Quality of Life: The Contested Rhetoric of Resource Allocation and End-of-Life Decision Making

Abstract

The term “quality of life” has a long history in the bioethics literature. It is usually used in one of two contexts: in resource allocation discussions in the hope of arriving at an objective measure of the worth of an intervention; and in end-of-life discussions as a concept that can justify the forgoing of life-sustaining treatment. In both contexts, the term has valid uses as it is meant to measure the efficacy of a treatment. However, the term has the unfortunate rhetorical problem that it often seems to be a judgment on the life of a human being. As such, it is highly inflammatory. We suggest that a return to a rhetoric that suggests a judgment on the treatment rather than the person is needed.

Keywords:

• bioethics
• disability
• quality of life
• resource allocation
• rhetoric
• substituted judgment

I. INTRODUCTION

As the population of the United States ages and the cases of chronic illness increases, quality of life questions become more important in decisions about medical treatment. The goal of the medical profession has been shifting from “curing” patients to helping patients live as comfortably as possible with a chronic condition. Since full recovery of an elderly or disabled patient is not likely, treatment decisions are being made based on the level of human functioning that can be restored in the patient. Will a patient be able to function independently in society, or will he or she live a life in which care is given in a relatively removed location? What are the life goals of the patient and how are these thwarted or advanced through a particular medical treatment? All of these questions are gathered under the ambiguous umbrella term, “quality of life.”

What is meant by the term, “quality of life?” A cursory glance at the medical and bioethical literature reveals that there is no easy answer to this question. Yet, because of our aging population and the increase in chronic illness, finding a way to clarify questions about quality of life becomes increasingly important.

II. THE STANDARD APPROACH: QUALITY OF LIFE AS RESTORATION OF NORMAL FUNCTIONING

We will first examine a very standard (and standardized) model of quality of life. Early quality of life research focused on the level of “normal” functioning restored to a patient after treatment (Musschenga, 1997, p. 14). Did the medical treatment allow the patient to function in society and perform everyday necessary tasks such as bathing, walking up stairs, or dressing? Physicians could observe a patient’s effectiveness at these tasks and determine if the medical treatment was working or that further treatment was warranted.

This “objective” model became more important as medicine became more focused on clinical protocols and economics (Morreim, 1992, p. 228). There is a growing repository of research that has been conducted on the effectiveness of many medical procedures. Physicians have long relied on this research when making decisions about treating a particular patient. If, according to the latest clinical research, less than 10 percent of patients more than 60 years old who undergo a certain medical procedure show an improvement in their health, then this is likely to weigh heavily in a physician’s treatment decisions.

Medical insurance companies, HMOs, and PPOs also focus on clinical research and the statistical effectiveness of medical procedures. The economic influence wielded by such groups factors greatly into treatment decisions. The onus is placed upon the physician to prove that a certain course of treatment is justified for the time, expense and resources it will use. Under this model, statistical data outweigh individual circumstances in decisions regarding medical treatment. Doubtless, this has sometimes left physicians and patients frustrated with the insurers.

Of course, this approach to quality of life is in terms of a medical model. On a medical model, problems such as disability are conceived as residing within the patient. Such problems need to be “treated” by medicine. Treatment efficacy is measured by the change in the patient. It is implicit within such a model that treatment should cease at some point, usually relatively quickly, because the patient gets “better.” In other words, the medical model operates most comfortably with an acute model of illness and quality of life is of interest in order to measure changes that are brought about by the catalyst of technology. The medical model and acute care dominate most approaches to quality of life.

There have been various attempts by medical professionals and ethicists to develop a standardized objective method for measuring quality of life. These measures, presumably, could be implemented to decide if a particular treatment will be effective in adding quality to a patient’s life. Some of these methods are controversial because they attempt to place a symbolic value on human life, whether it be economic value or health-related value (Beauchamp & Childress, 2001, p. 207). These measurements of quality of life are an attempt to allocate scarce medical resources more fairly. This is especially important from a public health perspective, which seeks to “. . . optimize the health status of the community’s population” (Morrow & Bryant, 1995, p. 1356).

