INTRODUCTION

The assessment of health-related quality of life (HRQL) among children is an area that has seen considerable growth over the past decade. Whereas clinicians and researchers have previously decried the paucity of available HRQL screening or research instruments for use with pediatric populations (e.g., Bradlyn, Ritchey, Harris, Moore, O'Brien, Parsons, Patterson, and Pollock, 1996), there have been more recent calls for a moratorium on new instrument development (Feeny, Barr, Furlong, Hudson, and Mulhern, 1999). Generic and disease-specific HRQL instruments are routinely used in certain clinical situations and are now more commonly included as part of pediatric research protocols. However, these HRQL clinical screening or research instruments have not been fully applied in end-of-life care for children and adolescents; thus, their utility and validity in this phase of care are essentially unknown. This is in part related to uncertainty about whether or not the definitions and the instruments of pediatric HRQL that emanate from either healthy or chronically ill pediatric populations can adequately represent the HRQL of dying children and adolescents. Despite this uncertainty, certain aspects of the HRQL clinical and research work completed in healthy or chronically ill pediatric populations do have relevance to the HRQL of dying children and adolescents.

HRQL is a means by which clinicians and researchers can comprehensively describe the impact of a particular disease or treatment on an individual (or group of individuals), beyond that which is typically conveyed by examination of survival statistics. The purpose of this appendix is to examine the role of HRQL in assessing the impact of end-of-life care with children and to describe the means and methods by which that function may be accomplished. Additionally, in light of the relative novelty of this field of inquiry, we will highlight gaps in the current scientific literature.

Health-related quality of life measurement has emerged as an important focus in clinical trials, clinical practice improvement strategies, and health care services research and evaluation (Fayers and Machin, 2000; Varni, Seid, and Kurtin, 1999a). Although health status, functional status, and health-related quality of life (HRQL) are terms that have often been used interchangeably, a meta-analysis suggests that health status and functional status most commonly are used to refer to the physical functioning dimensions of the broader HRQL construct, while HRQL additionally includes the psychosocial dimensions of emotional, social, and role functioning, as well as related constructs (Smith, Avis, and Assmann, 1999). HRQL instruments must be multidimensional, consisting at the minimum of the physical, mental, and social health dimensions delineated by the World Health Organization (WHO, 1948). While the importance of measuring HRQL in palliative care is increasingly recognized in the adult literature (Paci, Miccinesi, Toscani et al., 2001) the strengths and limitations of HRQL measurement in pediatric palliative care have not been adequately investigated.

WHAT IS HEALTH-RELATED QUALITY OF LIFE?

Numerous definitions have been proposed by investigators; however, most have as their conceptual foundation the World Health Organization's (WHO's) definition of health as “a state of complete physical, mental, and social well-being; not merely the absence of disease” (WHO, 1948). In the adult literature, these dimensions have, at times, been supplemented with dimensions of patient satisfaction and spirituality (e.g., Cella, 1997). In the child literature, clinicians and researchers have noted the importance of considering children's development stage and family impact on quality of life as well. For example, the Children's Oncology Group has adopted the following definition of HRQL: “a multidimensional construct, incorporating both objective and subjective data, including (but not limited to) the social, physical, and emotional functioning of the child and, when indicated, his or her family. HRQL measurement must be sensitive to changes that occur throughout development” (Children's Oncology Group, 2000). Note that HRQL is often described as having both an objective component (e.g., “what were you able to do today?”) and a subjective component reflecting patients' perceptions of their situation and abilities.

The definition that is given to HRQL does influence the measurement approach that will be used by investigators. For example, HRQL definitions may refer to an individual's satisfaction with his or her current level of functioning compared to an “ideal” (e.g., Cella and Cherin, 1988; Hinds, 1990; Vivier, Bernier, and Starfield, 1994), which would typically then require clinicians and researchers to assess satisfaction with current functioning on the one hand and ideal or desired functioning on the other. “Calman's gap” has been used to refer to the notion that HRQL is determined by the difference at a particular time between the hopes and expectations of an individual and that individual's current experience—to improve quality of life you must narrow the gap (Calman, 1984). The extent to which children can grasp the concept of comparing actual functioning to ideal functioning is not well understood however. This suggests that for young children in particular, this approach requires further investigation.

Other definitions of pediatric HRQL focus specifically on the impact of an illness or change in health on children or adolescents (Rosenbaum, Cadman, and Kirpalani, 1990), including any effect on their psychological status, social activity, physical symptoms, overall sense of well-being, or opportunity to have control over life choices (Bradlyn, Harris, Warner et al., 1993; Czyzewski, Mariotto, Bartholomew et al., 1994; Keith and Schalock, 1994). Definitions that reflect more psychological processes and states (e.g., Shumaker et al., 1990) typically rely on traditional psychometrically developed psychological measures of satisfaction and well-being.

WHAT IS HRQL IN END-OF-LIFE CARE, PARTICULARLY WITH CHILDREN?

HRQL is perhaps the ultimate outcome of interest in end-of-life care. When a child enters this phase, there is an implicit understanding that the quantity per se of that child's life is no longer the primary issue, but that the quality of the remaining time is the focus. Thus, the emphasis shifts when the child has been definitively diagnosed with a life-ending condition to one of the quality of the child's existence. The World Health Organization describes pediatric palliative care as “the active total care of patients whose disease is not responsive to curative treatment. Control of pain, of other symptoms, and of psychological, social, and spiritual problems, is paramount.” Further, the goal of palliative care is “the achievement of the best quality of life for patients and their families, consistent with their values, regardless of the location of the patient (WHO, 1998).

In a recently released policy paper on palliative care, the American Academy of Pediatrics (AAP) offered a more expansive definition that included an integration of curative treatment and palliation for children experiencing a life-shortening (but not immediately life-threatening) disease from the point of diagnosis forward (AAP, 2000). In adult literature about HRQL and palliative care, the concepts of spirituality, meaning, and life transitions are often included. Their frequent mention indicates that the subjective component of HRQL becomes a more primary focus in end-of-life care. Again, the extent to which this has been empirically studied in the pediatric literature is significantly limited. While acknowledging this limitation, we have chosen to adopt the AAP definition of palliative care as the basis for our comments here on pediatric HRQL and end-of-life care. This definition fits well with the unique aspects of many childhood illnesses (e.g., type, illness trajectory, response to treatment) and the evidence that children are generally treated more aggressively at end of life than are adults, perhaps without sufficient attention to the prevention and relief of distress and suffering.

Most definitions of palliative care emphasize the intent to alleviate disease- or treatment-related symptoms, but as noted by Doyle, Hanks, and MacDonald (1998), symptom alleviation is done to provide “the best quality of life for patients and their families.” Donnelly and Walsh (1996) note that palliative care patients are often multisymptomatic (median number of symptoms for adults is 11), with fatigue and lack of energy being most frequently reported; because of this, HRQL measures must be easy to complete.

Donnelly and Walsh (1996) argue further that HRQL questions need to be highly relevant to end of life and unambiguous in their meaning. Items regarding work and dating (for example) are inappropriate, and this and the overemphasis on the physical domain (rather than “poor” HRQL) may explain the low scores obtained by advanced cancer patients. They note that changes in HRQL need not parallel decline in physical ability and that for advanced cancer patients the frame of reference for evaluating HRQL changes and nonphysical issues emerge and become more important. These authors argue for global (as opposed to multidimensional) assessment—it is simpler and likely allows for increased compliance in those with poor stamina and allows the patient to define the construct.

Of course, short scales with more concrete descriptors do have the capability of being multidimensional. With children this may confer a benefit because we do not fully understand the limits of children's ability to deal with the more abstract concept of “health-related quality of life.”

For the purpose of this appendix, we are defining pediatric HRQL at the end of life as a multidimensional concept, reflecting physical, emotional, and social functioning of the dying child and his or her family. Our definition emphasizes that clinicians and investigators must take into account and incorporate developmental progressions and family functioning and impact. Finally, pending further empirical investigation, issues that are typically identified as important in general end-of-life care—for example, spirituality and life meaning—should be included to the extent that they are demonstrated to be developmentally appropriate.

