Advances in medicine and technology, in addition to a variety of other factors, have precipitated a dilemma for pediatricians and other health care workers: when is it justified to limit or refuse treatment for children whose parents want "everything done"? This article considers three aspects of this issue:
* What are the roles of the pediatrician and family concerning treatment of an ill child?
* When is it appropriate for physicians to refuse treatment?
* How can the process of conflict management and resolution be used?
THE ROLE OF THE PEDIATRICIAN AND FAMILY: WHO DECIDES?
Physicians are expected to balance the good of medical therapy against the risks of harm due either to the medical intervention or the disease. However, emphasis has recently shifted from the viewpoint that the physician is the essential decision maker toward increasing parental authority to the point where many physicians feel obligated to do whatever patients or their families request.
What are the options for the physician who believes that continued aggressive treatment is inappropriate? In 1983, the Presidential Commission considered this issue:
Although competent patients. . .have the legal and ethical authority to forego some or all care, this does not mean that patients may insist on particular treatments. The care available from health care professionals is generally limited to what is consistent with role-related professional standards and conscientiously held beliefs.1
In contrast, Lantos and colleagues have stated that it is the goals of the patient and family, not the physician's assessment of the efficacy of treatment that determine medical futility.2 Unquestioning assent to this parental role would lead to situations where parents are allowed to refuse treatments that are in the best interests of the child. There is general agreement that physicians should not accede to parental requests to withhold or withdraw treatment that is clearly beneficial to their child. Likewise, physicians also should not accept parental requests for treatments that will only inflict pain with no or very limited potential benefit. However, determination of this potential in infants and children is characteristically difficult. It has been suggested that physicians must continue or initiate new treatments with even a low likelihood of benefit out of respect for that (parental) choice.3 This is implied by the comments of Perelman and Post in their discussion of the case of Baby L: ". . .the central question is whether patients who cannot experience the richness of normal life have experiences that make continued existence from their own perspective better than no life at all."4 Such arguments, although compassionate, do not, in our opinion, give sufficient weight to the understanding that, for many such patients, the future is that of a chronic incapacitating illness with very limited chance of improvement but continued inestimable pain, suffering, and anguish.
As an infant becomes progressively ill, the prognosis changes, and reassessment of the appropriateness of care is indicated.5 Physicians may have to choose between prolonging a child's life and alleviating pain and suffering. The physician must not only take credit for saving the infant's life but also must assume responsibility for incurring injury and suffering. This can "only be justified by a realistic expectation that underlying medical problems will resolve if such life-sustaining therapy is instituted."5 This is more relevant as physicians can now maintain biologic life for extraordinary periods. The traditional view that parents alone possess the right to determine the best interests of their children cannot be supported in this era of advanced technology.6 It seems clear that a reversal of the current balance between autonomy and beneficence is needed to allow for a more appropriate decision-making process that can put into perspective the unrealistic expectations that technology has created.
REFUSAL OF TREATMENT: WHEN IS IT JUSTIHED?
When the seemingly best treatment decision for a child is an option that is not clearly "the right or good thing to do," the possibility of limitation of treatment should be discussed. Three general justifications for withholding, withdrawing, or limiting therapy exist7:
* the therapy in question will not work,
* the burdens of treatment outweigh the benefits (the quality of life is poor after treatment), and
* the burdens of the disease outweigh the benefits of continued suvival (the quality of life is poor prior to treatment).
When Therapy Will Not Work
What purpose is served by offering patients treatments that are known to be ineffective? If it is done to preserve the patient's autonomy, their autonomy does not extend to choosing useless procedures. If it is offered to give the family hope, then it is a cruel hope.8 Unfortunately, it is usually unclear whether the proposed therapy has little or no medical benefit. Therefore, most discussions of futility bring up the issue of medical uncertainty.
Consider the issue of cardiopulmonary resuscitation (CPR). Nurses, physicians, and families regularly fece the question of whether to attempt resuscitation. Over the past 15 to 20 years, a consensus has developed that CPR should be attempted in all hospitalized patients unless agreement has been reached that it would be inappropriate for a particular patient.9 Currently, nearly 60% of adult patients dying in hospital have a do not resuscitate (DNR) order,10 and only 7% of patients who survive CPR and leave the hospital suffer from permanent neurological impairment.11
Cardiopulmonary resuscitation and DNR decisions in children are more difficult. From the limited data available, neurologically intact survival rates of children after CPR are low, but not zero. Neonates, and particularly premature infants, generally fere worse than older children.12 In contrast to adults in whom cardiac disease may be primary, cardiac arrest in children often represents the end stage of a progressive physiologic deterioration and is frequently characterized by asystole, a refractory dysrhythmia at any age.
