Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Futile medical treatment: an analysis of physician decision-making
(USC Thesis Other)
Futile medical treatment: an analysis of physician decision-making
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zesb Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
FUTILE MEDICAL TREATMENT: AN ANALYSIS
OF PHYSICIAN DECISION MAKING
by
Lowell Carl Renold II
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Gerontology)
May 1996
Copyright 1996 Lowell Carl Renold II
UMI Number: 9636371
Copyright 1996 by
Renold, Lowell Carl, II,
All rights reserved.
UMI Microform 9636371
Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90007
This dissertation, w ritten by
Lowell. .Carl. Renold.....................
under the direction of h....i.?..... D issertation
Committee, and approved b y all its members,
has been presented to and accepted b y The
Graduate School, in partial fulfillm ent of re
quirements for the degree of
D O C T O R OF PH ILO SO PH Y
c.
Dean o f Graduate Studies
D a te 3/21/%
DISSERTATION COMMITTEE
( j .
Chairperson
Lowell Carl Renold II Jon Pynoos, Ph.D.
Futile Medical Treatment: An Analysis
of Physician Decision Making
An increasing awareness of modern medicine's inability
to produce quantitative and/or qualitative physiologic
change in many patients has lead to more fully developed
definitions of futile medical treatment. This study
investigated family physicians' decisions regarding futile
medical treatment for disabled patients.
One-hundred seventy physicians responded to a
nationwide mailed survey that included four vignettes that
described disabled patients. The hypothetical patients
ranged in age from 6 years old to 86 years old. Physicians
were asked to choose their treatment recommendations under
three conditions: (1) would they provide a specific
treatment for the patient; (2) why did they reach their
decision to provide or not provide treatment; (3) would they
feel it necessary to discuss their treatment decision with
the patient's surrogate; and (4) if the patient's surrogate
demanded that everything medically possible be done for the
patient, would this have influenced their decision.
Statistical analysis including ANOVA indicated that the
patient's age was a significant predictor of physician
decision making.
The study concluded that the older the patient the less
aggressive the treatments recommended by physicians. The
relationships between physicians' treatment recommendations
and their gender, age, religiosity and hospital affiliation
were also investigated. Physicians with high religiosity
scores recommended more aggressive medical treatments for
the youngest patient than less religious physicians.
The results of this research suggest that despite a
number of well-publicized attempts by researchers to develop
an objective, value neutral definition of futile medical
treatment as well as practice guidelines concerning such
treatment, physician decisions about futile medical
treatment are influenced by subjective patient and physician
factors.
Medical schools, professional medical associations,
allied health practitioners and gerontologists should
continue to educate the public about the most important
predictor of treatment outcome: functional status (Tilles,
1995) . It is also crucial that health care professionals
understand the significance of individual perceptions and
opinions regarding the elderly especially as they relate to
sensitive medical decisions.
DEDICATION
To my lovely wife Christine and my beautiful children
Nicole and Lowell.
ACKNOWLEDGMENTS
A research project of this size requires the support of
many people. My sincere appreciation goes to all who have
provided assistance and understanding throughout my graduate
studies and especially during the writing of this
dissertation.
William Lammers encouraged me to pursue a doctorate in
Gerontology. I shall forever value his counsel and
appreciate his scholarship, ideas and integrity.
Jon Pynoos, Chairman of my doctoral committee, deserves
special thanks for giving generously of his time, expertise,
and encouragement.
Professor Robert Stallings deserves particular
recognition for sharing his time and talent so generously in
this project.
To Alexander Capron for his insightful ideas and
assistance from the beginning.
My wife, Chris, and my two children, Nicole and Lowell,
assisted me in many ways to achieve this goal.
Finally, to my parents, Margaret and Lowell. Thank you
for your guidance and support in countless ways, for your
vision, understanding and mostly for teaching me how to
dream.
iv
TABLE OF CONTENTS
Page
DEDICATION ii
ACKNOWLEDGMENTS iii
LIST OF TABLES vi
CHAPTER
I. INTRODUCTION 1
Background 1
Objectives of the Study 11
Limitations 12
Assumptions 13
Definitions of Terms and Constructs 14
Organization of the Study 15
II. REVIEW OF THE LITERATURE 17
Background 17
The History of Family Medicine 17
Attempts to define Futility 19
Current research on futility 23
Study Questions 24
III. METHODS AND PROCEDURES 26
Research Design 26
Sample Selection 27
Methods and Procedures of
Data Collection 31
The Instrument 35
Variables 41
Scoring 44
Data Analysis 49
IV. RESULTS 53
Description of the Sample 53
Physician Decision Making 60
V
V. DISCUSSION 71
Summary of Results 71
Discussion of Results 75
Suggestions for Future Research 77
Conclusions and Policy Implications 83
SELECTED REFERENCES 75
APPENDIX A: Questionnaire on Futile
Medical Treatment 87
LIST OF TABLES
vi
Table Page
1 The U.S. Older Population, 1990-2050 6
2 Medicare Expenditures for elderly patients
during the last year of life 10
3 A Comparison of Survey respondents and
American Academy of Family Physician members 35
4 Personal Characteristics of Sample of Family
Physicians 57
5 Professional Characteristics of Sample of Family
Physicians 58
6 Physician Treatment Decision 61
7 Scheffe Analysis of Physician Treatment Decision 62
8 Physician Medical Futility Decision 64
9 Scheffe Analysis of Physician Medical Futility
Decision 64
10 Physician Discussion Decision 66
11 Pearson Correlation Coefficients 67
12 Patient A Pearson Correlation Coefficients 68
13 Regression Results for Patient A Futility
Decision 69
CHAPTER I
INTRODUCTION
Background
The concept of medical futility is not new to the
practice of medicine. References to futile medical treatment
are found as far back as Plato's classic philosophic work
The Republic. Although medicine as practiced by ancient
Greek physicians bares little resemblance to high-tech,
modern medicine, the idea that particular medical treatments
rendered to hopelessly ill patients can at some point, along
a spectrum of treatment goals, be considered futile at best
and immoral at worst is not a novel idea. In recent years,
as a result of several diverse and complex factors, a
heightened awareness of modern medicine's inability to
produce quantitative and/or qualitative physiologic change
in many patients has led to more fully developed definitions
of futile medical treatment as well as arguments both for
and against placing limitations on such treatment. It is,
however, only within the last decade that physicians, policy
makers and other health care professionals have been
exploring the issue of medical futility in significant depth
and detail. A computerized database search of the subjects
"futility," "futile" and "futile medical treatment"
generates a reference list originating only as far back as
the late 1980's, for example.
The concern and intellectual effort surrounding medical
futility is intense and extensive primarily because health
care is currently at the top of this nation's public policy
agenda. Decades of rising health care expenditures, a
massive federal deficit and an increase in persons reliant
on government health insurance have all pushed state and
federal health care programs to the brink of insolvency. A
number of health care reform policies currently being
debated by policy makers call for dramatic cutbacks in
expenditures and in some instances even the rationing of
health care services. Health care policy is now being
debated and discussed with a seriousness and intensity
perhaps not seen since the 1960's.
During the 1992 Presidential election, for example,
health care in America was a central and perhaps pivotal
election issue. Although there are many differing opinions
about why health care in America is currently high on the
public policy agenda, there are at least four major
variables that most policy-makers contend contribute to the
current debate.
The explosion of health care technology has certainly
created a vast array of moral, ethical and economic
dilemmas. Over the course of the last several decades,
remarkable advances have been made in surgical procedures,
artificial replacement devices, such as mechanical heart
3
valves and pace makers, and also in the area of pharmacology
which has for example, developed imunosuppressive drugs for
use on patients receiving transplanted livers, kidneys,
hearts and lungs with phenomenal success.
As a result of these advances, medical science now
possesses the ability to prolong vegetative life for a
seemingly indefinite period of time. Infants born with
diseases and deformities that would have proved fatal only a
few years ago are now kept alive by machinery and drugs
never dreamed possible by previous generations. A recent
example of one such infant, kept alive while suffering from
multiple birth defects which only a few years ago would have
proven fatal, is the case of Ryan Nguyen. In October of
1994 Ryan Nguyen was born suffering from multiple
deformities, not the least of which included malfunctioning
kidneys. Members of the treatment team caring for the child
refused to provide kidney dialysis and believed he would not
be an appropriate candidate for kidney transplantation. The
team felt that the child was too severely disabled to
benefit from extensive, costly, aggressive treatment.
Moreover, the team, in consultation with the hospital's
ethics committee, believed that providing anything more
intrusive or aggressive than "comfort care" would be immoral
(Capron, 1995).
The parents of Ryan Nguyen, however, challenged
attempts to withhold treatment from the infant and after a
number of legal maneuvers had the infant transferred to a
hospital that consented to provide the treatment required to
sustain the child's life. Although this incident raises a
number of interesting legal and moral issues, it also serves
as a reminder of the great technological advances made over
the last few decades. A report issued by the United Network
for Organ Sharing stated that in 1994 approximately 18,000
organ transplant operations were performed in the United
States and that as of 1996, over 44, 000 patients were
waiting for donor organs (Serafini, 1996).
The debate about whether or not Ryan Nguyen would
benefit from kidney dialysis and transplantation took place
in an environment in which transplantation surgery on
infants is not only technologically possible but is
occurring with increasing regularity.
At the other end of the age spectrum, patients who
would have certainly died only 30 years ago from
arteriosclerosis now endure one, two and even four bypass
grafts, surviving years and decades beyond surgery.
Recently, there have been a number of well publicized
examples of sports heroes, musicians and celebrities who,
after years of drug and alcohol abuse, received liver
transplants in almost routine fashion. Liver
transplantation surgery can cost well over $300,000. These
medical, technological and pharmacological advances all
contribute to the escalating cost of health care in America
(Callahan, 1987).
Demographic and social factors also add to this
nation's health care problems. America is, by many
accounts, the most diverse society in the world. Vast
differences in cultural values, beliefs and norms exist in
all states, cities and communities. Along with this
diversity there exists tremendous differences in wealth and
income distribution among its citizens. Some people, for
example, have excellent health care coverage and access
while others living in the same community have none. In
almost any public school in America, some of the children
will have no health care access, others publicly financed
health insurance and still others have the best health care
coverage and access available.
Additional demographic changes such as the aging of
America contribute to the health care debate. The United
States is experiencing an enormous population shift, a
transformation unprecedented in human history. The swelling
of America's older population with children born between the
years 1945 to 1960, the Baby Boom, has already resulted in
changes to several major federal policies such as Social
Security and Medicare.
Table 1 illustrates the rapid growth in the number of
persons over the age of 65 in the U.S. between the years
1900 to 1990. The table also shows that by 2040 it is
projected that one in five Americans will be over the age of
65.
Table 1
The U.S. Older Population, 1990-2050.
Number in thousands.
Year Number Percentage
1900 3, 084 4.0
1920 4, 933 4.7
1940 9, 019 6.8
1960 16, 560 9.2
1980 25,550 11.3
2000 34,921 13.0
2020 51,422 17.3
2040 66,988 21.7
From Population Projections of the United States, 1993.
The large number of elderly citizens is certain to
have a significant impact on America's health care system.
According to Ferrini and Ferrini:
Older people have a higher prevalence
of chronic illness than the rest of
the population, they have longer hospital
stays, visit physicians more often,
and are prescribed more drugs than younger
groups. Also, older people are the prime
users of long-term care institutions (Ferrini
and Ferrini, 1993).
Vocal critics of government funded health care programs
for the elderly, the largest of which is Medicare, argue
that unless federal spending on health care for the elderly
is dramatically reduced and the program restructured,
American taxpayers face certain fiscal doom. Medicare is
the largest health-care program for older people in the
United States. In 1990 Medicare expenditures exceeded $100
billion (Torres-Gil, 1992). Moreover, the cost of the
Medicare program grew 37 percent faster than the economy
during the five years prior to 1990 (Rosenblatt, 1991).
As a result of increased government spending on
programs for the elderly like Medicare, articles have
appeared in newspapers, magazines and journals that portray
the elderly as "Greedy Geezers" using up valuable, limited
health care resources. Philosopher Daniel Callahan, in his
bold 1987 book, Setting Limits: Medical Goals in a Aging
Society, went so far as to argue for the rationing of health
care services and resources based on age. Similar arguments
for aged-based rationing of health care resources have been
made by former Colorado Governor Richard Lamm (1993) and
philosopher Norman Daniels (1988).
In addition to the demographic transformation taking
place in the United States, economic changes are also
occurring. A report issued by Levy and Michel (1986) of the
Joint Economic Committee of the U.S. Congress indicated that
between the years 1951 to 1960 the rise in wages for an
average worker increased by 118 percent. Similarly, between
the years 1961 to 1970 the rate of wage increase was 108
percent. Between the years 1971 to 1990 however, the rate
of increase of real wages was a mere 16 percent (Thorson,
1995). It is important to note that it was during the
1960s, a period of rapid wage growth and unprecedented
economic expansion, that the two largest government health
care programs, Medicare and Medicaid, were implemented.
The decline in real wages, the shift from higher paying
manufacturing jobs to lower wage service jobs and the rise
of budget and trade deficits has resulted in increased
political pressure to cut expensive government health care
programs. Politicians are now considering ways to satisfy
constituents who are perceived as desiring lower taxes and
less government intrusion.