MODELS OF HEALTH-RELATED QUALITY OF LIFE FOR DYING CHILDREN AND ADOLESCENTS (CONCEPTUAL IMPLICATIONS)

In the past two decades, several models depicting health-related quality of life in children and adolescents experiencing chronic or serious illness have been proposed. Certain of these models are deduced from those developed with adults (Schipper, Clinch, and Powell, 1990), while others have been inductively derived from patient-specific pediatric groups (Hinds, 1990; Juniper, Guyatt, Feeny et al., 1996; Parsons and Brown, 1998; Testa and Lenderking, 1995). These models tend to reflect the domains of physical, emotional, and social health and the presence and intensity of symptoms; a few additionally reflect spirituality and family influence. Other models focus on the child's or adolescent's ability to function, presupposing that limitations in the abilities to complete role-related tasks or to achieve goals have a negative effect on HRQL (Barr, Furlong, Dawson et al., 1993; Feeny, Furlong, and Barr, 1998; Lansky, List, Lansky et al., 1987; Rosenbaum, Cadman, and Kirpalani, 1990). Yet other models have meaning of the illness experience as central to a child's or adolescent's HRQL (Haase and Rostad, 1994). The domains or components of these diverse models are not given differing weights, implying that the domains affect the child's or adolescent's HRQL equally or that our ability to identify meaningful weightings is not well developed at this time.

There are no published accounts of these HRQL models being applied to dying children and adolescents. This is somewhat surprising because HRQL is highly relevant in end-of-life care for children and adolescents. Determining a dying pediatric patient's HRQL could assist in diminishing patient and family acute suffering (such as that caused by unrelenting pain) as well as later suffering for the family survivors of the deceased child (such as that caused by regretting a treatment-related decision). Assessing the dying child's or adolescent's HRQL may also identify the patient or family who is at particular risk of a complicated end-of-life care situation that might be amenable to professional intervention. Additionally, maintaining a focus on the dying child's HRQL just as this was a focus during the child's earlier curative or life-sustaining care can reassure parents that their child will not receive any less care at this final stage of treatment.

Because HRQL has not been systematically studied in children and adolescents receiving end-of-life care, we do not fully understand the exact type and number of domains in a model of HRQL that would best represent this patient group. However, clinical reports and findings from a limited number of qualitative and descriptive studies suggest that an HRQL model applicable to children and adolescents needs to incorporate both child and family well-being, emphasize the presence and intensity of symptoms, and be sensitive to the duration and individuality of the dying process.

Dying children and adolescents express strong and at times urgent concerns for the well-being of their family members, clearly conveying that their family's HRQL is central to that of their own (Hinds, Oakes, Furman, 2001a; Hinds, Oakes, Furman et al., 2001b). It is hypothesized, then, that knowing their family's well-being is addressed contributes to a more positive HRQL for the dying child or adolescent. Likewise, the profound negative and life-changing effect on a family of a child member's dying and death has been repeatedly documented (Davies, Deveau, DeVeber, Howell et al., 1998; Gilliss, Moore, Martinson, 1997; Martinson, McClowry, Davies, Kuhlenkamp, 1994). Thus, it is reasonable to measure family members' HRQL in its own right.

Adolescents' descriptions of dying and those of parents of dying children convincingly convey the overwhelming ability of an intense and troubling symptom such as pain or nausea to diminish HRQL to unacceptable levels (Hinds, Bradshaw, Oakes, and Pritchard, in press). This latter report also suggests that domains in an HRQL model for dying pediatric patients do not have equal weight or influence. Although it is important to replicate this research, it may be that only one or a limited number of domains (i.e., symptom-related suffering) need to be assessed in critical end-of-life situations when the time and burden required for assessment are constrained. Models of HRQL for dying pediatric patients and their families may need to also reflect the influence of type of death and duration of illness if the death is illness related, parents' satisfaction with their ability to be “good parents” to their dying child, the ability of the family to maintain some degree of normalcy during the dying process, and the influence of family and child development and culture (Goh, Lum, Chan et al., 1999; Lantos, Berger, and Zucker, 1991; Rushton, 1994). Finally, clinical observations suggest that any model of HRQL for dying pediatric patients needs to reflect the fluidity of the dying process and the existence of a continual transition occurring for the patients, their family members, and their health care providers.

RECOMMENDED DOMAINS FOR HRQL IN END-OF-LIFE CARE

HRQL instruments typically reflect either a “generic” or a “disease-specific” approach to measurement. Generic instruments are appropriate for general population studies and for those situations in which disease- or treatment-related difficulties are not areas of primary interest. Disease-specific instruments are believed to contribute additional information that would not typically be collected via generic instruments, such as alopecia or lymphedema (in the case of cancer), or angina or shortness of breath (in the case of cardiovascular disease). While the relative merit of a generic versus a disease-specific approach to measuring HRQL is a matter of empirical inquiry (Patrick and Deyo, 1989), there are advantages to an integrated modular approach (Sprangers, Cull, Bjordal, Groenvold, and Aaronson, 1993; Varni, Seid, and Rode, 1999b). Disease-specific modules may enhance measurement sensitivity for health domains germane to a particular chronic health condition. A generic HRQL measurement instrument enables comparisons across pediatric populations and facilitates benchmarking with healthy populations but may not be sufficient in palliative care.

In addition to the generic physical, emotional, and social functioning domains delineated by the World Health Organization, the dimensions of role functioning and disease-specific symptoms are essential to include in the comprehensive measurement of HRQL in children, including ill children. Role functioning for children includes the ability to engage in the normal activities of daily living of childhood, which may be play, attending school, or participating in sports and recreational activities, depending on the age and developmental stage of the child. Disease-specific symptoms include pain, nausea, fatigue, and other effects of disease and treatment on child functioning.

Conceptually, disease-specific symptoms, which may be targeted for treatment intervention, are hypothesized to directly affect generic HRQL; that is, targeted symptoms are hypothesized to be causal indicators of HRQL. Thus, in structural equation terminology (Loehlin, 1998), disease-and treatment-related symptoms or problems are hypothesized to be causal indicators of the HRQL construct, while physical, emotional, social, and school (role) functioning are hypothesized to be effect indicators (Fayers and Hand, 1997). For example, in recent studies with the PedsQL (Pediatric Quality of Life Inventory), higher levels of pediatric cancer and rheumatology disease-specific symptoms were causal indicators of lower levels of the PedsQL 4.0 Generic Core Scales (Physical, Emotional, Social, and School Functioning) (Varni, Seid, and Kurtin, 2001), which are the effect indicators of generic HRQL (Varni, Burwinkle, Katz, Meeske, and Dickinson, 2002a; Varni, Seid, Smith, Burwinkle, Brown, and Szer, 2002c). For instance, higher levels of pain and fatigue were associated with lower levels of generic HRQL (Varni, Burwinkle, Katz, Meeske, and Dickinson, 2002a; Varni, Seid, Smith, Burwinkle, Brown, and Szer, 2002c).

These findings provide empirical support for the conceptualization that disease-specific symptoms are predictive of generic HRQL and may have implications for treatments designed to enhance HRQL in pediatric palliative care. Stronger causal evidence would result from a treatment intervention study in which an intervention tailored to target specific symptoms such as pain or fatigue results in decreased symptoms and subsequently higher generic HRQL. However, the selection of HRQL measures for an intervention must include those measures that have demonstrated responsiveness to change and, ideally, have been shown to have an impact on clinical decision-making (Varni, Seid, Knight, Uzark, and Szer, 2002b).

The following briefly delineates specific generic and disease-specific domains and symptoms that encompass the broader construct of health-related quality of life within pediatric palliative care based on findings from the pediatric and adult palliative care literature. Although many of the domains are consistent with existing conceptualizations of HRQL in general, the items that make up the scales measuring the domains may differ to some degree in palliative care or may differ at different points in the end-of-life trajectory (Stewart, Teno, Patrick, and Lynn, 1999).

