ETHICAL AND LEGAL ISSUES
I think in particular of two times when we made a particular request and were told that the ethics committee would be called. I’m now on the ethics committee at Children’s Hospital here and ... [I advocate there] that ethics committees not be viewed like the legal system. That’s not helpful or open to families but one other thing to be really frightened about.
Deborah Dokken, parent, 2001
Questions and disagreements about what constitutes appropriate medical treatment or limitations of treatment for infants and children with severe and often fatal medical problems are a staple of the bioethics literature. They are rarely—but then very visibly—the subject of litigation and legislation. Even so, when parents question or disagree with the health care team or hospital management, they may perceive some responses as legalistic and intimidating rather than constructive and compassionate.
One goal of palliative and end-of-life care is to minimize avoidable conflicts related to poor communication, cultural misunderstandings, deficient clinical care, and approaches to decision making that fail to assure families that they and the health care team are doing their best for the child. Such failures can haunt family members and clinicians long after a child’s death.
Not all conflict raises ethical or legal concerns, and ethical concerns about clinical decisionmaking certainly may arise in the absence of conflict. This chapter provides an overview of the ethical and legal context for decisions by clinicians, parents, and children about pediatric palliative and end-of-life care. It
describes several categories of decisions that have prompted ethical or legal questions and disagreements;
outlines the key ethical obligations of clinicians as commonly identified by clinicians and bioethicists;
discusses the potential for conflicts among ethical obligations;
identifies some strategies for preventing or resolving conflicts among those involved in decisions about a child’s care; and
reviews some ethical and legal questions about the care of children, including the status of children as decisionmakers and certain boundaries on parental authority.
This discussion focuses on clinicians’ decisions rather than on parental obligations or societal issues of resource allocation and burden sharing. The discussion does not consider a number of ethical and legal controversies. These include the declaration of brain death in anencephalic newborns, continuation or discontinuation of pregnancy following prenatal diagnosis of a fatal medical problem, use of certain alternative or complementary therapies at the end of life, and euthanasia. The chapter also does not discuss licensure issues, some aspects of which are considered in Chapter 6. Chapter 10 examines ethical issues in research involving children. Although this chapter does not include formal recommendations, the committee believes that the strategies for preventing and resolving conflict discussed here should be considered and tested as institutions develop the procedures and protocols recommended in Chapters 4, 5, and 6.
Many legal and ethical questions considered here are relevant for palliative and end-of-life care for both adults and children. Some issues are, however, unique to children. One example is whether schools serving medically fragile children must honor parental requests that cardiopulmonary resuscitation not be attempted for their child. Another involves getting permission from an abusive parent to withdraw life support when the parent will be charged with murder after the child’s death. Not unique but especially wrenching when a child is involved are questions about the ethics of using or forgoing certain medical interventions when disease is far advanced.
Although it is important to understand the limits that courts or legislatures have imposed on decisionmaking by clinicians or parents or both, these limits will not affect most decisions about palliative and end-of-life care for children. Similarly, the legal limits on children’s decisions about medical treatment are not inconsequential, but they should not constrain efforts by clinicians and parents to inform children about their condition (consistent with their intellectual and emotional maturity) and to determine and consider children’s treatment preferences. An assumption of this committee and most sources consulted by the committee is that litigation should be a last and rare resort when clinicians and family members disagree.
As with other aspects of palliative, end-of-life, and bereavement care, the legal system and ethical frameworks discussed here reflect this country’s history and dominant culture. Although the legal system of the United States and arguments based in biomedical ethics may sometimes confuse or surprise those raised in this country, people raised in other cultures may find them incomprehensible or possibly shocking. Particularly in legal matters, flexibility to consider cultural differences may be limited. Nonetheless, it is still both humane and prudent for clinicians, social service personnel, and legal counsel to be sensitive to differences in cultural values and experiences and to try to minimize conflict and family distress related to such differences. (See Appendix D for additional discussion of cultural diversity.)
TYPES OF DECISIONS
In recent decades, many legal disputes and ethical debates about care for infants and children have involved decisions to start or stop medical interventions. Other disputes focus on the limits of parents’ authority to decide about their child’s care and on whether and to what extent quality-of-life and financial considerations should influence decisions about life-sustaining treatments. This section briefly describes these decisions and the questions they have raised. Much of the initial discussion of some of these issues, for example, withholding and withdrawing life-support technologies, occurred in the 1970s and 1980s and predated the increased attention to the techniques and benefits of palliative care. Again, the issues are not necessarily unique to children but tend to be more emotionally charged and difficult for all involved.
Decisions About Who Decides
In general, parents have the legal right to make decisions about medical care for their child. This parental authority is occasionally challenged, usually when parents refuse treatment recommended by their child’s physician. In addition, adolescents have sometimes sought control over crucial decisions about their future, often in situations involving reproduction but sometimes involving care for far advanced illness. Sometimes they succeed. For example, in 1994, state officials unsuccessfully tried to force a 15-year-old boy, who had received two liver transplants, to take medications that he refused on grounds of unbearable pain (Penkower, 1996; see more generally McCabe, et al., 1996).1
For the situation in which a child’s participation in research is involved, federal regulations normally require both that parents give permission and that children also “assent” to participation. Chapter 10 discusses ethical issues related to the involvement of children in research.
That parents have the legal right to make decisions does not and should not, however, mean that parents and clinicians should simply exclude children from discussions and decisions about their care. Failing to provide children with information and the opportunity to discuss their fears, concerns, and preferences can isolate them and add to their anxiety and other distress. In Chapter 4, the committee recommended that children and adolescents be informed and involved in decisionmaking—consistent with their condition, maturity, and preferences and with sensitivity to the family’s culture and values. Involving the child and trying to see care through the child’s eyes is also an element of child- and family-centered care as discussed in Chapter 6. Particularly for adolescents, restricting participation in decisionmaking “may unnecessarily create tension where a therapeutic alliance is needed” (Burns and Truog, 1997, p. 73). Many medical care consent forms include a section documenting “assent” to care by adolescents.2
Care must, however, be taken not to give the false impression that parents will never override their child’s expressed wishes. Decisions by child patients—and by parents and adult patients—may be constrained by organizational or governmental policies or by environmental factors. For example, state laws may require emergency medical services providers to attempt resuscitation even for adults or children with DNR orders (Sabatino, 1999). Likewise, deeply held cultural values may lead parents to reject proposals to involve the child or even provide the child with information about his or her condition. Clinicians may respect these values but still be dismayed. Years of experience may temper such reactions. As one clinician observed “. . . more and more I’ve realized that it’s going to unfold the way that family needs it to unfold” (Mildred Solomon, Ed.D., Education Development Center, taken from interviews conducted for the Initiative for Pediatric Palliative Care, unpublished analysis by Hardart et al., 2002).
Although it is important to understand the ethical and legal context for decisionmaking, most disagreements about care are resolved informally through discussion and reflection. For example, the vignette about the Devane family in Chapter 3 described how an adolescent with recurrent cancer prevailed in refusing further burdensome experimental treatment (wanted by her physician and parents) and in living her remaining time doing what she wanted to do.
Decisions About Treatments
Stopping Versus Not Starting Treatment
Ethicists and judges generally agree that there is no meaningful ethical or legal difference between deciding not to initiate a treatment and deciding to discontinue a treatment (assuming that appropriate attention is paid to such matters as patient comfort) (see, e.g., President’s Commission, 1983; see also AMA, 1994a).3 Nonetheless, in the committee’s experience, both clinicians and parents may find it more emotionally difficult and morally challenging to stop a treatment once begun than not to initiate it in the first place.
Implementing a decision to stop a life-sustaining medical intervention may also be technically challenging as demonstrated by ongoing discussions of which strategy for removing a ventilator or breathing tube best meets patient and family needs (see Chapter 5). These needs may vary, depending on the patient’s condition and the family’s values. Some clinicians may even be reluctant to start such interventions when they might be useful for fear of not being able to stop them if they prove nonbeneficial. Thus, it is important for those responsible for such interventions to be thoroughly knowledgeable about all clinical and ethical aspects of initiating and stopping them.
