Medical futility: Predicting outcome of intensive care unit patients
by nurses and doctors—A prospective comparative study*
Sonia Frick, MD; Dominik E. Uehlinger, MD; Regula M. Zuercher Zenklusen, MD
Objective: First, to assess the pattern of the prediction of intensive care unit patients’ outcome with regard to survival and quality
of life by nurses and doctors and, second, to compare these predictions with the quality of life reported by the surviving patients.
Design: Prospective opinion survey of critical care providers;
comparison with follow-up for survival, functional status, and
quality of life.
Setting: Six-bed medical intensive care unit subunit of a 1,000bed tertiary care, university hospital.
Patients: All patients older than 18 yrs, admitted to the medical
intensive care unit for >24 hrs over a 1-yr period (December 1997
to November 1998).
Interventions: Daily judgment of eventual futility of medical
interventions by nurses and doctors with respect to survival and
future quality of life. Telephone interviews with discharged patients for quality of life and functional status 6 months after
intensive care unit admission.
Measurements and Main Results: Data regarding 521 patients
including 1,932 daily judgments by nurses and doctors were
analyzed. Disagreement on at least one of the daily judgments by
nurses and doctors was found in 21% of all patients and in 63%
of the dying patients. The disagreements more frequently con-
C
linical decision making is a
complex and dynamic process, accomplished by the experienced clinician as a subconscious and highly subjective gestalt
process (1). In intensive care unit (ICU)
practice, decisions concerning initiation,
continuation, extension, and termination
of diagnostic and therapeutic interventions are required daily. Decisions to
withhold or withdraw intensive life support to avoid futile medical treatment are
particularly delicate. Of course, the patient’s wishes— or, if the patient is no
*See also p. 646.
From the Medical ICU (SF), University Hospital,
Geneva, Switzerland; Division of Nephrology (DEU),
University Hospital of Bern, Switzerland; and Medicalsurgical ICU (RMZZ), Hôpital des Cadolles, Neuchâtel,
Switzerland.
Supported, in part, by grant 32-49585.96 from the
Swiss National Science Foundation (DEU).
Copyright © 2003 by Lippincott Williams & Wilkins
DOI: 10.1097/01.CCM.0000049945.69373.7C
456
cerned quality of life than survival. The higher the Simplified
Acute Physiology Score and the longer the intensive care unit
stay, the more divergent judgments were observed (p < .001). In
surviving and dying patients, nurses gave more pessimistic judgment and considered withdrawal more often than did doctors (p
< .001). Patients only rarely indicated bad quality of life (6%) and
severe physical disability (2%) 6 months after intensive care unit
admission. Compared with patients’ own assessment, neither
nurses nor doctors correctly predicted quality of life; false pessimistic and false optimistic appreciation was given.
Conclusions: Disagreement between nurses and doctors was
frequent with respect to their judgment of futility of medical
interventions. Disagreements most often concerned the most
severely ill patients. Nurses, being more pessimistic in general,
were more often correct than doctors in the judgment of dying
patients but proposed treatment withdrawal in some very sick
patients who survived. Future quality of life cannot reliably be
predicted either by doctors or by nurses. (Crit Care Med 2003;
31:456 –461)
KEY WORDS: medical futility; withdrawal; withholding; quality of
life; outcome; functional status; intensive care; doctors; nurses;
telephone interview
longer competent, his or her family’s or
surrogate’s wishes— have to be respected. In addition, clinical decision
making is based on medical evidence,
taking into account the probability of
survival (closely related to the premorbid
health status), type and severity of the
underlying disease, severity of the acute
insult, and intensity of the treatment (2–
5). Scoring systems, such as the Simplified Acute Physiology Score (SAPS) II,
Acute Physiology and Chronic Health
Evaluation III, or Mortality Probability
Models II, can provide important input
for human judgment, but their value is
very limited in the individual patient (6 –
11). The stage of an individual ICU patient, where the probability of survival
becomes dismal, is difficult to define
most of the time. The Study to Understand Prognoses and Preferences for Outcome and Risks of Treatment (SUPPORT)
showed that more accurate prognostic estimates did not influence treatment deci-
sion making as strongly as one would
expect (12–13). It has been postulated
recently that practice style may influence
the decision-making process much more
than precise estimation of survival (4,
14). Very little is known, however, about
the sensitivity and specificity of this approach (15), and unmeasured patient and
professional factors are very influential in
determining such decisions (5, 16 –17).
