The Case of Claire Conroy
An eighty-four-year-old non-ambulatory nursing home patient was confined to semi-fetal condition with "severe organic brain syndrome." She had necrotic gangrenous leg ulcers and no bowel control. The patient was not in a vegetative state. She followed movements with her eyes, showed some head and hand movements and smiled when massaged or when her hair was combed. Her life expectancy could not be predicted but it was thought to be "some months." (The patient died of natural causes while appeal was pending).
The patient's nephew, as her only relative, had been named guardian earlier, and requested removal of the nasogastric tube. The physician declined. The guardian sought a court order authorizing the withdrawal of tube.
The trial court granted the requested order, based on the belief that the patient was "reduced to a very primitive level." The court stated that it would be wrong to refuse nutrition if the patient could be manually fed.
The intermediate appeals court reversed the decision (though recognized it to be mooted by death):
Where death is impending but not yet imminent, and absent clear evidence of expressed patient's wishes, the State may authorize a guardian to withdraw life-sustaining treatment, but only under either of two "best interests" tests:
Artificial feeding was equated with artificial breathing by respirator, stating each to be "medical procedures."
(a) Without court order - if other two physicians concur on patient's condition and prognosis, guardian and physician may concur, in withdrawing life-sustaining treatment, if ombudsman also concurs and if spouse, parents, children or other next of kin concur. (They will be without criminal or civil liability if they acted without "bad faith.")
(b) With court order (not discussed in opinion, but surely follows in such instances where the
required concurrences are not present)
Commented on possible inadequacy of inquiry as to patient's earlier statements of treatment
Commented on inadequacy of collected "information concerning the benefits and burdens" and as to
pain suffered under alternate treatment options.
Article: “Must Patients Always Be Given Food and Water?”
By Joanne Lynn, MD and James F. Childress, PhD
The Hastings Center, October 1983
Joanne Lynn, M.D. is on the staff of the division of geriatric medicine, George Washington University, and was an assistant director of the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research.
James F. Childress, Ph.D., is commonwealth professor of religious studies and medical education, University of Virginia.
Many people die from the lack of food or water. For some, this lack is the result of poverty or
famine, but for others it is the result of disease or deliberate decision. In the past, malnutrition and dehydration must have accompanied nearly every death that followed an illness of more than a few days. Most dying patients do not eat much on their own, and nothing could be done for them until the first flexible tubing for instilling food or other liquid into the stomach was developed about a hundred years ago. Even then, the procedure was so scarce, so costly in physician and nursing time, and so poorly tolerated that it was used only for, patients who clearly could. benefit. With the advent of more reliable and efficient procedures in the past few decades, these conditions can be corrected or ameliorated in nearly every patient who would otherwise be malnourished or dehydrated. In fact intravenous lines and nasogastric tubes have become common images of hospital care.
Providing adequate nutrition and fluids is a high priority for most patients, both because they suffer directly from inadequacies and because these deficiencies hinder their ability to overcome other diseases. But are there some patients who need not receive these treatments? This question has become prominent public policy issue in a number of recent cases. In May 1981, in Danville, Illinois the parents and the physician of newborn conjoined twins with shared abdominal organs decided not to feed the children. Feeding and other treatments were given after court intervention, though a grand jury refused to indict the parents. Later that year, two physicians in Los Angeles discontinued intravenous nutrition to a patient who had severe brain damage after an episode involving loss of oxygen following routine surgery. Murder charges were brought, but the hearing judge dismissed the charges at a preliminary hearing. On appeal, the charges were reinstated and remanded for trial.
In April 1982, a Bloomington, Indiana, infant who had tracheoesophageal fistula and Down syndrome was not treated or fed, and he died after two courts ruled that the decision was proper but before all appeals could be heard.3 When the federal government then moved to ensure that such infants would be fed in the future,4 the Surgeon General, Dr. C Everett Koop, initially stated that there is never adequate reason to deny nutrition and fluids to a newborn infant.
While these cases were before the public, the nephew of Claire Convoy, an elderly incompetent woman with several serious medical problems, petitioned a New Jersey court for authority to discontinue her nasogastric tube feedings. Although the intermediate appeals court has reversed the ruling,5 the trial court held that he had this authority since the evidence indicated that the patient would not have wanted such treatment and that its value to her was doubtful.
In all these dramatic cases and in many more that go unnoticed, the decision is made to deliberately withhold food or fluid known to be necessary for the life of the patient. Such decisions are unsettling. There is not widespread consensus that sometimes a patient is best served by not undertaking or continuing certain treatments that would sustain life, especially if these entail substantial suffering.6 But food and water are so central to an array of human emotions that it is almost impossible to consider than with the same emotional detachment that one might feel toward a respirator or a dialysis machine.
Nevertheless, the question remains: should it ever be permissible to withhold or withdraw food and nutrition? The answer in any real case should acknowledge the psychological contiguity between feeding and loving and between nutritional satisfaction and emotional satisfaction. Yet this acknowledgement does not resolve the core question.
Some have held that it is intrinsically wrong not to feed another. The philosopher G.E.M. Anscombe contends: “For willful starvation there can be no excuse. The same can’t be said quite without qualification about failing to operate or to adopt some courses of treatment.”7 But the moral issues are more complex than Anscombe’s comment suggests. Does correcting nutritional deficiencies always improve patients’ well-being? What should be our reflective moral response to withholding or withdrawing nutrition? What moral principles are relevant to our reflections? What medical facts about ways of providing nutrition are relevant? And what policies should be adopted by the society, hospitals, and medical and other health care professionals?