A recently reported survey of clinicians involved in pediatric intensive care found that although 78 percent of physicians agreed that the decision not to start an intervention (such as mechanical ventilation) and the decision to stop it were ethically the same, only 57 percent of nurses did (Burns et al., 2001). Another recent survey at seven children’s hospitals or pediatric units of general hospitals found that majorities of nurses and residents and a near majority of attending physicians disagreed with this proposition of moral equivalence (Solomon et al., 2000c). Yet other research suggests that decisions about which interventions are forgone—and when—may not be supported by clear clinical or ethical rationales (see, e.g., Faber-Langendoen, 1996; Asch et al., 1999; Truog et al., 2000).
Clinicians’ attitudes and practices related to the use of life-sustaining medical interventions may be influenced by a number of factors including their cultural or religious values and their emotional response to a child’s grave illness or injury. Other factors include (1) ignorance or misunderstanding of the evidence about the benefits and burdens of specific interven-
tions at the end of life, (2) insufficient education in clinical ethics and reinforcement of ethics in day-to-day practice, and (3) feelings of helplessness related, in part, to inadequate knowledge of palliative strategies to prevent or relieve distress when life-sustaining interventions are forgone. This lack of palliative knowledge may contribute both to the continuation of nonbeneficial interventions and to needless suffering when interventions are halted, for example, when the removal of a breathing tube is managed with paralytic agents that do not relieve distress but prevent the patient from communicating it (Rushton and Terry, 1995; Truog et al., 2000; Henig et al., 2001).
With advances in medical technologies in the second half of the twentieth century has come increasing anxiety among clinicians, ethicists, and the informed public about the appropriate use of life-sustaining interventions, especially when a patient’s chance of survival is very low. Cardiopulmonary resuscitation is perhaps the best known of an array of such interventions that may be attempted (or forgone) when death is imminent.4 Other life-support measures include mechanical ventilation (for reasons in addition to respiratory arrest), mechanical provision of nutrients or fluids (artificial nutrition or hydration), blood transfusions, antibiotics, and dialysis.5
Decisions about life-sustaining interventions—especially resuscitation attempts—are often made in an atmosphere of crisis and even panic, but this is not always the case. For example, mechanical ventilation may be begun in a child with a progressive neuromuscular disease such as muscular dystrophy after extended discussion and profound reflection by the family, the child, and the care team on the benefits, risks, and burdens of this action.6 Still, even when parents know that their child has an invariably
fatal condition, such as Tay-Sachs disease, and have been counseled that cardiac or respiratory arrest can be expected when functional and cognitive deterioration is advanced and even when they understand the ultimate futility of attempting resuscitation, they may still—in their distress at that stage—call “9-1-1,” thereby triggering full-scale resuscitation efforts.
It is now generally accepted among clinicians, ethicists, policymakers, and the informed public that forgoing life-support measures is often appropriate as death approaches (see Chapter 5). Notwithstanding such consensus, the parents of a dying child may not accept a clinician’s assessment that life-support no longer has the potential to benefit the child and may insist that it be continued. Much of the discussion later in this chapter focuses on strategies for understanding the reasons for family rejection of physician counsel and for resolving disagreements without harm to the patient, the family, or the health care team.
Some controversy about forgoing specific interventions persists. For example, discussion continues about whether artificial nutrition and hydration at the end of life are morally or clinically different from other interventions and whether they should be maintained when other life-support measures are forgone (see, e.g., Nelson et al., 1995; Burck, 1996; Post, 2001; Gillick, 2001). Surveys of pediatric specialists have found that specialists are much more willing to forgo resuscitation or mechanical ventilation than artificial hydration given the same clinical situations (Nelson et al., 1995; see also Smith and Wigton, 1987). Other studies suggest that clinicians may believe mistakenly that artificial nutrition and hydration are always legally required in situations or jurisdictions in which that is not the case (Meisel et al., 2000). Although research findings are limited and inconsistent, some suggest that artificial nutrition or hydration when death is near may be more burdensome than beneficial in some circumstances (Finucane et al., 1999; Brody, 2000; see also Chapter 5). Again, whatever clinicians may believe and whatever the evidence may be, family values and understandings may differ.
Another controversy, noted earlier, involves whether schools will honor do-not-resuscitate (DNR) orders for medically fragile children, an increasing number of whom attend school (see, e.g., NEA, 1994; Rushton et al., 1994; AAP, 2000d, Hoffpauir, 2001). Some districts have refused to honor parents’ wishes, partly out of concern about the impact on staff and other students (see, e.g., Tucson Unified School District, 1996; Laramie County, 1996). Other districts have worked with parents on a case-by-case basis to develop a plan of care. State policies vary. In its discussion of this issue, the American Academy of Pediatrics (AAP, 2000d) does not make a blanket recommendation about what schools should do. Rather, it recommends that pediatricians and parents meet with relevant school personnel to explain the child’s medical condition and the goals of care and to hear the
concerns of other parents and school personnel, for example, about liability and effects on other children. It also urges that all parties be realistic, flexible, and ready to negotiate.
Interventions such as resuscitation and mechanical ventilation are usually intended to sustain life temporarily, for example, when a patient is incapacitated as a result of surgery or injury and when recovery or prolonged meaningful life is a realistic goal. They are not intended to cure or to alter the underlying disease. Other interventions such as surgery, chemotherapy, or radiotherapy are usually intended to cure the child’s medical problem or provide extended remission of disease.7 Decisions about these interventions sometimes arise in emergency situations, but they ordinarily follow a less urgent assessment of the child’s condition. Although the underlying ethical, legal, or clinical issues may be similar for decisions in both situations, the bias “to do something” (meaning something beyond providing comfort) seems most evident in an emergency.
Decisions About the Criteria for Decisions
Many ethical and legal criteria for decisions about end-of-life care are well accepted, although disagreement may exist about how to apply and weigh them in specific situations. Two criteria have, however, aroused particular controversy. One criterion involves the consideration of quality of life, particularly in decisions made on behalf of a patient who is not able to describe his or her own experience or preferences. The other criterion involves the consideration of organizational or societal costs in treatment decisions.
Quality of Life as a Criterion in Decisions
At both the societal level (when the well being of populations is the focus) and the clinical level (when the well being of individual patients and their families is the focus), there is more to health care decisionmaking than preserving life. As expressed by the American Academy of Pediatrics, “the goal [of palliative care] is to add life to the child’s years, not simply years to
the child’s life” (AAP, 2000g, p. 353). When physicians and others are identifying, explaining, and weighing the potential benefits and burdens of different courses of care, the implications for quality of life—not just the quantity of life—need attention. Chapters 4 and 9 and Appendix C discuss concepts and measures of quality of life (and their limitations) in more detail.
Judgments about quality of life involve both factual issues (e.g., what do data and experience suggest about a child’s prognosis with or without the treatment in question) and subjective elements (e.g., when has a child’s suffering become so great that it outweighs any benefits of the treatment). For infants and very young children as well as older children who are cognitively limited or comatose, judgments will rely on parents’ or others’ values and assessments.
Although ethicists, theologians, and clinicians may disagree among themselves about issues such as assisted suicide or euthanasia, most have come to agree that palliative actions intended to relieve suffering (i.e., protect the quality of life) are justified even when one unintended consequence or effect may be to hasten death (i.e., limit the quantity of life) (see, e.g., Cherny and Portenoy, 1994; Fleischman, 1998; Jonsen et al., 1998; Sulmasy, 2000; but see also Beauchamp and Childress, 1994, for a review of critiques of such arguments). Many professional organizations have also endorsed this perspective (see, e.g., AMA, 1994a; AAP, 2000g). As discussed in Chapter 5, the practice of terminal or palliative sedation involves this kind of “double-effect” reasoning.
Still, the role of quality-of-life considerations continues to prompt discussion (see, e.g., Kuhse, 1987; Beauchamp and Childress, 1994; Pellegrino, 2000). For example, are life-sustaining measures ethically or legally required for a permanently comatose infant? What about infants with profound impairment short of this state? Some take the position that, absent a double-effect situation, it is never acceptable to forgo life-sustaining treatments based on quality-of of-life considerations, regardless of family wishes. What are the obligations of clinicians to patients and families when unwanted interventions leave a patient profoundly impaired cognitively and physically? As described below, controversy over these questions has in a few instances prompted litigation and even legislation that has sought to preclude or restrict clinicians and families from considering quality of life in decisions about infants with severe birth defects.
