End-of-Life Care: Special Issues in Pediatric Oncology
Joanne M.Hilden, M.D.
The Cleveland Clinic Foundation
Bruce P.Himelstein, M.D.
University of Pennsylvania School of Medicine
The Children’s Hospital of Philadelphia
David R.Freyer, D.O.
DeVos Children’s Hospital, Grand Rapids, MI
Sarah Friebert, M.D.
Case Western Reserve University
St. Vincent’s Mercy Children’s Hospital
Hospice of the Western Reserve, Cleveland, Ohio
Javier R.Kane, M.D.
University of Texas Health Science Center
Christus Santa Rosa Children’s Hospital
Christus Santa Rosa Hospice
OVERVIEW: CHALLENGES UNIQUE TO THE PRACTICE OF PEDIATRIC PALLIATIVE CARE
Despite remarkable progress in the treatment of pediatric malignancy, 30 percent of children with cancer still die of their disease or its complications (Pizzo and Poplack, 1997). Cancer is the most common cause of nontraumatic death in children; 2,200 children die each year from cancer in this country (out of a total of 30,000 pediatric deaths annually). Although this is far fewer than the half-million adults who die, the premature death of a child is a unique tragedy.
This report examines the end-of-life care problems unique to pediatrics and suggests steps that could alleviate them. The solutions center on improved models of care and reimbursement structures, which represents the way toward “informed, shared decisionmaking” that can be accomplished only through time-consuming, detailed conferences. In the case of pediat-
rics, this is multilayered, involving the health care team, the parents, and the child patient. Models of care and reimbursement structures should value and recognize this.
Palliative care for children involves all-inclusive and compassionate care aimed at preventing and relieving suffering for those with life-threatening illness. Pediatric palliative care is family-centered care, with the child and family enwrapped in the center of a circle of professionals addressing spiritual, social, psychological, and physical needs. The prevalence of children living with active palliative care needs at a given time is estimated at 50 per 100,000 (Goldman and Christie, 1993).
Available resources designed for the care of adults with life-threatening illness do not fit the needs of dying children. Also, despite recent increases in interest in adult palliative care and hospice philosophy, a parallel increase in pediatrics has not occurred—80 percent of children dying with cancer in this country are still suffering, and their symptoms are not being adequately palliated (Wolfe, 2000).
Why are children with terminal malignancy suffering? First and fore-most, death in childhood is rare. As a result, medical, psychological, social, spiritual, and other practitioners for children are not likely to have much experience in palliative and terminal care. Then too, professionals providing quality end-of-life care to adults, including hospice staff, are not likely to have sufficient training to handle the complex physical, emotional, and psychological care of dying children and their families.
Children are not just small adults. The malignancies that afflict children differ substantially from the common adult cancers, and expectations of cure are much higher. These expectations for both families and treating professionals are realistically based on relatively better overall outcomes for children compared to adults with cancer (even for similar cancers). Heightened expectation of success leads to a reluctance of parents and health care providers to make a formal transition to non-cure-directed interventions.
Dying children defy the natural order, and pediatric providers are more likely to suffer a sense of failure when children die. Referral to an end-of-life program may be seen as abandoning hope, which may interfere with good communication and clinical care. Families and health care providers alike vary tremendously in their state of readiness for transition to an exclusively palliative approach in treating children, even when the definition of palliative care is well established and understood (Frager, 1996). Discussion of palliative care tends to be deferred, and an artificial distinction between curative and palliative care—when there should be continuity of care—is often made.
Communication across differing chronological ages, developmental levels, and decisionmaking capacities is a complex skill set to acquire and demanding to maintain. Children “have a right to be treated as developing
persons, as persons with a developing capacity for rationality, autonomy, and participation in health-care decision making” (AAP, 1995). At any given age, however, they may possess none, some, or all of the capacities necessary to participate in their own health care.
Children are extremely resilient and may rebound from multiple medical crises that would ordinarily be life-ending in an adult. Further, the ability of clinicians to predict timing of death is notoriously poor in the adult population and even worse when it comes to children, particularly those living with chronic illness for many years prior to death. Families and provider teams may be faced with waxing and waning palliative care needs and recurrent conversations over time about the transition to palliative care; this type of need does not fit neatly into the medical, psychological, spiritual, and economic framework established for adult end-of-life care.
Despite current practices to the contrary, extrapolation of adult-derived pharmacokinetic and pharmacodynamic data is often inappropriate and sometimes dangerous for children. Although recent government regulations may change the licensing requirements for new drugs to require pediatric labeling and indications, the rarity of death in childhood still mandates large and often cumbersome multi-institutional trials of symptom control measures for dying children.
Discussions and decisions surrounding end-of-life care have not consistently included the family and the child. In pediatric palliative care, only the individual child and family can determine what is best for them, based on their particular values and life experiences (Liben, 1996). Children need to participate in such discussions and decisions to the fullest extent possible, in order to achieve mastery and control over their own dying. Children have grief work to do and goodbyes to say, just as adults. Inadequate professional training in the ethical, moral, and legal implications of including children in their own care has the potential to rob children of their autonomy and to violate the concept of truth telling in medicine (Bartholome, 1993).
Finally, the death of a child is one of the most significant psychological stressors a person may ever face. The bereavement literature supports the notion that the risk of prolonged, complicated grief or pathological bereavement is substantial for the parents of a child who has died.
This chapter explores the issues in eight major areas:
education of providers,
education of children and families about the dying process,
special issues in communication: adolescents and assent,
delays in the initiation of palliative care for children,
fragmentation of palliative care services,
inadequate relief of pain and other physical symptoms,
research issues, and
reimbursement and regulatory issues.
EDUCATION OF PROVIDERS
Defining the Problem
Improving the quality of care and quality of life for dying children depends on improving the quality of education in pediatric palliative care. However, there are some basic impediments to teaching about death in childhood, including prognostic uncertainty, the move of pediatric residents to more outpatient experiences, and most importantly, the relative rarity of death in childhood (resulting in less provider experience and fewer opportunities for mentoring trainees). Some of these barriers, however, may represent educational opportunities (Sahler et al., 2000).
Training programs in the health professions have begun to pay increased attention to end-of-life issues, but the focus is on adults, with little content applicable to pediatrics. Examples include the American Medical Association’s Education for Physicians on End-of-Life Care (EPEC) curriculum (Emanuel et al., 1999) and the American Academy of Hospice and Palliative Care Medicine’s (AAHPM’s) “UNIPAC” self-study program for physicians who care for terminally ill patients and their families (AAHPM, 1998). Philosophically, adult and pediatric end-of-life care have much in common, but practical applications are clearly different for adults and children. There is currently no comprehensive end-of-life curriculum for pediatric palliative care, although a UNIPAC module for pediatrics is in preparation.
The lack of curriculum content specific to pediatric palliative care exists at all levels and across health care professions. No national standards for curriculum content in pediatric end-of-life care exist for schools of medicine, nursing, or social work, although more medical schools are including some aspects in pediatric clerkships. The Residency Review Committee “Program Requirements for Residency Education in Pediatrics” does not contain any specific language referring to palliative care or end-of-life care. The End-of-Life Nursing Education Consortium (ELNEC) Project from the American Association of Colleges of Nursing also lacks specific pediatric language, although nursing management courses including those specific to pediatric practice would fall within its curriculum guidelines (Ferrell et al., 2000). Recent revisions to accreditation standards from the Liaison Committee on Medical Education (LCME) include a standard on end-of-life care, but it is very broad in its scope, stating that “clinical
instruction should cover all organ systems, and must include the important aspects of preventive, acute, chronic, continuing, rehabilitative, and end-of-life care” (Accreditation Standards, 2000).
Although several academic centers have begun fellowship training programs in palliative care, there are no programs for pediatric palliative care. Reflecting this deficiency, the content outline for the AAHPM certification examination in hospice and palliative medicine lists children only as a special population under the subject heading “Death and Dying.”
Hospice and home nurse agencies care almost exclusively for adults, but in the absence of special services for children, by default, dying children are cared for by them as well. These providers need education to prepare them for children’s care or, at a minimum, ready access to consultation with experts in pediatric palliation. However, there are only a few programs that educate hospice providers in the unique aspects of caring for dying children (Brenner, 1993).
Although the regulatory language for hospice practice (from the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) does not specifically exclude children from consideration, there is a dearth of detailed information for programs that serve dying children (Accreditation Standards, 2000). A tool to address this problem is the Compendium for Pediatric Palliative Care, developed by the Children’s International Project on Palliative Care/Hospice Services (ChIPPS) and currently under review (Marcia Levetown, M.D.; personal communication, 2000). Once complete, it will be published by the National Hospice and Palliative Care Organization. The stated goal of the compendium is to “provide information that would enable a hospice with no pediatric experience to care for a child.” The effectiveness of this tool must be studied once it is released.
The American Society of Clinical Oncology surveyed adult and pediatric oncologists in 1999 regarding palliative and end-of-life care issues. Only 10 percent of pediatric oncologists who responded reported that they had formal courses in pediatric terminal care in medical school, and only 2.2 percent reported a rotation in a palliative care or hospice service. The most common method of learning about these topics reported by pediatric oncologists was “trial and error,” and many reported anxiety about having to work with dying children. These practitioners not only are treating children, but are the role models for future generations of both generalists and specialists, who often look to oncologists for expertise in end-of-life care.
Pediatricians and pediatric issues are underrepresented in national organizations and committees dealing with medical care and reimbursement. For example, although organizations such as the National Hospice and Palliative Care Organization-National Council of Hospice Professionals have some pediatric representatives on subcommittees, no subcommittee
addresses the unique educational, fiscal, clinical, regulatory, philosophical, and ethical needs of a pediatric hospice population. Similarly, although organizations dedicated to pediatric care have work groups devoted to end-of-life care (e.g., the Children’s Oncology Group [COG], the American Academy of Pediatrics [AAP], and Children’s Hospice International) and continue to advocate for the needs of dying children, unifying, collaborative national efforts to bring provider education to the forefront do not exist.
Finally, there are only a handful of quality textbooks targeted to pediatric palliative care, and the subject is underrepresented in the classic adult textbooks such as Supportive Care in Oncology (Weisman, 1998) or the Oxford Textbook of Palliative Medicine (Doyle et al., 1998). Examples of essential pediatric texts include Hospice Care for Children (Armstrong-Dailey and Goltzer, 1993), Care of the Dying Child (Goldman, 1994), and Cancer Pain Relief and Palliative Care in Children (WHO, 1998) as well as two explorations of the more spiritual aspects of childhood death, The Private Lives of Dying Children (Bluebond-Langer, 1978) and Armfuls of Time (Sourkes, 1995).