Physical functioning includes general ambulation and mobility, such as walking and running, as well as the ability to engage independently in the normal daily activities of childhood, such as sports and recreation, bathing, chores, self-care, and other physical tasks appropriate for age. Emotional functioning includes varying levels or intensity of depression, anxiety, anger, and worry. Social functioning includes interactions with friends, classmates, peers, and family members including siblings. School functioning involves the work of childhood and includes the ability to attend school, to concentrate at school, to remember what was presented in class, and to engage in homework and other school assignments. If the child is not able to attend school, it includes the ability to engage in school-related academic assignments such as take-home assignments and homework.

Fatigue is a multidimensional construct, and includes general fatigue (e.g., feeling too tired to start or finish things), sleep–rest fatigue (e.g., sleeping and resting a lot), and cognitive fatigue (e.g., hard to think quickly or problems with remembering). Fatigue was the most frequently reported symptom in a study of children with cancer at the end of life (Wolfe, Grier, Klar et al., 2000). Pain includes intensity (e.g., mild, moderate, severe), location (e.g., head, stomach), and the qualitative aspects of pain (e.g., sharp, pressing, horrible). Nausea includes sensations associated with treatments such as chemotherapy and may result in food not tasting appealing or in lack of appetite. Dyspnea or shortness of breath can be one of the more troubling symptoms in palliative care. Constipation can be associated with medications and the lack of activity. Cognitive functioning includes paying attention, problem solving, and remembering past and current events. Procedural anxiety is associated with medical procedures that are perceived as painful or threatening.

Other suggested domains include communication with physicians, nurses, and other health care professionals about feelings, questions, or health concerns important not only for parents, but also for children. Spirituality includes a belief system in which one attempts to understand the meaning and purpose of life and one's relationship to a higher power or to God. Spirituality may also be conceived as a coping strategy rather than an HRQL domain. For example, praying to God to make the pain and hurt go away may be conceived as a coping strategy (Varni, Waldron, Gragg, Rapoff, Bernstein, Lindsley, and Newcomb, 1996). However conceptualized, spirituality is considered an important construct in palliative care in adults. Whether and how to measure spirituality in pediatrics and how old children must be in order to assess their spirituality beliefs remain to be empirically determined.

Adult Versus Child HRQL Domains

The aforementioned HRQL domains and symptoms represent many of the dimensions that have been delineated in the adult palliative care literature and most from the pediatric palliative care literature (Richards and Ramirez, 1997; Stewart, Teno, Patrick, and Lynn, 1999; Wolfe, Grier, Klar et al., 2000). Not included were domains found in the adult palliative care literature that did not appear applicable to pediatrics, given issues of cognitive developmental stage concerns in children (Thompson and Varni, 1986). For instance, while the importance of the “existential” domain has been demonstrated in adult palliative care (Cohen, Mount, Bruera, Provost, Rowe, and Tong, 1997), the construct requires cognitive development that may be beyond most children.

Table C.1 provides an overview of representative child and adolescent HRQL instruments, with information regarding specific domain content. It can be seen that while most instruments include multidimensional assessment of physical, emotional, and social functioning, some of the “traditional” dimensions assessed in adult palliative care (e.g., spirituality, meaning of life) are generally absent.

TABLE C.1

Representative Pediatric HRQL Instruments.

USES OF HRQL

Richards and Ramirez (1997) identify several potential uses of HRQL information, including (1) clinical management—quantifying symptom severity and functional capacity can enhance the clinical care given to the patient; (2) clinical audit—estimations of the prevalence and severity of specific problems can be accomplished by aggregating patient data, which can then be used to help evaluate the outcomes of an intervention or policy; (3) clinical research—if optimizing HRQL is the goal of palliative care, then HRQL needs to be a primary end point in evaluating new treatments or approaches; and (4) resource allocation—limited resources make it essential that the cost-effectiveness of palliative care be measured and HRQL data provide the means by which treatment impact can be quantified.

In a 1997 Institute of Medicine (IOM) workshop (Teno et al., 1999), four applications of HRQL were noted: (1) clinical assessment, (2) quality improvement, (3) accountability, and (4) research. Each application carries with it a set of assumptions, which may or may not overlap with other applications. For example, quality improvement instruments typically reflect measures of process, while accountability focuses on outcomes.

IMPACT ON CLINICAL DECISION-MAKING

Measuring the quality of care at the end of life may help identify areas of care in which improvement efforts may be directed (Fowler, Coppola, and Teno, 1999). Quality-of-care measures may also be correlated with health-related quality-of-life measures (Seid, Varni, Bermudez et al., 2001). However, an alternative and potentially more direct approach to linking quality of care and HRQL is to investigate the impact of HRQL measurement on clinical decision-making. For example, how is clinical practice impacted by having these data available to health care providers?

The impact of a pediatric HRQL measurement instrument on clinical decision-making can be tested under the working hypothesis that HRQL measurement must occur at the point of service for each individual patient in order to improve health care outcomes (Wetzler, 2000). In pediatric primary care, point-of-service computer-based immunization tracking has been advocated to improve immunization rates (Adams, Conners, Mann, and Palfrey, 2000). In adult primary care, computer-generated feedback of HRQL screening findings, accompanied by problem-specific resource and management suggestions, resulted in subsequent improvements in patient mental health functioning (Rubenstein, McCoy, Cope et al., 1995).

In the absence of standardized HRQL screening information, primary care physicians often fail to recognize problems in patients' HRQL (Calkins, Rubenstein, Cleary et al., 1991). However, previous research with adult patients demonstrated that simply providing primary care physicians with HRQL screening information without specific resource and management suggestions was not sufficient to change either health care provider behavior or patient HRQL outcomes (Rubenstein, Calkins, Young et al., 1989). Thus, a standardized HRQL screening instrument appears to be a necessary but not sufficient condition for enhancing patients' health-related quality of life.

In a study with the PedsQL, families were given the PedsQL as they entered a Children's Hospital specialty clinic (Varni, Seid, Knight, Uzark, and Szer, 2002b). The completed PedsQL instruments were then either placed in the patient's chart or given directly to the pediatric specialist by the family. The pediatrician first went through his or her usual procedures, asking questions about the interim history before examining the child. The pediatrician then asked, “Is there anything else I should know before I look at the PedsQL?” At that point, the pediatrician scanned the completed instruments for “3s” (“often”) and “4s” (almost always”), asked about those items and either dealt with them during the clinic visit (with reassurance, confirmation of the problem being true for many other children, acknowledgment of how hard it is, and other simple explanations) or made a recommendation for an intervention. Interventions included having a social worker call the family, making a referral to a counselor, facilitating school accommodations, prescribing problem-specific pharmacological agents, referral to physical therapy, or referral to another member of the multidisciplinary team. Thus, the pediatrician had knowledge of and access to multiple resources and management strategies that are major components of multidisciplinary team care for many pediatric chronic health conditions.

In pediatric palliative care settings, it would be anticipated that a resource referral and management guide might further facilitate the process if the care was not provided in a multidisciplinary setting. Screening for problems in patient-perceived HRQL may also facilitate shared clinical decision-making (Higginson and Carr, 2001). When HRQL measurement instrument scores are available at the point of service, patient and parent perceptions of the child's physical and psychosocial health can inform clinical decisions by the health care provider (Varni, Seid, Knight, Uzark, and Szer, 2002b).

ILLUSTRATIVE MEASURES OF ADULT END-OF-LIFE HRQL

There are a number of HRQL instruments that have been investigated specifically for adults in end-of-life care. We have included a brief overview of several such instruments to illustrate the range of domains assessed and to provide specific examples of the more subjective domains (e.g., meaningfulness) that have gained acceptance in the care of adults.

McGill Quality of Life Questionnaire (Cohen, Mount, Strobel, and Bui, 1995). The McGill Quality of Life Questionnaire is comprised of four, factor-derived subscales: physical symptoms, psychological symptoms, outlook on life, and meaningful existence. The authors note that the McGill is remarkable for several reasons, including the fact that the existential domain is explicitly assessed, that positive contributions to HRQL are measured, and that the scale does not overly emphasize physical functioning and status. The importance of measuring the existential domain was supported by Cohen, Mount, Bruera, Provost, Rowe, and Tong (1997) who reported that the meaningful existence subscale correlated significantly with a single-item rating of overall quality of life. A recent comparison of the McGill to other HRQL instruments in an outpatient palliative care setting reported that patients preferred the McGill because of its comprehensive-ness and that the physical symptom and support information gathered by the McGill was a relative strength (Pratheepawanit, Salek, and Finlay, 1999).