Resources as a Criterion in Decisions
As health care costs have escalated since the 1960s, clinicians and others have become increasingly concerned about real or potential conflicts between clinicians’ responsibilities to individual patients and their obliga-
tions to support the prudent use of limited societal or institutional resources. One view is that, by and large, so much nonbeneficial treatment exists that eliminating it will preclude the need to limit or ration beneficial care. An alternative perspective is that control of health care costs also requires decisions to limit or ration care that is thought to offer some benefit but not enough to warrant its provision, given higher priorities. Disagreements often arise, however, over the judgment that a service is not beneficial.8 Many clinicians worry about the appropriateness of “bedside rationing.” 9 The traditional consensus in the clinical community has been that resources should not enter into physicians’ judgments about individual patients unless those judgments are governed by clear principles and procedures that follow public deliberation, reflect general community consensus, and are consistent with available scientific knowledge (see, e.g., Angell, 1985, Sulmasy 1992; AAP, 1996).
Ethical analyses may focus on obligations at an individual or a collective (organizational or societal) level. Regardless of the level of analysis, cultural diversity and sensitivity to individual and family differences must be factored into discussions, decisions, and practices.
Ethical Obligations at the Individual Level
At the individual level, ethical analyses generally focus on the obligations of clinicians, first, to individual patients and, second and less often, to those close to or responsible for a patient. Most analyses of clinical decisionmaking devote little attention to parental responsibilities, focusing instead on the responsibility of clinicians and others to guide and redirect those parents who are viewed as acting against their child’s best interests.
Notwithstanding children’s stage of development or legal status, most
issues related to their care can be considered within the same broad ethical frameworks that are widely applied to care for adults. During recent decades, general agreement has emerged on the core ethical obligations of clinicians to patients (see, e.g., Beauchamp and Childress, 1983; President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research [hereafter, President’s Commission], 1983; Jonsen and Toulmin, 1988; McCullough, 1988; Brennan, 1991). In common parlance,10 these obligations are to
do good (e.g., by relieving a child’s pain or providing emotional support),
avoid doing harm (e.g., by not providing life-sustaining treatments that impose burdens on a child without benefit),
respect patient autonomy (e.g., by generally attempting to consider care “through the child’s eyes” even though young children have no legal autonomy), and
treat patients equitably (e.g., by seeking to provide or arrange needed care for children regardless of a family’s ability to pay).
These ethical obligations may sometimes be in conflict and may not be equally weighed in practice. For example, a single action may involve both benefits and harms that have to be balanced in view of the patient’s overall circumstances and understood preferences. The obligations to avoid harm and to respect autonomy have been exhaustively considered in debates about forgoing life-sustaining treatment. In contrast, one recent study of such situations concluded that the value of providing effective palliative care when life-sustaining treatment is forgone “remains underanalzyed and needs more rigorous examination” (Burns et al., 2000, p. 3060).
Other obligations or other formulations of clinical and, more generally, medical ethics exist.11 Ethicists may, for example, cite the obligations of
“fidelity” and “professional integrity.” Fidelity is the responsibility of health care professionals to place the interests of their patients first. Professional integrity can be viewed as a broad obligation of physicians to act ethically in all their relationships and to be faithful to their moral values when they are challenged (see, e.g., Beauchamp and Childress, 1994; Alpers and Lo, 1999).
The clinician’s primary obligation is to his or her patient, and ethical discourse has focused on this obligation. In pediatrics, the principle that the “child and family are the unit of care” raises the issue of the clinician’s obligations to parents and other family members. These obligations do not appear to have been systematically interpreted using the ethical framework outlined above, although the potential for conflict between a child’s interests or preferences and a family member’s interests or preferences has been discussed (see, e.g., Randall and Downie, 1999).
A few observations about clinical ethics in this context can be offered. First, the communication strategies discussed in Chapter 4 and the consensus building strategies described below should usually help clinicians to do good for parents (e.g., by helping parents to feel they have done their best for their child); avoid harming parents (e.g., by helping them avoid choices they will later regret); and treating parents fairly (e.g., by providing understandable information and generally respecting their values). Second, in most situations, when clinicians fulfill their obligations to a child of doing good, avoiding harm, and treating fairly, they are doing likewise for parents because parents usually want to advance their child’s best interests. Third, in rare cases, as discussed further below, clinicians may conclude that doing good for the child requires opposing parental preferences and values.
Most discussions of clinical and medical ethics focus on the obligations of individual professionals rather than care teams. As discussed further below, different members of a child’s care team may sometimes have different views of what constitutes ethical care. Although physicians usually have the legal and professional authority to prevail, the persistence of unresolved and unacknowledged conflict can compromise the implementation of decisions, damage team members, and subtly (sometimes explicitly) undermine patient and family trust in team members. Some conflicts may reflect individual values, characteristics, and personalities; others may be rooted in different professional socialization and norms.
Ethical Obligations at the Collective Level
Clinicians may sometimes perceive their obligations to their patients individually to be in conflict with their obligation to do good or avoid harm for patients collectively. To cite a dramatic but atypical example, in emergency departments and critical care units, an unusual surge in injured pa-
tients may outstrip available personnel, space, and equipment. In response, clinicians and others have established triage protocols to guide decisions about how to allocate—that is, ration—treatment resources in such situations. Appendix B describes how prognostic tools (e.g., assessments of a patient’s likelihood of survival with or without treatment) have been developed to aid in formulating and applying such protocols (and for other purposes).
Debates about the appropriate use of limited resources arise across the spectrum of clinical decisions. Some involve routine elements of patient care, for example, how much physicians (and certain nonphysicians) should be paid for informing and educating patients compared to performing surgery or other procedures.
Once again, however, some of the most widely publicized disagreements about resource use focus on care for newborns who have conditions that—even with treatment—are incompatible with extended life or with neurological function beyond the most primitive level (Fost, 1999). One highly publicized case in the 1990s, that of “Baby K,” raised questions about the limits of parental demands on community resources—and on providers who are convinced that such demands violate their professional ethics and integrity (16 F 3d 590, CCA 4, 1994; Glover and Rushton, 1995). In this case, which is discussed in more detail below, the mother of an anencephalic child (who lacked all parts of the brain except the brain stem) insisted on repeated resuscitation. Federal trial and appellate courts held that emergency care could not be denied under the federal Emergency Medicine and Active Labor Act. (See, e.g., Clayton, 1995; Brown, 1996; and Maragakis, 1996, for critiques of this interpretation of the legislation, which was passed to prevent emergency department “dumping” of uninsured patients.)
Often, physicians may join with parents in a desire to “do everything” to prolong the life of a gravely ill child, including providing some treatments of marginal or no benefit. They may argue that such care provides emotional comfort and hope to the parents and allows all involved freedom from any guilt they might experience had they not pursued every option. In other cases, physicians may resist such treatment because they believe it harms the child, violates their clinical values, and misuses limited community resources.12
Indeed, the effort to define a quantifiable concept of “futile” treatment
has been driven in considerable measure by physicians who want more protection against family demands for treatments with no or virtually no potential for benefit (see, e.g., the perspectives in Zucker and Zucker, 1997; see also Schneiderman et al., 1990; Rushton and Hogue, 1993; IOM 1997; Avery, 1998; Goldstein and Merkens, 2000). While accepting the concept that medical treatments are futile in certain situations, some ethicists, clinicians, and researchers have concluded that the term is so variably and imprecisely used that, in general, it ought to be avoided (Beauchamp and Childress, 1994).13 Others have concluded, further, that it is impossible to craft a precise operational definition of futility that can reliably and validly govern subsequent real-world clinical decisions (see, e.g., Lantos et al., 1989; Truog et al., 1992; Waisel and Truog, 1995).14 Rather, assessing futility is a judgmental process requiring “judicious balancing” of effectiveness, benefit, and burden specific to each patient and treatment (Pellegrino, 2000). In addition, some argue that proposals to limit futile treatment may mask “prejudices about those who are disabled, who come from disadvantaged social groups, or who are dying” (AAP, 1996, p.150; see also Krakauer and Truog, 1997).
