THE PURSUIT OF GUIDELINES ABOUT FUTILE, MEDICALLY
INEFFECTIVE, AND INAPPROPRIATE CARE:
ATTITUDES AND DECISIONS
Douglas Humphrey Fernald
B.A., The Colorado College, 1989
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
This thesis for the Master of Arts
Douglas Humphrey Fernald
has been approved
Fernald, Douglas Humphrey (M.A.> Anthropology)
The Pursuit of Guidelines About Futile, Medically Ineffective, and
Inappropriate Care: Attitudes and Decisions
Thesis directed by Assistant Professor Kitty K. Corbett
This study identifies and assesses key issues and concerns expressed
by medical professionals and lay persons surrounding the care of those
who are critically ill or are near the end of their lives. As medicine and
society continue to change, medical communities and the larger
communities they serve face competing interests in intensive care
settings. Personal autonomy is a recognized right in the United States.
Professional autonomy in decision-making is an historical privilege.
These values conflict and challenge us to seek a proper balance for
medical decision-making. Within the medical community, some
advocate drafting guidelines for the use of intensive care through a
collaborative process involving medical professionals and diverse lay
persons. It is hoped that such a process will yield practical clinical
guidance shaped by professional knowledge and public attitudes and
beliefs regarding end-of-life medical care. While extensive debates
regarding futile or inappropriate care have taken place in the medical
ethics realm, few efforts assess critical sociocultural variables that
affect the process of deciding the proper locus of medical decision-making.
Through the use of questionnaires and focus group interviews in
collaboration with participant observation in a non-profit organization
pursuing guidelines, this exploratory study in the Denver Metropolitan
Area finds that many stakeholders acknowledge limits of medical care
and share strong opinions about inappropriate care and how care
decisions should be made. By seeking input from diverse members in the
area, we can develop an understanding of the critical issues and elements
in end-of-life decision-making, thereby facilitating the formulation of
contextually appropriate guidelines for the use of intensive care.
This abstract accurately represents the content of the candidate's thesis. I
recommend its publication.
Kitty K. Corbett
Arrangement of This Thesis..................8
2. MEDICINE IN AMERICA: A Short Review...........11
Technology, Corporations, and the Physician.11
Ethical Debates in Medicine: Futile Care
Laws and Codes: Documents of Change.........20
3. MEDICINE IN COMMUNITY.........................29
Whither the Lay Person?..................29
Thinking About Medicine Anthropologically...34
4. RESEARCH METHODS AND RESULTS..................43
Design and Sample..................43
Focus Group Interviews................. 66
Design and Sample.................66
5. MAKING DECISIONS ABOUT APPROPRIATE CARE......82
B. FOCUS GROUP GUIDE...........................105
Sincere thanks to the Colorado Collective for Medical Decisions, Inc. for its
support of this research effort.
The next thing (so at least it seems to me) is neither the construction of a universal
Esperanto-like culture, the culture of airports and motor hotels, nor the invention of some
vast technology of human management. It is to enlarge the possibility of intelligible
discourse between people quite different from another in interest, outlook, wealth, and
power, and yet contained in a world where, tumbled as they are into endless connection, it
is increasingly difficult to get out of each other's way.
-- Clifford Geertz, 1988
As we near the end of another millennium we face an expanding
array of decisions about living and dying. Late in this twentieth century,
death often becomes a protracted process rather than an event. It is a
process that involves more than just the patient, the doctor and the
family. Biomedical achievements of the last fifty years have enabled
medical professionals to intervene more effectively and more often in the
lives of critically ill patients, restoring them to health. Along with the
technological changes came social, political, and economic changes.
Despite the burden of dealing with death, patients, families and
healthcare professionals must now make more decisions with farther
reaching implications. In the last twenty years, however, we have begun
to question seriously the burdens of our medical marvels.
In dealing with end-of-life medical decisions, Americans find
themselves at an interesting intersection of medicine, technology,
economics, law and ethics. In a world that is increasingly
compartmentalized into professional specialties, technical jargon,
corporate concerns, and diverse subcultures, we have confounded our
abilities to communicate among the compartments. In talking casually
among acquaintances, we all hear tales of angst and disbelief experienced
in the doctor's office, in the hospital, or at the nursing home. One
wonders what are the bare essences of good medical care. As a society, we
have been slowly coming to terms with the complexities of contemporary
medical care. Locally, efforts are afoot to define a local version of
medicine in terms of appropriate medical care for those who are seriously
ill and may be near the end of their lives.
Twenty years ago Americans were rudely awakened to the
unfamiliar world of medical ethics when Karen Quinlan lost
consciousness and, through the wonders of medicine, began a long "life",
comatose and confined to a bed. Her parents fought to have her removed
from the respirator to allow her to die. In 1976 the New Jersey Supreme
Court ruled that Karen's father was indeed the legal guardian, and that
Karen's right to privacy, exercised by her father, entitled her to be
removed from the respirator. Despite the ruling and the removal of the
respirator, the parent's decision to maintain artificial feeding kept her
alive for another 10 years (Pence 1990:22). A careful reading of the case
raises numerous issues that are still relevant and often contentious.
Along with the variety of factors that influence our thinking religion,
family background, education Americans were introduced to some
unfamiliar concepts: informed consent, advance directives, autonomy,
extraordinary care, and the persistent vegetative state, among others.
Within the context of end-of-life medical care we challenge some of our
most fundamental, but least explored, notions of our bodies and our
In the face of the many recent changes that have taken place that
affect the type of medical care we receive, we question the wisdom of
technology and we question whether we need all of it all the time. And,
if we do not require any and all treatment, then we must establish some
appropriate methods for deciding those limits -- methods that are
medically sound and culturally appropriate. In other words, how do we
acknowledge and face the reality of death in the context of today's
medicine in today's social milieu?
In the twenty years since the Quinlan decision, physician
autonomy has eroded. Corporate agendas have changed how physicians
conduct business while court rulings have upheld a patient's right to
determine which treatments they will or will not elect. While there have
been ongoing debates about what is clinically appropriate care, there are
now debates about what is culturally appropriate care. Until recently
most of the debate took place among academics, medical professionals,
and courts. As this thesis will show, differing perspectives on the practice
of medicine and medical ethics can underscore important features of the
decision-making process that, heretofore, have been largely nascent
within policy discussions over end-of-life medical care. Exploring these
perspectives offers insight into those factors that stand out as critical to
the content, process, and consequences of the decisions we make near the
end of our lives.
Medical professionals have an established history in addressing the
dilemmas they face. Only recently, however, has the debate moved into
the public arena. Dwindling resources and eroding professional
autonomy have pushed the dilemmas into a larger forum asking society
to participate in decisions and responsibilities. Several years ago voters in
the state of Oregon were asked to decide how Medicaid money should be
apportioned by prioritizing certain medical procedures. Now, in
Colorado, a group of healthcare professionals and concerned citizens,
with the backing of most Denver-area hospitals, recently organized to
address issues of cost, technology and humane care. Known as the
Colorado Collective for Medical Decisions (CCMD, formerly GUIDe), this
multidisciplinary effort seeks public input on end-of-life medical care
with the hope of drafting guidelines for the use of intensive care. CCMD
is a non-profit organization composed of physicians, nurses, social
workers, chaplains, patient representatives, ethicists, and a handful of
Donald Murphy, a physician who heads CCMD, asserts that this
project will involve the public and the "final guidelines will ultimately
be shaped by healthcare professionals and the public" (Murphy and
Barbour 1994:328). One objective of the guidelines is to limit
care that is deemed inappropriate by the public and professionals.
What the guidelines actually say and how they will ultimately be
used is not precisely known. For CCMD, it is hoped that the product
derived from the collaborative process will lead to policy
recommendations for the use of intensive care in Colorado. Currently,
CCMD has several subcommittees drafting proposals for guidelines based
on current medical standards for care. These initial proposals attempt to
define care that is not medically indicated (Murphy and Barbour 1994). It
is not clear how these proposals will integrate public input regarding
inappropriate care or how the guidelines will be used with the public.
Judging from the CCMD Legal Subcommittee's statement on purpose, the
intent is to have the recommendations of the guidelines introduced as
standards for professional medical practice and included in efforts to
"draft new legislation, regulations, and policies, as applicable" (Murphy
and Barbour 1994:330).
Murphy predicts that, "Not only will our community agree on that
criteria [sic] for futile care, we shall agree on many criteria that help define
inappropriate intensive care, a much more difficult task" (1994:162).
However, even within the collective, there have been ongoing debates
over agendas, the pace of change, and appropriate terminology. At a
plenary meeting this fall, comments were solicited on a draft document
concerning a proposed definition of futility. In the course of discussion,
the plenary concluded that the concept of futility could be defined as
"medically ineffective" treatments, but with the additional understanding
that this was a declaration of professional goals of care and professional
criteria for meeting those goals, and did not attempt to speak for
Coloradans as a whole. In other words, implicit in the designation of
"medically ineffective" is that there is no underlying assumption about
whether the lay person would consider the treatment congruent with his
or her goals and his or her criteria for meeting those goals.
Although the CCMD project addresses many variables that could
figure into the final guidelines, its collaborative approach focuses on two
themes that call for exploration. One is medical in scope -- medically
ineffective care -- and the other is sociocultural in scope -- what diverse
community members believe to be appropriate care and decision-making
processes when faced with an end-of-life decision. The goal is to combine
these two themes into mutually-meaningful guidelines. The challenge,
however, is to combine them in a manner that acknowledges the
inextricable links and shared responsibilities between the medical
community and the lay community, while recognizing salient
differences. It may be that formal guidelines themselves are
inappropriate and unacceptable to Colorado residents even if there is
consensus on the content of the guidelines.
While definitions of medically ineffective care will be decided
among medical professionals, we should not expect them also to sort out
how lay persons decide what is appropriate and inappropriate and why.
The social scientist and the anthropologist in particular is in a unique
position to identify and evaluate concerns within the community and to
render practicable solutions for medical decision-making. By considering
how knowledge is derived from cultural and social processes, links across
the medical-lay spectrum may be established. We should, in the words of
Clifford Geertz, "enlarge the possibility of intelligible discourse between
people who are quite different from one another...." (1988:149).
Arrangement of This Thesis
Chapter Two lays out the historical background of the issues, with
emphasis given to relevant ethics debates, and the futile care debate, in
particular. These debates are primarily professional and confined to
medical and ethical realms. They are relevant because they control much
of the thinking in terms of policy decisions and how healthcare
professionals have addressed difficult care issues. Of particular concern to
facilitating understanding across lay-professional boundaries are the
issues of medically ineffective care, medically inappropriate care, and the
economic and social benefits gained by limiting "futile care".
Chapter Three addresses the sociocultural dimensions of the futile
care debate, highlighting contributions and criticisms from the social
sciences. While the medico-ethical literature identifies a need for public
input on policy matters, there are few studies that attempt to sort out the
various profiles of that public. This chapter draws on discussions in
medical anthropology to inform us about the variables that may be
important in the decision-making process and in the formulation and
implementation of guidelines, but may (or may not) be explicit in our
deliberations. Particular attention is given to the social construction of
knowledge, cultural diversity, and praxis how we approach the problem
of lay participation in defining the decision-making locus.
