properties and dimensions, which represent the most to least abstract concepts identified in the data, respectively. Axial coding reassembles the data broken down in open coding by identifying linkages among categories and their properties and dimensions. Selective coding comes into play later in the analysis as the grounded theory starts to crystallize and a central or core category is identified for the purpose of integrating and concisely explaining the grounded theory. Data Analysis in the Context of this Study The voluminous amount of data collected and my initial lack of familiarity with the coding processes in grounded theory development resulted in two distinct stages in my analysis of the data from this study. These two stages represent two related approaches to the coding process. The first stage covered the time period from before the end of data collection in April 1998 until September 1998. These months were devoted primarily to 104 the activity of open coding. The first identification of concepts emerged as I transcribed the data. I kept longhand notes of concepts and insights that arose during transcribing. The in-depth analysis, or fracturing of the data, began in earnest with the printed "hard" copy of a transcript, starting with the transcripts of interviews. The unit of analysis could be a word, phrase, sentence, or entire paragraph. Wide margins on either side of each page of transcripts enabled me to write code names and notes in the margins. Two and three letter codes were chosen to be semantically related to the concepts they represented, for ease of identification. Coloured pencils were used to label code names, with the text that corresponded to a particular code underlined in the same colour for easy reference. Codes that were related conceptually to each other were coded in the same colour. For example, all references to "hearing aids", code HA, were coded in dark green pencil, even though a range of related codes applied (e.g., "negative aspects of hearing aids" " H A N " , and "hearing aids as objects of mystery" " H A M " were both coded in dark green). Operational definitions of codes were not explicitly set from the outset, but were defined over the course of data analysis as more and more transcripts were coded. Descriptive codes that required little interpretation were the first to be assigned (Miles & Huberman, 1994, p.57). Codes that are more inferential and explanatory tended to emerge later in the process of open coding. For example, the code H A above is descriptive, while a code such as H U (influence of husband) may be considered inferential. Occasionally a code looked slightly different operationally in different parts of the data. I resisted the urge to tie down a definition prematurely, but let both definitions sit and eventually the data and analysis resolved the question of what code definition was more appropriate. For example, the code CO, was applied to text that reflected a participant's "concern for the opinions of others" and to text that reflected a participant's "concern for what others think of her". I was unable at first to decide whether these labels represented the same idea or two related but different ideas. As more data were analyzed, these two definitions merged into one ("concern for what others think of her") that seemed to fit all examples of the code CO. A l l the data from the study, that is, all transcripts of interviews (including those with family members and the audiologist), audiologic assessments, participant journals* and my own journal and notes, were coded as described above. A list of about fifty codes 105 was generated from the analysis. Extensive memos (see later section for more detail) on the codes were kept, with quotes by different participants documented to link particular concepts with the data. I created a profile of each participant that listed the particular codes that emerged from the data for that participant. This profile illustrated how the concept associated with a particular code manifested itself for a particular participant. I moved into the process of axial coding as links between codes emerged. Throughout my analysis I explored how a concept looked similar or different between and across participants. This is the "constant comparison method" associated with grounded theory analysis. This first stage of in-depth analysis was largely conducted away from my computer. Particularly as axial coding proceeded, I found it very helpful to draw diagrams, especially flow charts, to show the relationships between and among concepts. Diagrams are a form of visual memo (Strauss & Corbin, 1998, p. 217) that I used extensively throughout all three phases of coding (open, axial, and selective). In this first stage of analysis, I used diagrams to explore how the data addressed the research questions for individual participants. I found the creation of flow charts particularly helpful to model a participant's evolving relationship with her hearing and the help-seeking process. By the fall of 1998,1 had accumulated copious notes and diagrams, mostly on