Develop content for pediatric end-of-life care curricula in medical, nursing, chaplaincy, and social work training programs. The challenges facing creators of curricula include defining educational objectives; outlining the content of training; selecting teaching methods; exploring personal attitudes toward death, dying, and bereavement; promoting interdisciplinary collaboration; evaluating training; and defining the role and function of educators in pediatric palliative care (see Papadatou, 1997, for a discussion of challenges in creating a pediatric palliative care course).
Develop curricula with both traditional and alternative teaching methods. Standard didactic approaches are the tradition in many post-secondary education programs, but these approaches do not optimally address the emotional and psychological needs of students in a complex field such as pediatric palliative care. Alternative methods such as small group discussion, role playing, experiential learning by partnering with mentors, supervised clinical practice, and/or self-directed on-line learning may better suit training at all levels. Small studies have demonstrated the efficacy of nontraditional learning methods, including those derived from the psychology world, in altering attitudes of students regarding end-of-life issues (Razavi et al., 1988, 1991). At the Children’s Hospital of Philadelphia, a pilot study is under way to explore the role of an intensive, brief cognitive-behavioral intervention for staff in changing values and beliefs about pediatric palliative care.
Evaluate the effectiveness of educational materials and methods in pediatric palliative care. Curricula such as EPEC would be expected to be effective, but this cannot be assumed without appropriate evaluation. New outcome measures may be required to assess the skills of trainees and practitioners, incorporating assessments not only of knowledge about death and dying, but also of empathy, spiritual balance, educational capacity and effectiveness, or even business acumen.
Develop curricula that teach intact medical care teams the tenets of palliative care. Parents of seriously ill children have indicated that they strongly value the continuity of care achieved when the primary oncology team continues to care for their child through the time of death. While this delivers the desired continuity to parents, it has resulted in inadequate delivery of palliative care. Training of intact teams may improve the delivery of care as well as facilitate involvement of the necessary professionals.
Create and fund pediatric palliative care training and fellowship programs. Effective training in pediatric palliative care will depend upon many factors, including exposure to a wide array of clinical materials relevant to the field of study; good mentorship; well-defined evidence-based curricula; and the availability of a suitable academic environment to support the study of related fields such as bioethics, adult palliative care, epidemiology and biostatistics, or pharmacology. “Centers of Excellence” in pediatric palliative care should be created in which training and fellowship programs can offer education to adult hospice workers who care for the occasional dying child.
Add appropriate end-of-life content to general pediatric, pediatric subspecialty, and hospice and palliative medicine certifying examinations. Content on end-of-life care should be added not only to general pediatric board examinations, but also to subspecialty certifications such as intensive care, cardiology, neurology, and neonatology, which along with oncology have the most pediatric deaths. Hospice and palliative care practitioners must have a minimum fund of knowledge in order to provide comprehensive and compassionate care to dying children.
Add language to home health and hospice regulations specifically mandating competencies in pediatric end-of-life care. Providers should have a minimum fund of knowledge regarding medical, physiological, emotional, and developmental issues of the dying child. The pediatric UNIPAC curriculum (when available) could be made mandatory for providers who will care for children, and successful completion of the curriculum could be a criterion for individual and institutional licensure.
Develop national collaborative efforts to advocate for education in pediatric end-of-life care. To meet the educational needs of professionals caring for dying children, pediatric palliative care training sessions should
be offered in conjunction with national meetings of organizations that care for children with life-threatening illness (e.g., COG, AAP, and the American Society of Pediatric Hematology/Oncology). For adult hospice providers, pediatric end-of-life curricula should be offered at national meetings of the relevant professions. Collaboration between programs and interested individuals dedicated to pediatric palliative care (e.g., members of the American Association of Hospice and Palliative Medicine, the National Hospice and Palliative Care Organization, and Children’s Hospice International), as well as funding for these collaborations, must be a national priority.
EDUCATION OF CHILDREN AND FAMILIES ABOUT THE DYING PROCESS
Defining the Problem
Many families who are navigating the health care system during the treatment of their child’s cancer often joke that they should receive honorary medical or nursing licenses. This comment, although somewhat tongue-in-cheek, underscores the complexity of the tasks of children and families facing life-threatening illness. From the time of diagnosis to the time of cure or death, families must assimilate an overwhelming amount of information, function as advocates for their child and themselves, make informed decisions (often without adequate information), and negotiate ever-changing systems for delivery of care (including insurance plans). All this must be accomplished while continuing to work or care for other family members at home. These tasks become increasingly burdensome and difficult when a child’s prognosis is not good, and families must balance quality-of-life issues with their drive and need to “leave no stone unturned” in pursuing treatment options. In this scenario, provision of accurate, up-to-date, and comprehensive information in an understandable manner is even more crucial.
It is at the stage of diagnosis of a life-limiting prognosis that families are faced with seemingly dichotomous treatment options. The availability of Phase I and II clinical trials for pediatric oncology patients offers continued “aggressive” therapy with a small chance of physical or life-prolonging benefit to the patient, and possibly with altruistic benefits. At the same time, palliative care options need to be discussed so that optimal symptom management can preserve patient comfort and dignity. It is this simultaneous provision of potentially curative and palliative medicine that currently escapes us.
If resources for providers of pediatric end-of-life care are lacking, the availability of educational materials for affected children and their families
is even further behind. Some examples of information required by families to make good decisions include understanding the diagnosis and prognosis; the likely effects of the disease on the patient; other relevant physical or emotional problems likely to impact the course of illness; other symptoms likely to occur; what death will look and be like with and without artificial interventions; uses and interactions of medications; the availability of pharmacologic and nonpharmacologic interventions to ease suffering; the availability of professional and nonprofessional resources to aid the family; physical modifications and facilities to make home or transportation more accessible; and what changes in functional status are likely to occur. In short, families need a complete appreciation of the effects of a life-threatening illness on the physical, psychological, spiritual, and practical dimensions of care.
An additional educational need regards advance directives. With adults, there is at least a chance that end-of-life wishes will have been considered before being faced with a life-threatening illness, but this almost never occurs in the pediatric setting. Most pediatricians and even pediatric sub-specialists are not skilled in discussing advance directives. Research clearly demonstrates that patients and families prefer to be guided in these discussions by practitioners they trust (Frager, 1996; Whittam, 1993). Resources to make decisions concerning withdrawal of life-sustaining treatment (including nutrition and hydration), and covering principles of palliative care and issues of medical futility, are not currently available to pediatric patients and their families. Although there are books available to parents describing leukemia and other cancers, there is not much available to families to help prepare them for the medical and psychosocial details of the death of their child.
Develop protocols for use by interdisciplinary teams to explain disease and prognosis in terms that families and patients can understand. Content of discussions should be spelled out and be accompanied by delivery of the material in written form for later review.
Develop materials for child patients at every developmental level and their families with disease-specific information, prognosis, palliative care terminology and options, and clinical trial terminology and options.
Develop resources that detail expected physical changes toward the end of life.
Involve parents and older adolescents (when appropriate) in national organizations developing policy for pediatric hospice and palliative care standards and reimbursement.
ADOLESCENTS AND ASSENT
Defining the Problem
Caring for Adolescents
Adolescents facing death have palliative care needs substantially different from those of younger children or adults. The unique psychosocial issues for dying adolescents—which relate to the normal developmental tasks of this time of life—include greater focus on physical appearance, reversal of developing independence, lack of control, loss of self-confidence, social isolation, disruption of future plans, and desire to be listened to by their care providers (Carr-Gregg et al., 1997). At initiation and during provision of palliative care, communication with adolescents requires particular sensitivity to the concerns characteristic of this age group.
Adolescents with cancer do not have proportionate access to clinical trials sponsored by national pediatric oncology cooperative groups, a possible factor in the relatively lower survival rates observed in this age group (Bleyer et al., 1997). It is not clear whether similar differences exist for adolescents in their access to pediatric palliative care services or in their qualitative experience while receiving services.
Issues of Assent and Consent
Children who are developmentally capable of participating in their own health care decisionmaking are often prevented from doing so. Historically, children have been declared legally and ethically incompetent to participate in decisions about their own health care. Except for circumstances involving mature or emancipated minors, decisions regarding health care for children under age 18 generally are made by surrogate decisionmakers, usually parents. However, some health care providers (and most ethicists and palliative care professionals) believe that children who have reached the age of assent and are capable of expressing a preference should be given choices and have their wishes respected. This is especially true in the area of end-of-life care, when quality, not quantity, of life is the main focus. After all, who better can decide what constitutes quality of life for an individual than that person? Leikin (1993) writes:
…if a minor has experienced an illness for some time, understands it and the benefits and burdens of its treatment, has the ability to reason about it, has previously been involved in decision making about it, and has a comprehension of death that recognizes its personal significance and finality, then that person, irrespective of age, is competent to consent to forgoing life-sustaining treatment.
In pediatrics, consent actually amounts to authorization by the parents for treatments and procedures, reflecting the assumption that parents are the most authentic spokespeople for their children. However, most children are capable of consent after age 14, by which time, with normal development, they possess full decisional capacity and flexible thinking (Brock, 1989). “Assent” refers to a child’s agreement with the proposed treatment. Although it is not a term defined in law, assent respects children as individuals with developing capacities for participation in health care decisionmaking. Conversely, “coercion” describes an essentially paternalistic act of forcing participation in treatment or research, which should be avoided.
Assent in pediatric practice consists of four basic elements:
demonstrating respect for the child as a patient and as a developing person by assisting the child to develop an appropriate awareness of illness;
disclosing the nature of the proposed intervention and what the child is likely to experience (truth telling);
assessing the child’s understanding of information and the factors influencing his or her evaluation; and
demonstrating respect for emerging autonomy and the development of decisionmaking capacity by soliciting expressions of willingness on the part of the child to accept the intervention (Bartholome, 1993).
The American Academy of Pediatrics Committee on Bioethics recommends that assent for treatment should be obtained from the pediatric patient when developmentally appropriate and should be binding when used in the research setting (AAP, 1995). However, guidelines from the National Institutes of Health (NIH) Office for Protection from Research Risks (OPRR) state that assent or dissent is conditional on parental permission if participation in research is potentially beneficial to the child, in which case parental permission overrides the child’s dissent (OPRR, 1991).