McMaster Quality of Life Scale (Sterkenburg, King, Woodward, 1996). The McMaster Quality of Life Scale (MQLS) is a multidimensional scale assessing physical, emotional, social, and spiritual domains. It is comprised of 32 items regarding physical symptoms, functional status, social functioning, emotional status, cognition, sleep and rest, energy and vitality, general life satisfaction, and meaning of life. There is a parallel proxy form to be completed by family and staff; the rating scale includes a “cannot tell” response for proxies and a “does not apply,” which can be used for symptoms that may not be relevant for certain patients. Responses to the MQLS are based on a 24-hour time interval to best capture acute changes. The authors report that the scale was reliable and valid with adult palliative care patients, with acceptable intra- and interrater reliability and internal consistency. The mode of administration was found to be important—patients who completed the scale verbally rated their HRQL lower than those who were able to complete it in written format. The scale was found to be sensitive to change and not influenced by age or sex. Although proxy ratings were correlated, systematic differences were found, and the authors recommend that proxy ratings be avoided whenever possible.

Missoula-VITAS Quality of Life Index (Byock and Merriman, 1998). This scale is also focused on an individual's end of life. This 25-item scale is constructed in a manner that allows respondents to differentially weight each dimension and to interpret the items according to their own experiences. The authors report appropriate internal consistency findings and note that total scores correlated with patient-provided global HRQL ratings and another HRQL instrument, although not with observer-rated functional status.

Although the information presented above is representative of investigations with adults only, there are several potentially important factors to consider in future research with children and adolescents. As we have noted throughout this discussion, the extent to which children and adolescents perceive existential and highly abstract domains (such as meaningfulness and spirituality) to be important at this stage of life is largely unknown. While it makes intuitive sense that these domains might have substantial merit among this group of patients, we do not have a good understanding of how to describe and measure them, whether in fact patients describe them as important, or the relationship of these factors to cognitive development.

The adult literature also provides continued demonstration of the difficulties associated with the use of proxy informants. The discrepancy among different proxy informants, as well as the apparent differences between patient and proxy information, underscores the importance of treating these data with caution.

THE HEALTH-RELATED QUALITY OF LIFE OF FAMILIES

This is a potentially significant, but poorly understood, area of inquiry. While we have noted the impact of caring for a dying child on the HRQL of parents and siblings, these investigations are generally at the level of the individual and not at the level of the family as a unit. A review of the existing literature does not reveal any instruments that adequately assess the core dimensions of HRQL as outlined above (i.e., physical, emotional, and social well-being). However, there are instruments that assess family functioning, and one or more of the core HRQL dimensions, to at least some extent. For example, the Quality of Life Index (McCubbin, Olson, Lavee, and Patterson, 1985) assesses the family's overall feeling of satisfaction along a number of dimensions, including schools, health care, and relatives. The Impact on Family Scale assesses the financial burden, social restriction, personal strain, and coping strategies of parents (Stein and Reissman, 1980). However, further investigation is necessary to fully develop a conceptualization and working definition of “family health-related quality of life” as opposed to the more generic concept of “family quality of life.”

ISSUES IN THE ASSESSMENT OF PEDIATRIC HRQL AT END OF LIFE

In selecting (or developing) HRQL instruments for pediatric palliative care, an important explicit goal should be to identify (or develop and test) brief measures for the broadest age group empirically feasible, specifically including child self-report for the youngest children possible. Thus, the measures selected to assess HRQL in pediatric palliative care should emphasize the child's perceptions, including the youngest children empirically possible. It is very important that the items chosen for inclusion be initially derived from the measurement properties of the child self-report scales, while the parent proxy report scales should be constructed to directly parallel the child self-report items. This approach helps ensure that the voices of children are heard in matters relating to their HRQL.

Although it is likely that children and adolescents will have different concerns related to their HRQL, the instruments selected should reflect those items that are of universal concern across age groups. Attempts to keep the wording, and thus the content, of items as similar as possible across parallel forms, while being sensitive to developmental differences in cognitive ability, facilitate the evaluation of differences in HRQL across and between age groups, as well as the tracking of HRQL longitudinally.

There are several palliative care situations that should be identified as presenting particular challenges to the assessment of HRQL and many of the issues raised throughout this appendix. Specifically, assessment of neonates, particularly those in the intensive care setting, is an uncharted area of HRQL. Given that the child is clearly unable to serve as a respondent to typical HRQL measures, investigators and clinicians must address the systematic incorporation and applicability of symptom-based scales (e.g., pain observations) and caregiver proxies as critical sources of information to best understand the infant's HRQL.

An additional situation that has received little attention is that of the acute, immediately life-threatening injury (e.g., motor vehicle accident). This situation is particularly problematic in light of the short and immediate time frame for action, the lack of any prior contact and longitudinal understanding of the patient's preferences, and in some instances, the unavailability of a qualified proxy (such as the parents), to provide a historical context to understanding the patient.

BARRIERS TO STANDARDIZED HRQL MEASUREMENT

Using standardized instruments to measure the HRQL of children in palliative care may meet with some resistance by clinicians who feel that the completion of these measures represents too great a burden for very ill children and their parents. Oftentimes, a more qualitative approach is sought with the belief that such methods are less burdensome and intrusive than standardized quantitative methods. While understandable, similar concerns have been seen in the past in regards to pain assessment in children and HRQL measurement in children with newly diagnosed cancer. These concerns were met with brief instruments (to reduce respondent burden) developed with focus groups and cognitive interviews (to hear the voices of children and their parents) and by a careful attention to the methodological details involved in establishing the reliability and validity of these instruments. Subsequent to the development, field-testing, and documentation of the measurement properties of these standardized quantitative instruments, clinical trials have been conducted targeting pain management and enhancing HRQL through symptom reduction using these quantitative instruments as outcome measures. Thus, standardized HRQL instruments have the potential to improve the lives of children facing life-threatening conditions by providing the systematic documentation of efficacy of treatment interventions designed for pain and symptom relief. In palliative care, standardized measures have the potential to increase the accountability and the quality of the care provided by allowing comparison of health care institutions or practitioners and utilizing that information to inform consumers and aid quality improvement efforts (Teno, Byock, and Field, 1999).

Clearly, there are challenges to be faced using standardized measures in clinical trials, clinical practice, and quality assurance efforts (Jordhoy, Kaasa, Fayers, Ovreness, Underland, and Ahlner-Elmqvist, 1999). For example, systematically addressing these barriers has the real potential to improve the health-related quality of life of children with life-ending conditions.

Differences among observers also represent a barrier to assessment. For example, Wolfe et al. (2000) found poor agreement between parent retrospective reports of their children's symptoms at end of life and physician reports. Additionally, although fatigue was frequently reported as a source of suffering for these children, it was infrequently addressed by the treatment team. Finally, Wolfe et al. (2000) found that more active involvement by the treating physician was rated with higher parent ratings of their child's comfort during the last month of life. Hilden et al. (2001) surveyed the pediatric members of the American Society of Clinical Oncology regarding end-of-life care. Although few had received formal training in palliative care, the overwhelming majority (91 percent) perceived themselves as competent in managing pain for children at the end of life and felt that most of their patients did not die in pain, providing further support for the lack of synchrony between parent and physician perceptions. A recent survey by Burns et al. (2001) reported that while the majority of providers reported views that were in agreement with consensus positions adopted by national organizations, nurses were less likely than physicians to agree that families were well informed and ethical issues were adequately discussed in their intensive care units. This suggests once again that there may be substantial disagreement among the perceptions of providers from different disciplines and family members.