paper, to accompany my colourfulry coded transcripts. However, I was unsure how to proceed to develop theory from the many codes and linkages I had amassed. At this point, I took a break from data analysis and turned my attention to writing a first draft of my literature review. In the spring of 1999,1 began what I call the second stage of my data analysis, which was facilitated by my introduction to a very methodical, step by step approach to grounded theory data analysis outlined by Grams (1994). I started anew with my data, using a more microscopic lens than in my first phase of analysis. This involved, first, a word-by-word analysis of the first five or so pages of the first interview of this study. Each idea represented in a line of text was given a code name to describe "what this is about". Figure 3.2 illustrates this process of open coding, using a segment from the first interview Figure 3. 2 Example of Open Coding (Phrases in superscript font represent conceptual codes) C: /we i n c , u d e s h u , n c o r a m e n t s don't socialize all that much anvwav h e a r i n s c h a l l e n ^ Hfestylcreduced socializing ^ ^ ^ y ^ persona, expenence y Q u ^ ^ ladies I talk to in the laundry room or something else some of them arc.have very quiet voices e x t e r n a l a t t r i b u t i o n s I have to ask them to repeat c o p i n g s t r a , e g y . . and I jus t d o w n p l a y i n g h e a r i n g c h a l l e n g e s tell them I don't think I'm hearing that well {laughs} anymore c o p m g *rategy(u-huh) and know they're very Understanding c a u s e they're a l l m y . . m y ^ c o h o r t effect: comparison with cohort members I: {laughing} they probably have the hearing (that's right) ..the same C: Exactly c o h o r t e f f e c t . . . .one problem or another c o h o r t m e m b e r s h a v e s i m i l a r p r o b l e m s . . .anyway this is an over fifty-five b u i l d i n g c o h o r t e f f e c t : s u r r o u n d e d b y c o h o r t m e m b e r s ; l i f e s t y l e I: U-huh{C laughs}..u-huh..u-huh C' SO they're Very Understanding response o f friends/cohorts: c ° h ° r t members relate to hearing problem I: U-huh....adt' .getting that's.that's the main., that's your .strategy then I guess is just (yeah) to tell people..that..(yeah) c o p ' n g s t r a t e g y there's no other.. .d'y when you miss something do you ever ask people to...(repeat it?) c o p i n g s t r a t e g y..rephrase something or (repeat it?) e o ^ " ^ repeat it (yeah) or ..yeah C: I jUSt S p l a y i n g hearing challenge g a y copingstrategy ^  ( u _ h u h ) Q r J d i d n > t you c ° P ' n 8 s t r a t e 8 y o r s o m e th ing like that it's a (u-huh) and (1 sec) people are pretty good you know r e s p o n s e o f W e n d s / c o h ° r t s {C small laugh} I: And that always does the trick Q. QJj yeS C 0 P' n ^ s t r a t e Sy : works in her view I: Yeah..good C: Yeah (u-huh) I mean these are friends f H e n d s v s s t r a n g e i s s o you're not going to ..(u-huh)..friends and family they're not going to be irritated by that r e s p o n s e o f family/friends/cohorts 107 with participant Cheryl (1-1-18). In this example, the conceptual codes follow (in superscript font) the idea they represent. The analysis of these first pages resulted in a list of 46 codes or concepts, which did not at this point break down easily into categories, properties or dimensions. I continued my line-by-line analysis for the entire first interview.. As suggested by Grams (1994, p. 13), I did not belabour the code names adopted at this point; the primary goal was to capture the meaning of the word, phrase or sentence as it was used by the participant. Open coding of the section of transcript in Figure 3.2 resulted in the following list of codes: includes hu in comments, downplaying hearing challenges, lifestyle: reduced socializing, personal experience, external attributions, coping strategy, response of friends/cohort members, cohort effect: comparison with cohort members, cohort members have similar problems, cohort effect: surrounded by cohort members; lifestyle, response of friends/cohort members: cohort members relate to hearing problem, response of friends/cohort members, coping strategy: works in her view, friends vs strangers, response of family/friends/cohort. It may be seen that most of the codes from this piece of text address two main ideas or concepts: "cohort members" and "coping strategies". These codes were all included in a preliminary framework that was developed from all the codes labeled in this first interview. The process of axial coding began to group some codes with others. The framework reflected this grouping of codes into categories, properties and dimensions. In some cases, codes seemed to fit equally well in two or more categories; in those cases, they were included in both categories and marked with an asterisk to indicate that they were present in more than one part of the framework. The framework at this point consisted of nine pages. One small portion of this framework is shown in Table 3.2 to illustrate the process of axial coding and how codes relating to