These demographic, social and economic changes along
with decades of high levels of defense spending have
resulted in increases to the Federal budget deficit.
According to Kingdon (1995), in times of excessive budget
deficits, a nation's large social programs, Medicare and
Medicaid for example, become the prime targets of policy
makers eager to reduce federal expenditures. The fiscal
realities of the 1980s and 1990s have forced policy makers
to take a critical look at both public and private health
care expenditures.
Additionally, a great deal of attention is now focused
on the elderly who, although in 1996 are only 13% of the
U.S. population, account for over one third of total health
care spending in the U.S. Lubitz and Riley (1993) note that
research into the amount and types of medical services
rendered in the period before death to elderly patients has
created the popular impression that a greater share of
resources is being devoted to dying patients than in the
past.
Lubitz and Riley (1993) analyzed data from the
Continuous Medicare History Sample to determine the
distribution of Medicare payments in the last year of life,
according to the number of days before death. Table 2
illustrates the results of their findings and reveals that
in both 1976 (the first year of complete data) and 1988 (the
latest data) almost half of all Medicare costs in the last
year of life were incurred in the last 60 days of life, and
about 40% were in the last 30 days.
10
Table 2
Medicare Expenditures for elderly patients
during the last year of life.
Days of Life 1976 1988
1 to 30 40% 40%
31 to 60 12% 13%
61 to 90 6% 5%
91 to 120 4% 4%
121 to 150 4% 4%
151 to 180 4% 4%
181 to 210 4% 4%
211 to 240 4% 4%
241 to 270 3% 3%
271 to 300 3% 3%
301 to 330 3% 3%
331 to 360 3% 3%
Source: Lubitz and Riley: 1993.
It is in the context of this larger health care debate
that questions about medical futility have arisen. Although
a number of recent, well publicized articles have attempted
to argue that the denial of futile medical treatment to
hopelessly ill patients can be an objective, impartial
exercise, others reject this contention. The medical
futility debate is, by nature, about resource allocation and
scarcity, access to service and rationing. Any serious
discussion of medical futility must consider the basic goals
of the medical profession. The concept of medical futility
brings to opposition the conflicting social and medical
principles of patient autonomy and the "rescue principle."
Medical futility affects the patient/physician relationship
as well as the principle of informed consent. This study
will examine the attitudes of Family Practice Physicians
concerning futile medical treatment for severely disabled
patients with life threatening conditions. Specifically,
this study will attempt to determine if physician and
patient characteristics affect decisions concerning futile
medical treatment, or, whether these decisions are arrived
at through objective, impartial calculation.
Objectives of the Study
The purpose of this study is to determine if there are
significant differences in physician treatment decisions for
younger disabled patients with life-threatening conditions
compared to older patients with similar medical problems.
"Treatment decisions" include whether or not the physician
would provide a specific treatment and also the rationale
for providing or not providing the treatment.
Justifications for providing or not providing treatments
represent a continuum which ranges from definitely
beneficial to definitely futile. Moreover, this study will
attempt to identify certain demographic, background and
organizational variables which have a measurable impact on
treatment decisions by family practice physicians for
disabled patients with life-threatening conditions.
Specifically, this study focuses on the relationships
12
between these variables and the decisions of physicians who
are members of the American Academy of Family Physicians.
A nationwide, random survey has provided the data base
for this study. The study used four vignettes which
describe disabled patients ranging in age from 6 to 75 with
life-threatening conditions. This study considered the
relationship between a patient's age, the characteristics of
the physician (age, gender, religiosity, etc.), the type of
hospital the physician practices in (teaching vs. private),
and the treatment recommendation for the particular patient;
whether or not the physician believed the treatment was
futile or beneficial; if the physician would discuss the
treatment with the patient's family/surrogate; and if
family/surrogate demands for aggressive levels of treatment
would influence the physician's treatment decisions. The
study examined individual physician decisions as well as
patterns of decisions found among physicians at specific
types of hospitals.
Limitations
This study was a nationwide survey designed to analyze
physician attitudes about futile medical treatment for
severely disabled patients with life-threatening conditions.
The survey was mailed to a random sample of 400 Family
Physicians from a total population list of 14,965. Although
surveys are widely used in social science research as a
13
means of data collection, they are not without their risks
(Isaac and Michael, 1983). The following are some of this
study's limitations.
First, it is possible that the survey was completed by
respondents who were accessible and cooperative. Surveys
may arouse response sets and are vulnerable to over-rater or
under-rater bias (the tendency of respondents to give
consistently high or low ratings).
Second, the availability and use of other predictor
variables may change the regression equation derived from
this study.
Third, because this model was based on data collected
from Family Physicians obtained from a sample provided by
the American Academy of Family Physicians, the results of
this study apply only to that population. The implications
of this study might be useful however, when analyzing
practice characteristics of other physician specialty
groups.
Assumptions
There are several principles relevant to this study of
medical futility. Although not exhaustive, these principles
represent the major presuppositions pertinent to research
utilizing a nationwide, random sampling technique.
First, survey respondents provided answers to questions
about hypothetical patients that can be generalized to real-
14
life, practice situations. The survey results contain
demographic, biographic, and professional variables that may
be used to predict physician decision making regarding
futile medical treatment for disabled patients with life-
threatening conditions.
Randomization will produce research results that are
common and generalizable to physicians in other geographic
regions of the U.S. Moreover, statistical analysis of the
data produced meaningful results that can be replicated at
a future date.
Definitions of Terms and Major Constructs
The term "futile medical treatment" is defined as a
medical intervention that cannot achieve the goal of the
intervention no matter how frequently or regularly it is
administered. The term futile implies failure to achieve a
desired result. Although there is considerable debate
surrounding exactly when and what types of interventions are
futile, in this study, the term is meant to describe, in its
most literal sense, treatments that are not medically
efficacious.
The term "disabled patients with life-threatening
conditions" is meant to describe the group of hypothetical
patients in this study. Disabled means incapacitated by
illness, injury, or wounds. The hypothetical patients are
imperiled, catastrophically ill and have significant
15
handicaps. The terms all describe a class or group of
patients with severe physical and/or mental disabilities.
The term "provide" is intended to include the
management, authorization, and administration of a specific
treatment or treatment regime by a physician for a
particular patient.
The term "discuss" is used to describe the process by
which a physician provides counsel, advise, information,
facts and opinions regarding patient prognosis and diagnosis
to either the patient's family or surrogate.
The term "decision" means the act or process of
deciding. A decision is a determination arrived at after
consideration. Physicians in this study are asked to make
decisions about patient treatment options after taking into
consideration professional standards as well as patient
facts.
The term "influence" in this study is used to describe
the action on the part of the patient's family or surrogate
to affect, direct and/or control the provision of patient
treatment options.
Organization of the Study
This dissertation is composed of five chapters.
Chapter I introduces the study, identifies the research
objectives and questions, lists the limitations and
assumptions of the study, defines terms which are unique to
16
this research, and explains the study's organization.
Chapter II reports findings from a review of the literature.
Chapter III describes the methods to collect and analyze the
data. Chapter IV presents and explains the findings.
Chapter V is reserved for the summary, conclusions, and
recommendations.
CHAPTER II
REVIEW OF THE LITERATURE
Background
The debate about futile medical treatment has a
relatively brief history. It is only within the last few
years that medical professionals and policy makers have
attempted to define medical futility as well as discuss the
moral, social and political implications of withholding
treatment to severely disabled patients. Moreover,
attempts have been made to provide decision making protocols
to health care professionals in various clinical situations
where futile medical treatment decisions may arise. One
specific physician specialty group likely to encounter
decisions about futile medical treatment is the family
practice physician.
The History of Family Practice Medicine
Family medicine is a nationally recognized field of
professional specialization. In order for a physician to
become board certified in family medicine, he or she must
serve a residency period usually no shorter that 3 years and
pass a national exam. Family medicine has been recognized
as a professional specialization in the United States for
over 30 years. Family physicians provide basic health care
to all members of the family with an emphasis on prevention
and psychosocial issues (Ferrini and Ferinni, 1993).
18
Membership in the American Academy of Family Physicians, the
nation's largest professional organization of its type,
numbers over 14,500.
Over the last several decades, federal policy has
attempted to provide limited incentives to individuals
interested in pursuing careers in Family medicine. Title
VII of the Public Health Service Act, "which authorizes
funding for departments of family medicine and for
residencies in family medicine, is the major explicit
federal source of such support"(Budetti, 1993).
Family physicians, as defined by the Social Security
Act, are the primary medical case managers in our society.
Decreasing interest in family medicine as a field of
specialization has lead to several recent health care reform
proposals that examine incentives that would encourage more
medical students to pursue careers in family medicine. In
the future, as a result of demographic changes, physicians
certified in family medicine will likely require increased
training in geriatric medicine. Currently, requirements for
training in geriatrics for family physicians are minimal.
Physicians need additional education and training about
chronic illnesses and the special needs of the elderly which
include quality of life issues (Ferinni and Ferinni, 1993).
19
Attempts to define Futile Medical Treatment
Several authors have wrestled with basic concepts
related to futile medical treatment. At the core of the
futility debate is a definitional question. How is futile
treatment defined? How does a physician recognize when
treatment should be denied to a patient on the grounds of
futility? What is the physician's obligation regarding the
disclosure of the decision to withhold treatment?
Schneiderman, Jecker and Jonsen (1990) contend that
medical treatment that leaves unaltered a state of permanent
unconsciousness and/or does not end dependence on intensive
medical treatment should be considered futile and should
therefore not be administered regardless of
patient/surrogate demand. They define futility as: "Any
effort to achieve a result that is possible but that
reasoning or experience suggests is highly improbable and
cannot be systematically produced" (Schneiderman, Jecker and
Jonsen, 1990) . These authors posit a twofold test to
determine if a particular medical treatment is futile: A
quantitative test whereby "physicians conclude (either
through personal experience, experiences shared with
colleagues, or consideration of reported empiric data) that
in the last 100 cases, a medical treatment has been useless,
they should regard that treatment as futile"; and a
qualitative test whereby a physician attempts to determine
20
if a given treatment "merely preserves permanent
unconsciousness or fails to end total dependence on
intensive medical care (Schneiderman, Jecker and Jonsen,
1992)
In contrast to the view of Schneiderman, Jecker and
Jonsen, Lantos, et al., (1989) suggest that physicians
should not withhold medical treatment based on an ambiguous
futility definition and that, because the principle of
medical futility lacks concrete ethical and professional
guidelines, physicians should arrive at treatment decisions
in conjunction with patients or their surrogates. The
authors reject the Schneiderman, et al., argument that
physicians can arrive at conclusions concerning futile
treatment based, in part, on a quantitative measure. Lantos
et al., attempt to address the issue that futility may have
different and perhaps ambiguous meanings to medical
professionals. They admit, however, that some medical
therapies may indeed, have no beneficial effect for
particular patients. They resolve this conflict by
presenting a model for determining when treatments with low
efficacy should be withheld. The model, unlike that
proposed by Schneiderman, et al., includes consideration of
available resources as well as potentially achievable goals
as determined by the physician and the patient.
21
Physician Practice and Futility
A number of articles have recently appeared in
physician journals in which authors argue that
cardiopulmonary resuscitation (CPR) should not be used on
patients suffering from terminal illness or who have "poor
quality of life" (Blackball, 1987; Faber-Langendoen, 1991).
CPR, originally conceived of in 1960 by Kouwenhoven, was
intended for selective use on patients with acute illness.
As a result of widespread training among healthcare
professionals of all specialties, i.e. nurses, physicians
and paramedics, CPR has been expanded to include a wide
range of patients, even the terminally ill. According to
Faber-Langendoen (1991), in most acute care hospitals, CPR
at the time of clinical death has become the standard of
care. Blackhall (1987), moreover, argues that not only is
CPR inappropriate in a number of these cases, but that by
offering CPR to hopelessly ill patients, physicians are
falsely optimistic and offering unrealistic hopes for
miracles. CPR, according to Blackhall (1987), should not
even be considered in patients with metastatic cancer.
Offering CPR to these patients represents bad faith because
doing so implies a potential benefit when there is none.
Recent studies suggest that physicians tend to
discriminate against patients with cancer by not offering
CPR when compared to patients with different end-stage
22
diseases with similar prognoses. These studies suggest that
physicians are less likely to offer CPR in all terminally
ill cancer patients irrespective of cancer type (Faber-
Langendoen, 1991).
Futility and Physician/Patient discussion
Capron (1994) contends that physicians have, within the
last few years, utilized the futility principle in response
to the growth of patient autonomy. He cites as an example
the recent decision of the United States Court of Appeals
for the Fourth Circuit regarding "Baby K". The court ruled
on appeal that the hospital and physicians treating Baby K,
a severely disabled anencephalic infant, were required to
continue care requested by Baby K's mother even though the
hospital and treating physicians believed the treatment
futile. Capron (1994) argues that in this case, and others
like it, the futility principle is not only unnecessary but
dangerous. He argues that futility only superficially
provides an objective measure by which physicians can
discontinue patient treatment. The measure serves as an
excuse for physicians to not discuss treatment options with
patients or their surrogates. Capron also believes that
health care administrators and policy makers are using the
futility argument to ration health care resources. He
refutes the notion that by not providing futile medical
treatment resources will be saved. Capron (1994) cites a
23
recent study which found only minor savings resulting from
withholding futile treatment. He believes that the moral
cost of withholding medical treatment far outweighs any
economic benefits.
