@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Applied Science, Faculty of"@en, "Nursing, School of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Steele, Rose Guiller"@en ; dcterms:issued "2009-07-02T23:01:57Z"@en, "1999"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Children with neurodegenerative, life-threatening illnesses account for a significant proportion of children requiring palliative care. Most of their care is provided at home by their families. Yet, there is a paucity of research that examines families' experiences when a child is dying at home. This grounded theory study generated a contextually-grounded description of families' experiences of living with a child who has a neurodegenerative, life-threatening illness, and how those experiences changed over time; the impact on families; and families' perceptions of factors that influenced their ability to care for their children. Data were collected from eight families through observations and audiotaped interviews. Families moved through a process of navigating uncharted territory as they lived with their dying child. This process occurred within the broader context of the acute, curative health care system and the sociocultural environment. The process was initiated by a precipitating event, such as one parent recognizing a deterioration in the child's motor skills. The emotions of fear, uncertainty, and grief gave impetus to the process. These emotions were very strong in the beginning. They gradually declined over time, but increased with subsequent precipitating events. Families lived much of their lives on plateaus of relative stability where they often felt alone and isolated from health care professionals. Inevitably, periods of instability originated in subsequent precipitating events in the process that led to families dropping off the plateau. Living with a dying child had a profound physical, emotional, and financial impact on families. The process of navigating uncharted territory was characterized by four dimensions - entering unfamiliar territory, shifting priorities, creating meaning, and holding the fort. Parents used strategies within these dimensions to manage the physical, cognitive, and emotional work arising from the situation. These dimensions and strategies were constrained and facilitated by four intervening conditions - relationships with health care providers, availability of information, gender differences, and communication between parents. The theory generated in this study contributes to an understanding of families' experiences as they live with a dying child. This theory provides a framework from which appropriate, family-centred interventions can be developed to facilitate optimal palliative care."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/10004?expand=metadata"@en ; dcterms:extent "15545135 bytes"@en ; dc:format "application/pdf"@en ; skos:note "N A V I G A T I N G U N C H A R T E D T E R R I T O R Y : E X P E R I E N C E S O F F A M I L I E S W H E N A C H I L D H A S A N E U R O D E G E N E R A T I V E L I F E T H R E A T E N I N G I L L N E S S by R O S E G U I L L E R S T E E L E B . S c . N . , University o f Toronto, 1991 M . S c , University o f Toronto, 1994 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (School o f Nursing) W e accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A M a r c h 1999 © R o s e G . Steele, 1999 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 Kf c W g i^-igj The University of British Columbia Vancouver, Canada Date f W c ^ ?>0^ DE-6 (2788) NAVIGATING UNCHARTED TERRITORY: EXPERIENCES OF FAMILIES WHEN A CHILD HAS A NEURODEGENERATIVE LIFE THREATENING ILLNESS Abstract Children with neurodegenerative, life-threatening illnesses account for a significant proportion of children requiring palliative care. Most of their care is provided at home by their families. Yet, there is a paucity of research that examines families' experiences when a child is dying at home. This grounded theory study generated a contextually-grounded description of families' experiences of living with a child who has a neurodegenerative, life-threatening illness, and how those experiences changed over time; the impact on families; and families' perceptions of factors that influenced their ability to care for their children. Data were collected from eight families through observations and audiotaped interviews. Families moved through a process of navigating uncharted territory as they lived with their dying child. This process occurred within the broader context of the acute, curative health care system and the sociocultural environment. The process was initiated by a precipitating event, such as one parent recognizing a deterioration in the child's motor skills. The emotions of fear, uncertainty, and grief gave impetus to the process. These emotions were very strong in the beginning. They gradually declined over time, but increased with subsequent precipitating events. Families lived much of their lives on plateaus of relative stability where they often felt alone and isolated from health care professionals. Inevitably, periods of instability originated in subsequent precipitating events in the process that led to families dropping off the plateau. Living with a dying child had a profound physical, emotional, and financial impact on families. The process of navigating uncharted territory was characterized by four dimensions -entering unfamiliar territory, shifting priorities, creating meaning, and holding the fort. Parents used strategies within these dimensions to manage the physical, cognitive, and emotional work arising from the situation. These dimensions and strategies were constrained and facilitated by four intervening conditions - relationships with health care providers, availability of information, gender differences, and communication between parents. The theory generated in this study contributes to an understanding of families' experiences as they live with a dying child. This theory provides a framework from which appropriate, family-centred interventions can be developed to facilitate optimal palliative c iv Abstract Table of Contents List of Figures Acknowledgements CHAPTER I Problem Background to the Problem Statement of the Problem Purpose Summary CHAPTER II Conceptual Framework 9 Perspective on Grief 9 Conceptualizations of Grief 10 Chronic Sorrow 13 Anticipatory Grief 13 Perspective on Children and Life-Threatening Illnesses 14 Children's Knowledge About Death 15 Communicating With Children About Death 17 Siblings 19 Perspective on Families 21 Systems Theory 22 Boundaries 24 Family Research 24 TABLE OF CONTENTS ii iv x 1 1 6 7 7 Perspective on Coping Individual Coping Family Coping Summary C H A P T E R III Literature Review Impacts of Families' Experiences Impact on Family Structure and Patterns of Interaction Emotional Impact 46 Physical Impact 48 Financial Impact 48 Spiritual Impact 50 Summary 51 Factors Influencing Family Experiences 51 Location of Care 51 Communication 54 Levels of Family Functioning 57 Conclusion 62 Research Questions 64 Definition of Terms 64 C H A P T E R IV Research Design and Implementation 66 Symbolic Interactionism 66 Setting 68 Selection of Participants 69 The Sample 7j Data Collection and Procedures <72 Data Analysis g j Scientific Rigour gy Ethical Considerations gy Summary gg vi CHAPTER V Findings 90 Context of the Illness Experience 90 Acute, Curative Health Care System 92 Sociocultural Environment 99 Health care providers 99 Extended family and friends 101 Impact of visible differences 103 Navigating Uncharted Territory 104 Emotions as Momentum for the Process 106 Physical Impact on Families 109 Emotional Impact on Families 111 Navigating Uncharted Territory: A Basic Social Process 115 Entering Unfamiliar Territory 115 Initial Precipitating Event 116 Diagnostic Process 117 Seeking Information 121 Sharing Information 126 Shifting Priorities 127 Going Into Slow Motion Focusing on Child Putting Life on Hold Getting Rid of Excess Baggage Picking up the Badge of Disability Creating Meaning Taking One Day at a Time Refraining the Experience Reaffirming Faith Participating in Research Holding the Fort Living by the Clock Co-ordinating Services Providing Physical Care Promoting the Child's Health Forming New Relationships Subsequent Precipitating Events Dropping off the Plateau Intervening conditions Relationships With Health Care Providers Availability of Information Gender Differences Grieving Medication and feeding regimes Information needs viii Caring for self Communication Between Parents Summary CHAPTER VI Discussion Stress, Coping, and Adaptation Frameworks Individual Coping Family Coping The Illness Trajectory Gender, Communication, and Culture Family-Centred Care Impact of the Experience Burden Chronic Sorrow Fatigue Family Management Style Summary CHAPTER VII Summary and Implications 224 Summary 224 Implications for Practice, Education, and Research 229 Practice and Education 229 Research 237 Conclusion 239 185 185 186 189 191 196 200 208 208 211 212 217 221 References ix 241 Appendixes 260 Appendix 1: Family Member Information Sheet 261 Appendix 2: Consent Form: Adult Family Member 262 Appendix 3: Parental Consent Form: 111 Child 264 Appendix 4: Parental Consent Form: Sibling 266 Appendix 5: Initial Questions for Interviews 268 Appendix 6: Letter to Families Requesting Feedback on Draft Analysis 269 X LIST OF FIGURES Figure 1: Navigating Uncharted Territory 91 Figure 2: Dropping off the Plateau 169 ACKNOWLEDGEMENTS xi It is difficult to find the right words at a time like this. I am full of emotions that cannot be fully expressed in mere words. Creating knowledge is a wonderful, heady experience. But it is also challenging, time-consuming, and, at times, frustrating. I have been on my own personal journey in the process of creating knowledge. Yet, I could not have completed this journey without the help and friendship of some very important people. I owe my heartfelt thanks and this dissertation to the following people: To the families who shared their experiences with me, who welcomed me into their lives, and allowed me the privilege of telling their stories. I admire them all very much and thank them for being such wonderful teachers. Their willingness to share their experiences, even when it caused them personal pain, left me in awe. My hope is that this dissertation will help make their lives, and the lives of similar families, somewhat easier. To my excellent dissertation committee. My supervisor, Dr. Betty Davies, RN, Ph.D., whose support and encouragement have been vital throughout my journey. Betty continually guided me with her wisdom, knowledge, and genuine caring. Her faith in me helped me through the challenging times, and her unbridled joy in creating knowledge was infectious and inspiring. My committee members, Dr. Joan Bottorff, RN, Ph.D. and Rhea Arcand, RN, M.N., who were a critical part of my journey. Each provided unfailing support, yet continually challenged me to raise my level of thinking. I could always count on Rhea and Joan to ask the difficult questions and to point out the gaps in my logic. Joan offered insightful and helpful suggestions that encouraged me to move beyond description and towards a truly theoretical formulation of families' experiences. Rhea's administrative and clinical background enhanced our discussions and added a welcome dimension to my dissertation. Without them, I would not have gained the valuable insights that promoted depth in my work. To my friends and colleagues for their interest and encouragement. I especially want to thank Colleen Varcoe and Susan Dahinten. Colleen was always there for me when I needed her. Her support and understanding knew no limits, even when she was deeply immersed in her own work. I appreciate and value Colleen's comments on my work as much as I value her friendship. Though Susan moved on in her professional career and left the province, she continued to provide strength through her quiet support. Somehow, she always knew how and when to ask me if I was working on my dissertation. Both Colleen and Susan helped to keep me on track. To my own family who had faith in me and gave me their unremitting love and support. My teenage daughters, Laura and Nicola, who helped me through the long years of university and kept me grounded in reality. Somehow, they managed to hang in there, even when I was distracted or short-tempered. Finally, to my greatest supporter, my husband David who never ceased telling me how proud he is of my accomplishments. We did it together. 1 Chapter I: Problem Background to the Problem Over the past 30 years, the field of palliative care has become fairly well established with adults. Pediatric palliative care, however, is relatively new. It is only in the last 10-15 years that pediatric palliative care has received clinical and, even more recently, research attention. The basic principles of the palliative care movement are similar for both the adult and the pediatric populations: (a) to provide physical, emotional and spiritual care to the sick person and the family, (b) to promote the best quality of life, and (c) to provide bereavement care. As suggested by the leading pioneer in pediatric palliative care, Mother Frances Dominica from England, the role of pediatric palliative care is to offer friendship, support, and practical help to families (Dominica, 1987). In addition, complex symptom management is an integral part of pediatric palliative care programs (Goldman, 1998; McGrath, 1998). Ideally, palliative care should be introduced early in the child's progressive, life-threatening illness (PLTI), preferably at the time of diagnosis (Frager, 1996). This early link with palliative care services allows an opportunity to promote palliative care principles, facilitates family-centred care that is responsive to the changing needs of families, and may help prevent crisis-oriented management. As the family's needs dictate, attention should shift from acute to palliative care that includes respite and terminal care, and then to bereavement care. Thus, health professionals not only provide bereavement care, but often work with grieving families for a considerable period of time before the child dies. In adult palliative care, most patients have cancer and are expected to die within six months of admission to palliative care. Conversely, children in pediatric palliative care suffer from a wide variety of diseases and syndromes. About 20% of the children may have cancer, but many have progressive neuromuscular or neurodegenerative conditions, such as Battens disease or one of the leukodystrophies, that will eventually cause their death, though not necessarily in the next six months (Ashby, Kosky, Laver, & Sims, 1991; B. Davies & Howell, 1998; Goldman, 1996). Consequently, the time between diagnosis and death is variable in these children. Many children live for a prolonged period of time, though they will not reach adulthood. Families often require respite care during this protracted period of time. The pioneers in pediatric palliative care (Dominica, 1987; Hunt, 1986) emphasized the importance of respite care for these families. While there has been an increasing focus on complex symptom management in the past few years (Goldman, 1998; Liben, 1996; McGrath, 1998), little has changed in the commitment to respite care as an integral component of pediatric palliative care (B. Davies & Howell; Goldman, 1998). Respite care is acknowledged as essential for families (B. Davies & Howell; Goldman, 1996). It provides them with some relief from the unremitting, 24-hour care of their ill children. Respite care is usually provided for a relatively short period of time on an intermittent basis to allow families to have a break. The numbers of children with a prolonged terminal illness are low when compared with those in the adult population. For example, there were 346 children who died from terminal illnesses in British Columbia over the three year period 1988-1990. At any one time, there are over 200 children in the province who are living with progressive, life-threatening illnesses (B. Davies, 1992). One estimate from the United Kingdom indicates that 1:1000 children may be affected by life-threatening illnesses (Goldman, 1996). Although the current numbers are relatively small, these children may pose substantial management problems and may use a disproportionate amount of in-patient health resources (Caring Institute of the Foundation for Hospice and Home Care, 1987). Additionally, the numbers of children with a PLTI are projected to increase as the incidence of life-threatening diseases rises and as advances in technology and medicine reduce mortality rates for conditions from which children previously died (Broome, 1998; B. Davies & Howell, 1998). 3 Home care, rather than in-patient care, is strongly supported for pediatric palliative care (Lauer, Mulhern, Hoffmann, & Camitta, 1986; Stevens, 1998). Most of the care for these children is typically provided at home by their families over an extended period of time that may last years rather than months (Burne, Dominica, & Baum, 1984; Goldman, 1998; Stevens). The home provides a secure, familiar, and comfortable environment for the child (B. Davies & Howell, 1998). It allows parents to feel more in control and is less disruptive to family life (Martinson, 1993). However, the impact of caring for a child with a PLTI can be substantial. Care is often provided to the detriment of the individual caregivers and to the family itself (Folden & Coffman, 1993; Stevens). Children exist within a family, so the child's illness affects the whole family. In turn, the family's response to the illness has an important impact on the ill child (Friedman, 1998; Leonard, Enzle, McTavish, Cumming, & Cumming, 1995). Therefore, experts agree that care of these children must also include their families. The ill child, parents, and siblings must be seen as the central focus of care. They must also be involved in all aspects of care and decision making (Martinson, 1993). Parents want and expect to take care of their child. They may want to be in control even more so than families of terminally ill adults (B. Davies, 1992). Families of dying children tend to be large because parents and grandparents are still alive, siblings live at home, and friends are part of the larger affected group. So meeting the needs of these families often requires more time, skill, and expertise than is needed for families of older patients. Family stress is at its highest when a member is entering or leaving the family system and homeostasis or equilibrium is disturbed (Friedman, 1998). Thus, families with a dying child are at risk for suffering the emotional and physical effects of stress. The increased media attention on such families in recent years has documented some of the stresses families face, as well as the consequences. For example, a mother in Canada was charged with starving her ill child to death, and in another family, the father was convicted of manslaughter in the death of his severely 4 disabled child. The problems faced by these families have been thrust into the public eye and have led to increasing debate over how professionals can best support families. Although no morbidity statistics are available, caring for a child with a PLT I at home has been associated with emotional, psychological, and financial distress in parents (Davies, 1996; Diehl , Moffitt, & Wade, 1991; Stein & Woolley, 1990). Parents may be anxious, suffer from insomnia, and be socially dysfunctional; some may start to use or increase their use of tranquillisers, anti-depressants, or cigarettes to relieve their anxiety (Stein & Woolley). Parents also struggle with trying to balance the demands of their i l l child with the needs of healthy siblings. In addition, the i l l children may experience emotional problems, such as anxiety and unhappiness. Siblings also experience distress when a brother or sister is dying. Indeed, in some families, siblings may have more problems than the dying child (Stein & Woolley). Emotional and behavioural problems, such as difficulty in dealing with their school work or increased aggression or withdrawal, may occur. Bakke and Pomietto (1986) noted that health professionals want to support grieving families. Yet, they are often unaware of the specific elements of support, such as the type of information needed or the location of care, that would be most beneficial for families. Additionally, health professionals, such as nurses, often view themselves as advocates for patients and families. A s such, professionals must be capable of presenting information and alternatives to families so families can make the choice that is most appropriate for them. Anecdotal data suggest that although families often require substantial attention, information, and support, there are important differences between the perceptions of families and staff in how well these needs are met. For example, bereaved families often say, \"If only they had told me to expect...\" (that the child would hemorrhage, or what would occur at the time of death, and so forth). Yet, healthcare professionals wi l l point out that those issues and information had been discussed repeatedly during the child's illness. When provided with clear evidence that 5 the issues were discussed and information given, families tend to say, \"Oh, yes. I'd forgotten that.\" They cannot retrospectively identify what else the health care professionals could have done to have made the experience any easier. Little is known about the kinds of information that are useful, when families hear or do not hear information, what kinds of support are desirable, or how that support should be given. There may be times or stages when \"teachable moments\" occur in the lives of families, or families may give cues that could alert health care professionals to intervene in a particular way. No research is currently available to identify such events. Family involvement in care of a child with a PLTI is complicated by the fact that death and dying are generally taboo subjects in the United States of America (USA) and Canada (Fulton, 1994). In these North American societies, people seldom think about death anymore, especially a child's death. North Americans are reticent to talk about the meaning of death in human life, about appropriate responses to death, or about the place of rituals and communal bonds in the face of death. There is no general culture for dealing with issues around death. Society hides death, such as with the almost exclusive use of funeral homes in North America (Walsh & McGoldrick, 1988). North Americans often seem to act as if death is just another disease to overcome. The logic that death is a constant enemy is at odds with the belief of many clinicians that death is part of living. Moreover, the USA has been described as unusual among developed countries in its attempts to try to conquer death (Callahan, 1995), though Canada also fits with this description. Some countries seemingly have less difficulty in caring for the dying -they do not struggle with the clinical, moral, and legal problems seen in the USA and, to a somewhat lesser extent, in Canada. In North American culture, physicians often do everything that is possible, but when they are unable to do more then they leave the patient to nurses (Callahan). Many professionals have a particularly difficult time in identifying a child as terminally il l . They refrain from moving from curative to palliative care at an appropriate time (B. Davies & Steele, 1996; O'Gorman, 1998). Although the palliative care movement has led to 6 many changes in the care of dying patients, too many health professionals remain afraid of death. Therefore, professionals cannot always provide the support families need because of their own insecurities and fears. Over the 20th century, death has moved from the realm of the sacred to the secular, from the private to the public, and from the natural to the artificial (Callahan, 1995; O'Gorman, 1998). Medicine takes the view that death can be averted or treated, as can be seen in the rapid increase in technology and intensive care units. Whereas people used to die at home with their families, dying in hospitals and being cared for by professionals has become the norm. Since people are no longer confronted by the death of older family members at home, most young adults have gained little experience of death and dying. Therefore, young parents of children with life-threatening illnesses have not had to deal with many deaths before, and almost certainly not a pediatric death (B. Davies, 1992). This lack of experience with death and dying can exacerbate a situation that is already fraught with real and potential difficulties. In addition, families are usually developmentally and chronologically young, and they may need considerable attention, skill, and time to help them survive the experience (B. Davies). Statement of the Problem There is a lack of knowledge about the experiences of families where a child has a progressive, life-threatening illness. Unlike adult palliative care, a substantial percentage of children in pediatric palliative care die from neurodegenerative, life-threatening illnesses (NLTIs). Health professionals have barely begun to document the effects on families of caring for a child with an NLTI. Little is known about the most appropriate interventions for these families. Retrospective accounts of a child's illness have not always been accurate or even possible. Therefore, in addition to retrospective data collection, a prospective approach is important to capture family experiences as they evolve. 