the idea of "cohort members" (as shown in Figure 3.2; italicized in Table 3.2) were incorporated into the beginning framework. The concepts related to "cohort members" are retained, even though some of these code names evolved to other names as more of the transcripts were analyzed. In Table 3.2, the terms categories, properties, and dimensions are put in parenthesis to indicate that, at this See Appendix L for the transcription key. 108 early stage, this is how the concepts listed were classified; the final grouping of categories, properties and dimensions in the theoretical framework was very different. In the final theoretical framework, for example, the concept labeled "cohort effect" was ultimately renamed "cohort members/friends" and was no longer a category but a dimension of a property labeled "influences", which in turn was related to a category labeled "self-assessing". Other concepts that were classified as dimensions in the early stages of analysis ultimately became sub-dimensions. Not shown in Table 3.2 are the page numbers that were attached to code names to indicate where in the data an example of a particular code could be found. Table 3.2 Preliminary Theoretical Framework Illustrating Axial Coding. Italics highlight codes from Fig. 3.2 that relate to the idea 'cohort members' "Categories" "Properties" "Dimensions" COHORT EFFECT/INFLUENCE OF COHORT M E M B E R S comparison to cohort members not hearing as well as hu not "giving in" to h aids response of cohort members accepted P's hearing not surprised by P's hearing empathy by/with cohort members concern for what they think of her surrounded by cohort members hearing aids may hear similarly have similar problems if TV too loud if she spoke too loudly lives in "over 55" bldg cohorts may have an aid positive aspects 109 As more data were coded, ideas repeated themselves and new ideas surfaced less frequently. At this point it was possible to move from a kne-by-line analysis to sentence-by-sentence or even paragraph-by-paragraph analysis, with the analysis becoming a search for new ideas. Nevertheless, the analysis remained methodical until every interview, audiologic assessment, and journal was analyzed. Codes were grouped together as open and axial coding interacted in the Ongoing analysis. By the end of this second stage of analysis, the list of all of the concepts identified in the data was a hefty seventeen pages long. As I worked with this list, I realized that many of the ideas represented were variations of the same idea. For example, the following entries were among those listed under the category "approaches to coping", "trying to decipher key word", "asking for the 'last of that"', "watching faces closely", and "daughter says P uses visual cues a lot". In asking the question "What is this about?" I realized that these ideas were all part of a bigger idea that I labeled "piecing together". "Piecing together" became a dimension of the property "reacting to hearing challenges" and the ideas listed above became sub-dimensions. Later in the data analysis I reviewed the codes I had identified in the first stage of data analysis and was heartened to discover that there was great similarity and overlap in the codes generated in both phases of analysis. This corroboration of findings increased my confidence that my findings are well-grounded and defensible. Themes In the process of coding data, making links between concepts, articulating the findings, and building theory, it became evident that some concepts were repeated, with slight variation, and these concepts seemed to fit equally well into more than one category or property in the emerging framework. Such concepts are called themes (Morse & Field, 1995), as they tend to appear, with variations, in different parts of the theory. Themes tend to be abstract and thus usually emerge later in the data analysis process. I kept track of concepts that were thematic through memos ("relationship memos"), highlighting the thematic label in upper case letters in the memo for easy retrieval so that I could easily track how a certain theme was manifested in different properties or dimensions. 110 Figure 3.3 contains two excerpts drawn from my memo writing. These two memos illustrate how the theme of "control" looks different in two examples of analysis, the dimension "effect of aging" in the first example, and the dimension "relationship with husband" in the second example (with dimensions italicized in examples). Figure 3.3 Two Examples from Memos on the Theme "Control" June 15/99 Tliinking again about CONTROL, this time in relation to effect of aging. It seems to me that there is a perceived (conscious or not) relationship between aging and loss of control, and to the extent that one maintains control in one's life, then one is not aging, or is not displaying the effects of aging, loss of control being one of these effects. So, maybe because hearing loss and hearing aids are equated with getting older, then de facto, they are equated with a loss of control in one's life. July 28/ 99 P7's major trigger for booking the hearing