As illustrated in the following case, a dying child will often show progressive deterioration despite aggressive care, and it may become quite clear that resuscitation efforts as a last-ditch measure would be ineffective. Thus, while we do not have data to accurately predict the outcome of attempted resuscitation in all patients, the prognosis for some patients may be quite clear.
Patient E was a 3 -year-old male who suffered a near-drowning episode on Christmas day. Three months after the event he had made no neurologic recovery and was in a vegetative state when pneumonia, respiratory failure, and shock developed. At the parents' request and against the physician's judgment, he was intubated and ventilated, and received cardiovascular support. Despite these measures, the child remained hypotensive, acidotic, and anurie. Although conferences with parents had occurred regularly throughout his hospitalization, they intensified when it became clear that despite the aggressive support the child might succumb to the current infection and shock. After lengthy discussions, it was decided that further resuscitation measures would not be added. He was allowed to die - 14 weeks after the submersion accident occuned.
If any treatment will not work, the decision to initiate or withdraw it is the physician's responsibility, and patients do not have the right to insist on its use. In the litigious environment prevalent today, ceding decision-making responsibility entirely to patients or their family members may seem the path of least risk to physicians. This practice may lead to the continuation of mechanical ventilation, cardiopulmonary resuscitation, or other therapies past the point of any benefit if a family insists that caregivers "do everything." On the basis of two ethical principles, physicians are morally obligated not to offer or provide treatments they believe to be worthless.13,14 First, on the grounds of beneficence, treatments that are of no clear benefit to the patient are not ethically justifiable. Second, on the grounds of justice, deathprolonging treatments represent a misuse of diminishing, expensive health care resources.
What should happen when it is unclear whether treatment is life-sustaining or useless? In such cases, the responsibilities of physicians and the rights of patients will be best balanced in a process of shared decision making aimed at consensus. Physicians, patients, and their families must come to terms with the inability of medicine to postpone death indefinitely.
When the Burdens of Treatment Outweigh the Benefits
Proportionate treatment is that which, in the view of the patient or family, has at least a reasonable chance of providing benefits to the patient that outweigh the burdens attendant to the treatment. Benefits exist when treatment allows, at the very least, a remission of symptoms enabling a return toward a functioning, integrated existence. Thus, even if a proposed course of treatment might be extremely painful or intrusive, it would still be a proportionate treatment if the prognosis was for complete cure or significant improvement in the patient's condition. On the other hand, a treatment course that is only minimally painful or intrusive may nonetheless be considered disproportionate to the potential benefits if the prognosis is virtually hopeless for any significant improvement in condition.15
A decision to withhold or withdraw a treatment from a patient is fundamentally an ethical decision; it is a question about obligations or duties. When a patient is competent, there is consensus that he or she has a nearly absolute right to decide what shall be done. In present-day society, adults are presumed to be able to decide for themselves if life extension is a good thing; minors are not.16 However, society is now recognizing that certain minors have capacities to consent on their own for certain medical treatments. If a minor has experienced an illness for some time, understands the illness and the benefits and burdens of its treatment, has the ability to reason, and has a comprehension of death that recognizes its personal significance and finality, then that person, irrespective of age, is competent from an ethical standpoint to consent to forego life-sustaining treatment.16 Consider the following example illustrating this point:
A 14-year-old female with neuromuscular disease was hospitalized with acute respiratory insufficiency and cor pulmonale following an episode of pneumonitis. Although severely incapacitated and wheelchair dependent, she was intelligent and articulate. On admission, she was obtunded, and at the parents' request she was intubated and mechanically ventilated. Questions as to whether long-term ventilator dependency was likely and a value judgment as to whether long-term ventilator dependency was acceptable were raised the first hospital day.
Six weeks after admission, the child was awake, alert, and able to communicate, but ventilator dependent. Although the parents wanted to enroll her in a chronic home mechanical ventilation program, the patient repeatedly and consistently expressed a desire not to be ventilated. Numerous sessions with different individuals established quite clearly that total dependence on a machine was unacceptable to the patient. After long debate, lasting 1 more month, her requests were respected and ventilator support was removed. She died the same day surrounded by friends, family, and articles of personal importance to her.