Legal and ethical debates about the appropriate age of consent for medical treatment or research participation are interesting and important but oversimplify the issues when it comes to caring for adolescent patients. In the first place, decisionmaking capacity itself is not a static phenomenon; it can be intermittent or fluctuating, and it may vary over time with changes in clinical condition. Secondly, people are not static either: like adults, adolescent patients vary significantly in their ability to comprehend what is happening to them. Care providers, therefore, need to be attentive to changing competence in adolescent patients (Friebert and Kodish, 1999).
Develop and promote a structure for communication between clinical staff and family, specifically including the child patient, who should be part of the decisionmaking process whenever possible. The goal is to set up the expectation that a child will be as fully informed as possible, so that when tough decisions come along, the child can participate. An example of this is the “Final Stage Conference” (Nitschke et al., 1997), used at the Children’s Hospital of Oklahoma since the 1970s, in which a consistent approach is employed at the time of a child’s cancer relapse to communicate essential information regarding disease status, prognosis, and care options. The child is routinely included in the discussion (with the parents’ permission), which is tailored to his or her developmental understanding. Available investigational and palliative therapies and expectations for the terminal course are described. In the experience of the authors, the Final Stage Conference has been effective at conveying essential information, enhancing participation of the child and family in reaching a sound decision, facilitating dialogue within the family unit, and maintaining the family’s trust.
Several disease- and treatment-related characteristics of children with cancer are relevant when considering discontinuing active therapy (Freyer, 1992). These include
the medical experience of the child,
the nature of pediatric cancer therapy,
the unpredictability of treatment responses,
parental and/or physician biases, and
the necessity of palliative care.
For children and adolescents capable of expressing their values and preferences, the use of “modified substituted judgment” (substituted judgment is a legal concept for surrogates’ making decisions for previously competent adults) is recommended for enacting decisions consistent with their wishes (Freyer, 1992). This means that parents can apply their child’s stated values when decisions are required. When combined with traditional guidelines for end-of-life decisions (such as benefit-burden analysis), the consistent application of these guidelines for children appears to enhance provider-patient or family communication. Clinical studies are required to confirm this.
Develop standards for decisionmaking capacity (including advance directives) in the pediatric population based on developmental level or
“illness competency.” Pediatric patients should be assessed individually on their desire and ability to participate in decisionmaking, regardless of chronological age, according to a set of standards that have been validated with input from pediatricians, ethicists, legal counsel, and developmental psychologists.
DELAYS IN THE INITIATION OF PALLIATIVE CARE FOR CHILDREN
Defining the Problem
During the care of a child whose cancer becomes refractory to therapy, there comes a point when it is appropriate to initiate palliative care. This transition point is somewhat arbitrary and lacks a universally accepted definition, but it can be considered as the time at which the goals of palliative care become more important than other treatment end points. Defining the exact time is difficult, however, because the transition from anticancer to palliative therapy is gradual for most children, and palliative care may be appropriate very early, in conjunction with potentially curative treatment.
Delaying the initiation of palliative care results in (1) losing the opportunity to promote palliative care principles to the patient and family; (2) being less able to tailor palliative care to the evolving needs of the patient; (3) crisis-oriented management, which exacerbates the sense of vulnerability and helplessness; (4) absence of a framework for preventive, proactive interventions or decisionmaking; and (5) difficulty in supporting the family’s strengths and capacity to cope and in the maximizing quality of the remaining time (Frager, 1996; Goldman, 1996; Vickers and Carlisle, 2000). This is consistent with a recent study in which terminally ill children with cancer were more likely to be described by their parents as peaceful and calm during their last month of life if their hospice care decision occurred earlier in the course of their illness (Wolfe et al., 2000).
Several barriers can prevent timely initiation of palliative care. First, the transition from anticancer therapy to palliation is usually gradual, making the decision point ambiguous for starting palliative care. With each relapse, the prognosis for cure decreases. Although the use of second-line (retrieval) and investigational therapy is often available and reasonable, the need for control of physical and psychological symptoms increases with time as the patient’s condition deteriorates.
Second, the traditional framework for making the transition to palliative care deals poorly with this reality, such that health care decisions involve exclusive, all-or-none use of either anticancer therapy or palliative care, separated by a clean break from one to the next at some discrete point in time (Frager, 1996). If families and providers are forced to choose be-
tween investigational therapy and palliative care, the result is often delayed implementation of the latter. The study by Wolfe and colleagues (2000) referred to earlier did not assess whether palliative care had been delayed for the 103 children included but did report anticancer treatment very late in the course of disease: Phase I and II investigational drug trials had been received by 24 percent and 38 percent of the patients, respectively, and half of the children had anticancer treatment during the last month of life, which consisted of a bone marrow transplant for 22 percent.
A third barrier to initiating palliative care for some patients is that appropriate services are not available, particularly home-based palliative care. This type of care can be delivered in the inpatient setting (and may be preferred for some patients), but the family’s home is the preferred location for most dying children (Collins et al., 1998; Goldman, 1996). Several successful models of home-based palliative care for children have been described (Martinson, 1993a), yet a recent survey by Children’s Hospice International (1998) indicates relatively few organized pediatric palliative care services in operation. Adult hospice providers are not experienced with children; parents or providers are understandably reluctant to transition care to such providers. Consequently, it usually falls to the treatment team at the pediatric oncology center to coordinate palliative care. An advantage of this is the continuity through established relationships, but a potential disadvantage is the tendency to delay initiation of palliative care, in part because most pediatric oncologists lack formal training in end-of-life care (Hilden et al., 2001). (See also the the sections on reimbursement and education.)
Fourth, the needs and beliefs of parents may be responsible for delayed implementation of palliative care for children (Children’s Hospice International, 1998; Nitschke et al., 2000; Whittam, 1993). Compared with adults, death in children is considered unnatural and especially tragic. It is difficult for most parents not to equate stopping cancer treatment with “giving up” on their child, resulting in continued treatment beyond significant hope of cure or recovery. Nor are physicians invulnerable to the difficulties in refocusing the therapeutic goal from cure to comfort care (Whittam, 1993). Leading obstacles to providing hospice services to children were an association of the hospice concept with death rather than life enhancement, lack of clarity regarding when to refer, and physician reluctance in making the referral (Children’s Hospice International, 1998). Additionally, cross-cultural beliefs and practices may influence attitudes toward death and discourage the use of palliative care for some dying children (Die Trill and Kovalcik, 1997; Sagara and Pickett, 1998). In any case, the need for parents to choose aggressive curative therapy for their child is real, not “unrealistic.” Our systems of care must accommodate this reality, rather than trying to change parents.
Fifth, the lack of parental awareness and accurate knowledge about palliative care may prevent its early initiation. There is little evidence regarding parental knowledge of optimal care standards for dying children, but in the Children’s Hospice International (1998) survey, parents’ lack of familiarity with hospice services was rated as an important obstacle to referral. The education and training of pediatric physicians is acknowledged to be deficient in the skillful provision of palliative care (Khaneya and Milrod, 1998; Sahler et al., 2000). It seems reasonable to assume that this deficiency would result in parents’ lack of awareness of the option.
Finally, cost may deter early implementation of home-based palliative care. Even though it appears substantially less expensive to care for dying children at home than in a hospital (Martinson, 1993b), the cost of home care may be a barrier for some patients, especially those whose care is publicly funded (Children’s Hospice International, 1998; Schweitzer et al., 1993). In addition, “hidden” costs to the family of a child dying at home, in terms of lost wages and out-of-pocket expenses for nonmedical supplies, may be a barrier. Furthermore, some insurance programs do not cover the concurrent provision of anticancer treatments and palliative care (Whittam, 1993).
Disseminate to health care professionals clear, widely accepted clinical criteria for determining when palliative care should be initiated for a child with cancer. The physician and nursing groups from COG, APON (Association of Pediatric Oncology Nurses), and ASPHO (American Society of Pediatric Hematology/Oncology) should collaborate on the development of these criteria.
Fund research to determine referral patterns for palliative care services. If we are to test the hypothesis that earlier referral to palliative care services will improve quality of life for dying children, baseline data are needed, including variation according to geographical area and third-party payer status.
Develop a new model for the transition to palliative care, permitting a gradual blending of anticancer and palliative therapies, with the latter becoming more dominant as the child’s course proceeds (Frager, 1996).
Increase the availability of satisfactory home palliative care services for children, recognizing that multiple models have been used successfully and that there is no single model that is best for all pediatric oncology centers, communities, or regions. Rural areas are in particular need of improved access to pediatric home palliative care services. In such areas, where the maintenance of a centralized, traditional palliative care program is not feasible, an effective model has utilized trained, community-based
home care providers working in close collaboration with a coordinating palliative care service for children (Martinson, 1993a). Large-scale development of this approach requires mechanisms to ensure appropriate training and funding for the direct care providers, who will be experienced with adult rather than pediatric patients.
Develop formal training in palliative care for oncology teams. This issue is covered in a separate section on provider education. Avoiding delay in referral for palliative care should be included in the educational content of the training programs.
Organize efforts to improve coverage of pediatric palliative care services, especially those provided in the home, where cost-effectiveness has been demonstrated (Martinson, 1993b). This issue is covered in a separate section on reimbursement.
FRAGMENTATION OF PALLIATIVE CARE SERVICES
Defining the Problem
Delivery of palliative care to children routinely involves multiple professionals from various disciplines and spans two months or longer (Children’s Hospice International, 1998; Collins et al., 1998; Wolfe et al., 2000). Ideally, the services should be coordinated and seamless, but in practice, lack of coordination leads to fragmentation and poorly timed delivery. The magnitude of this problem is unclear from the literature, but its importance is implied by palliative care guidelines from SIOP (International Society of Paediatric Oncology) (Masera et al., 1999) and the International Work Group on Death, Dying, and Bereavement (1993), which explicitly state the need for a well-coordinated system of care for these patients.