test with the audiologist is her husband's hearing and their relationship (relationship with husband), and I wonder if she put off booking the test until she convinced him to also get his hearing tested and he agreed to go with her. Here the issue of CONTROL arises again. Writing Memos Throughout the data collection and analysis processes I wrote memos, both in longhand directly on interview transcripts and in a designated notebook, and, later in the analysis, on computer. Memos were of three categories: method memos, profile memos and relationship memos. Method memos documented the procedures followed and decisions made in analyzing the data, and the questions and dilemmas encountered in analysis. Profile memos discussed the data or interpretations thereof for a particular study participant. Relationship memos were the most numerous, especially later in the analysis, and were an important vehicle in the process of theory-building. They included insights on interpreting findings, making connections or links among concepts, and notes about I l l themes. An example of a method memo, a profile memo, and a relationship memo are illustrated in Figures 3.4, 3.5, and 3.6, respectively. Memos served three major purposes. First, they were an analytical device to record insights about the data and to help depict relationships among concepts. Second, memos provided a record of the analytical process; a way to document the stages of theory development. Third, later in the analysis, memos served as a springboard for writing about the findings and testing the theory. Figure 3.4 Example of Method Memo May 20/99 In analyzing the second five pages of PI-1,1 find that I sometimes have more than one code I want to attach to words or phrases. What I do in these situations is attach both codes and try not to get too hung up on how and what I name them at this point. Figure 3.5 Example of Profile Memo June 26/99 It occurs to me that for P2, it was really the system that failed her (see 2-4-8) by both first audiology test in 1990 saying that a hearing aid would not help and then her doctor delaying a referral for a hearing test on more than one visit, and all the time P tliinking she needed a hearing aid. Figure 3.6 Example of Relationship Memo June 24/99 What comes first: hi incorporated as part of one's identity and then it's easier to accept a ha (a facilitating factor in ha trial) OR the ha as instrumental in incorporating hi in one's identity, or is it a bit of both? 112 Presenting Findings The findings of this study are presented in the form of a written explication of the grounded theory associated with a theoretical (conceptual) framework. The writing process in grounded theory research is evolutionary: it is through writing about the findings that the theory that has been developed is refined. Expressing one's ideas in writing forces one to be very clear in defining and explaining concepts and their linkages, and provides insights into any breaks in logic or points of weakness in the developing theory. As Miles and Huberman (1994, p. 299) point out, reporting the findings of a qualitative study is not separate from the analysis, it is analysis. As mentioned above, memos were a great facilitator in writing about findings, particularly at the beginning of the writing process. As well, three oral presentations I made at conferences on portions of the findings "forced" me to write about and articulate these findings, thereby helping me to clarify aspects of the developing framework and theory. A very systematic approach to presenting findings was adopted, working through the different levels of abstraction in the framework (Grams, 1994). After introducing the framework and the core category, each category formed the basis for the substantive writing. Each category was defined, its properties were named, and a detailed explanation of every dimension associated with each property was provided. Every dimension was illustrated by a quotation that best exemplified the dimension under discussion and served to connect the theory with the data and ground it to the words of participants. Linkages between and among dimensions, properties and categories were explicated. This process was repeated for each category within the framework. Themes were defined, explained and illustrated toward the end of the writing process. Morse (1994, p.231) states that one approach to qualitative writing is "to write... as though the reader is solving the puzzle with the researcher." The methodical approach to presenting the findings just described should provide the reader with a clear logical path to follow. The reader should understand what each of the categories, properties and dimensions in the framework mean, and how different parts of the framework fit together. 