Other potential barriers to effective end-of-life care were also identified by Hilden et al. (2001). Almost 50 percent indicated that they waited for the family to initiate discussions regarding advance directives and other communication difficulties. This is interesting in light of a finding with adult palliative care cancer patients (Detmar, Aaronson, Wever, Muller, and Schornagel, 2001) that psychosocial issues are the most vague in terms of who should initiate the discussion. Although patients and physicians reported it was important to discuss a wide range of issues, the most comfortable topics were around issues of physical functioning. This suggests that psychosocial issues may be underreported by patients and under-identified by physicians, leading to a degree of unrecognized morbidity.

In 1997, the IOM convened a workshop to review quality-of-care and quality-of-life assessment at the end of life (Teno, Byock, and Field, 1999). At that time, it was proposed that instruments should have clinical meaning, be practical, and possess adequate psychometric properties. Additionally, the instruments should assess dimensions that were important to those facing death, should include both patient and family perspectives, and should examine both process and outcome of end-of-life care. In pediatric HRQL, we have a body of literature that sufficiently identifies instruments that are psychometrically sound and practical, but what is most sorely absent is the clear delineation of domains or topics that are relevant to child and adolescent patients.

Important domains have been identified for adult patients at end of life, however. For example, Steinhauser et al. (2001) surveyed seriously ill patients, recently bereaved family members, and health care providers. They reported that pain and symptom management, preparation for death, achieving a sense of completion, decisions about treatment preferences, and being treated as a “whole person” were rated as important by all respondents. Several items were significantly more important to patients than providers, including being mentally aware, having funeral arrangements planned, not being a burden to others, helping others, and coming to peace with God. Although there were many items with substantial agreement across raters, it is important to note that there were differences in patient preferences compared to those of health care providers.

CULTURAL ADAPTATIONS AND LANGUAGE BARRIERS

This is a poorly studied area in pediatric HRQL in general and in pediatric end-of-life care in particular. A number of pediatric HRQL instruments are available in a variety of language translations; for example, the Health Utilities Index and the PedsQL are both available in Spanish and French versions. The inclusion of a broad range of eligible patients in HRQL is important for a number of reasons, including the potential generalizability-limiting effects when investigators fail to include non-English-speaking patients in a clinical trial, in addition to the ethical obligation to offer treatment broadly. However, investigators are challenged to provide scales that have been through rigorous forward and backward translation processes in order to demonstrate cross-cultural equivalence and cross-instrument equivalence. The simple “word-for-word” literal translation of concepts related to support and the meaning of life is not likely to successfully convey the conceptual underpinnings of these terms.

METHODS AND IMPLEMENTATION— MEASUREMENT IMPLICATIONS OF A HRQL MODEL FOR DYING CHILDREN AND ADOLESCENTS

In recent years, the recommended approach for measuring HRQL has been to include a generic measure of health status and a disease-specific measure within the same study or trial and to have the patient complete both forms (Scott and Garrood, 2000). This approach has been used as a way to define and address needs of unique patient groups while also producing responses that may be used to compare one patient group with other groups. Results have been used to compare different treatments, including toxicities and survival, to monitor health status over time, and to identify needed health policy changes (see, e.g., Cramer and Spilker, 1998; Levi and Drotar, 1998). Exceptions to this recommended approach are very likely to be made when measuring HRQL of dying children and adolescents because of the concern with patient burden and the need to give priority to those measures that can sensitively detect change in clinical conditions rather than those that yield data for population comparisons. Future research is needed to identify the utility of different assessment schedules and combinations of instruments, for example, assessing generic HRQL monthly in combination with weekly assessment of symptom-specific items.

It must be acknowledged that in certain situations, children may be unwilling or unable to provide the necessary information for an assessment of their HRQL. As noted in the previous discussion regarding proxy reporting of HRQL, the patient's report is far and away the perspective of choice. In many situations, it has been demonstrated that proxy reporting is not equivalent and, in the case of internal states such as mood and pain, may actually underrepresent significant problem areas. Investigators should consider the use of instruments such as visual analogue scales in situations where completion of a longer HRQL scale may be too burdensome.

In those situations in which proxy report is the only information available, a family member or health care provider who has closely observed and interacted with the dying child and who can accurately interpret the child's physiological and behavioral responses may be the most clinically useful source of estimating that child's HRQL acutely and over time. In such cases, strategies should be implemented to maximize the sensitivity of the family member or health care provider to changes in the dying child's status while avoiding fatiguing that person. For example, the sensitivity of health care providers is heightened when the same individuals are regularly assigned to care for the dying child. In addition, their efforts to document and share their astute clinical observations of the child's responses with other staff can further benefit the child for those times when the provider is not available to provide direct care to the dying child.

MEASUREMENT REQUIREMENTS IN PURPOSE OF ASSESSMENT

Any measure of HRQL must be both reliable and valid for the purpose for which it is being used. In general, reliability refers to the stability or accuracy (i.e., extent to which it is free from error variance) of the score(s), while the validity of an instrument reflects the extent to which it actually measures the intended construct. Additionally, HRQL instruments that are used in clinical trial settings must also be responsive to changes in clinical status in order to be considered useful. However, the purpose for which the assessment is being conducted is also important to consider, since there are unique requirements depending upon whether an instrument is being used for program evaluation purposes, for individual patient clinical decision-making, or for policy determinations.

A NOTE ON HIDDEN MORBIDITY IN PEDIATRIC PALLIATIVE CARE

Previous research in pediatric health care investigating psychosocial health as the “new hidden morbidity” demonstrated the continuing under-identification of psychosocial problems in routine pediatric practice (Costello, Edelbrock, Costello, Dulcan, Burns, and Brent, 1988). It seems reasonable to expect that the underidentification of psychosocial health problems demonstrated in pediatric primary care (Costello, Edelbrock, Costello, Dulcan, Burns, and Brent, 1988) and in pediatric tertiary care for children with chronic health conditions (Varni and Setoguchi, 1992) is also true in pediatric palliative care (Wolfe, Grier, Klar et al., 2000).

In adult patients' communication with their physicians during outpatient palliative care visits, it was demonstrated that patients' self-reported HRQL was the most powerful predictor of discussing HRQL issues with their physicians (Detmar, Muller, Wever, Schornagel, and Aaronson, 2001), with lower HRQL patients being more likely to discuss these issues with their physicians than those with higher HRQL. Even in patients experiencing serious HRQL problems, emotional functioning and fatigue were not addressed approximately 50 percent of the time. These findings suggest the value of systematic assessment of HRQL concerns utilizing screening methods akin to laboratory tests. Similar to laboratory tests for biological disease, screening for HRQL morbidity in a patient population requires a standardized test with established reliability and validity (Varni and Setoguchi, 1992).

To tackle this hidden psychosocial morbidity, HRQL measures may serve as standardized screening instruments for identifying physical and psychosocial health concerns from the perspectives of both the child and the parent at the point of service (Varni, Seid, Knight, Uzark, and Szer, 2002b).

Consistent with a serial screening approach is the recommendation from the American Academy of Pediatrics that an integrated model of palliative care include pain and symptom management at diagnosis and throughout the course of the condition, regardless of the ultimate outcome. From this perspective, all children diagnosed with a potentially life-threatening condition should be screened on a regular basis for HRQL concerns and provided with appropriate palliative therapies based on these serial screenings. In this way, all children, regardless of the potential for cure, would be managed with optimal health-related quality of life as a vital and essential health outcome goal.

CONCLUSIONS

Health-related quality of life is the outcome of uppermost concern in end-of-life care for children and adolescents. When children or adolescents begin this phase of care, there is an implicit understanding that the quantity of their lives is no longer the primary issue; instead, the quality of their remaining time is the principal focus of care efforts. Regrettably, insufficient knowledge is available regarding how children and adolescents die. As health care providers focus more on HRQL of dying children and adolescents through formal clinical assessments and research, medical knowledge and understanding of this phase of care will expand. Physical and emotional care of dying children and adolescents will be improved, and evidence-based justifications for resources that could benefit dying pediatric patients and their family members will be more readily available.

