Palliative Care at the End of Life

Consider the following advice given to parents whose children are dying in hospitals: “If your child has to die, he can die peacefully. You can make sure he is free of pain. You can make sure that everyone has a chance to say good-bye” (Hilden & Tobin, 2003, p. 3). To offer parents this kind of unconditional assurance (i.e., “You can make sure . . . ”) dismisses the confusing and disturbing realities of actually having to care for a child when it becomes increasingly apparent that curative intent is failing and staff begin to question how best to proceed. The complexity of symptom control in various clinical conditions sometimes precludes children from having peaceful deaths. However, when you read findings from the few palliative care studies that exist (and there are few that consider children as participants), issues of pain management and psychosocial support at the end of life do not seem to be so difficult to resolve. These findings promote ideas that, when satisfactory end-of-life care is not achieved, it is because mistakes were made, staff were inadequately trained, and children thereby were made to suffer unnecessarily. Such ways of thinking in turn lead bereft parents to feel guilty at not having empowered themselves to have taken greater control in the care of their child and to have done the right thing for their child. Although mistakes occur, staff can be better trained, and children might unnecessarily suffer, there are very few guarantees of a comfortable way of dying from medical causes. Most textbooks and studies about end-of-life care simply ignore the messy realities and uncertainties, particularly as they pertain to children and their families. The Report to the Board of Directors of the American Psychological Association from its Working Group on Assisted Suicide and End-of-Life Decisions (2003) raised a clarion call to document publicly what it is like, in practical day-by-day terms, for people who die in hospitals and how it affects endof- life decisions for the staff, patients, and families. We all prefer to die quickly, without protracted suffering and pain and without humiliation. Deaths during sleep are particularly preferred.

Interventions for Cancer Late Effects and Survivorship

It is now generally accepted that the diagnosis of many pediatric cancers and their treatments result in significant and long-lasting neurocognitive, psychological, and psychosocial impairments and difficulties. The current status of research in this field has been addressed by other chapters in this text. We would, however, like to emphasize at the onset of our chapter that we firmly believe pediatric cancer is truly a family affair. The effects of the diagnosis of a life-threatening illness and its often-chronic treatment not only result in significant impact on the child’s or adolescent’s neuropsychological and psychological state, but also cause psychological ramifications for the parents, siblings, and extended family members. In healthy, well-functioning families, this major life obstacle can serve as an impetus to rally family members in support of the child. When this happens, interventions for late effects are beginning to be identified as effective and of potential benefit. This field, however, is clearly in its infancy. If the family is chaotic and struggling with relationship issues, the prognosis is less positive. Our clinical observations of these relationships are supported both by preliminary data from studies conducted by our research group and others, and by published manuscripts in the field of pediatric traumatic brain injury (Yeates et al., 1997, 2001). In one of the only studies investigating the impact of familial variables on psychosocial and neuropsychological outcome in pediatric brain tumor patients, the results are extremely consistent with the traumatic brain injury population (Carlson-Green, Morris, & Krawjecki, 1995). Reduced maternal dependence on external coping resources, higher parental socioeconomic status, dual-parent families, and familial cohesion were all identified as improving long-term outcome in this population, as documented by intellectual and behavioral integrity. The late effects of pediatric cancer and its treatment are physical, cognitive, psychological, and social. When multiple effects are present, they can be expected to result in a synergistic impact not only on the child, but also on other family members. The important point is that late effects should not be viewed in isolation or summated but should be appreciated for their interrelatedness.

Neuropsychological Late Effects

This chapter reviews neuropsychological late effects associated with childhood cancer and its treatment. The study of late effects presupposes that patients are long-term survivors, if not permanently cured, of their disease. Late effects are temporally defined as occurring after the successful completion of medical therapy, usually 2 or more years from the time of diagnosis, and it is generally assumed that late effects are chronic, if not progressive, in their course. This definition serves to separate late effects from those effects of disease and treatment that are acute or subacute and time limited, such as chemotherapy-induced nausea and vomiting or temporary cognitive changes induced by cancer therapy. Research interest in neuropsychological outcomes, as well as neurological and other functional late effects, has shown an increase commensurate with improvements in effective therapy. For example, 30 years ago when few children were cured of acute lymphoblastic leukemia (ALL), questions related to the ultimate academic or vocational performance of long-term survivors were trivial compared to the need for improved therapy. In contrast, today more than 80% of children diagnosed with ALL can be cured, and issues related to their quality of life as long-term survivors have now received increased emphasis. There is at least comparable attention to neuropsychological status in primary brain tumors. We first provide a brief medical background on the two most frequent forms of childhood cancer, ALL and malignant brain tumors, followed by a review of the current neuropsychological literature. The literature review provides an in-depth analysis of the types of cognitive impairments observed and known or suspected risk factors for impairments. When neurobiological substrates are known, particularly from neuroimaging studies, they are discussed. Finally, we conclude the review with sections that discuss current recommendations for a core battery of neuropsychological assessment of survivors and recommendations for future research. Approximately 20,000 children and adolescents under the age of 20 years were diagnosed with cancer in 1999 (Steen & Mirro, 2000). The most commonly diagnosed cancer in this age group is ALL, a malignant disorder of lymphoid cells found in the bone marrow that migrates to virtually every organ system, including the central nervous system (CNS), via the circulatory system. ALL accounts for one fourth of all childhood cancers and 75% of all cases of childhood leukemia (Margolin, Steuber, & Poplack, 2002).