7 Purpose The purpose of this study was to enhance understanding of the experiences of families with a child who has an NLTI. The knowledge generated by this study and its contribution to developing theory will be used to promote effective palliative care of these families. Specific aims of the study were to: (1) describe families' perceptions and experiences of living with a child who has a neurodegenerative, life-threatening illness, and how those experiences changed over time; (2) describe the impact on the family of living with a child who has a neurodegenerative, life-threatening illness; and (3) describe families' perceptions of the factors that influenced their ability to care for their child who has a neurodegenerative, life-threatening illness. Summary Increasing numbers of children with progressive, life-threatening illnesses are being cared for at home by their families. Health professionals can provide support to families, but they do not have a clear understanding of what kind of support is appropriate or how to provide that support to these families. Living with a child who has a PLTI may have a profound impact on families. Many families may experience psychological, physical, financial, and social disruptions that may affect their quality of life. Health care professionals are in a position to provide interventions that will promote the health and welfare of families, if they know the type of interventions to offer. Although palliative care has been developing rapidly in the past decade, there is still a dearth of research in this field, particularly in the pediatric area. It is imperative that suitable interventions are uncovered and developed to provide guidance to practitioners. This research study focused on the experiences of families when a child has a PLTI, specifically a neurodegenerative illness. A relatively high percentage of children requiring palliative care suffer from neurodegenerative illnesses, yet very little research has been completed in this area. 8 The findings of this research provide a foundation for identifying interventions that will enable health care professionals to provide the most appropriate care to such families. The remainder of this dissertation follows a fairly standardized format. The conceptual framework that underlies the study is introduced in Chapter II. This framework sets the context in which the research questions were asked. Related literature is reviewed in Chapter III to provide an understanding of what is currently known about families where a child has a progressive, life-threatening illness, especially a neurodegenerative disease. In Chapter IV, the research design, procedures, and implementation are explained. The findings of the study are presented in Chapter V. The experiences of families when a child has a neurodegenerative, life-threatening illness are described and conceptualized through the process of navigating uncharted territory. This newly developed theory is placed within the larger context of current research and literature in Chapter VI. Finally, the study is summarized and implications for practice, education, and research are offered in Chapter VII. 9 Chapter II: Conceptual Framework A conceptual framework underlies a study problem and supports the rationale for conducting a study. It describes the perspective from which one views the world, and gives guidance for understanding the context in which one studies phenomena. A conceptual framework is not synonymous with a theoretical framework. In this study, grounded theory methodology was used to develop a contextually-grounded theory about the experiences of families when a child has a neurodegenerative, life-threatening illness (NLTI). Clearly, the use of a previously developed theoretical framework would have been inappropriate for this purpose. However, researchers do not begin a study without some understanding of the context within which the research questions will be asked. Four major concepts formed the context for this study. The first concept concerned children and life-threatening illnesses, the second concept was related to families, and the third concept was coping. The focus of this study was where these three concepts intersected within the broader context of grief. In this chapter, the researcher's perspective of these four major concepts will be described in order to delineate the context in which the study was conducted. Perspective on Grief Families with a child who has a progressive, life-threatening illness live with the threat of death and dying every day. Death means the loss of many things - the loss of this life, of self as a person, of tomorrow, of those we love, and of all we know. In families with a dying child, losses are felt by every member of the family. Not only must the parents prepare themselves for their child's death, but the ill child and siblings must also deal with this reality. Therefore, it is important to understand the relevant theories of death and dying that affect both families and those who provide care to them. In this study, grief was thought of as a process, rather than as stages. Families were not expected to move through pre-determined stages. Instead, the researcher sought understanding about the process of grief. 10 Conceptualizations of Grief In early conceptions of grief, clinicians and theorists viewed grief as linear and time-bound (Miles & Demi, 1994). Terms such as grief, grieving, and mourning were used synonymously. Uncomplicated grief was seen as a progression through stages of shock and disbelief; developing awareness of loss; and prolonged recovery. The work of uncomplicated grief was expected to take 4-6 weeks to complete (Miles & Demi; Teel, 1991). If mourning was delayed, chronic, exaggerated, or masked, then the grief response was said to be abnormal. Consequently, the extended grieving typical of families of a dead child was seen as pathological and requiring psychiatric assistance or counselling. Although there had been a number of theories proposed about grief, little attention was paid to death and dying by professionals and lay people until Elisabeth Kubler-Ross published On Death and Dying in 1969 (Kubler-Ross, 1969). This book was instrumental in bringing the staged model of grief into public and professional awareness. According to Kubler-Ross, people go through five different stages when they are faced with dying: denial and isolation; anger; bargaining; depression; and acceptance. Although she wrote that, \"These means will last for different periods of time and will replace each other or exist at times side by side\" (p. 122), many professionals and lay people viewed the stages as sequential. Consequently, patients were expected to move through these stages until they finally accepted their death. Many professionals and lay people continue to work today with this notion of linearity and time-bound grief. In recent years, however, the linear model has been questioned by some clinicians and theorists (Cowan & Murphy, 1985; Worthington, 1989; 1994). Questions have been raised particularly in relation to the length of time considered \"normal\" to continue grieving, and the assumption that failure to achieve resolution of grief is abnormal. The five-stage model attributed to Kubler-Ross is not supported by research and has been called superficial, inadequate, and misleading (Corr, 1993). There has been no demonstration of the existence of the stages, nor evidence that people 11 actually move from stage one through to stage five (Buckman, 1998). Instead, humans cope in many ways, not just these five. There are no prescriptive or obligatory ways in which people should cope with death and dying. There is no \"right\" or \"wrong\" way to die. Coping with dying is a richer and more complex process than merely progressing through stages. Buckman (1998) proposed a three-stage model in which he argued that individuals facing death react in ways that are consistent with their character and their previous coping patterns. Reactions are, therefore, not a product of the diagnosis or the stage of the dying process. In addition, Buckman proposed that progress through the dying process is marked by resolution of the elements in emotions that are resolvable, rather than by a change in the type of emotions. His model incorporates three stages: initial (facing the threat); chronic (being ill); and final (acceptance of death). When emotions in the first stage are resolved, then an individual moves into the second stage. Not everyone reaches this second stage. Buckman suggests that those individuals require professional help. In the third stage, emotions are less intense and the individual accepts impending death. While this model is an improvement on Kubler-Ross' model (Kubler-Ross, 1969), it continues to present a mechanistic and unidimensional approach to a multidimensional process. Many theorists and clinicians now conceptualize grief as a process, not as stages (Attig, 1991; 1996; Cowles, 1996; B. Davies & Eng, 1998; Gyulay, 1989). According to Attig (1996), grief is the specific emotion that one feels in reaction to loss. This emotion engenders helplessness and passivity because the emotion grief is essentially irrational. Normally, emotions are motivating. But grief cannot motivate because there are no recognizable means to the end desired - wishing will not prevent death or return the deceased person to life. Attig suggested that the depth and power of grief may derive from this lack of motivational force. Furthermore, in contrast to the passive, standard conceptions, Attig viewed the process of grieving as an active, complex process, that presents challenges and opportunities. An active process may be more 12 accurate descriptively and may capture more of the complexity of bereavement than a more passive view. It may also be more powerful in promoting self-understanding and in providing better direction for those who would help the dying and the bereaved. There is some empirical research to support a dynamic conceptualization of grief. Focus group participants from a variety of cultural backgrounds were asked to draw on their own personal experiences of grief. Their definitions of grief were similar to those in the professional literature (Cowles, 1996). Irrespective of their culture, they viewed grief as a dynamic, pervasive, and highly individualized process. Participants agreed that culture or cultural heritage is a key component of the context in which people respond to actual or potential health problems. They also believed that, though mourning rituals or traditions may be culturally defined and proscribed, each person experiences grief in his or her own way. In addition, grief occurs universally across all age groups. In contrast to the professional literature that often emphasizes temporal and behavioural limits to grief, the research participants did not impose limits. Neither were they judgmental about others' experiences of grief. Others' experiences were viewed as simply different, hot as wrong. Most agreed that the progression of grief is unpredictable and that changes over time vary. In addition, grief does not necessarily have a particular end-point. McClowry, B. Davies, May, Kulenkamp, and Martinson (1987) provided further support for this view of grief. They used grounded theory methods to study 49 families who were interviewed 7-9 years after a death following childhood cancer. Many parents and siblings were still experiencing pain and loss at the time of interview. Despite similarities in experiences related to the initial loss of the child (i.e., in that it had created an \"empty space\"), three patterns of grieving responses were noted: \"getting over it;\" \"filling the emptiness;\" and \"keeping the connection.\" Families stated that time lessened the pain. In addition, the sense of emptiness changed over time. The \"empty space,\" however, was likely to recur at specific times such as holidays and anniversary dates like the child's date of birth or death. 13 Chronic Sorrow Recurring waves of grief or chronic sorrow have been observed in parents of children with mental or physical disabilities. This pervasive, recurrent sadness was first characterized by Olshansky (1962) as a normal response to disruptions in anticipated normalcy. Clinicians and researchers have reported that families' ongoing losses may also be associated with feelings such as sadness, anger, depression, confusion, fear, anxiety, embarrassment, and isolation that vary in intensity among different persons, across situations, and at different times within an individual (Fraley, 1990; Lindgren, Burke, Hainsworth, & Eakes, 1992; Phillips, 1991; Teel, 1991). Further research has expanded the concept of chronic sorrow to individuals experiencing a variety of loss situations, as well as to their family caregivers (Eakes, 1993; Eakes, Burke, Hainsworth, & Lindgren, 1993; Hainsworth, Eakes, & Burke, 1994; Lindgren, 1996). Chronic sorrow is based on the premise that people may experience recurrent episodes of grief following a loss. The cyclical nature of chronic sorrow means that each episode of grief wanes over time. Periods of happiness are interspersed between episodes of grief, so grief does not usually become incapacitating (Lindgren et al.). When a child has an NLTI, a family experiences losses over many years as the child's condition worsens. Their experiences of grief associated with these losses are not well understood. Moreover, parental feelings of recurrent grief may resurface during situational or developmental crises experienced by their children. Families may require additional professional assistance at those times to deal with the resultant recurrent grief. Anticipatory Grief The extended period of illness experienced by children with NLTIs may provide families with time to prepare for the death of their child. Kubler-Ross (1983) proposed that when families have time to mourn losses as they occur, then the seriousness and frequency of psychological difficulties following the child's death will be minimal. However, as noted by B. Davies and Eng (1993), the value and function of anticipatory grief has not been conclusively demonstrated. One 14 difficulty with the concept of anticipatory grief is that it leads to the assumption that an expected death is easier to deal with than an unexpected death. Yet, this view may be erroneous, since the expected death of a child appears to have as severe an impact on parents as does a sudden bereavement (Hill, 1994). In addition, many clinicians have observed that, despite education and anticipatory grief, families are rarely prepared for the reality of their child's death. B. Davies (1993) noted that parents and siblings, who have accepted the inevitable outcome of the child's illness, often make comments after the death such as, \"But I didn't think he would die until he graduated\" or \"I thought he would wait until I got home from school\"(p.l41). It appears that anticipatory grieving may have a role to play in families' grieving. However, that role remains unclear. It cannot be expected, therefore, that anticipatory grieving will reduce the need for assistance among families with a child who has an NLTI. In summary, current theorists are replacing the early conceptions of grief as linear and time-bound with a new conceptualization of grief as an active, dynamic process that may continually evolve yet never reach a conclusion. This newer conceptualization of grief was adopted as a conceptual basis for the current study of families where a child has an NLTI. The concepts of chronic sorrow and anticipatory grief were also included in the conceptualization in anticipation that they may have relevance for working with these families. Perspective on Children and Life-Threatening Illnesses Children were central in this study. It is critical, therefore, to consider varying perspectives on children and life-threatening illnesses. Different perspectives will lead to alternative conceptualizations that could, in turn, lead to different research designs and data collection methods. In this study, children were considered active participants in the family. It was assumed that, providing they could communicate, children who were touched by life-threatening illnesses would be capable of thinking about and discussing death and dying. 15 Children's Knowledge About Death Early assumptions about how to deal with children who faced life-threatening illness were based on how well children developed their concept of death. Development of this concept was thought to parallel Piaget's sequence of periods of cognitive development (Piaget, 1960). Adults, therefore, thought that death meant separation to infants and toddlers. For pre-schoolers in the stage of preoperational thought, death would then be viewed as both separation and departure. Death would be like sleep, and dead people could feel when people visited their graves. These children would sometimes think that death was reversible. In the period of concrete operational thought, age 6-10 years, death would be seen as real, irreversible, and something that happened to other people. This group would tend to personify death and to think of it in terms of an outside agent. As children got older, and especially with the development of formal operational thought, death would be viewed as the end of life. It would involve irreversible decomposition of the body. In late adolescence, death would also be conceived as the end of personal time, as a cruel personal blow, and as the loss of newly developed intimate relationships (Bartholome, 1995). Based on the views that were prevalent early in the development of knowledge about children and death, professionals advocated a closed approach to protect children (Stevens, 1993; Waechter, 1971; 1985). Parents were told to maintain a sense of normality, to shield the child from knowing how seriously ill they were, and to remain cheerful and pretend that everything would be alright. However, some clinicians challenged the protective view (Bluebond-Langner, 1978; Spinetta, 1974; Waechter, 1985). They noted that ill children were anxious and knew about their disease. These clinicians observed that children who had the opportunity to discuss their illness, prognosis, and concerns seemed less anxious than children who were deprived of this opportunity. One study of children with life-threatening illness provides support for these observations (Waechter, 1971; 1985). Death anxiety among 64 children between the ages of 6 16 and 10 years was studied through the use of a projective test and anxiety scale. The children (matched for sex, age, race, social class, and family background) were divided into four groups: children with chronic disease who were expected to die; children with chronic disease, but a good prognosis; children with a brief illness; and non-hospitalized, well children. Only two children with fatal disease had been told their diagnosis and had discussed their concerns about death with their parents. Death imagery was found in 63% of the stories told by all of the children. Most of the children seemed aware of their prognosis despite efforts to protect them from knowledge about their disease. Those children with fatal illness who had more opportunity to discuss their fears and concerns also expressed less specific death anxiety. Although there are now over 50 studies of children's understanding of death, some inconsistencies exist among studies (Faulkner, 1993). Some of the discrepancies in results are related to methodological variations, such as some studies had small sample sizes or did not include a full age-range of children. In addition, many studies were limited to white, middle-class, suburban children and so, may not be applicable to other children. Difficulties also arise when trying to compare studies using disparate data collection methods including standardized interviews, descriptions of death-related pictures, and spontaneous or directed play. Although standardized interviews were most commonly used, variations in the difficulty and specificity of questions prevent direct comparison of results across studies. Inconsistencies also result from the difficulty in defining the phases that children pass through as their concept of death develops. Although level of cognitive development has been broadly linked to a child's understanding of death in some studies (Koocher, 1974), not all researchers agree (White, Elsom, & Prawat, 1978). Faulkner suggested that part of the difficulty arises because Piaget did not directly examine how children learn about death. Thus, researchers have had to extrapolate which cognitive skills a child needs to understand the components of a mature concept of death. 17 It appears that information about a particular child's understanding of death can only be obtained by discussing the issue with the individual child (Faulkner, 1993). Evidence suggests that a child's concept of death is related to his or her own experience with illness and its treatment, rather than only to the child's chronological age or stage of cognitive development (Reilly, Hasazi, & Bond, 1983). Children who have experience with death understand death in more depth and at an earlier age than children without this experience (Reilly et al.). In addition, the family's culture and environment, and children's wide variation in individual rates of development (physically, emotionally, cognitively, and intellectually) are also said to affect the child's developing concept (DeSpelder & Strickland, 1996). Communicating With Children About Death Professionals need to understand how a child's approach to death is related to developmental capacities, as well as to innate differences and past experiences. Even children as young as four years of age may be aware of the seriousness of their illness. In recent times, families tend to be more open in their communication, although actually discussing death may still be avoided. Goldman and Christie (1993) studied children over three years of age and their families (N=31) to determine whether they discussed the child's impending death together. Twenty six of the children had been told that their disease had recurred. In 28 families with siblings over three years of age, 75% had told the siblings about the impending death. Goldman and Christie also interviewed staff members (N=22) in the oncology unit about their attitude to an open approach in talking about death and dying and how often it occurred. Factors that influenced communication included the length of time the child had been ill and the family's communication style. Children who had been ill the longest were more aware of the severity of their illness. Each family with an open communication style acknowledged the impending death. All staff agreed or strongly agreed that open communication is the preferable approach. Some qualified it, however, with the proviso that parents should also be happy with this approach. Staff 18 thought that 10-80% (median of 45%) of children talked about their own death with their families. A lack of communication skills on the part of both staff and families may make it difficult to discuss death with the ill child. Additionally, maintaining openness may become increasingly difficult with relapse and impending death. Despite what many people think, children are not miniature adults. A child's reality is different from that of an adult. Children do not understand the world from an adult's perspective (Bartholome, 1995; Faulkner, 1993). They do not have a well-developed sense of self. Moreover, they live in a world where fantasy and reality are not well divided. Children cannot understand the concepts of chance and probability, so everything has an immediate and proximate cause. They are also egocentric. They relate everything that happens around them to their own feelings, ideas, and actions. Children may use concretistic thinking to sort out their world. They have their own sense of time. They have no sense of physical or personal time. Time is not a linear progression, but is collapsed into the present (Bartholome). Children also listen more to what adults do than to what they say, so non-verbal language is important. In turn, professionals and families need to look for cues to what the child wants to talk about. Moreover, they must be prepared to accept fears and anxieties related to loneliness, mutilation, and death that seem irrational from an adult perspective, because the fears will be quite real to the child (Faulkner). Based on clinical experience and available research, Stevens (1998) provided guidelines for caregivers of seriously ill and dying children. According to Stevens, these children know a lot about their situation, so it is ineffective and damaging to conceal information from them. Questions should be answered truthfully, especially since the child often knows the answer and is just seeking confirmation. Children will often accept information about their prognosis in a matter of fact way. Those who do not ask questions may be trying to protect their parents (Bartholome, 1995; Stevens). Infants who are too young to have a concept of death still need physical relief and comfort. Young children may have difficulty in expressing their fears of 19 separation. They will respond to the anxiety around them. Discussing fears and promoting family togetherness is believed to be important for children of 2-7 years. Expressive play with art and music may be helpful in encouraging children to share their fears and feelings (Gray, 1989). Children of 7-12 years of age may fear abandonment, destruction, and body mutilation. It is suggested that they need people to be truthful with them and to foster their sense of control over their deteriorating body. Friends are also important and should be encouraged to visit. One of the tasks of adolescence includes separation from parental control. Adolescents often struggle with