Efforts to define futile care have also been motivated by the expectation or hope that the application of such a definition in practice could help control health care costs. Two studies of pediatric intensive care in single institutions concluded that only a small percentage of patients met any one of several definitions of futility and that their care generally involved relatively limited resources (Sachdeva et al., 1996; Goh and Mok, 2001).
The allocation of resources for medical care is just one element of a broader set of issues of about the allocation of community resources and responsibilities. Families of a seriously ill or injured child often must shoulder heavy financial burdens, sometimes extending to job loss, home loss, bankruptcy, and homelessness. Even if they have private health insurance, it will usually not cover all the home health care and nonmedical home care and other services needed by a child and family. Medicaid and other gov-
ernment programs may cover more services but, in most cases, only after a family’s resources are virtually exhausted.
Ethics in Practice
Notwithstanding the moral passion and intellectual energy devoted to ethical analyses and positions, the greater challenge may be in matching practice to principles, even when the principles are, by and large, not in dispute. Financial constraints and time constraints may undermine ethical practice, for example, when clinicians face appointment schedules that leave them little time for the kind of consensus-building discussions described below.
Violations of ethical practice and avoidable patient and family suffering also may arise from missteps involving what should be routine organizational procedures and actions. For example, despite years of attention, patients and families cannot be guaranteed that decisions made in advance about the use of life-sustaining medical treatments will be honored because information about advance directives may not follow the patient, for example, during a transfer from home to hospital. Such failures prompted the Oregon effort, described in Chapter 6, to develop tools and implement procedures that have increased the likelihood that patients will have their wishes honored (Tolle et al., 2002).
In some cases, health care professionals and health care organizations may be constrained by licensure or other regulatory restrictions from providing care they believe appropriate. For example, hospices may be restricted not just by Medicare and Medicaid coverage limits but also by licensure from providing a broad range of palliative services to all adults and children in need. Eliminating or revising these restrictions would take state and national action.
Further, despite various educational efforts, it is clear that physicians and other care team members sometimes misunderstand both the evidence base and the ethical context for life-support interventions and are not properly prepared—intellectually or emotionally—to inform and advise patients and families. Even with a correct understanding of science and ethical principles, some clinicians may lack the skills and attitudes to provide compassionate and effective communication with seriously ill children and their families. Education is not sufficient to change practice but it is necessary to provide requisite skills and it can shape attitudes and values.
Consistent with studies of care for adults, studies of pediatric care have documented considerable variation in attitudes about and use of life-sustaining technologies in real or hypothetical situations (see, e.g., Levetown et al., 1994; Randolph et al., 1997, 1999; Keenan et al., 2000; see also Asch et al., 1995; Fins, 1999; Breen et al., 2001; Puntillo et al., 2001). Variations in
attitudes can lead to variations in decisions and practices, although variations in knowledge, habits, and organizational protocols also may contribute. Patients and families may, as a result, be unable to expect consistency in care, fairness, and freedom from avoidable suffering.
Creating organizational and societal environments for ethical practice requires persistent, cooperative efforts to change virtually every dimension of medical care—including professional education, arrangements for institutional governance, mechanisms for professional collaboration, patient-clinician decisionmaking, information and performance evaluation systems, and financing principles and policies (see, e.g., Rushton and Brooks-Brunn, 1997; Brodeur, 1998; Jonsen et al., 1998). Although not identical in purpose or method, the principles and strategies for integrating ethical principles into organizational culture and routine professional practice have much in common with those guiding many of the continuous quality improvement initiatives discussed in Chapter 6.
The following section focuses on strategies for resolving conflicts. Some of these strategies, for example, the development and successful application of clinical practice guidelines, also can help in transforming statements of principles into effective practice.
STRATEGIES FOR PREVENTING OR RESOLVING PARENT– CLINICIAN AND OTHER CONFLICTS ABOUT CLINICAL CARE
As noted earlier, conflicts cannot always be avoided. Conflicts may even be productive or beneficial in some situations, for example, when parents pursue an issue in disagreement rather than capitulating to a course of action and later regretting their silence. The same point may hold when members of the health care team challenge and argue with each other. In the face of disagreement, overly confident or assertive clinicians may become more sensitive to parental concerns and values. Nonetheless, extended or severe conflict about care of a gravely ill or injured child may be very destructive to all involved and may subject children to needless suffering, for example, as clinicians and family argue about treatments.
Strategies to prevent or resolve conflicts about clinical care can operate at the individual level. That is, they come into play when a particular dispute arises or the potential for dispute exists, but they do not establish general rules or formal precedents to govern subsequent cases. Conflict resolution strategies can also operate at the organizational or system level, for example, in the form of ethics committees or ethics consultation protocols. In addition, research can be useful in preventing or resolving disputes, for example, by clarifying the benefits and burdens of specific medical interventions in different clinical situations. Although some efforts have been made to document the implementation and results of strategies to
prevent or resolve conflict, more research is needed to assess their strengths and limitations and guide adjustments to produce better results.
Continued Discussion and Involvement of New Parties
At the individual level, attempts to resolve conflicts involving specific clinicians, parents, and sometimes children typically involve continued discussion of the situation. Following the identification of a conflict, new participants may become involved in the discussion. Thus, a child’s care team may meet as a group with the family and perhaps the child to discuss a disagreement that has arisen during discussions involving the family and the child’s main physician. Sometimes, during such a team meeting, a nurse, social worker, or other physician may recognize that despite an individual physician’s best efforts to communicate clearly, the family lacks or misunderstands crucial clinical or other information and terminology. The conflict, in essence, turns out to be not so much about differences in values as about differences in information or understanding of the “facts” of a situation or differences in the way language is understood. When organizations make family conferences or protocols for counseling routine elements of pediatric care, they become system-level strategies for preventing or managing conflict (see, e.g., Hansen et al., 1998; Curtis et al., 2001).
Team conferences without family members present may also be employed to deal with conflicts among care team members, for example, when physicians and nurses disagree about the use of a life-sustaining intervention. If such conflicts are routine, then system-level approaches may be appropriate and could include cooperative development of practice guidelines based on systematic assessments of evidence and consensus-building processes.
Counseling and Consultations
Some conflicts may be less about facts or values (e.g., whether it is ever morally acceptable to withhold medically provided hydration) than about emotional issues of power, loss, distrust, guilt, anger, or fear of abandonment. Psychological counseling (usually involving the patient or family but sometimes involving a physician or other clinician) may defuse these emotions and, thus, the conflict.
Sometimes, ethics consultants (see discussion below of ethics committees) are brought in to help resolve conflicts through continued discussion or, at minimum, to help provide a framework for the further discussion. Sometimes discussions may focus on information and exploration of val-
ues, but they may also uncover and address issues of trust, fear, anger, or other emotions, including personality conflicts.
Consensus Building Techniques
At the request of the American College of Physicians–American Society for Internal Medicine (ACP-ASIM), three experts in adult palliative care proposed a consensus-based approach to decisionmaking for those who are unable to make decisions about their own care (Karlawish et al., 1999; see also Hoffman, 2001). The authors also offer suggestions about how to handle situations when discussion has not resulted in a consensus about the care of a patient who cannot make decisions about his or her own care. These suggestions, which should be tested further, include the following:
postponing decisions and recommending that those involved take more time to think about and discuss concerns and goals;
seeking interim steps such as a time-limited trial of a medical intervention rather than insisting on an all-or-nothing decision;
continuing to identify and understand each participant’s views on the goals of medical care for the patient and the care options for achieving those goals;
bringing in a trusted third party such as an ethics or palliative consultant or religious adviser; and
avoiding language or actions that personalize conflicts, turn decisionmaking into a power struggle, or attack the religious, cultural, or other values of the participants.
Regardless of the specifics, discussion strategies place a premium on communication skills and advance care planning as discussed in Chapter 4. Poor communication skills as well as failures of empathy and compassion are undoubtedly behind some of the disputes with parents that require interventions of the kinds discussed here. Based on the guidance developed from the ACP-ASIM, Table 8.1 presents examples of discussion steps and illustrative language that can be employed to guide discussion toward consensus. Because the original guidance focused on adult care, the text has been slightly altered. Again, further assessment of these strategies is desirable.