Factors affecting physician decision making
Several studies have examined the specific ethical
implications of medical technologies that preserve life
while prolonging patient suffering. Crane (1975) studied
this dilemma and concluded that physicians categorize
chronically or terminally ill patients in terms of the
clinical considerations of the illness as well as the degree
to which the patient is capable of interacting with others.
Crane's study implies that patient ability to engage in
gregarious functions is a significant predicator of how
aggressively the patient is treated (Crane, 1975).
Current research on futile medical treatment
Curtis, et al., (1994) conducted a study based on
interviews with 44 internal medicine resident physicians at
the University of Washington in Seattle. Subjects were
interviewed to identify the reasons used when Do Not
Resuscitate (DNR) orders were written. Researchers
attempted to distinguish when and how often physicians used
quantitative futility as a justification to write a DNR
order verses using qualitative futility as a justification.
The results of the study indicated that physicians used
quantitative futility as a rationale for writing DNR orders
in 51 percent of the cases studied (Curtis, 1994).
Physicians used qualitative futility as a rationale for DNR
orders in 42 percent of the cases studied. The study
determined, however, that in only 57 percent of the cases in
which qualitative futility was used as a basis for writing a
DNR order on a patient able to communicate did the physician
actually discuss the treatment decision. The researchers
also identified race as a significant predictor when
quantitative futility was used as a reason for writing a DNR
order.
Study Questions
The objectives for this study and the review of the
literature lead to the following specific study questions:
1. Is patient age related to a physician's decision
to provide medical treatment?
2. Is patient age related to a physician's
determination that treatment for the patient is
futile?
3. Are individual physician differences (age, gender,
religiosity) related to treatment and futility
decisions for disabled patients with life-
threatening conditions?
Are there institutional (type of hospital)
differences in treatment and futility
decisions for disabled patients with life-
threatening conditions?
26
CHAPTER III
METHODS AND PROCEDURES
Research Design
In order to test the hypotheses regarding futile
medical treatment for disabled patients, this study used a
repeated measures design. A repeated measures design
exposes each subject to multiple treatments. This multiple
exposure to treatments allows the researcher to determine if
multiple variations exist within the same sample subject
(Kidder and Judd, 1994). The repeated measures design
affords a number of research advantages: (a) because factors
are varied within subjects, fewer subjects are required to
produce valid results; (b) statistical strength is increased
because subjects are counted repeatedly; (c) greater
precision is gained by using each subject as his or her own
comparison (Kidder and Judd, 1994).
There are, however, a number of limitations to the
repeated measures design. Because questions are asked
numerous times, multiple exposure to the questions under
study is inevitable. Participants are often aware of the
research questions under consideration, and this may affect
the way they answer. Another limitation is that the
sequence in which questions are posed may affect subject
responses. Subjects may become tired and begin anticipating
27
questions and answers by the time they receive the second,
third or forth treatment.
In order to anticipate and alleviate these potential
problems, three major steps were taken in designing the
survey. First, the length of the survey was kept to a
minimum. A pilot survey utilized six vignettes with six
treatment options which totaled 14 pages of text. Both the
vignettes and treatment options were reduced in an attempt
to deal with both respondent fatigue and question
anticipation. Next, the background questions asked of the
physicians were also kept to a minimum. Only those
questions that have been reported relevant in previous
research on physician decision making were included in the
survey. Finally, with the assistance of a panel of
physicians, patient background information provided in the
vignettes was reduced to a minimum level. Only information
specifically relevant and necessary for the physician to
make a judgment about patient treatment was included in the
vignette.
Sample Selection
A random sample of 400 Family Physicians was drawn from
a list of 14,965 Family Physicians complied by the American
Academy of Family Physicians. The Academy's mailing list is
generated and updated daily by their Computer Services
Division. Requests for membership lists of this type are
28
reviewed by the Academy and must meet a number of criteria
before they are released for research purposes. The Academy
requires that the survey instrument be designed to produce
valid and reliable results and that these results should
have the potential to significantly add to the body of
knowledge in the medical field, particularly family
medicine. Moreover, the survey must produce a positive
benefit for the membership of the Academy.
The survey utilized four vignettes which describe
hypothetical patients. All of the patients described are
seriously ill. A number of the treatment options considered
in the survey require the physician to consider the use of
sophisticated medical equipment and the application of
complex medical treatments. It was appropriate, therefore,
to include in the sample only hospital-based family
physicians currently practicing medicine. The .American
Academy of Family Physicians' computerized mailing list is
perhaps the most up-to-date source of family physicians
practicing in the United States. The Academy is able to
randomly select physicians from the membership list
utilizing a number of criteria specified by the researcher.
It was therefore possible to obtain a random sample for use
in the study with the following two specific criteria: 1)
all physicians must be hospital based; and 2) physicians
must be currently practicing family medicine.
Family medicine is the field of medicine in which the
physician a) serves as a physician of first contact with
families and with patients of all ages and provides a means
of entry into the health care system; b) evaluates the
patient's total health needs, provides personal medical care
within one or more fields of medicine, and refers the
patient, when indicated, to appropriate sources of care
while preserving the continuity of care; c) assumes
responsibility with the patient for comprehensive and
continuous health care and acts as a leader or coordinator
of others providing health services; and d) considers the
patient's total health care within the context of his or her
environment, including the community and the family or
comparable social units (Title 42 CFR 57.1603 pg. 345-46).
Family physicians were selected as the most appropriate
sample group for a number of reasons. Family physicians
are: 1) the physicians most likely to come in contact with
patients from all age ranges; 2) the physicians most likely
to be a primary health care provider; and 3) the physicians
most likely to make real-life decisions regarding day to day
patient treatment. While other physician specialty groups
would have provided interesting information regarding the
treatment of elderly patients (geriatricians, for example),
it would have been difficult to compare treatment decisions
across a variety of age ranges from that sample population.
30
Geriatricians are unlikely to have extensive experience with
patients under the age of 55-unlike family physicians who
regularly treat patients from across the age spectrum.
Sample Size
The size of the sample required to achieve the
necessary results was determined through the use of a power
analysis. The power analysis is used to determine the
sample size required "in order to detect any effects due to
the independent variables, given (a) the size of the effect
of these variables, (b) the type of statistical test to be
utilized, and (c) the level of significance of the study"
(Rudestam and Newton, 1992). This test allows the
researcher to determine the likelihood of committing a Type
II error in which a false null hypothesis is not rejected.
This means that an effect existed but was not detected by
the study. Rudestam and Newton (1992) state that "power is
equal to one minus the probability of a Type II error.
Thus, if a probability of a Type II error is .15, power is
.85 (1-.15 = .85)(p. 65)". The following table illustrates
the number of subjects required for a study based on a level
of significance of .05 and a desired power of .80. Both of
these percentages are considered to be generally acceptable
criterion.
31
Effect size Total N Required
Small
Medium
Large
786
128
52
Based on the Power test using .05 as a level of
significance and a desired power of .80, with medium effect
size, the goal of the survey was to produce a minimum of 128
valid responses. Response rates for mailed surveys can vary
anywhere between 10 percent to 50 percent. A middle
response rate of 25 percent was assumed given the nature of
the sample population. Physicians are assumed to be
somewhat more likely to respond to a survey about practice
characteristics than would members of the general population
responding to a mailed survey about shopping, for example.
An initial pilot of 40 family physicians in Orange County,
California yielded a 39% response rate. Based on a power
analysis, using 128 as the goal for total valid surveys
required, and assuming a 25%-39% response rate, it was
determined that 400 surveys would be mailed nationwide to
family physicians.
Based on a thorough review of previous studies that
examined the issue of medical futility, it was determined
that a nation-wide, cross-sectional, random sample of family
physicians would provide the most useful data for analyzing
research questions about medical futility. The survey of
Family Practice Physicians utilized vignettes with a Likert
Methods and Procedures of Data Collection
32
item scale to investigate the question of medical futility.
The method of using vignettes provided useful information
for several reasons.
Numerous researchers have utilized vignettes and case
histories to determine physician attitudes about patient
treatment and discussion of treatment options (Otten, 1985 &
Crane 1975). Crane (1975) notes that:
One justification for the use of case histories
to assess physicians' attitudes toward
these issues is that the technique resembles
to some extent the tests which physicians take
in order to become board certified. These
examinations also present typical cases
and ask the physician to indicate what treatments
he would use. (p. 23).
Moreover, according to Kidder and Judd (1994), the use
of a questionnaire with a Likert item scale has the
following distinct advantage: Each individual response is
assumed to be related to the underlying construct (in this
case, the futility decision) and scaling techniques can
combine them into a single overall measure of the construct.
The survey was pilot tested February 2, 1995. The
pilot test, as stated was administered to forty (40) family
practice physicians in Orange County, California. A
reliability coefficient was used as a test of reliability
for the four vignettes used in the survey. The vignettes
were also tested separately for reliability.
33
The initial mailing, including a cover letter was
mailed to 400 physicians on March 9, 1995. Eight days
later, a reminder postcard was mailed. Three weeks after
the reminder, a letter and replacement questionnaire was
sent. After another three weeks, a final letter and
replacement questionnaire was sent. This survey method
combines techniques described by Dillman (1978) and
Creswell (1994). The following timeline was used in the
administration of the survey:
Procedure Date Response
1) First mailing: March 9, 1995 105
2) First Reminder postcard: March 17, 1995 21
3) Second Reminder Package: April 6, 1995 50
4) Third Reminder Package: April 27, 1995 0
Response bias is an important concern for research
utilizing the survey method. Response bias is the question
of whether the participation of nonrespondents would have
statistically changed the overall results of the research.
Response bias was analyzed using a wave analysis. Responses
to selected items by Week 1, Week 2, Week 3, and so forth
will be studied (Leslie, 1972). This procedure assumes that
respondents who return surveys in the final weeks of the
response period are "almost" nonrespondents. If their
responses are not different from those responding in earlier
34
weeks, the argument that no response bias occurred is
strengthened.
One-hundred twenty-six (126) family physicians
responded to the first survey mailing and reminder postcard.
In an effort to increase the response rate, two-hundred non
respondents were randomly selected and sent a second
complete mailing that included a new survey questionnaire.
Fifty of the two hundred responded to the second request.
In order to analyze if response bias had occurred, the
sample was divided into two groups: early and late
responders. An Analysis of variance (ANOVA) was then
performed using these two samples. The ANOVA determined
that no statistically significant difference existed between
the early and late responders.
In order to further assess if differences existed
between survey respondents and non-respondents, a comparison
of responding physicians and the general membership of the
American Academy of Family Physicians was conducted. It is
assumed that because the sample population was randomly
drawn from the Academy's membership roll, non-respondents
have similar personal and professional characteristics as
the general membership of the American Academy of Family
Physicians. Therefore, comparison of survey respondents to
the general membership will assist in determining if
differences exist between respondents and non-respondents.
35
Table 3 compares the age and gender of respondents and the
general membership of the Academy. The results of the
comparison offer additional evidence that respondents to the
survey are similar to the general membership of the Academy
in terms of age and gender.
Table 3
A comparison of Survey respondents and American Academy of
Family Physician members.
Description Respondents Academy Members Difference
Age
31 - 35 18% 13% 5%
36 - 40 24% 21% 3%
41 - 45 28% 23% 5%
46 - 50 12% 16% 4%
51 - 55 5% 8% 3%
56 - 60 2% 6% 4%
61 + 11% 13% 2%
Sex
Male 81% 80% 1%
Female 19% 20% 1%
Source: American Academy of Family Physicians: 1996
Based on this analysis, the two groups were combined into a
single group for the purpose of data analysis.
The Instrument
The research instrument utilized in the study of
medical futility was a mailed paper and pencil survey in
which study participants were mailed a questionnaire ten
pages in length. The questionnaire was accompanied by a
36
letter of introduction, a return postage-paid envelope and a
pre-paid postcard that was to be mailed separately from the
survey. The purpose of the postcard was to track survey
respondents while maintaining their confidentiality and
anonymity. The introductory paragraph of the survey
provided the participant with a brief summary of the
research topic as well as instructions for proper survey
completion. The paragraph also acknowledged that the
vignettes, while containing some clinical information, were
nevertheless hypothetical and required the physician to
answer based on the limited amount of information.
Mailed surveys, while useful in measuring attitudes,
certain abilities, aptitudes and knowledge of a particular
topic or field, among other things, are subject to a number
of limitations. Mailed surveys are, for example, unable to
completely simulate real life situations. Kerlinger (1973)
argues that surveys are useful, however, when assessing the
relative incidence, distribution and interrelations of
naturally occurring phenomena. Naturally occurring
phenomena refers to the processes of life as they occur
(Kidder and Judd). Surveys are useful when manipulation of
variables is not possible for either ethical, financial or
practical reasons.
Another limitation of mailed surveys is that in-depth
explanation of underlying motives, feelings and attitudes is
37
difficult to assess because face-to-face questioning and in-
depth probing of respondents is not possible. The
relationship of independent variables to dependent variables
can often be more easily explored in personal interviews and
true experiments than in impersonal, mailed surveys.