In our effort to find answers to these questions, we will concentrate upon the care of patients who are incompetent to make choices for themselves. Patients who are competent to determine the course of their therapy may refuse any and all interventions proposed by others, as long as their refusals do not seriously harm or impose unfair burdens upon others.8 A competent patient’s decision regarding whether or not to accept the provision of food and water by medical means such as tube feeding or intravenous alimentation is unlikely to raise questions of harm or burden to others.
What then should guide those who must decide about nutrition for a patient who cannot decide? As a start, consider the standard by which other medical decisions are made: one should decide as the incompetent person would have if he or she were competent, when that is possible to determine, and advance that person’s interests in a more generalized sense when individual preferences cannot be known.
The Medical Procedures
There is no reason to apply a different standard to feeding and hydration. Surely when one inserts a feeding tube, or creates a gastrotomy opening, or inserts a needle into a vein, one intends to benefit the patient. Ideally, one should provide what the patient believes to be of benefit, but at least the effect should be beneficial in the opinions of surrogates and caregivers.
Thus, the question becomes: is it ever in the patient’s interest to become malnourished and dehydrated, rather than to receive treatment? Posing the question so starkly points to our need to know what is entailed in treating these conditions and what benefits the treatments offer.
The medical interventions that provide food and fluids are of two basic types. First, liquids can be delivered by a tube that is inserted into a functioning gastrointestinal tract, most commonly through the nose and esophagus into the stomach or through a surgical incision in the abdominal wall and directly into the stomach. The liquids used can be specially prepared solutions of nutrients or a blenderized version of an ordinary diet. The nasogastric tube is cheap; it may lead to pneumonia and often annoys the patient and family, sometimes even requiring that the patient be restrained to prevent its removal.
Creating a gastrostomy is usually a simple surgical procedure, and, once the wound is healed, care is very simple. Since it is out of sight, it is aesthetically more acceptable and restraints are needed less often. Also, the gastrostomy creates no additional risk of pneumonia. However, while elimination of a nasogastric tube requires only removing the rube, a gastrostomy is fairy permanent and can be closed only by surgery.
The second type of medical intervention is intravenous feeding and hydration, which also has two major forms. The ordinary hospital or peripheral IV, in which fluid is delivered directly to the bloodstream through a small needle, is useful only for temporary efforts to improve hydration and electrolyte concentrations. One cannot provide a balanced diet through the veins in the limbs: to do that requires a central line, or a special catheter placed into one of the major veins in the chest. The latter procedure is much more risky and vulnerable to infections and technical errors, and it is much more costly than any of the other procedures. Both forms of intravenous nutrition and hydration commonly require restraining the patient, cause minor infections and other ill effects, and are costly, especially since they ordinarily require the patient to be in a hospital.
None of these procedures, then, is ideal; each entails some distress, some medical limitations, and some costs. When may a procedure be foregone that might improve nutrition and hydration for a given patient? Only when a procedure and the resulting improvement in nutrition and hydration do not offer the patient a net benefit over what he or she would otherwise have faced.
Are there such circumstances? We believe that there are; but they are few and limited to the following three kinds of situations: 1. The procedures that would be required are so unlikely to achieve improved nutritional and fluid levels that they could be correctly considered futile; 2. The improvement in nutritional and fluid balance, through achievable, could be or no benefit to the patient; 3. The burdens of receiving the treatment may outweigh the benefit.
When Food and Water May Be Withheld
Futile Treatment. Sometimes even providing “food and water” toa patient becomes a monumental task. Consider a patient with a severe clotting deficiency and a nearly total body burn. Gaining access to the central veins is likely to cause hemorrhage or infection, nasogastric tube placement may be quite painful, and there may be no skin to which to suture the stomach for a gastrostomy tube. Or consider a patient with severe congestive heart failure who develops cancer of the stomach with a fistula that delivers food from the stomach to the colon without passing through the intestine and being absorbed. Feeding the patient may be possible, but little is absorbed. Intravenous feeding cannot be tolerated because the fluid would be too much for the weakened heart. Or consider the infant with infarction of all but a short segment of bowel. Again, the infant can be fed, but little if anything is absorbed. Intravenous methods can be used, but only for a short time (weeks or months) until their complications, including thrombosis, hemorrhage, infections, and malnutrition, cause death.
In these circumstances, the patient is going to die soon, no matter what is done. The ineffective efforts to provide nutrition and hydration may well directly cause suffering that offers no counterbalancing benefit for the patient. Although the procedures might be tried, especially if the competent patient wanted them or the incompetent patient’s surrogate had reason to believe that this incompetent patient would have wanted them, they cannot be considered obligatory. To hold that a patient must be subjected to this predictably futile sort of intervention just because protein balance is negative or the blood serum is concentrated is to lose sight of the moral warrant for medical care and to reduce the patient to an array of measurable variables.
No Possibility of Benefit. Some patients can be reliably diagnosed to have permanently lost consciousness. This unusual group of patients includes those with anencephaly, persistent vegetative state, and some preterminal comas. In these cases, it is very difficult to discern how any medical intervention can benefit or harm the patient. These patients cannot and never will be able to experience any of the events occurring in the world or in their bodies. When the diagnosis is exceedingly clear, we sustain their lives vigorously mainly for their loved ones and the community at large.