Discussions of the kind outlined in Table 8.1 take time, which is often in short supply as clinicians respond to health systems, hospitals, hospices, private insurers, Medicaid programs, and other state programs that are trying to control costs. Nonetheless, investments in careful initial communications with families can help limit subsequent investments in discussions
TABLE 8.1 Structuring Discussions to Reach Consensus About Care for Patients Who Lack Decision-Making Capacity
Identify the main participants
“We need to make some decisions about the care of [child’s name]. Is everyone here who could help us think through what we should do?
Invite the participants to narrate
“Can you tell me how she’s changed, how how the patient has reached this things have gone for all of you?” Or “I stage know I’ve been caring for [child’s name] for some time, but it helps me if you can tell me how she’s changed, how things have gone for each of you.”
Teach the decisionmakers about the expected clinical course of the patient’s disease
“[Child’s name] has an incurable, progressive, and ultimately fatal disease. I can’t say for sure when she’ll die . . . but given [the situation], we shouldn’t be surprised when she does.”
Advocate the patient’s quality of life and dignity
“We ought to care for her in a way that makes us confident that after she’s gone, we can say we did the best for her.”
Provide guidance on the basis of existing data and clinical experience
“For patients like [child’s name], feeding with a tube does not significantly reduce the risk for pneumonia. On the basis of my experience, a speech therapist may give us some useful hints on ways to feed her that will allow her to continue to eat by mouth.”
SOURCE: Adapted from Karlawish et al., 1999, for the American College of Physicians— American Society of Internal Medicine. Used with permission.
to provide fuller information, correct misunderstandings, and defuse conflicts over goals of care. In certain situations, they may also reduce the potential for expensive litigation. Nonetheless, when capitation payments for physician services or fee-for-service payments for physician office visits are unreasonably low, as is often the case with Medicaid, and when payment policies limit or preclude payment for counseling, team conferences, and other kinds of communication, financial incentives clearly do not support the communication strategies described here and in Chapter 4. Thus, just as research, professional education, and organization structures have to support good clinical practice, so must financing policies.
Organization- or System-Level Strategies
The discussion in the preceding section focuses on the resolution of disputes at the individual level but also points to the need for organizational- and system-level responses to prevent or resolve conflicts. If distrust related to ethnic or religious differences appears to be at the heart of some disputes, are there organizational or community sources of distrust or misunderstandings that should be investigated and responded to by health care organizations and their communities? If misunderstanding of facts or terminology is a consistent problem, can training programs and protocols be created to help clinicians communicate more successfully or might public education programs be helpful?
Ethics committees and similar groups constitute a system-level effort to assist in the resolution of disputes about clinical care (see, e.g, Fletcher, 1991; AAP, 1994b, 2001d; AMA, 1984, 1994b; Dugan, 2001). They developed as hospitals struggled with the increasingly difficult questions that advances in medical technology have created. In 1983, the President’s Commission endorsed the creation of hospital ethics committees. Such committees are now part of the institutional fabric of most American hospitals. The Joint Commission on the Accreditation of Healthcare Organizations (JCAHO) accepts such committees as one approach to meeting certain of its requirements, for example, having a “functioning process to address ethical issues” (JCAHO, 1998, Standard RI.1). A number of organizations have set forth guidelines for the creation, composition, and operation of ethics committees. Box 8.1 presents the recommendations of the AAP.
To a considerable degree, ethics committees have focused on articulating processes and providing consultative resources for resolving disputes without necessarily establishing guidelines or rules to address the substance of such disputes. For example, committee consultants may be useful in identifying when a persistent conflict in values may be best handled by transferring responsibility for a child’s care to another physician or trying to arrange for the child’s transfer to another institution. More generally, they may set forth processes for attempting to determine when a dispute is grounded in misunderstanding of facts or terminology and when it is grounded in true differences in values about the goals of care or the benefits of treatments. Box 8.2 summarizes the elements of such a process.
Ethics committees may also propose and try to develop professional consensus on institutional policies for specific issues such as DNR orders. In addition, committees usually have an educational role in teaching other
SOURCE: AAP, 2001d.
SOURCE: AMA, 1997, Policy E2.037.
members of the hospital community about the application of ethical concepts in clinical situations.
A number of questions have been raised about the goals of ethics committees and their ability to meet the goals set for them (see, e.g., Povar, 1991; Nelson and Shapiro, 1995; Gillon, 1997; Casarett et al., 1998; Mitchell and Truog, 2000). Some questions involve who (e.g., nurses, social workers) can bring questions to these committees. Others concern whether committees have sufficient resources and expertise to provide ethics consultations.
Questions about the independence of ethics committees have also arisen. In an environment of fiscal stress and resource restrictions, ethics committees run the risk of becoming and being perceived as partisan institutional agents (see, e.g., Kelly et al., 1997; Howe, 1999).
Draft standards for ethics consultations have been developed by a task force that included the Society for Health and Human Values, the Society for Bioethics Consultation, and several other organizations (SHHV/SBC, 1998). Research on the consequences of ethics consultations or ethics committee involvement in decisions about patient care is limited, but some studies suggest a positive role (Dowdy et al., 1998; Schneiderman et al., 2000). Further research may guide these committees by identifying individual- and system-level factors that affect family perceptions about end-of-life decisions. For example, in retrospective interviews, Abbott and colleagues (2001) found that family members identified several sources of comfort and support, including pastoral care, prior discussion of treatment preferences, and lenient visiting rules.
Protocols for Communication and Decisionmaking
Organizational- or system-level strategies for preventing and resolving conflicts take varied forms. Box 8.3 summarizes one example of an organization-level strategy to reduce clinician–family conflict in the intensive care unit (ICU) by improving teamwork and communication with families of patients nearing death. Clinicians, administrators, and ethicists developed the approach in part to reduce the potential for certain troubling ethical dilemmas and in part to reduce institutional exposure to liability based on patient or family dissatisfaction with care. An evaluation of this process is under way.
The discussion below considers two other organizational or system approaches to preventing or resolving conflicts about clinical care: research-and evidence-based practice guidelines and public policy (litigation and legislation). In addition, although individual clinicians are obligated to consider how they may improve their skills and attitudes, corresponding action at the organizational and system levels also must be considered, for ex-
In combination with other strategies to either prevent or manage conflicts, several adult intensive care units associated with Harvard University have been cooperating to develop and implement a strategy to reduce conflicts between clinicians and families about aggressive use of life-sustaining technologies for patients nearing death. They began with a period of intensive data collection—both personal interviews and surveys—to better understand the experience of patients and families and to identify their characteristics, concerns, and areas of dissatisfaction. Based in part on this information and on information from physicians and nurses, they then developed a four-step process to prevent conflicts.
Other elements of the strategy include developing palliative care guidelines and developing a procedural approach to assessing when care is “futile.”
SOURCE: Summarized from Truog, 2000. Quoted material used with permission.
ample, in the form of continuing education programs, changes in undergraduate and graduate medical education, competency testing, and funding to support these initiatives.
As discussed throughout this report, the procedures and policies of managed care organizations, insurers, and cost-conscious hospitals or other providers can contribute to conflicts and misunderstandings. System-level finance reform is essential, although it will never provide all the resources that health care professionals and families want. Some conflict over resources is inevitable.
Strengthening the Knowledge Base for Decisionmaking
Better scientific knowledge will neither prevent nor resolve all debates and disputes. For example, in the Baby K case mentioned earlier, clinical facts were not central to the mother’s insistence on repeated resuscitation of her infant with anencephaly. Likewise, as discussed in Chapter 6, efforts to synthesize scientific knowledge in the form of guidelines for clinical practice are important but implementation of guidelines cannot be assumed in the face of contrary institutional, financial, and cultural influences (Rushton and Brooks-Brunn, 1997).
Nonetheless, doing good is easier when practice is guided by research demonstrating what works and what does not work to produce desired outcomes—whether those outcomes are cure for a disease or relief from suffering. Scientific research can also help defuse some controversies by challenging contradictory factual premises (explicit or implicit) on different sides of a dispute.