Reliability
Reliability is the consistency of the findings of a
survey. A reliability test is used to determine if a
question produces the same results when it measures twice an
individual believed not to have changed in the time between
measurement. Correlation analysis was used to determine the
reliability of the survey items. An average item-total
correlation analysis was performed using the results of
Pilot II. Average item-total correlation is done by
correlating each item with the total score and averaging
those correlation coefficients. This gives a measure of how
much the answer to each item agrees with the sum of answers
to the other items. The test measures whether the
instrument assesses the same variable with each additional
item. A high reliability score would indicate that
respondents maintained the same relative positions on the
survey items. The reliability score analysis indicated an
acceptable item-total correlation of .91.
38
Internal Validity
Internal validity is the ability of the survey to
predict what it was designed to predict. The question of
internal validity is perhaps the most central issue related
to survey research. Internal validity is the conclusivness
with which the effects of the independent variables are
established as opposed to the possibility that confounding
variables may have caused the observed results. Surveys are
less likely to be as internally valid than are true or
quasi-experiments. Surveys, if properly constructed, can
however, yield highly valid results.
Face Validity
In order to assure the face validity of the instrument
used to measure medical futility, a number of steps were
taken in the design phase of the survey. First, a thorough
review of relevant journal articles and monographs resulted
in the identification and isolation of a number of
independent and dependent variables. For example, a recent
survey by Schneiderman, et al., (1993) asked patients with
life threatening illnesses a series of questions about
preferences for life saving medical treatment. Treatment
options included cardiopulmonary resuscitation (CPR),
artificial nutrition/hydration and hospitalization for
pneumonia. Patients were asked to respond on a 5 point
scale. Physicians' personal preferences for self-treatment
39
(if they were in similar situations) were then compared to
those of their patients to determine if physician attitudes
influenced their perceptions of patient preferences.
Gillick et al., (1993) also surveyed physicians and
nurses to examine personal preferences for life sustaining
treatment at the end of life. Treatment options included
CPR, mechanical ventilation, artificial nutrition/hydration,
major surgery, kidney dialysis, chemo therapy, minor
surgery, invasive diagnostic tests and pain medication.
Physicians and nurses were presented with 6 scenarios and
were asked to assume they were the patient under
consideration and in conditions ranging from coma to being
homebound and dependent.
A third study by Pearlman et al., (1994) examined
physician application of quantitative and qualitative
futility principles. This research included interviews
conducted on 44 internal medicine residents.
Finally, an evaluation by a group of judges to
determine if the measures in the survey measured what they
purported to, was conducted using a panel of four family
physicians. The panel of physicians concluded that each of
the vignettes, treatment options and potential responses
were professionally appropriate.
40
External Validity
By choosing a nationwide, random sample of family
practice physicians, the external validity of the study was
increased. External validity is the generalziablity of
research findings to the larger population from which the
sample was drawn. The answers provided by the respondents
in this study and the results provided by data analysis
should be generalizable to other physicians across the
United States.
Structure
The survey was cross-sectional. Respondents are asked
questions at a single point in time. The other type of
survey, a longitudinal design, examines attitudes, opinions,
practices, etc., along a number of points in time. Although
a longitudinal survey of physician attitudes about medical
futility across time would likely produce interesting and
useful information, longitudinal designs are usually
prohibitively expensive and tracking and maintaining contact
with participants across months and years can prove
problematic.
Two random pilot test were conducted to isolate
important variables and also determine potential response
rates. The first pilot test included eight treatment
options and six vignettes. This pilot survey was sent to
family physicians practicing in Orange County, California.
41
During this phase of the research, personal interviews were
conducted with a number of physicians. These interviews
resulted in modifications to the survey which included
limiting both the number of treatment options and vignettes
to four. The second pilot survey also used a random sample
of forty family physicians practicing in Orange County,
California. Response rates for the surveys were 42 percent
for Pilot I and 41 percent for Pilot II.
Variables
The dependent variables in this study are "provision
decision" (PD), "medical futility decision" (FD),
"discussion decision (DD)" and "family/surrogate effect"
(SE). Provision decision is meant to measure whether or not
the physician would provide or not provide the specific
treatment option listed. Medical futility decision is meant
to measure the physician's evaluation of the validity and
clinical efficacy of the treatment choice. The treatment
choices represent a continuum of possible treatment options
from more acute treatments (CPR) to less acute treatments
(major surgery).
The dependent variable, discussion decision, measures
whether or not the physician believes it is necessary to
discuss his/her decision regarding providing or not
providing a given treatment with the patient's surrogate or
parent. Family/Surrogate effect is meant to measure the
42
extent surrogate demands that "everything medically possible
be done for the patient" have on physician decision making.
Treatment options included in the survey include CPR,
Artificial Nutrition/Hydration, Kidney Dialysis and Major
Surgery. These treatments were selected for inclusion in
the study for two primary reasons. First, the treatments
listed in order, CPR through Major Surgery, represent a
continuum of urgency. Decisions involving the various
treatments can take place over the course of a few seconds
or a few weeks. The need for CPR in a patient experiencing
cardiac arrest is immediate, for example. A patient in
cardiac arrest in need of CPR may live for only minutes. A
patient in need of Artificial Nutrition/Hydration may live
for hours or days. Patients requiring Kidney Dialysis may
live for weeks. Finally, patients requiring major surgery
may live for weeks or months.
Secondly, each of the treatments listed are currently
viewed by many in the medical and legal communities as
controversial. Each of the treatments present certain
ethical dilemmas for practitioners whether it is because of
its potential cost, dialysis for example, or because of the
disputed efficacy of the treatment. Blackhall (1987), for
example, in an article entitled "Must we Always use CPR?,"
argues that "CPR is a desperate technique that works
43
relatively infrequently, and in many types of patients,
virtually never".
A simple data sheet was used to collect the background
characteristics of the physicians such as age, sex, and
religious affiliation. Age is used as an independent
variable in the analysis and is analyzed as a continuous
variable. Gender and religious affiliation are included and
considered categorical variables. Hospital affiliation is
also measured as a categorical variable with five mutually
exclusive categories: university based teaching hospital;
community based teaching hospital; community based public
hospital; community based private hospital and other.
Religiosity is measured as a continuous, interval-level
variable using a single forced choice question. Respondents
are asked about the role of religion in their life.
Responses are scored from (1) extremely non-religious to (8)
very important. Higher scores suggest a more religious
person while lower scores indicate a person who attaches
less importance to the role of religion. All of the
demographic variables and the respective scores were based
on similar variables and scores found in the 1994 General
Social Survey.
The independent variable patient age reflects
assumptions regarding the patient's chronological age. This
variable is represented by the four vignettes. The four
44
vignettes represent patient age as an ordinal variable with
four labels. Ranking the vignettes from youngest to oldest
would produce the following order:
Vignette Age Rank
A 6 Youngest
B 36
C 75
D 86 Oldest
Each vignette contained clinical information believed
to represent real clinical situations and conditions. Each
vignette contains information evaluated by a panel of judges
who tested for construct validity. The panel included an
immunologist specializing in the care and treatment of
patients affected by HIV disease, a family physician
specializing in public health, and a family physician
specializing in internal medicine. The judges found the
vignettes to be clinically accurate and appropriate to the
research questions. In order to control for any spillover
effects, the vignettes were presented to the subjects in
random order.
Scoring
Four vignettes were provided to survey participants.
Each vignette described a terminally ill patient, his
underlying condition, age and gender as well as the
statement that the surrogate/parents had not previously
expressed their views about the extent of medical treatment
desired for the patient. The vignettes were followed by a
series of treatment options. Using a Likert format,
respondents were asked to indicate if they would provide the
treatment listed to the patient. Possible responses and
scores were as follows:
Definitely No Probably No I Don't Know Probably Yes Definitely Yes
-2-1 0 1 2
The decision to withhold CPR from patients experiencing
cardiac arrest generally occurs if a "Do Not Resuscitate"
order has been included in the patient's chart. As a
result, the survey question listing CPR as a treatment
option is worded in a slightly different manner than are the
questions about other treatment options. For CPR, the
physician is asked: if in order to prevent CPR from being
administered to the patient, would they write a Do Not
Resuscitate Order (DNR). The same response choices,
"Definitely No" to "Definitely Yes", are provided to the
physician. Scoring for this question is reversed, however.
The question asked if the physician would write an order to
prevent CPR. This means that an affirmative response
equates to a withholding of the treatment. By reversing the
score, responses to the question about CPR can be more
easily compared to other treatment options.
46
Based on the coding strategy used in the survey, the
lower the overall score, the less likely the physician is to
provide the treatment options for the patient. The higher
the overall score, the more likely the physician is to
provide the treatment options.
The next question asked the physician why they reached
their conclusion about providing or not providing the
specific treatment to the patient. Responses and scoring
were as follows:
Definitely Fut. Prob. Fut. I Don't Know Prob. Bene. Definitely Bene.
-2 -1 0 1 2
Based on this coding strategy, the higher the overall
score, the less likely the physician is to view a treatment
option as futile. The lower the score, the more likely the
physician is to view the treatment option as futile.
Discussion decision is also measured using the Likert
format. This question asks if the physician would feel it
necessary to discuss the treatment decision, to provide or
not provide, with the patient's parent/surrogate. Possible
responses and scores were as follows:
Definitely No Probably No I Don't Know Probably Yes Definitely Yes
-2-1 0 1 2
Based on this coding strategy, the higher the overall
score the more likely the physician is to discuss the
treatment decision with the patient/surrogate, the lower the
47
score the less likely the physician is to discuss the
treatment decision.
Family/Surrogate effect, the final question, is
also measured using the Likert format. This question asks
the physician if the patient's parent/surrogate demanded
that everything medically possible be done for the patient,
would this have influenced your decision to provide (or not
provide) any of the following treatments? Possible
responses and scores were as follows:
Yes No
2 -2
Based on this coding strategy, the higher the overall
score, the more likely the physician would be influenced by
the patient's family/surrogate about his/her decision to
provide or not provide the specific treatment listed. The
lower the score, the less likely the physician would be
influenced about his/her decision to provide or not provide
the specific treatment listed.
Each of the individual responses are assumed to be
related to the underlying constructs (in this case, the
provision decision, futility decision, discussion decision
and family/surrogate effect).
In summary, each survey yielded four scores-the
Provision Decision, the Futility Decision, the Discussion
Decision, and the Surrogate Effect. Each of these scores
48
measures a different dependent variable. The potential
range of the scores is as follows:
Highest Lowest
Vignette 1 Provision Decision 8 -8
Futility Decision 8 -8
DiscussionDecision 8 -8
Surrogate Effect 8 -8
Vignette 2 Provision Decision 8 -8
Futility Decision 8 -8
DiscussionDecision 8 -8
Surrogate Effect 8 -8
Vignette 3 Provision Decision 8 -8
Futility Decision 8 -8
DiscussionDecision 8 -8
Surrogate Effect 8 -8
Vignette 4 Provision Decision 8 -8
Futility Decision 8 -8
DiscussionDecision 8 -8
Surrogate Effect 8 -8
Likert scaling, like the one employed in this study, is
an attitude scale that poses statements and asks a
respondent to indicate how much he or she agrees or
disagrees with each statement. A middle answer of "I don't
know" is also usually included in this type of survey. The
scale score on a Likert questionnaire is determined by
summing all responses related to a particular item. The
scale score is then interpreted as representing the subjects
attitude on the particular construct under investigation.
According to Kidder and Judd (1994) , the basis for this
interpretation is that the probability of agreeing with
49
favorable items and disagreeing with unfavorable ones
increases directly with the degree of favorability of the
subject's attitude. "A subject who is highly favorable will
respond favorably to many items and be given a high score, a
subject who is indifferent will be given a middle score and
a subject who is unfavorable will respond unfavorably to
many items and given a low score" (p. 234).
Likert scales offer the researcher a number of
advantages. First, Likert scales are relatively easy to
construct. Second, they can be used to measure a number of
dependent variables in a single survey. Likert scaled
surveys are generally more internally valid than other
scaled surveys of similar length. Last, the range of
choices offered on Likert scales may give respondents more
comfort than scales using dichotomous (i.e. agree or
disagree) choices.
Data Analysis
The following 9 hypotheses will be tested:
1. Younger patients receive more aggressive medical
treatment from physicians than older patients.
2. Physicians are less likely to consider medical
treatment for younger patients futile than
they are for older patients.
3. Physicians are more likely to discuss treatment
decisions with younger patients/surrogates
than with older patients/surrogates.
4. There is a relationship between futility decision
and discussion decision.
5. Older physicians determine treatments for patients
are futile less often than do younger physicians.
6. Female physicians determine treatments for
patients are futile less often than do male
physicians.
7. Physicians practicing in teaching hospitals
judge that treatments for patients are futile less
often than do physicians practicing in non
teaching hospitals.
8. Physicians practicing 10 or more years judge that
treatments for patients are futile less often than
do physicians practicing less than 10 years.
9. More religious physicians judge that
treatments for patients are futile less often than
less religious physicians.