While these considerations probably indicate that continued artificial feeding is best in most cases, there may be some cases in which the family and the caregivers are convinced that artificial feeding is offensive and unreasonable. In such cases, there seems to be no adequate reason to claim that withholding food and water violates any obligations that these parties or the general society have with regard to permanently unconscious patients. Thus, if the parents of an anencephalic infant or of a patient like Karen Quinlan in a persistent vegetative state feel strongly that no medical procedures should be applied to provide nutrition and hydration, and the caregivers are willing to comply, there should be no barrier in law or public policy to thwart the plan.9
Disproportionate Burden. The most difficult cases are those in which normal nutritional status or fluid balance could be restored, but only with a severe burden for the patient. In these cases, the treatment is futile in a broader sense – the patient will not actually benefit from the improved nutrition and hydration. A patient who is competent can decide the relative merits of the treatment being provided, knowing the probable consequences, and weighing the merits of life under various sets of constrained circumstances. But a surrogate decision maker for a patient who is incompetent to decide will have a difficult task. When the situation is irremediably ambiguous, erring on the side of continued life and improved nutrition and hydration seems the less grievous error. But are there situations that would warrant a determination that this patient, whose nutrition and hydration could surely be improved, is not thereby well served?
Though they are rare, we believe there are such cases. The treatments entailed are not benign. Their effects are far short of ideal. Furthermore, many of the patients most likely to have inadequate food and fluid intake are also likely to suffer the most serious side effects of these therapies.
Patients who are allowed to die without artificial hydration and nutrition may well die more comfortably than patients who receive conventional amounts of intravenous hydration.10 Terminal pulmonary edema, nausea, and mental confusion are more likely when patients have been treated to maintain fluid and nutrition until close to the time of death.
Thus, those patients whose “need” for artificial nutrition and hydration arises only near the time of death may be harmed by its provision. It is not at all clear that they receive any benefit in having a slightly prolonged life, and it does seem reasonable to allow a surrogate to decide that, for this patient at this time, slight prolongation of life is not warranted if it involves measures that will probably increase the patient’s suffering as he or she dies.
Even patients who might live much longer might not be well served by artificial means to provide fluid and food. Such patients might include those with fairly severe dementia for whom the restraints required could be a constant source of fear, discomfort, and struggle. For such a patient, sedation to tolerate the feeding mechanisms might preclude any of the pleasant experiences that might otherwise have been available. Thus, a decision not to intervene, except perhaps briefly to ascertain that there are no treatable causes, might allow such a patient to live out a shorter life with fair freedom of movement and freedom from fear, while a decision to maintain artificial nutrition and hydration might consign the patient to end his or her life in unremitting anguish. If this were the case a surrogate decision maker would seem to be well justified in refusing the treatment.
Inappropriate Moral Constraints
Four considerations are frequently proposed as moral constraints on foregoing medical feeding and hydration. We find none of these to dictate that artificial nutrition and hydration must always be provided.
The Obligation to Provide “Ordinary” Care. Debates about appropriate medical treatment are often couched in terms of “ordinary” and “extraordinary” means of treatment. Historically, this distinction emerged in the Roman Catholic tradition to differentiate optional treatment from treatment that was obligatory for medical professionals to offer and for patients to accept.11 These terms also appear in many secular contexts, such as court decisions and medical codes. The recent debates about ordinary and extraordinary means of treatment have been interminable and often unfruitful, in part because of a lack of clarity about what the terms means. Do they represent the premises of an argument or the conclusion, and what features of a situation are relevant to the categorization as “ordinary” or “extraordinary”?12
Several criteria have been implicit in debates about ordinary and extraordinary means of treatment; some of them may be relevant to determining whether and which treatments are obligatory and which are optional. Treatments have been distinguished according to their simplicity (simple/complex), their naturalness (natural/artificial), their customariness (usual/unusual), their invasiveness (noninvasive/invasive), their chance of success (reasonable chance/futile), their balance of benefits and burdens (proportionate/disproportionate), and their expense (inexpensive/costly). Each set of paired terms or phrases in the parentheses suggests a continuum: as the treatment moves from the first of the paired terms to the second, it is said to become less obligatory and more optional.
However, when these various criteria, widely used in discussions about medical treatment, are carefully examined, most of them are not morally relevant in distinguishing optional from obligatory medical treatments. For example, if a rare, complex, artificial, and invasive treatment offers a patient a reasonable chance of nearly painless cure, then one would have to offer a substantial justification not to provide that treatment to an incompetent patient.
What matters, then, in determining whether to provide a treatment to an incompetent patient is not a prior determination that this treatment is “ordinary” per se, but rather a determination that this treatment is likely to provide this patient benefits that are sufficient to make it worthwhile to endure the burdens that accompany the treatment. To this end, some of the considerations listed above are relevant: whether a treatment is likely to succeed is an obvious example. But such considerations taken in isolation are not conclusive. Rather, the surrogate decision maker is obliged to assess the desirability to this patient of each of the options presented, including nontreatment. For most people at most times, this assessment would lead to a clear obligation to provide food and fluids.