Quality of Life in Childhood Cancer: Meaning, Methods, and Missing Pieces

It is only in the last three decades that the quality of the lives of children and adolescents treated for cancer and their families has become a major focus in the field of pediatric oncology. This shift from helping families to tolerate arduous treatments and prepare for early death is a result of advances in treatment and survival rates for most pediatric disease categories. One result of this paradigm shift is that quality of life (QOL) has become a critical construct within the field of pediatric oncology. The construct of QOL was initially developed for use with adult populations and was based on the definition of health generated in 1948 by the World Health Organization (WHO): “a state of complete physical, mental, and social well being, and not merely the absence of disease or infirmity.” Although there remains no universally adopted definition of QOL, the WHO’s definition of QOL as an “individual’s perceptions of their position in life in the context of the culture and value system in which they live and in relation to their goals, standards, and concerns” is frequently employed (WHO, 1993). This definition includes several domains that are considered central to the QOL construct: physical, mental/emotional, and social. This initial construct has been expanded with adult populations to include physical symptoms and functioning, functional status (i.e., ability to participate in daily and life activities), psychological functioning, and social functioning (e.g., Ware, 1984). This more expansive definition is referred to as health-related quality of life (HRQOL). HRQOL emphasizes the impact of health on one’s QOL but looks further to include other domains of life functioning that are also potentially affected by health/illness states (Jenney, 1998). The HRQOL construct was initially developed for populations of adults living with chronic illness to assess the impacts of illness/injury/disability, medical treatment, or health care policy on an individual’s life quality (for reviews, see Aaronson et al., 1991; Patrick&Erikson, 1993; Speith&Harris, 1996). Over time, there have been modifications and developments in the construct, approaches to measurement, and the measures themselves (Wilson & Cleary, 1994).

Psychosocial and Behavioral Issues in Stem Cell Transplantation

Stem cell transplantation (SCT) or bone marrow transplantation (BMT) has evolved from a heroic, experimental therapy of last resort to a standard therapy for many high-risk leukemias and the preferred first option after leukemic relapse (Sanders, 1997; Santos, 2000; Treleaven & Barrett, 1998; Wingard, 1997). The indications for SCT have widened to include a number of other malignant disorders, including lymphomas, solid tumors, and even brain tumors, as well as to a growing number of nonmalignant disorders (Meller & Pinkerton, 1998; Santos, 2000; Treleaven & Barrett, 1998). The growth of bone marrow registries that allow for wider use of unrelated donor transplants and developments in stem cell selection techniques that allow for haplotype transplants using mismatched family donors, including parents, have greatly increased the availability of SCT as a viable treatment option for seriously ill children (Mehta & Powles, 2000). At the same time, advances in supportive care have led to improved survival outcomes and thus to a rapidly growing number of long-term survivors of SCT (Santos, 2000; Treleaven & Barrett, 1998). Yet, despite the extraordinary medical and technical advances that have saved the lives of many children, SCT remains a high-risk medical procedure involving a prolonged and physically demanding treatment regimen that can challenge the coping capacities of patients and their families. Psychosocial research in pediatric SCT has progressed more slowly, but available studies indicate that SCT is a stressful experience that can have a negative impact on the social functioning, self-esteem, and general emotional well-being of survivors (Barrera, Pringle, Sumbler, & Saunders, 2000; McConville et al., 1990; Parsons, Barlow, Levy, Supran, & Kaplan, 1999; Phipps, Brenner, et al., 1995; Phipps & Barclay, 1996; Rodrigue, Graham-Pole, Kury, Kubar, & Hoffman, 1995; Stuber, Nader, Yasuda, Pynoos, & Cohen, 1991; Vannatta, Zeller, Noll, & Koontz, 1998). A number of studies have also focused on parental response to SCT (Barrera et al., 2000; Kronenberger et al., 1998; Manne et al., 2001, 2002; Phipps, Dunavant, Lensing, & Rai, 2004; Rodrigue et al., 1996; Streisand, Rodrigue, Houck, Graham-Pole, & Berlant, 2000). Much of the literature to date has focused on long-term outcomes in survivors, particularly neurocognitive and academic outcomes.