Correlation analysis, multiple regression analysis and
analysis of variance (ANOVA) are the most practical
statistical methods for analyzing these hypotheses for a
number of reasons. First, because the study utilized a
repeated measures design in which participants were subject
51
to a number of treatment conditions, comparison of multiple
means was required. ANOVA allows the researcher to test the
relationship between the dependent variable and the
circumstances under which it is measured when the same
dependent variable is measured repeatedly across
respondents.
The first ANOVA was used to test the significance of
the main effect of Patient Age on the physician's Provision
Decision (PD). The second ANOVA attempted to test the
significance of the main effect of Patient Age on the
physician's medical Futility Decision (FD). The third and
fourth ANOVAs attempted to test the significance of the main
effect of the Patient Age on the physician's Discussion
Decision (DD) and Patient Age and Family/Surrogate Effect
(SE) .
Other hypotheses concerning physician decision making
and medical futility require the use of correlation and
regression analysis. In this study about medical futility a
number of professional and personal physician
characteristics are included as independent variables.
Hypotheses concerning these variables are tested using
correlation analysis to determine if relationships exist
between these independent variables and the dependent
variables under consideration.
Correlation analysis, while useful in determining the
extent of relationship between two variables, does not
provide information about the nature of that relationship.
Regression analysis supplies variance measures which allow
the researcher predictive capability. Both simple and
multiple regression analysis will be used in this study.
"Simple regression analysis is used to predict an object's
value on a criterion variable, given its value on one
predictor variable. Multiple regression is used to predict
an object's value on a criterion variable when given its
value on each of several predictor variables" (Kachigian,
1992). The goal of this phase of the analysis is two fold:
1) determine if one or more independent variables has a
statistically significant effect on the dependent variables
2) determine the nature of the statistically significant
effect on the dependent variables.
53
CHAPTER IV
RESULTS
Description of the Sample
Data for analysis of the research questions regarding
physician decision making for disabled patients was obtained
from an original random sample of 400 Family Physicians
acquired from the American Academy of Family Physicians.
One hundred-seventy six subjects responded to the first and
second mailings of the questionnaire on medical treatment
for disabled patients with life threatening conditions.
Four subjects were dropped due to incomplete responses. Two
other subjects were dropped because the respondents
identified themselves as currently only practicing emergency
medicine. These six surveys were omitted, reducing the
total number of usable surveys to one-hundred seventy.
A description of the personal and professional
characteristics of the responding family physicians are
found in Tables 4 and 5.
A majority (70 percent) of family physicians responding
to the study were between the ages of 30 to 45 years old.
The mean was 43.8 years of age, the median 42 years of age
and the mode 41 years of age. This high percentage of
respondents between the ages of 30 to 45 is due in part to
the increased number of physicians trained in the late
1970's and early 1980's. Eighty-one percent of the
54
respondents were male. A majority (62 percent) of the
sample was raised either Baptist (13 percent), Catholic (31
percent) or Protestant (18 percent). The next largest
religious denominations were Methodist (10 percent) and
Jewish (6 percent). Buddhist, Quaker, Hindu, Lutheran and
Presbyterian accounted for a combined total of 17 percent of
the survey respondents.
Responses to the question about current religious
identification showed a decrease in religious affiliation
with a total of 47 percent currently identifying themselves
as either Baptist (12 percent), Catholic (19 percent) or
Protestant (16 percent). Buddhist, Quaker, Hindu, Lutheran
and Presbyterian remained relatively unchanged at 15
percent. The most significant change occurred in the None
category. Five percent responded None when asked for the
denomination raised while 26 percent responded None when
asked about their current denomination.
Fifty-eight percent of the responding physicians
identified themselves as being Very non-religious (30
percent) or Extremely non-religious (28 percent). Twenty
percent identified themselves as being either Extremely
religious (4 percent), Very religious (7 percent) or
Somewhat religious (9 percent). Twenty percent identified
themselves as being neither religious nor non-religious (6
percent) or Somewhat non-religious (14 percent).
55
Ninety-one percent of the physicians responded that
they were currently in full-time clinical practice. Forty-
eight percent of the respondents have been in practice
between 1 to 10 years. Thirty-four percent have been in
practice between 11 to 20 years. Eighteen percent have been
in practice for 21 or more years. The high percent of
survey respondents with less than twenty years of practice
is expected given the high percentage of physicians between
30 to 45 years of age.
Twenty-nine percent of the respondents are practicing
in a community based private hospital. Twenty-eight percent
are in a community based public hospital. Twenty six percent
are practicing in a either a community or university based
teaching hospital. Of the remaining 17 percent of
respondents who answered "other" to this question, a
majority (92 percent) were active duty officers practicing
in either a U.S. Army or U.S. Navy hospital.
Patient type is presented in quartiles. Table 4 shows
that when asked to estimate the percentage of patients the
physician sees who are members of an HMO, a majority (65
percent) estimate that between 1 percent to 25 percent of
their patients are HMO members. When asked to estimate the
percentage of patients who are members of a Fee-For-Service
type of insurance plan, a majority (80 percent) estimate
that between 1 to 50 percent of their patients participate
in a Fee-For-Service type of insurance plan. Finally, when
asked to estimate the percentage of patients who are
recipients of publicly funded healthcare, i.e. Medicare, a
majority (59 percent) estimate that between 1 to 50 percent
of their patients are recipients of publicly funded
healthcare.
Physicians were also asked to estimate the percentage
of patients they regularly see in a number of age
categories. Table 4 presents the results of this question
with responses aggregated by quartile. When asked to
estimate the percentage of patients between the ages of 0 to
18, 81 percent of the physicians surveyed estimate that
between 1 percent to 25 percent of their patients are in
this age category. The results also indicate that forty-
eight percent of the physicians surveyed estimate that
between 26 percent to 50 percent of their patients are
between the ages of 19 to 64 years old. For patients age 75
to 84, 94 percent of the physicians surveyed estimated that
only 1 to 25 percent of their patients fall into this age
category. Finally, the table also shows that 98 percent of
the physicians surveyed estimate that between 1 to 25
percent of their patients are age 85 and older.
57
Table 4
Personal Characteristics of the Sample of Family Physicians
Description
Age
30 - 35 years
36 - 40
41 - 45
Percentage n
18 30
24 41
28 47
46 - 50 12 21
51 - 55 5 9
56 - 60 2 4
60 + 11 18
Sex
Male 81 138
Female 19 32
Religious Denomination Raised
Baptist 13 22
Catholic 31 53
Jewish 6 10
Methodist 10 17
Protestant (not otherwise listed) 18 30
None 5 8
Other * 17 20
Religious Denomination Current
Baptist 12 20
Catholic 19 32
Jewish 6 10
Methodist 6 11
Protestant (not otherwise listed) 16 28
None 26 44
Other* 15 25
(*Buddhist, Quaker, Hindu, Lutheran, Presbyterian)
Role of Religion
Extremely religious 4 7
Very religious 7 10
Somewhat religious 9 15
Neither religious nor non-■relig. 6 11
Somewhat non-religious 14 24
Very non-religious 30 51
Extremely non-religious 28 48
Don't Know 2 4
58
Table 5
Professional Characteristics of the Sample of Family
Physicians
Description Percentage n
Nature of Professional role
Full-time clinical Practice 91 154
Part-time clinical Practice 9 16
Years in Practice
01 - 10 48 82
11 - 20 34 58
21-30 6 10
31 - 40 12 20
Hospital Type
University Based Teaching Hospital 9 16
Community Based Teaching Hospital 17 29
Community Based Public Hospital 28 47
Community Based Private Hospital 29 49
Other 17 29
Patient Type
Members of an HMO
01 - 25% 65 110
26 - 50% 27 45
51 - 75% 5 9
76 - 100% 5 6
Members of a Fee-For-Service Insurance Plan
01 - 25% 52 88
26 - 50% 38 64
51 - 75% 5 9
76 - 100% 5 9
Recipients of Publicly Funded Healthcare, i.e. Medicare
01 - 25% 21 35
26 - 50% 38 65
51 - 75% 15 26
76 - 100% 26 44
59
Table 5
Professional Characteristics of the Sample of Family
Physicians (continued)
Description
Patient Age Estimate
00 - 18 years
01 - 25%
26 - 50%
51 - 75%
76
-
100%
19 - 64
01 - 25%
26 - 50%
51 - 75%
76 - 100%
65 - 74
01 - 25%
26 - 50%
51 - 75%
76 — 100%
75 - 84
01 - 25%
26 - 50%
51 - 75%
76 - 100%
Age 85 +
01 - 25%
26 - 50%
51 - 75%
76
-
100%
Percentage n
81 138
19 32
0 0
0 0
24 41
48 82
16 27
12 20
78 133
20 34
1 2
1 2
94 160
6 10
0 0
0 0
98 167
2 3
0 0
0 0
Physician Decision Making for
Disabled Patients
Hypothesis 1
The first hypothesis stated that younger patients
receive more aggressive medical treatment from physicians
than older patients. A repeated measures ANOVA was used to
investigate this hypothesis. An analysis of the means and
standard deviations of physician treatment decisions in
Table 6 shows an inverse relationship between patient age
and treatment decision with the exception of Patient B. The
patient in this vignette is infected with HIV disease. An
analysis of the means in Table 6 shows that this patient
received the lowest treatment score of all the patients.
Patient A, a 6 year old child, received the highest
treatment score with a mean score of -1.211. It should also
be noted that Patient C, a 70 year old male in a Persistent
Vegetative State, received a lower treatment score than did
Patient D, an 86 year old man with respiratory
decompensation. The means in Table 6 also suggest that
treatment decisions for the two older patients cluster
together with physicians recommending a lower level of
treatment than for the youngest patient.
61
Physician Treatment
Table 6
Decisions
Patient Age Mean SD
Means and Standard Deviations
Patient A 6 -1.211 3.961
Patient B 30 -3.011 3.682
Patient C 70 -2.776 3.664
Patient D 86 -2.752 3.509
Source of
Variation
SS df MS
Between Groups 347.923 3 115.974 8.434*
Within Groups 9295.482 676 13.750
Total 9643.406 679
_____
An examination of the Table 6 ANOVA results indicates
that the treatment means are significantly different. In
summary, there is a significant difference in the
physicians' decision to provide treatment to the patients
described in the vignettes. While the ANOVA test determines
if treatment means differ significantly, it does not,
however, evaluate which means differ significantly. Several
statistical procedures have been developed to make all
possible comparisons between individual pairs of means. One
such technique is the Scheffe test. This is an appropriate
post-hoc test. Multiple Scheffe tests were conducted to
determine which treatment means differed significantly.
62
Table 7 shows all of the possible comparison combinations of
the treatment means and the results of the Scheffe analysis.
Table 7
Scheffe Analysis of Physician Treatment Decision
Mean 1 Mean 2 Plus Minus
A to B -1.211 -3.011 2.461 1.138
*
A to C -1.211 -2.776 2.226 0.903
*
A to D -1.211 -2.752 2.202 0.879
*
B to C -3.011 -2.776 0.426 -0.896
B to D -3.011 -2.752 0.402 -0.920
C to D -2.776 -2.752 0.638 -0.685
*Significant
As illustrated in Table 7, differences in the means of
physician treatment decisions differed significantly between
patients A and B, A and C and A and D. This means that
patient A, the youngest patient, would receive significantly
more aggressive treatment when compared to the other three
patients. These statistical findings allow us to accept the
first hypothesis.
Hypothesis 2
The second hypothesis stated that physicians are less
likely to consider medical treatment for younger patients
futile than they are for older patients. This was again
investigated with a repeated measures ANOVA. A lower
futility score is interpreted to mean that physicians
believe the treatment is more futile. A higher futility
score is interpreted to mean physicians believe the
treatment is less futile. An analysis of the means and
63
standard deviations of the physician's futility decisions in
Table 8 again show an inverse relationship between patient
age and the medical futility decision with the exception of
Patient B. Again, according to the vignette, Patient B is
infected with HIV disease and based on an analysis of the
means in Table 8, this patient receives the lowest medical
futility score of all the patients. Patient A, the 6 year
old child, receives the highest medical futility score with
a mean of -2.623. It should be noted that again, Patient
C, a 70 year old male in a PVS, receives a lower medical
futility score than does Patient D, an 8 6 year old man with
respiratory decompensation. The means in Table 8 also
suggest that medical futility decisions for the two older
patients cluster together with physicians more likely to
conclude that treatment is medically futile for these older
patients than for the youngest patient.
64
Table 8
Physician Medical Futility Decisions
Patient Age Mean SD
Means and Standard Deviations
Patient A 6 -2.623 3.743
Patient B 30 -4.123 3.232
Patient C 70 -3.823 3.407
Patient D 86 -3.788 3.515
Source of Var.
SS df MS F
Between Groups 223.133 3 74.377 6.142*
Within Groups 8185.394 676 12.108
Total 8408.528 679
*p<.05
An examination of the Table 8 ANOVA results indicates
that the treatment means are significantly different.
Multiple Scheffe tests were conducted to determine which
treatment means differed significantly. Table 9 shows all
of the possible comparison combinations of the treatment
means and the results of the Scheffe analysis.