Chapter Four defines the methods and samples used in this study.
Questionnaires and focus group interviews were used to assess attitudes
about limits on medical care and identify issues critical to the decision-
making process. Because of the complexity of the subject in question,
survey data would by themselves be inadequate for exploring the issue of
"medical futility" and its proper place in decision-making. Focus groups
allow for open-ended discussions that underscore the factors contributing
to one's personal decisions. This chapter also presents interesting
findings from the questionnaires and the focus group interviews.
Chapter Five addresses the implications of this study.
Consideration is given to setting policy agendas, integrating the futile care
concept with sociocultural exigencies, and areas of future research.
Chapter Six, a conclusion, presents the author's notes and thoughts
on the future of medical care in the United States, giving consideration to
ethical and anthropological concerns coupled with the findings of this
MEDICINE IN AMERICA: A SHORT REVIEW
Yet it is as evident in itself, as any amount of argument can make it, that ages are no more
infallible than individuals -- every age having held many opinions which subsequent ages
have deemed not only false, but absurd; and it is certain that many opinions, now general,
will be rejected by future ages, as it is that many, once general, are rejected by the present.
John Stuart Mill, 1859
Technology. Corporations, and the Physician
Healthcare professionals do not arrive on the scene de novo. Many
of their beliefs and standards of practice are the result of social and
historical processes that bear reviewing here. Historically, physicians
made the decisions about the type and length of care of their patients.
They were often closely tied to the community and families of the patient
and, when no effective interventions could be given, the doctor was at
the bedside consoling the patient. Beginning in the 1930s, however, the
role of the physician began to change along with the advent of powerful
and effective medicines and with the increasing concentration of
professional practice within hospitals.
Following World War II, dramatic advances in pharmacology led
to the development of antibiotics. Physicians could seemingly work
miracles with new antibiotics, especially penicillin. Through interviews
with well-established physicians, Sharon Kaufman reveals to us that
these medical advances were initially accidental, resulting from a
combination of technology and the daily demands of doctors striving to
help patients. One physician who practiced in the early 1940s felt
antibiotics were an "exciting new development" and noted that, "The
treatment of infections up to that time had almost been nil: bedrest,
nutritious diet, and quiet" (Kaufman 1993:172).
This transformation, Kaufman notes, was a move from "care" to
"cure" (1993:316). As new procedures and pharmaceuticals became
increasingly effective, efforts were aimed at treating an objectified,
physical body, leaving the subjective, psychosocial body to the side.
Physicians could now intervene over a relatively short interval, and,
with some certainty, radically alter the course of a patient's life.
Physicians continued to make most of the decisions, but the consequences
of their decisions were much more dramatic. Further, as the training of
physicians became more sophisticated and more technical, the average
person was in no position to question the physician's decision.
New technological wonders, in effect, changed the way we die in
America. With the scientific ability to sustain indefinitely the life of the
most physiologically compromised patient, we begin to question
technology, physicians, and the goals of medicine. Though we have new
abilities to sustain life, do we also have the need or desire to do so? And
who decides where the limits are? For society to come to terms with the
changes, it must undergo some cultural changes, and as many point out,
culture change for the masses is a slow process and one that does not
progress along rational, scientific lines.
Along with the technological changes, there were transformations
within the structures of healthcare delivery. As Paul Starr describes the
transformation (1982:352-363), increased national investment in research
and hospitals following WWII had the effect of concentrating power and
money within empire-like medical centers. Gone were the general
practitioners working in rural or inner-city areas with close ties to the
communities they served. Physicians were now an important source of
revenue for the large medical centers and had to be considered in terms of
their ability to generate profits. By the 1980's, many medical centers were
turned from non-profit institutions to for-profit corporations. This trend,
Starr predicted, would lead to the deterioration of physician autonomy
(1982:444-446). An important consequence of a transformation to
corporate medicine is the focus on economics as the "bottom line." This
is of concern because the motivation to treat or not to treat might be
unduly influenced by whether the treatment will result in a profitable
reimbursement to the corporation. In an age of shrinking government
payouts, increasing medical costs, and edgy insurance companies, one
solution is to seek out treatments that can be eliminated without harm to
the patient. While cost is sometimes downplayed in debates about
"rationing" or limiting medical care, it certainly adds fuel to a well-stoked
Ethical Debates in Medicine: The Futile Care Concept
Ethical debates inform us about one part of the medical culture that
controls much of the thinking in the medical community. William
Benitz, a neonatologist who has written on decision-making in intensive
care, makes the following observation:
Whether by design or by default, experience in dealing with ethical
issues leads to the adoption of ideas or behaviors that guide
subsequent considerations. In addition, even naive participants
(such as students or parents who unexpectedly find themselves in
these circumstances) come with tacit assumptions about the nature
of the decision-making process. Denial of the utility or necessity of
these conceptual frameworks cannot negate their existence but can
only obscure their influence. [1993:282]
To restate this, healthcare professionals are strongly influenced by the
language and experiences of their professional lives. Discussions on
various ethical topics, such as autonomy, beneficence, and justice, form
part of the framework for dealing with difficult care decisions. To what
extent the lay person shares this knowledge is unknown. Furthermore,
the lay person may talk about these same ethical issues -- and may agree
with the professional view using a different vocabulary.
Ethics debates attempt to weigh the benefits and burdens associated
with each potential choice. With end-of-life care, the debate tries to weigh
the benefits sustaining life against the potential costs.
The primary debate in ethics which concerns this research is the
concept of futile care. Framing the discussion in terms of this concept
highlights two important themes that must be considered: care that is
medically effective and care that is culturally appropriate. The first theme
concerns professional duties and limitations, including the principles of
beneficence and informed consent, the practical limits of medical
interventions, and the physician's own moral and ethical standards a
culture that guides the provider's thinking. The second theme concerns
autonomy, or one's right to determine which care is appropriate
according to one's own moral and ethical standards the culture that
guides the consumer's thinking. When these two themes are
incongruent, decision-making becomes a more difficult process.
The Hastings Center Guidelines define a treatment as
physiologically futile if it is "clearly futile in achieving its physiological
objective and so offers no physiologic benefit to the patient" (Cranford
and Gostin 1992:307). Based on this definition, designating a treatment as
futile is a clinical determination based on the knowledge that the
treatment cannot reasonably be expected to have a demonstrable
physiological effect in achieving a stated goal. For instance, prescribing
antibiotics for a viral infection would be considered medically ineffective.
While such a practice may have psychosocial dimensions (ie, acquiescing
to a patient's insistence for "pills" just to get rid of them), it does not work
toward achieving the objective of destroying the virus. Ostensibly,
deeming a treatment medically ineffective is based on a body of empirical
and statistical knowledge, is value-free, and is reasonable for the
healthcare team to make unilaterally. After all, at some level, we accept
that physicians are experts on these matters.
However, the debate over futile care continues because at the
bedside a determination of "futile" can be imbued with non-medical and
non-scientific judgments. A member of a healthcare team may consider
care to be futile with respect to strong personal feelings about the
objectives of medicine in general. A number of commentators have
noted that quality of life, resource allocation, and harm to the patient may
encourage physicians to label some treatments as futile (Cranford and
Gostin 1992; Brennan 1992; Veatch and Spicer 1992). When futile care
begins to be defined by these standards it becomes misleading. That is, a
treatment may have an effect, but one which the physician may consider
to be of no benefit to the patient because, for instance, it would not
improve the quality of life of that patient. There is no common standard
definition of quality of life that would satisfy all patients or their families.
The debate over futile care is framed within specific social and
historical contexts. Without strong notions of autonomy, physicians
could return to paternalism and make unilateral decisions about care.
Without the technology, these difficult decisions would not have to be
made. With unlimited funds, patients who demand unlimited care
would not drain resources from the collective "piggy bank." Autonomy
as a value is not going to go away anytime soon; technology is not going
to revert to a pre-WWII level; and payors are tightening their belts
everywhere. In sum, society as well as individuals must comment on
inappropriate or futile care.
We would do well to consider a broader perspective in this debate
and ask how such determinations of medically ineffective care are a
benefit to society. Obviously, cost is a major concern. In the narrowest
fiscal sense, most people would agree that it is sensible to stop the use of
medically ineffective treatments that cost a good deal of money but are
extremely unlikely to have an effect in achieving a specified goal. But
there must be other justifications for moving the debate into more public
forums. Legitimacy of the profession may be one of the motivations.
Fear of litigation may be another. Or it may be that in sharing in the
formulation of policy, the non-professional members of a community
also share in responsibilities regarding the outcomes of their decisions.
While the debate over futile care is still unresolved, it is important
because, as Benitz notes, "experience in dealing with ethical issues leads
to the adoption of ideas or behaviors that guide subsequent
considerations" (1993:282). Ethics are standards of right and wrong that
attempt objectively to promote sound medical practice. Ethical codes of
practice are mainstays in the Western medical tradition and will continue
to frame thinking on difficult care issues.
The word futility continues to be problematic. To simplify, I use
the phrase "medically ineffective" to talk about those interventions that
do not achieve a specific physiological object. This says nothing about the
appropriateness (which is a value-laden term) of an intervention, in
terms of community attitudes and beliefs about goals of medicine. That
is, care that is deemed medically ineffective may be considered
appropriate by members of the community. Conversely, care that is
deemed medically effective may be considered inappropriate by the
community. Herein lies one of the main objectives of this study: to
determine what members of the community think are appropriate
treatments for people who are near the end of their lives.
Law and Codes: Documents of Change
While the futile care concept figures prominently in ethics debates,
it represents only one portion of the change in the professional culture of
biomedicine. Cultural changes can be measured by any number of
variables. They could be our clothing, our use of slang, how many hours
we watch television, or the national divorce rate. In a society where laws
are presumed to reflect the wishes of a constituency rather than the
whims of a few, legislation becomes a marker of cultural change. In
professional realms, codes of ethical practice reflect the collective view of
many individuals attempting to meet new demands.
Since the 1950s, there has been an increase in official governance of
ethical practice that recognizes the rights of the individual to make
decisions about treatments they receive. In 1974, the United States
Congress established the National Commission for the Protection of
Human Subjects of Biomedical and Behavioral Research. The three main
principles it identified are respect for persons (autonomy), beneficence,
and justice. (Levine 1991:208). Autonomy refers to an individual's right
to make decisions free from external constraints. Beneficence, as applied
to physicians, requires them to help others by preventing or removing
harm. Justice has been interpreted in numbers of ways, but generally
refers one's right to be treated according to norms of fairness that apply to
all citizens. Justice is often described in terms of distributive justice
which "refers to distribution of primary social goods, such as economic
goods and fundamental political rights" (Beauchamp and Waters 1994:26).
The American Medical Association addresses these and other
ethical issues in its statement on the "Fundamental Elements of the
Patient-Physician Relationship." (JAMA 1990) The statement focuses on
the rights of the patient to receive information, make decisions, assure
confidentiality, maintain continuity of care, and receive adequate care
regardless of ability to pay. There are some portions of the statement that
should be highlighted. It speaks of a collaborative effort and a "mutually
respectful alliance", "costs of appropriate treatment", and a right to care
"dependent on society providing resources so that no patient is deprived
of necessary care because of an inability to pay for the care" (JAMA
1990:3133). These phrases recognize the patient as an important element
of the decision-making process and asks society to bear some of the
burdens of medical care. The individual patient's wishes, however, seem
to be of greater concern in end-of-life care decisions.