Recent studies indicate that fragmentation of care may contribute to the distress of dying children and their families. In a recent study, parents of children with cancer who had died at home suggested better coordination of care as a way of improving palliative care services (Collins et al., 1998). In the same study, some families reported difficulty in arranging for readmission of children to the hospital for control of symptoms. Conflicting information from caregivers constitutes another form of fragmented care and has been associated with increased pain and non-pain-related suffering in children who died of cancer; the same was true for lack of involvement by the primary oncologist (Wolfe et al., 2000). Similarly, parents of children dying of cancer express a strong need to feel cared for and connected to their treatment team but often experience feelings of abandonment as death nears (James and Johnson, 1997).
Regardless of where care is delivered, the lack of a designated coordinating entity contributes most to fragmented care. In its absence, few prac-
ticing physicians can fill this gap (Hilden et al., 2001; Khaneya and Milrod, 1998; Sahler et al., 2000). In current medical care structures, providers are tied to particular sites and services (i.e., cardiology or intensive care unit). A family will see different chaplains, social workers, and providers at different sites and on different shifts, and often loses contact with these individuals upon discharge. Even when palliative care or pain teams are involved, their communication over time and sites is discontinuous. Furthermore, in rural regions the wide geographic distribution of dying children makes it untenable for them to be aided by palliative care programs based in metropolitan centers.
Unfortunately, fragmentation also occurs where established palliative care services are available. Although it has not been studied formally, one reason seems to be undefined channels of communication between the family and care providers. In most instances, direct care problems (e.g., development of new symptoms) should be addressed to a home care team. However, families may communicate these problems to personnel more familiar to them at the treatment center. In these situations, fragmentation can result from a failure of one health care team to give prompt notification of the clinical problem to the other. Although less common nowadays, the inability to provide certain interventions in the home setting (e.g., local blood bank policies proscribing transfusions outside a licensed hospital) also can result in fragmentation of care and the need to travel from home to the medical facility. Poor working relationships or communication channels between in-home care providers and the hospital may result in difficulty gaining readmission for management of difficult symptoms during the final stage.
There are some models in place or in the planning stages to remedy these problems. At Boston Children’s Hospital the Pediatric Advanced Care Team (PACT) coordinates the essential elements of end-of-life care across the continuum of inpatient, outpatient, and home care settings for children with limited life expectancy. The goals of PACT are to improve family and caregiver communication, lessen pain and suffering, and emphasize meaningfulness during the end-of-life period. Interventions have focused on four main areas: (1) patient care, (2) education, (3) bereavement, and (4) outreach. As of 2000, the program had consulted on 80 patients, and the experience so far suggests that caregivers and families value the service. In a survey of providers following each consult, all physicians and nurses and 93 percent of psychosocial clinicians found the consults helpful (Wolfe, 2000).
The Pediatric Palliative Care Project at Children’s Hospital and Regional Medical Center in Seattle, Washington, is another model, begun in 1998 and funded by the Robert Wood Johnson Foundation (http://www.seattlechildrens.org/pedpalcare/). It is evaluating the use of symptom control algorithms, a decisionmaking and charting tool, and a case man-
ager who works with payers. In the first 18 months of the project, 60 children were referred for consideration and 20 enrolled in the program (the others received consultation or referral to appropriate services). The results of these evaluations are not yet available, but one initial impression is that at the time of referral, children are often at a point where symptom control is complex, requiring more sophisticated symptom control algorithms (Beth Forbes, personal communication, 2000).
Children’s Hospice International has issued a call for proposals for its “Program of All-Inclusive Care for Children” (PACC), modeled after the adult Program for All-Inclusive Care of the Elderly (PACE), which is being supported by a congressional appropriation. According to the program description (www.chionline.org), “PACC will offer and manage all health care, medical, social services and support services needed by families to care for children diagnosed with life threatening and potentially life-limiting conditions, from the time of diagnosis through end of life care, and support their families, including bereavement care.” The first grant cycle is limited to five states but will expand as more funding is available.
The PACC concept, with the addition of a care coordinator (trained to oversee specific communication content), is being tested with adult patients in the National Advanced Illness Coordinated Care (NAICC) program. Pilot results from the NAICC experience are encouraging. In a Department of Veterans Affairs Medical Center in New York State, NAICC patients had an 85 percent completion rate for advance directives, compared to 22.5 percent for a diagnostically matched control sample, and a 90 percent documentation rate for final stage of disease discussions, compared to 40.7 percent for control patients. Also, average inpatient costs were $1,923 less per patient per year for Advanced Illness Coordinated Care Program (AICCP) patients. In addition, providers at sites around the country are highly satisfied with the NAICC training they have received and report meaningful changes in their end-of-life care practices. Several commercial insurers have agreed to reimburse six NAICC visits per patient at selected sites (Daniel Tobin, M.D., personal communication, 2000). A NAICC model for children is under development.
Develop comprehensive pediatric palliative care services that include a family-oriented, relationship-centered focus. This should include programs to provide continuity of care over transitions that now result in fragmentation. New roles, such as care coordinators, should be developed and evaluated.
Define essential features of a pediatric palliative care service (Goldman, 1996; Martinson, 1993a):
Continuous (24-hours per day, seven days a week) access to care providers able to make regular and unscheduled home visits (this will usually be a registered nurse)
Continuous access to pediatric palliative care experts for management suggestions and continuity
Ability to deliver all reasonable palliative interventions in the home setting without administrative or financial restrictions (e.g., transfusion of blood products, availability of pharmacy and durable pediatric medical supplies)
Respite care for the family
Immediate access to hospital or inpatient hospice facility if needed for symptom control
Bereavement care during and after the death of a child
Develop initiatives to address the special challenges to pediatric hospice and palliative care faced in rural or other underserved areas. At least one model for pediatric home palliative care has been demonstrated to be effective in this setting (Martinson, 1993b). “Home Care for the Child with Cancer,” was a nursing research study initiated in 1976 with funding from the National Cancer Institute to demonstrate the feasibility and benefits of home care for dying children. In that project where approximately 50 percent of the children lived in rural areas, successful use was made of nurses recruited from the patients’ communities to provide direct in-home care in collaboration with palliative care experts from the urban center. This model should be implemented on a larger scale (Martinson, 1993b).
Develop partnerships between palliative care centers and community-based primary care physicians. An evolving role has been described for the primary care physician in pediatric palliative care (Howell, 1993; Wessel, 1998), although its cultivation on a large scale will require systematic education initiatives, as well as the active support of established palliative care centers. There is significant potential for these providers to improve continuity of care.
INADEQUATE RELIEF OF PAIN AND OTHER PHYSICAL SYMPTOMS
Defining the Problem
Despite our best efforts, approximately 30 percent of all children diagnosed with a malignant disease die (Pizzo and Poplack, 1997). Ready access to skilled palliative care is clearly the standard of care for adults dying from malignant disease (Council on Scientific Affairs, 1996), but there is cur-
rently no national standard for access to quality palliative care for dying children.
There is considerable evidence that pain is a common symptom in children with terminal cancer (Wolfe et al., 2000). Much effort has gone into developing resources for providing adequate pain relief (Hain, 1997; Schrechter, 1990; WHO, 1998), but they have not been disseminated adequately. For example, of second-year pediatric residents informally surveyed as they started their oncology rotation at the Children’s Hospital of Philadelphia, fewer than one-quarter were familiar with the World Health Organization pain ladder (Bruce Himelstein, M.D., unpublished observation, 2000).
Pain therapy in childhood is also limited by the lack of pediatric labeling for drugs that might benefit children; for example, newer long-acting opioid preparations. Investigator-initiated studies are often difficult to carry out in the pediatric palliative care setting, because patients may be geographically separated. There are also inherent difficulties in performing such studies of patients who are largely homebound. The lack of clinical trials for children is not limited to palliative care and has been recognized as a general problem. A partial remedy that is in place is a provision of the Food and Drug Administration (FDA) Modernization Act of 1998, which offers incentives to pharmaceutical companies with drugs under patent to obtain pediatric data in exchange for an additional six-month exclusivity. This leaves the problem of appropriate labeling for off-patent drugs, as well as any patented drugs that industry decides not to test.
Symptoms other than pain are actually more troublesome, according to parents and physicians (Wolfe et al., 2000). Fatigue, dyspnea, poor appetite, constipation, nausea, vomiting, and diarrhea are less successfully managed than pain, as reported by parents (but not by physicians). This finding is bolstered by reports by pediatric oncologists, who do not rate their skills in treating non-pain symptoms highly and who are more anxious treating difficult non-pain symptoms (Hilden et al., 2001). Unfortunately, little information is available about the incidence of these symptoms in children with cancer (Hain et al., 1995).
The evidence base for interventions for symptoms such as delirium, cough, dyspnea, somnolence, anxiety, and anorexia in the terminally ill pediatric population is also very poor. Without essential data from well-designed clinical trials, practitioners of pediatric palliative care are left to extrapolate from adult studies or to practice anecdotal medicine. Quality of care in any other medical subspecialty under these conditions would be considered substandard. The situation is even worse for interventions to improve nonphysical symptoms. Much has been written about the psychological, spiritual, and emotional aspects of dying, including several out-
standing studies of dying children and their families (reviewed in Stevens, 1998a, 1998b), but few interventions that might alleviate distress have been tested adequately.
Late referrals to skilled practitioners of end-of-life care may also play a role in the undertreatment of symptoms, but no one has studied the simple epidemiology of time of referral to time of death in pediatrics (although such data are available for adults).
Develop care models that integrate palliative care specialists, symptom control specialists, and psychosocial services into the mainstream of pediatric oncology care. Team care has been shown in the adult hospice literature to decrease pain and symptom severity, as well as to improve cost-effectiveness (Hearn and Higginson, 1998; Mercadante, 1999). This will involve the education of physicians simply in terms of the prompt and appropriate use of pain specialists.
Research to assess the efficacy of care models. Pediatric oncology clinical trials units, in particular the recently formed Children’s Oncology Group (a merger of the Children’s Cancer Group, Pediatric Oncology Group, National Wilms’ Tumor Study Group, and Intergroup Rhabdomyosarcoma Study Group), have a central responsibility to improve the quality of life of children with cancer. Trials of novel models of care to improve quality of life for children with terminal malignancy are critical.
Research to assess the efficacy of established and innovative symptom control interventions. The pediatric oncology trials groups (COG, and/ or the Pediatric Pharmacology Research Unit funded by the National Institute of Child Health and Human Development) carry out symptom control studies with new agents. Some therapeutic questions may require nationwide trials to enroll enough terminally ill children. Trials should address both physical and nonphysical (e.g., psychological, emotional, spiritual) conditions.