But sometimes, as we have indicated, providing food and fluids through medical interventions may fail to benefit and may even harm some patients. Then the treatment cannot be said to be obligatory, no matter how usual and simple its provision may be. If “ordinary” and “extraordinary” are used to convey the conclusion about the obligation to treat, providing nutrition and fluids would have become, in these cases, “extraordinary.” Since this phrasing is misleading, it is probably better to use “proportionate” and “disproportionate” as the Vatican now suggests,13 or “obligatory” and “optional.”
Obviously, providing nutrition and hydration may sometimes be necessary to keep patients comfortable while they are dying even though it may temporarily prolong their dying. In such cases, food and fluids constitute warranted palliative care. But in other cases, such as a patient in a deep and irreversible coma, nutrition and hydration do not appear to comfort the staff and family.14 And sometimes the interventions needed for nutrition and hydration are so burdensome that they are harmful and best not utilized.
The Obligation to Continue Treatments Once Started. Once having started a mode of treatment, many caregivers find it very difficult to discontinue it. While this strongly felt difference between the ease of withholding a treatment and the difficulty of withdrawing it provides a psychological explanation of certain actions, it does not justify them. It sometimes even leads to a thoroughly irrational decision process. For example, in caring for a dying, comatose patient, many physicians apparently find it harder to stop a functioning peripheral IV than not to restart one that has infiltrated (that is, has broken through the blood vessel and is leaking fluid into surrounding tissue), especially if the only way to re-establish an IV would be to insert a central line into the heart or to do a cutdown (make an incision to gain access to the deep large blood vessels).15
What factors might make withdrawing medical treatment morally worse than withholding it? Withdrawing a treatment seems to be an action, which, when it is likely to end in death, initially seems more serious than an omission that ends in death. However, this view is fraught with errors. Withdrawing is not always an act: failing to put the nest infusion into a tube could be correctly described as an omission for example. Even when withdrawing is an act, it may well be morally correct and even morally obligatory. Discontinuing intravenous lines in a patient now permanently unconscious in accord with that patient’s well-informed advance directive would certainly be such a case. Furthermore, the caregiver’s obligation to serve the patient’s interests through both acts and omissions rules out the exculpation that accompanies omissions in the usual course of social life. An omission that is not warranted by the patient’s interests is culpable.
Sometimes initiating a treatment creates expectations in the minds of caregivers, patients, and family that the treatment will be continued indefinitely or until the patient is cured. Such expectations may provide a reason to continue the treatment as a way to keep a promise. However, as with all promises, caregivers could be very careful when initiating a treatment to explain the indications for its discontinuation, and they could modify preconceptions with continuing reevaluation and education during treatment. Though all patients are entitled to expect the continuation of a particular mode of care.
Accepting the distinction between withholding and withdrawing medical treatment as morally significant also has a very unfortunate implication: caregivers may become unduly reluctant to begin some treatments precisely because they fear that they will be locked into continuing treatments that are no longer of value to the patient. For example, the physician who had been unwilling to stop the respirator while the infant, Andrew Stinson, died over several months is reportedly “less eager to attach babies to respirators now.”16 But if it were easier to ignore malnutrition and dehydration and to withhold treatment for these problems than to discontinue the same treatments when they have become especially burdensome and insufficiently beneficial for this patient, then the incentives would be perverse. Once a treatment has been tired, it is often much clearer whether it is of value to this patient, and the decision to stop it can be made more reliably.
The same considerations should apply to starting as to stopping a treatment, and whatever assessment warrants withholding should also warrant withdrawing.
The Obligation to Avoid Being the Unambiguous Cause of Death. Many physicians will agree with all that we have said and still refuse to allow a choice to forego food and fluid because such a course seems to be a “death sentence.” In this view death seems to be more certain from malnutrition and dehydration than from foregoing other forms of medical therapy. This implies that it is acceptable to act in ways that are likely to cause death, as in not operating on a gangrenous leg, only if there remains a chance that the patient will survive. This is a comforting formulation for caregivers, to be sure, since they can thereby avoid feeling the full weight of the responsibility for the time and manner of a patient’s death. However, it is not a persuasive moral argument.
First, in appropriate cases discontinuing certain medical treatments is generally accepted despite the fact that death is as certain as with nonfeeding. Dialysis in a patient without kidney function or transfusions in a patient with severe aplastic anemia are obvious examples. The dying that awaits such patients often is not greatly different from dying of dehydration and malnutrition.
Second, the certainty of a generally undesirable outcome such as death is always relevant to a decision, but it does not foreclose the possibility that this course is better than others available to this patient.17 Ambiguity and uncertainty are so common in medical decision making that caregivers are tempted to use them in distancing themselves from direct responsibility. However, caregivers are in fact responsible for the time and manner of death for many patients. Their distaste for this fact should not constrain otherwise morally justified decisions.
The Obligation to Provide Symbolically Significant Treatment. One of the most common arguments for always providing nutrition and hydration is that is symbolizes, expresses, or conveys the essence of care and compassion. Some actions not only aim at goals, they also express values. Such expressive actions should not simply be viewed as means to ends; they should also be viewed in light of what they communicate. From this perspective food and water and not only goods that preserve life and provide comfort; they are also symbols of care and compassion. To withhold or withdraw them – to “starve” a patient – can never express or convey care.