One well-known attempt at measuring quality of life is the quality-adjusted life-years (QALYs) model. QALYs are based on a scale from 0 to 1. A score of “0” signifies death and a score of “1” signifies a year of healthy life. Patients with chronic conditions are assigned a number within this range, depending upon the severity of their condition. Decisions to treat a patient or not depend upon how much the treatment in question can improve the patient’s score on the QALY scale. “The value of particular health outcomes depends on the increase in the utility of the health state and the number of years it lasts” (Beauchamp & Childress, 2001, p. 209). One important critique of the QALY model is that, while it may effectively measure improvements in a patient’s physical state, it may not truly represent a patient’s quality of life as the name promises. There are many other factors beyond physical health that should be weighed when considering a patient’s quality of life. This realization has given rise to using the more precise term health-related quality of life (HRQL) (Beauchamp & Childress, 2001, p. 210).

Health-related quality of life (HRQL) has emerged as a popular phrase in medical literature. However, there are differences of opinion among medical researchers regarding the definition of HRQL, as well as the suitability of HRQL as a true measure of a patient’s “quality of life.” Some researchers claim that HRQL is what is commonly meant by the phrase “quality of life.” “Health-related quality of life is increasingly recognized as a major outcome in patient care and clinical research and as a term embraces the concepts of functional status, health status, quality of life, and well-being” (Spitzer et al., 1995, p. 1511). Others insist that HRQL research puts too much emphasis on health status and a patient’s ability to function, which is not what most people mean by quality of life.

HRQL research contains an underlying assumption that a person’s physical abilities are directly related to their quality of life. This is, according to some researchers, faulty logic because, “This view fails to acknowledge the interconnectedness of health status with other aspects of existence such as changes in income, work status, personal relationships, coping strategies, responsibilities, self-image, and customary modes of being” (Leplege, 1997, p. 47). Still other researchers admit that quality of life is a very general concept and no instrument has yet been developed that can completely measure a patient’s quality of life.

A more focused approach to quality of life must be taken when conducting quality of life research and some have attempted accordingly to refine HRQL. “One specific focus is health-related quality of life (HRQL), which refers to the measurement of the quality of those aspects of a person’s life that impact directly upon their health” (Dempster & Donnelly, 2000, p. 46). These aspects include physical, social and psychological functioning. However, there is no consensus among researchers regarding how each of these should be measured.

We can see that part of what is at issue is whether the medical model is an adequate framework for assessing quality of life. Of course, it is true many of the things that contribute to a person’s quality of life have nothing to do with health directly. Even more to the point, many of the health-related matters that contribute to quality of life produce no actual change in the patient. That is, a wheelchair for a person who cannot walk and a van that is modified for a wheelchair produce no immediate measurable change in the health of the patient. Yet, they are clearly related to the person’s quality of life. And it is odd if health-related quality of life measures should fail to capture the change these produce since the impediments to the person’s quality of life that these assistive devices remedy, are health-related. Again, the measure seems to capture too little, and it is not just anything that is left out. As we noted earlier, the rhetoric of the medical model of quality of life seems predisposed to making quality of life something that resides in the person and fails to capture the possibility that quality of life is a complex set of interactions and relations.

III. NON-STANDARD OR SUBJECTIVE APPROACHES TO ASSESSING quality of life

Another way to assess a patient’s quality of life is to ask him or her. Many quality of life researchers have proposed this more “subjective” approach in recent medical literature (Carr & Higgenson, 2001, p. 1357; Leplege & Hunt, 1997, p. 47; Musschenga, 1997, p. 20). The support for patient autonomy in Western medicine has propelled subjective quality of life judgments into the limelight. The subjective model is often dependent upon a patient’s life goals and her ability to meet them (or not) once she has been inflicted with a medical problem. From a medical viewpoint, a patient’s biological health may be improving, but the patient may still judge his or her quality of life to be poor because he or she is unable to achieve the life goals he or she aspired to prior to her illness. Still another patient may be very satisfied with his or her health progress and therefore judge his her quality of life to be excellent.