Epidemiological research to determine the incidence and prevalence of symptoms in children with life-threatening illness.
State and federal programs to improve access to palliative care services and appropriate use of symptom control measures. Recent FDA rulings clearly support the desire to obtain pediatric data with new drugs and biologicals. Federal legislation, such as that advocated by Children’s Hospice International, recently passed to support the development of demonstration model programs, is needed to provide quality care for children with life-threatening illness.
REIMBURSEMENT ISSUES IN PEDIATRIC PALLIATIVE CARE
Defining the Problem
The biggest problems related to reimbursement are, first, payment for time spent communicating with parents and children and second, payment for palliative and hospice care for dying children.
Reimbursement for Physician-Family and Patient Communication
Excellent communication between physicians and parents, and between physician and the child patient, is essential to excellent cancer care for all children, but even more so for children who die. This communication is even more time-consuming than it is for adult patients, because of the complexity of communicating with children and the need to communicate complicated information to parents who naturally cling to the hope for cure. It is well established that “cognitive services” are generally poorly reimbursed compared to physical medical interventions, and this phenomenon is exaggerated in pediatrics, not only because of the stated complexity, but also because pediatricians are penalized for the fact that their actual patient—the child—is not always present during long discussions with parents. Moreover, billing codes do not distinguish among the subgroups of seriously ill pediatric patients with chronic disease, cognitive impairment, and/or complex life-threatening illnesses, for whom multiple specialty physicians are involved and for whom advance care planning and coordination of care among multiple physicians and services are labor intensive.
Reimbursement rules are discussed here in the context of dying children, but the principles and difficulties are relevant throughout the course of a child’s cancer, regardless of diagnosis or prognosis. This discussion is based on reports from practitioners who belong to the Children’s Oncology Group Principle Investigators (COG PIs; responders representing 44 out of 125 institutions), polled specifically for this project, in the complete absence of published data on reimbursement rates and amounts either from previous research or from insurers (who view the information as proprietary).
Prolonged Physician Services
Billing codes for “prolonged physician services” would seem to be the appropriate codes for conversations with patients and families pertinent to death and dying. These conversations take considerable time in pediatrics, since they involve working with children at various developmental ages as well as with parents. So for pediatrics, the “talk time” can be at least twice
that of working with an adult patient. In the real world of pediatric medical care, these conversations are separate from time spent managing the child’s complex medical problems.
The prolonged services codes are intended to be used in conjunction with office visit codes or inpatient visit codes (CPT codes 99201–99215, 99241–99245, 99301–99350). These office or inpatient visits are described as appropriately billed for counseling time if the time spent counseling is more than 50 percent of the physician-patient interaction. The total time for these codes goes up to 40 minutes, and this time most often is consumed doing the exam and reviewing charts and test results, so pediatricians should be able to rely on the prolonged service codes to bill additional time for counseling
There are specific codes for prolonged physician service with direct face-to-face contact (99354–99357), which are described as follows (Current Procedural Terminology [CPT], 2000):
Codes 99354–7 are used when a physician provides prolonged service involving direct (face-to-face) patient contact that is beyond the usual in either the inpatient or outpatient setting. This service is reported in addition to other physician service, including evaluation and management service. 99354 or 99456 are used to report a total duration of prolonged service of 30–60 minutes on a given date. 99355 or 99357 are used to report each additional 30 minutes beyond the first hour.
These descriptions suggest that the codes should cover the extended communication necessary in pediatric care. Although the CPT lists these as billable in 30-minute increments with a maximum of 3 hours total, the reality is that they are relatively poorly reimbursed (and often denied outright by payers) relative to surgical procedures or physical exams, even though they represent a large percentage of time spent by physicians in caring for children with advanced illness. The COG PIs reported as follows: 40 percent do not even bill these codes because experience has shown they will not be reimbursed; 60 percent do bill them, and of these, 25 percent are rejected and 75 percent reimbursed (many are not able to collect information about how much is collected), with practices receiving a median of $91.00 for the first hour of service (range $47.00 to $144.00) and a median of $75.00 for an additional half-hour (range $21.00 to $142.00).
The codes for prolonged physician service without direct (face-to-face) patient contact are 99358 and 99359. These services are described as appropriate for non-face-to-face time spent by physicians providing “evaluation and management services at any level” and billable for continuous or discontinuous time in 30- minute increments with no stated maximum. However, the Health Care Financing Administration considers these two codes to be bundled into office and hospital visits. Thus most payers do not
reimburse them, despite the example given in the CPT 2000 appendix citing the use of these codes to counsel family members without the presence of the patient. Therefore lengthy counseling time is not being reimbursed at all. COG PIs reported as follows: 77 percent do not bill these because they will not be reimbursed; 23 percent do bill them, reporting amounts paid as low as $10.00–30.00.
Lack of reimbursement or poor reimbursement for the prolonged services codes is a serious problem for pediatrics, but the total nonreimbursement of non-face-to-face time codes is an even greater problem for several reasons. First, the parents of seriously ill children often wish to discuss treatment issues—particularly issues regarding the possible death of the child—without their child present. Second, when the child is an infant or toddler, his or her face-to-face presence is irrelevant to the discussions, and children most often are not present for medically complex conversations. Third, interdisciplinary team management is discouraged by this system. Often, when the physician is with the parents, child life or social work staff are counseling the child patient. There are no codes for the latter staff to charge, so their services are bundled into the physician or team services. As a result, the charges that providers code must support the services of the entire team working with the child and family. The American Academy of Hospice and Palliative Medicine training guide for physicians (Storey and Knight, 1996) states that current reimbursement systems “discourage significant patient-physician interaction by selectively reimbursing for brief, procedure-related visits.”
Representative examples from the practices of the authors are presented described below.
A young girl with a brain tumor had a suspicious lesion on magnetic resonance imaging (MRI) at the end of therapy. The child was scheduled for a biopsy, under anesthesia; the oncology pediatric nurse practitioner did the history and physical that morning. The neurosurgeon performed the biopsy and frozen section confirmed the malignancy that day. An hour later, the pediatric oncologist spent 90 minutes with the parents, reviewing the biopsy results, discussing their sadness and fear, and going over with them the treatment options as well as strategizing about how to tell the 10-year-old child. The oncology billing office stated that there was no way for the pediatric oncologist to bill for this time. It was not face-to-face with the patient, and 99238 or 99239 codes are denied in that state. Codes 99354– 99357 could not be used as extended time codes, since the oncologist had not done the history and physical and thus, had not had any face-to-face contact with the child, which is required for those codes. Thus, 90 minutes of time, highly valued by the family, was not reimbursed at all.
A hospital visit of intermediate complexity, which included assess-
ment of liver enlargement in the context of chemotherapy for acute lymphoblastic leukemia, and 20 minutes of counseling the family, took place on the same day that the pediatric oncologist did a spinal tap on a child (delivering chemotherapy). The hospital visit code was denied as “global to the 96450” (the spinal tap code). Thus, a complex hospital visit, with the total time for the day at 60 minutes, including a procedure as well as extensive counseling, was reimbursed at $161.00. Compare this to local reimbursement rates for upper gastrointestinal endoscopy ($325.00), sigmoidoscopy ($118.30), mammogram ($94.30), and tonsillectomy/adenoidectomy ($390.00).
Even when these codes are reimbursed, the rate is not sufficient to support a clinical practice that includes a multidisciplinary team, despite its value to patients and families:
A clinic visit (99215) with extra counseling time, totaling 80 minutes, was reimbursed at $64.62 with the prolonged services code denied.
A hospital visit with management of a complex medical problem was reimbursed at $78.96 (recall that this allows up to 35 minutes before a prolonged services code is allowed to be charged). The prolonged services code 99356 was reimbursed at $96.32, adding up to $143.58 for 90 minutes of time. Physicians will be hard pressed to run a clinical service at that rate, let alone utilize the multidisciplinary psychosocial team needed by these families.
Hospital Discharge Services
The care of children with advanced illness requires a multidisciplinary team approach, but reimbursement for the contribution of nonphysician providers is extremely poor. Teams depend on staff such as nurse practitioners to do a great deal of teaching at the time of discharge. This teaching can prevent readmissions and increases the parents’ sense of control and efficacy. Payers routinely deny nurse practitioner charges for discharge services, despite their being licensed and credentialed to provide this service.
Case Management Services
Case management services (listed in CPT 2000) consist of team conferences and telephone calls, which play a huge role in the management of children with advanced illness. They are reimbursed inadequately or not at all.
TEAM CONFERENCES Billing codes exist for medical conferences by a physi-
cian with the interdisciplinary team present (codes are 99361 for 30 minutes and 99362 for 60 minutes). These conferences are described as follows (CPT, 2000): “Medical conference by a physician with interdisciplinary team of health professionals or representatives of community agencies to coordinate activities of patient care (patient not present).” These services are reimbursed very poorly despite their significant contribution to the care of children with advanced illness. COG PIs report that 75 percent no longer bill this code because of nonreimbursement. Of the 25 percent who do bill this, 64 percent get some reimbursement but few could state amounts. The median reported was $75.00 (range $40.00 to $250.00).
TELEPHONE CALLS The billing codes for phone management by a physician are described as follows: “telephone call by a physician to patient or for consultation or medical management with other health care professionals,” from “simple or brief” to “complex or lengthy to coordinate complex services of several different health professionals working on different aspects of the total patient care plan.” This excellent description notwithstanding, these codes are reimbursed poorly or not reimbursed at all. The National Association of Children’s Hospitals and Related Institutions (NACHRI) pediatric oncology study group recently conducted a study of time spent in telephone care (NACHRI, 1999). In a four week study, oncology practices logged 18 to 84 hours of calls. COG PIs report that 98 percent of them do not bill for telephone calls since they are not reimbursed. The few who billed reported receiving no reimbursement.
Care Plan Oversight Services
Physicians are responsible for the medical care (prescriptions, pain management, nutrition and fluid management, infection management, and management of symptoms to the time of death) of children dying at home in the care of hospice teams. It is well established that parents strongly desire the physician who treated their child’s illness to oversee care through the dying process (Liben, 1996; Martinson, 1995). These physicians most commonly are not employed by the hospice and so must bill for professional services according to the usual codes. Most supervision of care takes place over the phone, in a mode compatible with that described by care plan oversight codes.