Legislatively, the exercise of personal autonomy in medical
decision-making was strengthened through the Patient Self-
Determination Act of 1991. The law was designed to encourage the use of
advance directives, persuade healthcare professionals and institutions to
honor advance directives, and encourage education of patients and
physicians on the subject of end-of-life care decisions. This legislation
followed in the wake of the United States Supreme Court 1990 decision
that effectively ruled that because Nancy Cruzan (who was in a persistent
vegetative state) had not made clear declarations about her treatment
preferences in advance, the hospital could not be forced to withdraw care
that might result in her death (Beauchamp and Waters, 1994:384).
In general, attention to the needs and wishes of the patient, at least
in a theoretical sense, take precedence when healthcare teams must make
difficult end-of-life care decisions. Three guidelines shape those
decisions. First, healthcare teams should follow advance directives when
the instructions are applicable. Second, if no advance directive is
available for the situation at hand, then they must defer to substituted
judgment, which means the appointed surrogate should make a decision
based on their knowledge of what the patient would have wanted. Third,
if the surrogate has no knowledge of the patient's wishes, then a decision
should be made that promotes and protects the patient's best interests
Advance directives, or treatment directives, are documents that
allow individuals to express their preferences in the event they become
incompetent and unable to make choices about their care. A living will is
a legal document stating a person's treatment preferences in the event of
future incompetence and terminal illness. Proxy appointments such as a
durable power of attorney, enable a person to assign decision-making
rights to a specific individual (a surrogate) who will likely act in
accordance with the patient's wishes. These documents are supported by
constitutional, statutory and common law and must be recorded by
institutions providing care. They can be effective means to ease the
decision-making burden of family members and healthcare professionals.
However, some ethicists (Emanuel and Emanuel 1993) point out
that such instructions are often inadequate. First, instructional directives
may not cover all the clinical circumstances and, second, those few who
have such directives, may not see them honored by physicians. A large
study which recently concluded its second phase found interesting yet
discouraging results. Known as SUPPORT, this study undertook a 2-year
prospective study designed to "improve end-of-life decision-making and
reduce the frequency of a mechanically supported, painful, and prolonged
process of dying" (SUPPORT 1995:1591). Despite training interventions
that were supposed to improve communication with families or patients
and improve medical outcomes, the study found that patients
experienced no improvement in discussions with healthcare
professionals, high percentages of patients still experienced pain (as
reported by families), and physicians had no better knowledge of their
Why did the study fail to improve end-of-life decision-making?
One consultant to the project suggests that any number of subjective
assessments on the part of the physician could be implicated, including a
physician's preconceived notions about a patient's preferences (Lo 1995).
This is a reminder that despite autonomy being the rule, if no discussion
about preferences ever took place, or if those discussions were ignored by
the physician, physicians shoulder a good deal of responsibility in the
perpetuation of medically ineffective and medically inappropriate care
As an alternative to physician autonomy in instances where
advance directives are rendered ineffective, Emanuel and Emanuel argue
that there is a need to provide supplemental directives developed by a
"community of patients" (1993:10). Initially, such guidelines would be
based on treatment goals and preferences expressed by patients who are
already "enrolled" in a healthcare facility. This is an interesting proposal
because it would methodically gather input on patient preferences from
those who actually face difficult decisions. The authors contend that this
level of community "retains its primary focus on the importance of
patient self-determination" (1993:10);
At this time in American history, personal autonomy carries
weight in ethical decision-making. Central to this ethical principle is the
patient's ability to have access to all information necessary to make an
unencumbered decision. To ensure that patients can exercise that right,
they need to be as fully informed about their options as is reasonable.
Following the publicity of the medical atrocities of World War II, the
Nuremberg Code outlined the core of what became referred to in the late
1950s as "informed consent" (Levine 1991).
As much as it can, the informed consent rule facilitates a decision-
making process by requiring that the patient be fully informed as to the
risks and benefits of treatment. This is, of course, a complex issue. How
well informed the patient really is and how involved the patient is in
making the decision are open to debate. Jay Katz argues that legal
. challenges have failed to make significant inroads into a medical
tradition that customarily withholds a good deal of information from
patients. In the Canterbury v. Spence ruling in 1972, the court asserted a
patient's right to self-determination, but established that the standard for
disclosure of information was based on the physicians' "objective"
assessment of "what a prudent person in the patient's position would
have decided if suitably informed" (Katz 1994:147). In effect, this ruling
asks the doctor to make a judgment about patients' needs (general), rather
than assess each individual's need (specific) for information. Katz argues
that the court should have considered the individual patient's plight and
"required physicians to learn new skills: how to inquire openly about
their patients' concerns, doubts, and misconceptions about treatment."
As an alternative, Katz asks us to consider each individual. In a
multicultural community, each individual may have widely differing
needs. Robert Levine has noted that the principle of informed consent
may not be universally valid given differing perspectives on personhood
(1991:210). Furthermore, the three principles noted above that guide end-
of-life decision-making -- advance directives, substituted judgment, best
interest overemphasize the individual patient's wishes and directs little
attention to broader considerations.
We should ask: What information is important and relevant?
How is it understood? Beneficence for whom -- the patient, the family,
the community, society? Where is the family and the community in all
of this? Who is the family? Who is "the community"? And, if
autonomy is the rule and informed consent is a means to uphold the
rule, why is there a problem in end-of-life care?
Certainly, there is more to these questions than simply coming up
with a formula to balance the aforementioned principles of autonomy,
beneficence, and justice. There is a problem because we have reached a
point where many realize the limits of medicine and limits of the human
body. Unrestrained autonomy, however, can demand any and all care. If
this places unreasonable demands (and I would argue that it does) on
healthcare teams and others' hard earned money, then there is a problem.
One person's opportunity is another's loss of opportunity. How do we
encourage limits to autonomy without encroaching on fundamental
individual freedoms? In essence, there is a need to reconstruct our
culture in a manner that shares responsibility for setting and abiding by
policies that seek to limit medically ineffective treatments for the
seriously ill patients who are near the end of their lives.
The objective of this chapter was to bring to this discussion the idea
that our culture has faced profound changes in medicine and to
underscore some of the important features of the professional culture
that has dealt with these changes over many years. Through legislation
and ethics debates, professionals are attempting to work through some of
the difficulties of our time. However, we must assess how these complex
discussions have real and tangible consequences for the average person.
Most of us who live long enough will likely be faced with the death of
someone who is close to us. Debates in professional and academic circles
do affect the behavior of those who will treat our friends and family,
which in turn affects the choices we as individuals and we as a society
MEDICINE IN COMMUNITY
Whither the Lay Person?
To this point, academic, legal and professional considerations have
been discussed. They are complex discussions and require some
background reading just to acquire a working vocabulary. Nevertheless,
lay people face difficult medical decisions every day without the benefit of
training in medical terminology or ethical precepts. Unfortunately, it is
not clear how they reach a decision and what is really important to the
decision-making process. In matters of life or death, medical objectives
can become muddled by personal values and ideations about one's roles
and responsibilities to the patient and to society. If neither patients nor
physicians are to be entirely trusted with the decision, then there must be
some sizeable middle ground to work with.
In recognizing the economic, technological, physical, and cultural
limits of medical care, the professional realm is starting to turn to face the
communities it serves (e.g., Colorado Collective for Medical Decisions,
Oregon Health Decisions). This raises a number of important questions.
Who is the community? What do they know about futile care? Do these
things matter? Clearly they do matter or the debate over futile care would
have been resolved long ago. Moreover, given the increasingly diverse
character of this country, we would be naive to claim that what many
medical professionals hold to be ethical practice is universal. We would
also be naive in hurling difficult questions to an unsuspecting public.
"Which is more important, autonomy (your rights), beneficence (doctor's
duty), or justice (fairness to all Americans)?" Even if we could get
answers to such a question, what would they tell us? However, in an age
of personal liberty, if we fail to consult the masses at some level, the
legitimacy of the biomedical enterprise might be questioned by more than
an insignificant few.
The dimensions of the impact of cultural variables have only
recently received much attention. Nancy Jecker et al. (1995) addressed the
issue of a caregiver who is a member of the dominant US culture and a
patient who is from a different culture. Such differences can have
profound effects on the effectiveness of treatments and communication
between physician and patient or decision-maker. Yet, she acknowledges
that the "dearth of coverage of cross-cultural issues in bioethics itself
reflects the general paucity of scholarship in this area" (Jecker et al.
One study found that physicians are poor predictors of patients'
preferences for life-sustaining interventions, though it suggested that
more discussion with the patients about end-of-life care might improve
the predictive accuracy of physicians (Druley et al. 1993). Cultural
variables may well play a factor in this. Another study was conducted to
assess specifically the effects of ethnicity and race on attitudes toward
advance directives, life-prolonging treatments, and euthanasia. The
findings showed significant differences among ethnic groups regarding
physician-assisted dying and circumstances under which they would want
the physician's assistance in dying (Caralis 1993).
Within metropolitan Denver, it is not known to what degree
various communities of professionals and lay persons are united or
divided on issues involving intensive care. Likewise, it is not known
what issues specific sub-groups may identify as most important and how
much variability exists on attitudes and beliefs within these sub-groups.
Discontinuities across any lines may prove important in drafting
guidelines designed to serve a bounded, but diverse community.
In researching the diversity of attitudes and concerns in a large
urban setting, special attention must be given to the principles of
autonomy, beneficence, and justice. While discussion of these principles
in anthropology focus largely on problems associated with their
application, two key issues are raised (Marshall 1991). First, informed
consent may be difficult to obtain when cultural or linguistic variance
prevents understanding of the issue, or when the final product of
qualitative research methods are unknown. Second, we must consider
the potential harm to the participants in more ways than strictly
physiological outcomes. Failure to solicit their opinions or failure to
include diverse opinions in policy agenda discussions could ultimately be
harmful to an individuals claim to autonomy and perhaps beneficence.
Certain responsibilities are owed to our informants. Patricia Marshall
points out that there may be conflicting principles at work in researching
sensitive issues (1991). For medical anthropologists (and others), there
are dangers in collecting data. There are no guarantees as to how the
information should be used. How the anthropologist uses the
information may be restricted by other considerations: who funds the
research, how much proprietary control is granted to the anthropologist,
and who the beneficiaries of the research might be. In essence, what is the
ultimate role of the anthropologist? There are differing positions within
the discipline as to how strongly anthropologists should be advocates for
their research participants. In this study, the proper role, so it seems to
me, is to follow a research design that promotes frank discussions of the
issues across the lay-professional boundaries and collects data in a
manner that allows as many voices to be heard as possible. Furthermore,
if the objective of developing guidelines is to include heretofore unheard
opinions and attitudes, then a proper role would be to see that these
opinions and attitudes are systematically included in policy discussions.