A Prospective and Retrospective View of Pediatric Hematology/Oncology

How does one help a family whose child has been diagnosed with a life-threatening illness? It is a deceptively simple question with complicated answers. This brief chapter is not meant to be a history of biopsychosocial pediatric oncology, and it does not cover every theme. The explosion of studies on children with cancer over these past decades (Pizzo & Poplack, 2001) renders a retrospective look formidable and subjective. The sole purpose of this retrospective examination into the earliest beginnings is to place into context some of the main themes that have appeared over the past years, so that they can serve as a foundation for our recommendations for future intervention and research in the field. That is our assigned task. Much of the review reflects personal respective experiences beginning in the late 1960s. The chapters that form this volume, written by many of the most experienced psychosocial researchers who have brought the field so far forward over these many years, are the state of the art, tell us where we have been most recently, and tell us in greater detail where we are at the moment. Where does our psychosocial history begin? What have we done these past many years to help the children and their families cope with the illness and its treatment? With due awareness of the subjectivity and inevitable unfairness of our venture, we undertake the task with due apologies for any omissions that may occur in this retrospective review. As we begin to look in some detail at the main themes formed over the past four decades, we place our review into the context of four preambles: a multidisciplinary and international effort; an alliance between physicians and parents; research and service; and a sharing of the research wealth with economically struggling countries. From the earliest years, the effort to care for the child with cancer has been multidisciplinary, multi-institutional, and international, involving a highly cooperative and collaborative effort of physicians, nurses, psychologists, social workers, and allied health care professionals working together across national borders.

Genetic Issues

The hallmark of genetic medicine is that medical concerns of one individual are germane not only for that person, but also for their offspring and future generations. Pediatric cancer patients, survivors of childhood cancers, their parents, and their siblings all have concerns that are likely to be increasingly affected by advances in genetics. It has been estimated that between 10% and 15% of pediatric cancers are either hereditary or familial in origin (Quesnel & Malkin, 1997). Advances in genetics will help identify pediatric cancers that have a hereditary origin as well as genetic syndromes, identified by other phenotypic features, which put the mutation carrier at increased risk for some types of cancer. Knowledge about hereditary cancer syndromes will help determine the risk of a second malignancy in pediatric survivors, risks of cancer in siblings, and of course, cancer risk for future offspring. To accurately estimate the risk of second primary malignancy in survivors, it is crucial to be able to identify and separate those survivors who do and do not have cancer of hereditary origin because the latter group typically faces excess risk for developing other cancers that are part of the inherited cancer syndrome. Genetic testing is currently available for several cancer syndromes that occur in childhood, most of which are relatively rare. In individual cases, predictive genetic testing may improve the likelihood that a cancer may be discovered in its early and most treatable stages. Reproductive technologies that involve genetic analysis may be important for some survivors who wish to avoid having a child with an inherited predisposition to cancer. Also, pharmacogenomics, the matching of pharmacologic treatments to patients based on genetic analysis of likely efficacy or susceptibility to adverse side effects, will likely become an important way in which genetic medicine affects pediatric cancer patients through reducing the burdens of cancer treatment. The ultimate goal of the translation of genetic findings into pediatric oncology practice is the development of targeted strategies to prevent the development of cancer. With genetic advances come questions about the ethical application of genetic technology.

Loss and Grief

There are many types of loss, but in most Western cultures, the death of a child is considered the most difficult loss because of the symbolic meaning and value associated with having children (Rubin & Malkinson, 2001). In such cultures, the death of a child is considered “out of order” and shatters basic expectations regarding the sequence and predictability of events (Rando, 1983; Schmidt, 1987). The loss of a child challenges the evolutionary role of the parent as “protector” and may result in feelings of despair, isolation, and guilt (Finkbeiner, 1996). This reaction to losing a child is perhaps related to the lower mortality rate experienced in many Western cultures. Cultures with higher infant mortality rates may view the significance of a child’s death differently (Eisenbruch, 1984a). There are also differences in how people of various ethnic backgrounds experience the loss of a child within the United States (Kalish & Reynolds, 1976). Despite the extensive history of research and writings on loss and bereavement, there is a dearth of controlled studies specific to bereavement in the pediatric oncology population. Ethical and methodological challenges may account for the limited research in this area. In addition, the increase in the survival rate for pediatric oncology patients over the past several decades has resulted in an emphasis on the study of coping and adjustment of survivors. In the United States, mortality rates associated with pediatric cancers have been declining for over a quarter century. Between 1975 and 1995, the overall decline in mortality was nearly 40% (Ries, 1999). Still, an estimated 1,500 deaths were expected in 2003 among children diagnosed with cancer between the ages of birth and 14 years, indicating that clinicians in this field are still frequently confronted with anticipatory grief and subsequent bereavement issues for patients and families (American Cancer Society, 2003). The current chapter provides a brief overview of relevant bereavement literature in the context of describing bereavement in pediatric oncology and introduces a model of coping with bereavement suited to describing the range of reactions to the loss of a child.