The decision to ventilate this child initially was correct because the exact wishes of the child could not be determined and because it was not clear that long-term ventilator dependency would be the outcome. It is generally worse to make an error in judgment that results in death than an error that results in life.17 A ventilator-dependent child can, as in our case, be withdrawn from the ventilator and allowed to die if further evidence and reflection leads to that conclusion. Therefore, ventilation should be initiated until there is time to review the facts, prognosis, preferences, and other ethical considerations. Although it was psychologically more difficult to withdraw treatment in this child to whom family and staff had become attached, it was ethically preferable from all perspectives because the decision was based on better data, a clearer prognosis, and knowledge of the child's desires. Although there was initial disagreement within the family, outside consultation and active involvement by the ethics committee resolved the disputes and led to the decision that the child's desires should be respected. The final decision was arrived at through process.
When patients are not competent or are too young to have their desires known, family members (usually parents) or friends, and sometimes court-appointed guardians serve as surrogates to represent the patient's best interest in the decision-making process with health professionals. It must be emphasized that the central principle guiding decisions in these cases should be the best interests of the child and not mere parental preference.17 In the majority of cases, parental desires are consistent with others' opinions about what course of treatment would be in the child's interest. In a minority of cases parents and the health care team disagree, and dilemmas arise that must be confronted and resolved. Families, health care institutions, and health care professionals must work together to make decisions for patients who lack decision-making capacity.
When The Burdens of the Disease Outweigh the Benefits of Continued Survival
In the April 15, 1990 issue of the New England Journal of Medicine, Paris et al presented the case of Baby L18
Baby L, as presented, was a 2 -year-old girl who, because of fetal and neonatal problems, suffered severe hypoxic-ischemic encephalopathy. She was blind, deai and quadriplegic, and the extent of her environmental interaction was to respond to painful stimuli with a grimace, stiffening, and withdrawal. She suffered from intermittent seizures. The infant underwent a gastrostomy at the age of 1 month, a fundoplication at 4 months, and a tracheostomy at 7 months. She had frequent hospitalizations for pneumonia. The case concerned one of these hospitalizations, and the issue involved the mother's request for full treatment, including mechanical ventilation and cardiovascular support if necessary. The physicians caring for the infant believed that this course of action would be inhuman and against the best interests of the child. They further stated that even if the court issued an order for mechanical ventilation, they would not comply, for that order would violate their professional ethical obligations.
This case raises the question of whether continued survival in the child's painful baseline state is in the child's best interest. The issue is not whether mechanical ventilation will work to reverse respiratory failure or even whether subsequent care will be too burdensome. The real issue is whether it is in the child's best interest to return to her baseline state should respira' tory failure occur.
The most difficult clinical, ethical, and legal issues arise when the question is asked whether to discontinue aggressive life support or to allow individuals living with damaged brains or nervous systems to die.19 It is the responsibility of the physician, with the support and consensus of the entire health care team, to recommend a course of action for the individual patient. As health care professionals, we must resist the urge to institute desperate remedies for desperate patients and families.20 Unless a reversal or amelioration of the underlying condition can be expected, treatments should be questioned; in many cases, such treatments may be disproportionate or inhumane. Physicians should not be slaves to technology, and patients should not be its prisoners.21
Treatment is said to be futile when it will not, with reasonable medical certainty, preserve a physiological function necessary to maintain life.14 A health care professional has no obligation to offer, begin, or maintain a treatment which, in his or her best judgment, will be physiologically futile.22
As all health care professionals realize, situations occur in which all that medicine can offer is the prolongation of a dying curve or the fixation of a child's course of development at levels far below activities and abilities that give meaning to human existence. However, unless the child's death is proximate and treatment will fail to reverse the physiologic disturbance, a recommendation to withdraw or withhold treatment involves a value judgment about the child's interest. In such instances, a physician should not make unilateral treatment decisions. Utility or futility of any particular treatment under these circumstances can be judged only relative to the selection of a goal. When parents are requesting a treatment, a low likelihood of benefit may be sufficient to mandate physician respect for the choice.3 Considerable care must be exercised in applying this rationale because it may be all too easy to assume that a life of mental retardation or significant physical disability itself constitutes an excessive burden for the child. Medical futility is often confused with despair about quality of life.4
If a physician concludes that treatment harms the child by inflicting pain and suffering but offers no appreciable benefit for the child, then the physician is justified in refusing to provide that therapy despite parental demand. The physician is then obligated to reach an agreement with parents as outlined below. In some instances, transfer of care may be required.