Daniel Callahan notes that, "Since there are no commonly accepted
norms available to provide a wholly independent standard, there must be
continuing public discussion and debate about the balance between good
process and good outcomes" (Bioethics 1991). Recognizing the need for
the proper location of such debate, Troyen Brennan suggests that
developing public forums to discuss these issues "removes the debate on
public, positive rights from the bedside and places it in the appropriate
context" (1992:338). Defining that context is difficult and draws our
attention to the wider sociocultural forces that have shaped the current
mode of medical decision-making.
Thinking About Medicine Anthropologically
To an already voluminous professional literature on treatment
dilemmas, more needs to be added. Ethics debates, the move to curative
medicine, the institutionalization of medical care, and the changing locus
of decision-making authority have been considered in terms of their
impact on the behavior of physicians. If public input is considered
important in drafting guidelines for intensive care use, then an
accompanying literature that ranges outside the medical profession must
also exist. We must know not only what attitudes and opinions people
have, but we must attempt to know which of the attitudes and opinions
matter most in the decision-making process. When we have some idea
of this we can begin the business of reconstructing both the cultural and
social authority needed to influence how healthcare professionals and lay
Paul Starr, in the Social Transformation of American Medicine
addresses a number of features of our healthcare system that are relevant
to the local transformation of medicine in Denver. The introduction
looks at the origins of physician authority in the United States in terms of
"underlying changes in material life and social organization" (1982:9).
Authority, therefore, incorporates two sources of effective control:
legitimacy and dependence. The former rests on the subordinates'
acceptance of the claim that they should obey; the latter on their
estimate of the foul consequences that will befall them if they do
He goes on to say that authority relations undergo change and the
legitimacy of the authority may be in doubt. Medical professionals,
especially physicians, appear to be in such a phase now, and must turn to
persuasive arguments to bolster the legitimacy of their own autonomy.
Starr traces a history of the rise of corporate medicine, concluding
that the "bottom line" approach to medicine has taken away some
physician autonomy. While the corporate factor plays an expanding role
in decision-making, more aggressive claims to individual (as opposed to
professional) autonomy would seem to play a role in creating social
One notion I find helpful in thinking about the transformation of
medical decision-making is the concept of authoritative knowledge
defined by Brigitte Jordan in her book, Birth in Four Cultures:
It is the knowledge that within a community is considered
legitimate, consequential, official, worthy of discussion, and
appropriate for justifying particular actions by people engaged in
accomplishing the tasks at hand. [1993:154]
She emphasizes that authoritative knowledge is not the knowledge of
people in positions of authority. "To the extent that such persons are
members of a community of practice, they will share the local version of
authoritative knowledge with other members" (1993:154). That is, we, as
consumers, are complicit in the construction of authoritative medical
knowledge. This concept holds out the possibility that the transformation
of this knowledge need not be in one particular direction or dominated by
one particular group. Further, Jordan's insight narrows our scope from
large macro-processes to community-level micro-processes. We want to
consider how perceived authority and real authority influence
individuals' lay and professional -- decision-making behavior.
The strong assertion of individual rights to choose may be one
means by which that character of the authoritative knowledge has
changed. Certainly one area that has changed what constitutes
appropriate actions by physicians is increased malpractice litigation in this
country. Donald Murphy estimates that "20 to 50 percent of some doctors'
charges are for defensive medicine interventions used to fend off
potential litigators" (1995). And, it is not difficult to find someone who
thinks that such practices are outrageous and that litigation in this
country has gone to an extreme with jury awards to disgruntled plaintiffs.
If conventional wisdom presides, changes in what constitutes appropriate
care are conceivable. Economics, however, are incomplete in describing
how people act in the formulation of rules and regulations.
Some in medical anthropology speak more directly about how
power relations and political agendas are influential in the social
construction of medical knowledge and our understanding of the body
(Lock and Scheper-Hughes 1990, Rhodes 1990). Others invite us to
consider suffering in a broader narrative that recognizes an individual's
"local cultural orientations", rather than narrowly defining the suffering
in terms of a prescribed nosology (Kleinman 1988).
In looking closely at individuals and the networks to which they
are tied, Kleinman informs us about how individual behaviors are
affected by their personal histories and how these histories impact the
relationship, and ultimately the outcome, of doctor-patient interactions.
In particular, he speaks of the consequences of differing ways of
expressing our symptoms and disorders. The word illness refers to the
culturally shaped "complaints" that patients and their families bring to
the practitioner. The practitioner "reconfigures the patient's and family's
illness problems as narrow technical issues, disease problems" (1988:5).
This medicalization of illness tends to ignore the important psychosocial
dimensions of the illness. Kleinman's insights should be expanded to
look at the medicalization of death. To the physician who narrowly looks
at the physical body in terms of biology, there is a tendency to ignore what
death means to the family. The clinical details of the person may play an
insignificant role in the decision-making behavior of the individuals
who must cope with the possibility of loss. The medicalization of death
may be of further consequence in that death is viewed as a negative
clinical outcome "assessed solely through the rhetoric of improvement of
disease processes" (Kleinman 1988:6). This might encourage
overtreatment, not out of greed or malice, but out of the trained
inclination to try something that might keep a patient alive or "improve
the disease process."
Other anthropological perspectives on how we behave during
illness concern decision-making in terms of the "therapy management
group" and how the social arrangements of this group can affect the
decision-making and care processes (Janzen 1987). Many times, families
must make difficult decisions without the benefit of the patients' input.
How family members interact, who comprises the family, and who has
greater authority (legal or social) become important questions in end-of-
life care dilemmas.
One viewpoint that has received considerable attention over the
last decade is that of critical medical anthropology. Merrill Singer, one of
its chief proponents, defines the critical method as:
A theoretical and practical effort to understand and respond to
issues and problems of health, illness, and treatment in terms of
the interaction between the macrolevel of political economy, the
national level of political and class structure, the institutional level
of the health care system, the community level of popular and folk
beliefs and actions, the microlevel of illness experience, behavior,
and meaning, human physiology, and environmental factors.
Recently, the critical medical perspective has begun to focus on issues of
praxis and how anthropologists become advocates within the
communities they research, giving particular attention to the political
and economic dynamics that intervene in that community (Singer 1993,
1994, 1995). The political and economic imbalances within a community
will likely shape the decision-making process for the care of critically ill
patients, even if professional-lay guidelines are drafted. However, given
that these variables along with most others in the decision-making
process are poorly understood and not clearly identified, an expressly
political-economic approach may not be relevant or appropriate at this
early and undeveloped stage of discussions with the medical and lay
Finally, we should consider the concept of medical pluralism in
this country because it draws attention to the fact that many people who
will find themselves in hospitals or clinics do not share a heritage of
American biomedicine and may find much of it alien and inappropriate.
Medical anthropology attempts to describe shared meanings within
communities and how these meanings are shaped by historical, political,
economic, and social relationships and how the various meanings are
distributed across demographic lines (Kleinman 1995; Singer 1995).
While efforts by ethicists and medical professionals are essential to
developing effective dialogue among various stakeholders, medical
anthropology is in a unique position to offer a more holistic approach
that seeks to identify socially constructed meanings of medicine. Such an
approach offers insight into non-medical perceptions about the goals of
medical care and helps define what are acceptable outcomes in terms of
sociocultural variables and not in terms of strict clinical variables, or
rational ethical variables.
In end-of-life decision-making, we must look at what is important
in terms of both process and content. Guidelines seek to establish
stronger claims to professional and social legitimacy to enable healthcare
workers to make decisions more easily without re-enacting the entire
futile care debate for each potentially difficult case. Moreover, such
guidelines should have relevant medical content according to
professional and lay needs, while encouraging a decision-making process
that is effective in terms of the communities they serve.
For any of these debates medical, ethical or anthropological to
impact the lay person and the healthcare professionals in the social
reconstruction of authoritative knowledge for end-of-life care, we must
know what and how we think about intensive care use. The main point
of considering these and other views is that how people decide is based
largely on what they perceive to be true and correct. Those notions are
socially constructed and it may not be immediately apparent why an
individual advocates one treatment while disdaining another. In large,
metropolitan areas, what boundaries define a community and its "local
version" of appropriate practices are difficult to ascertain. The following
pages assess both the content and process of end-of-life medical decision-
making in the Denver area.
Arthur Kleinman succinctly states that the "job" before us is "to
situate a bioethical problem in that local ethos in order to understand
what is at stake for the participants, what is contested, and thereby offer a
cultural formulation of conflicting ethical priorities" (1995:1672). Once
this has been done, he goes on to say, we must compare "local and
professional bioethical standards," negotiate the differences, and, when
both parties cannot resolve the difference, they "should specify the nature
of the problem for further review." Given the dearth of cross-cultural
research on end-of-life and intensive care decision-making, we are in the
very early stages of the process.
RESEARCH METHODS & RESULTS
This study identifies and assesses key issues and concerns expressed
by the medical community and lay persons in Denver surrounding the
care of those who are near the end of their lives. Questionnaires were
used to assess attitudes about the treatment preferences and decision-
making authority of individuals of diverse backgrounds who have been
introduced to the topic of end-of-life medical dilemmas. Focus group
interviews were used to generate open-ended discussions regarding
intensive care scenarios.
Design and Sample
A questionnaire (see Appendix A) related to end-of-life care and
medical decision-making was developed by a committee from the
Colorado Collective for Medical Decisions (CCMD), with limited input
provided by this author. The questionnaire was given out to participants
who were part of diverse groups that were introduced to the CCMD and
were given a brief slide presentation with an accompanying script. The
script introduces the concept and objectives of CCMD and describes the
various ethical and treatment dilemmas facing healthcare professionals
In addition to demographic questions, the subjects were asked
about their attitudes towards medical costs, rights of patients and doctors,
care for terminally ill patients, medical guidelines, and use of technology.
The questionnaires included statements related to end-of-life medical care
and medical decision-making and asked respondents to state how often --
always, most times, sometimes, rarely, never they thought they would
agree with the statement. Frequency distributions, chi-square statistics,
and rank correlations were calculated where appropriate.
The sample was obtained by members of CCMD who contacted
various civic, professional and educational groups inviting them to an
introduction to CCMD. The groups represent a cross-section of
metropolitan Denver and include hospital ethics committees, patient
support groups, nurses associations, religious organizations, seniors
groups, university ethics classes, and several national service clubs.
Identifying groups that would be interested was a difficult task. The
groups were contacted over a period of five months, relying on contacts
within the CCMD network. A total of 220 questionnaires were completed
As shown in Table 4.1, 28% of the respondents were male and 72%
were female. Most of the respondents were married (63%), with only 15%
reporting they were single, 11% reporting they were widowed, and 8%
reporting they were divorced. Overall, most respondents fell into the
middle to high annual household income categories, with 42% in the
$21,000 to $49,999 range and 38% in the $50,000 to $100,000 range. Sixteen
percent had annual household incomes above $100,000 with male
respondents representing a notably disproportionate percentage of the
highest income category -- 52% male versus 48% female.