Posttraumatic Stress and Posttraumatic Growth in Childhood Cancer Survivors and Their Parents

Few would disagree that a diagnosis of childhood cancer is shocking and frightening to the child and to the child’s family. In addition to the sudden and dreadful diagnosis, the intensive and often lengthy treatment is extremely stressful. The combination of diagnosis and subsequent treatment frequently leaves children and parents feeling helpless. However, life-threatening illness has historically been categorized as stressful but not “traumatic” in the way that interpersonal violence or natural disasters are traumatic. In fact, the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) (American Psychiatric Association [APA], 1987) stated that “chronic medical illness” was specifically excluded as a potential precipitant of a formal psychiatric diagnosis of posttraumatic stress disorder (PTSD). This was important because PTSD is the only diagnosis in the DSM that includes a cause or precipitating event within the diagnostic criteria rather than simply offering a description of typical symptoms. Clinical observations challenging the perception that cancer was not sufficiently “traumatic” to precipitate the symptoms of PTSD began to be described in the psychiatric literature in the 1980s and prompted a series of field trials in preparation for the fourth edition of the DSM. A group of 24 adolescent cancer survivors and their mothers were interviewed, and findings revealed that a substantial minority of both groups met criteria for a diagnosis of PTSD (Alter et al., 1996). Life-threatening medical illness was subsequently included as a potential precipitating event for PTSD in the text of the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) in 1994. The accompanying text specified not only that the person experiencing the medical life threat might respond with posttraumatic stress but also that this might also be a precipitant for family members. This allowed a common framework for researchers to investigate posttraumatic stress responses to medical life threat in adult patients, spouses of patients, and childhood patients, their siblings, and their parents. In this chapter, we review the growing body of literature that has emerged regarding the epidemiology, correlates, and predictors of posttraumatic stress symptoms in those exposed to medical life threat.

Cancer and Blood Disorders in Childhood: Biopsychosocial-Developmental Issues in Assessment and Treatment

Cancer, sickle cell disease (SCD), hemophilia, and other blood-related and immunologic disorders represent some of the most complex medical conditions of childhood. All involve a diagnosis that is directly associated with a genetic risk that is interpreted within a complex family system. All have complex biology involving multiple organ systems, and all are potentially fatal. All involve treatment that is demanding, both biologically and behaviorally. All inevitably alter the normal course of development, often during critical periods in the lives of children and families. All have potential significant economic and social consequences that include costs of treatment, indirect costs associated with disease management, and potential long-term costs associated with disability. All have potential long-term effects of treatment that may involve additional new diseases or disabilities. Surprisingly, however, hematologic and oncologic diseases of childhood have one other commonality; despite the complexity and high potential for devastating biologic, psychosocial, family, and economic consequences, all have affected individuals and families who do not experience these devastating consequences and in fact demonstrate a biologic and psychologic resiliency that defies conventional wisdom. Understanding the complex interactions among genetic risk; biology of disease; effectiveness and outcome of treatment; child and family coping, adjustment, and resilience; developmental trajectories; and community support is the challenge for investigators and clinicians during this century, particularly as basic advances in diagnosis and treatment result in anticipation of probable survival for the vast majority of children with these conditions. It is for this reason that these diseases of childhood are frequently considered from a biopsychosocial perspective (Engel, 1980), although we argue that this term must be expanded to incorporate developmental complexity, particularly when applied to children. Since Engel (1980) first proposed a biopsychosocial model of illness as a conceptual model for understanding and treating functional gastrointestinal disorders, our understanding of chronic illness has increasingly incorporated this perspective. The biopsychosocial model recognizes that illness, and one’s experience of illness, occurs through a dynamic interaction among biologic, psychologic, social, and environmental factors, all of which overlap as potential causes and maintenance factors of symptoms associated with the illness.