Individual values and beliefs have an important role in defining quality of life. Therefore, health care providers have to focus on the person who is sick or who has a chronic illness and what is important to that individual; they cannot focus solely on treating the illness. (Erlen, 1996, pp. 78–79)

Opening the discussion to include patient views of their own quality of life is an important advancement. According to some researchers, this is the only way to gain a valid measure of quality of life. “A true assessment of quality of life can only be achieved using weights for individual patients” (Carr & Higginson, 2001, p. 1359).

Looking to patients’ self-assessments of quality of life leads one to ask, “How do patients evaluate their own quality of life?” Some recent papers by Carr and colleagues proposed that patients make quality of life assessments by comparing their expectations of health with their experience. Every human being will perceive some change in his or her experience of health, whether because of the onset of chronic illness, or due to the normal aging process. Expectations will also change throughout a lifetime, and may shift rather dramatically during the course of an illness.

“When expectations are matched by current experience, there is no quantifiable impact on quality of life. Whenever the experience of health falls short of expectations there is an impact” (Carr, Gibson, & Robinson, 2001, p. 1241). Over time, if an illness is chronic, a patient may be able to lower his or her expectations toward more realistic goals, thus having the effect of raising his or her quality of life.

Still another group of people asked to make quality of life assessments are those who are closest to a patient, usually a spouse, sibling, or parent. These health care proxies are called upon to make decisions about medical treatment when the patient is unable to do so. The assumption in these cases is that, in the absence of input by the patient, the proxy has the best understanding of what is in the patient’s best interest. There have been some studies conducted that test the validity of this assumption. These studies show that third parties often rate the patient’s quality of life much lower than the patient does (Gerhart, Koziol-McClain, Lowenstein, & Whiteneck, 1994; Gill & Feinstein, 1994). However, at least in the case of end-of-life treatments, surrogates may do a reasonable job of guessing the patient’s wishes (Sulmasy, Heller, & Terry, 1994).

The “problem” regarding subjective quality of life measures is that these measures may have a difficult time serving the resource allocation function that drove the development of objective quality of life measures. That is, if each person is the only judge of how a treatment affects his or her quality of life, then we cannot say in advance whether we should fund all instances of treatment-x for condition-y. Only each person can tell us whether such a treatment is worthwhile for him or her. However, it may be possible to generalize from a sample of subjective assessments.

Perhaps resource allocation decisions cannot be individually subjective but may be subjective across a class of persons. That is, perhaps persons who are unable to walk can tell us which treatments and what kinds of assistance typically improve their quality of life most. Because persons who do not have a particular disability or long-term illness underestimate the quality of life of persons who do relative to their subjective judgments, what subjective judgments essentially do is define the class of relevant opinions in calculating efficacy. Such a model allows that any individual person with an illness or disability may have completely idiosyncratic assessments but does not base allocation decisions on that particular assessment, only on the general assessment.

Subjective assessments of improvements in quality of life hold promise for allocation decisions within a group that suffers from a particular illness. That is, it seems justifiable to consult a group of persons who suffer from a particular illness and find out how they assess the relative merits of various treatments and support services in terms of how they add to their quality of life. And, such a subjective model is likely to help us advance beyond the medical model since a subjective assessment of quality of life allows for improvement without that improvement necessarily being an improvement in one’s medical condition. It is when medicine and medical researchers alone frame the questions that results are likely to be restricted to medical interventions.

However, it is less promising to ascertain how treatments and services for different groups compare based on subjective assessments of quality of life. That is, although subjective assessments may show a particular treatment or service to be of little or no value, it would be extremely difficult to know how two moderately regarded treatments for different conditions actually could be justifiably ranked based on these relative assessments. Nevertheless, in assessing the strengths of the subjective view of quality of life, we have also uncovered what is most typically objectionable about quality of life measures.