There are various codes for care plan oversight services (99374, 99375, 99377, 99378, 99379, and 99380). As an example, 99377 is described as follows:
Physician supervision of a hospice patient (patient not present) requiring complex and multidisciplinary care modalities involving regular physician development and/or revision of care plans, review of subsequent reports
of patient status, review of related laboratory studies, communication with other health care professionals involved in a patient’s care, integration of new information into the medical treatment plan and/or adjustment of medical therapy, within a calendar month; 15–29 minutes.
Half an hour per month grossly underestimates the time spent in this activity, especially as symptom control needs escalate at the end of life. It is not practical for physicians to make daily home visits, but daily phone calls are common. These codes are very poorly reimbursed despite the heavy responsibility of managing a dying child at home in the care of a home health agency. While the CPT codebook reflects this complexity, the codes are rejected if billed more than once a month, for time spent up to an hour. COG PIs report that 85 percent did not bill for these codes because of total lack of reimbursement. Of the few who did bill and knew the amount paid, the median was $66.00.
Reimbursement of Hospice and Palliative Care Services
Few children with life-threatening conditions get comprehensive palliative care. In the United States, hospice rules and payments are influenced strongly by the federal government, particularly the Medicare hospice benefit, which requires relinquishing reimbursement for potentially life-prolonging treatment in favor of just palliative treatment (Kinzbrunner, 1998; Vermillion, 1996). Every child suffering from chronic, life-threatening and terminal illness needs palliative care interventions, whether or not the family has given up on a cure. Unfortunately, current hospice admission and reimbursement practices are not consistent with optimal palliative care of seriously ill patients (Field and Cassel, 1997a).
Hospice Admission Guidelines
Hospice is a mechanism for delivering care when the goal to achieve comfort overrides the goal to prolong life. It has become synonymous in the United States with palliative care, mainly because of rules established under the Medicare program requiring patients to choose between potentially life-prolonging treatment and palliative care under hospice. Similarly, current guidelines encourage the provision of expert palliative care to children suffering from serious illnesses and their families only if they agree to be enrolled in hospice. It is very difficult, however, to admit pediatric oncology patients to hospice.
The great majority of terminally ill children and their parents are not willing to sign on to hospice service for a variety of reasons. To accept hospice care, both parents and physicians must have a minimal level of “uncertainty” regarding the child’s prognosis and must have exhausted all
other resources. Accepting hospice means that families have abandoned hope for a cure and have begun to focus on the fact that the child will eventually die. This acceptance of the reality of death is one of a series of steps in a palliative care approach. Unfortunately, acceptance of terminal illness is also a prerequisite for receiving palliative care services through hospice. This is a no-win situation in which acceptance of death is both a prerequisite for eligibility and an expected outcome of palliative care intervention.
The Medicaid hospice benefit provides reimbursement for expert pain and symptom management, grief and bereavement counseling, pastoral care counseling, and home nursing care. Experience in clinical practice, however, shows that children with chronic, life-threatening and terminal illnesses need these services long before they become eligible to receive hospice care (Kane et al., 2000). This includes patients whose disease is likely to be incurable but who continue to be treated with the intent to cure (e.g., with a Phase I agent), even though the chance that treatment may be of benefit is very small and may entail significant toxicity.
Reimbursement Practices in Hospice and Palliative Care
Hospice financing mechanisms are often inimical to quality palliative care. Hospice organizations—which are usually reimbursed at a fixed per diem rate—will not deliver some expensive services for fear of jeopardizing the solvency of their programs (Field and Cassel, 1997b). From a financial perspective, the “ideal” hospice patient is the one who lives the longest with minimal interventions in both personnel time and pharmacotherapy. Thus, hospice is placed in the situation of favoring admission of patients who will not receive medical interventions that may prolong their life. Also, it is more likely to refuse admission to patients with a disease known to be incurable but who continue potentially life-prolonging interventions. This is particularly detrimental for the pediatric cancer patient. Examples where current guidelines may interfere with effective palliative care include the child receiving therapy for a highly malignant disease for whom long-term disease-free survival is known to be unlikely at diagnosis, the child enrolled in a Phase I or II study or its equivalent, and the child with disease refractory to experimental or conventional treatment who most likely will die from his or her disease but whose parents continue to hope for cure.
A clinical case may serve to illustrate this situation. “AJ” was a 20-year-old patient with acute lymphoblastic leukemia who relapsed in the bone marrow for the fourth time. He had received care from his oncology team for eight years. Without a bone marrow transplantation, the prognosis for long-term survival was poor, and this was known by the medical team, patient, and family. However, there was no suitable bone marrow
donor available: he had no siblings, and there were no unrelated donors who matched sufficiently well. Nonetheless, the patient and his family continued to choose treatment with curative intent hoping that his leukemia would stay in remission while the search for a marrow donor continued. At the time of his fourth relapse, he was still refusing hospice services—but he was not in denial of reality. He had decided he wanted all medical interventions necessary to keep him alive for as long as possible, but he refused artificial life support, mechanical ventilation, and cardiopulmonary resuscitation. He had signed an out-of-hospital do-not-resuscitate (DNR) form and discussed advance directives, choosing his mother as a surrogate decisionmaker if he was unable to make his own decisions. At one point in this process, he was placed in a difficult dilemma: choose treatment for your leukemia (oral palliative chemotherapy, transfusion of blood products) or hospice services at home. Hospice refused to admit him because of the high cost of care related to palliative interventions necessary to prolong his life, which he believed continued to be of good quality. This young man eventually died in the hospital from a serious infection. He deserved the benefit of palliative interventions but never received hospice care. This seems to be the case for the majority of pediatric oncology patients who die from progressive disease.
Another problem encountered in pediatrics is the choice parents sometimes have to make between hospice care and home nursing services for their ill child. Children who are eligible for services under state Medicaid waiver programs for medically fragile children (potentially any child with a life-limiting illness) are generally not eligible for hospice benefits under Medicaid or private insurers. This reality puts parents in a no-win situation, forcing them to choose between hands-on nursing support hours versus the holistic family-centered care provided by a supportive care or hospice and palliative care program. Most chronically or terminally ill children would benefit from some hours of home nursing support, as defined by the Medicaid criteria. However, most hospice and palliative care programs do not provide continuous nursing support in the home, unless they’re paired in a contractual arrangement with a home care company or unless they have chosen to bear the high cost of maintaining a home care license within a hospice.
“Bridge programs” have attempted to circumvent this issue by covering a severely ill child in a home care model until the child is strictly “hospice eligible.” Reimbursement for these comprehensive services is sparse at best, and usually means that hospice and palliative care programs are following families as “self- pay” until the very end of the child’s life or until the child is placed in a residential facility when home care needs are too great. Under these circumstances, the only part of the multidisciplinary care plan that is potentially reimbursable is physician billing, and the myriad other impor-
tant services provided by the rest of the hospice or palliative care team go financially unrewarded. This reality creates something of a paradox. Hospice and palliative care organizations often recognize the importance of establishing a separate pediatric team because of the issues unique to pediatric end-of-life care. However, the ability to lose money on a pediatric program, due to poor reimbursement and high numbers of self-pay (usually meaning “no pay”) patients, puts undue pressure on smaller programs and necessitates high levels of private, philanthropic support. The responsibility for providing specialized pediatric palliative care services, therefore, generally falls on agencies that are large enough to absorb financial losses on the pediatric side. Such programs generally have as part of their mission statements a commitment to providing quality end-of-life care to children and do so primarily as a community service or as a marketing tool, neither of which is necessarily sustainable.
In addition to putting families and providers in a difficult situation, the current system is also internally inconsistent. It is acceptable for patients with third-party insurance and no financial eligibility for Medicaid to tap into Medicaid waiver programs for nursing support. However, Medicaid patients cannot have both, forcing families to choose between nursing support and hospice or palliative care services. The justification for this policy is to prevent duplication of services by excluding patients receiving waiver support from being eligible for the hospice benefit and other Medicaid programs. The “duplication of service argument” applies only to low-income patients and should be amended.
Even in situations where pediatric patients are enrolled “on the benefit,” the contracted rate of reimbursement pays for only a fraction of the services provided. Patients who are on “full benefit” are entitled to all of the core services modeled after the Medicare hospice benefit, but this does not include creative arts therapies (e.g., art, dance and movement, music, and drama therapies), nor does it provide bereavement services beyond 13 months, too short for many families who lose children to illness. Longer follow-up is unlikely to be reimbursed by any formal mechanism other than philanthropic support. Similarly, spiritual care, and child life and expressive therapy are particularly important in the pediatric arena as families struggle with a child’s terminal illness. The literature abounds with examples of the importance of art, music, and movement therapy for dying children and their families, particularly siblings, in supporting effective coping and grieving. Yet these disciplines are not reimbursed at all in the current benefit system.
Apart from the issues discussed above, pediatric palliative care faces an additional challenge. While no direct reimbursement for palliative care is available for any patient, adult palliative care programs are able to capture some revenue through physician billing under different codes (e.g., for
symptom management) than those used by the primary care provider or other specialists involved in the patient’s care. In pediatrics, however, this revenue stream is difficult to capture without the services of a pediatric hospice or palliative care physician. Some hospitals and acute care institutions allow internal medicine and family practice physicians to see pediatric patients (especially adolescents), which affords pediatric patients the benefit of hospice or palliative care expertise and allows practitioners to bill for their services. However, these patients are then not receiving specialized pediatric care. The likelihood of finding expertise in pediatric palliative care and hospice is currently low and drops even further when patients are not within a hospital setting.
One additional problem in pediatric hospice care is the shortage of contracted beds for symptom control or hospice admissions within acute care settings. In large hospitals, adult hospice programs have little trouble securing a few beds to be used for their patients. In pediatrics, however, hospitals are reluctant to commit beds to hospice care because they may be empty much of the time but could be used for other admissions if available.
Set a minimum reimbursement rate for physician communication time with parents and with patients. Legislation may be required to prevent denial of payment for this basic service.
Enforce payment for telephone time and care plan oversight time.
Enforce reimbursement of team conferences with patients and families at a rate that reflects participation of the multidisciplinary team.
Develop and test palliative care codes for reimbursement of physicians and other health care providers for pediatric palliative care interventions offered based on the goals of medical care:
Palliative care codes: services offered for patients with incurable disease receiving treatment with the intent to prolong a life of good quality
Hospice care codes: palliative care in which the primary focus of treatment is end-of-life comfort care
Bereavement care codes: palliative care in the form of grief and bereavement counseling for surviving family members
Develop patient evaluation and management codes for reimbursement of palliative care services in different settings. Include funding mechanisms for the entire multidisciplinary team, including mental health and bereavement workers.