113 This concludes the discussion of the analytic method. The study design, data collection and analysis processes adhered to strict and consistent ethical standards which are described in the next section. Ethical Considerations Issues of Confidentiality and Informed Consent Ethical issues of informed consent, privacy, and confidentiality of data were adhered to as written in the consent form that was signed (before any data was collected) by all participating women (Appendix M), each family member who was interviewed (Appendix N) and the audiologist (Appendix O). Because the booking secretary mentioned my name to potential participants before my initial telephone contact with them, and because this initial contact occurred right after their audiology appointment was booked, it was important to make clear to these women that I was not affiliated with the audiology facility and that my study and their involvement in it was not part of their upcoming audiology appointment. As stated previously, each participant was assigned a code name for use in this dissertation as well as in all other publications and presentations related to this research. Only the researcher can connect a participant's pseudonym with her real name. A l l personal identifiers have been removed from any quotes used in presentations or publications. Care has also been taken not to reveal the identity of any family members of participants, the audiologist, or the name of the audiology business. Family members and the audiologist are referred to by a single initial in all quotes published or presented. As the consent forms indicate, informants were advised that they had the right to withdraw from the study at any time, that they had the right to request that I withdraw from observing their audiologic assessment, and that their participation, or lack thereof, would in no way affect the care they received regarding their hearing from the participating audiologist or any other hearing professional. 114 The Role of Researcher versus Clinician I made no effort to hide from participants the fact that I am an audiologist. I made every effort, however, to avoid crossing into the role of clinician when participants occasionally asked me questions related to my audiology expertise, for example, questions about what to expect from the test or what certain results meant. At the last interview I did explain and discuss the basic audiologic findings to a few participants who had expressed in the interviews their sense of confiision regarding their test results. I made the decision to briefly step into my "audiology identity" after weighing the pros and cons of this action and concluding that the repercussions for my research were very minimal, given it was the end of participants' involvement with the study. Furthermore, I felt that such action would not negatively impact these women or their course of audiologic rehabilitation. I felt uncomfortable at the prospect of parting company with a participant who might continue to seriously misinterpret her test results. Criteria for Assessing Quality Scientific rigour in quantitative research is tied to criteria of reliability, validity, and generalizability. These criteria have been referred to tongue-in-cheek as the "holy trinity.. .for all true believers in science" (Kvale, 1995). Qualitative research cannot be assessed using these same criteria because of fundamental differences in philosophy that affect every aspect of a qualitative study beginning with the choice of research question and ending with the analysis of findings. In fact, many respected qualitative researchers eschew the term "rigour" in favour of the term "trustworthiness" as the umbrella encompassing the criteria for assessing quality (Lincoln, 1995). I do not advocate the continuation of a rigid dualistic approach to empirical and post-empirical research, but I do wish to reiterate three of the major differences between quantitative and qualitative research to highlight the need for different criteria to assess each approach. First, qualitative research is naturalistic inquiry. Whereas a quantitative study controls variables as much as possible, a qualitative study is interested in how a study question plays out in the real-world or natural setting. Second, the nature of the data in qualitative research are 115 largely the words of mformants, unlike the results of quantitative inquiry that reduce to statistics and graphs. A third difference is related to the first two. Quantitative research reifies objectivity, whereas qualitative inquiry acknowledges the influence the researcher and participants exert upon each other. Neither research paradigm disputes the need to adhere to the canons of "good science", but the debate continues, both within and across qualitative paradigms, as to the exact criteria that should be used to judge the merits of qualitative research. In the opinion of some researchers (e.g., Lincoln, 1995), these criteria are still emerging, as qualitative enquiry establishes its legitimate place in research. The following discussion describes the criteria of auditability, credibility, and fittingness as applied to this study, and the ways in which the study design, data collection, and data analysis contributed to the rigour or tmstworthiness of this research. Auditability Reliability in quantitative research refers to the consistency, stability, and dependability