It is easy to see why quality of life is a contentious term in the resource allocation literature. It sounds as if one is making a judgment about another person’s life and judging that person to have failed a test with the result that they are deemed unworthy of treatment. Of course, this is a skewed characterization as it is not the patient’s quality of life that is directly supposed to be at issue. It is the efficacy of a treatment or intervention in raising the patient’s quality of life that is important. However, we have seen that the critics of quality of life clearly have made two important points.

1. Because quality of life is meant to be an objective measure, it can lead to skewed calculations if it is based on views regarding the quality of life of persons with disabilities of which the general public has little experience. In fact, it aids in framing all questions in terms of a medical rather than a social model of illness and disability.

2. The term quality of life tends to focus the attention on the person and not on the treatments or interventions being offered. Comfort with such a focus is distinctively un-American, as we shall demonstrate more clearly in the next section. Of course, the place that disability advocates fear the slippery slope ends is in inhumane quality of life judgments in end-of-life care.

IV. END OF LIFE CARE AND THE RHETORIC OF QUALITY OF LIFE

The ethical and legal consensus regarding forgoing medical treatment in the United States is based upon the concept of informed consent and has thereby sought to avoid one person making quality of life judgments about another. There has always been an implicit desire in our American ethos to avoid quality of life considerations in making such decisions. After all, we are each entitled to the “pursuit of happiness.” The concept of happiness is something to be defined primarily by each individual, not by the state. Since the state is meant to safeguard life and liberty in addition to the pursuit of happiness, there is something contradictory about anyone saying that another person’s life is not of high enough quality to merit the state’s protections. Of course, we can each judge our own happiness and the means to our self-determined ends.

The doctrine of informed consent, an ethical and legal amalgamation of principles and practices, asserts that the individual has a virtually unlimited right to refuse medical treatment. Because each person can determine his or her own concept of the good life, one can refuse medical treatments that he or she judges not to contribute to one’s well being (Faden & Beauchamp, 1986). And, one can even judge that one’s own quality of life is low enough that extending that life through medical treatment is not warranted. Although others can try to persuade one differently, they are not empowered to override the refusal of a person of a sound mind who makes such a decision. Unfortunately, the paradigm instance of treatment refusal, that is, a competent conscious patient who makes an informed refusal, can be difficult to adapt to all relevant situations and have led many to advocate quality of life judgments as a supplement (Drane & Coulehan, 1995; Munetz, Lidz, & Meisel, 1985).

The doctrine of informed consent rests upon the assumption that a person’s wishes are knowable and accessible. However, in many, many cases, this assumption has proven unfounded in practice. Wishes can prove unstable and patients may have difficulty translating their values into particular treatment choices (Forrow, 1994; Kuczewski, 1996). Furthermore, the patient’s capacity to make a decision is often in question, the patient may have left an advance directive but its applicability to the present situation may not be clear, either because the circumstances of the current situation may not have been anticipated or the patient’s personality has been so altered by injury that the warrant for applying the directive is in question. Finally, there are also cases such as those of babies, young children, and persons with congenital mental disorders who have never attained decision-making capacity. In all of these cases, relying solely on the patient’s self-determination is either very difficult or impossible.

In the absence of clear wishes from a patient with decision-making capacity, it is natural to ask the question “what kind of life” the patient has and whether the treatment contributes anything to this life. As a result, we often hear people asking whether the patient’s quality of life justifies further intervention. Of course, people often do not speak in complete sentences or whole thoughts. As is often the case, the term “quality of life” comes front and center and the reference to interventions falls away. We hear people say “what about his quality of life?” And, of course, the expressions concerning quality of life can suggest that that life is not worth living. It is this usage of quality of life terminology that is especially controversial.

As with quality of life judgments in resource allocation, the main concern in end-of-life care is that persons with power, in this case physicians, will deem the lives of persons with disabilities not worth living (Smith, 2000). This has fostered the controversy known as the “futility” debate (Helft, Siegler, & Lantos, 2000).