When further therapy has been determined to be inappropriate by family and health care professionals, then efforts should be directed to creating an environment that allows peaceful death.23
CONFLICT MANAGEMENT AND RESOLUTION: THE PROCESS
Disagreements about "what should be done" are inevitable and often lead to conflict. When infants and children are involved, the conflict is often between the physician and parents. It is sometimes possible to prevent such conflict, but more often it is necessary to manage and resolve it. It may be easier to manage and resolve conflict if we can understand the reasons behind it.24
Some conflicts arise because the two parties are using different fects in their calculation of the benefits and burdens of a particular modality. The family may be hearing insufficient fects, incorrect fects, or conflicting "facts" from different members of the health care team. Conflicts also may occur, however, even when there is agreement about the fects, but differences in value judgments. These value differences may be identified or unrecognized and may come from cultural or religious dogma or tradition, or from personal or professional experience. A third source of conflict is different emotions. The "detached" physician may feel that he or she is being more objective then the parent who is overwhelmed with fear, anger, guilt, grief, or denial. Physicians also should be aware that they are not immune to the same emotions and reactions in their professional preferences.
Conflict management begins with prevention. It seems trite to say that the foundation for conflict management is good communication, but it is always helpful to remind ourselves that communication is not what is said, but what is heard. If the parents perceive that the physician is insensitive in his or her approach to them, they are more likely to question how much compassion he or she is able to direct toward their child. Callous communication of a poor prognosis and recommendation for withdrawal of treatment may be met with resistance that goes on to conflict, while compassionate communication of the same information and recommendation may be more easily accepted.25
Another important facet of good communication is identification of one individual who is the primary source of information for the parents. That person must be clear in his or her own mind regarding the fects of the situation and must transmit the recommendations of the team clearly, tactfully, and compassionately. Even when there are professional differences of opinion about the best plan, having one individual who can candidly discuss the situation with the parents may lessen the conflictual tone. Similarly, once conflict has arisen, having a single resource may lessen its impact and help in its resolution.
It is helpful in trying to avert or manage conflict to concentrate on two general principles. The first is the need to clarify the goals of treatment. Is everyone involved still aiming for life prolongation or have some shifted their goal to patient comfort? Agreement on the goal of treatment often dissipates conflict that is focused on a specific treatment modality. The second general principle of conflict management is shared decision making. If the parents can be assured that the goal of discussion is to come to a mutually agreeable decision rather than to unilaterally impose a paternalistic decision, they may feel some strength in that degree of control and therefore be less adversarial.
Resolution of established conflict can sometimes be facilitated by seeking other opinions. Sometimes a second medical opinion may lend weight to either the position of the health care team or the family. If the conflict has escalated to the point that the parents are antagonistic to the entire hospital, getting an outside opinion, in person or by telephone, may help to defuse the charged atmosphere.
Consultation from a clinical ethicist or an ethics committee may likewise be beneficial. If such a consultation is able to enhance communication, identify ethically permissible options, clarify goals, and encourage joint decision-making, the parents may then feel that they have been heard by another forum and may be willing to follow the professional recommendations. Alternatively, the health care team may come to a clearer understanding of the family's position and may be willing to follow their request. Often the resolution that comes out of such a consultation is agreement of both parties to pursue a time-limited trial of the intervention under discussion, followed by reassessment of whether it has achieved the goals mutually agreed to. When the focus of the conflict is whether a particular course of action is legally permissible, consultation with hospital counsel may clarify the situation and resolve the conflict. Going to court to ask a judge to make a health care decision is the least desirable method of resolving conflict. Discussion and consultation will most often lead to resolution of conflict between physician and parents. The resolution may involve continuation of aggressive care, agreement on a limitation of treatment plan, or consultation from or transfer of the patient to hospice care.
When Conflict Cannot Be Resolved
There are times when conflict cannot be resolved. In some instances, parental demands for continued aggressive care cannot possibly achieve the patient's best interests. In other situations, the pediatrician reaches the conclusion that he or she cannot provide the specific treatment modality demanded or cannot pursue the goal chosen by the parents because it is outside the bounds of good medical practice. Sometimes the physician must exercise his or her professional responsibility and say "We don't do that here." When it must be said, it must be said gently and with compassion rather than with frustration or anger.
Once the pediatrician has told the parents that he or she is unwilling to provide the requested service, only two options remain. In most instances, he or she should offer to transfer the child's care to another pediatrician and offer to help the parents find one who is willing to accept the responsibility of pursuing the parental demands. If no one can be found who is willing to accept the transfer, the pediatrician has no other choice but to go to court to ask a judge to disqualify the parents as health care decision makers for the child and appoint a guardian who can authorize limitation of treatment.