On the questionnaire, respondents were asked to state their
religious preference if they had one, or indicate that they had no religious
preference. Responses were assigned to one of five categories. No
Characteristics of the Respondents
n, % of row*, Total
19-40 14 (22%) 49 (78%) 63 (32%)
41-60 21 (28%) 54 (72%) 75 (38%)
61-88 20 (33%) 41 (67%) 61 (31%)
Married 44 (34%) 86 (66%) 13 0 (63%)
Single 7 (23%) 23 (77%) 30 (15%)
Widowed 1 ( 4%) 22 (96%) 23 (11%)
Divorced 3 (13%) 14 (87%) 17 ( 8%)
Other 2 (40%) 3 (60%) 5 ( 2%)
Annual Household Income
<$21000 1 (11%) 8 (89%) 9 ( 4%)
$21000-$49999 18 (22%) 65 (78%) 83 (42%)
$50000-$100000 19 (26%) 55 (74%) 74 (38%)
>$100000 16 (52%) 15 (48%) 31 (16%)
Caucasian 43 (24%) 138 (76%) 181 (90%)
Other 12 (57%) 9 (43%) 21 (10%)
Catholic 8 (31%) 18 (69%) 26 (13%)
Protestant 8 (15%) 44 (85%) 52 (26%)
Jewish 8 (27%) 22 (73%) 22 (11%)
Other 11 (31%) 25 (69%) 36 (18%)
No preference 22 (35%) 40 (65%) 62 (31%)
Percentages may not equal 100% due to rounding
**Total male respondents: n=57 (28%); Total female: n=149 (72%); Total N=206
religious preference was stated most frequently at 31% of the total.
Mainline Protestants (Methodists, Lutherans, Episcopalians, Baptists, and
Presbyterians) comprised 26% of all respondents and Catholics comprised
an additional 13%. Eleven percent of respondents were Jewish, while
18% specified a religious preference other than the ones named here and
were assigned to the category of Other.
At 90% of the total, Caucasians account for an overwhelming
majority of the respondents surveyed. Only 10% of the respondents
classified themselves as Native American, African American, Hispanic,
Asian American, or Other and are reported here in the collapsed variable
Data are grouped into three broad categories: (1) rights of
stakeholders, (2) cost concerns, and (3) care for the seriously or terminally
In the first category, the respondents were asked to consider three
statements related to the rights of doctors, patients, and insurance
companies or the government with respect to care or treatment decisions
(Table 4.2). When asked how often they agreed with the statement that
doctors have the right to override a patient's decision for treatment, most
respondents (71%) rarely or never agreed that doctors have this right.
Responses to Statements Related to Rights of Parties Involved
Percent (%) of respondents* who agree with the statement...
Statement Always Most times Sometimes Rarely Never
Patients1 personal physicians have the right to override patients' decisions about medical treatment < 1 3 26 45 26
Patients should have the right to any and all treatment, even if survival or cure is unlikely 7 14 36 37 6
Insurance companies and government have the right to limit care in to order to reduce costs 2 8 44 26 19
Twenty-six percent of the respondents sometimes agreed that doctors
have this right, while less than 4% of the respondents mostly or always
agreed that doctors have this right. While they recognized the limits
doctors have, many respondents seemed to recognize limits to the rights
of patients to demand any treatments, even when survival or cure is
unlikely. Forty-three percent of the respondents answered that rarely or
never should patients have the right to "any and all treatments even
when survival or cure is unlikely," while 36% thought patients should
sometimes have this right. When it comes to the government's or
insurance companies' right to limit care to reduce costs, 19% never agreed
with this statement, 26% rarely agreed to this statement, 44% sometimes
agreed with this statement, while 8%.agreed with this most times and 2%
Cost of medical care was addressed by three different statements
(Table 4.3). Most respondents (57%) sometimes agreed that "It is
acceptable to limit some medical treatments because of their expense."
Fifteen percent rarely agreed and 9% never agreed. Sixteen percent agreed
most times that it is acceptable, while only 4% always agreed to this.
When asked how they would respond to the statement that the
cost of treatments should be discussed with patients before treatment is
decided, 50% agreed with this always and 35% agreed with this most
times. Twelve percent agreed with this sometimes, while only 3% rarely
or never agreed with this. A similar statement, should cost be a concern
when giving medical care, received different treatment by the
respondents. Only 14% always agreed with this, 29% agreed with this
most times, and 40% agreed with this sometimes. Seventeen percent
rarely or never agreed with the statement.
Responses to. Statements Related to Medical Costs
Percent (%) of respondents* who agree with the statement...
Always Most times Sometimes Rarely Never
It is acceptable
to limit some medical
treatments due to
their expense 4 16 57 15 9
Costs of treatments should be discussed with patients before treatment is decided 50 35 12 1 2
Cost should be a concern when providing health or medical care 14 29 40 9 8
The remaining statements on the survey concern medical care for
seriously or terminally ill patients (Table 4.4). In a series of questions
concerning limits on medical care, most respondents agreed that care for
the terminally ill should be limited. Sixty-three percent agreed that at
least most times it is right to "limit medical treatment for terminally ill
patients," another 26% agreed it is sometimes right to limit treatments for
terminally ill patients. Yet, 11% agreed that rarely or never is it right to
limit care for terminally ill patients.
Responses to Statements Related to Terminally
or Seriously 111 Patients
Percent (%) of respondents* who agree with the statement...
Statement Always Most times Sometimes Rarely Never
It is right to limit medical treatment for terminally ill patients 16 48 26 7 3
Doctors should take all measures to keep terminally ill patients alive < 1 4 21 45 30
Keeping a person alive in a permanent unconscious state using machinery is acceptable 1 1 10 38 50
All medical efforts should be used for premature babies even if chance for survival is minimal 3 9 26 48 14
Healthy seniors should get same aggressive medical treatment as healthy younger patients 40 38 18 4 <1
Persons with irreversible mental decline should get the same aggressive treatment as others 3 14 32 41 10
Medical guidelines would be helpful in deciding when to stop extraordinary care for terminally ill patients 32 49 16 2 1
Society encourages inappropriate health care for terminally ill patients 5 45 41 8 2
Medical professionals encourage prolonged treatment for the hopelessly or terminally ill 1 13 49 25 13
When subjects were asked if they agree that "doctors or nurses
should take all measures to keep terminally ill patients alive" less than
5% percent thought most times or always should this happen. Also, only
21% agreed that this should happen sometimes, while 75% agreed that
rarely or never should doctors and nurses take all measures to keep
terminally ill patients alive.
When the case was more specific regarding care for patients for
whom recovery is unlikely, respondents were more likely to agree to
limit care. To the statement, "Keeping a person alive in a permanent
unconscious state by relying on machinery is acceptable," 50% never
agreed with this, 38% rarely agreed with this, while 10% sometimes
agreed with this. Less than 3% agreed to this most times or always.
A clear majority of respondents thought that "medical guidelines
would be helpful in deciding when to stop extraordinary care for
terminally ill patients." Thirty-two percent agreed that guidelines would
be helpful always, 49% agreed that guidelines would be helpful most
times, and 16% agreed that guidelines would be helpful sometimes. Only
3% agreed guidelines would be helpful rarely or never.
Most respondents agreed that society and medical professionals
encourage prolonged or inappropriate care for terminally ill patients. Of
all respondents, 86% agreed that sometimes or most times society
encourages inappropriate healthcare for the terminally ill. Similarly, 71%
agreed that sometimes or most times medical professionals encourage
prolonged treatment for the hopelessly or terminally ill.
While the survey data gives us a general sense of what percentages
of people in Denver think about a number of different issues related to
end-of-life care and the rights of various parties, some of the findings are
difficult to interpret.
In an attempt to identify which demographic variables might be
important in a respondent's level of agreement with a particular
statement, measures of association (see Tables 4.5, 4.6, and 4.7) were
calculated for selected statements in each of the three categories outlined
in Tables 4.2,4.3, and 4.4.
The first statement from Table 4.2 says, "Physicians have the right
to override patients' decisions about medical treatment." Over one
quarter of the respondents would never agree with this statement.
However, Chi-square and Cramer's V calculations, Table 4.5, found only a
weak association between religious preference and the degree of
agreement on the statement. Respondents who had no religious
preference were more likely to agree rarely, but were not more likely than
References to Statements on Rights by Demographic Characteristics
HOW OFTEN RESPONDENTS AGREED
DEMOGRAPHIC %Some %Rarely %Never # of X2 Cramer1s
CHARACTERISTIC times Cases value p-value V *
Caucasian 29 43 28
Non-Caucasian 24 57 19 200 1.57 0.46 N.S.
Income < $50,000 28 51 21
Income > $50,000 30 37 33 196 4.87 0.09 N.S.
Male 32 42 26
Female 27 46 27 204 0.43 0.81 N.S.
Ages 19-40 32 49 19
Ages 41-60 21 48 31
Ages 61-88 34 36 31 197 5.88 0.21 N.S.
Catholic 23 42 35
Jewish 30 40 30
Mainline Prot. 39 39 22
Other 42 31 28
No preference 17 60 23 218 15.33 0.05 0.19
*N.S. = not significant at the 0.05 level; no Cramer's V was calculated.
other religious groups to never agree. From this it is difficult to assess
what factors might influence those 26% of respondents who would never
agree that physicians have a right to override patient's decisions.
Two statements in the second category from Table 4.3 address cost
using different language. The first statement in that table says, "It is
acceptable to limit some medical treatments because of their expense."
Fifty-seven percent would agree with this sometimes, while only 20%
would agree with this most times or always. Chi-square tests, Table 4.6,
demonstrate that non-Caucasians were less likely to agree with this
statement (p < 0.01) as were those with annual household incomes of less
than $50,000 (p < 0.02). A similar statement in the same category says,
"Cost should be a concern when providing health or medical care."
Though quite similar to the previous statement, only 40% would now
agree sometimes, while 43% would agree with this most times or always.
As with the previous statement lower income and non-Caucasian ethnic
backgrounds were associated with rarely agreeing with the statement (p <
0.01 and p < 0.05, respectively).
When Cramer's V was calculated for these significant associations,
all returned values of less than 0.31, suggesting that the strength of the
association between income and ethnic background is relatively weak.
References to Cost Statements by Demographic Characteristics
HOW OFTEN RESPONDENTS AGREED
DEMOGRAPHIC %Most %Some- %Rarely # of %2 Cramer's
CHARACTERISTIC times times Cases value p-value V *
Caucasian 21 59 20
Non-Caucasian 5 40 55 201 12.96 0.00 0.26
Income < $50,000 12 58 31
Income > $50,000 26 56 18 192 8.33 0.02 0.21
Male 17 60 23
Female 26 49 26 200 2.56 0.28 N.S.
Ages 19-40 10 65 26
Ages 41-60 28 55 16
Ages 61-88 17 55 28 194 9.19 0.06 N.S.
Catholic 36 40 24
Jewish 19 56 26
Mainline Prot. 17 58 25
Other 11 61 28
No preference 20 61 19 214 7.56 0.48 N.S.
*N.S. means not significant at the 0.05 level; no Cramer's V was calculated.
TABLE 4.6, continued
HOW OFTEN RESPONDENTS AGREED
DEMOGRAPHIC %Most %Some- %Rarely # of %2 Cramer's
CHARACTERISTIC times times Cases value p-value V *
Caucasian 43 42 15
Non-Caucasian 33 29 38 201 7.01 0.03 0.19
Income < $50,000 30 42 29
Income > $50,000 55 36 9 196 18.66 0.00 0.31
Male 51 33 16
Female 39 42 19 208 2.31 0.31 N.S.