During the last few years, aspects of
quality of life (QOL) more and more entered into the decision-making armamentarium. However, QOL is a very personal perception, and prospective
estimation of future QOL is a major challenge (18 –20).
A large body of literature has been
published in the last few years about the
assessment of QOL (21–23), and several
questionnaires have been developed and
validated for intensive care patients (24 –
27). Most of these questionnaires evaluate QOL 3– 6 months after ICU discharge
Crit Care Med 2003 Vol. 31, No. 2
and are of no help for the daily prospective decision-making process during ICU
stay.
Different appreciation of survival
chances or future QOL of a patient can
lead to disagreements about management
strategies, eventually resulting in conflicts between physicians and nurses,
healthcare worker frustration culminating in burn-out syndrome, and impaired
patient care. Several studies have shown
various results concerning disparity of
nurses’ and physicians’ attitudes (28 –32).
In our personal experience, survival
chances and future QOL of ICU patients
are often perceived more negatively by
nurses than by physicians. We are not
aware of any study comparing prospective
estimation of future QOL assessed by
healthcare personnel during ICU stay
with actual QOL as assessed by former
patients some time after ICU discharge.
We therefore designed a prospective
study to investigate the pattern and reliability of judgment concerning the outcome of ICU patients by critical care
nurses and doctors, respectively, comparing their judgments to the patients’ individual appreciation of QOL and functional status 6 months after ICU
admission.
METHODS
Study Protocol. During 1 yr, from December 1997 to November 1998, all patients admitted for !24 hrs to our six-bed adult medical ICU were included in the study. The ICU
was one of two identical units representing the
medical ICU of an European University Hospital of medium size with 1,000 hospital beds,
!35,000 admissions per year, and 250 medical
beds. The patients were admitted via the emergency department or the wards or by transfers
from other hospitals. There was no specific
coronary care unit in the hospital.
The study design consisted of a prospective
daily prognostic judgment on the patient’s
eventual outcome by ICU nurses and doctors.
The following questions had to be answered by
both nurses and doctors: “First, do you consider treatment futile with regard to a) survival, and b) QOL, and second, do you think
that treatment should be withdrawn?” Possible answers were “yes,” “no,” or “questionable”
for each question. Medical futility was defined
as medical strategy that would not accomplish
the goals of survival on one hand and acceptable QOL on the other hand. QOL was defined
as the presumed satisfaction of an individual
patient with regard to his or her physical,
psychological, and social situation. The questions had to be answered daily for each patient.
A special questionnaire was added to the patient’s chart at the day of admission to the ICU
Crit Care Med 2003 Vol. 31, No. 2
and remained there throughout the ICU stay,
always visible and accessible to all ICU personnel. Questions were answered daily by nurses
and doctors, usually after the morning medical ward rounds. The assessments, once noted,
were not changed and were not subject to
discussion between the healthcare providers
in order to respect each answer as such.
Nurses’ judgments were made by the nurse
in charge of the patient during the day shift,
which involved 30 nurses overall. All nurses
were either ICU-certified personnel or were
undergoing their postgraduate education of 2
yrs’ duration. Moreover, they had a professional, nonintensive care experience of !5 yrs.
Physicians’ judgment was made by the staff
physician who directed the daily medical
rounds. To cover 365 days, ten staff physicians
with variable ICU experience were involved; in
about 60% of the days the judgment was made
by the head of the unit. Whenever feasible,
physicians’ discussions with the patients and
their families were performed in the presence
of the nurses. No withdrawal, withholding, or
major modification of treatment was performed on QOL arguments alone or without
consent of the patient or, if the patient was no
longer competent, his or her surrogate.