Our purpose in this discussion has been to facilitate the focus on HRQL of dying children by offering a definition of HRQL in dying children, proposing an HRQL conceptual model that includes the well-being of the dying child and the family, and identifying HRQL measurement challenges and strategies. A review of this literature suggests a number of areas in which we have a poor or limited understanding of how to best understand the HRQL of children during this stage of life. In particular, there is limited knowledge regarding how family HRQL should be conceptualized and assessed and the extent to which abstract concepts such as meaningfulness of life and spirituality should be, or are able to be, included in our conceptualization for children. Although researchers will continue to benefit from the work carried out with adults who are dying, it is imperative that our efforts to better understand children who are dying move forward quickly so that children who face this challenge can receive the highest quality of care.

REFERENCES

• Achenbach, T.M., McConaughy, S.H., and Howell, C.T. (1987). Child/adolescent behavioral and emotional problems: Implications of cross-informant correlations for situational specificity. Psychological Bulletin, 101 , 213–232. [PubMed: 3562706]
• Adams, W.G., Conners, W.P., Mann, A.D., and Palfrey S. (2000). Immunization entry at the point of service improves quality, saves time, and is well-accepted. Pediatrics, 106, 489– 492. [PubMed: 10969092]
• American Academy of Pediatrics (2000). Committee on Bioethics and Committee on Hospital Care. Pediatrics, 106(2), 351–357. [PubMed: 10920167]
• Barr, R.D., Furlong, W., Dawson, S., Whitton, A.C., Strautmanis, I., Pai, M., Feeny, D., and Torrance, G.W. (1993). An assessment of global health status in survivors of acute lymphoblastic leukemia in childhood. American Journal of Pediatric Hematology/On cology, 15, 284–290. [PubMed: 8328641]
• Bradlyn, A.S., Harris, C.V., Warner, J.E., Ritchey, A.K., and Zaboy, K.A. (1993). An investigation of the Quality of Well-Being Scale with pediatric oncology patients. Health Psy chology, 12, 246–250. [PubMed: 8500455]
• Bradlyn, A.S., Ritchey, A.K., Harris, C.V., Moore, I.M., O'Brien, R.T., Parsons, S.K., Patterson, K., and Pollock, B.H. (1996). Quality of life research in pediatric oncology: Research methods and barriers. Cancer, 76, 1333–1339. [PubMed: 8826959]
• Brunelli, C., Constantini, M., Di Giulio, P., et al. (1998). Quality of life evaluation: When do terminal cancer patients and health care providers agree? Journal of Pain and Symptom Management , 15, 151–158. [PubMed: 9564116]
• Burns, J.P., Mitchell, C., Griffith, J.L., and Truog, R.D. (2001). End of life care in the pediatric intensive care unit: Attitudes and practices of pediatric critical care physicians and nurses. Critical Care Medicine, 29, 658–664. [PubMed: 11373439]
• Byock, I.R. and Merriman, M. (1998). Measuring quality of life for patients with terminal illness: The Missoula-VITAS quality of life index. Palliative Medicine, 12, 231–244. [PubMed: 9743822]
• Calkins, D.R., Rubenstein, L.V., Cleary, P.D., et al. (1991). Failure of physicians to recognize functional disability in ambulatory patients. Annals of Internal Medicine , 114, 451–454. [PubMed: 1825267]
• Calman, K.C. (1984). Quality of life in cancer patients—An hypothesis. Journal of Medical Ethics, 10, 124–127. [PMC free article: PMC1374977] [PubMed: 6334159]
• Cella, D. (1997). FACIT Manual. Center on Outcomes, Research, and Education, Evanston Northwestern Healthcare and Northwestern University.
• Cella, D.F. and Cherin, E.A. (1988). Quality of life during and after cancer treatment. Com prehensive Therapy, 14(5), 69–75. [PubMed: 3292140]
• Children's Oncology Group. (2000). Guidelines for including quality of life endpoints in COG clinical trials. Unpublished document.
• Cohen, S.R., Mount, B.M., Bruera, E., Provost, M., Rowe, J., and Tong, K. (1997). Validity of the McGill Quality of Life Questionnaire in the palliative care setting: A multi-centre Canadian study demonstrating the importance of the existential domain. Palliative Medi cine, 11, 3–20. [PubMed: 9068681]
• Cohen, S.R., Mount, B.M., Strobel, M.G., and Bui, F. (1995). The McGill Quality of Life Questionnaire: A measure of quality of life appropriate for people with advanced disease. A preliminary study of validity and acceptability. Palliative Medicine, 9(3), 207– 219. [PubMed: 7582177]
• Costello, E.J., Edelbrock, C., Costello, A.J., Dulcan, M.K., Burns, B.J., and Brent, D. (1988). Psychopathology in pediatric primary care: The new hidden morbidity. Pediatrics , 82, 415–424. [PubMed: 3405677]
• Cramer, J. A. and Spilker, B. (1998). Specific populations and specific diseases. In: J.A.Cramer, editor; and B. Spilker, editor. (eds.), Quality of Life and Pharmacoeconomics: An Introduction (2nd ed., pp. 151–191 ). Philadelphia: Lippincott-Raven Publishers.
• Czyzewski, D.I., Mariotto, M.J., Bartholomew, K., LeCompte, S.H., and Sockrider, M.M. (1994). Measurement of quality of well being in a child and adolescent cystic fibrosis population. Medical Care, 32, 965–972. [PubMed: 8090048]
• Davies, B., Deveau, E., deVeber, B., Howell, D., Martinson, I., Papadatou, D., Pask, E., Stevens, M. (1998). Experiences of mothers in five countries whose child died of cancer. Cancer Nursing, 21(5), 301–311. [PubMed: 9775480]
• Detmar, S.B., Aaronson, N.K., Wever, L.D.V., Muller, M. and Schornagel, J.H. (2001. a). How are you feeling? Who wants to know? Patients' and oncologists' preferences for discussing health-related quality of life issues. Journal of Clinical Oncology, 18, 3295– 3301. [PubMed: 10986063]
• Detmar, S.B., Muller, M.J., Wever, D.D., Schornagel, I.H., and Aaronson, N.K. (2001. b). Patient–physician communication during outpatient palliative treatment visits: An observational study. Journal of the American Medical Association , 285, 1351–1357. [PubMed: 11255393]
• Donnelly, S. and Walsh, D. (1996). Quality of life assessment in advanced cancer. Palliative Medicine, 10, 275–283. [PubMed: 8931062]
• Doyle, D., Hanks, G.W.C., and MacDonald, N. (1998). Introduction. In: D. Doyle, editor; , G.W.C. Hanks, editor; , and N. MacDonald, editor. (eds.), Oxford Textbook of Palliative Medicine. Oxford: Oxford University Press.
• Fayers, P.M. and Hand, D.J. (1997). Factor analysis, causal indicators and quality of life. Quality of Life Research , 6, 139–150. [PubMed: 9161114]
• Fayers, P.M. and Machin, D. (2000). Quality of Life: Assessment, Analysis, and Interpreta tion . New York: Wiley.
• Feeny, D., Furlong, W., and Barr, R.D. (1998). Multiattribute approach to the assessment of health-related quality of life: Health utilities index. Medical and Pediatric Oncology Supplement, 1, 54–59. [PubMed: 9659947]
• Feeny, D., Barr, R.D., Furlong, W., Hudson, M., and Mulhern, R. (1999). A postscript to the international workshop on assessing health-related quality of life in children with cancer. International Journal of Cancer , Supplement12, 155. [PubMed: 10679889]
• Fowler, F.J., Jr., Coppola, K., and Teno, J. (1999). Methodological challenges for measuring quality of care at the end of life. Journal of Pain and Symptom Management, 17, 114– 119. [PubMed: 10069151]