Ages 19-40 44 41 14
Ages 41-60 52 36 12
Ages 61-88 30 45 25 198 8.19 0.09 N.S.
Catholic 42 50 8
Jewish 41 35 24
Mainline Prot. 39 40 21
Other 44 36 19
No preference 46 41 13 219 5.00 0.76 N.S.
*N.S. means not significant at the 0.05 level; no Cramer's V was calculated.
This does not help us determine why more people would agree more
often to the second statement, than the first. Subtleties in language may
be important if respondents distinguish between medical treatments and
medical care. Respondents may also perceive cost to be a concern in
general (perhaps, prescribing generic versus brand-name drugs), but this
does not mean that cost is more often an acceptable basis on which to
limit care. "Concern" may be too vague a term. The word "limit" itself
may affect how one assesses the statement. For some procedures,
limiting them could mean withholding altogether, using only for a
certain period of time, using them only on certain individuals, or
reducing their use based on cost.
In the third category regarding treatment for terminally ill patients
(from Table 4.4), respondents were asked to consider nine different
statements, three of which are examined in more detail, Table 4.7. First,
"Doctors and nurses should take all measures to keep terminally ill
patients alive," had 75% of respondents agreeing rarely or never and 21%
agreeing sometimes. A test for association shows that older respondents
(ages 41-60 and 61-88) rarely or never agree with this statement (p < 0.02),
while younger respondents (ages 19-40) are more likely to agree
sometimes or rarely, but less likely to never agree. That is, younger
References to Statements About Seriously or Terminally 111 by Demographic Characteristics
HOW OFTEN RESPONDENTS AGREED
DEMOGRAPHIC %Some- %Rarely %Never # of %2 Cramer' s
CHARACTERISTIC times Cases value p-value V *
Caucasian 25 46 29
Non-Caucasian 29 38 33 202 0.46 0.79 N.S.
Income < $50,000 30 37 33
Income > $50,000 22 51 27 197 4.27 0.12 'N.S.
Male 21 47 32
Female 28 44 29 206 0.90 0.64 N.S.
Ages 19-40 32 54 14
Ages 41-60 19 49 32
Ages 61-88 28 33 39 199 13.11 0.01 0.21
Catholic 31 42 27
Jewish 43 30 27
Mainline Prot. 29 39 33
Other 22 64 14
No preference 16 46 38 220 17.30 0.03 0.20
*N.S. means not significant at the 0.05 level; no Cramer's V was calculated.
TABLE 4.7, continued
STATEMENT DEMOGRAPHIC CHARACTERISTIC HOW OFTEN %Some- times RESPONDENTS AGREED %Rarely %Never # of Cases %2 value p-value Cramer's V *
Caucasian 13 39 49
Keeping a Non-Caucasian 10 40 50 199 0.13 0.94 N.S.
alive in a Income < $50,000 16 34 51
permanent Income > $50,000 10 44 47 194 2.91 0.23 N.S.
11 39 51
relying on Female 13 39 48 203 0.28 0.87 N.S.
acceptable Ages 19-40 21 51 29
Ages 41-60 8 35 57
Ages 61-88 10 29 60 196 16.67 0.00 0.21
Catholic 12 39 50
Jewish 14 45 41
Mainline Prot. 18 32 50
Other 19 33 47
No preference 5 41 54 217 7.80 0.45 N.S.
*N.S. means not significant at the 0.05 level; no Cramer's V was calculated.
TABLE 4.7, continued
It is right
i * ** ** \
HOW OFTEN RESPONDENTS AGREED .
DEMOGRAPHIC %Most %Some- %Rarely # of %2 Cramer's
CHARACTERISTIC times times Cases value p-value V *
Caucasian 64 28 8
Non-Caucasian 48 19 33 201 13.14 0.00 0.26
Income < $50,000 56 29 15
Income > $50,000 68 25 8 196 3.93 0.14 N.S.
Male 68 23 9
Female 61 28 12 205 1.04 0.60 N.S.
Ages 19-40 48 38 13
Ages 41-60 69 23 8
Ages 61-88 70 20 10 198 9.38 0.05 0.15
Catholic 50 31 19
Jewish 59 31 10
Mainline Prot. 65 23 12
Other 53 36 11
No preference 73 20 7 218 8.79 0.36 N.S.
*N.S. means not significant at the 0.05 level; no Cramer's V was calculated.
respondents may be inclined to favor more treatment when compared to
older respondents, but the association is weak (V = 0.21). Religious
preference was associated with rarely agreeing, but not with never
agreeing (p < 0.04). The association was weak, as well (V = 0.20).
To the statement "Keeping a person alive in a permanent
unconscious state by relying on machinery is acceptable," 88% percent of
respondents would rarely (38%) or never (50%) agree with this. Only 10%
would sometimes agree with this. Again, the chi-square test shows that
those in the two older age categories were less likely to agree with this
statement, while the those in the youngest age category were more likely
to agree with the statement (p < 0.01). Other demographic variables were
not associated with responses.
Finally, in response to the statement that "It is right to limit
medical treatment for terminally ill patients," ethnic background and age
category were weakly associated (V < 0.27) with how people answered.
Non-Caucasians were more evenly distributed between mostly or rarely
agreeing, while Caucasians demonstrated a stronger preference for
limiting treatment in more cases. Older age categories also showed a
stronger preference for limiting treatment in more cases of terminally ill
With few demographic variables to explain how people respond, a
number of questions remain unresolved. While we can say that most
respondents support limits on medical care and view some limits as
appropriate, for the approximately 20% who agree to limits sometimes,
there are few specifics that can be elucidated. What does it mean that 26%
of respondents agree that sometimes it is right to limit care for the
terminally ill? Or that 57% agree that sometimes it is acceptable to limit
care due to expense? Under what circumstances will they agree always or
never agree? Are there particular treatments that would influence their
decisions? Are there financial details that would be important in their
Interestingly, when a Cronbach's alpha correlation was calculated
on three statements related to limiting care for the terminally ill, no
strong association was found. The first three statements listed in Table 4.3
were considered because they ask for respondents to consider limits to
care for patients who are terminally ill, but use different language and are
more or less specific. A low coefficient (a = 0.456) indicates there are
inconsistencies with how the same respondents answer similar questions.
This could have several implications for how we interpret the
respondents' understandings or interpretations of differing language.
We would have expected that those who agreed that it is right to
limit care for the terminally ill would not agree that doctors and nurses
should take all measures to keep terminally ill patients alive and they
would not agree that it is acceptable to keep a permanently unconscious
person alive using machinery. We would also expect the opposite to be
true. Instead, we find that the relationship between these variables is not
very strong, and that making assumptions about general beliefs about
limiting medical treatments may not be correct for more specific cases.
Because the relationship is weak, we cannot predict that those who feel
strongly about one statement also feel equally strong on a related
statement. What do we make of this? Certainly, there are limitations
with the survey and there may be internal inconsistencies within the
structure of the statements. The numbers, however, do suggest
something more interesting: that language, phrasing and details of a case
may play important roles in how individuals assess the appropriate range
of treatments. The two statements regarding limits to medical care
analyzed above could be interpreted in many ways, if we imagine subtly
different assumptions being made by the respondent. As Redelmeier, et
al. argue, many patients do not make decisions in an ideal fashion by
gathering information, weighing the costs and benefits, then selecting the
optimal choice. Instead, people reach judgments based on "simplifying
heuristic rules and search until they find an acceptable solution" (1993;
72). They further reason that such a process can lead to inappropriate
decisions being made by the patient regarding clinical, scientific
For instance, when one is in an unconstrained setting and can
think freely about a scenario, one might think that a terminally ill patient
who is very near death would not derive enough benefit from some kind
of organ transplant to justify such practices. But, when that same
respondent is in a real clinical setting and faces their own death or the
death of someone close, they may feel that transplants are justifiable
because there is some hope for recovery and that we should not abandon
the patient. Additionally, the length of time before death may be very
important in how a person responds. At one, three, or six months prior
to death, a patient or family may think differently about their options and
the benefits and burdens. Or they may simply think that it is a doctor's
responsibility to do everything unless the patient or family tells them
A survey like the one used for this study does not allow us to assess
which variables are critical to the respondents' decision-making. In
contrast to these somewhat confusing results from the questionnaires, the
focus group interview phase of this project, despite being qualitative,
yielded clearer, more consistent responses to our questions.
Design and Sample
In addition to the surveys, focus group interviews were conducted
to allow for open-ended discussions in response to three healthcare
scenarios that consider care for someone who is not likely to survive.
Questions were designed to elicit responses to the scenarios with regard to
appropriate medical care, medical decision-making, and the feasibility of
guidelines that would outline what constitutes appropriate care. The
focus group interviews were also conducted with the intent of comparing
professional and lay perceptions regarding appropriate care for the same
scenarios. A focus group guide (see Appendix B) was developed by the
author and a professional mediator who works closely with CCMD. The
content of the guide was carefully followed to ensure that each group
heard the same information, scenarios and questions. The focus groups
addressed two topics: (1) what is considered appropriate care for the given
scenarios, and (2) what issues or concerns are essential to the decision-
making process. Participants were first briefly introduced to the purpose
of the CCMD project and this thesis. Approximately 20 minutes were
devoted to the discussion of each scenario. The interviews were tape
recorded to ensure accuracy. The scenarios are presented in Figure 4.1.
Scenario One: A close relative in her late 80s has
suffered a severe stroke and has been in a coma for
six weeks. In the opinion of her doctors there is no
chance of her recovering her awareness or her ability
to think, reason, or remember. Her current physical
condition is likely to decline slowly over time. She
requires others to feed, bathe and dress her. She
may live in this condition for several years.
Scenario Two: A close relative is 23 years old and
was in a serious automobile accident. She has been
in a coma for six weeks. In the opinion of her
doctors there is no chance of her recovering her
awareness or her ability to think, reason, or
remember. Her current physical condition is likely
to decline slowly over time. She requires others to
feed, bathe and dress her. She may live in this
condition for several years.
Scenario Three: A small, hard working family has an
extremely premature newborn. This infant was born so
prematurely and weighed so little, the chance of
survival is less than 1 in 10. If this baby
survives, she would have a 90% chance of suffering
severe developmental and physical problems.
FIGURE 4.1 - Focus Group Scenarios
Following the reading of each scenario, the participants were asked
how much medical treatment should be given. They were then asked
what other factors might be important, if guidelines would be helpful,
and how disagreements should be handled should they arise between or
within the healthcare team and the family.