the paradox of wanting support and yet wanting to meet challenges by themselves. They may prefer to confide in a peer group, especially with others in a similar situation. As with younger children, adolescents may be more concerned about their family and friends than about themselves. They tend not to be afraid of death so much as afraid of dying. Professionals are encouraged to listen carefully to understand the adolescents' perceptions of their own illness and prognosis. Privacy and a sense of independence is important to an adolescent, as is peer contact and support. Offering choices may give adolescents a sense of being in control, and may reduce anger, frustration, depression, and anxiety. Writing poems, letters, or a journal or drawing can release pent-up emotions. Many adolescents want to know that they will be remembered. Creating a permanent record to leave behind, such as video, tape, or photograph, may be valuable (Stevens). While these guidelines are helpful, further research is needed to evaluate the efficacy of these approaches. Siblings. Surviving siblings are particularly vulnerable. They should not be forgotten (Gibbons, 1992). They are also children and so their reality is that of a child. Consequently, siblings may have their own version of what caused the child's illness. They may have misconceptions about the nature of the illness or about the hospital clinic and treatment program (Sourkes, 1980; Stevens, 1998). Sourkes also found in her study that siblings may have a fear of developing the 20 same illness. Yet at the same time, they may feel guilty and ashamed because of their relief that they do not have the illness. In addition, siblings may have ambivalent feelings about the ill child, such as envy or shame. They may be so stressed by the illness that their schoolwork and relationships suffer (Sourkes; Stevens). B. Davies and Martinson (1989) reported from their research with siblings of children who had died of cancer that emotional realignment within the family, family separation, and the ill child's therapeutic regimen can be very stressful for siblings. These stressors may result in negative consequences such as feelings of loneliness, resentment, rejection or neglect, frustration, fear, and sadness. Some clinical observations suggest that parents may be unable to respond to siblings for a variety of reasons (Gibbons). Parents may be unavailable or detached, may cease family celebrations, may feel inadequately prepared to discuss death with siblings, may think siblings are too young to understand, or may want to protect siblings. Moreover, siblings may interpret a lack of communication as an accusation that somehow they are at fault. The lack of communication may isolate siblings even more. Some parents may become overprotective of siblings. Or, they may withdraw from parental involvement because they are afraid to lose this child too (Gibbons). B. Davies (1988) reported that the family is critical in creating the environment in which siblings respond to the death of a child since children are dependent on their parents (and adults) for information and support. They often do not have any other source of help. In her study of 34 families 2-36 months following the death of a child, B. Davies used semi-structured interviews and standardized instruments to examine family environments. She discovered that siblings in the more cohesive, active, and religious families had fewer behavioural problems. B. Davies suggested that families who participated more in social activities were promoting adaptation in siblings. Social support was also noted as important for the siblings. B. Davies suggested that professionals could contribute to sibling adaptation by encouraging and facilitating family cohesion, having information sessions with the family and not just with the parents, asking 21 family members to assist in the care of the dying child, and treating the patient and the family as one unit for care. In conclusion, professional understanding of how children develop their concept of death has changed over the years. It is now recognized that a child's perception of reality and concept of death is not the same as an adult's. Some theorists and clinicians recognize the importance of open communication instead of refusing to discuss death. Siblings are also children. They are vulnerable when a child is dying and should not be forgotten. Clinical observations and available research have provided an important foundation for the development of guidelines to assist professionals in supporting ill children and siblings (Stevens, 1998). Few of these recommendations, however, have been adequately evaluated. Perspective on Families Family means different things to different people. There are numerous definitions of family, such as those that emphasize relationships through blood ties, adoption, guardianship, or marriage. Other definitions focus on social or emotional relationships between at least two people in a single household. Earlier definitions of family reflected the traditional nuclear and extended family. Definitions have become broader in the past 10-15 years, and now include a variety of possibilities, such as single-parent families, blended families, and same-sex families (Friedman, 1998; Hanson & Boyd, 1996). In this study, the family was viewed as a complex unit that is composed of individual members, but that has distinct attributes of its own. Ideally, the family would be self-defined and include whoever the family says it includes. The family unit in this study, however, was limited by the researcher to those members with whom health care professionals most often interact and who usually play a key role in pediatric palliative care - the ill child, parents, and siblings. There are also several different approaches to family nursing. These perspectives include the family as the context for individual development; the family as the unit of care; the family as 22 a system; and the family as a component of society (Friedman, 1998; Hanson & Boyd, 1996). In this study, the family was viewed as an interactional system in which the whole is more than the sum of the parts. It was assumed that systems data are needed in order to provide care to all family members, as well as to the family as a unit (Feldstein & Rait, 1992; Friedman). Family systems theory provided a useful guide for examining families' interactional relationships (Friedman; Jassak, 1992), and for directing the process of data collection. The following overview of systems theory further explicates the perspective on families that formed part of the context for this study. Systems Theory Systems theory explains the behaviour of complex, organized systems. It is a way of looking at the world in which objects are interrelated with one another, von Bertalanffy (1968; 1975) recognized living systems as a complex organization of many parts. He sought to account for the seemingly animate behaviour of non-living systems in terms of complex inter-relationships and transactions among components. Holism is a key feature in systems theory. A system must be understood as a whole. It cannot be comprehended by examining its individual parts in isolation from each other. There are properties of the system that do not derive from the components and only emerge at the systemic level. Human systems are self-reflexive. Thus, they can examine their own system and set goals for themselves. Communication is extremely important and humans behave according to the meaning a thing has for them (von Bertalanffy). Transactional patterns, rather than individual responses, are of interest in families. Family processes can be understood as a product of the entire system, instead of any one person. These overall patterns of interaction within families, rather than each separate member's behaviour, illuminate family processes and reflect the family system (Forchuk & Dorsay, 1995). When one examines family systems within a hierarchy of systems (subsystems are smaller than the family unit and suprasystems are larger than the family), then one can explore questions in more depth 23 and can explore how family systems change. Morphogenesis is the process of family change by the whole family system, rather than by just a single family member. First-order change is when individual behaviour changes without the completion of a positive feedback loop. Whereas second-order change is a higher more major level of change that can alter the system itself. A system is a set[s] of elements standing in interrelation among themselves and with the environment (von Bertalanffy, 1968; 1975). A system, therefore, is either a set of interrelationships or a set of components, depending on the level of analysis. In family research, individual family members are the components. They are interdependent, and behaviours of the components exhibit mutual influence. That is, each member's behaviour affects every other member of the family. Hierarchies and the issue of power are important in families; for example, a parental subsystem may exert more power than an offspring subsystem. Systems require inputs, throughputs, and outputs that are interrelated by rules of transformation that take place in systems. Families may become immobilized because they do not have a family rule to use in dealing with a particular situation, such as when a child is dying. Equifinality is the ability of a system to achieve the same goals, but through different routes. Goal-oriented behaviour results from communication of information among components and from how this communication is organized, especially through feedback loops (Friedman, 1998). Feedback loops are closed circuits or paths in which information goes from one part, through others, and back to the origin. A negative loop operates to restore or maintain equilibrium (homeostasis), while a positive loop increases deviation from the status quo. If the structure of the system stays the same, then it is morphostatic; if the structure changes, then it is morphogenetic (von Bertalanffy, 1968; 1975). Although overall control of the behaviour of the system as a whole does not reside in one component, a single component can become increasingly important in determining the behaviour of the whole (progressive centralization). Thus, a dying child may exert unusual influence on the family. 24 Boundaries. Boundaries are an essential concept in systems theory. If one can identify components of a system, then one is really drawing boundaries around them. A boundary defines the system and marks the interface between the system and its environment. Human systems are embedded in the environment, so one can only gain understanding of families if one takes into account the sociocultural, historical, political, and economic contexts (Friedman, 1998). Boundaries can be determined by persons inside or outside the system. However, there may not be consensus about the boundaries because determining a boundary requires self-reflexivity and some degree of judgement. Systems are said to be open or closed depending on their boundaries. Boundaries are impermeable (closed to interchange from outside the system), semi-permeable (allowing some interchange), or amorphous or transparent (offering no impediments to interchange). Al l families are open systems, although in varying degrees. One needs to determine what or who permeates the boundaries. In many family assessments, boundaries are operationalized as: (1) degree of permeability or, from another perspective, of cohesiveness; and (2) emotional connectedness among family members (Klein & White, 1996). The rigidity of the family structure affects how smoothly families can adjust to entrances and exits across the family boundaries. When there is boundary ambiguity, such as might occur as a child is dying, then uncertainty about who is in or out of the family system may be immobilizing. Whether one believes that it is the person per se or their roles in the system that are important to the family system makes a difference to the meaning of a family member's death. If one believes in the importance of roles, then someone else will be able to fill the empty role. However, if the individual person is more important, then no one can fill the empty place. Family Research In pediatric palliative care, the unit of care is the family and not just the individual child. Clinicians and researchers recognize that patients exist within a family system and also within a 25 myriad of social systems (Clarke, 1995; Friedman, 1998; Gulla, 1992). Until recently however, families have been neglected in health care research. There is little research and guidance available to assist health professionals in providing family-level care (B. Davies, Reimer, & Martens, 1994; Friedman). While family-focused care often implies that health professionals must view the family solely as a unit, it is important to note that one cannot view the unit as a whole without giving consideration to the individuals who make up the unit (B. Davies, Reimer, Brown, & Martens, 1995; Friedman). The definition of family must be explicit in family research. Further, articulation of a conceptual framework guides research design and methodology decisions (Boyd, 1996). Data from families should be collected and analyzed in a manner that is consistent with the researcher's conceptualization of the family (Copeland & White, 1991; Gilliss & Davis, 1992). In this study, family was defined as a complex unit composed of individual members, but with distinct attributes of its own. Accordingly, data needed to be collected from individual members, as well as from the family as a unit. In addition, analysis had to take both the individual and the family unit perspectives into account. Family researchers have primarily focused on individuals, rather than the entire family unit (Broome, 1998; Leahey & Wright, 1987). The underlying assumption has been that if one knows an individual's role, such as mother or father, then one can determine the behaviours of the unit (Gilliss & Davis, 1992). Researchers have tended to make inferences about the family based on only the mother's experiences (Birenbaum, 1995; Kristjanson, 1992). While Birenbaum suggested that, at times, one member of the family may be the best informant about a family phenomenon, relying on reports of a single individual or subset has the potential to present a biased view of the family when the family unit is the focus of study (Lynn, 1995). In this research study, the views of the individual members and the family as a whole were all considered significant. Therefore, all members of the family were encouraged to participate to 26 the maximum extent possible, recognizing that participation was restricted by age, developmental stage, and cognitive ability. Obtaining accurate information from the family about the family can pose a problem for researchers. The information obtained from each source of data may be different and there may be a lack of agreement among the family members. One may find a lack of correspondence between individual reports and those of the family as a whole. However, although one may get different views, they may be complementary to one another. Uphold and Strickland (1989) suggested that obtaining information from multiple members of a family is beneficial and performs a kind of triangulation that gives a better picture of the whole. Triangulation, either using multiple data sources or multiple methods to obtain data, may improve the researcher's ability to find out about the families' experiences (Woods & Lewis, 1992). Both participant observation and interview methods of data collection were used in this current study. Sources of data also included individual family members as well as the family as a group. A criticism of this approach centres around the meaning of triangulation and the implicit expected outcomes. Triangulation implies that one is focusing on a central point, in this case \"the family experience,\" in an attempt to discover \"the truth.\" However, there may be multiple \"truths\" within any family. The experiences of less powerful members of the family, such as children, have tended to be ignored in the past. Although involving children may require creative methods of data collection such as the use of drawings or poems, they are able to and should be encouraged to contribute. Family research needs to value the contribution of all members of the group, and to view their perceptions of the situation as being equally relevant (B. Davies et al., 1994). If different members report different \"facts,\" the researcher should not then decide which person carries more weight and is more \"truthful\" in representing the family experience. Instead, as in this study, specific \"facts\" should be viewed as part of the data. 27 Research examining the experience of the family as a whole must reflect the dynamism in family systems, and must use measures that portray families as entire units (Broome, 1998; Ferketich & Mercer, 1992; Woods & Lewis, 1992). However, when one's level of inquiry involves the family as more than the sum of its parts, no adequate tool exists to measure the whole. It has been noted that though family theorists and therapists are concerned with family systems and family psychopathology, no widely accepted concepts of relational and systemic pathology have yet been developed (Cowan, Cowan, & Schulz, 1996). Complex systemic models that combine conceptual, measurement, and statistical techniques require further investigation (Cowan et al.). In quantitative research particularly, questions are raised about how to combine individual family member data into scores that reflect the family as a unit (Clarke, 1995; Ferketich & Mercer; Uphold & Strickland, 1989). For example, is a summative score derived from individual scores an accurate representation of a family's score? Can one obtain a score about a family simply by aggregating the individual family member's score on a given instrument? Is the maximised family score a good indicator of impacts on the family as a whole? What value is there in highlighting differences in scores between family members as the central concern of a study? These and similar questions raise issues that continue to be debated. There are unique weaknesses and strengths in each approach (Uphold & Strickland). Currently, no adequate answers exist to these questions. Again, one must base the choice of sources of data on the purpose of the study, the theoretical basis for the study, and the population of interest. Since the definition of family and its boundaries is often artificially and externally imposed and may not reflect the boundaries that would be meaningful to the family, Kristjanson (1992) suggested that research based on families' own definitions of their membership may be useful for some research questions. Moreover, family research is transactional, thus one should examine patterns of family behaviours (Kristjanson). However, the context of the data collection must also be taken into account because individuals may respond to the same question in 28 different ways if asked individually or in a family group (Astedt-Kurki & Hopia, 1996; Kristjanson). Birenbaum (1995) suggested that grounded theory methodology, involving data collection from both the individual family members and from the group as a whole, is an efficient way of collecting and analyzing family-level concepts. Grounded theory, involving data collection from both individual family members and from the family as a whole, was the method chosen for this study. Families differentiate and carry out their functions through subsystems within the family, such as spousal, parental, and sibling subsystems. Loss, actual and potential, is a major transition that disrupts patterns of interaction and requires family reorganization (Martinson, McClowry, B. Davies, & Kuhlenkamp, 1994; Walsh & McGoldrick, 1988). \"Off-time\" losses, such as the death of a child, are often viewed as the most painful ones. However, although members of a family may be affected by the terminal illness, one cannot assume that all members are adversely affected (Birenbaum, 1995). Members of the family are often not affected in the same way, nor do they respond in a similar fashion. Birenbaum called for research to identify who is affected and why, because some individuals seem capable of buffering themselves against an event. Within a family, the unique constellation of relationships probably affects the impact of the death on the family members and on the family as a whole. Therefore, one must examine the relationships pertaining to each family member, as well as examining the family group as a whole and distinct entity. Families do not exist in isolation from the larger community. Particularly as a child's illness progresses, families may require increasing external help and support (Jefidoff & Gasner, 1993). Families must then seek treatment, information, and services for the child and themselves outside of the family, while concurrently managing these systems. For example, families in Jefidoff and Gasner's study needed to discuss their fears about the child's death. Yet, they also needed to prepare other family members and friends for the imminent death. Clinicians have 29 observed that the interaction patterns of family members, their interactions with others outside the family system, such as friends and neighbours, and the degree of social support may influence a family's ability to cope with the experience of having a terminally ill member (Ross-Alaolmolki, 1985). The interaction between the family system and the health care delivery system may also be important. It may have a direct bearing on how patients and families come to understand their roles during illness (Speedling, 1983). Hospital norms and regulations may implicitly and explicitly affect the family's access to information. The attitudes of personnel may influence the family's ability to develop competence in performing the required roles. Larger systems, such as hospital wards and physician-family relationships, may foster dependency on the medical system and preoccupation with caregiving. Thus, families may be prevented from developing new and flexible roles (Rosman, 1988). Speedling suggested that, \"Observing families as they live through the process [of illness] can provide vital information about what they need at different stages of the process and how the current system for delivering care affects the quality of coping\" (p. 75). The current study used this type of observation to uncover such information. To summarize, little research has been from the perspective of the whole family. Most research has been retrospective and has involved only the parents, and sometimes only the mother, to provide a family perspective. Yet, children perceive the world differently from adults. Thus, the children's perspectives must also be sought. Moreover, the literature is unclear in offering guidance to assist health professionals in providing family care and conducting family research. A central issue in family research concerns how individual reports on experiences can be used to provide insights about the family unit. Recently, researchers have suggested that the views of all family members should be treated as equally important, and differences between accounts should be viewed as further data for a higher level of analysis. 