• Gilliss, C.L., Moore, I.M., and Martinson, I.M. (1997). Measuring parental grief after childhood cancer: Potential use of the SCL-90R. Death Studies, 21(3), 277–287. [PubMed: 10169695]
• Goh, A.Y., Lum, L.C., Chan, P.W., Bakar, F., and Chong, B.O. (1999). Withdrawal and limitation of life support in pediatric intensive care. Arch Disabled Child , 80, 424–428. [PMC free article: PMC1717910] [PubMed: 10208946]
• Guyatt, G.H., Juniper, E.F., Griffith, L.E., Feeny, D.H., and Ferry, P.J. (1997). Children and adult perceptions of childhood asthma. Pediatrics, 99, 165–168. [PubMed: 9024440]
• Haase, J. and Rostad, M. (1994). Experiences of completing cancer treatments: Child and adolescent perspectives. Oncology Nursing Forum , 21, 1483–1492. [PubMed: 7816675]
• Harris, C.V., Bradlyn, A.S., Aronoff, S.A., Spieth, L.E., Warner, J.E., and Pinzone, H. (1994). Quality of Life in Cystic Fibrosis. North American Cystic Fibrosis Conference. Orlando, FL.
• Higginson, I. and Carr, A. (2001). Using quality of life measures in the clinical setting. British Medical Journal, 322, 1297–1300. [PMC free article: PMC1120388] [PubMed: 11375237]
• Hilden, J.M., Emanuel, E.J., Fairclough, D.L., Link, M.P., Foley, K.M., Clarridge, B.C., Schipper, L.E. and Mayer, R.J. (2001). Attitudes and practices among pediatric oncologists regarding end-of-life care: Results of the 1998 American Society of Clinical Oncology survey. Journal of Clinical Oncology, 19, 205–212. [PubMed: 11134214]
• Hinds, P.S. (1990). Quality of life in children and adolescents with cancer. Seminars in On cology Nursing, 6, 285–291. [PubMed: 2274726]
• Hinds, P.S., Bradshaw, G., Oakes, L., and Pritchard, M. (In press). Children and their rights in life and death situations. In: R. Kastenbaum, editor; , J. Crissman, editor; , and M. Kearl, editor. (eds). Encyclopedia of Death and Dying. New York: Macmillan Reference USA.
• Hinds, P.S., Oakes, L., and Furman, W. (2001. a). End of life decision making in pediatric oncology. In: B. Ferrell, editor; and N. Coyle, editor. (eds.), Textbook of Palliative Nursing ( pp. 450– 460 ). New York: Oxford University Press.
• Hinds, P.S., Oakes, L., Furman, W., Quargnenti, A., Olson, M.S., Foppiano, P., and Srivastava, D.K. (2001. b). End-of-life decision making by adolescents, parents, and healthcare providers in pediatric oncology. Cancer Nursing, 24 , 122–136. [PubMed: 11318260]
• Hinds, P.S., Oakes, L., Quargnenti, A., Furman, W., Sandlund, J.T., Bowman, L., Olson, M.S., Heideman, R. (1998). Challenges and issues in conducting descriptive decision-making studies in pediatric oncology: A tale of two studies. Journal of Pediatric Oncol ogy Nursing, 15(3 Suppl 1), 10–7. [PubMed: 9700237]
• Janicke, D.M., Finney, J.W., and Riley, A.W. (2001). Children's health care use: A prospective investigation of factors related to care-seeking. Medical Care, 39, 990–1001. [PubMed: 11502956]
• Jordhoy, M.S., Kaasa, S., Fayers, P., Ovreness, T., Underland, G., and Ahlner-Elmqvist, M. (1999). Challenges in palliative care research; recruitment, retention and compliance: Experience from a randomized controlled trial. Palliative Medicine, 13, 299–310. [PubMed: 10659099]
• Juniper, E.F., Guyatt, G.H., Feeny, D.H., Ferrie, P.J., Griffith, L.E., and Townsand, M. (1996). Measuring quality of life in the parents of children with asthma. Quality of Life Research, 5, 27–34. [PubMed: 8901364]
• Langeveld, J.H., Koot, H.M., Loonen, M.C.B., Hazebroek-Kampschreur, A.A.J.M., and Passchier, J. (1996). A quality of life instrument for adolescents with chronic headache. Cephalalgia, 16, 183–196. [PubMed: 8734770]
• Lansky, S.B., List, M.A., Lansky, L.L., Ritter-Sterr, C., and Miller, D.R. (1987). The measurement of performance in childhood cancer patients. Cancer, 60, 1651–1656. [PubMed: 3621134]
• Lantos, J.D., Berger, A.C., and Zucker, A.R. (1991). Do-not-resuscitate orders in a children's hospital. Critical Care Medicine, 21, 52–55. [PubMed: 8420730]
• Lauer, M.E., Mulhern, R.K., Schell, M.J., et al. (1989). Long-term followup of parental adjustment following a child's death at home or in the hospital. Cancer, 63, 988–994. [PubMed: 2914304]
• Lauer, M.E., Mulhern, R.K., Wallskog, J.M., et al. (1983). A comparison of parental adaptation following a child's death at home or in the hospital. Pediatrics, 71, 107–112. [PubMed: 6848957]
• Levi, R. and Drotar, D. (1998). Critical issues and needs in health-related quality of life assessment of children and adolescents with chronic health conditions. In: D. Drotar, editor. (ed.), Measuring Health-Related Quality of Life in Children and Adolescents (pp. 3–24). Mahwah, NJ: Erlbaum.
• Loehlin, J.C. (1998). Latent Variable Models: An Introduction to Factor, Path, and Structural Analysis . Mahwah, NJ: Erlbaum.
• Martinson, I.M., McClowry S.G., Davies B., and Kuhlenkamp E.J. (1994). Changes over time: A study of family bereavement following childhood cancer. J Palliative Care, 10(1), 19–25. [PubMed: 8035252]
• McCorkle, R. and Young, K. (1978). Development of symptom distress scale. Cancer Nurs ing, 1:373–378. [PubMed: 250445]
• McCubbin, H.I., Olson, D.H., Lavee, Y., and Patterson, J.J. (1985). The family paradigm album. St. Paul: University of Minnesota.
• Mulhern, R.K., Lauer, M.E., and Hoffman, R.G. (1993). Death of a child at home or in the hospital: Subsequent psychological adjustment of the family. Pediatrics, 71, 743–747. [PubMed: 6835757]
• Paci, E., Miccinesi, G., Toscani, F., et al. (2001). Quality of life assessment and outcome of palliative care. Journal of Pain and Symptom Management, 21, 179–188. [PubMed: 11239736]
• Parsons, S.K. and Brown, A.P. (1998). Evaluation of quality of life of childhood cancer survivors: A methodological conundrum. Medical and Pediatric Oncology Supplement, 1, 46–53. [PubMed: 9659946]
• Patrick, D.L. and Deyo, R.A. (1989). Generic and disease-specific measures in assessing health status and quality of life. Medical Care, 27, S217–S233. [PubMed: 2646490]
• Pratheepawanit, N., Salek, M.S., and Finlay, I.G. (1999) The applicability of quality of life assessment in palliative care: Comparing two quality of life measures. Palliative Medi cine, 13, 325–334. [PubMed: 10659101]
• Richards, M. and Ramirez, A. (1997). Quality of life: The main outcome measure of palliative care. Palliative Medicine, 11, 89–92. [PubMed: 9156103]
• Rosenbaum, P., Cadman, D., and Kirpalani, H. (1990). Pediatrics: Assessing quality of life. In: B. Spilker, editor. (ed.), Quality of Life Assessments in Clinical Trials . ( pp. 205–215 ). New York: Raven Press.
• Rubenstein, L.V., Calkins, D.R., Young, R.T., et al. (1989). Improving patient function: A randomized trial of functional disability screening. Annals of Internal Medicine, 111, 836–842. [PubMed: 2683917]
• Rubenstein, L.V., McCoy, J.M., Cope, D.W., et al. (1995). Improving patient quality of life with feedback to physicians about functional status. Journal of General Internal Medi cine, 10, 607–614. [PubMed: 8583263]
• Rushton, C. (1994). Moral decision-making by parents of infants who have life-threatening congenital disorders . PhD Nursing, Catholic University of America, School of Nursing.
• Salmon, P., Manzi, F., and Valori, R.M. (1996). Measuring the meaning of life for patients with incurable cancer: The life evaluation questionnaire. European Journal of Cancer, 32A(5), 755–760. [PubMed: 9081350]