Why is providing food and water a central symbol of care and compassion? Feeding is the first response of the community to the needs of newborns and remains a central mode of nurture and comfort. Eating is associated with social interchange and community, and providing food for someone else is a way to create and maintain bonds of sharing and expressing concern. Furthermore, even the relatively low levels of hunger and thirst that most people have experienced are decidedly uncomfortable, and the common image of severe malnutrition or dehydration is one of unremitting agony. Thus, people are rightly eager to provide food and water. Such provision is essential to minimally tolerable existence and a powerful symbol of our concern for each other.
However, medical nutrition and hydration, we have argued, may not always provide net benefits to patients. Medical procedures to provide nutrition and hydration are more similar to other medical procedures than to typical human ways of providing nutrition and hydration, for example a sip of water. It should be possible to evaluate their benefits and burdens, as we evaluate any other medical procedure. Of course, if family, friends, and caregivers feel that such procedures affirm important values even when they do not benefit the patient, their feelings should not be ignored. We do not contend that there is an obligation to withhold or to withdraw such procedures (unless consideration of the patient’s advance directives or current best interest unambiguously dictates that conclusion); we only contend that nutrition and hydration may be foregone in some cases.
The symbolic connections between care and nutrition or hydration adds useful caution to decision making. If decision makers worry over withholding or withdrawing medical nutrition and hydration, they may inquire more seriously into the circumstances that putatively justify their decisions. This is generally salutary for health care decision making. The critical inquiry may well yield the sad but justified conclusion that the patient will be served best by not using medical procedures to provide food and fluids.
A Limited Conclusion
Our conclusion - that patients or their surrogates, in close collaboration with their physicians and other caregivers and with careful assessment of the relevant information, can correctly decide to forego the provision of medical treatments intended to correct malnutrition and dehydration in some circumstances - is quite limited. Concentrating on incompetent patients, we have argued that in most cases such patients will be best served by providing nutrition and fluids. Thus, there should be a presumption in favor of providing nutrition and fluids as part of the broader presumption to provide means that prolong life. But this presumption may be rebutted in particular cases.
We do not have enough information to be able to determine with clarity and conviction whether withholding or withdrawing nutrition and hydration was justified in the cases that have occasioned public concern, though it seems likely that the Danville and Bloomington babies should have been fed and that Claire Conroy should not.
It is never sufficient to rule out "starvation" categorically. The question is whether the obligation to act in the patient's best interests was discharged by withholding or withdrawing particular medical treatments. All we have claimed is that nutrition and hydration by medical means need not always be provided. Sometimes they may not be in accord with the patient's wishes or interests. Medical nutrition and hydration do not appear to be distinguishable in any morally relevant way from other life-sustaining medical treatments that may on occasion be withheld or withdrawn.
**REFERENCES – please see PDF file for references for this article**
Article: Caring or Starving? The Case of Claire Conroy
By Richard A. McCormick
America, April 6, 1985
Richard A. McCormick, S.J. is Rose F. Kennedy Professor of Christian Ethics at Georgetown University, Washington, D.C.
The courts have set a trend toward wider discretion in withdrawing or withholding treatment from terminally ill patients. This progression involves some obvious dangers.
Claire Conroy was an 84-year-old nursing home resident. She suffered from irreversible physical and mental impairments including arteriosclerotic heart disease, diabetes, and hypertension. She could neither speak nor swallow and was fed by a nasogastric tube. Her movements were very limited though she could smile or moan in response to some stimuli. She was restricted to a semifetal position and lacked control of her excretory functions. Thomas C. Whittemore, Miss Conroy’s nephew and guardian, requested that the nasogastric tube be removed from his awake but severely demented aunt. The application was opposed by Miss Conroy’s guardian “ad litem” (for purposes of litigation).
At trial, two physicians testified that Miss Conroy would die of dehydration in about a week after removal of the nasogastric tube. They also concurred in the opinion that her death would be painful. One physician regarded the nasogastric feeding as optional medical treatment. Miss Conroy’s own physician, however, believed that removal of the tube would be unacceptable medical practice. The trial court (Judge Reginald Stanton, Feb. 2, 1983) decided to permit removal of the tube because Miss Conroy’s life had become intolerably and permanently burdensome.
This decision was appealed by Miss Conroy’s guardian ad litem, but she died while the appeal was pending. However, the appellate division considered the matter too important to be left unresolved. It reversed the trial court’s judgement and stated that removal of the nasogastric tube would be tantamount to killing her. A guardian’s decision, the court argued, may never be used to withhold nourishment from an incompetent patient. As the court worded it: “The trial judge authorized euthanasia [homicide]… If the trial judge’s order had been enforced, Conroy would not have died as the result of an existing medical condition, but rather she would have died, and painfully so, as the result of a new and independent condition: dehydration and starvation. Thus she would have been actively killed by independent means.”
Mr. Whittemore took the question to the New Jersey Supreme Court, the same court that had decided the Karen Quinlan case. The court released its decision Jan. 17, 1985. After acknowledging the right of a competent adult to decline medical treatment – a right embraced within the common-law right of self-determination – the court addressed the rights of the incompetent. It noted: “The right of an adult who, like Claire Conroy, was once competent, to determine the course of her medical treatment remains intact even when she is no longer able to assert that right or to appreciate its effectuation.”