30 Perspective on C o p i n g The individual coping literature is wel l established, but family coping remains a developing concept (Birenbaum, 1991; C o w a n et al . , 1996). F a m i l y coping is complex and difficult to measure. For example, the members o f a family who participate in the family's coping process are at various developmental levels. H o w , then, can one use similar data col lect ion methods when fami ly members may range in age f rom infants to the elderly? Simi lar issues have been raised about developing instruments to assess fami ly coping. For example, can one describe fami ly cop ing s imply by aggregating the individual fami ly member's styles o f coping scores? C a n issues o f power within families be accounted for so that chi ldren have a voice? O r should responses be weighted to reflect the status o f individuals within the fami ly? Despite the emphasis on fami ly as the unit o f analysis, the perceptions o f individuals continue to be the focus o f fami ly coping measurement (Birenbaum, 1991; B r o o m e , 1998; C o w a n et al . , 1996). M o s t are designed as individual , self-report instruments to assess perceptions about the fami ly as a unit (Kristjanson, 1992). Thus , they provide relational-level data that are not necessarily meaningful when exploring interactional relationships (Sul l ivan & Fawcett, 1991). In addit ion, fami ly coping instruments are often just administered to parents (Birenbaum). The role o f children in family coping, therefore, is often ignored. Because o f the continuing use o f individuals to represent the family, it is important to identify what is known about individual coping. Th is section wi l l begin with examination o f one model o f individual coping that dominates current theorizing in the f ield o f stress and coping. Limitat ions o f this model for the current study wi l l be identified prior to examining a family model that is prominent in the existing fami ly coping literature. Individual C o p i n g T h e transactional mode l o f coping proposed by Lazarus and F o l k m a n (1984) is a model o f individual coping that is wel l known in health care and in the stress f ield. In this mode l , two 31 processes, cognitive appraisal and coping, mediate the person-environment relationship. These processes then determine adaptational outcomes. Cognitive appraisal determines whether and to what extent a particular transaction or series of transactions between the person and the environment is stressful. There are three types of cognitive appraisal: primary appraisal; secondary appraisal; and reappraisal. Primary appraisal allows the person to classify the transaction as irrelevant, benign-positive, or stressful. Secondary appraisal involves the evaluation of what might and what can be done. Secondary appraisal takes into account which coping options are available, the outcome expectancy, the efficacy expectancy, and the consequences of using a particular strategy. Reappraisal is \".. .a changed appraisal on the basis of new information from the environment, which may resist or nourish pressures on the person, and/or information from the person's own reactions\" (Lazarus & Folkman, p.38). Reappraisal indicates that appraisal of the situation may change over time. Many factors affect cognitive appraisal. Person and situational variables are inter-dependent. Among the most important person factors are commitments and beliefs. They help to determine the saliency of an encounter in relation to well-being. They also shape the person's understanding of an event and, therefore, his or her coping efforts. Commitments also influence appraisal through their impact on vulnerability. Lazarus and Folkman (1984) state that, \"The greater the strength of a commitment, the more vulnerable the person is to psychological stress in the area of that commitment\" (p.58). Beliefs about personal control and existential beliefs are particularly relevant to appraisal. Situational variables are interdependent. Together with personal variables, they contribute to an appraisal of harm/loss, threat or challenge. Situation variables include novelty, predictability, event uncertainty, temporal factors, ambiguity, and timing of the stressful events in relation to the life cycle (Lazarus & Folkman). Coping is affected by cognitive appraisals. Lazarus and Folkman (1984) define coping as \".. .constantly changing cognitive and behavioural efforts to manage specific external and/or 32 internal demands that are appraised as taxing or exceeding the resources of the person\" (p. 141). Coping functions are emotion-focused or problem-focused. These functions may occur simultaneously, and may either facilitate or impede the coping outcomes. Coping is a process that implies effort. No strategy is better than another one. When judging effectiveness, consideration must be given to the context and the frame of reference. Effective coping must be judged from the client's perspective, rather than only from the researcher's opinions or values about how the person should be coping. Effective coping is not equated with mastery over the environment. The dynamics and the change that characterize coping as a process are the function of the continual appraisals and reappraisals of the changing person-environment relationships. The major adaptational outcomes are social functioning, morale, and somatic illness. Each outcome can be conceptualized as having both a short-term and a long-term component, with short-term outcomes reflecting coping with a specific encounter. Relationships between the short- and long-term outcomes are complex in that the nature of the short-term outcomes are not necessarily associated with similar long-term outcomes. The relationships between the three major adaptational outcomes are also complex. Good functioning in one area may be at the expense of functioning in another area (Lazarus & Folkman, 1984). The basic assumptions of this transactional model of coping (Lazarus & Folkman, 1984) maintain that coping is effortful on the part of an individual; is situationally focused; and is accessible to the consciousness, so can be reported. This model, therefore, can be used to describe an individual's response to an acute episode or event. Families who are living with a child with an NLTI may face acute episodes, but they also face persistent demands that are part of their daily lives. How, then, can a set of behaviours and cognitions be described as coping, as opposed to ordinary living? Moreover, Gottlieb (1997) suggested that coping associated with recurrent or persistent life difficulties may be more or less effortful or automatic depending on when questions are asked. Coping efforts eventually become incorporated into one's daily life. 33 They are no longer associated with a specific stressor. Individuals, therefore, may not be aware that they continue to use a particular way of coping. Consequently, the Lazarus and Folkman model has limited applicability in research with families living with a child who has an NLTI. First, reliance on the individual's perspective fails to explore the family's perspective. Second, coping is only part of a family's life. Using this framework would restrict the scope of the study. Thus, important aspects of families' lives may be overlooked. Finally, the possibility that individuals routinize some coping efforts, and so are unable to consciously identify them, suggests that the researcher may disregard or fail to uncover certain coping efforts employed by families if the focus is on conscious coping efforts. Given that individual coping has not been shown to reflect a family process, individual coping frameworks were not viewed as appropriate for this current study. In keeping with a family systems view of the family, theoretical frameworks about family coping that are based on systems theory were reviewed. The tenets of one major family coping framework are discussed in this section to highlight its strengths and limitations for this study. Family Coping The Resiliency Model of Family Stress, Adjustment, and Adaptation is a family coping framework that is widely used in health care (McCubbin & McCubbin, 1993). This model builds on the Double ABCX model of family stress theory developed by McCubbin and Patterson (1982), which in turn was expanded from the original family stress theory proposed by Hill (1949). In earlier renditions of McCubbin and Patterson's model (1982), \"A\" denoted a stressful event with associated hardships; \"B\" represented the family's physical, psychological, material, social, spiritual, and informational resources; \"C\" demonstrated the family's subjective definition of the stressful event; and \"X\" denoted a crisis, and the amount of disruptiveness or incapacitation within the family due to the stressful event. The models are based on a number of assumptions: (1) the family is a system; (2) unexpected or unplanned events are perceived as 34 more stressful than other events; (3) stressful events within the family are more disruptive than those outside of the family; (4) lack of previous experience with a stressor leads to greater perceived stressfulness; and (5) ambiguous stressor events are more stressful than nonambiguous ones. In the Double ABCX Model, the \"C\" factor was expanded to \"CC\" and was defined as the family's perception of the original stressor and of the pile-up of other stressors (\"AA\"), plus the family's perceptions of its resources (\"BB\"). The concept of coping was incorporated into the Double ABCX Model and was perceived as an adaptive process in families wherein resources, perceptions, and behaviour interacted (McCubbin & Patterson, 1982). Coping included both cognitive and behavioural strategies. Altering the meaning of the situation was viewed as one way in which families coped with situations where demands exceeded capabilities. In addition, a more generalized meaning construct was introduced to the Double ABCX Model. This construct was a sense of coherence (SOC) which was based on the salutogenic model of health proposed by Antonovsky (1979) in answer to the prevailing pathogenic model. Antonovsky suggested that it is important to examine how people successfully cope with stressors and what influence that successful coping has on health. He developed the SOC and defined it as a global orientation which sees the world as more or less comprehensible, manageable, and meaningful. According to Antonovsky, the SOC is a determinant variable that affects health consequences and other aspects of well-being. The Double ABCX Model was eventually renamed the Resiliency Model of Family Stress, Adjustment, and Adaptation to emphasize adaptation as the central outcome of the stress process (McCubbin & McCubbin, 1993). This emphasis on potential positive outcomes was consistent with the influence of Antonovsky's work (1979; 1987) on individual and family resilience, and the results of other researchers' studies of resilience. Two levels of meaning were differentiated. Situational meanings were defined as the individual's and family's subjective 35 definitions of demands and capabilities, as well as the relationship between them. Global meanings were viewed as transcending any given situation and reflected a more stable set of cognitive beliefs about the family's relationship to society and the relationship among family members. The goals of the family unit's behavioural responses were to eliminate stressors, manage hardships, resolve intrafamilial conflict, and develop resources to facilitate family adaptation over time (McCubbin & McCubbin). There are two phases in the Resiliency Model (McCubbin & McCubbin, 1993). The Adjustment Phase starts with an illness stressor that interacts with a family's vulnerability. This family vulnerability to life changes and pile-up of stressors interacts with the family's type and established patterns of functioning. These components then interact with the family resistance resources, such as the quality of communication between husband and wife. This, in turn, interacts with the family's appraisal of the illness stressor and its severity. Finally, interaction continues with the family's problem solving and coping strategies. The result of these interactions is either good adjustment with positive outcomes, or a maladjusted crisis situation. Family crisis may be viewed as an expected outcome of the numerous stressors that often accompany chronic illness. It requires a change in patterns of family functioning that denotes the beginning of the Adaptation Phase of the Resiliency Model (McCubbin & McCubbin). The adaptation phase is also determined by a number of interacting components. The pile-up of stressors (\"AA\") interacts with the family's resiliency that is partly determined by the family's type and newly instituted patterns of functioning. These components interact with the family's resources and capabilities (\"BB\") which are supported by family and friends (\"BBB\") and the family's appraisals. The family's situational appraisal (\"CC\") then interacts with the family's schema appraisal (\"CCC\") and creates meaning. Finally, the resource and appraisal components interact with the family's problem-solving and coping strategies to facilitate adaptation. Family adaptation is the outcome of efforts to achieve coherence and functioning in a 36 family crisis situation. These efforts are directed at achieving balance and fit at both the individual-family level and the family-community level. Some families will become well-adjusted over time, while others will remain in a maladjusted crisis situation (\"XX\"). According to McCubbin and McCubbin (1993), families are more likely to adapt successfully to an illness stressor if they are less vulnerable because fewer other stressors or changes are occurring concurrently; if they have more adaptive patterns of functioning, such as more emotional closeness and more flexibility in roles, boundaries, and rules; if they define the situation positively and feel they can master or control it; and if they have good coping and communication skills. Further, a family's SOC is a mediating coping factor between family stressors and adaptation. The higher a family's sense of coherence, the more likelihood of successful adaptation (McCubbin & McCubbin). The Resiliency Model of Family Stress, Adjustment, and Adaptation (McCubbin & McCubbin, 1993) is a model that is undergoing continual revision in response to research evidence. It provides a somewhat useful framework in which to examine family experiences of a child's progressive, life-threatening illness. However, the framework has some significant limitations for this population that must be addressed. First, most of the research on the model has been done with healthy families (Danielson, Hamel-Bissell, & Winstead-Fry, 1993). Recent research has begun to examine families where a member has a disability or chronic illness (Patterson & Leonard, 1994; Patterson, Leonard, & Titus, 1992), but there is no research on progressive, life-threatening illnesses. In addition, since there is very little research at all in pediatric NLTIs, one must be open to discovering the potential scope of families' experiences. The use of a theoretical framework may bias researchers and may prevent or restrict them from seeing the richness and the full extent of families' experiences when a child has an NLTI. Second, the model is based on enhancing understanding of family adjustment and adaptation responses to stressful situations. While this goal is laudable, the emphasis on responses to 37 stressors and coping strategies has limited meaning when one is trying to understand a family's experiences of living with a child who is dying. Understanding experiences encompasses coping, but also includes the fuller and richer description and theoretical formulation of family life. Finally, one of the major concepts in the model, the resiliency concept of SOC, is based on research with individuals. Family global meanings have not yet been operationalized and tested. It was, therefore, premature and constraining to use the Resiliency Model as a framework in this current study which examined the experiences of families as a whole, rather than the perception of only one family member. Summary The conceptual framework for this study was based on particular perspectives on children and life-threatening illnesses, families, and coping. The focus of this study was where these three perspectives intersected within the broader context of grief. Grief was viewed as an active, dynamic process that may continually evolve, yet never reach a conclusion. This more recent understanding of grief challenges professionals to reconsider the needs of families when a child has an NLTI. In addition, the kind of interventions that will be most helpful should also be reconsidered. The concepts of chronic sorrow and anticipatory grief were also considered potentially relevant to families where a child has an NLTI. Children, both ill children and siblings, were considered active participants in the family. It was assumed that, providing they could communicate, children who were touched by life-threatening illnesses would be capable of discussing issues related to death and dying. Open communication, rather than hiding death, was considered important. Professional understanding of how children develop their concept of death has changed over the years. It was recognized that a child's perception of reality and concept of death is not the same as an adult's. The vulnerability of siblings when a child is dying was acknowledged. 38 In this study, systems thinking informed the perspective on families. The family was viewed as a complex unit that is composed of individual members, but that has distinct attributes of its own. The unit of care was seen as the family and not just the individual child. The unit as a whole could not be viewed without considering the individuals who made up the unit. The contribution of all members of the group was valued. Their perceptions of the situation were viewed as being equally relevant. This perspective on family guided research design and methodological decisions. Little research has been conducted from the perspective of the whole family. Most family research has been retrospective and has involved only the parents, and sometimes only the mother. Currently, there is little research and guidance available to assist health professionals in providing family-level care. The final concept underlying this study was coping. The individual coping literature is well established, but family coping remains a developing concept. Family coping is complex and difficult to measure. Despite the emphasis on family as the unit of analysis, the perceptions of individuals continue to be the focus of family coping measurement. Neither individual nor family coping frameworks were perceived as appropriate for this current study. The emphasis on responses to stressors and coping strategies had limited meaning because it was considered that understanding family experiences encompasses coping, but also includes a fuller and richer description and theoretical formulation of family life. Chapter III: Literature Review Some researchers (Davies, 1996; Martinson, 1993; Parker, 1996; Stein, Forrest, Woolley, & Baum, 1989; Stein & Woolley, 1990) have begun to investigate the impact on families of caring for a child with a progressive, life-threatening illness (PLTI) at home. However, existing research in pediatric palliative care is minimal. There is little known about families' experiences when a child has a PLTI. Even less has been written about experiences with a neurodegenerative, life-threatening illness (NLTI). Only two studies were found that focused specifically on families of children with neurodegenerative disorders (Davies; Parker). Given the paucity of specific research, the literature review was expanded to include research in the related area of family experiences when a child has a chronic illness or disability. It was assumed that there are some similarities between these families and families where a child has a PLTI. For example, families of chronically ill children also provide most of their child's care at home. An important difference, however, is that families of children with PLTIs must prepare themselves for their child's death. Moreover, the ill child and siblings must also deal with this actuality. Only limited direction, therefore, can be taken from the extensive literature on chronic illness and disability. The available research does suggest that a child's illness has a profound impact on every dimension of family life. Families may be affected emotionally, physically, financially, and spiritually. Moreover, their structures and ways of organization may be permanently altered. The numbers of children with a prolonged terminal illness are low when compared with those in the adult population. There were, for example, 346 children with terminal illnesses who died in British Columbia over the three year period 1988-1990. At any one time, there are over 200 children in the province who are living with progressive, life-threatening illnesses (B. Davies, 1992). In Ontario, a home-based, pediatric palliative care program established in a major teaching hospital only admitted an average of 16 patients a year over a period of eight years (Kopecky, Jacobson, Joshi, Martin, & Koren, 1997). Yet, one estimate from the United Kingdom 40 indicates that 1:1000 children are affected by life-threatening illnesses (Goldman, 1996). Approximately 20% of these children have cancer, but many have progressive neuromuscular or neurodegenerative conditions that will eventually cause their death, though not necessarily within a few months or even a few years (Ashby et al., 1991; B. Davies & Howell, 1998; Goldman). Families, therefore, provide care for extensive periods of time. Although the current numbers are relatively small, these children often pose substantial management challenges (Chambers, Oakhill, Cornish, & Curnick, 1989; Hain, Patel, Crabtree, & Pinkerton, 1995; Hunt & Burne, 1995; Hunt, 1990; Kopecky et al.), and may use a disproportionate amount of in-patient health resources (Caring Institute of the Foundation for Hospice and Home Care, 1987). Additionally, the numbers of children with a PLTI are projected to increase as the incidence of life-threatening diseases rises and as advances in technology and medicine reduce mortality rates for conditions from which children previously died (Broome, 1998; B. Davies & Howell). In this chapter, the literature is reviewed to provide an overview of what is known about families' experiences when a child has a PLTI. First, the potential consequences of living with a child who has a PLTI are detailed to illustrate the possible impacts of an NLTI on families' experiences. Second, factors that may influence the experience are examined to provide a context for understanding families' experiences. Research from the extensive chronic illness and disability literature is used to supplement findings from pediatric palliative care studies. Gaps in knowledge are identified, especially in relation to neurodegenerative illnesses. Particular attention is paid to the experiences of families at home because most of the care for these children occurs at home. Impact of Families' Experiences There is limited knowledge available about the potential consequences of living with a child who has a PLTI. The impact of families' experiences has not been well documented. Indeed, the majority of available research only appeared in the literature within the past few 41 years. This beginning research tentatively suggests, however, that families' experiences have five major impacts. Living with a child who has a PLTI may affect a family's structure and patterns of interaction and may cause emotional, physical, financial, and spiritual disruptions. Research in each of these areas is presented to provide an understanding of the potential impacts of families' experiences when a child has an NLTI. Impact on Family Structure and Patterns of Interaction A child's life-threatening illness may disrupt family patterns of interaction, require family reorganization, and pose shared adaptational changes. Some researchers have found that changes in family roles become necessary in order to manage the child's illness (Clarke-Steffen, 1997; Gravelle, 1997; Stein et al., 1989; Stein & Woolley, 1990). Clarke-Steffen interviewed mothers, fathers, ill children, and siblings in a longitudinal, grounded theory study of seven families in which a child was recently diagnosed with cancer. Two semi-structured interviews were completed with each individual, then the family was interviewed as a group in the final interview. She reported that family roles were drastically altered in some families. Some role changes were planned, but many others occurred spontaneously. Typical role changes included mothers changing jobs or stopping work to stay at home with the ill child. Fathers were more likely to rearrange their schedules or take time off work when needed. The bulk of the ill child's care fell to mothers, while fathers often took responsibility for the siblings and increased their household chores. Siblings were sometimes asked to spend more time on the ill child's care and to take on more responsibilities in the household. Similar findings were reported by Gravelle who used a phenomenological approach in her study with eight families of children with a range of PLTIs. Eight mothers and three fathers participated in an unstructured interview, and five of them completed second interviews. Mothers gave up work, sometimes unwillingly, to care for the child. Some felt that they had no other choice. Mothers who assumed their new role unwillingly were often resentful and angry. Stein and Woolley also reported that mothers were 42 usually the primary caregivers. Although fathers generally continued working, they often used vacation time or took time off work to attend to the child's needs. Some mothers felt particularly housebound because of exhaustion, concern about leaving the child safely in someone else's care, and worry about finding suitable child care. Moreover, these parents also feared leaving their children in case they died unexpectedly. Siblings were sometimes expected to assume extra responsibilities. Consequently, siblings had less free time and fewer friends than their peers (Stein & Woolley). In many families, the daily routines of family life may be disrupted because families need to follow routines imposed by the child's needs (Parker, 1996; Stein & Woolley, 1990). Families, therefore, may suffer from a lack of spontaneity in their lives. They may be unable to do anything on the spur of the moment. Plans may be interrupted because of the child's changing needs. When Parker interviewed five married couples and two married women, she discovered that parents of children with a rare NLTI called tuberous sclerosis complex regretted this lack of spontaneity in their lives. Further, these parents felt that their lives were incomparable with the lives of other families because of this disruption. Stein and Woolley also noted that the complex medical needs of the ill children took precedence over family routines. In one of the first studies in pediatric palliative care (Stein et al., 1989), parents (n=25 families) in the first free-standing pediatric hospice in the world, Helen House in England, were found to experience more psychological difficulties, marital stress, financial, and employment problems than would have been expected from normal samples. Researchers used standardized instruments and semi-structured questionnaires in this retrospective study to examine parents' perceptions of the care offered and the impact of chronic and life threatening illnesses. Only parents were included in the study, although they were asked to respond to questions about the ill children and siblings. In 20 families, the child was still alive. Five out of 25 families experienced significant marital problems. A prospective study was also done in the same hospice to examine 43 parents' perceptions of the care offered and the impact of a child's life threatening