• Schipper, H., Clinch, J., and Powell, V. (1990). Definitions and conceptual issues. In: B. Spilker, editor. (ed.), Quality of Life Assessment in Clinical Trials . ( pp. 11–24 ). New York: Raven Press.
• Scott, D.L. and Garrood, T. (2000). Quality of life measures: Use and abuse. In: Bailliere's Clinical Rheumatology ( pp. 663–687 ). Harcourt Publishers Ltd. [PubMed: 11092795]
• Seid, M., Varni, J.W., Bermudez, L.O., et al. (2001). Parents' perceptions of primary care: Measuring parents' experiences of pediatric primary care quality. Pediatrics, 108, 264– 270. [PubMed: 11483786]
• Shumaker, S.A., Anderson, R.T., and Czajkowski, S.M. (1990). Psychological tests and scales. In: B. Spilker, editor. (ed.), Quality of Life Assessments in Clinical Trials. New York: Raven Press.
• Smith, K.W., Avis, N.E., and Assmann, S.F. (1999). Distinguishing between quality of life and health status in quality of life research: A meta-analysis. Quality of Life Research, 8, 447–459. [PubMed: 10474286]
• Sprangers, M.A.G., Cull, A., Bjordal, K., Groenvold, M., and Aaronson. N.K. (1993). The European Organization for Research and Treatment of Cancer approach to quality of life assessment: Guidelines for developing questionnaire modules. Quality of Life Re search, 2, 287–295. [PubMed: 8220363]
• Stein, R.E.K. and Reissman, C.K. (1980). The development of an Impact-on-Family scale: Preliminary findings. Medical Care 18:465–472. [PubMed: 7401703]
• Steinhauser, K.E., Christakis, N.A., Clipp, E.C., McNeilly, M., Grambow, S., Parker, J., and Tulsky, J.A. (2001). Preparing for the end of life: Preferences of patients, families, physicians, and other care providers. Journal of Pain and Symptom Management, 22(3), 727– 737. [PubMed: 11532586]
• Stephens, R.J., Hopwood, P., Girling, D.J., and Machin, D. (1997). Randomized trials with quality of life endpoints: Are doctors' ratings of patients' physical symptoms interchangeable with patients' self-ratings? Quality of Life Research, 6(3), 225–236. [PubMed: 9226980]
• Sterkenburg, C.A., King, B., and Woodward, C.A. (1996). A reliability and validity study of the McMaster Quality of Life Scale (MQLS) for a palliative population. Journal of Palliative Care, 12, 18–25. [PubMed: 8857243]
• Stewart, A.I., Teno, J., Patrick, D.L., and Lynn, J. (1999). The concept of quality of life of dying persons in the context of health care. Journal of Pain and Symptom Management, 17, 93–108. [PubMed: 10069149]
• Stromgren, A.S., Groenvold, M., Sorensen, A., and Andersen, L. (2001). Symptom recognition in advanced cancer. A comparison of nursing records against patient self-rating. Acta Anaesthesiol Scand, 45(9),1080–1085. [PubMed: 11683656]
• Stuber, J.L. and Bursch, B. (2000). Psychiatric care of the terminally ill child. In: H.M. Chochinov, editor; and W. Breitbart, editor. (eds.), Handbook of Psychiatry in Palliative Medicine. Oxford: Oxford Press.
• Teno, J., Byock, I., and Field, M. (1999). Research agenda for developing measures to examine quality of care and quality of life of patients diagnosed with life-limiting illness. Journal of Pain and Symptom Management, 17, 75–82. [PubMed: 10069147]
• Testa, M.A. and Lenderking, W.R. (1995). Quality of life considerations in AIDS clinical trials. In: D.M. Finkelstein, editor; and D.A. Schoenfeld, editor. (eds.), AIDS Clinical Trials ( pp. 150– 159 ). New York: Wiley-Liss.
• Thompson K.L. and Varni, J.W. (1986). A developmental cognitive–biobehavioral approach to pediatric pain assessment. Pain, 25, 282–296. [PubMed: 3748588]
• Varni, J.W., Burwinkle, T., Katz, E.R., Meeske, K., and Dickinson, P. (2002. a). The PedsQL in pediatric cancer: Reliability and validity of the Pediatric Quality of Life Inventory Generic Core Scales, Multidimensional Fatigue Scale, and Cancer Module. Cancer , 94, 2090–2106. [PubMed: 11932914]
• Varni, J.W., Katz, E.R., Colegrove, R., and Dolgin, M. (1995). Adjustment of children with newly diagnosed cancer: Cross-informant variance. Journal of Psychosocial Oncology, 13, 23–38.
• Varni, J.W., Katz, E.R., Seid, M., Quiggins, D.J.L., Friedman-Bender, A., and Castro, C.M. (1998). The Pediatric Cancer Quality of Life Inventory (PCQL): I. Instrument development, descriptive statistics, and cross-informant variance. Journal of Behavioral Medi cine, 21, 179–204. [PubMed: 9591169]
• Varni, J.W., Seid, M., Knight, T.S., Uzark, K., and Szer, I.S. (2002. b). The PedsQL 4.0 Generic Core Scales: Sensitivity, responsiveness, and impact on clinical decision-making. Journal of Behavioral Medicine 25(2), 175–193. [PubMed: 11977437]
• Varni, J.W., Seid, M., and Kurtin, P.S. (1999. a). Pediatric health-related quality of life measurement technology: A guide for health care decision makers. Journal of Clinical Out comes Management, 6, 33–40.
• Varni, J.W., Seid, M., and Kurtin, P.S. (2001). The PedsQL 4.0: Reliability and validity of the Pediatric Quality of Life Inventory Version 4.0 Generic Core Scales in healthy and patient populations. Medical Care, 39, 800–812. [PubMed: 11468499]
• Varni, J.W., Seid, M., and Rode, C.A. (1999. b). The PedsQL: Measurement model for the Pediatric Quality of Life Inventory. Medical Care, 37, 126–139. [PubMed: 10024117]
• Varni, J.W., Seid, M., Smith, T., Burwinkle, T., Brown, J., and Szer, I.S. (2002. c). The PedsQL in pediatric rheumatology: Reliability, validity, and responsiveness of the Pediatric Quality of Life Inventory Generic Core Scales and Rheumatology Module. Arthritis and Rheumatism 46(3), 714–725. [PubMed: 11920407]
• Varni, J.W. and Setoguchi, Y. (1992). Screening for behavioral and emotional problems in children and adolescents with congenital or acquired limb deficiencies. American Jour nal of Diseases of Children, 146, 103–107. [PubMed: 1531284]
• Varni, J.W., Waldron, S.A., Gragg, R.A., Rapoff, M.A., Bernstein, B.H., Lindsley, C.B., and Newcomb, M.D. (1996). Development of the Waldron/Varni Pediatric Pain Coping Inventory. Pain, 67, 141–150. [PubMed: 8895242]
• Vivier, P.M., Bernier, J.A., Starfield, B. (1994). Current approaches to measuring health outcomes in pediatric research. Current Opinion in Pediatrics, 6(5), 530–537. [PubMed: 7820199]
• Weitzner, M.A., McMillan, S.C., and Jacobsen, P.B. (1999). Family caregiver quality of life: Differences between curative and palliative cancer treatment settings. Journal of Pain and Symptom Management, 17, 418–428. [PubMed: 10388247]
• Wetzler, H.P. (2000). The future of health status assessment. American Journal of Managed Care, 6, 121–123. [PubMed: 11009743]
• Wolfe, J., Grier, H.E., Klar, N., et al. (2000). Symptoms and suffering at the end of life in children with cancer. New England Journal of Medicine, 342, 326–333. [PubMed: 10655532]
• World Health Organization. (1998). Cancer Pain Relief and Pallative Care in Children . Geneva: World Health Organization.
• World Health Organization. (1948). Constitution of the World Health Organization Basic Document . Geneva: World Health Organization.
• Zwinderman, A.H. (1992). Statistical analysis of longitudinal quality of life data with missing measurements. Quality of Life Research, 1, 219–224. [PubMed: 1301130]

Footnotes

*

Robert C. Byrd Health Sciences Center, West Virginia University.

†

Children's Hospital and Health Center, San Diego, University of California, San Diego School of Medicine.

‡

St. Jude Children's Research Hospital.