There are rare occasions when the pediatrician should not offer to transfer the care to another physician. If he or she believes that following the parents' demand will cause the child unnecessary and uncontrollable suffering, he or she has a moral obligation to the patient to seek the patient's good over the parental demands. In such circumstances, he or she may elect to go directly to the judge with a request for humane limitation of treatment.
To our knowledge, only one case in the United States has been taken to court by pediatricians requesting judicial intervention to override parental demands for life-prolonging treatment.18 Because of the very serious nature of such a move, all health care workers involved with the care of children should observe the note of caution with which the authors of that report conclude:
The issues raised in overriding a family's request for potentially life-saving treatment are complex and difficult. They ought to remain difficult. It is only with the greatest of caution that physicians should consider overturning the presumption of treatment and respect for the family's wishes. The potential for abuse demands that safeguards be designed to ensure that decisions are openly made, principled, and focused on the best interests of the patient. The minimal requirements should include agreement among health care workers, the concurrence of an ethics committee, openness to a second opinion, and a comprehensive note in the patient's chart detailing all the factors considered in the decision.18
1. President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Deciding to Forego Life-Sustaining Treatment: A Report on the Ethical, Medical, and Legal Issues in Treatment Decisions. Washington, DC: Government Printing Office; 1983.
2. Lantos JD Singer PA, Walker RM, et al. The illusion of futility in clinical practice. Am J Med. 1989:87:81-84.
3. Fleischman AR. Point-counterpoint - physicians' refusal of requested treatment. J Perinatal. 1990;10:407-408.
4. Perelman RH, Post NC. Point-counterpoint-physicians' refusal of requested treatment. J Perinatal. 1990;10:412-414.
5. Stevenson DK, Benitz WE, Rhine WD, Ariagno RL. Point-counterpointphysicians' refusal of requested treatment. J Perinatol. 1990;10:408-409.
6. Nathenson G. Point-counterpoint - physicians' refusal of requested treatment. J Perinarol. 1990;10:414.
7. Tomlinson T, Brody H. Ethics and communication in do-not-resuscitate orders. N Engl J Med. 1988;318:43-46.
8. Blackhall LJ. Must we always use CPR! N Engl J Med. 1987:317:1281-1285.
9. Paris JJ, Reankm FE. Moral, ethical and legal issues in the intensive care unit. J Intensive Care Med. 1991;6:175-195.
10. Johnson PV, McNamee M, Campion EW. The do not resuscitate order- a profile of its changing use. Arch Intern Med. 1988;148:2373-2375.
11. Bedell SE, Delbonco TL1 Cook EF, Epstein FH. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med. 1984;310:1089-1093.
12. Lantos JD Miles SH, Silverstein MD, Stocking CB. Survival after cardiopulmonary resuscitation in babies of very low birth weight. Is CPR futile therapy? N Engl J Med. 1988;318:91-95.
13. Weil MH, Weil CJ, Rackow EC. Guide to ethical decision-making for the critically ill: The three R's and QC Crit Core Med. 1988;16:636-641.
14. Tomlinson, T, Brody H. Futility and the ethics of resuscitation. JAMA. 1990;264:12761280.
15. Guidelines on the Termination of Life-Sustaming Treatment and the Care of the Dying. A Report by the Hastings Center. Bloomington, Ind: Indiana University Press; 1987.
16. Leikin S. A proposal concerning decisions to forego life-threatening treatment for young people. J Pediatr. 1989;115:17-22.
17. Farteli PM, Feet NC. Long-term mechanical ventilation in pediatrie respiratory failure: medical and ethical considerations. Am Rev Respir Dis. 1989;140:S36-S40.
18. Paris JJ, Crone RK, Reardon E Physicians refusal of requested treatment: the case of Baby L N EnglJ Med. 1990-322:1012-1015.
19. Roy DJ, Verret S, Roberge G Death, dying, and the brain. Prim Care. 1986; 13:367378.
20. Moore FD. The desperate case: CARE (costs, applicability, research, ethics). JAMA. 1989;261:1483-1484.
21. Angeli M. Prisoners of technology: the case of Nancy Cruzman. N Engl J Med. 1990;322:1226-1228.
22. Task Force on Ethics of the Society of Critical Care Medicine. Consensus report on the ethics of foregoing life-sustaining treatments in the critically ill. Crit Core Med. 1990;18:1435-1439.
23. Safar P, Winter R Helping to die. Crit Cor« Md. 1990;18:790-791.
24. Nelson LJ, Nelson RM. Ethics and the provision of futile, harmful or burdensome treatment for children. Crit Care Med. In press.
25. Gert B, Culver CM. Moral theory in neurologic practice. Semm Neurol. 1984;4(1):914.