Clearly a substitute decision-maker or proxy must be called upon to function at this point. May proxy ever decide that life-sustaining treatment may be withheld or withdrawn from an incompetent but not comatose patient? The court responded in the affirmative and proposed three tests or standards corresponding to three different situations. First, there is the “subjective standard,” under which life-sustaining treatment may be withheld or withdrawn “when it is clear that the particular patient would have refused the treatment under the circumstances involved” (emphasis added). This clear intent can be concluded from written directives (living will) or oral statements made to family, friends, or health providers. It might also derive from a durable power of attorney or appointment of a proxy with authorization to make medical decisions on the patient’s behalf.
Second, there is the “limited objective test.” Life-sustaining treatment may be withheld or withdrawn from a patient like Claire Conroy when there is trustworthy evidence that the patient would have refused the treatment and the proxy is satisfied that the burdens of the patient’s continued life with the treatment outweigh the benefits of that life for the patient.
Finally there is the “pure objective tests.” Under this test the burdens of the patient’s life with treatment should clearly and markedly outweigh the benefits the patient derives from life. Furthermore, the unavoidable and severe pain of the patient’s life with treatment must be such that continued life-sustaining treatment would be inhumane.
In elaborating its decision, which was a reversal of the appellate division’s judgement that cessation of artificial feeding was a killing act, the New Jersey Supreme Court made several interesting points. First, it stated that the record in the Conroy case did not satisfy the standards prescribed by the opinion. Second, it rejected several distinctions as analytically unhelpful in this case: the distinction between actively hastening death and passively allowing a person to die; the distinction between withholding and withdrawing; the distinction between ordinary and extraordinary means. For instance, the court viewed the active-passive distinction as “elusive” and “particularly nebulous” where withholding or withdrawing life-sustaining treatment is concerned. It stated: “In a case like that of Claire Conroy, for example, would a physician who discontinued nasogastric feeding be actively causing her death by removing her primary source of nutrients; or would he merely be omitting to continue the artificial form of treatment, thus passively allowing her medical condition, which includes her inability to swallow, to take its natural course?”
Third, the court stated clearly that artificial feeding by nasogastric tube or intravenous infusion is equivalent to artificial breathing by a respiratory. In other words, it is a medical procedure and should be provided or withheld according to the criteria applicable to medical procedures.
Finally and very importantly, it stipulated a procedure to be followed in cases like that of Claire Conroy. The person (e.g. family member, guardian, physician) who believes that withholding or withdrawing life-sustaining treatment corresponds to the patient’s wishes or would be in her/his best interests must notify an ombudsman. Those with contrary beliefs should do the same. The ombudsman is to treat every such notification of withholding or withdrawing as a possible abuse. Two physicians unaffiliated with the nursing home and with the attending physician must confirm the patient’s medical condition and prognosis.
The decision is, in a sense, a linear descendent of a previous case involving withdrawing or feeding. Clarence Herbert underwent surgery for closure of an ileostomy at Kaiser Permanente Hospital, Harbor City, Calif., in 1981. Shortly after successful completion of the surgery, Herbert suffered cardiorespiratory arrest. He was revived and immediately placed on life-support equipment. Within the following three days it was determined that Mr. Herbert was in a deeply comatose state from which he was unlikely to recover. Tests performed by several physicians indicated that he had suffered severe brain damage, leaving him in a vegetative state that was likely to be permanent.
At that time Mr. Herbert’s physicians, Dr. Robert Nejdl and Dr. Neil Barber, informed his family of his condition and the extremely poor prognosis. The family then drafted a written request to the hospital personnel stating that they wanted “all machines taken off that are sustaining life.” Dr. Nejdl and Dr. Barber complied and removed Mr. Herbert from the respiratory. He continued to breathe. After two more days, the two physicians, after consulting with the family (though the record is a bit hazy here), ordered removal of the intravenous line and nasogastric tube that provided hydration and nourishment. Shortly thereafter Mr. Herbert died.
Dr. Nejdl and Dr. Barber were accused of murder by the Los Angeles District Attorney. Los Angeles Municipal Judge Brian Crahan dismissed the case. It was reopened (May 5, 1983) by Superior Court Judge Robert A. Wenke on the grounds that the dismissal was erroneous. The Herbert case received widespread publicity. The implications of Judge Wenke’s decision were stated simply by Dr. Barber: “No doctor will take a patient off a respirator now.”
The matter eventually reached the Court of Appeal. On Oct. 12, 1983, Judge Lynn Compton exonerated Dr. Nejdl and Dr. Barber of any unlawful conduct. In the course of this opinion, the court made several interesting and important points. First, Judge Compton noted that even though life-support devices are self-propelled, still each drop of IV fluid is “comparable to a manually administered injection or item of medication.” Hence disconnecting such devices is “comparable to withholding he manually administered injection.” Second, the court viewed intravenous nourishment and fluid as “being the same as the use of the respirator.” Third, medical nutrition and hydration resemble medical procedures rather than typical ways of providing nutrition and hydration. Hence they are to be evaluated in terms of their burdens and benefits. Finally, since the court viewed the physicians’ actions as omissions rather than affirmative actions, the resolution of the case depends on whether there was a duty to continue to provide life-sustaining treatment. The court asserted that there is no such duty once the treatment is useless. And it was useless in Herbert’s case because it merely sustained biological life with no realistic hope of a return to a cognitive, sapient state. Thus, continued use of life sustainers was “disproportionate.”