of a test. In contrast, qualitative research does not seek to replicate results because it holds that findings cannot be repeated. As Strauss and Corbin (1998) point out, reproducing social phenomena, the area of interest to qualitative researchers, can be difficult because it is almost impossible to replicate the exact conditions under which data were collected and to control all the variables that may influence the findings. However, given similar conditions, the same general rules of data collection and analysis, and the same theoretical perspective of the original researcher, other researchers should arrive at the same or very similar explanations of the phenomena under study. Guba and Lincoln (1981) propose that auditability, not reliability, be the criterion of rigour or trustworthiness relating to the consistency of qualitative findings. Sandelowski (1986) identifies two important aspects of auditability: A study and its findings are auditable when another researcher can clearly follow the 'decision trail' used by the investigator. In addition, another researcher could arrive at the same or comparable but not contradictory conclusions given the researcher's data, perspective, and situation, (p.33) 116 In this study, auditability was sought by attention to consistency in data collection, data analysis and interpretation of findings. Following the collection of data from the first two participants, the study design was reviewed and approved for its perceived ability to be feasibly implemented across all subsequent study participants. Data were collected in a consistent fashion for each participant, with attention paid to following the interview guide, thereby ordering the areas of inquiry similarly across participants. My prolonged contact with each participant over many interviews spanning many weeks provided the time necessary to conduct a thorough exploration of the research questions. Moreover, having several interviews with each participant allowed for topics to resurface in different interviews, sometimes frequently. I constantly compared a participant's responses across interviews to gauge how consistent they were over time for the same question or for similar or related questions. Prolonged contact with participants also permitted me to verify my interpretation of her response from a previous interview or to probe further any answers that I perceived to be incomplete or inconsistent. Pursuing inconsistent responses was important in determining whether the source of variation reflected a change in the participant's opinion on a question over the time course of the study. Data triangulation, the use of a variety of data sources to crosscheck findings (Patton, 1990), was an important tool utilized in this study in the quest for auditability. For example, I compared a participant's answers in interviews with the written entries she made in her journal and my observations of her behavior. Data from the audiologic assessment also permitted me to compare a participant's comments to me or in her journal with her responses to the audiologist's questions, either on the "client intake history form" or in conversation with the clinician before or after the hearing test. The over-riding goal was not to prove or disprove the consistency of individual participants, but rather to look at patterns of interaction in responses and behavior and to be able to understand incongruencies if and where they occurred. Another form of data triangulation consisted of the various "permutations" of the interview format. Each permutation of conversational dyad: researcher and participant, researcher with family member (occasionally with participant present), participant and audiologist, and researcher with audiologist, provided a unique perspective on the 117 research questions. As well, as mentioned previously, my interviews with the seven women afforded the opportunity to play the role of participant-observer as well as interviewer, insofar as I was able to observe and record how each woman's hearing impacted her ability to converse one-on-one. I was also able to observe and record conversational "breakdowns" and some of the strategies these women utilized to cope with their hearing challenges and to repair conversational breakdowns. Similarly, my interviews with a spouse who was hard-of-hearing afforded the opportunity to observe aspects of his behaviour that could potentially affect the dynamics between him and his wife, the participant. In data analysis, attention to detail in transcribing the raw data of interviews led to consistency in this first step of data reduction, as I strove to reproduce as faithfully as possible the utterances of each participant. A lengthy time spent immersed in data analysis and interpretation spanned two slightly different approaches to coding that were compared and found to corroborate each other in the identification of concepts leading to the development of theory. The data analysis procedures described in this chapter and the presentation of findings in upcoming chapters should provide the reader with a "decision tr