Table 9
Scheffe Analysis of Physician Medical Futility Decision
Mean 1 Mean 2 Plus Minus
A to B -2.623 -4.123 2.120 0.879
*
A to C -2.623 -3.823 1.820 0.579
*
A to D -2.623 -3.788 1.785 0.543
*
B to C -4.123 -3.823 0.320 -0.920
B to D -4.123 -3.788 0.285 -0.956
C to D -3.823 -3.788 0.585 -0.656
*Significant
65
As illustrated in Table 9, differences among the means
of physician medical futility decisions differed
significantly between patients A and B, A and C and A and D.
This means that treatment for patient A, the youngest
patient, would have been judged the least futile when
compared to the other three patients. These statistical
findings allow us to accept hypothesis number 2.
Hypothesis 3
The third hypothesis stated that physicians are more
likely to discuss treatment decisions with younger
patients/surrogates than with older patients/surrogates.
This hypothesis was also investigated using a repeated
measures ANOVA. As illustrated in Table 10 the results of
the repeated measures ANOVA determined no significant
difference between a physician's discussion decision and
patient age. Therefore, this hypothesis is rejected.
66
Table 10
Physician Discussion Decision
Patient Age Mean SD
Means and Standard Deviations
Patient A 6 7.235 3.719
Patient B 30 6.9 7.262
Patient C 70 7.117 4.317
Patient D 86 7.052 4.369
Source of
Variation
SS df MS F
Between Groups 9.964 3 3.321 .0675
Within Groups 3324.059 676 4.917
Total 3334.024 679
Hypothesis 4
The fourth hypothesis stated that there is a
relationship between a physicians futility decision and
discussion decision. This hypothesis was investigated using
correlation analysis. As illustrated in Table 11 the
results of the correlation analysis determined no
significant difference between physician futility decision
and the discussion decision. These statistical findings
allow the rejection of hypothesis number 4.
67
Table 11
Pearson Correlation Coefficients
Treatment Decision Patient A 1
Discussion Decision Patient A .29339
Treatment Decision Patient B 1
Discussion Decision Patient B .14212
Treatment Decision Patient C 1
Discussion Decision Patient C .22629
Treatment Decision Patient D 1
Discussion Decision Patient D .25051
Hypotheses 5,6,7,8,9
Hypotheses 5, 6, 7, 8, and 9 each stated that there is
a relationship between a physician's medical futility
decision and physician age; physician gender; hospital type;
years of practice and physician religiosity. Correlation
analysis for each of the four patients determined that no
significant relationships exist between the independent and
dependent variables. Table 12 shows that only modest
correlations exist between medical futility decision and
hospital type (.17) and medical futility decision and
physician religiosity (.18) for patient A.
68
Table 12
Patient A Pearson correlation coefficients.
Futility A
Age
Futility A
Gender
Futility A
Yrs Practice
Futility A
Religiosity
Futility A Age
1
-0.078 1
Futility A Gender
1
0.055 1
Futility A
Hospital type 0.172
Futility A Hospital Type
1
1
Futility A Years Practice
1
-0.041 1
Futility A Religiosity
1
0.184 1
In an attempt to further analyze the extent and nature
of the relationships between the dependent variable medical
futility decision and the independent variables physician
age, physician gender, hospital type, years of practice and
physician religiosity multiple regression analysis was
performed. A total of 4 separate regression models were
used in this analysis— one for each patient. Of the 4
regression models run, only one model yielded significant
results.
Table 13 shows the results of the regression model for
Patient A which used Futility Decision as the dependent
variable and physician age, physician gender, hospital type,
69
years of practice and physician religiosity as the
independent variables. The table shows that only physician
religiosity is significant (t=2.93).
Table 13
Regression Results for Patient A Futility Decision
Regression
Statistics
Multiple R 0.260
R Square 0.067
Adjusted R 0.039
Square
Standard 3.469
Error
Observations 170
Analysis of
Variance
df SS MS
Regression 5 143.345 28.669
Residual 164 1974.654 12.040
Total 169 2118
Coefficients Standard t Statistic
Error
Intercept -4.760 2.792 -1.704
Physician -0.006 0.075 -0.082
Age
Physician 0.202 0.705 0.287
Gender
Hospital 0.369 0.224 1.644
type
Years of -0.028 0.071 -0.392
Practice
Role of Rel. 0.172 0.058 2.935*
*p<.05
Based on this analysis, it is concluded that physicians
with higher religiosity scores are more likely to conclude
that treatment for Patient A, a 6 year old, is beneficial
than are physicians with lower religiosity scores.
71
CHAPTER V
DISCUSSION
Summary of Results
The primary objective of this study was to answer
questions about the relationship between patient age and
physician decision making regarding futile medical
treatment. Individual physician characteristics were also
investigated to determine if they had any affect on
physician decisions. In order to meet the objectives of the
study, nine hypotheses were formulated and tested. Data for
analyzing the hypotheses was obtained from a nationwide
mailed survey. Four hundred family physicians were randomly
selected to participate in this study. One hundred seventy
usable responses were acquired from three separate mailings
to the family physicians. Names of physicians were drawn
from a membership list compiled by the American Academy of
Family Physicians.
The survey instrument included four descriptive
vignettes. Each vignette provided patient information which
included patient age, gender, medical condition, and also
identified the patient's legal decision maker. The patients
were described in the vignettes with the following serious
medical conditions: Patient A, a 6 year old male was in a
Persistent Vegetative State, the result of near drowning;
Patient B, a 30 year old male, had HIV disease and a t-cell
72
count of 55; Patient C, a 70 year old male, was also in a
Persistent Vegetative State, the result of a stroke; and,
Patient D, an 86 year old male suffered from dementia and
had interstitial lung disease as a complicating factor.
From the descriptions physicians should conclude that each
hypothetical patient is in extremely poor physical shape.
Moreover, each of the descriptions provided contain
information consistent with current medical practice.
In addition to patient background information,
respondents were provided with a list of medical
interventions potentially required by each patient. These
interventions included CPR, Artificial Nutrition/Hydration,
Kidney Dialysis and Major Surgery. For each vignette,
physicians were asked the following questions related to the
separate treatment options which were CPR, Artificial
Nutrition/Hydration, Kidney Dialysis and Major Surgery:
1) Would they provide the treatment;
2) Why they decided to provide or not provide the
treatment;
3) Would they discuss their decision with the
patient's surrogate/family;
4) If the surrogate/family demanded that "everything
medically possible be done for the patient", would
that have an effect on their treatment decision.
73
Responses to these questions yielded scores for four
dependent variables: provision decision, futility decision,
discussion decision, and family/surrogate effect. A total
score was calculated for each of the four treatment options
(CPR, Artificial Nutrition/Hydration, Kidney Dialysis and
Major Surgery) and for each question (would they provide the
treatment, why, etc.). A total survey score which added
individual vignette scores was calculated and provided four
separate repeated measures of each of the dependent
variables.
It was also hypothesized that a number of other
variables might influence physician decisions regarding
futile medical treatment. These variables included
physician age, physician gender, physician hospital
affiliation and physician religiosity. Each of these
variables were explored for potential affect on the
dependent variables.
Using repeated measures ANOVA, data derived from the
surveys was used to test the first three hypotheses. In
summary, these hypotheses were:
1. There is a relationship between a physician's
treatment decision and patient age.
2. There is a relationship between a physician's
medical futility decision and patient age.
74
3. There is a relationship between a physician's
discussion decision and patient age.
Statistical analysis supported hypotheses 1 and 2 at
the .05 significance level. Hypothesis number 3 was
rejected.
The fourth hypothesis stated:
4. There is a relationship between futility decision
and discussion decision.
Statistical analysis found no significant correlation
between these two variables.
Physician age, physician gender, hospital affiliation,
years of practice and physician religiosity were also used
as independent variables for hypotheses 5 through 9. These
hypotheses stated:
5. Older physicians determine treatments for patients
are futile less often than do younger physicians.
6. Female physicians determine treatments for
patients are futile less often than do male
physicians.
7. Physicians practicing in teaching hospitals
judge that treatments for patients are futile less
often than do physicians practicing in non
teaching hospitals.
75
8. Physicians practicing 10 or more years judge that
treatments for patients are futile less often than
do physicians practicing less than 10 years.
9. More religious physicians judge that
treatments for patients are futile less often than
less religious physicians.
Pearson correlation coefficients were calculated for
each of these hypotheses and revealed only modest
correlations for Patient A. In an attempt to further
analyze these hypotheses, regression analysis was utilized.
Regression analysis concluded that physician religiosity was
a significant independent variable in predicting "Futility
Decision" for Patient A. That is, physicians with higher
self-rated religiosity scores were more likely to view
treatment to Patient A as beneficial than physicians with
lower religiosity scores.
Discussion of Results
Data from this study support the hypothesis that
patient age significantly affects a physician's decision to
provide specific treatments to severely disabled patients
with life-threatening conditions. The youngest patient in
the study received the highest score for each of the
treatment provision questions. The two oldest patients
received consistently lower treatment recommendation scores
than patient A. Patient B, who suffered from HIV disease
76
received the lowest score of all the patients. This is most
likely due to the nature of the patient's disease and not
his age. A number of physicians noted that the risk of
infection when treating this patient, who had HIV disease,
was significantly higher than any of the other patients. It
can be concluded from this study that physicians, like many
in the general population, maintain stereotypical attitudes
about patient age and treatment outcome (Thorson, 1995).
The results of this study also support the hypothesis
that patient age significantly affects a physician's
futility decision. Despite the efforts of researchers who
promote an objective standard by which treatments can be
judged futile, physicians appear inclined to conclude that
treatments for older patients are more likely futile than
for younger patients. The notion that older patients are
less likely to benefit from a given treatment, that they
have lived long enough, or that resources are better spent
elsewhere, is probably quite common among family physicians
who treat large numbers of elderly patients but have little
training in geriatric medicine.
The study was unable to substantiate the idea that
physicians are less likely to discuss treatment decisions
with a patient's surrogate/family when they determine that
the treatment is futile. Capron argues that physicians use
the futility argument when they do not want to discuss
77
difficult treatment situations with a patient's
surrogate/family (Capron, 1994). Additional research is
needed to further explore this question.
The study also determined that physicians with higher
self-rated religiosity scores were more likely to view
treatment as beneficial for Patient A than were less
religious physicians. The following is representative of
responses provided by physicians with high self-rated
religiosity scores:
Male respondent, 68 years old: I believe
nutrition/hydration is a basic human right.
Suggestions for Future Research
Future research on the topic of futile medical
treatment should address a number of specific issues. Some
of the more interesting anecdotal responses provided by the
participants in this study focused on the issue of computer
assisted decision making. The following statements
represent a sample of respondent comments on this subject:
Male respondent, 51 years old: Many decisions
are difficult to make because of the legal
climate which exists. Each hospital should
have an independent means of making a Do Not
Resuscitate decision based on the patient's
current conditions and past medical history.
Perhaps, a nationwide computer program which would
make such decisions for us would be required in
order to decrease the legal liability.
78
Male respondent, 68 years old: The only
way futile treatments will be stopped is
if patients and/or surrogates receive detailed,
objective information on the outcome/efficacy
of a given treatment.
The idea that a large, computerized database which
would provide physicians and ultimately family members with
information related to treatment outcomes, is not
necessarily a novel idea. Since the early 1970's physicians
and researchers have been attempting to develop computer
programs designed to assist physicians and family members
with difficult decisions regarding potential treatments.
The Apache system is an example of this type of computer
program. The Apache system, which began in 1978 with a
grant from the Health Care Financing Administration, began
collecting data on patients in the intensive care units of
over 200 hospitals.
The Apache program is an attempt to predict the
survival odds of a number of medical procedures including
coronary bypass surgery, mechanical ventilation, artificial
nutrition and hydration based on 17 physiologic variables,
including blood pressure, respiratory rate, temperature,
pulse, white blood count, urine output, age and underlying
medical condition.
Current research on physicians who utilize computer
assisted decision making programs, reveals that many of
these physicians incorrectly identify age as being a
79
significant predictor of patient outcome (Oliver, 1995).
These findings lend support to this study on medical
futility.
Additional research is needed that explores both
physician willingness to relinquish decision making
authority for legal or financial reasons as well as patient
deference to such methods. Moreover, training and research
is needed that investigates physician use of age versus
functional status as a predictor of treatment outcome.
Future research can also investigate the association
between family variables and treatment decisions for
disabled patients with life-threatening conditions. Family
variables such as socioeconomic status and education might
be used in a study that utilizes vignettes. In light of
this study's findings that indicates the age of the patient
affects the futility decision, a researcher could hold the
age of the patient constant and vary the family resources to
determine the possible effects of family factors on futile
medical decision-making. The research question is: Do
family variables such as socioeconomic status and education
influence a physician's futility decision? A second
question might be: Do surrogate/parent's intelligence,
verbal abilities, and level of cooperation influence a
physician's futility decision?
80
A third area where additional research is likely to
yield interesting results is a study of physician
perceptions and opinions regarding the allocation of finite
medical resources. The following responses represent a
sample of physician statements regarding resource allocation
and the treatment of severely disabled patients:
Female respondent, 39 years old: We will
soon have limited medical access because of
limited medical funds-our funds nationwide
should be focused on patients who have a
chance to resume some "quality of life”.