For example, in the United States, the 1980s saw considerable discussion of the appropriateness of correcting physical defects in newborns with significant mental retardation and other severe physical deficits. In one notable case, after parents would not approve surgical repair of a correctable defect in a child with Down syndrome and after a judicial challenge to their decision failed, politicians responded with the so-called Baby Doe regulations, toll-free telephone hotlines for people to report similar cases, and subsequent legislation to require treatment of handicapped newborns except in specified situations (see, e.g., Pless, 1983; Lantos, 1987; AMA, 1992; Weir, 1992; Caplan et al., 1992). Since this episode, research has indicated that many affected infants have a better prognosis than clinicians and parents previously assumed (Teddell et al., 1996; State et al., 1997; Amark and Sunnegardh, 1999). That is, the infants often have reasonably good prospects for many years of dependent but apparently enjoyable life.
Clinical practice guidelines or protocols represent one focused system-level strategy to create a credible, authoritative, evidence-based framework to guide individual patient care decisions (IOM, 1990a, 1992). Procedures vary in rigor and credibility, and methods for developing guidelines continue to be debated and refined. Generally, the more rigorous processes for guideline development bring together clinicians, methodologists, and sometimes consumers, ethicists, and others to define the issues at stake, identify and evaluate relevant scientific evidence or facts, and set forth statements about appropriate care that are based on an explicitly described combination of evidence, clinical judgment, and values.
Recent international guidelines on cardiopulmonary resuscitation, for example, assessed a considerable body of research that attempts to link resuscitation outcomes (e.g., survival to hospital discharge, neurological
function) to patient characteristics (e.g., diagnosis, age). These guidelines note that cardiac arrest in children, unlike adults, is uncommon and is rarely a sudden event. It typically results from other than a primary cardiac cause and often is the final event associated with progressive shock or respiratory failure related to trauma, respiratory or neurological disorders, sepsis, or unexplained causes (sudden infant death syndrome [SIDS]) (AHA, 2000a). Survival is uncommon, and children who survive are often neurologically devastated. The guidelines note these dismal outcomes and urge the development of a consensus definition of when resuscitation would be futile. The guidelines mention only two specific circumstances in which not initiating resuscitation is indicated: (1) when patients have a clear advance directive asking health care workers not to begin resuscitation in the event of a cardiac arrest and (2) when patients show signs of irreversible death such as rigor mortis, decapitation, and dependent lividity or postmortem hypostasis (purple coloration from pooling of blood in dependent body areas) as agreed upon by a consensus of the medical community. The guidelines emphasize the lack of rigorous research to support many common elements of resuscitation and identify many areas for further research.
To the extent that research helps to reduce disagreement among clinicians, this will be a benefit in itself. Further, reducing disagreements among clinicians should also reduce clinician–patient or clinician–family conflicts that are stimulated or reinforced by evident variations in clinician views and practices.
Throughout this report, the limited knowledge base for much pediatric palliative and end-of-life care is documented. Chapter 10 includes recommendations and directions for research to strengthen the knowledge base for effective palliative and end-of-life care for infants, children, and adolescents. The recommendations focus on knowledge to improve clinical care, but better knowledge is also important to inform ethical and legal decisionmaking.
Litigation and Legislation
Disputes about which values should prevail in a patient’s care may end up in court when other conflict resolution approaches fail and death or another outcome has not intervened to make the conflict moot. The judicial system provides a socially sanctioned process for resolving individual disputes within a framework of statutes, regulations, and case law (i.e., precedents established by prior decisions). As courts have increasingly faced disputes that involve highly technical issues, judges and policymakers have struggled with questions about the ability of judges and juries to understand and weigh scientific and technical information, often presented by experts who differ in their presentation and assessment of this information.
In general, litigation tends to be a costly, disruptive, and unpredictable vehicle for resolving conflicts that is stressful for all parties.
Statutes or administrative regulations constitute a different kind of system-level response to controversies about clinical decisionmaking. The development of statutes or regulations may or may not take scientific evidence into account and may or may not attempt to reflect or create clinical or community consensus about an area of disagreement. For example, state laws about adolescent decisionmaking “form a patchwork quilt of rights and limitations” that neither reflects nor contributes to a coherent view of adolescent capacity to make medical decisions (Oberman, 1996, p. 127).
Oregon offers an example of the rare jurisdiction that set out—not without problems and controversies—to employ careful and explicit strategy to (1) develop community consensus on priorities for medical care; (2) use clinical research and judgment to assess the relative benefits of treatments for common medical problems; (3) cover the most beneficial treatments within predetermined spending levels; and (4) expand health coverage to more people using the savings from reductions in services of marginal or no benefit (see, e.g., Bodenheimer, 1997; IOM, 1997). The priority-setting framework explicitly included comfort care (e.g., hospice, pain management) among the essential services (Cotton, 1992). Oregon’s strategy has provoked continuing political, ethical, and analytic debates as well as legal challenges that have limited its application to and beyond the state’s Medicaid program. State budget problems have limited the move toward universal coverage (see, e.g., Rojas-Burke, 1999). Despite, or perhaps as a result of, the attention paid to the Oregon approach, it has not been replicated by other states.
Debates about rationing care often focus on expensive versus inexpensive health care services. The real issue, however, is not the expense per unit of service but, rather, the expense per unit of benefit (for example, years of life or days free from pain). High-volume services with low unit costs are less dramatic but not necessarily less important than very expensive, low-volume services. Systems must inevitably make trade-offs among alternative ways of using available resources to benefit large groups (their members), and different systems have made different choices. Some of the implicit means of rationing potentially beneficial services rely on price or inconvenience. The next section discusses legal issues related to several kinds of disputes that can arise in the treatment of children.
Although the committee views litigation as a last resort in cases of conflict about care for children with life-threatening medical problems, situations will arise that make recourse to the courts appropriate or un-
avoidable. Litigation often takes a significant emotional and financial toll on all involved and can result in decisions and precedents that have unanticipated repercussions far beyond the original case. The following discussion reviews various legal issues related to decisions about medical care for children.
Parental Refusal of Treatment
As medical care becomes more complex, so do the types of legal problems arising out of parental refusals of treatment (Holder, 1983). Some refusals stem from religious convictions; for example, some Jehovah’s Witness members object to blood transfusions for children as well as adults (Stanfield et al., 2000). Other refusals are based on a parental view that a child with a serious illness (such as cancer) should not be subjected to the side effects of treatment and that “alternative therapies” such as laetrile offer as much benefit as chemotherapy without the side effects (see, e.g., Schiff v. Prados, 112 Cal Rptr 2d 171, 2001; Marshall v. Sackett, 907 SW 2d 925, Tex 1995; Green v. Truman, 459 F Supp 342, DC Mass 1978; see also Faw et al., 1977; Horwitz, 1979). Some parents, who accept the fact that their child is probably dying, want to stop painful or other unpleasant treatments to prolong the child’s life in an effort to provide as much peace as possible (Gerben v. Holsclaw, 692 F Supp 557, DC ED Pa 1988; see also Nealy, 1995).
Court intervention, based on the legal principle that failure to obtain adequate medical care for a child is a violation of state child neglect laws, is always an option if physicians consider it appropriate. Until recently, when a physician testified that a child would die or be permanently disabled if treatment were not provided at once, the court order would be automatic. This is not so uniformly true today, but it is still clearly the case in situations where parents refuse treatments that would likely cure or substantially modify the course of the child’s illness.
Religious conviction is never a defense for a refusal to provide medical care for a child (see, e.g., In the Matter of D.R., 20 P 3d 166, Okla Civil Appeals, 2001; Hoang v. State, 250 Ga App 403, 2001; AAP 1997d). For example, if a 3-year-old child of a Jehovah’s Witness is in an automobile accident, needs a blood transfusion to which the parents object, and is expected to recover completely, a court order would be issued in any court in the country (Jehovah’s Witnesses of Washington v. King County Hospital, 390 US 598, 1968). Similarly, if a child had acute appendicitis but his parents refused to consent to any medical care at all, a court order would be issued. When the need for life-saving treatment is urgent, a physician or
hospital that provides such treatment without a court order and over parental objection will not be liable, as long as it is clear that there was no time to apply for and receive a court order (HCA, Inc. v. Miller, 36 SW 3d 187, Tex App 2000).