These two cases have a key difference. Clarence Herbert was judged to be in a permanent vegetative state. Claire Conroy was not. She was incompetent but not comatose. Of those in a permanent vegetative state, the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research wrote in Decisions to Forgo Life-Sustaining Treatment, 1983, 190: “Most patients with permanent unconsciousness cannot be sustained for long without an array of increasingly artificial feeding interventions – nasogastric tubes, gastrostomy tubes, or intravenous nutrition. Since permanently unconscious patients will never be aware of the nutrition, the only benefit to the patient of providing such increasingly burdensome interventions is sustaining the body to allow for a remote possibility of recovery. The sensitivities of the family and of care-giving professionals ought to determine whether such interventions are made.”
A footnote to this last sentence notes that it can be anticipated that courts will grant requests to withhold or withdraw further treatment, including IV drips, from such patients. And that is just what the court did in the Herbert case. But the New Jersey Supreme Court also did the same thing in principle to the Conroy case, and for some of the same reasons. That is, both courts regard feeding by IV lines and nasogastric tubes as basically medical procedures to be judged by a burdens-benefits calculus. Furthermore, both courts (to a lesser degree the New Jersey Supreme Court) allow “quality-of-life” components to function in determining the best interests of the patient.
There is another feature of these decisions that could easily be missed. Both represent reversals of lowe courts that held that withholding or withdrawing nutrition-hydration from comatose and/or incompetent patients would constitute murder. I think it fair to say that these judicial disagreements reflect the state of ethical discussion that preceded them and still surrounds them.
One of the first shots in this discussion was fired by hospice nurse Joyce V. Zerwekh. She argued in Nursing, January 1983, that it is not always more merciful to administer IV fluids to a dying patient. There are both beneficial and detrimental effects associated with dehydration and the judgement must be individualized.
Since the Zerwekh study, the literature has piled up impressively. For instance, Kenneth Micetich, MD, Patricia Steinecker, MD, and ethicist David Thomasma (all of Stritch School of Medicine, Loyola University, Chicago) concurred that IV fluids may not be morally required under a threefold condition: 1) the patient must be dying: “Death will be imminent (within two weeks) no matter what intervention we may take.” 2) The patient must be comatose. Comatose patients would experience no pain, thirst, etc. 3) The family must request that no further medical procedures be done in the face of impending death (Archives of Internal Medicine, 1983).
James Childress of the University of Virginia and Joanne Lynn, M.D., of George Washington University carried the matter a step further. They argued that there are cases, even though relatively few, when it is in the best interests of patients to be malnourished and dehydrated. They listed three situations: 1) The procedures that would be required could be considered futile. 2) The improvement in nutritional and fluid balance, though achievable, could be of no benefit to the patient (e.g. persistent vegetative state). 3) There are cases where the burdens to be borne in receiving the treatment may outweigh the benefit. Terminal pulmonary edema, nausea, and mental confusion may be more likely in some patients as a result of artificial hydration and nutrition (Hastings Center Report, October 1983).
Even more recently a group of distinguished clinicians advocated the withholding of parenteral fluids and nutritional support from severely and irrevocably demented patients, and occasionally, from elderly patients with permanent mild impairment of competence, a group characterized as “pleasantly senile” (New England Journal of Medicine, 1984).
Such voices have not gone unchallenged. Daniel Callahan (Hastings Center) agrees that it is morally licit to discontinue feeding in the circumstances noted by Lynn and Childress. Yet he is profoundly uneasy with that conclusion. The feeding of the hungry, whether they be poor or physically unable to feed themselves, is “the most fundamental of all human relationships” (Hastings Center Report, October 1983). It is, he argues, extremely dangerous to tamper with so central a moral emotion. There remains a deep-seated revulsion at stopping feeding even under legitimate circumstances. As I read him, Mr. Callahan would respect that revulsion and continue feeding as “a tolerable price to pay to preserve – with ample margin to spare – one of the few moral emotions that could just as easily be called a necessary social instinct.”
Gilbert Meilaender of Oberlin College carries this a step further. He argues that the withdrawal of nourishment from permanently unconscious patients involves us in “aiming at their deaths.” This we should never do. Nor does their permanent comatose state mean that it is useless to feed them. In these cases, feeding remains care for the embodied person, and it is dualistic to think otherwise. Nor is the care “in any strict sense medical treatment.” It treats no particular diseases; rather “it gives what all need to live” (Hastings Center Report, December 1984).
A physician (Mark Siegler) and an attorney (Alan J. Weisbard) are deeply troubled by the emerging literature justifying withholding or withdrawing of hydration and nutritional support (Archives of Internal Medicine, January 1985). They reject the idea that anyone (physicians, families, courts) can properly make such judgements for the incompetent. Therefore, they want to reverse the stream of this literature and offer several arguments to bolster this reversal. First, patients will be protected against diagnostic errors, inadequate treatment and unscrupulous (e.g. for financial reasons) care. Second, physicians will not be compelled to make ad hoc, quality-of-life judgements. Third, the medical profession will be protected against the gradual dilution of its dedication to the welfare of the patients. Finally, society will benefit by rejecting this practice because it bears the seeds of unacceptable consequences (e.g. devaluation of the unproductive).