illness (Stein & Woolley, 1990). Twenty-one families from Helen House were matched with a control group who were not receiving or intending to receive hospice services. The groups were individually matched for the child's diagnosis, age, length of illness, and social class. At the time of referral to the hospice, hospice families perceived less overall support compared with the control group (Stein & Woolley). Similar difficulties to those found in the previous study (Stein et al.) were experienced in both the hospice group and the control group (Stein & Woolley). Parents also worried about their child's symptoms, the course of the illness, and about caring for the child if their own health failed (Stein & Woolley). Family members may focus on the ill child to the exclusion of virtually everything else, including their relationships with each other, extended family, and friends (Gravelle, 1997). During their research at Helen House, Stein et al. (1989) and Stein and Woolley (1990) discovered that families of children with a PLTI had no time for social and leisure activities. Parents were unable to do anything as a couple or as a family when the focus was on the ill child. Difficulties such as marital discord, sibling rivalry, inattention to other children, and loss of relationships with extended family and friends often resulted. Marital difficulties have also been reported by other researchers (Thoma, Hockenberry-Eaton, & Kemp, 1993; Whyte, 1992). Some parents in Parker's study (1996) worried about the effect their child's illness would have on their marriage. They had heard about higher divorce rates in families with ill children. Martinson et al. (1994), however, suggested that while illness can push parents apart, it can also strengthen their love and commitment. The findings of this research on family structure and patterns of interaction provide important insights on the impact of a child's life-threatening illness on family experiences. However, little is known about how changes in family structure and interactions occur over time. Research evidence when a child has an NLTI is particularly limited. Similar findings have been 44 reported in studies of families with chronically ill children (Diehl et al., 1991). Parents of children with medically complex needs also reported that extended family members were often afraid of the child (Diehl et al.). This fear may help account for the decreased family contact reported by many families. In some families caring for chronically ill children, parents shared tasks according to their skill and comfort level. In a study of 30 families, it was reported that during difficult periods of caring for chronically ill children the majority of parents (63%) could focus better on the daily care required when they were able to rely on one another (Clements, Copeland, & Loftus, 1990). There were some families (n=9) where the mother assumed responsibility for care of the household as well as for the ill child. While acknowledging that the burden was heavy, one mother said, \"They (men) don't have the inner strength that women do....I think that a woman is a lot stronger than a man\" (p. 159). Another described her feelings of resignation about her husband's lack of participation in the child's physical care, \"He has never participated in the treatments or anything like that. I have asked him, but he said that he didn't know how to do it....knowing his character and everything, he probably wouldn't be able to do a good job of it anyway\" (p. 159). Single parents (n=2) found caring for the child to be an overwhelming burden. Families drew on many supports to help them. External supports included the health care team, grandparents, friends, informal and formal support groups, and the church. Inner strength came from religion and prayer. Forty percent of the parents reported receiving significant support from their ill child. Although limited, there is some evidence that families may be able to reduce the impact of their child's illness by reinforcing the family structure (Lesar & Maldonado, 1996; Patterson, Jernell, Leonard, & Titus, 1994; Ray & Ritchie, 1993). In a study of 48 parents of children with HIV infection (n=31) and uninfected perinatal HIV exposure (n=17), Lesar and Maldonado used standardized measures and semi-structured interviews to assess parental coping behaviours and 45 their mediating effect on the impact of HIV infection on the family system. Eleven participants were foster or adoptive parents, five were relatives, and the remainder were birthparents. These authors suggested that parents who emphasize doing things together as a family and who strengthen family relationships successfully cope with the impact of HIV infection by facilitating family cohesiveness. Caution should be used when interpreting these results, however, because most of the questionnaires were completed by only one parent, usually the mother. Patterson et al. (1994) arrived at a similar conclusion in their study of 48 medically fragile children and their families at home. They determined the behaviours of home care providers that contributed to positive and negative relationships with families. In addition, they identified the child, parent, and community factors that were associated with strain in the parent-professional relationship. Parents completed a number of quantitative instruments, as well as open-ended questions about their experiences and perceptions of caring for their medically fragile child. Family cohesion, family organization, and support from the community reduced the amount of strain parents reported having with home care providers. Whyte (1992) also reported that caring for a child with cystic fibrosis affected the interaction patterns and the coping responses of families. Where there was synchrony between the partners, family functioning remained strong. Whyte found that families needed to move through the transition of seeing themselves as a \"normal\" family to accepting that they were a family with a health problem. Otherwise, the relationship became very strained. The findings from research with families of chronically il l children provide important suggestions about the potential impact of a child's life-threatening illness on family structure and patterns of interaction. However, little is known about the transferability of this research to families where a child has an NLTI. Emotional Impact The emotional impact on families of caring for a child with a life-threatening illness, including cancer, cystic fibrosis, HIV/AIDS, and NLTIs, is beginning to be documented (Bluebond-Langner, 1996; Clarke-Steffen, 1997; Davies, 1996; Gravelle, 1997; Parker, 1996; Stein et al., 1989; Stein & Woolley, 1990; Wiener, Theut, Steinberg, Riekert, & Pizzo, 1994). Most parents are anxious and worried (Parker; Stein et al.; Whyte, 1992). Wiener et al. examined four dimensions of psychological functioning in 101 parents of HIV-infected children. Scores were generally high on both state and trait anxiety. In addition, almost 25% of the parents were moderately to severely depressed, and another 15% were mildly depressed. Other parents have been found to suffer anxiety, insomnia, and social dysfunction as they struggled to balance the demands of their ill child with the management of everyday living (Stein & Woolley). These parents started or increased their use of tranquillizers, anti-depressants, or cigarettes as they tried to relieve their anxiety. Parents also worried about their child's symptoms, the course of the illness, and about caring for the child if their own health failed (Stein & Woolley). Fear, anger, and feelings of sorrow and loss are also common (Gravelle; Parker; Whyte). In a quantitative study of fathers whose child was chronically ill (n=23), one of the most common stressors was worrying about the child's health (Cayse, 1994). Cayse also found that fathers wondered about what the child's future would likely be, and were concerned for their wives' health. Because some of the PLTIs are genetic, some parents may feel responsible for causing the child's illness. The guilt and anguish that may arise from this sense of responsibility has been noted in research with families where a child has an NLTI (Davies, 1996; Hunt & Burne, 1995; Parker, 1996) or cystic fibrosis (Bluebond-Langner, 1996; Whyte, 1992), and in HIV-infected parents who infected their child (Wiener et al., 1994). Many of the NLTIs are inherited through maternal genes, so mothers may be more likely than fathers to describe a sense of guilt. It may be important to identify which parent feels guilty so appropriate interventions can be implemented. 47 The tendency of researchers to report data as family-level findings when they have been obtained from a single member of the family may obscure this important element. For example, Davies mailed an investigator-developed questionnaire to 15 families of children with NLTIs. She asked families to describe the factors that influenced how they coped with their child; the strategies they used to cope and how these had changed over time; and what supportive resources they had been told about or used. Unfortunately, although Davies states that families completed the questionnaire, there is no indication of who actually filled out the questionnaire. It is unclear if the results reflect the perceptions of only one member of a family, either the mother or the father, or the family as a unit. One cannot determine exactly who reported feelings of guilt. There is some evidence that ill children themselves may also experience emotional and behavioural problems, such as anxiety and unhappiness, that, in turn, may further add to the emotional burden parents carry (Stein & Woolley, 1990). Some siblings may also experience anger, sadness, ambivalence towards the ill child, and reduced attention from parents (Ashby et al., 1991; B. Davies & Martinson, 1989; Sourkes, 1980; Stein et al., 1989; Whyte, 1992). In fact, in some families, researchers have reported that siblings may have more problems than the dying child (Spinetta, 1981; Stein & Woolley). Some siblings may worry excessively about their own health, and may be afraid of becoming ill like their brother or sister. Siblings may also have emotional and behavioural problems such as difficulty in dealing with their school work, difficulties in establishing and maintaining relationships with peers, and increased aggression or withdrawal (Gallo, Breitmayer, Knafl, & Zoeller, 1993; Stein et al.). Other authors have also reported nonspecific psychological problems in siblings (Burne et al., 1984; Stein & Woolley). However, mothers in one study recognized that siblings who were well adjusted were minimally affected by the illness (Gallo et al., 1993). Furthermore, several researchers have suggested that living with a chronically ill child may also promote emotional development in siblings. For example, an increase in personal maturity and moral values has been noted in some siblings 48 (Walker, 1993). The inconsistency of the literature related to siblings of children with chronic physical and cognitive disabilities appears not to have changed since Faux's (1993) review. Faux suggested that siblings may be at increased risk depending on the interactions of such variables as age, birth order, and gender. Physical Impact A few authors remarked, almost in passing, that parents of children with PLTIs were exhausted (Gravelle, 1997; Martinson, 1993; Stein & Woolley, 1990). In addition, exhaustion has been observed in mothers of chronically ill children (Stewart, Ritchie, McGrath, Thompson, & Bruce, 1994) and has been recognized as having a significant impact on their lives (Gravelle), although this has not been well described. Clearly, this is an area that requires further exploration. Only one study provides some insight into the possible physical impacts of caring for a child with an NLTI (Leonard, Johnson, & Brust, 1993). The health of mothers of children with disabilities was reported to decline. These parents often encountered such difficulty in managing that they recognized the need for more help. In particular, lifting and positioning the child was reported as physically exhausting for many mothers. More research is clearly needed to understand the physical impact of caring for children with NLTIs. Financial Impact Families face many financial costs that are invisible to others, and yet may be a drain on family resources. Sometimes, parents, especially fathers, have been reported to feel they have to work overtime to ease their financial burden (Clarke-Steffen, 1997). Often, particularly in a country such as Canada where government health plans ensure a basic level of coverage, people assume that all costs are covered by insurance. Most insurance plans, however, do not pay for incidental costs. The cumulative effect of these costs may take a toll on families. Some researchers in pediatric palliative care have identified high cumulative, monetary costs related to buying medications, special diets and equipment, or frequent attendance at health 49 care facilities with the associated costs of travel, food, and telephone calls to the family left at home (Birenbaum & Clarke-Steffen, 1992; Stein et al., 1989; Stein & Woolley, 1990). In addition, they have reported indirect costs, such as mothers reducing their work hours or giving up their jobs to take care of the child, that result in a loss of family income at the same time that financial costs are increasing (Parker, 1996). Some fathers refused promotions or transfers, while others became unemployed or bankrupt (Birenbaum & Clarke-Steffen; Stein et al.; Stein & Woolley). In addition, those parents with insurance may be unable to change jobs because new insurance companies may not cover them. Even when families have insurance coverage, either by private coverage as some do in the USA or through government health plans as in Canada, families may still incur great financial costs when a child is ill. Schweitzer, Mitchell, Landsverk, and Laparan (1993) noted that, though families usually want to keep their child at home, the cost of nonreimbursed expenses may be enough to exclude many families from participation in a home-based program. These researchers used two sources of data to estimate direct costs (such as wages of health care workers, medication, and special clothing and supplies that would not have been purchased otherwise) and indirect costs (including lost wages where a caregiver lost a job, gave up leisure time, or took vacation, and volunteer care by friends, neighbours, and other volunteers). Retrospective data on provider utilization and duration in an American home-based pediatric hospice were collected for 177 families. Twenty-seven of these families provided prospective data through telephone interviews and expenditure diaries. Weekly incidental expenses averaged $42, and average indirect expenses amounted to $140 per week. The total average cost profile for home care episodes was $4,808. Direct personnel costs averaged $2,850, direct incidental costs were $446, and indirect costs amounted to $1,478. For many families of chronically ill children, a high percentage of weekly income may be spent on out-of-pocket expenses. Studies have consistently shown that these expenses may be a 50 significant drain on family resources (Diehl et al., 1991; Ray & Ritchie, 1993; Walker, Epstein, Taylor, Crocker, & Turtle, 1989; Youngblut, Brennan, & Swegart, 1994). Out-of-pocket expenses reported in one study of 910 parents included travel expenses, parking fees, and child care for other children, as well as other expenses (Walker et al., 1989). Diehl et al. (1991) also reported a lack of financial assistance to middle income people and difficulty in finding out what assistance might be available. Spiritual Impact The impact of caring for a chronically ill child on experiences related to spirituality and faith have been described by a few researchers. There is evidence that many parents seek out spiritual guidance and support (Davies, 1996). Spirituality and faith have provided both emotional and network support for many parents. Faith has been shown to be a key factor in a family's ability to keep an ill child at home (Davies, 1996) and in promoting positive adaptation among families of children with disabilities (Bennett, DeLuca, & Allen, 1996; Youngblut et al., 1994). Bennett focused on positive adaptation and investigated the informal resources that parents use across the family life cycle. Interviews with 12 parents of children with disabilities ranging in age from 15 months to 30 years revealed that religious beliefs were particularly supportive. Faith in a higher power has been associated with higher levels of coping and hope in family caregivers of ill adults (Herth, 1989; Herth, 1993; Kaye & Robinson, 1994). It has also been reported as an effective coping strategy in a number of different illness situations (Cayse, 1994; Lesar & Maldonado, 1996; Schwab, 1990; Steele & Fitch, 1996). Some families of chronically ill children report receiving inner strength from religion and prayer (Clements et al., 1990; Rivera, Stoner, & Groothuis, 1996). A strong belief in God may provide caregivers with the strength and support they need to continue caregiving. 51 Summary It is not unrealistic to expect that families may undergo many changes when a child has a terminal illness. Normal patterns of interaction may be disrupted, and families may need to learn to adapt to new challenges and changes. There is evidence to suggest that family structure may be at risk of deteriorating as energy, time, and resources are directed towards the ill child. Living with a child who has a PLTI may also cause emotional, physical, financial, and spiritual disruptions. However, it appears that some families may be able to cope with changes better than other families. For example, families where parents share tasks and rely on one another may have less difficulty in dealing with the changes and with the care of the ill child. Clearly, a better understanding of the impact of caring for a child with an NLTI is needed. The situation of caring for such a child presents some unique challenges that we are only beginning to understand. We need to know more about these impacts and challenges to learn more effective ways of supporting these families. Factors Influencing Families' Experiences Families' experiences do not occur in isolation. Instead, they are influenced by a number of factors that may be internal or external to the family. Some factors that are thought to influence the experience include the location of care, communication among family members and between families and professionals, and level of family functioning. These factors will be examined to provide a context for understanding families' experiences. Location of Care The physical environment in which care occurs may affect the family. While there are no pediatric studies that have asked dying children their preferences about location of care, it is believed that most children would prefer to be at home (Kopecky et al., 1997). The importance to families of the location of care is demonstrated in studies involving three different programs 52 providing home-based palliative care for children with terminal illnesses (Duffy et al, 1990; Lauer & Camitta, 1980; Martinson, 1993; Martinson et al., 1986). Duffy et al. (1990) surveyed eight parents of children with central nervous system tumours who had died while receiving home-based palliative care. Their overall satisfaction with the program was high compared to the care they had received prior to admission to the program. The component of the program that was most satisfactory was being able to care for the child at home with the support of professionals. In another study (Lauer & Camitta, 1980), families of children with cancer who received home care (n=42) reported that they wanted the child at home. In addition, the medical and nonmedical financial burdens of inpatient and outpatient hospital care were reduced when the child died at home. The Home Care for the Child with Cancer Project is a comprehensive study that was conducted from 1976 to 1980 in three large American institutions (Martinson, 1993; Martinson et al., 1986). The purpose of this project was to evaluate the feasibility of home hospice care for families. Primary care was provided directly by families with strong nursing support. Nurses made home visits and were available on call 24-hours-a-day, 7-days-a-week. Fifty-eight of the 64 children referred to the project over the reported 2-year study period died; 46 of them at home, 11 in hospital, and 1 en route to hospital. The children had a wide range of cancer diagnoses. Their parents participated in semi-structured interviews one month post-death to review the process of home care, parental assessment of care, and financial aspects of home care. Martinson (1993) reported that parents whose child died at home had fewer negative and more positive comments about their child's home care, compared with those whose child had died in hospital. Parents noted that fulfilling their child's wish to get home had positive effects. They perceived the child to be happier, more secure, and relaxed once home and no longer separated from family and friends. Children were also able to live more normally and have more control over their lives. The illness was very disruptive for parents, but they believed that home care 53 reduced this disruption and allowed them to have more control. Parents also enjoyed being able to care for their own child. They valued being constantly nearby. Negative aspects of home care included stress, exhaustion, difficulty in keeping well-meaning visitors away, and the need for more follow-up and management of medication. Invasion of family privacy has been identified as an important issue for some families caring for a chi ld at home (Leonard, Brust, & Patterson, 1991; Patterson et al., 1994). Families are often dependent on home care professionals. Yet, the regular presence of others in the home may increase the strain on families. Two major themes that emerged from analysis of open-ended questions in the Patterson et al. study were related to home care staffs competence, and attitudes of nurses towards the child, parents, and the rest of the family. The greatest difficulties in parent-professional relationships were experienced by mothers when they received more help from home health aides, as opposed to professional nurses or home helpers. One of the challenges noted by this study was the need for professionals to balance supporting the family in a non-intrusive manner while providing sufficient support. Findings from these studies must be interpreted cautiously because the studies were retrospective, and from the perspective of parents, usually the mothers. Consequently, the results may be biased by the recollections and experiences of only one member of the family. They may not reflect the family's experience. Despite the limitations of the previous research studies and the potential drawbacks of home care, results suggest that many families want to care for their i l l child at home wherever possible. Finding ways to support families requires an in-depth understanding of families' experiences and how they manage at home. Yet, there is a paucity of research that examines families' experiences at home when one of their children has a PLTI . Although it appears that some families want to have their child at home, we only have a cursory understanding of the impact on family life of caring for a child with a PLTI . 54 Communication Although open and effective communication may be important for families as they care for their ill child (Clarke-Steffen, 1997), problems related to communication have been identified among family members (Clarke-Steffen; Whyte, Baggaley, & Rutter, 1997), and between families and health care professionals (James & Johnson, 1997). Issues related to communication have been addressed by researchers in the context of families' experiences. Sharing information with the ill child and siblings can be especially difficult for some parents (Clarke-Steffen). Parents often struggle to find a balance between protecting children by withholding information and providing sufficient information to maintain a trusting relationship. Sometimes, parents may have discrepant views of the situation. Even when they try to address their differences, these issues are often a source of ongoing conflict. A qualitative study of 16 families where a child had congenital heart disease, asthma, diabetes mellitus, or cystic fibrosis revealed similar findings (Whyte et al.). The investigators reported that most families had difficulty maintaining good communication, even when parents had strong communication skills prior to the child's illness. This block in communication between parents contributed to strain in marriages and affected parents' ability to continue caring for their child. Other researchers have also noted that the spousal relationship in similar contexts has an important influence on parents' mental health (Deatrick & Knafl, 1990; Nagy & Ungerer, 1990; Ray & Ritchie, 1993). In some cases, communication about the child's condition may be unclear between health care professionals, such as when a child's status changes from active treatment to palliative care (B. Davies & Steele, 1996). Researchers have noted that this lack of communication may create gaps in the support services provided to patients and families (Davies, 1996; Singleton, 1992). Moreover, the bureaucracy involved in obtaining services can be very frustrating for parents (Gravelle, 1997). Faulty communication may also cause anxiety in parents that impedes satisfactory care. In her study of families where a child had an NLTI, Davies reported that 55 families were often followed by multiple services within the hospital. Having one person, who knew them and their child, co-ordinate their care, facilitated their connection to the hospital and made it easier for families. Until that link was in place, they felt isolated, alone, and unsure of their ability to care for their child at home. Other researchers have also identified the need for a competent case manager to ensure adequate communication (Diehl et al., 1991; Gravelle; Whyte, 1992; Woolley, Stein, Forrest, & Baum, 1991). In the U.K, the term \"cornerstone carer\" was used by Woolley et al. during their research at Helen House (Stein & Woolley, 1990). These researchers defined a \"cornerstone carer\" as the person who provided central support or who enlisted or co-ordinated appropriate help. Eleven out of 45 families in this study did not have a \"cornerstone carer.