Failure to provide medical care to a child can be prosecuted as a criminal offense. Convictions of manslaughter and even murder (Commonwealth of Pennsylvania v. Nixon, 2000, 563 Pa 425, 761 A 2d 1151, 2000; Commonwealth of Pennsylvania v. Barnhart, 345 Pa Super 10, 497 A 2d 616, 1985) have been upheld when a child died without medical care. In these cases, the state must prove by “substantial medical evidence” that the illness or accident would not normally have been fatal if the child had been treated with usual and appropriate interventions.
When cure for a child’s medical problem is unlikely or impossible, however, courts have been increasingly willing to allow parents to make decisions based on their subjective analyses of risks and benefits (Newark v. Williams, 588 A 2d 1108, Del 1991; In the Matter of Matthews, 225 A.D. 2d 142, 650 NYS 2d 373, 1996). If another physician can be found who agrees with the parents and testifies that she or he will assume responsibility for the care of the child after the original physicians declined to follow the parents’ wishes, all courts will permit parents to remove their child to the other physician, even if the original physicians are convinced that the new physician’s therapies are outside any accepted medical standard (In re Hofbauer, 47 NYS 2d 648, 419 NYS 2d 936, 393 NE 2d 1009, 1979). If a child’s life is not immediately threatened, even if the underlying condition is desperately serious, most courts’ longstanding practice is to refuse to order that high-risk therapies be given over parental objection (In re Hudson, 1955, 13 Wash 2d 673, 126 P 2d 765, 1942; In re Seiferth, 309 NY 80, 127 NE 2d 820, 1955).
It is one thing to treat a child in the hospital under court order when the child will then leave the hospital to return to normal life at home. It is quite another to obtain a court order to administer chemotherapy or other treatment over parental objections when the parents’ cooperation is required to bring the child for continuing outpatient services. If their objections are sufficiently adamant, they can leave the state with the child (In re Chad Green [In re: Custody of a Minor], 379 NE 2d 1053, 393 NE 2d 836, 1979). This raises the prospect that if a court order is obtained, the child will have to be either hospitalized during the entire period or removed under court order from his parents and placed in foster care. To inflict this on a family and a child at a time of suffering and perhaps impending death is something that should be considered only under the most extraordinary circumstances and when there are strong reasons to believe that the achievable goal is long-term remission or a potential cure.
In sum, a physician confronted with parents who, for whatever reason, are “uncooperative” may assume correctly that she or he can get a court order to treat the child. The ultimate question is whether such a step will, on balance, benefit the child. In many situations, it will not.
Parental Insistence on Treatment
The converse physician–parent conflict arises when the physician feels that further aggressive life-prolonging interventions for a dying child are futile or will cause suffering in excess of any potential benefit and the parents refuse to accept that decision (Paris et al., 1990). Some but not all of these conflicts arise when parents, for religious or other reasons, will not accept the concept of brain death.
If a patient (child or adult) is, by standard medical criteria, brain dead (see Chapter 1), no permission is required to pronounce the patient dead (see, e.g., Lovato v. District Court, 198 Colo 419, 601 P 2d 1072, 1979; Alvarado by Alvarado v. New York City Health and Hospitals Corporation, 145 Misc 2d 687, 547 NYS 2d 190, 1989). Even if the family objects, the physician may sign the death certificate and then remove the respirator (thus demonstrating that the patient was dead before life support was removed) (see, e.g., Law v. Camp, 116 F Supp 2d 295, 2000, 2001 WL 868354, CCA 2, July 27, 2001).
In some circumstances, however, physicians may find it advisable to obtain a court order before terminating life support over parental objections. One of these circumstances is when the child’s condition is or may be the result of abuse inflicted by the parent (In re L.H.R., 253 Ga 439, 321 SE 2d 716, 1984; In re M.D., 758 A 2d 27, DC CA 2000.). When a parent will face a murder or manslaughter charge as soon as the child is pronounced dead, he or she will almost never consent to withdrawal of life support (AAP, 2000e). Since the circumstance of pronouncing the child dead will be the major issue in the parent’s trial, it is important to have a clear record of the circumstances, including a judge’s order to terminate life support—even if the child meets all criteria for brain death (Truselo v. Carroll, 2000 WL 33324536, Del Fam Ct 2000; In re Tabatha v. Ronda R., 252 Neb 687, 564 NW 2d 598, 252 Neb 864, 566 NW 2d 782, 255 Neb 818, 587 NW 2d 1909, 1998; In re Haymer, 115 Ill App 3d 349, 450 NE 2d 940, 1983; see also Massie, 1993; Fleming, 1999).
In most litigation involving parental insistence on treatment, however, the child is not brain dead but is in a persistent vegetative state. The physician feels that in the absence of any hope of recovery, continued maintenance of the patient on life support is a mistake. In contrast, the parents are simply thankful their child is “not dead” and refuse to allow termination of
life support (see, e.g., Velez v. Bethune, 219 Ga App 679, 466 SE 32d 627, 1995; Moore, 1995). Even if they agree with the physician’s judgment that there is no hope of improvement, they may take the position that if the child is not dead, he or she is still alive and should be treated. In one recent case (Burks v. St. Joseph’s Hospital, 227 Wisc 2d 811, 596 NW 2d 391, 1999), a woman delivered a baby at 22 weeks of pregnancy. She was told (as she would be in virtually all hospitals in the country) that her 7 ounce daughter, even though she was breathing at birth, had no hope of survival. Although she asked that the baby be taken to the Newborn Special Care Unit, she was told that extensive efforts at resuscitation were inappropriate, and the baby died 21/2 hours later. She sued the hospital and the court held that she had a right to pursue the claim. Whether she will win and recover damages has not yet been determined.
The best known of case of this type was the case of Baby K, which has already been mentioned (In re Baby K., 16 F 3d 590, CCA 4, 1994; see also, In the Matter of Infant C., 1995 WL 1058596, Va. Cir November 17, 1995). Baby K was anencephalic but, for reasons still unclear, was put on a respirator in the delivery room. Her mother refused to permit a DNR order or removal of the respirator. She firmly believed that God would heal her child. Baby K was eventually discharged to a nursing facility that accepted babies. She was readmitted to the hospital several times for treatment of respiratory distress. Finally the hospital asked for an order from the court stating that it did not have to provide extraordinary medical treatment to this hopeless case. The court found that the federal Emergency Treatment and Active Labor Act, which requires any hospital to provide essential care to emergency admissions, applied to Baby K and the hospital could not refuse to treat her in the emergency department. The trial judge’s decision was upheld when the hospital appealed to the United States Court of Appeals for the Fourth Circuit. Baby K finally died in the pediatric nursing home at age 14 months, still on full life support (Fletcher, 1997).
Thus, it is increasingly clear that before a physician may terminate life support on any patient, adult to newborn, when the parent or next of kin objects, she or he should assume that it is necessary to ask a court for an order. Refusal of such a request should not, however, come as a particular surprise (Massie, 1993). An ethics committee consultation may help mediate the issues and might persuade the parents to change their minds, particularly if the basic problem is the parents’ lack of trust in the physician. Nonetheless, the findings of an ethics committee have no legal standing and cannot be used alone as the basis for termination of life support.
Based on their experience, many physicians recognize that even very small children know when they are very sick. As discussed in Chapter 4, they are often far more aware of death than adults may realize.
Medical and nursing care of seriously ill children includes helping them to achieve a developmentally appropriate understanding of their illness and making sure that they know what to expect from tests and treatments. Their views should be taken seriously, but these views may or may not be the deciding factor in therapeutic decisionmaking.15
Preadolescent children are rarely, if ever, asked if they want the medical care their physician and parents decide is best for them. No one asks a 6-year-old if he wants an injection. He is told that he is going to get one and what it will feel like. (A child can still be offered choices such as which arm to use and can still be advised on what he or she can do to make it hurt less.)
For adolescents, the picture is more complicated. Under English common law, a minor was emancipated if he (not she) was a young man who was not subject to parental control or regulation. In all aspects of his life, he was considered to be a legal adult and could buy and sell property, sign contracts, get married, or do anything else adults could do.16 The American legal system adopted the concept. In twenty-first century America, an emancipated minor is one who is married, is in the military (a much less frequent occurrence than it was when the age of majority was 21 instead of 18), or is self-supporting and living away from home. If a minor’s marriage is dissolved, he or she remains emancipated. In addition to these categories of emancipated minors, many states have enacted statutes providing other contexts in which a minor (with or without a court order) is emancipated and, thus, whose parents have no further legal responsibilities for him or her.