Given the sensitive nature of the subject matter, it was not known
how individuals would respond to an invitation to participate, but we
were quite warmly received. In general, people were interested in the
topic and wanted to share their ideas. The sample of the professional
groups were assembled through connections within CCMD. The lay
group of retired men was assembled through an extended chain of leads
stemming from a close acquaintance to the author. In total, five focus
groups were conducted, two with clinical social workers, one with
physicians, one with patient representatives, and one with a group of
retired men. Each group had at least five participants and no more than
Although the focus group questions had been re-worked
numerous times before the first interview, there was no clear notion of
how people would respond to questions that are difficult and many times
quite personal. I was initially struck by two things: one, how little
prodding the participants needed and, two, the overall consistency of
responses. Some of the consistency may be from the fact that the
professional groups are exposed to these scenarios regularly. Yet, in the
lay group, the consistency was maintained and their responses were
direct. As it turned out, many of the participants had dealt with
situations similar to the first scenario. Even so, the candor was surprising
given all of the aforementioned literature and debates on end-of-life
This is not to say the decisions they made were easy or without
discomfort. Among the professional groups where the participants did
not know each other, or knew each other only on a limited basis,
discussions began rather quietly. Answers were relatively short and
opinions appeared to be reserved. However, once the boundaries were
tentatively assessed, the participants became more comfortable, talked
more loudly, and seemed to express stronger opinions. When a
particular issue of common concern was voiced it often was seized by
others and had the catalytic effect of opening up other areas of concern.
In presenting Scenario One, an interesting pattern developed as
more focus groups were conducted. Prior to reading the scenarios, the
participants were informed that the information in the scenario is all they
have to go on. Yet, in every group there was an expressed need for more
details on the case before they answered the first question. Are there
advance directives? Are there family members involved? Is she on a
respirator? When no more information was given, a number of
participants would first defer to hypothetical advance directives. Absent
those directives, they would give an answer. Often the answer was
framed in terms of an "I" statement: "I know what I would want to
happen...." Or as one physician said, "I would want to be given a fatal
dose of morphine." A similar pattern followed the presentation of the
Scenario Two. These two themes suggest that in hypothetical discussions
consideration of the patient's desires figure prominently in how much
medical treatment should be given in the first scenario. Did the patient
tell us in a treatment directive what her wishes are? If not, this is what
my wishes would be.
The discussion of the first question in the first scenario did not
vary hugely from one group to the next. Comfort measures were
overwhelmingly voiced as the proper treatment. Comfort care was
variously defined by the respondents as "pain control", "warmth", "no
bed sores", "no dry or cracked lips". One social worker commented that
comfort measures are sometimes aimed at comforting the family as well.
One of the retired men said that, "Quality of life issues control my
thinking." This factor was reiterated in the two social workers groups as
Among those in the patient representatives focus group, one
participant indicated that cost might become an issue, because
interventions other than comfort care would be costly and of questionable
benefit. A number of individuals, lay and professional, also talked in
terms of "extraordinary" care or "heroic" measures. Most respondents
judged that the patient's condition was irreversible. Based on that
assessment, quality of life considerations, and what they would want for
themselves, nothing more than comfort care was considered appropriate.
Contrast these reasons for limiting care with the more complex
medical ethical perspective that would argue that more aggressive care
would be medically ineffective and therefore not appropriate. Some
ethicists might go on to argue that more aggressive treatment might be
harmful to the patient which would violate the physicians' principle of
beneficence. For this scenario, it appears that comfort care is medically
and culturally appropriate, though it is important to carry with us the
language used to defend this choice in the focus groups. It was not the
language of ethicists, though the connotation of the two are similar in
terms of what it means regarding the care of the patient. The point here
is that speaking in terms of "medical futility" may unnecessarily
cloud the issue for lay persons who understand that aggressive care will
not benefit the quality of life of the patient at any point in the future.
Two other questions were asked regarding the first scenario that
defined more specifically a change in the clinical situation and what
would now be appropriate action. Respondents were first asked to
comment on what they thought if the patient developed pneumonia and
doctors and nurses treated her aggressively. They were then asked what
they thought if the patient's heart stopped beating and she stopped
breathing and the healthcare team used CPR to revive her. To both
questions, responses were unanimously against these interventions.
Again, respondents judged her condition to have no chance of
At this level, one of the physicians raised the issue of medical
futility. Another countered that families might not see it this way. If the
relative developed pneumonia, they might not consider whether it is
appropriate or inappropriate, but they simply are not ready to make a
decision, "so there is pressure to keep things going." Reaching that
decision seemed to be another matter.
The next set of questions focused more on the decision-making
process. One question asks who should make decisions about the type of
care the woman in this scenario should receive. The order in which the
responses came showed surprising consistency. First came the disclaimer
about the "best scenario" or the "ideal situation" in which the patient had
previously indicated what her preferences were. Then, in most instances,
the family, if available, was identified as the ultimate decision-maker.
Other elements were cited, such as proxy law, finances, ethics committees,
or as one of the outspoken members of the retired men's group put, "A
panel of three off the street," but he added, "Not three doctors."
Participants consistently emphasized that the family must
ultimately make the decision about the care of the patient. While many
acknowledged that the physician is essential to the process, the ultimate
decision resides with the family. Furthermore, the family arose in
different contexts. Several times participants mentioned that the family
really needs support in what ever decision they make. Part of that
support, it was explained, comes in the form of education and making
available to them all of the information they need.
The focus group interviews suggest that once everyone in the
decision-making process has the benefit of everyone else's knowledge and
attitudes, there would appear to be more ease and greater consensus
across all the parties involved. In the course of discussion, living wills or
other advance directives were continuously deferred to in the hope that
the patient would have made the decisions about what they considered
In two separate groups it was plainly stated that in the first two
scenarios, the "family becomes the patient" and the "family is the
patient." This meant that the patient is essentially "gone" and efforts
should be directed toward helping the family make its decision. Even
among the physicians, it was made clear that if there is family available to
make the decision, then it really is their decision to make. Of course, as
one physician pointed out, a family cannot force a doctor to provide
the care if it is counter to his or her professional and moral judgment.
Although there was general consensus within the groups, each
group was asked how they thought disagreements should be resolved.
Again, after deferring to advance directives, the ultimate decision came
down to the family. One of the retired men pointed out, however, that
the family must have some scientific knowledge as a "caution to those
who might make rash decisions." The importance of relevant
information seemed to be highlighted in several ways.
Among the professionals, ethics committees were identified as a
key element in the process. Most viewed the role of the ethics committee
as one of "information sharing." Ethics committees should not make the
decisions about care. They are there to make sure everyone, including the
various healthcare professionals involved, hear the same set of facts.
In two of the professional groups, one with social workers and the
other with physicians, many thought that part of the problem with
families reaching consensus about the care of a relative is that there is
often different information being communicated and understood. One
physician noted that sometimes, "You get two or three [physicians] saying
different things." Another physician added that consensus is relatively
easy to reach, "once we get everyone to communicate and know the same
set of facts." She continued by saying, "You have to make all of those
three people get in the same room."
Scenario Two directed discussion toward age and guidelines.
Again, there was consensus among the responses regarding appropriate
care. Initially, responses seemed to be quite strong and definite; "no
difference" from the 80 year old relative given the same conditions.
However, as discussion moved along, the concession was made in one of
the social workers' groups and in the men's group, that there may be a
longer waiting period before the actual decision is made. There is
something qualitatively different about a 23 year old and an 80 year old, in
that the 23 year old has "their [sic] whole life ahead of them," but if the
doctors are to be believed, then the treatment is no different.
The second question related to Scenario Two asked if the
participants thought guidelines would be helpful. No definition was
given as to what a guideline would be or how one might be used. The
initial responses were that guidelines would be helpful in decision-
making. It is important to note, however, that there were concerns
regarding the guidelines. Many expressed concern about the word itself.
Some found it potentially threatening and thought it carried overtones of
a mandate for care decisions. In these discussions it was pointed out that
guidelines might be very appropriate for professionals who must make
clinical decisions. For the family, however, it would be better to talk
about "options for care" that outline what different interventions mean
medically. It was also pointed out that guidelines could be an important
way to support families by demonstrating that this is what others think
and this is how others who have been through this have handled
Guidelines for professionals may be helpful in getting their points
across to the family by being attuned to the informational needs of the
family. In this capacity guidelines may serve to highlight both the most
appropriate medical protocol and the most appropriate communication
protocol. Based on the focus group discussions, guidelines for medical
care should focus on two areas: educating families about the medical
interventions and outcomes associated with the case, and educating
healthcare professionals about appropriate methods for facilitating the
Scenario Three presented more difficulties for the respondents in
all the groups. Discussions quickly turned from considerations of what is
appropriate care to considerations of the factors that need to be addressed
before any decision can be made.
There was some sense that competing interests of autonomy (the
family's right to decide) and justice (long term costs to us all) were at
work here. In the men's group, they agreed that the family had the right
to choose, but contradicted this with a comment that, "at some point
society must decide" what is appropriate. In the professional groups,
family were considered to have a right to choose, but with the disclaimer
that they may have to assume all the responsibilities of their choice,
including any financial burdens.
Cost was raised as a more important factor in Scenario Three. Some
felt that if care were to go on for many months, then cost would definitely
come into the decision-making process. In the physicians' group, one
doctor noted that at some point someone might say, "O.K., it's time to pay
up." This, of course, raises the question of when that threshold has been
crossed, and when we as a community can say, "O.K. it's time to pay up,
our community feels it can no longer support this child, the doctors have
done all they can reasonably be expected to do and must devote their
Tied in with issues of responsibility in Scenario Three was how
apparently ill-prepared families are. Several participants felt that the
parents of a child push for care but fail to consider the longer term
consequences of having to support a child that will probably have many
difficulties. One physician felt that the only long-term support that we
gave them was in the form of "cops and jails to lock them up."
Knowledge of the outcomes of extreme prematurity might make families
reconsider their short term decisions when the long term consequences
are taken into account. However, these longer term consequences may
not be important to the parents in the early stages of decision-making
who face the potential loss of their child.
If we revisit the Karen Quinlan case, the New Jersey Supreme
Court made its ruling in favor of Ms. Quinlan on the basis of the patient's
right to privacy. Nevertheless, her parents were faced with decisions
other than removal of the respirator. We cannot know what burdens, if
any, Karen carried, but we can be certain that her parents bore great
burdens in sustaining her life by choosing not to withdraw artificial
feeding. Regardless of the medical outcome, the family is burdened with
the results, whether life or death.
While the focus groups emphasized autonomy with respect to the
family, some consideration of the burdens to the family were noted. One
patient representative confessed that in dealing with his father's death he
had to consider his and his family's own well-being over and above his
father's declarations that he "wanted everything." The son felt
"everything" in his case was futile and everyone would be better off not
prolonging his death.
I have presented above some of the interesting issues and attitudes
disclosed in the focus group interviews. Before placing these in the
broader context of the reconstruction of authoritative knowledge and
how they impact the design and implementation of guidelines, a short
summary of the general focus group findings might be helpful:
Consensus that only comfort care is appropriate for Scenario
One and Scenario Two.
Families are the ultimate decision-makers, regardless.
Information and communication among all decision-making
parties is essential
Guidelines would be helpful as information regarding options
the family has.
"Guidelines" as a word suggests something stronger than just
"options". There was concern that they would become
Family dynamics are important as there may be disagreement
among family members.
Quality of life was often considered in the judgment of
Most hoped that advance directives were in place in Scenario
One and Scenario Two.
In Scenario Three, long term care for the infant was a major
concern if the infant were to live.
Also in Scenario Three, the burden on society if they must pay
for long term care of infant became an issue.