Male respondent, 50 years old: My philosophy
over the years is more and more becoming that
patients with hopeless prognoses and futures
should be left in the hands of "nature" or "God”,
and our precious (and dwindling or at
least finite) health care resources be
restricted to those recipients most likely to
benefit from their application. America must
learn to bury its dead-not sustain them
artificially and indefinitely.
Male respondent, 56 years old: Old people
without quality of life should be allowed to die
of "natural" causes, with relative dignity
and without expenditure of large amounts of
scarce resources.
In the last several years a great deal of attention has
focused on rising health care costs in the U.S. and
considerable debate now concerns the issue of managed health
care. Managed care plans— run by doctors and hospitals,
insurers and managed care companies— are emerging all over
the country (Serafini, 1995). Research that probes both
81
physician attitudes regarding futility decisions and
perceptions about dwindling resources as well as how
futility decisions may differ by delivery method e.g.,
managed care vs. independently operating physicians, should
produce interesting and timely results.
Additional research should also address physician
concerns about infection risks to health professionals when
treating persons suffering from AIDS. It is interesting to
note that for Patient B, a 30 year old male with AIDS, a T-
cell count of 55 and additional complications, the
recommendation to not offer treatment is fairly widespread
among responding physicians.
Despite over a decade of professional education
concerning universal precautions as well as a policy
statement by the American Medical Association regarding the
duty of physicians to offer treatment to AIDS patients,
concerns over potential infection risk appears to be fairly
prevalent among physicians responding to the survey. The
following comments are a sample of statements concerning
infection risks posed by AIDS patients by two physicians who
were reluctant to treat Patient B:
Male respondent, 71 years old: The potential of
infectious experiences from the procedures
limits options. My emphasis would be for help in
the grief process— allowing death to occur
naturally.
82
Male respondent, 31 years old: Surgery and CPR
pose risks of infection to others. I may drag my
feet on these issues in terminal patient with
AIDS.
Additional research should also examine personal and
professional physician characteristics such as religiosity
and geographic region of practice to determine how these
variables may affect treatment decisions for AIDS patients.
A majority of physicians stated that it was important
to consult with parents or surrogates regarding treatment
decisions. In this regard, a number of physicians felt that
educating family members/surrogates regarding futile
treatments was important. The following response is an
example of this idea:
Male respondent, 48 years old: Although
I would take all family desires into
consideration on these judgment calls, I
strongly believe that educating the family
along the lines of "letting go" and "allowing
a loved one to die" would enhance the situation
and reduce the stress around the decisions for
those items inquired about.
Future research can further consider physician
communication with surrogates and family members. The type
of information included in these discussions, the duration,
the participants, the questions asked and issues addressed
are all important factors that require additional
investigation.
83
Conclusions and Policy Implications
This study has shown that a number of factors influence
physician decisions regarding medical treatment for disabled
patients with life threatening conditions. First, the data
indicate that patient age significantly affects a
physician's decision to actually provide medical treatment.
Second, patient age significantly affects a physician's
perception regarding the efficacy of a specific medical
treatment. Third, it was determined that the independent
variable physician religiosity significantly influences
physician futility decisions for the youngest patient.
The results of this research suggest that despite a
number of well publicized attempts by researchers and
practitioners to develop an objective, value neutral
definition of futile medical treatment as well as practice
guidelines concerning such treatment, physician decisions
about futile medical treatment are influenced by subjective
patient and physician factors.
It is important that medical schools, professional
medical associations, allied health practitioners and
gerontologists continue to educate the public about the most
important predictor of patient outcome: functional status
(Tilles, 1995). As the U.S. approaches the year 2020 when,
as it is projected, more than 50 million citizens will be
over the age of 65, it is crucial that health care
84
professionals understand the significance of individual
perceptions and opinions regarding the elderly especially as
they relate to sensitive medical decisions.
As U.S. policy makers continue to debate reform for
both the Medicare and Medicaid health care programs, it is
important that the major issues surrounding futile medical
treatment be addressed. Managed care programs designed
specifically for Medicare participants are expanding
throughout the U.S. but fundamental questions regarding the
role of "gatekeepers" and "utilization review committees",
particularly regarding their perceptions of futile medical
treatment, deserve continued attention. Aging advocates
should be concerned that in these new delivery systems the
elderly might be denied certain medical treatments as
physician gatekeepers, responding to business concerns i.e.,
cost savings, ration health care on the grounds that
treatment for some elderly patients is either qualitatively
or quantitatively futile.
Moreover, a recent study investigating patient-
physician communication revealed some fairly discouraging
information. The SUPPORT Study (1995), designed to
understand and improve end-of-life communication and
decision making and reduce the frequency of mechanically
supported, painful and prolonged dying, concluded that even
when an explicit effort is made to increase physician-
85
patient contact, nearly half of all Do-Not-Resuscitate
orders are written in the last 2 days of life. Also, family
members reported that half of the patients in the study who
were able to communicate in their last few days spent most
of the time in moderate or severe pain. Despite the best
efforts of the physicians and nurses in the study to
understand patient preferences regarding treatment options,
results indicate that simply enhancing opportunities for
more patient-physician communication may be inadequate to
change established practices.
Finally, it is important to note that none of the
hypothetical patients included in this research had a
durable power of attorney for health care. Many states
allow individuals to choose another person to make health
care decisions for them if they are unable to speak for
themselves. This person has the legal authority to make
decisions about what types of medical care the patient
should receive. A durable power of attorney allows an
individual to write down their health care wishes regarding
receiving or not receiving specific treatments (California
Medical Association, 1992). From a public policy
perspective, the durable power of attorney can be useful
when questions regarding futile medical treatment are
decided. If older patients clearly express their
preferences regarding a variety of treatment options through
86
a durable power of attorney, questions concerning futile
medical treatment may be minimized and also less perplexing.
87
SELECTED REFERENCES
Blackhall, L. J. (1987). Must we use CPR? New England
Journal of Medicine, 317/ 1281-84.
Bloom, A. (1968). The Republic of Plato. New York: Basic
Books.
Budetti, P. (1993). Achieving a Uniform Federal Primary
Care Policy: Opportunities Presented by National
Health Reform. Journal of the American Medical
Association. Vol 269 (4). pg. 498-501.
California Medical Association (1992). Your Health Care:
Who Will Decide When You Can't.
Callahan, D. (1987). Setting Limits: Medical goals in an
Aging Society. New York: Simon and Schuster Inc.
Publishers.
Capron, A. (1994). Medical Futility: Strike Two. Hastings
Center Report, 24J5), 42-43.
Capron, A. (1995). Baby Ryan and Virtual Futility. Hastings
Center Report, 2J5(2), 20-21.
Crane, D. (1975). Decisions to treat critically ill
patients: a comparison of social versus medical
considerations. Milbank Memorial fund Quarterly, 53,
1-33.
Creswell, J. (1994). Research Design: Qualitative and
Quantitative Approaches. Thousand Oaks: Sage.
Curtis, J. and Pearlman, R. et al. (1995). Use of the
medical futility rationale in do-not-attempt-
resuscitation orders. Journal of the American Medical
Association. 273;124-128.
Daniels, N. (1988). Am I My Parents' Keeper? An Essay on
Justice Between the Young and the Old. New York:
Oxford University Press.
Dillman, D. (1978). Mail and telephone surveys: The total
design method. New York: Hohn Wiley.
88
Faber-Langendoen, K. (1991). Resuscitation of patients with
metastatic cancer: is transient benefit still
futile? Archives of Internal Medicine, 151, 235-9,
Ferrini, A and Ferrini, R. (1993). Health in the Later
Years. Madison: Brown and Benchmark.
Gillick, M. R., et. al. (1993). Medical Technology at the
End of Life: What would Physicians and Nurses Want for
Themselves? Archives of Internal Medicine, 153(22),
2542-7.
Isaac, S and Michael, W. (1983) . Handbook in research and
evaluation. San Diego: EDITS Publishers.
Kachigan, S. (1992). Statistical analysis: An introduction
to univariate and multivariate statistics for the
social sciences. New York: Radius Press.
Kerlinger, F. (1977). Foundations of behavioral research
(3rd ed.). New York: Holt, Rinehart and Winston.
Kidder, C and Judd, L. (1994). Research Methods in Social
Relations. Fort Worth: Holt, Rinehart and Winston,
Inc.
Kingdon, J. (1995). Agendas, Alternatives, and Public
Policies. 2nd Edition. New York: HarperCollins.
Lamm, R. (1993). "Intergenerational Equity in an Age of
Limits: Confessions of a Prodigal Parent," in Gerald
Winslow and James Walters (eds.), Facing Limits: Ethics
and Health Care for the Elderly, Boulder, CO: Westview
Press.
Lantos, J. D., Singer P. A., Walker R.M., et al. (1989). The
illusion of futility in clinical practice. American
Journal of Medicine, 87, 81-84.
Leslie, L. (1972). Are high response rates essential to
valid surveys? Social Science Research,!, 323-334.
Levy, F. and Michel, C. (1986). "The Economic Future of the
Baby Boom" (Paper presented at the first annual
conference of Americans for Generational Equity,
Washington, D.C., 10 April 1986).
89
Lubitz, J. and Riley, G. (1993). Trends in Medicare
payments in the last year of life. New England Journal
of Medicine. 328:1092-6.
Oliver, S. (1995). What are my chances, doc? Forbes. July
31. pg. 136-137.
Otten, H. (1985). Professional judgments of neglect in child
self-care: A field experiment. Unpublished doctoral
disseration, University of North Carolina at
Greensboro.
"Overview of Entitlement Programs: 1992 Greenbook" (1992).
Committee on Ways and Means, U.S. House of
Representatives. Washington, D.C.: U.S. Government
Printing Office.
Population Projections of the United States, 1993, by the
U.S. Bureau of the Census, Washington, D.C.: U.S.
Government Printing Office.
Rosenblatt, R. (1991). "Bankruptcy of Part of Medicare
Feared." Los Angeles Times, May 18.
Rudestam, K and Newton, R. (1992). Surviving Your
Dissertation: A Comprehensive Guide to Content and
Process. Newbury Park: Sage Publications.
Schneiderman L. J., & Jecker N. S. (1993). Futility in
Practice. Archives of Internal Medicine, 153,437- 441.
Schneiderman, L. J., Jecker N. S., Jonsen A. R. (1990).
Medical futility: its meaning and ethical implications.
Annals of Internal Medicine, 112, 949-954.
Serrafini, M.W., (1995). Not Your Father's HMO. The
National Journal, Oct. 21.
Serrafini, M.W., (1996). The National Journal.
Support Principal Investigators. (1995). A Controlled Trial
to Improve Care for Seriously 111 Hospitalized
Patients. Journal of the American Medical Association.
274:20.
Thorson, J. (1995). Aging in a Changing Society. New York:
Wadsworth Publishing.
90
Tilles, J. (1995). Personal interview. Conducted April,
1995 at the University of California, Irvine-Medical
Center.
Title 42. United States Code of Federal Regulations.
57.1603.
Torres-Gil, F. (1992). The New Aging: Politics and Change
in America. New York: Auburn House.
APPENDIX A
QUESTIONNAIRE ON MEDICAL TREATMENT
FOR DISABLED PATIENTS WITH LIFE-THREATENING
CONDITIONS AND ACCOMPANYING LETTERS
92
Dear Family Practice Physician:
I am writing to request your participation in the data
collection for my research on Medical Treatment for Disabled
Patients. This research is being conducted on a nation-wide
basis and will be used to complete my doctoral dissertation
at the University of Southern California, Leonard Davis
School of Gerontology.
It is my hope that you will take twenty minutes from your
hectic schedule and complete this survey. The results
should provide interesting information about physician
decision making regarding medical treatment for disabled
patients. I have enclosed a postcard with your packet.
Please drop the postcard in the mail when you complete the
survey. This will be my way of knowing who did participate,
and I will not contact you further with reminders. Your
survey instrument remains confidential. Also, the postcard
will ask if you are interested in receiving the results of
the study.
It will be appreciated if you will complete the
questionnaire prior to April 29, 1995 and return it in the
stamped envelope enclosed. Other phases of this research
cannot be completed until the questionnaire data are
analyzed. Your comments concerning the issues addressed in
this research are welcomed. I am thanking you in advance
for your cooperation, support and prompt response.
Sincerely yours,
Carl Renold
93
Dear Family Practice Physician:
A few weeks ago I mailed you a questionnaire on physician
decision making regarding medical treatment for disabled
patients. I hope that you will participate with other
family physicians across the country and complete this
survey. The data collected with this instrument will be
used to complete my doctoral dissertation at the University
of Southern California, Leonard Davis School of Gerontology.
The survey will take approximately twenty minutes to
complete. The results should provide interesting
information about physician decision making for disabled
patients. In the event that your questionnaire has been
misplaced, a replacement is enclosed. I have enclosed a
postcard with the packet. Please drop the postcard in the
mail when you complete the survey. This will be my way of
knowing who did participate, and I will not contact you with
further reminders. Your survey instrument remains
confidential. Also, the postcard will ask if you are
interested in receiving the results of the study.
Please join with your colleagues and complete this
questionnaire by May 30, 1995. Other phases of this
research cannot be completed until the questionnaire data
are analyzed. Your comments concerning the issues addressed
in this data are welcomed. I am thanking you in advance for
your cooperation, support, and response.