Adolescents now are being allowed to make decisions about their own medical care and other areas of their lives at a level that would have been astonishing 50 years ago (Holder, 1987; Caldwell v. Bechtol, 724 SW 2d 739, Tenn 1987; AAP, 1995a; Hartman et al., 2000). Beginning about 1950 with the epidemic of venereal diseases in adolescents, physicians realized that if a teenager knew that his or her parents would be contacted to provide consent for treatment, thus discovering what their son or daughter had been doing, the adolescent would forgo treatment, thus spreading the problem. As a result of physician influence, by the mid-1960s all states had enacted legislation permitting minors to be treated for venereal disease without parental notification, and today almost all states have similar laws allowing confidential treatment for drug and alcohol problems. At the time, some states began enacting minor treatment statutes giving a minor of a specified age (usually 16, but in some states as young as 14) the right to consent to any medical treatment on his or her own without parental consent.
Even in the absence of state statutes allowing minors to consent to treatment, it has been more than 40 years since a court awarded damages to parents against a physician who treated a child without their consent, as long as the adolescent patient gave a knowledgeable consent to the procedure. These court rulings have become known as the “mature minor” rule. The standard accepted by courts in virtually all states is that if a young person (14, 15, or older) understands the nature of the proposed treatment and its risks and expected benefits as well as an adult could, he or she may give consent. The maturity of the particular adolescent and the gravity of the illness must be factored into the assessment.
One might assume that if a patient may consent to treatment, then that patient may also refuse it. In fact, it is very unusual for a court in this country to allow an adolescent to refuse treatment that is clearly necessary (e.g., an appendectomy) even if it is not incontrovertibly life saving (see, e.g., Rosato, 1996; Skeels, 1990; Lonowski, 1995; Bidari, 1996; Penkower, 1996; Derish et al., 2000). In a few cases, all of which involve adolescent Jehovah’s Witnesses and their refusal of blood transfusions, courts allowed adolescent patients to refuse transfusion based on a demonstration that they understood the nature of the illness, could articulate their own religious objections to the blood transfusion (i.e., were not merely reflecting parental pressure), and understood quite clearly the permanence of death. In at least one case involving a 16-year-old with leukemia, the child died (In re E.G., 133 Ill 2nd 98, 549 NE 2d 322, 1989; Belcher v. Charleston Area Medical Center, 188 W. Va 105, 422 SE 2d 827, 1992; Traugott and Alpers, 1997). Many more cases, however, have held in identical or similar circumstances that while the right to refuse treatment is available for any mentally competent adult, no such right exists for minors (Novak v. Cobb
County—Kennestone Hospital Authority, 849 F Supp 1559, DC Ga, 74 F 3d 1173, CCA 11 1996).
Except for West Virginia, states do not permit a minor, even if legally emancipated, to create a legally enforceable living will, durable power of attorney, or other statement that he or she would want treatment stopped in case of serious illness or accident (Hawkins, 1992).17 This does not mean that the physician should neglect to find out what the child’s views on continuing treatment may be if the long-term prognosis is not good (AAP, 2000g). Those views should still be taken into account when decisions are made (In re Chad Swan, 569 A 2d 1202, Maine 1990).
When parents and an adolescent patient disagree, the physician is the patient’s, not the family’s, advocate. If the prognosis is poor and the patient has “had enough,” the physician is professionally obligated to do everything she or he can to persuade the parents to let the child’s views control. Therapies that the physician considers inadvisable may not be required by distraught parents. On the other hand, if the prognosis is good but the patient does not want to continue therapy, the physician’s responsibility is to understand and respond to the child’s fears and help her or him through the treatment.
Although there may be instances in which parents are bitterly divided on the wisdom of continuing life-sustaining treatment for a child who is likely to die, the committee located only one case on the question. A terminally ill 13-year-old girl was in a coma. Her mother agreed with her physicians that a DNR order was the appropriate next step. Her father adamantly disagreed and wanted her treated as aggressively as possible. The hospital turned to the court for guidance, and the trial court and state supreme court each held that both parents had to agree before a DNR order could be written (In Re Jane Doe, 262 Ga 389, 418 SE 2d 3, 1992).
In cases of this sort, the parent who wishes the child to continue “living” is likely to be upheld by a court. If the parents are divorced, the legal custodian is likely to prevail in such a conflict (Durfee v. Durfee, 87 NYS 2d 275 NY 1949).
Children Who Are Wards of the State
State officials must approve medical care provided to children who are wards of the state, for example, children in foster care or correctional facilities. State officials and policies, which are not always in writing, determine whether health care providers can obtain consent for measures such as DNR orders or removal of mechanical ventilators. Although committee members had considerable experience with state refusals to grant consent for such measures, written information on policies and their application is very limited. This leaves physicians in a difficult position and potentially encourages substandard care and preventable suffering.
The Newborn with a Correctable Defect or Extreme Prematurity
The care of newborns with severe handicaps or extreme prematurity became a political issue during the 1980s following the well-publicized case of a baby with Down syndrome and a tracheoesophageal fistula who died because his parents refused to consent to surgery. An anonymous person tried to get a court to order surgery, but the local juvenile court held that the baby’s parents had the right to make the decision. The Indiana Supreme Court concurred, and the baby died while an appeal to the United States Supreme Court was being sought.
The Department of Health and Human Services (HHS) then issued regulations requiring treatment of all handicapped infants except those who would inevitably die of their irreparable conditions (Shapiro, 1984). These regulations applied only to children with birth defects who had not yet had their first birthday. They did not apply, for example, to a healthy newborn severely injured in an automobile accident on the way home from the hospital or to a 13-month-old child with birth defects. The regulations were struck down, republished, and finally declared unconstitutional by the United States Supreme Court in 1986 (Bowen v. American Hospital Association, 476 US 610, 1986). In the meantime, however, the American Academy of Pediatrics and HHS reached agreement on regulations requiring any state seeking federal funds for child abuse programs to agree to enact state regulations to require that handicapped infants be treated except in certain specific circumstances. Those circumstances are that the infant was irreversibly comatose, the treatment would be “inhumane,” the treatment would be “futile,” or the provision of treatment would merely prolong the infant’s dying (Child Abuse Amendments of 1984, P.L. 98-457).
Since the 1980s, the federal government has not intervened on behalf of any allegedly neglected newborn. As was the case before the “Baby Doe” regulations, state courts have dealt with conflicts arising when physicians wish to override a parent’s decision to withdraw a treatment from a baby
(Stahlman, 1990). The federal investigations and rules may have, at least for a while, made neonatologists more fearful of forgoing life-sustaining interventions that they truly believed to be unwise (Kopelman et al., 1988).
High-profile litigation, regulations, and legislation about decisions at the end of life are uncommon but may create anxiety and fear among clinicians that have the potential to distort the way physicians inform and counsel patients and families (see, e.g., Kopelman et al., 1988). Some of this anxiety and fear is based on misunderstandings of statutes, regulations, and judicial decisions, although some may also be based on excessive caution on the part of hospital or other legal counsel and management. Efforts by professional societies, health professions educators, and organizations to improve end-of-life care may help correct erroneous views and overcome undue caution. These groups may also help to promote wider discussion and evaluation of strategies for preventing, resolving, and managing conflicts about care for dying patients.
Not all conflicts can be avoided, but they can be handled in ways that increase or decrease the potential for damage to the involved parties. Strategies include developing evidence- and consensus-based guidelines for care, improving communication skills, fostering sensitivity to cultural differences, and developing organizational procedures for identifying and defusing potential conflicts and promoting trust. Although the emphasis of these strategies will often be on physicians and parents as decisionmakers, as recommended in Chapter 4, children should be involved in discussions about their care consistent with each child’s intellectual and emotional maturity and preferences and with sensitivity to family cultural background and values.
The next chapter discusses directions for improving health professions education in palliative, end-of-life, and bereavement care. It notes that education in these issues is often restricted to ethical concerns and argues that competence in palliative and end-of-life care cannot be achieved without guided clinical experience and training that is backed by a good understanding of available scientific evidence.