\" Of those 11 families, 9 had children with NLTIs. Families where a child has an NLTI may, therefore, be particularly vulnerable to communication difficulties. Families appear to have a variety of experiences in communicating with health care professionals. In a retrospective, qualitative study of 12 parents of 8 children who had died from cancer, parents reported feeling cared for when professionals spoke with them and treated them with respect, compassion, and sensitivity (James & Johnson, 1997). Conversely, a professional's failure to listen to parental concerns caused frustration for parents. Moreover, parents felt abandoned by caregivers who physically and emotionally withdrew from them. Parents in Parker's study (1996) also reported feelings of frustration, anger, and hopelessness when health care professionals refused to take parents' concerns seriously. Researchers have determined that it may not be easy for families to communicate with health professionals (Clarke-Steffen, 1997; Diehl et al., 1991). For example, Diehl et al. used focus group interviews to identify parents' perceptions of their needs when a child has medically complex needs. These parents identified the health care system as a major source of distress and frustration because of its fragmentation of care. In addition, parents reported a feeling of \"them against us.\" They felt that professionals, especially doctors, did not listen to parents or give them 56 the respect they deserve. Parents thought they needed to be assertive and informed because professionals seemed uncomfortable with the child, did not spend enough time with the parents, and blamed the parents for the child's behaviour even though the child might be sick. Professionals who did a good job were the exception. Parents' need for honest and appropriate information so they can manage their situation has been well documented in studies, and includes those families who have children with cancer (Clarke-Steffen, 1997; James & Johnson, 1997), cystic fibrosis (Bluebond-Langner, 1996), NLTIs (Davies, 1996), and other chronic or life-threatening illnesses (Diehl et al., 1991; Gravelle, 1997; Walker et al., 1989). Information may be empowering (Clarke-Steffen; Gibson, 1995; Gravelle), and allow families some control in a situation where they have little control. It may also help to reduce uncertainty (Cohen, 1995; Galloway & Graydon, 1996; Selder, 1989). Yet, some researchers have shown that information needs are not always met (Clarke-Steffen; Walker et al., 1989). In a mailed survey with a convenience sample of parents whose children had various chronic illnesses (n=910, 38% return), communication and information needs were ranked as important, though frequently not met (Walker et al.). Many parents wanted more communication with health professionals. They needed to obtain information about programs and services, financial planning for the future, the child's particular health problem, financial aid and insurance plans, and the treatment prescribed. Barriers faced by families when looking for information may include complex medical language, reluctance of some professionals to disclose information, and negative attitudes of those providing the information. Moreover, in situations where there is little information about a disease entity, as is the case with NLTIs, the task of providing the information parents need becomes more difficult. Thus, even when professionals are willing to share what they know, there may be no information available. In such cases, the scarcity of information has been seen to contribute to increased uncertainty and frustration in these families (Davies; Parker, 1996). 57 Problems in communication have also been attributed to discrepancies that arise when parents need to talk about their feelings related to the child's death and dying and family members and friends are unwilling to discuss these issues (Diehl et al., 1991). Our understanding of the complexities underlying communication problems for families has been extended by Jefidoff and Gasner (1993) in their study of Israeli parents of a dying child. They found that even when families can discuss their fears and talk about the child's imminent death, they may be unable to articulate the more unconscious and complex needs, such as those related to understanding changes in family dynamics that may occur in the final stages of the disease or describing feelings about the funeral and burial, unless they have the assistance of skilled hospice staff. The investigators suggest that families' ability to recognize and deal with these potential stressors puts them at higher risk for further problems. The findings from this study (Jefidoff & Gasner) must be interpreted cautiously however, because the researchers did not say how many families or parents were included in the study. Moreover, the results from this Israeli study may not be generalizable to other cultures. In summary, issues related to communication appear to play a significant role in shaping families' experiences when they are caring for an ill child. Our understanding, however, remains limited because of the paucity of research that has addressed these issues. Further, the complexity of the dynamics involved in communication needs to be explored in order to learn how to provide effective care to families of children with NLTIs. Levels of Family Functioning The concept of family functioning has received considerable attention from family theorists, family therapists, and others who focus on family care (Friedman, 1998; McCubbin & McCubbin, 1989; McCubbin & McCubbin, 1993; Olson, Russell, & Sprenkle, 1983), but only minimal attention in palliative care. It is not improbable that a family's ability to adapt to the changes imposed by the child's illness and impending death may depend in a large part on their level of functioning. One group of researchers (B. Davies, Spinetta, Martinson, McClowry, & Kulencamp, 1986), however, have made an important contribution to our understanding of family functioning in the context of pediatric palliative care. In their study of 111 families whose child had died from cancer, the investigators conducted semi-structured interviews and rated family functioning using a five-point scale. The global rating of each family's overall level of functioning was derived from the researchers' judgements based on clinical assessment of the family during the interview, the family's responses to the interview questions, and how the family compared to other families. The dimensions of functioning were viewed as occurring along a continuum of functionality and were not judged as good or bad. Families at the extreme ends of functionality were compared and contrasted. The more functional families freely discussed the dead child, the illness and death context, and the family's responses since the death. They shared views with other members of the family and were able to disagree without conflict. More functional families recognized that others were also grieving, and that differences in grieving were alright. They were empathetic and supportive. On the other hand, some families did not have freedom of expression and one member frequently spoke on behalf of another. Members of these less functional families focused on their individual grief and did not empathize with others, or even recognize others' grieving or differences in grieving. Functional families were aware of a process occurring. They could describe their feelings about events, such as the funeral, rather than only being able to concentrate on the content, such as the location of and attendance at the funeral. They anticipated special events, such as Christmas, as being difficult and tried to prepare constructively. They could also see how they had changed. Less functional families, on the other hand, made no plans to cope with difficult times, did not share their feelings, and did not seem to change. They seemed resigned to everything as if they could not change it (B. Davies et al., 1986). 59 Functional families also acknowledged their grief and questioned how they could cope with the sadness of their child's death, while seeing that something positive had come out of the experience. Returning to a normal life which included frivolity was important to functioning families, and they integrated their experience into their life. They wondered if they could or should have done something different to help their child. But, they were not regretful. They saw these feelings as part of the mourning process. They continually changed their focus of hope as the child's condition deteriorated. Thus, although they still hoped for a miracle, they were realistic about the prognosis. These families remembered the good and the bad things about their child. In contrast, other less functional families suppressed their natural feelings and often made comments related to religion that reflected a belief that they could not question events because it would show a lack of faith. Some had difficulty in accepting that their child was dying. They often fought the diagnosis till the end without acknowledging the inevitability of death. They often idealized their dead child and seemed to forget that he or she had once been a normal child who misbehaved on occasion. Less functional families regretted what they had done or not done that may have contributed to their child's death. Their lives were based on their experience of having a child die. Everything else was seen as frivolous compared to that experience. Functional families were flexible in their roles (B. Davies et al., 1986). They acknowledged that one child was no longer there and did not expect another child to fill the gap. Less functional families were often rigid about roles and expected another child to take over the roles of the dead child. Some parents imposed the roles, some family members self-imposed them, and some parents expected surviving children to assume the characteristics of the dead child. Gender-related role rigidity was more obvious in less functional families in which the father would not express grief over the dead child. More functional families were able to ask for and use help from a variety of sources. They could also turn down unwanted help graciously. Other families were uncomfortable or lacked confidence to ask for help and felt very alone. Functional families recognized their loss and the feelings of sadness, anger, and depression that accompanied it. But, they were able to reorganize their life and move on. They knew that they were not and could not be the same as.they were before their child died. Others did not want to reorganize and tried to keep everything as it had been. They remained sad, depressed, or angry and could not move on. Functional families accepted that they were vulnerable and admitted the pain they suffered. They always tried, though, to see the good things in life as well. Lower functioning families defended their feelings of vulnerability. They could not give themselves permission to laugh because they believed that the enormity of their loss prevented them from ever having fun again. Functional families also reassessed their lives in view of their experience and questioned their previously held beliefs. Quality of life and person-related values often became more important to these families. Although they accepted their child's death as fate, they also vented their anger at God. They were able to continue to believe in God even though they did not like what had happened to them. Other families accepted the hand of fate without question and saw God as all good or all bad. They either never questioned why or they left their religion. Expression of sadness and tears was accepted by functional families as part of the grief process. Conversely, dysfunctional families often equated sadness with craziness. Feelings might come out too strongly at other times of stress though, such as when a family pet died. Functional families had low control over their grief. They discussed their feelings and shared their sadness. They were sensitive to others. Dysfunctional families had a high control of their grief. Yet,' nonverbal behaviour was often an indicator of underlying emotions. The more functional families were open, flexible, and adaptive. They could drop old friends who did not share their new perspectives and they made new friends who did. They could move on with their lives. They often became reinvolved with children with cancer to help others benefit from their experience. Less functional families were bitter about old friends who could not cope with the child's death. 61 They did not want to make new friends. If they became involved in healthcare issues, it was only to help themselves work through their difficulties (B. Davies et al., 1986). In another study (B. Davies et al., 1995), family level of functioning was also related to how families managed the transition of fading away when one of their adult members was dying with advanced cancer . In this three phase study, the researchers used grounded theory methodology to understand how families experience their lives when one of their members is dying. Twenty-three families in total were interviewed. Participation came from 66 family members, including the terminally ill adults. Again, dimensions of functionality occurred along a continuum. Some families used past experiences as opportunities for growth and learning, while others dwelled on the negative aspects (B. Davies et al., 1995). In some families, members described a broad range of feelings and acknowledged vulnerabilities and paradoxical feelings. Others avoided talking about the impending death and were often angry. Some families identified and discussed problems through mutual support and togetherness. They resolved the problem by taking control of the situation. These families learned from the problem and its solution. Other families concentrated on the emotional issues without trying to find a solution. They were often bitter and blamed others for the situation. As a group, they showed consensus. But in individual interviews, they often blamed other members of the family for the problem. Families who used multiple resources and support were usually satisfied with the result of the help. Some other families were reluctant to seek or accept help and were often dissatisfied with the help they received. Some families recognized each member for their valuable contributions to the patient's well-being and concentrated on keeping the patient well cared for without seeking attention for themselves. In contrast, in other families, each individual seemed to focus on their own needs and not recognize the contributions or needs of others. Flexibility in allocating responsibilities 62 usually resulted in families who were happy with their decisions. Conversely, families who tried to hold on to old roles could not adapt easily. Members sometimes felt obligated to provide care, rather than wanting to do it. Some families tolerated differences inside and outside the family, even though they did not necessarily agree with a particular view. Where families were in agreement about the family identity, they could provide a supportive environment for their members. Some families did not have a family identity. Instead, they only had personal characteristics. The \"truth\" was often different in the group compared with individual interviews. There was often a history of abuse in these families. These families often had difficulty in tolerating differences outside the family as well (B. Davies et al., 1995). Clearly, there are differences in how families deal with situations. How these differences affect provision of care is not as clear. Negotiation and mediation may be critical when working with families who are not cohesive. Further research is necessary to understand how the dimensions of functioning affect how and when services, options, and resources should be offered. This understanding is necessary to develop the range of interventions that may be required by families with varying degrees of functionality. Conclusion There is a dearth of research and knowledge to guide practitioners in providing optimal care for families where a child has a PLTI, particularly a neurodegenerative disease. While the available research provides an important beginning, the findings must be viewed as tentative because of small sample sizes and the scarcity of studies. In addition, the retrospective nature of some of the studies may have resulted in important information being forgotten or under-reported. Knowledge about families where a child has a PLTI is steadily increasing, but very little research has focused on families of children with NLTIs. While researchers investigating children with chronic illnesses provide some support for these findings, one must be careful of 63 drawing firm conclusions. Any differences between families of children with NLTIs and those with chronic diseases may not be apparent in the limited research that is available. The research to date provides a cursory understanding of the impacts of families' experiences when a child has a NLTI, and the factors that influence those experiences. Research indicates that all members of the family may be adversely affected by a child's life-threatening illness. Most of the available research, however, has been from the perspective of the parents, and sometimes only the mother. There have been few attempts to capture the views of the ill child and his or her siblings. Yet, relying on adults to identify impacts on children, whether ill children or siblings, may neglect the children's fears and anxieties and may fail to capture their experiences. It is critical that all members of a family, including the ill child, are asked about their experiences. A number of the studies with families of children with a PLTI used quantitative methods to obtain data (Lesar & Maldonado, 1996; Martinson et al., 1986; Nagy & Ungerer, 1990; Stein et al., 1989; Stein & Woolley, 1990; Thoma et al., 1993; Wiener et al., 1994). Some of these instruments were developed for the studies and underwent little psychometric testing, so their reliability and validity are unknown. Others were standardized psychosocial instruments that have limitations when completed by such families. Many psychosocial scales were developed and normed on college students or psychiatric patients. They may, therefore, have little relevance to families where a child has a PLTI. In addition, most instruments designed for healthy families or families of chronically ill children have not been subjected to validity and reliability testing in families where a child is dying. Thus, limited confidence can be placed in the results obtained from such measures. Finally, pre-existing instruments can only capture discrete parts of an experience. Families would need to complete many more instruments than were used in these studies in order to fully describe their experience. 64 The focus on rich, detailed descriptions in some of the qualitative studies has uncovered important aspects of families' experiences when a child has a life-threatening illness. Unfortunately, the studies of families where a child has an NLTI are very limited in their descriptions. There is a need for more in-depth studies to uncover the experiences of families as they live with a child who has an NLTI. Moreover, it is important that all members of a family have the opportunity to contribute their perspective. Retrospective studies that are done after a child has died may fail to reveal the depth and breadth of families' experiences. Instead, research is needed during the experience itself. Qualitative research methods may be the most effective way of capturing the full richness of families' experiences when a child has an NLTI. Research Questions Specific research questions that guided this study were: 1) What is the family's perception and experience of living with a child who has a neurodegenerative, life-threatening illness? 2) How does the family's perception and experience change over time? 3) What is the impact on the family of living with a child who has a neurodegenerative, life-threatening illness? 4) What factors do families perceive enhance or undermine the family's ability to care for the child with a neurodegenerative, life-threatening illness? Definition of Terms For the purposes of this study, some terms were defined as follows: Diagnosis Period: The beginning of living with a child's neurodegenerative, life-threatening illness; when a family first learns about the child's illness. This period of time lasted from initial diagnosis until one month after diagnosis. Family: A complex unit that is composed of individual members, but that has distinct attributes of its own. Ideally, the family is self-defined and includes whoever the family says it includes. 65 The family unit in this study was limited to those members with whom health care professionals most often interact and who usually play a key role in pediatric palliative care - the ill child, parents, and siblings. Ill Child: A child, from infancy to 17 years of age, who had been diagnosed with a neurodegenerative, life-threatening illness. Neurodegenerative Life-Threatening Illness: Any disease that led to progressive destruction of the nervous system and was expected to cause premature death in childhood or early adulthood. Pediatric Palliative Care: Active care to children and their families when cure is no longer possible. Alleviation of symptoms, psychological, spiritual, and social problems is paramount. The goal of pediatric palliative care is to assist children and their families to achieve the greatest possible quality of life. Pediatric palliative care encompasses respite, terminal, and bereavement care. Respite Care: Support provided to families of a child with a neurodegenerative, life-threatening illness that relieved them of the continuous 24-hour care of the child for a period of time. Respite care is provided in and out of the child's home. Respite Period: The time from one month after diagnosis until one month prior to the expected death of the child during which the family lives with and manages the child's illness. Terminal Period: The end of living with a child's neurodegenerative, life-threatening illness; when the child was expected to die within one month. 66 Chapter IV: Research Design and Implementation Theory is critical to the development of any practice discipline such as nursing. The generation of knowledge for practice provides evidence on which to base interventions. The focus of this study was on generating a detailed, contextually-grounded description and theoretical explanation of the experiences of families with a child who has a neurodegenerative, life-threatening illness (NLTI). The choice of research design was made on the premise that the research design in any study must be suitable for the state of existing knowledge about the questions being asked (Siegel, 1983). Little is known about family experiences when a child has an NLTI. In addition, families themselves are the most pertinent source of information about their experiences. Qualitative methods are most appropriate when asking questions dealing with subjective experience and perceptions with situational meaning. Grounded theory is a qualitative research method that is particularly suited to family research because of its focus on social processes (Daly, 1992). Therefore, grounded theory had the potential to develop theory about family experiences. Consequently, the primary analytical method used in this study followed the procedures of grounded theory (Glaser & Strauss, 1967; Strauss & Corbin, 1990). Symbolic Interactionism The design of a grounded theory study is guided by the underlying symbolic interactionist theory in which shared meaning creates human behaviour. Symbolic interactionism is a theoretical perspective that is often used to guide research and practice in the family field (Klein & White, 1996). Symbolic interactionism characterized the theoretical perspectives of scholars such as pragmatists including Mead, James, Dewey, and Pierce, and sociologists including Thomas, Park, Cooley, and Wirth (Lindesmith, Strauss, & Denzin, 1975). In the early 1930s, Mead (1934) delineated an ontological position that was later articulated as symbolic interactionism by Blumer (1969). The symbolic interaction school of thought relies heavily on the pragmatic philosophical tradition which has its roots in the belief that a phenomenon only has 67 meaning if it can be directly or indirectly applied to a specific situation. In addition, it argues that an individual is in constant interaction with the environment and chooses which stimuli to respond to (Burr, Leigh, Day, & Constantine, 1979). Interactionists believe that understanding of human behaviour requires learning about the beliefs and values that individuals get from interacting with others. The best way to understand the behaviour of humans is to understand the meanings and values that occur in people's minds and that cause behaviour. However, these mental variables are affected by antecedent variables, such as feedback from intimate interactions; perceptions of generalized conditions in the social situation; and contextual variables like social status or social norms. In addition, the physical self is different from the social self. The \"I\" part refers to the unpredictable and spontaneous parts that make a person unique, whereas the \"Me\" part is determined by learned social roles. Because the \"I\" part of the self will influence what goes on in the mind, there is always an unpredictable component to the mentalistic variables (Burr et al). Unlike the approach of many social scientists, symbolic interactionists do not assume that culture is composed of commonly shared signs and symbols that are uniformly understood. Symbolic interactionists believe that one cannot understand a person's behaviour unless one understands the meaning of an event from the individual's perspective. Thus, cultural meanings and social behaviour are inexorably entwined. How a situation is interpreted is based on the meaning the situation holds for the individual. Meaning arises from a number of variables including past experiences, language, educational level, ability, and cultural background (Klein & White, 1996). Symbols are used to constitute meaning, and meaning is negotiated through symbols. Humans decide how to behave on the basis of learned symbols and their beliefs about the importance of meanings. Therefore, individuals act according to shared meaning, and behaviour is influenced by the meaning of the ideas in the mind. 68 Guided by symbolic interactionism, a researcher needs to explore the meaning given to situations by the people involved. One cannot assume that meaning is shared by all of the participants. A researcher must try to understand the world from the participant's perspective. Thus, objective measures are not sufficient to obtain information. Instead, formal and informal interviews with participants are required. In addition, one cannot observe the part of the self that is \"I,\" only the part that is socially defined as \"Me.