Create contractual arrangements for a small number of beds in pediatric hospitals to be used for children needing respite or symptom control care, with appropriate reimbursement.
RESEARCH NEEDS IN PEDIATRIC ONCOLOGY END-OF-LIFE CARE
The research base in pediatric oncology for dying children is deficient in every area, directly affecting the quality of the care that children receive. In addition to the important work of developing and testing new treatments to reduce the number of children who face death from cancer, research must focus on the children who do, in fact, die. Results of research in the following areas could be of direct benefit in the treatment of dying children.
Describe current end-of-life care practice patterns. The descriptive information available for the care of dying children is inadequate (e.g., very little information on the use of hospice, the use of palliative care teams or pain services, incidence of death at home versus the hospital or other inpatient facility, use of advance directives or DNR orders and the duration these are in effect before death, use of psychosocial multidisciplinary teams, and patient and family satisfaction with these services).
Create tools to assess the quality of pediatric end-of-life care. A single instrument (questionnaire) has been validated for pediatrics, consisting of a two-hour interview, which is not practical for broad use (Wolfe, 2000). New instruments (either completely new or adapted from adult models) for widespread use are needed.
Evaluate pediatric models for provider-patient-family communication. Research is needed to define communication models that prevent psychological harm to parents and children and that result in a sense of control and efficacy for parents. Research questions include how to communicate bad news effectively, how to discuss withdrawal of therapies, how to teach what to expect as their child dies, how to communicate goals of care that incorporate both curative and palliative therapies without a feeling of “giving up,” and how to facilitate communication across treatment sites.
Create and evaluate comprehensive parent educational materials, including, for example, what to expect during withdrawal of support, what will be experienced during an expected death at home, and how to advocate for symptom control.
Evaluate models of decisionmaking that are family centered, and emphasize the involvement of the child.
Evaluate models of care that address the needs of siblings.
Investigate the barriers to optimal symptom control in pediatric oncology practice.
Develop and validate symptom (and suffering) assessment tools for the pediatric population to be used for both research and treatment.
Initiate clinical trials in symptom control within the Children’s Oncology Group.
Incorporate symptom control algorithms into COG clinical trials.
Develop models of care incorporating the principles of palliative care throughout the mainstream of medical therapy of seriously ill pediatric oncology patients, from the time of diagnosis to the time of death.
Investigate the impact on care delivery of barriers to optimal care.
Develop and evaluate a “mobile medical record” for palliative care content that will follow the patient across various treatment sites.
Assess the needs of medical providers caring for dying children.
Develop palliative care codes for reimbursement of physicians and other health care providers for pediatric palliative care interventions offered across settings based on the goals of medical care.
Investigate the financial implications of care models developed as discussed above.
Develop education strategies for providers.
PEDIATRIC ONCOLOGY END-OF-LIFE CARE: FUTURE DIRECTIONS
The education of providers must be adapted to meet the unique needs of those caring for dying children, and must include both traditional and nontraditional teaching methods. The recommendations below include mechanisms for accountability.
Develop educational materials for families that facilitate the most complete understanding of the child’s condition and prepare them as much as possible for what will occur as the child dies. In this manner, patients and families will be empowered to participate in treatment decisions.
Facilitate the involvement of children in their treatment decisions (assent).
Develop and evaluate models of oncology care that incorporate palliative care principles and facilitate continuity of care by providers educated in pediatric end-of-life care.
Develop strategies to address the inadequate relief of pain and other symptoms in pediatric oncology patients nearing the end of life.
Institute regulatory and reimbursement policies that adequately address the complexity and time involved in caring for children with advanced illness.
Develop research initiatives that will assess current practice patterns, evaluate models of care delivery, evaluate models of communication and decisionmaking, study methods of symptom control, evaluate the feasibility and cost-effectiveness of new reimbursement models, and evaluate innovative educational initiatives.
AAHPM (American Academy of Hospice and Palliative Medicine). 1998. Hospice/Palliative Care Training for Physicians: UNIPACs. Dubuque, Iowa: Kendall/Hunt Publishing Company.
AAP (American Academy of Pediatrics), Committee on Bioethics. Informed consent, parental permission, and assent in pediatric practice. Pediatrics 1995; 95:314–317.
Accreditation Standards, 2000. Liaison Committee on Medical Education.
Angell, M., Caring for the dying-congressional mischief (editorial). N Engl J Med 1999; 341: 1923–1925.
Armstrong-Dailey A, Goltzer SZ (eds). 1993. Hospice Care for Children. New York: Oxford University Press.
Bartholome WG. Care of the dying child: the demands of ethics. Second Opin 1993; 18:25–39.
Bennet DS. Depression among children with chronic medical problems: a meta-analysis. J Pediatr Psychol 1994; 19:149–169.
Bleyer WA. The U.S. pediatric cancer clinical trials programmes: international implications and the way forward. Eur J Cancer 1997; 33:1439–1447.
Bleyer WA, Tejeda H, Murphy SB, Robinson LL, Ross JA, Pollock BH, et al. National cancer clinical trials: Children have equal access; adolescents do not. J Adolesc Health 1997; 21: 366–73
Bluebond-Langner M. 1978. The Private Lives of Dying Children. Princeton, NJ: Princeton University Press.
Brenner P. 1993. The volunteer component. In: Hospice Care for Children, Armstrong-Dailey A, Goltzer SZ (eds). New York: Oxford University Press, pp.198–218.
Brock DW. 1989. Children’s competence for health care decisionmaking. In: Children and Health Care: Moral and Social Issues, Kopelman LM, Moskop JC (eds). Boston: Kluwer Academic Publishers, pp. 181–212.
Carr-Gregg MRC, Sawyer SM, Clarke CF, Bowes G. Caring for the terminally ill adolescent. Med J Aust 1997; 166:255–258.
Castro O, Gordeuk VR, Dawkins F. Letter to the editor. N Engl J Med 2000; 342:1049–50.
Chevlen E. Letter to the editor. N Engl J Med 2000; 342:1049–1050.
Children’s Hospice International. 1998. 1998 Survey: Hospice Care for Children. Executive Summary Report. Alexandria, VA.
Collins JJ, Grier HE, Kinney HC, Berde CB. Control of severe pain in children with terminal malignancy. J Pediatr 1995a; 126:653–657.
Collins JJ, Kerner J, Sentivany S, Berde CB. Intravenous amitriptyline in pediatrics. J Pain Sympt Manage 1995b; 10:471–475.
Collins JJ, Grier HE, Sethna NF, Wilder RT, Berde CB. Regional anesthesia for pain associated with terminal pediatric malignancy. Pain 1996a; 65:63–69.
Collins JJ, Geake J, Grier HE, Houck CS, Thaler HT, Weinstein HJ, Twum-Danso NY, Berde CB. Patient-controlled analgesia for mucositis pain in children: a three-period crossover study comparing morphine and hydromorphone. J Pediatr 1996b; 129:722–728.
Collins JJ, Stevens MM, Cousens P. Home care for the dying child. A parent’s perception. Australian Family Physician 1998; 27:610–614.
Collins JJ, Dunkel IJ, Gupta SK, Inturrisi CE, Lapin J, Palmer LN, Weinstein SM, Portenoy RK. Transdermal fentanyl in children with cancer pain: feasibility, tolerability, and pharmacokinetic correlates. J Pediatr 1999; 134:319–323.
Collins JJ, Byrnes ME, Dunkel IJ, Lapin J, Nadel T, Thaler HT, Polyak T, Rapkin B, Portenoy RK. The measurement of symptoms in children with cancer. J Pain Sympt Manage (in press).
Cooper MG, Keneally JP, Kinchington D. Continuous brachial plexus neural blockade in a child with intractable cancer pain. J Pain Sympt Manage 1994; 9:277–281.
Council on Scientific Affairs, American Medical Association. Good care of the dying patient. JAMA 1996; 275:474–478.
Department of Health and Human Services, Food and Drug Administration. 1998. Regulations requiring manufacturers to assess the safety and effectiveness of new drugs and biological products in pediatric patients. Federal Register p. 66632.
Die Trill M, Kovalcik R. The child with cancer: influence of culture on truth-telling and patient care. Ann NY Acad Sci 1997; 809:197–210.
Doyle D, Hanks GWC, MacDonald N (eds). 1998. Oxford Textbook of Palliative Medicine. Oxford: Oxford University Press, pp. 1013–1117.
Emanuel EJ, Weinberg DS, Gonin R, Hummel LR, Emanuel LL. How well is the Patient Self-Determination Act working? An early assessment. Am J Med 1993; 95:619–628.
Emanuel L, von Gunten C, Ferris F. 1999. The education for physicians on end-of-life care (EPEC) curriculum. Princeton, NJ: Robert Wood Johnson Foundation: Princeton.
Faulkner KW. Talking about death with a dying child. AJN 1997; 97:64–69.
Fehder WP, Sachs J, Uvaydova M, Douglas SD. Substance P as an immune modulator of anxiety. Neuroimmunomodulation 1997; 4:42–48.
Ferrell B, Grant M, Coyne P, Egan K, Paice J, Panke J. 2000. End-of-life nursing education consortium (ELNEC) project. American Association of Colleges of Nursing.
Field MJ, Cassel CK (eds). 1997. Approaching Death. Improving Care at the End of Life. Washington, DC: National Academy Press.
Frager G. Pediatric palliative care: building the model, bridging the gaps. J Palliat Care 1996; 12:9–12.
Fredrikson M, Furst CJ, Lekander M, Rotstein S, Blomgren H. Trait anxiety and anticipatory immune reactions in women receiving adjuvant chemotherapy for breast cancer. Brain Behav Immun 1993; 7:79–90.
Freyer DR. Children with cancer: special considerations in the discontinuation of life-sustaining treatment. Med Pediatr Oncol 1992; 20:136–142.
Friebert SE, Kodish ED. Kids and cancer: ethical issues in treating the pediatric oncology patient. Cancer Treat Res 2000; 102:99–135.
Goldman A, Christie D. Children with cancer talk about their own death with their families. Pediatr Hematol Oncol 1993; 10:223–31.
Goldman A. Home care of the dying child. J Palliative Care 1996; 12:16–19.