Cost was more of a concern when the outcome of the patient
was thought to be more ambiguous.
Now we must consider how these opinions and concerns figure into the
content and process of establishing guidelines for the use of intensive care
and in what capacity the guidelines can be implemented if developed.
MAKING DECISIONS ABOUT APPROPRIATE CARE
The reality is, that one dissenting responsible person gets the veto. You can have ten kids
who all say 'Yes, let her go,' and the eleventh kid says, 'No, don't.' The reality is, is that
the dissenter wins.
A pediatrician responding in a focus group
Any attempt to set policy agendas for deciding inappropriate care
must recognize the context in which these policies are made. This means
recognizing the broader social and historical processes, evaluating local
versions of what constitutes the appropriate definition of care and where
the decision-making rights are perceived to lie, and considering the
individual and situational contexts of end-of-life decision-making.
As Starr points out, corporate decisions will likely decide some of
the policy matters for intensive care use. However, his perspective
focuses largely on the loss of physician autonomy. From the changes in
laws and codes noted in Chapter 2 of this paper, and from the strong
consensus among focus groups, it is clear that families are perceived to
have the right to make autonomous decisions. Corporate policies that
infringe upon this right will be impugned by many who feel this is
inappropriate, if not illegal. The corporate factor may be
disproportionately large, when we consider the power imbalance that
might exist between heavily subsidized corporate lobbies and an
uninitiated and poorly organized lay community. Singer may be correct
that political and economic agendas need to be assessed as primary in
However, if we think in terms of the legitimation of decision-
making privileges through the reconstruction of healthcare professionals'
cultural and social authority, then public input becomes a potent tool in
While the focus groups and questionnaire data (see Table 4.2)
suggest that a portion of our community believes in unrestrained liberties
to receive medical treatment, there are burdens that come with that
decision. Family relations can become difficult, more care may mean
making more decisions, and individual and community resources can
become strained. A consensual process of deciding what is and what is
not appropriate end-of-life medical care highlights this burden and opens
the door for sharing responsibility in the decision-making process and
outcome. This may be idealistic, but at least the process of dialogue
between groups with differing backgrounds is a move in the direction
that much of the professional literature calls for. Again, in the focus
groups, the exchange of information and communication was identified
as an important issue.
Many focus group informants held the opinion that better
information for patients and families was critical to facilitating decisions.
Many also thought that guidelines might play an important role in this
process by providing families with a specific document or set of ideas that
reflect the collective thinking of the medical community. One doctor
plainly stated that the guidelines to her meant allowing discussions about
end-of-life care to take place with her patients and their families; to "find
out what it means to them." This is something altogether different from
a guideline that would set the standard for which procedures are or are
not allowed. Noted earlier, there were some strong reservations about
the word guidelines, which suggested to some a mandate. They suggested
that guidelines be labeled as options or information. Also, some believed
that guidelines could provide emotional support by giving the families
and healthcare teams some sense that the community was behind their
decision and sharing in the responsibility. However, this was not to say
that the guidelines were the ultimate authority. That authority was
clearly expressed to lie with the family.
This is an interesting paradox. Although there was consensus with
regard to appropriate care in Scenarios One and Two, there was also
consensus that this should not bind other families or individuals to the
same treatment decision. This takes us back to Kleinman, who argues
that each encounter between doctor and client takes on its own
dimensions, differing across the cultural spectrum. What each of us
brings to the negotiation for care will change the process of deciding on
care. The individual context of decision-making seems to matter. Can
guidelines as mandates operate outside this context?
This raises further questions about content. If guidelines are not
mandates, but are informational pieces, then the content should be
presented in an appropriately understandable manner. If the guidelines
are intended to facilitate the decision-making process, then the content
should include information about the most important contingencies of
the process itself.
As a mandate for care that attempts to forge consensus across
diverse stakeholder groups, a good deal more work lies ahead in drafting
such guidelines. From the questionnaire data it is not clear what level of
consensus can be achieved in Denver. The population of the survey was
relatively homogeneous and yet there is considerable ambiguity and
For instance, when 26% of respondents would sometimes agree
that physicians have the right to override a patient's decision, what does
it mean? What particulars would indicate if the physician had the right
or not? If they knew more about the limits of medicine in general, would
these 26% think differently? These questions reflect limitations of our
survey instruments, surely, but they also highlight the recognition by
informants that decisions are contingent and negotiated. Again,
guidelines as mandates might violate the need for context specific,
The focus group data does suggest that when the details of a case
are known, the decisions are much clearer. There was consensus among
all groups that only comfort care for the 80 year old relative who was in a
coma was appropriate and that the family should ultimately make the
decision. Perhaps a more sophisticated questionnaire would deliver
more accurate data on the finer points of medical decision-making. Even
so, these may be subject to differing interpretations based on one's beliefs,
background knowledge, and assessment of the patient's medical
While the clinical specifics might facilitate a better understanding
for individual decision-making, the focus groups made evident the point
that in dealing with end-of-life care, we are dealing with much more than
the physical body. Two statements highlighted this: "The family becomes
the patient" and "The family is the patient." When cure seems quite
remote in many cases, the caring becomes an important element. In
terms of medically ineffective care, some argue that because the patient
will not recover, the hospital's staff are discharged from any clinical
responsibility to prolong life using any and all means (Mitchell et al.
1993). We are now dealing with a family and are now dealing with a
variety of personal and cultural variables, which are beyond the strict
clinical purview of the physicians. We must now recognize that we may
be dealing with several individuals coming into the process with various
backgrounds and beliefs.
The ethics literature is largely silent on the issue of families. Most
of the discussion concerns the patient and the physician, with the family
being considered only as an attenuation of the patient. John Hardwig is
one commentator who squarely addressed the issue in his article "What
About the Family" (1990). In this, Hardwig points out that the phrase
"the patient is the family" draws attention to the need to discuss the
family, but tends to treat the family as one patient. He further notes that
some patients consider their family interests as part of their interests,
while others do not. The difficult task is to turn from individualistic
thinking to more communitarian thinking. Recall that the focus group
participants commonly deferred to the patient's preferences or their own
preferences, but did not consider how these decisions and the outcomes
would ultimately affect the family.
The dimensions of the family in medical decision-making are
poorly known, but may well become important in reducing the use of
medically ineffective care. Within the focus groups family matters were
frequently raised as complicating factors in the decision-making process.
Many times poor communication between family members was to blame,
or a single dissenting voice made the decision. Or, simply, the burden of
making a decision under conditions of uncertainty delayed a decision or
encouraged a decision for more aggressive care. Even if we consider
Janzen's therapy management groups (1987) and Kleinman's assessment
of conflicting models of illness and disease (1988), there is no systematic
way we can account for individual preferences within a family.
Consistently, however, many participants thought that information in
the form of guidelines would serve the purpose of informing family
members equally and easing some of the burden of decision-making.
In many studies, more questions are raised than are answered.
This one is no exception. At some point, however, practical implications
must be considered and questions put aside. The futile care debate will
continue to simmer and guidelines for medically ineffective care will be
faced perhaps with legislation, by insurance companies, or through
education. All of these avenues face wary professional and lay
communities. Keeping in mind some of the findings of this study,
guidelines must be handled carefully if they are to move the futile care
debate forward with practicable results.
New legislation that has the appearance of taking away freedoms
will probably invite public rebuke or political backlash. Furthermore,
new laws are often interpreted differently by various stakeholders who
assume differing responsibilities. Consider the Baby Doe Rules instituted
by the Reagan Administration in 1983. These rules were established to
prevent the mistreatment of handicapped infants who were otherwise
treatable with routine surgery. Differing interpretations of the rules led to
considerable unrest within the medical community. In part, the policy
was considered to be of questionable benefit to all the babies considered
under the rule, but mostly it fostered a witch-hunt mentality with so-
called "Baby Doe Squads" (Pence 1990:142). The fear of prosecution under
the rules encouraged medically ineffective care for babies who doctors felt
had no chance of survival, but also created intense anger among
pediatricians (Pence 1990:143). Should policies regarding end-of-life
medical care become legislated without a clear understanding by the
public, it seems likely that claims to autonomy will conflict with the
stated policies. In other words, if the public does not see a benefit to
codified guidelines for intensive care use, then attempts to implement
the guidelines will be hampered by differing lay and professional
One could make the argument that by legislating limits on
medically ineffective care, we are removing the burden of decision-
making from the family and removing the burden of cost from society.
While these are logical arguments, they do not account for the current
cultural context and certainly do not acknowledge the attitudes of a good
percentage of survey respondents who feel patients sometimes (36%) or
most times (14%) have the right to any and all treatment. Additionally,
focus group participants, either professional or lay, suggest the ultimate
decision is the family's.
The focus groups tell us that more medical information would
help families decide among themselves what is appropriate and helps
build consensus between families and healthcare professional. One
physician in the context of a focus group made the following comment:
My short experience is that if you give people facts and
information, then allow them the opportunity to tell you what
those facts mean to them, they're usually very good at being clear
about that. And help them sort through what some of the issues
are....The tragedy is when those discussions are not allowed to
Unfortunately, we cannot say whether more information will change a
family's expectations of medicine before they are faced with a difficult
If we return briefly to the concept of authoritative knowledge, to
some extent we should expect the family to share some of the same ideas
as the physicians. It might be reasonable to assume that they would
understand that there are clinical limitations to medicine and that, by
entering into this biomedical setting, they share some responsibility to
recognize those limits which have been agreed upon by the medical
community. But this understanding will not come about unless the
limitations of the possible interventions have been clearly set out before
the family in terms they will understand.
There is still the matter of appropriate and inappropriate care.
Establishing guidelines through a consensual professional-lay approach is
desirable if input from the community is dutifully integrated into the
recommendations. Unfortunately, the results of the questionnaire data
do not provide clear information regarding the content of the guidelines
with regard to what people feel are appropriate or inappropriate
procedures. Furthermore, we have not determined from the survey how
people think about the process of deciding this issue. The focus groups,
however, provide a platform regarding content and process from which
we can pursue specifics. From the focus groups it is apparent that the
subjects felt strongly that with regard to someone who has no chance of
recovery only comfort care should be given. However, across the
spectrum of possible cases, it is not known what community members
agree are appropriate limits to care. In Scenario Three, regardless of what
was appropriate or inappropriate, discussion centered on the proper locus
of decision-making and cost. Here, as in the legal and ethical realms
discussed earlier, autonomy appears to be the dominant rule.
Ultimately, if guidelines are established we should not assume that
process and outcomes will necessarily improve. Recall earlier criticisms
regarding advance directives and the SUPPORT study (1995)
demonstrating that even training aimed specifically at improving
communication and end-of-life care had no significant impact.
Furthermore, educating the public through open dialogue and informing
them through guidelines may prepare them, but it is no guarantee of an
individual's decision-making behavior.
Making policy decisions about futile, medically ineffective, or
culturally inappropriate intensive care use must appeal to both
professional and lay concerns. Moving between the two realms requires
translating between the highly developed language of ethics, codes, and
biology and the highly individualized, context-rich language of concerns,
fears, hopes, and desires. Achieving consensus requires a negotiation of
terms. From this study, it appears that more work is needed before
meaningful negotiations can take place.