\" Consequently, one must again obtain subjective information by discovering the person's perspective. Informal and formal interviews were, thus, an integral part of understanding and studying families in this current research study. Since meaning is constructed through interaction and meaning creates behaviour, then behaviour must be studied at the interactional and at the symbolic level. In order to understand human behaviour, a researcher must examine interactions in context. Therefore, behaviour needs to be observed in natural settings. Participant observation can be used to look for interaction, patterns of interaction, and their consequences. In addition, analysis of the interactions must include the symbolic level, so informal questioning of participants must be done immediately after an interaction to clarify the participants' self-definitions and shared meanings. A researcher cannot assume the meaning of an action or interaction, but must check out her assumptions and seek information directly from the participants. In studying families in which a child has an NLTI, participant observation was important to provide data about the process of living with such a child, and about how families managed caring for their ill child. Setting Data collection took place in the homes of families whose child had an NLTI. The purpose of this research was to understand how families managed their lives with such a child, as well as how their lives were affected by the illness. The researcher offered to accompany families during their daily lives in order to gain this perspective. Families consistently explained 69 that their lives revolved around the ill child in the home. Interviews and supplementary observations, therefore, all took place where the families suggested - in their homes. Selection of Participants Families were recruited for this study through two sites: a pediatric hospice care program and a children's hospital. Both sites were located in the Lower Mainland of British Columbia. Families were eligible for the study during the period of time in the child's illness when they may require respite,care. For the purposes of this study, this period was defined as lasting from one month after diagnosis until one month prior to the expected death of the child. Where a child had just recently been diagnosed, families were excluded because it was anticipated that they would lack the depth of experiences that would make them key informants for this study. Those families in which the child had reached the terminal period of the NLTI were also excluded because of the potential stress of this period on families. In addition, it was anticipated that those families may have different needs and experiences from families requiring respite care. For the purposes of this study, family was limited to those members with whom health care professionals most often interact and who usually play a key role in pediatric palliative care - the ill child, parents, and siblings. Each member of the family who could communicate verbally and or in writing was asked to participate. The ill child was part of the study in whatever capacity was possible, such as being observed during care. Al l of the ill children were precluded from full participation because of profound mental impairment or an inability to communicate. Families were not excluded from the study because the ill child was unable to participate, nor because one parent or siblings were non-existent, unavailable, or unwilling to participate. This research included those children, parents, and siblings who were willing and able to participate. Initial contact with families was made by the hospice and hospital staff. Families were considered potential participants if they met the following eligibility criteria: 70 1) The ill child had been diagnosed with a neurodegenerative, life-threatening illness. Where an illness is not life-threatening, families may face different issues. Neurodegenerative illnesses may create unique circumstances for families. 2) The ill child was 17 years of age or younger. The context of this research was pediatric palliative care. 3) Families were living in the period of time in the child's illness when they may require respite care, as defined in this study. At least one month had elapsed since diagnosis and death was not expected by clinical personnel for at least one month following participation. The specific context of this research was the respite component of pediatric palliative care. 4) At least one adult, probably a parent, who lived with the child and was involved in the child's care was willing to participate in the study. Single parent families were not excluded. 5) Those adult family members who wished to participate were able to read, write, speak, and understand English. Informed consent and participation in interviews required comprehension of written and verbal English. 6) Those siblings who wished to participate were able to communicate verbally and or in writing in English. From about age seven years, minor children needed to give their assent in addition to parental consent. Adult siblings were eligible for participation. 7) The family was emotionally and physically capable of participating in the study. The researcher and the clinical personnel were to make this judgement. Those families who were considered at risk were not approached. As data analysis and collection proceeded, theoretical or purposive sampling was used to ensure that the sample was appropriate. Families were deliberately selected for their potential contributions to the developing theory, and according to the theoretical needs and direction of the study. This approach ensured that the evolving theory was representative of the phenomena being investigated, and allowed for examination of the full range of variation in emerging 71 concepts in order to develop theory. To extend the conditions under which actions and interactions occurred, interviews occurred with families who required varying amounts of respite care. In addition, some children were deteriorating slowly while deterioration was occurring rapidly in others. Data collection continued until the data were saturated, that is until no new data were obtained from interviews and the phenomena were richly described. The Sample Twenty-nine participants were interviewed or observed in this study. Of the eight families who participated, two families had two children who were affected by an NLTI. Therefore, 10 sick children in total were observed. Six of the children were boys and the ages ranged from 3-13 years. At the time of the study, the children had been diagnosed for two and a half to six years. None of these children could be interviewed because they were unable to communicate verbally due to the nature of their illnesses. While all the children had neurodegenerative illnesses that would eventually lead to their death, the actual diagnoses will not be named to facilitate confidentiality of families. Most of these illnesses affect very few children throughout the world, perhaps less than 100 for each diagnosis. Indeed, some of these children represent the only known case of a particular illness in British Columbia. Identifying their diagnoses would effectively identify the children and families. While the different illnesses are manifested in many ways, common attributes such as changes in verbal ability, changes in motor skills, and the proliferation of feeding disorders meant that these families experienced many of the same opportunities and challenges as they lived with the illness. There were only three families in which the ill child had siblings. Usually, parents chose to have no more children once they learned about their child's diagnosis because these NLTIs are often genetic. Out of four siblings, three, all female, had been born before the child was diagnosed. The other sibling was male, and the siblings' ages ranged from 2-9 years. 72 In the majority of families, the parents were married and lived together, although there were difficulties in most marriages. One couple had been married but was now separated, another couple was also separated but had never been married, and in the eighth family, a single father cared for his child alone. The mother was not involved in that family's life at all. The length of the marriages ranged from 8-15 years. Parents' ages ranged from 28-48 years and most had completed high school, although educational levels varied from grade six to a university degree. Occupations of the parents included those who were currently unemployed, often because they were caring for the ill child, blue and white collar workers, and professionals. Socio-economic diversity was apparent with annual family incomes ranging from $11,000 to $112,000. Four families subsisted on incomes of less than $15,000 per year, while the other families earned $50,000 or more. However, all families reported a substantial drop in actual or anticipated income due to the child's illness. Geographic diversity was restricted to the Lower Mainland of British Columbia, despite efforts to contact families in other parts of the province. However, location of the tertiary children's hospital is important for families, so families often choose to live close to the hospital. While urban and suburban settings were represented, it was somewhat difficult to distinguish what difference geographical location might make. Some families in the suburbs felt they were very close to the city, yet others who lived in the same area stated that the trip to the city was a major event because it was so far away. Of the eight families who participated in this study, all but one family identified with the dominant Caucasian Canadian culture. The other family came from an East Indian background. Data Collection and Procedures In accordance with grounded theory methodology, the main method of data collection was in-depth interviews with families supplemented by participant observation. Data collection was facilitated through the pediatric hospice care program and the children's hospital. Initial data 73 collection occurred over a period of one year. During the following year, the evolving analysis was shared with the families. After the first interviews and observations were completed with all eight families, each family was sent the preliminary analysis by mail. A letter was enclosed inviting families to contact the researcher if they had any comments. The letter also stated that the researcher would call each family (see Appendix 6). One family called immediately on receipt of the package. A second interview took place with this family in the family home. Al l other families were called by the researcher. Subsequent interviews took place with these families by telephone. Two families requested that they receive major iterations of the evolving theory so they could make comments. The data were sent by mail and these families were interviewed by telephone on a few occasions. The other families simply requested a copy of the completed research report. Initial contact with potential participants was made by clinical staff, all of whom were nurses. The researcher met with staff to explain the study, to outline the eligibility criteria, and to answer any questions staff had. Ongoing communication with staff was critical in ensuring accrual of participants into the study. Other researchers (B. Davies et al., 1995; McCorkle, Packard, & Landenburger, 1985) have suggested that physicians, someone designated by the physician such as a nurse, or a family member may act as a barrier to recruitment of potential participants. Further, B. Davies et al. suggested that nurses in particular may be reluctant to tell families about a study because they often assume the role of patient and family advocate, and may try to protect the families from a seemingly stressful imposition. Sometimes it may be unclear whether a person is really unsuitable for a study or if the advocate's attitudes and prejudices have influenced the decision (Ling & Perm, 1995). Therefore, to facilitate accrual, the researcher attempted to have frequent face-to-face communication with the professionals involved in recruitment and to maintain a high profile. However, many of the contacts actually occurred by telephone and electronic communications. Despite the researcher's best intentions 74 and attempts to encourage staff participation in research by discussing why and how research is done, by pointing out the benefits of research, and by highlighting the ways that patients' and families' rights are safeguarded (Raudonis & Kirschling, 1992), some difficulties in recruitment did occur at the pediatric hospice program. It appeared that some families were not offered the opportunity to take part in the study because staff considered them too stressed to participate. Yet, parents told the researcher that only families themselves had the right to choose what they did and did not participate in. Parents held strong beliefs that staff should not make those kind of decisions without consulting the families. While recognizing that participation in palliative care research presents potential problems, automatically disqualifying terminally ill patients and their families from research may be viewed as paternalistic (Mount, Cohen, MacDonald, Bruera, & Dudgeon, 1995). Mount et al. argue that restricting someone's rights to take part in research is devaluing and being disrespectful of that person. In addition, it suggests that they are incapable of autonomous decision making, participating in society, giving to others, or finding purpose and meaning (Mount et al.). Bruera (1994) suggests that, at least with adult patients, 80-90% of palliative patients and their families who are approached to take part in a study may consent. However, there is a perception and belief that clients should be protected from the potential harm of research, especially where children are involved. Yet, this protectionism silences the voices of vulnerable people (Aranda, 1995). As found in this current study, researchers suggest that participants often consent so that they can give something back and help others in similar circumstances (Aranda; Kristjanson, Hanson, & Balneaves, 1994). Moreover, even though the trauma associated with talking about dying experiences may be painful, it may not be harmful. In the study by Aranda, some patients and their families were upset and crying. Yet, they wanted to continue in the research because it gave them a chance to talk about a loved one when no one else was willing to listen anymore. The parents in this current study sometimes reacted in the 75 same manner. T h e potential value o f participating in research should not be undervalued. Furthermore, the participants in a study act as teachers to the researcher, and ultimately to other health professionals. T h e y are teaching lessons that cannot be learned otherwise. Changes within the pediatric hospice program, including changes in contact personnel and establishment o f a Research Rev iew committee after the research had been approved and initial data col lect ion had started, and misperceptions among the management team about research with palliative care families slowed recruitment. It was never clear to the researcher how many families in the program met the study criteria as the numbers shifted over time. A s best as could be ascertained, two families were asked to participate, but refused because they were too busy. In addit ion, two other families left the program during the study per iod so the staff wou ld not approach them. There are no available data on how many other families were not approached by staff. A t the request o f the staff and despite the researcher's desire for consultation, families who met the eligibil ity criteria for the study were identified solely by the cl inical staff. Initial contact with potential participants was made by the personnel who identified the fami ly as suitable for the study. The nurse explained to the child's parent that a nursing study was in progress, and that agreeing to talk with the researcher d id not commit the fami ly to participating in the study. T h e researcher was clearly differentiated f rom the care provider, and verbal consent was obtained to a l low the nurse to forward the family member's name and telephone number to the researcher. The researcher first contacted parents by telephone. Af ter the study had been explained to the parents, they were invited to participate in the study. A l l fami ly members were encouraged to participate. T h e y were informed regarding the protection o f their rights, and the decision to participate was entirely their own choice. W h e n parents agreed to participate, an appointment was made for interviews and participant observations. The researcher asked the parents to 76 suggest the most appropriate time and place for her to come and observe them as they managed their daily lives. All parents wanted the interviews and observations to be done in the family home. Some parents were interviewed on the same day as their spouse, while others were only available on separate days. The family interview occurred after the individual members' interviews were completed. At the first appointment, the process of informed consent was explained again. The participants signed duplicate copies of a consent form, and retained one copy. Children could not legally give consent. Provincial laws specify a legal age of consent, usually with some exceptions, and competence is determined based on maturity. Therefore, parents signed a consent for minor children. However, children who were able to communicate also needed to give assent for participation. Al l of the ill children and one of the siblings were unable to communicate. The researcher obtained assent from all siblings who were interviewed, although the parents legally signed the consent. Assent meant that the sibling showed understanding of the purpose of the research, knew what she could expect and what was expected of her, and was willing to participate. Consent was obtained from parents for observations of the children who could not be interviewed. When consent had been obtained, the researcher conducted in-depth, open-ended interviews with families to obtain the participants' perspectives. Each family member who agreed to participate was interviewed individually, and then the family was interviewed as a group. This order of interviews allowed family members to say what they wanted in confidence before meeting in the larger group (B. Davies et al., 1995). The researcher was also more prepared for anything that might come up in the group interview since she could anticipate areas of potential difficulties for individuals. There is little guidance in the literature for how to approach multiple sources of qualitative data in family research. B. Davies et al. assumed that all members' perceptions of the situation were equally relevant and contributed to the whole picture of the 77 family, whether or not similar information was obtained in individual and group interviews. A similar assumption was made in this study. Confidentiality can be difficult to maintain when case studies and direct quotations are used in reports, so participants were told about the small risk of public disclosure (Larossa, Bennett, & Gelles, 1981). All interviews ended with a question that asked the participants what it had been like for them to take part in the interview. Interviews lasted from a half hour to three hours, depending on the age, physical, and psychological condition of the participants. Demographic information was collected from each informant to aid in describing the sample. Interviews were tape-recorded and subsequently transcribed. The researcher transcribed the majority of the interviews. Individual family informants were asked to tell the researcher about their experiences when a child has a neurodegenerative, life-threatening illness (NLTI). Both current and retrospective information was sought. The interviews began with an open-ended prompt such as, \"Tell me about your life since (child) was diagnosed with (an NLTI). What has this experience been like for you?\" The purpose of this approach was to elicit the informant's perspective with as few prompts as possible. Other broad, open-ended questions that guided the first interviews included, \"Tell me about a typical day. How does your day start? What happens next? Is a typical day today the same as it was a month (or year) ago?\" \"What is the same?\" \"What has changed?\" \"What has helped you care for (child)?\" \"Can you tell me about times when something or someone has made it easier.\" \"What has made it difficult for you to care for (child)?\" \"Have there been times that have been particularly difficult?\" \"Can you give me an example.\" \"What is life like for you now?\" One young sibling provided some data by drawing a picture. The researcher then asked her about the meaning of her picture. This discussion was recorded in field notes and on audiotape. This technique has previously been successfully used with children as young as three years and ten months (Dowden, 1995). 78 In family group interviews, the focus was on the family unit. Questions, therefore, were reworded to reflect the family experience. Again, both current and retrospective information was sought. A t the beginning of the family interview, families were reminded about the focus of the individual interviews and were advised that some of the questions in the family interview were similar to those in the individual interview. However, the focus of the individual interview was on how the individual was managing the situation and how the individual saw the family managing. In contrast, the family interview was to determine how families managed these kinds of situations. Families were advised that some of the responses in the family interview might be the same as in the individual interviews, but other responses might be different. Questions included, \"Tel l me about your family's life since (child) was diagnosed with (an NLTI ) . What has this experience been like for your family?\" Other broad, open-ended questions that guided the first interviews included, \"Tel l me what you do together as a family.\" \"Is this different from before (child) was sick?\" \"What is the same?\" \"What is different?\" \"Why might that be?\" \"What has helped your family as you live with (child's) illness?\" \"Can you tell me about times when someone or something has made it easier.\" \"What has made it difficult for your family to care for (child)?\" \"Have there been times that have been particularly difficult?\" \"Can you give me an example.\" \"What is life like for your family now?