Physicians sometimes believe it is best for a patient to have a do-not-resuscitate order that is in effect in the case of cardiac arrest or to withhold or withdraw certain treatment that is judged ineffective. When families (assuming the patients are unconscious) disagree, many physicians and ethicists have argued for their right to impose a unilateral DNR order. It is unilateral because it is without the shared decision making that is the ideal of informed consent (Lidz, Appelbaum, & Meisel, 1988). In other words, futility theorists argue that if a treatment is “futile,” a physician has no obligation to offer it (Layson & McConnell, 1996).

Futility can be defined in many ways. The least controversial definition is that a treatment simply won’t work. Vasopressors won’t maintain the blood pressure, resuscitation will not restart the patient’s heart, and so on. This kind of physiological futility is fairly uncontroversial (Youngner, 1988). The more controversial definitions include a quality of life judgment by the physician. That is, some argue that we can define a series of conditions such as “continued biologic life without conscious autonomy” or “conditions requiring constant monitoring, ventilatory support, and intensive care nursing” (Schneiderman, Jecker, & Jonsen, 1990, p. 952). These particular authors exclude conditions such as severe mental impairments or continuous technological dependence if the person can be discharged from the hospital for periods.

It’s not hard to see the problem. Although only the most extreme vitalists would argue that society has an absolute obligation to provide every means of medical intervention to a person who is permanently unconscious, there is still a sense of concern about others beside the patient making the judgment that that life is somehow less worthwhile (Sprangers & Aaronson, 1992). It is because of this basic discomfort that we commonly pursue attempts to base decisions on interpretations of the patient’s prior values, however difficult such an enterprise may be.

Some prominent ethicists have argued that we cannot avoid the concept of quality of life (McCormick, 1978). That is, whenever a patient himself or herself makes a judgment that he or she should forgo a life-sustaining intervention, or we make such a judgment concerning a newborn or an unconscious person whose wishes are sketchy, we have implicitly said that the quality of life is too low to warrant the burdens imposed by the intervention. Note this formulation. For it is not necessarily the case that their quality of life makes them somehow unworthy of the intervention itself. This would seem to be the crux of the matter in the often-vitriolic controversies about quality of life.

Traditional Catholic moral theology from which the United States drew its understanding of forgoing life-sustaining treatment (Kelly, 1991, pp. 1–15) focused on the burdens and benefits of treatment, not on the life of the person herself. These burdens and benefits can be considered in terms of pain and suffering as well as intrusions on the dignity of the person. Unfortunately, this approach has also succumbed to a variety of misunderstandings such as becoming overly medicalized and key terms such as “ordinary” and “extraordinary” becoming attached to particular kinds of treatments or become synonymous with such red herrings as “usual” and “unusual” (Beauchamp & Childress, 2001, pp. 123–125).

Discussions about forgoing life-sustaining treatment are usually conducted in terms of life, death, and pain and suffering. Treatment is thought to extend life and terminating it to result in death. Burdens of the treatment are usually judged in terms of pain inflicted. Benefits are either conceived in terms of life extension and/or providing some additional opportunity for the patient’s physical condition to improve. When a discussion of forgoing life-sustaining treatment becomes reasonably sophisticated, some attempts at assigning probabilities to improvement of the patient’s condition are made. When quality of life is brought up, it is often in regard to the worthwhileness or lack of worthiness of having the patient’s life extended.

Of course, the hospice movement has tried to counter this by conceiving of a quality life or a quality death in broader terms. Considerations of what technology impedes the patient and family from achieving, for example, the intimacy of physical contact, taking the patient out of the ICU, are relevant (Byock, 1997). The issue is not simply whether a ventilator inflicts pain but that it also leads to the sedation of the patient, an unresponsiveness that can be a sacrifice of valuable interactive time. The question becomes whether an intervention is proportionate to the needs of the patient and his or her family, not whether the patient’s life is worthwhile. But can we reinterpret the rhetoric of quality of life to foster this broader vision of our medical technology or has quality of life become irretrievably linked to a medical model of care? We believe that little good is likely to result from trying to salvage the rhetoric of quality of life.