This is a sampling of the literature that preceded and still surrounds the Claire Conroy decision. Where does it leave us? With several key issues and a final caution:
When artificial nutrition and hydration are medical procedures or not is often confused by the introduction of terms such as hunger and thirst. People can be denied artificial nutrition and hydration without experiencing hunger and thirst. Conversely, they can feel hunger and thirst without being malnourished or dehydrated. The usage “feed the hungry and thirsty” in this context tends to predispose us to regard artificial nutrition and hydration as something other than medical procedures.
On the other hand, if the patient is viewed as nondying, and artificial provision of nutrients is not a medical procedure but rather an instance of “normal, ordinary care,” then omission of such nutrients could be more suspect.
Let me put it this way. Those who contend that withdrawing hydration and nutrition involves us in “aiming at death” or being involved in “the direct causal responsibility for death” (Siegler, Weisbard) must be consistent and apply their analysis to the competent patient. Is a competent patient who refuses a nasogastric tube or a gastrostomy guilt of suicide (“aiming at his own death”)? Most of us would and should answer: “It all depends.” If the patient can be tided over a transitory illness and returned to normal life, the treatment would surely be obligatory and refusal of it, other things being equal, would be suicidal. If however, this is not the case and the artificial feeding is foreseeably permanent, the treatment would be morally optional. Being such, it would not necessarily involve death-aim. It need involve only a thoroughly Christian assertion that there are values greater in life than living, that we all retain the right to decide how we shall live while dying.
However, an ambiguity remains. What is to count as burden, and correlatively as a benefit? If a patient’s life can be prolonged, but only in a comatose state, is that a benefit to the patient? Or if a treatment will preserve life, but only a pain-racked, incompetent life, is that a benefit?
The issue here is this: In weighing the burdens-benefits of a treatment, is it the burden of the treatment only (e.g. its pain, expense etc.) that is legitimately considered, or may we include in the assessment the burden of continued existence itself? In other words, may the quality of life preserved be a proper dimension of the calculus?
The President’s commission answered this last question in the affirmative when it defined the patient’s best interest broadly to “take into account such factors as the relief of suffering, the preservation or restoration of functioning, and the quality as well as the extent of the life sustained.” Both the Herbert court and the Conroy court did the same. For instance, the Conroy court (New Jersey Supreme Court) acknowledged that “although we are condoning a restricted evaluation of the nature of a patient’s life in terms of pain, suffering and possible enjoyment under the limited-objective and pure objective tests, we expressly decline to authorize decision-making based on assessments of the personal worth or social utility of another’s life, or the value of that life to others.”
Some continue to attempt to finesse this extremely difficult and delicate issue by conceptualizing decisions in terms of “ordinary” and “extraordinary” means. But it will not work. Such terms only disguise the quality of life component unavoidably present in some of these decisions. For this reason nearly every recent commentator would agree with the President’s commission when it stated: “The claim, then, that the treatment is extraordinary is more of an expression of the conclusion than a justification for it.”
My own opinion on these issues is that the permanently comatose and some noncomatose but elderly incompetent patients may be classified broadly as dying; that feeding by IV lines and nasogastric tubes is a medical procedure; that its discontinuance need not involve aiming at the death of such patients; and that the burden-benefit calculus may include, indeed often unavoidably includes, a quality-of-life ingredient, providing we draw the line at the right place.
And that brings us to the caution. The Claire Conroy case and the decision of the New Jersey Supreme Court may appear to be isolate instances. That is not the case. There are many thousands of nursing home residents like Claire Conroy. They are a terribly vulnerable population. They are elderly and often suffer from crippling disabilities. They often are without surviving relatives. Physician play a more limited role in nursing homes than in hospitals, and patient advocacy is correspondingly less intense. Furthermore, as the New Jersey Supreme Court notes, nursing homes are often afflicted with industry-wide problems that make them a very troubled and troublesome component of the health care system. And all of this at the very time when there are economic and social pressures on health care delivery. Together these factors may make it extremely difficult to keep patients’ best interests at the heart of these decisions. In other words, the potential for abuse is enormous.
This suggests the wisdom of Daniel Callahan’s statement at a spring conference on these problems in 1984: “The time to curtail abuses in the future is to begin now in trying to go through those steps that will draw lines very carefully.” We have moved from Quinlan (persistent vegetative state – removal of respirator, nasogastric tube, and IV lines) to Conroy (incompetent but noncomatose – removal of nasogastric tube). The progression is obvious, and obviously dangerous, unless we draw clear lines based on clear criteria. If we do not, we will not long be confined within the limits set forth in cases like Conroy.
Technology can help us or hurt us, individually and societally, and in ways that are awesome in their implications. The recent technological revolution in methods of hydration and nutritional maintenance is a case in point. That is why the ultimate caution was well stated by the New Jersey Supreme Court: When evidence of a person’s wishes is equivocal, or the best interests determination doubtful, “it is best to err, if at all, in favor of preserving life.”
CONROY: Discussion Questions
How would the argument be made if Claire Conroy had never been competent?
employed regarding proxy consent to withholding/withdrawing food and nutrition: the "subjective
standard," the "limited objective test" and the "pure objective test." Under what conditions would
each be used?
Annas, George J. 1985. "From Quinlan to Conroy." Hastings Center Report 15 (April): 24-26.
McMahan, Jeff. 1993. "Killing, Letting Die, and Withdrawing Aid." Ethics (January): 251-279.
Steinbrook, Robert and Bernard Lo. 1988. "Artificial feeding-Solid Ground, Not a Slippery Slope." NEJM318 (February 4): 286-290.