Goldman A (ed). 1994. Care of the Dying Child. Oxford: Oxford University Press.
Hain RDW, Patel N, Crabtree S, Pinkerton R. Respiratory symptoms in children dying from malignant disease. Palliat Med 1995; 9:201–206.
Hain RDW. Pain scales in children: a review. Palliat Med 1997; 11:341–350.
Hearn J, Higginson I. Do specialist palliative care teams improve outcomes for cancer patients? A systematic literature review. Palliat Med 1998; 12:317–332.
Hilden JM, Emanuel EJ, Fairclough DL, Link MP, Foley KM, Clarridge BC, Schnipper LE, Mayer RJ. Attitudes and practices among pediatric oncologists regarding end-of-life care: results of the 1998 American Society of Clinical Oncology survey. JCO 2001; 19: 205–212.
National Hospice Organization. 1997. Hospice Fact Sheet.
Howell DA. 1993. Role of the primary physician. In: Hospice Care for Children, Armstrong-Dailey A, Golzer SZ (eds). New York: Oxford University Press, pp. 172–183.
Hunt A, Joel S, Dick G, Goldman A. Population pharmacokinetics of oral morphine and its glucuronides in children receiving morphine as immediate-release liquid or sustained-release tablets for cancer pain. J Pediatr 1999; 135:47–55.
International Work Group on Death, Dying, and Bereavement. Palliative care for children. Death Studies 1993; 17:277–280.
James L, Johnson B. The needs of parents of pediatric oncology patients during the palliative care phase. J Pediatr Oncol Nurs 1997; 14:83–95.
Kane J, Barber RB, Jordan M, Tichenor KT, Camp K. Supportive/palliative care of children suffering from life-threatening and terminal illness. Am J Hosp & Palliat Care 2000; 17:165–172.
Kart T, Christrup LL, Rasmussen M. Recommended use of morphine in neonates, infants and children based on a literature review. Part 2-Clinical use. Paediatr Anaesthes 1997; 7:93– 101.
Kazak AE, Blackall G, Himelstein B, Brophy P, Daller R. Producing systemic change in pediatric practice: An intervention protocol for reducing distress during painful procedures. Fam Syst Med 1995; 13:173–185.
Kazak AE, Penati B, Brophy P, Himelstein B. Pharmacologic and psychologic interventions for procedural pain. Pediatr 1998; 102:59–66.
Khaneya S, Milrod B. Educational needs among pediatricians regarding caring for terminally ill children. Arch Pediatr Adolesc Med 1998; 152:909–914.
Kinzbrunner BM. Hospice: 15 years and beyond in the care of the dying. J Palliat Med 1998; 1:127–137.
Kosary Cl, Ries LAG, Miller B et al. (eds). 1995. Cancer in children. In: SEER Cancer Statistics Review, 1973–1992. U.S. Dept of health and Human Services, 1995. NIH publication number 96–2789. Bethesda, MD: National Cancer Institute, pp. 455–465.
Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics, 1999. CA Cancer J Clin 1999; 49:8–31.
Leikin S. The role of adolescents in decisions concerning their cancer therapy. Cancer 1993; 71(suppl):3342–3346.
Levy MH. Doctor-patient communication: the lifeline to comprehensive cancer care. ASCO Educational Book. 1998; 195–202.
Liben S. Pediatric palliative medicine: obstacles to overcome. J Palliat Care 1996; 12:24–28.
Manne SL, Jacobsen PB, Redd WH. Assessment of acute pediatric pain: do child self-report, parent ratings, and nurse ratings measure the same phenomenon? Pain 1992; 48:45–52.
Martinson IM. Hospice care for children: past, present, and future. J Pediatr Oncol Nurs 1993; 10:93–98.
Martinson I. 1993. A home care program. In: Hospice Care for Children, Armstrong-Dailey A, Golzer SZ (eds). New York: Oxford University Press, pp. 231–47.
Martinson IM. Improving care of dying children. West J Med 1995; 163:258–262.
Masera G, Spinetta JJ, Jankovic M, Ablin AR, D’Angio GJ, Van Dongen-Melman J, et al. Guidelines for assistance to terminally ill children with cancer: a report of the SIOP working committee on psychosocial issues in pediatric oncology. Med Pediatr Oncol 1999; 32:44–48.
McCarthy AM, Cool VA, Petersen M, Bruene DA. Cognitive behavioral pain and anxiety interventions in pediatric oncology centers and bone marrow transplant units. J Pediatr Oncol Nurs 1996; 13:3–12.
Mercadante S. Pain treatment and outcomes for patients with advanced cancer who receive follow-up care at home. Cancer 1999; 85:1849–1858.
NACHRI (National Association of Chilren’s Hospitals and Related Institutions). 1999. Patient Hematology and Oncology FOCUS Groups.
Nitschke R, Wunder S, Sexauer CL, Humphrey GB. The final-stage conference: the patient’s decision on research drugs in pediatric oncology. J Pediatr Psychol 1997; 2:58–64.
Nitschke R, Meyer WH, Sexauer CL, Parkhurst JB, Foster P, Huszti H. Care of terminally ill children with cancer. Med Ped Oncol 2000; 34:268–270.
Office of Protection from Research Risks (OPRR). 1991. Protection of human subjects. Department of Health and Human Services. National Institute of Health. Title 45, Code of Federal Regulations, Part 46. Appendix 4, p. 16.
Pain Relief Promotion Act of 1999. H.R. 2260. 106th Congress, 1st session.
Papadatou D. Training health professionals in caring for dying children and grieving families. Death Studies 1997; 21:575–600.
Pizzo P, Poplack D (eds). 1997. Principles and Practice of Pediatric Oncology. Philadelphia: Lippincott-Raven.
Razavi D, Delvaux N, Farvaques C, Robaye E. Immediate effectiveness of brief psychological training for health professionals dealing with terminally ill cancer patients: a controlled study. Soc Sci Med 1988; 27:369–375.
Razavi D, Delvaux N, Farvacques C, Robaye E. Brief psychological training for health care professionals dealing with cancer patients: a one-year assessment. Gen Hosp Psychiat 1991; 13:253–260.
Ross JA, Severson RK, Pollock BH, Robison LL. Childhood cancer in the United States: a geographical analysis of cases from the pediatric cooperative clinical trials groups. Cancer 1996; 77:201–207.
Ryan ND. Psychoneuroendocrinology of children and adolescents. Psychiatr Clin North Am 1998; 21:435–441.
Sagara M, Pickett M. Sociocultural influences and care of dying children in Japan and the United States. Cancer Nurs 1998; 21:274–281.
Sahler OJZ, Frager G, Levetown M, Cohn FG, Lipson MA. Medical education about end-of-life care in the pediatric setting: principles, challenges, and opportunities. Pediatr 2000; 105:575–584.
Schechter NL, Altman AJ, Weisman SJ (eds). Report of the consensus committee on pain in childhood cancer. Pediatr 1990; 86(suppl):813–834.
Schweitzer SO, Mitchell B, Landsverk J, Laparan L. The costs of a pediatric hospice program. Public Health Reports 1993; 108:37–44.
Shir Y, Shenkman Z, Shavelson V, Davidson EM, Rosen G. Oral methadone for the treatment of severe pain in hospitalized children: a report of five cases. Clin J Pain 1998; 14:350–353.
Sirkiä K, Hovi L, Pouttu J, Saarinen-Pihkala UM. Pain medication during terminal care of children with cancer. J Pain Sympt Manage 1998; 15:220–226.
Sourkes B. 1995. Armfuls of time: the psychological experience of the child with a life-threatening illness. Pittsburgh, PA: University of Pittsburgh Press.
Staats PS, Kost-Byerly S. Celiac plexus blockade in a 7-year old child with neuroblastoma. J Pain Sympt Manage 1995; 10:321–324.
Stevens MM. 1998. Psychological adaptation of the dying child. In: Oxford Textbook of Palliative Medicine , Doyle D, Hanks GWC, MacDonald N (eds). Oxford: Oxford University Press, pp. 1045–55.
Stevens MM. 1998. Care of the dying child and adolescent: family adjustment and support. In: Oxford Textbook of Palliative Medicine, Doyle D, Hanks GWC, MacDonald N (eds). Oxford: Oxford University Press, pp. 1058–75.
Storey P, Knight C. 1996. UNIPAC six: Ethical and legal decision making when caring for the terminally ill. Gainesville, FL: AAHPM.
Teno JM, Casey VA, Welch L, Edgman-Levitan, S. Patient focused, family centered end-of-life medical care: views of the guidelines and bereaved family members. Draft manuscript for Third Woods Hole conference on measuring quality of life and quality of care at life’s end. June 2000.
Vermillion J. 1996. The referral process and reimbursement. In: Hospice and Palliative Care Sheehan DC, Forman WB (eds). Sudbury, MA: Jones and Bartlett Publishers, pp 11–20.
Vickers JL, Carlisle C. Choices and control: parental experiences in pediatric terminal home care. J Ped Onc Nursing 2000; 17:12–21.
Weisman SJ. 1998. Supportive care in children with cancer. In: Principles and Practice of Supportive Oncology. Berger A, Portenoy RK, Weissman DE (eds). Philadelphia: Lippincott-Raven, pp. 845–52.
Wessel MA. The role of the primary pediatrician when a child dies [editorial]. Arch Pediatr Adolesc Med 1998; 152:837–838.
Whittam EH. Terminal care of the dying child: psychosocial implications of care. Cancer 1993; 71:3450–3462.
WHO (World Health Organization) and International Association for the Study of Pain. 1998. Cancer Pain Relief and Palliative Care in Children. Geneva: World Health Organization.
Wolfe J. Personal communication to Joanne Hilden, 2000.
Wolfe J, Grier HE, Klar N, Salem-Schatz S, Emanuel EJ, Weeks JC. Physician-assisted suicide and euthanasia: experiences and attitudes among parents of children who have died of cancer. Proc ASCO 1999; 18:577a.
Wolfe J, Grier HE, Klar N, Levin S, Ellenbogen J, Salem-Schatz S, Emanuel EJ, Weeks JC. Symptoms and suffering at the end of life in children with cancer. N Engl J Med 2000; 342:326–333.
Yee JD, Berde CB. Dextroamphetamine or methylphenidate as adjuvants to opioid analgesia for adolescents with cancer. J Pain Sympt Manage 1994; 9:122–125.