Sincerely yours,
Carl Renold
Futile Medical Treatment: An Analysis of
Physician Decision Making
Phase II
1995 Survey
Lowell Carl Renold II, Researcher
Doctoral Candidate at The University
of Southern California
Leonard Davis School of Geronotology
Los Angeles, California 90027
95
PHYSICIAN SURVEY
The purpose of this survey is to examine physician decision making regarding
medical treatment for disabled patients. Below are four vignettes that describe seriously ill
patients with life threatening conditions. For each vignette, please read the patient
background information and presenting conditions before answering each question. While
it is recognized that more medical information would normally be available when
determining patient treatment options, please answer the questions based on the specific
detail presented. It is hoped that the information gathered from this survey will provide
useful information about physician decision making.
Patient "A" Patient A, a 6 year old male, is in a Persistent Vegetative State (PVS), the result of near
drowning. The patient has been in the Persistent Vegetative State for 4 months. Physicians believe the
patient has no realistic chance of regaining consciousness. The child's parents have not previously
expressed their views about the extent of medical treatment desired for the patient. Assuming no other
medical problems exist and LEAVING ASIDE ALL OTHER CONSIDERATIONS....
1) In order to prevent Cardiopulmonary Resuscitation from being administered to the patient if he
experienced cardiac arrest, would you write a Do Not Resuscitate (DNR) order?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Cardiopulmonary Resuscitation? Please complete
the following sentence: Because in this case I believe Cardiopulmonary Resuscitation is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to write or not write a DNR order)
with the patient's parents?
Definitely No Probably No I don't know Probably Yes Definitely Yes
2) If the patient could not swallow and was unable eat, would you provide the patient with Artificial
Nutrition/Hydration?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Artificial Nutrition/Hydration? Please complete
the following sentence: Because in this case I believe Artificial Nutrition/Hydration is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide
Artificial Nutrition/Hydration) with the patient's parents?
Definitely No Probably No I don't know Probably Yes Definitely Yes
96
3) If the patient experienced renal failure, would you provide the patient with Kidney Dialysis?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Kidney Dialysis? Please complete the following
sentence: Because in this case I believe Kidney Dialysis is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide Kidney
Dialysis) with the patient's parents?
Definitely No Probably No I don't know Probably Yes Definitely Yes
4) If the patient required Major Surgery such as removing the gall bladder or part of the intestines,
would you provide the patient with Major Surgery?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Major Surgery? Please complete the following
sentence: Because in this case I believe Major Surgery is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide Major
Surgery) with the patient's parents?
Definitely No Probably No I don't know Probably Yes Definitely Yes
5) If the patient's parents demanded that eveiything medically possible be done for the patient,
would this have influenced your decision to provide (or not provide) any of the following treatments?
Cardiopulmonary Resuscitation: Yes No_____
Artificial Nutrition/Hydration Yes No_____
Kidney Dialysis Yes No_____
Major Surgery Yes No_____
Comments:
97
Patient "B" Patient B, a 30 year old male with Acquired Immune Deficiency Syndrome (AIDS), has a
T-cell count of 35 and Kaposi's sarcoma plaques on both hands. The patient is currendy demented and
non-responsive. Physicians feel that the patient has no real chance of surviving more than 2-3 months.
The patient's partner has durable power of attorney for health care and is achng as the legal surrogate.
Neither the padent nor surrogate have previously expressed their views about the extent of medical
treatment desired for the padent. Assuming no other medical problems exist and LEAVING ASIDE ALL
OTHER CONSIDERATIONS....
1) In order to prevent Cardiopulmonary Resuscitation from being administered to the padent if he
experienced cardiac arrest, would you write a Do Not Resuscitate (DNR) order?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Cardiopulmonary Resuscitation? Please complete
the following sentence: Because in this case I believe Cardiopulmonary Resuscitation is...
Definitely Fudle Probably Fudle I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to write or not write a DNR order)
with the padent's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
2) If the patient could not swallow and was unable eat, would you provide the patient with Artificial
Nutrition/Hydration?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Artificial Nutrition/Hydration? Please complete
the following sentence: Because in this case I believe Artificial Nutrition/Hydration is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide
Artificial Nutrition/Hydration) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
3) If the patient experienced renal failure, would you provide the patient with Kidney Dialysis?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Kidney Dialysis? Please complete the following
sentence: Because in this case I believe Kidney Dialysis is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
98
Would you feel it necessary to discuss your treatment decision (to provide or not provide Kidney
Dialysis) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
4) If the patient required Major Surgery such as removing the gall bladder or part of the intestines,
would you provide the patient with Major Surgery?
Definitely No Probably No 1 don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Major Surgery? Please complete the following
sentence: Because in this case I believe Major Surgery is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide Major
Surgeiy) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
5) If the patient's surrogate demanded that everything medically possible be done for the patient,
would this have influenced your decision to provide (or not provide) any of the following treatments?
Cardiopulmonary Resuscitation: Yes No_____
Artificial Nutrition/Hydration Yes No_____
Kidney Dialysis Yes No_____
Major Surgery Yes No_____
Comments:
99
Patient "C" Patient C, a 70 year old married, male, has been in a Persistent Vegetative State (PVS),
the result of a stroke, for 4 months. Physicians believe the patient has no significant chance of regaining
consciousness. The patient's wife is acting as the surrogate decision maker. Neither the patient nor
surrogate have previously expressed their views about the extent of medical treatment desired for the
patient. Assuming no other medical problems exist and LEAVING ASIDE ALL OTHER
CONSIDERATIONS....
1) In order to prevent Cardiopulmonary Resuscitation from being administered to the patient if he
experienced cardiac arrest, would you write a Do Not Resuscitate (DNR) order?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Cardiopulmonary Resuscitation? Please complete
the following sentence: Because in this case I believe Cardiopulmonary Resuscitation is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to write or not write a DNR order)
with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
2) If the patient could not swallow and was unable eat, would you provide the patient with Artificial
Nutrition/Hydration?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Artificial Nutrition/Hydration? Please complete
the following sentence: Because in this case I believe Artificial Nutrition/Hydration is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide
Artificial Nutrition/Hydration) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
3) If the patient experienced renal failure, would you provide the patient with Kidney Dialysis?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Kidney Dialysis? Please complete the following
sentence: Because in this case I believe Kidney Dialysis is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
100
Would you feel it necessary to discuss your treatment decision (to provide or not provide Kidney
Dialysis) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
4) If the patient required Major Surgery such as removing the gall bladder or part of the intestines,
would you provide the patient with Major Surgery?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Major Surgery? Please complete the following
sentence: Because in this case I believe Major Surgery is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide Major
Surgery) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
5) If the patient's surrogate demanded that everything medically possible be done for the patient,
would this have influenced your decision to provide (or not provide) any of the following treatments?
Cardiopulmonary Resuscitation: Yes No_____
Artificial Nutrition/Hydration Yes No_____
Kidney Dialysis Yes No_____
Major Surgery Yes No_____
Comments:
101
Patient "D" Patient D, an 86 year old male, is elemented and nonresponsive to caregivers and family
members. The patient has a long histoiy of interstitial lung disease and is admitted to the hospital with
respiratory decompensation secondary to bacterial pneumonia. The patient's daughter has durable power
of attorney for health care and is acting as legal surrogate. Neither the patient nor surrogate have
previously expressed their views about the extent of medical treatment desired for the patient. Assuming
no other medical problems exist and LEAVING ASIDE ALL OTHER CONSIDERATIONS....
1) In order to prevent Cardiopulmonary Resuscitation from being administered to the patient if he
experienced cardiac arrest, would you write a Do Not Resuscitate (DNR) order?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Cardiopulmonary Resuscitation? Please complete
the following sentence: Because in this case I believe Cardiopulmonary Resuscitation is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to write or not write a DNR order)
with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
2) If the patient could not swallow and was unable eat, would you provide the patient with Artificial
Nutrition/Hydration?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Artificial Nutrition/Hydration? Please complete
the following sentence: Because in this case I believe Artificial Nutrition/Hydration is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide
Artificial Nutrition/Hydration) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
3) If the patient experienced renal failure, would you provide the patient with Kidney Dialysis?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Kidney Dialysis? Please complete the following
sentence: Because in this case I believe Kidney Dialysis is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide Kidney
Dialysis) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
102
4) If the patient required Major Surgery such as removing the gall bladder or part of the intestines,
would you provide the patient with Major Surgery?
Definitely No Probably No I don't know Probably Yes Definitely Yes
Why did you reach this conclusion regarding Major Surgery? Please complete the following
sentence: Because in this case I believe Major Surgery is...
Definitely Futile Probably Futile I don't know Probably Beneficial Definitely Beneficial
Would you feel it necessary to discuss your treatment decision (to provide or not provide Major
Surgery) with the patient's surrogate?
Definitely No Probably No I don't know Probably Yes Definitely Yes
5) If the patient's surrogate demanded that everything medically possible be done for the patient,
would this have influenced your decision to provide (or not provide) any of the following treatments?
Cardiopulmonary Resuscitation: Yes No_____
Artificial Nutrition/Hydration Yes No_____
Kidney Dialysis Yes No_____
Major Surgery Yes No_____
Comments:
103
1)
2)
3)
What is your age?
Sex? M
What type of hospital do you practice in?
University Based Teaching Hospital
_ Community Based Teaching Hospital
Community Based Public Hospital
Community Based Private Hospital
Other (Please Describe)
4) Please estimate the percentage of patients you see in the following categories:
Members of an HMO
Members of a Fee-For-Service type of Insurance Plan
Recipients of Publicly Funded Health care, i.e., Medicare, Medicaid, VA
5) What is the Nature of your professional role?
Full-time Clinical Practice
Part-time Clinical Practice
Not Performing Clinical Duties
6) How many years have you been in practice?
7) What rf"gion, if any,
were you raised in?
Baptist
Buddhist
Catholic
Quaker
Hindu
Jewish
Lutheran
Methodist
Muslim
Protestant (not otherwise listed)
8)
Presbyterian
None
Other (please specify)
What religion, if any,
do you currently practice?
Baptist
Buddhist
Catholic
Quaker
Hindu
Jewish
Lutheran
Methodist
Muslim
Protestant (not otherwise
listed)
Presbyterian
None
Other (please specify)____
104
(Continued)
9) Would you describe yourself a s (please circle only one)
Extremely religious
Very religious
Somewhat religious
Neither religious nor non-religious
Somewhat non-religious
Very non-religious
Extremely non-religious
Don't know
10) Please estimate the percentage of patients you regularly see in the following age
categories:
Ages 0-18 %
Ages 19-64 %
Ages 65 - 74 %
Ages 75-84 %
Ages 85 + %
Thank you for completing this survey. If you have any additional comments, please feel
free to use this space. Your remarks are important.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
An analysis of postacute treatment and outcome differences between Medicare fee-for-service and managed care
PDF
Evidence of age-based rationing of health care to the elderly: The effect of age and death on the use of health care services by elderly enrollees in an HMO
PDF
The self-justification and image management strategies of physicians in policy compliance treatment situations.
PDF
Statistical analysis of health care expenditure and health status in the well-elderly study
PDF
The use of feedback to modify medical house staff test ordering behavior.
PDF
Patient life in a nursing home: an experiential study
PDF
The transprofessional model of terminal care: Reforming end-stage care in HIV/AIDS.
PDF
Two essays on linear regression models: Solving shortest length least-squares problems and the determination of birth outcome
PDF
Procreative intentions in rural China: The relation with sex compositions of existing children, sex preferences, and social-economic status
PDF
The impact of state abortion policy on selective abortion ratios in the United States
PDF
Clinical capital equipment acquisition: A study of executive decision-making
PDF
Health social workers' perspectives on practice with people with disabilities in the era of the Americans with Disabilities Act: An exploratory-descriptive study
PDF
Elevated inflammation in late life: predictors and outcomes
PDF
Receipt of and perceived difficulty in accessing needed services among HIV-seropositive men and women
PDF
Business and medicine at the crossroads: An ethical assessment of for-profit managed care.
PDF
State legislation concerning individuals with dementia: An evaluation of three theoretical models of policy formation
PDF
The effect of rate regulation and price competition on labor efficiency in ancillary and nursing service areas of hospitals in California, 1983-1991
PDF
Hospital versus home delivery in Lesotho: A comparison of mountain and non-mountain locations
PDF
Physicians' use of compliance-gaining message strategies: Their effect upon patient medical communication satisfaction and patient adherence to treatment regimen instructions
PDF
Legislative intent and aging policy in Los Angeles County: the relative effects of age versus need in public conservatorship
Asset Metadata
Creator
Renold, Lowell Carl, Ii
(author)
Core Title
Futile medical treatment: an analysis of physician decision-making
Degree
Doctor of Philosophy
Degree Program
Gerontology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Gerontology,health sciences, health care management,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c17-99559
Unique identifier
UC11354962
Identifier
9636371.pdf (filename),usctheses-c17-99559 (legacy record id)
Legacy Identifier
9636371.pdf
Dmrecord
99559
Document Type
Dissertation
Rights
Renold, Lowell Carl, Ii,
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
health sciences, health care management