V. THE RHETORIC OF QUALITY OF LIFE

It is clear that the term quality of life is inflammatory. Quality of life suggests that one person can or will be making a judgment about the value, meaning, happiness, and/or worthwhileness of another’s life. Terminology that carries such connotations is unlikely to be helpful in producing consensus on sensitive issues such as resource allocation and end-of-life care.

We do not mean to suggest that all of the uses of quality of life measures are illicit. As we noted, it is helpful to evaluate technology in terms of whether and how much it improves the lives of people. After all, what else is technology for? Nevertheless, the language of quality of life focuses the discussion on the life of the person when what we are interested in is the worthiness of the treatment, intervention, or assistance being provided. Because treatments can only be evaluated in relation to their effect on the patient, the intervention-patient dyad is indivisible. We cannot analyze one without consideration of the other. Nevertheless, our language tends to facilitate an emphasis on one side of the dyad or the other. Quality of life creates an inordinate focus on the patient side. We seek a return to terminology that focuses on the treatment side.

We believe that, in general, we need to reinvigorate the use of terms such as ordinary and extraordinary care or proportionate and disproportionate care. This terminology began to fall from general favor in the United States with the famous case of Karen Ann Quinlan and continued its decline into the early 1990s. In general, the rhetoric of extraordinary treatment was thought to focus too heavily on the treatment side of the dyad. In fact, the rhetoric suggested that one could identify a treatment modality such as a ventilator or feeding tube as ordinary or extraordinary in and of itself, apart from considerations of the effect of the treatment on particular conditions (Meisel, 1995, pp. 481–486). Nevertheless, this is clearly not what the terminology was meant to do. By saying a treatment is ordinary in a particular instance is a statement about the moral obligation the patient has to himself or herself and that society has to support such care. Of course, these misunderstandings are also likely to be frequent and common if the use of the terminology is revived.

Any rhetoric will be misunderstood from time to time, perhaps frequently. Such is the case with quality of life. The question regarding any rhetoric is whether it generally fosters greater understanding or obscures, misleads, and divides so frequently that it should be discarded. We have tried to show that we believe the rhetoric of quality of life to have proven unhelpful overall and its misunderstandings to be contrary to our self-understanding as Americans. As a result, we counsel a return to treatment-centered terminology such as ordinary and extraordinary care, proportionate and disproportionate treatment. History has shown this latter rhetoric to be subject to frequent misinterpretation as well. However, these misunderstanding still end up being errant judgments on medical treatments and interventions, not the lives of our fellow humans.

In sum, we believe that replacing the rhetoric of quality of life with that of ordinary and extraordinary care, proportionate and disproportionate treatment, will have these advantages.

• First, and foremost, the return to the rhetoric of ordinary and extraordinary care returns the emphasis of any discussion to the treatment side of the patient-treatment dyad and an analysis in terms of the burdens and benefits of treatments, not a judgment on the patient’s life.

• Second, the rhetoric of “care” as in “ordinary care” suggests that bioethical deliberation should not only be concerned with medical treatments for specific conditions but should include all interventions, assistive devices, and social supports that contribute to, are proportionate means to, achieving one’s life aims. The rhetoric of care is simply less medical in nature and suggests a social model of evaluating need.

Of course, there are also things that a change in terminology will not accomplish in and of itself.

• We cannot expect the change in terms to provide an answer to which treatments and supports are worthwhile in which situations. We cannot be spared the hard work of consensus building in regard to these difficult choices.

• Similarly, we cannot expect that the language of ordinary and extraordinary care will be able to avoid having to seek a precarious balance objective and subjective determinations in allocating resources. Capturing all of the important phenomena in determining the worthiness of certain supports requires attention to subjective elements while allocation decisions must be made in the aggregate. Terminology will not spare us. In fact, terminology that does not mask either the subjective or objective elements is preferable as it will not mislead us.

Of course, we are well aware that changing rhetoric is a long cultural process and not a simple act of will. But it is a worthwhile endeavor as it promises to lead to a more civil discourse and allows us to respect for human dignity while we hold difficult discussions and make the difficult decisions in the use of technology.