\" As the data analysis progressed, interviews with individuals and families became more specific. Questions were then developed depending on the theoretical requirements of the study. Observations of the family were also used to get to know the family better. Observations were both supplementary and complementary to interviews and were made before, during, and after the interviews. These observations of interactions, such as among the family members or between the family and care providers, lasted for half an hour to three hours. The researcher observed families on between one and three occasions, depending on the families' wishes. Participation in family activities was generally inappropriate as most of the activities revolved 79 around providing care to the ill child. However, where possible, the researcher joined in family activities. For example, she shared lunch with one family, walked to school with another family, sat on the floor with one family as the parents played with their children, and observed and spoke with other people who were present in the family home. The influence of the researcher on families was minimized by the sheer volume of activities that families needed to perform. Families were usually so involved in caring for their child that the researcher's presence was secondary and seemingly insignificant. Observations focused on how families managed the care of their child while also managing their daily lives; what supports were available to families; how effective those supports appeared; and how care was provided to these families. In addition, the researcher asked questions and carried on informal brief conversations with families and others following each interaction to clarify, validate, and extend observations. Other people involved in these interactions included visiting nurses, care attendants, and child care workers. Verbal rather than written consent was obtained from each person who was observed but not interviewed. Brief field notes were made in each setting. These notes were expanded and recorded on paper or audiotape as soon as possible after the observation period. Full, expanded or transcribed notes were entered into a computer program for data management. As the data analysis progressed, observations focused on checking emerging concepts from the data. An attempt was also made to negate, validate, or extend those concepts. Interviews and observations of families can be particularly difficult because of aspects of the family that create special circumstances. Larossa et al. (1981) noted the importance of informed consent and the risk/benefit equation in qualitative research involving families. Since it is never possible to design a totally risk free study, nor is it usually possible in qualitative research to be fully aware of the complete nature of the project or potential outcomes, one should view consent as an ongoing process. Therefore, the researcher asked each participant at every 80 observation and interview if he or she wished to continue with the study or to withdraw with no repercussions. Consent was obtained on an ongoing basis at each stage in data collection. Informed consent is also difficult to obtain because the pervasiveness of family life means that the range of salient issues will be extensive, and that seemingly unrelated, and often unanticipated, issues may merge during the study period. Thus, the researcher cannot forecast the exact nature of the research. Moreover, family life also has a certain inaccessibility because one cannot become a part of the family, so interviews may be the most feasible and most efficient method of collecting data from families. However, the unstructured nature of interviews can almost coerce respondents into answering, even if they would not normally talk about a subject. Conjoint interviews can be especially difficult, since they may provide less opportunity to exercise informed consent (Larossa et al., 1981). The researcher was sensitive to these issues, and interviewed each participant individually prior to conjoint interviews. Hence, the researcher became aware of issues that one or more of the family members did not wish to talk about in the group, and could guide the interview away from those areas to protect the family member. The natural setting of the home may provide a rich context for research, yet the informality of the setting can also encourage friendliness, trust, and self-disclosure. The researcher may be treated as a guest and people may be lulled into doing and saying things they do not want recorded in a study (B. Davies et al., 1995; Larossa et al., 1981). Serendipity can also play a large role with resultant questions about invasion of privacy and consent. Additionally, the researcher is in a position of power, and a power-dependency relationship may develop, or a skilled interviewer may encourage vulnerable people to answer questions that would not normally be answered. Finally, the resemblance of qualitative family research to therapy may lead to role ambiguities. Participants may disclose more to \"their therapist\" than they would to a researcher. Both participants and field workers may be uncomfortable when role confusion is present. According to Larossa et al., the researcher can prevent or alleviate this 81 confusion by acting like a researcher, rather than like a friend. However, B. Davies et al. noted that clinical researchers may have great difficulty in maintaining a researcher role. Awareness of these potential difficulties, critical reflexivity, and constant examination of actions and interviewing techniques helped the researcher to reduce the risk of misuse of her position. Data Analysis Analyzing data of this nature is always intellectually and emotionally challenging (B. Davies et al., 1995). The interviews involved discussions about death and dying in children and portrayed family disruptions that caused the researcher to experience sadness and pain. The researcher needed opportunities to talk about and to deal with the emotional impact of the data. Her dissertation supervisor, and to a lesser extent her dissertation committee, provided a source of support when needed. In addition, the researcher discussed her experiences with colleagues who have conducted research in sensitive areas. B. Davies et al. noted that research of this nature should not be a solitary endeavour because of the potential for emotional reactions to the data. Therefore, the researcher analyzed the data in collaboration with her dissertation supervisor. This collaboration was an opportunity to gain greater insight into the data because of dual perspectives in analysis. In all discussions, anonymity of the participants was respected. Management and analysis of transcribed field notes and interviews were facilitated by the use of a word processor computer program. Each transcript was entered sequentially, and all interviews and observations pertaining to a family were kept in unique directories. Transcripts were coded directly on the computer, and codes were embedded in the transcripts using a bold style to facilitate easy compilation of a code list. In addition, data could be searched for particular codes in different transcripts simply by using the \"Find\" feature in the word processor program. Sections of data from different transcripts could then be copied to a new file for ease of retrieval. Code lists were dated and were compiled for every transcript, and for each time a transcript was coded. Codes were linked to the appropriate transcripts. Each recoded transcript 82 was dated and numbered to facilitate retrieval. Over time, code lists were merged and used for further analysis. Al l prior code lists were retained in order to keep track of the development of codes and categories. Memos were written whenever the researcher was theorizing or worrying about anything. Participants' code numbers and the date were included on each memo. The researcher wrote out her thoughts and included the code names that she was thinking about when she felt the need to write a memo. Al l memos were carefully linked to the appropriate transcript. As data analysis progressed, the researcher drew diagrams to capture the relationships between categories and subcategories. Hard copies of all transcripts, code lists, memos, and diagrams were kept in binders for easy access during data analysis. Data analysis was concurrent with data collection, since analysis guided the questions for future data collection. Theoretical sampling decisions were made on an ongoing basis, depending on the needs of the study. For example, as data were analyzed, the researcher sought to interview families who did not live in Vancouver, were non-Caucasian, had more than one ill child, had siblings, or lived in varying socioeconomic circumstance. The coding procedures laid out by Strauss and Corbin (1990) were used to analyze the data. These procedures build, rather than test theory. They give rigour to the process, help break through the researcher's biases and assumptions, and provide the grounding, build the density, and develop the sensitivity and integration needed to generate a rich, tightly woven, explanatory theory that closely approximates the reality it represents (Strauss & Corbin). The researcher moved between one type of coding to another in a given coding session, especially from open to axial coding, and continually asked many and diverse questions about the phenomenon under study. Constant comparative analysis of each transcript and field notes was used to check for commonalties and differences both within and across transcripts. Data from individuals were compared and contrasted with both similar and dissimilar transcripts from individuals, such as a mother being compared with another mother and contrasted with a father. Group data were 83 compared with other group data, and were also contrasted with individual data. In addition, transcripts involving one family (observations, individual family member interviews, and the group interview) were compared and then contrasted with data from other families. Thus, commonalties and differences within and between individual and family data were discovered. \"Open coding\" was the first level in data analysis (Strauss & Corbin, 1990). Open coding is the process of breaking down, examining, comparing, conceptualizing, and categorizing data. Initially, open coding was non-judgmental and only named the ideas that were in the data set. Each transcript was broken down into units and open coded two or three times, at least in the early stages of analysis. Transcripts were examined line by line to identify codes or words that captured the meaning of the events (Strauss & Corbin). Once a number of transcripts had been open coded a few times, the researcher conceptualized the data by giving similar phenomena the same conceptual name. There were many conceptual labels, so these concepts were then grouped into preliminary categories to reduce the number of units with which the researcher needed to work. The conceptual name given to a category by the researcher reflected the data that it represented. Questions were continually asked of the data to develop categories in terms of their properties (characteristics or attributes) and dimensions (locations of a property along a continuum). This process formed the basis for making relationships between categories and subcategories, and later between major categories. The next level of analysis involved \"axial coding\" where the researcher made connections between a category and its subcategories (Strauss & Corbin, 1990). Categories were linked and developed by asking questions and making comparisons. The researcher asked questions about the type of relationship, in terms of the conceptual labels, and looked at the data for evidence, incidents, and events that supported or refuted the questions. Evidence of differences and variation added density and variation to the theory. The researcher continued to look for other properties of categories and the dimensional location of each incident that was coded. Patterns in 84 the data were noted to provide the basis for selective coding, and concepts and relationships were constantly verified with the data. A coding paradigm (Strauss & Corbin, 1990) was used to specify a category (phenomenon) in terms of the conditions that gave rise to it: causal conditions, context, intervening conditions, action/interactional strategies, and resultant consequences. A phenomenon is the central idea or event towards which a set of actions/interactions is directed to manage or handle it, or to which the set is related. In order to discover a phenomenon, the researcher asked questions such as, \"What is it that these data are referring to? What is the action/interaction all about?\" Causal conditions are the events or incidents that lead to the occurrence or development of a phenomenon. Words such as when; while; since; because; due to; and on account of were often cues to causal conditions, although the researcher sometimes needed to check back from the phenomenon to find the causal (antecedent) condition. The context provides a specific set of properties that pertain to a phenomenon and shows how the phenomenon is handled under a specific set of perceived conditions, whereas intervening conditions are the broader structural context pertaining to a phenomenon that act to facilitate or constrain the action/interactional strategies taken within a specific context. These conditions need to be managed. Action/interactional strategies are purposeful and goal-oriented, and are done for some reason in response to or to manage a phenomenon. Words to look for included action-oriented verbs or participles. Finally, consequences are the outcomes of the action and interaction. Consequences were events, happenings, or responsive actions/interactions that were actual or potential and that occurred in the present or in the future. The consequences of one set of actions sometimes became part of the conditions in another sequence. \"Selective coding\" is the process of selecting the core category (central phenomenon around which all the other categories are integrated) by systematically relating it to other categories, validating those relationships, and filling in the categories that need further 85 refinement and development (Strauss & Corbin, 1990). This is not a linear process, since the researcher must move back and forth between \"steps.\" Selective coding is similar to axial coding, but occurs at a more abstract level. At this level, the researcher tried to identify the story by writing a memo that captured the essence of the story in a few sentences. Then the researcher moved beyond description to conceptualization and told the story analytically. The central category was named with a high level of abstraction, and the chosen conceptual label fitted the story it represented. Questions that the researcher used to identify the central category included, \"What phenomena are reflected repeatedly in the data? How can the findings be summarized? What essential message about this area does the researcher want to pass on to others? What can be considered important about this area and why?\" The core category was developed in terms of its properties and then dimensions. Al l other categories were then related to the central category in subsidiary categories by means of the paradigm: conditions, context, intervening conditions, strategies, and consequences. When certain categories were consistently emerging and no new information was being discovered, then theoretical saturation was reached and data collection ceased. However, refining the theory continued into the writing phase and necessitated the researcher returning to the participants to collect further data to fill any gaps that existed in the theory (Strauss & Corbin, 1990). The developing theory was grounded by validating it against the data. The researcher looked for a general fit, not necessarily a perfect fit in every case, and modified the theory until a general match was made. When the theoretical framework held up to scrutiny and the conditions and processes were built in and accounted for, then the researcher went back to the categories and filled in any missing detail to give conceptual density and increased conceptual specificity. During development of the theory, the researcher also attempted to identify the process that was involved in the families' experiences by linking action/interactional sequences. Retrospective data obtained from families were used in this analysis. The researcher noted 86 changes in conditions that influenced action/interaction over time, the action/interactional responses to that change, and the consequences that resulted from that action/interactional response. She then described how those consequences became part of the conditions influencing the next action/interactional sequence. Process is often a difficult idea to grasp, because identifying process involves an in-depth examination of and incorporation of changed action/interaction into the analysis, as this varies over time and in response to changes in conditions. Because process explains growth and development, or the lack thereof, then the researcher needed to account for or explain any variation in the rate of movement, or why a passage downward might reverse itself or be interpreted, and with what consequences. Analysis was aided by the use of memos. In grounded theory, memos have a specialized meaning - they are the written records of analysis related to the formulation of theory, and they represent the researcher's abstract thinking about the data (Strauss & Corbin, 1990). Memos help the researcher to move away from the data to abstract thinking, and then to return to the data to ground the abstractions in actual data. Memos also show the theory developing step-by-step, and are crucial in developing a dense, well integrated theory. Memo writing began at the inception of the research project and continued until the final writing. Different types of memos were used to facilitate analysis. Memos related to codes contained the actual products of the three types of coding, such as a category with some of its properties and underlying dimensions. Theoretical memos were theoretically sensitizing and contained the products of inductive or deductive thinking about categories, their properties, dimensions, relationships, and variations (though they were provisional until validated with data). Operational memos contained decisions and directions to the researcher regarding sampling, questions, possible comparisons, and leads to follow. 87 Scientific Rigour Qualitative research must be rigorous, but does not use the same criteria of scientific rigour as quantitative research. Criteria include truth value or credibility, applicability or fittingness, consistency or auditability, and neutrality or confirmability (Hall & Stevens, 1991; Sandelowski, 1986; Schutz, 1994). Rigour was maintained by following the procedures of grounded theory (Strauss & Corbin, 1990). Credibility, fittingness, and trustworthiness of the data were ensured by constantly checking to ensure that the coding categories portrayed the data as a whole, and by checking with families to ensure that the interpretation made sense to them. In addition, presentation of findings to colleagues in the field provided further support for the theory. Theoretical sampling ensured the relevance and comprehensiveness of the data, and adequacy was obtained by assuring that all categories were saturated. Auditability was facilitated by consistency in data collection, format, and analysis, and in consistent routines for recording and reflection. The development of the analysis is demonstrated in this final written research report that reflects the complete research process. Ethical Considerations Ethical clearance was sought and obtained from the University of British Columbia, Behavioural Sciences Screening Committee, and from the board and or ethics committee of each of the facilitating institutions. Each informant was given an explanation of the study. They were told that their participation was voluntary and that they could withdraw from the study, refuse to answer any questions, or terminate interviews or observations at any time. Participants had an opportunity to ask questions before consenting to join the study. Written informed consent was obtained from each participant. In the case of minor children, parents signed the consent form, but children who were cognitively able also gave their assent to participation. In addition, ongoing consent was obtained verbally at each interview or observation. Informants were assured that all information would be confidential, and only the researcher would have access to 88 the identifying information. Data were kept in a locked cabinet that could only be accessed by the researcher. Selected portions of the data were shared with the researcher's dissertation committee for purposes of illustration and analysis, however, anonymity was preserved. All participants were asked if they wanted to receive any of the results, and in what format. Some chose to review their personal interview transcripts, while others requested a summary article. All parents asked for a copy of the final research report. Transcribed interviews were mailed to participants who requested a copy of their own interview. No individual's interview was given to another member of the family. Each family member's interview was mailed in a separate envelope. The envelope was addressed to one particular individual and contained only that individual's transcript. Parents confirmed that confidentiality would then be maintained. All identifying information was removed from transcripts, field notes, and demographic information. Code numbers were assigned to each participant. The master list of code numbers and names was kept separately from the data in another locked drawer that was accessible only to the researcher. At the conclusion of the study, all identifying information and audiotapes will be destroyed. Transcribed data will be retained for educational and other research purposes, subject to further ethical approval according to standard procedures. Summary This research was designed as a grounded theory study that focused on the experiences of families with a child who has a neurodegenerative, life-threatening illness. A symbolic interactionist perspective guided the method. The study was conducted in a natural setting; in the homes of families who agreed to participate. Interviews, supplemented by observations, were used to enhance understanding of the experiences of families with a child who has a neurodegenerative, life-threatening illness. Every member of the family was encouraged to participate. Interviews were conducted with individuals, and then with the family as a whole. This study generated a detailed, contextually-grounded description and theoretical explanation of 89 t he e x p e r i e n c e s o f t h e s e f a m i l i e s . T h e k n o w l e d g e g e n e r a t e d b y t h i s s t u d y a n d i ts c o n t r i b u t i o n to d e v e l o p i n g t h e o r y w i l l b e u s e d to p r o m o t e e f f e c t i v e p a l l i a t i v e c a r e o f f a m i l i e s w i t h a c h i l d w h o h a s a n e u r o d e g e n e r a t i v e , l i f e - t h r e a t e n i n g i l l n e s s . T h e g r o u n d e d t h e o r y tha t e v o l v e d t h r o u g h i n t e r p r e t a t i o n o f t he d a t a i n t h i s s t u d y w i l l b e p r e s e n t e d i n t he f o l l o w i n g c h a p t e r . I n c h a p t e r s i x , t he p r o c e s s that f a m i l i e s w e n t t h r o u g h w i l l b e d i s c u s s e d a n d p l a c e d w i t h i n t h e c o n t e x t o f t he c u r r e n t r e s e a r c h a n d l i t e ra tu re . T h e s t u d y w i l l b e s u m m a r i z e d i n t he f i n a l c h a p t e r , a n d i m p l i c a t i o n s f o r p r a c t i c e , e d u c a t i o n , a n d r e s e a r c h tha t a r o s e from t h i s s t u d y w i l l b e i d e n t i f i e d . 90 Chapter V: Findings Navigating Uncharted Territory: Experiences of Families When a Child has a Neurodegenerative, Life-Threatening Illness The purpose of this chapter is to present the grounded theory that evolved through interpretation of the data in this study. This grounded theory of the experiences of families when a child has a neurodegenerative, life threatening illness (NLTI) is the product of multiple phases of analysis. The basic social process of navigating uncharted territory (see Figure 1) characterized the process that these families went through. Navigating uncharted territory was comprised of four main dimensions: entering unfamiliar territory; shifting priorities; creating meaning; and holding the fort. Each dimension will be explored in-depth later in this chapter. Experiences do not occur in isolation. Instead, they occur within a sociocultural context that can influence and constrain a person's ability to act in certain ways. Explicating the context of an experience is very important therefore, because it provides a framework within which the experience can be understood. Consequently, the context of the illness experience will be described prior to presenting the grounded theory. In addition, certain intervening conditions can inhibit or facilitate the strategies people use within a specific context. Four intervening conditions that moderated the families' experiences in this study, relationships with health care providers, availability of information, gender differences, and communication between parents, are described following presentation of the theory. Discussion of the findings and their relationships to current literature will take place in a subsequent chapter. Context of the Illness Experience The onset and diagnosis of a child's neurodegenerative, life threatening illness (NLTI) affected not only the child, but also enveloped the whole family in the illness experience. Moreover, the illness experience evolved in and was created by the interaction between a 91 1-s u 'tJ -3 TO \"t2 3 I-H > C/3 OH O t a. i s : •a '> o £ \\ 2 .Pi > a. o I ~ £3 > s < . ^ H u . O •fc! TJ s • a s T3 C U c O o u I ?60 .£ - » ° •£ .2 so <3 c o -a JS-C c3 o J S 2 & o <5 o \" U > .a O w i - >^ — OH JS OH J 3 CL, o o 2 u .sr S3 J S a> S .S o o •a. e o .2 ;£ o oo £ .£ is O O u c o — .5 = 3 o M 2 c -= 1 § P H o « J S _ 60 ~ •a bp a. E « 3 a o „ oo .£ S .S ts K s y u OH o 8 o ID 60 •a CS: J S c e 60 ,o 60 _o c ts ek £ 'S £ Se intbi Sh infoi 60 o *• '5b ra ra S3 c O ra • rH . s 4—* o , a. 60 P , .5 * too OH ( H 5S H H OH U S-H ra 3 t .2 u o C c o + H