Outcome Measures. Survival was assessed
for all patients at ICU discharge, at hospital
discharge, and at 6 months. QOL and functional status as judged by the patients were
assessed by telephone interviews 6 months
after ICU admission. The telephone interviews
lasted !10 mins and were performed by one of
the authors who had not had any previous
contact with the patients. To assess functional
status, all patients were asked about their independence with regard to three distinct activities of daily living: personal body care,
household management, and outdoor mobility. Patients who were independent with respect to all three activities were considered
“completely independent,” those who were
limited with one or two of the three activities
were considered “partially dependent,” and
those who needed help with all three activities
were considered “fully dependent.” QOL was
assessed by asking the patients open questions
such as, “How are you? How are you pleased
with your present life considering all important aspects? How would you judge then the
overall quality of your present life?” Patients
were allowed to explain in their own words
how they felt. At the end of the interview, the
investigator summarized the patient’s statements and let the patient estimate his or her
QOL as good, fair, or bad. The technique of
assessment of QOL and health status by phone
interview as compared with assessment by another QOL measure, the Sickness Impact Profile (33), has recently been validated (unpublished data).
Data collected from the patient charts included patients’ gender, age, SAPS II, diagnosis on admission, length of ICU stay, hospital
stay, and mortality. They were analyzed first as
one patient population and post hoc stratified
into survivors and nonsurvivors. The nonsurvivors were further divided into patients who
had died in the ICU (ICU deaths), patients who
had died on the ward (in-hospital deaths), and
patients who had died after hospital discharge
(out-hospital deaths). Patients were further
grouped into those who received divergent
and those who received nondivergent judgments. The different subgroups were compared with regard to admission diagnosis, age,
SAPS II, and length of ICU and hospital stay.
The reliability of nurses’ and doctors’ judgment for all patients was evaluated by comparing their predictions with the actual outcome and patients‘ subjective assessment of
their well-being.
The study was approved by the Institutional Review Board and the Local Ethics
Committee.
Statistical Analysis. The software package
Systat was used for data analysis (Systat 8.0;
SSPS, Evanston, IL). Methods included nonparametric statistics (Mann-Whitney U statistics for two levels and Kruskal-Wallis analysis
of variance) and 2 " 2 and 2 " k cross-tables,
respectively, for the comparison of frequencies
between groups. Values are given as median
and ranges.
RESULTS
Overall Results. Initially 547 patients
were included; 26 were lost to follow-up,
and data of 521 patients could be analyzed. Overall data of these 521 patients
are given in Table 1. In 79% of the patients, nurses and doctors agreed in all
their 969 daily judgments, whereas in
21% of the patients, discordant predictions were given on at least one of the 963
daily judgments. There were significantly
more divergent opinions on patients who
eventually died than on survivors (p #
.001) and on in-hospital deaths compared
with out-hospital deaths (p # .01; Table
2). The patients judged differently had a
higher SAPS II score (p # .001) and a
longer ICU stay (p # .001) than those
judged identically. The age of the patients
did not influence the degree of concordance between nurses and doctors (Table
3). Admission diagnosis was a determinant of the frequency of divergent judgment: There were hardly any differences
in judgment by nurses and doctors about
patients with unstable angina or acute
myocardial infarction, whereas patients
hospitalized for cardiopulmonary arrest,
cardiogenic shock, sepsis, or septic shock
were prone to induce divergent judgment
(Table 4).
Judgment of Outcome of Dying Patients by Nurses and Doctors. The 45
patients who died in the ICU spent a total
of 284 days in the ICU and therefore un457
Table 1. Overall Data
All patients
Survivors
ICU deaths
In-hospital deaths
Out-hospital deaths
All deaths
n (%)
Age in Yrs,
Median
(Range)
SAPS II
Median
(Range)
ICU Days,
Median
(Range)
Hospital Days,
Median
(Range)
Daily
Judgments,
n
521 (100)
409 (78.5)
45 (8.6)
49 (9.4)
18 (3.5)
112 (21.5)
64 (18–95)
62 (18–91)
68 (19–90)
68 (24–95)
72 (28–86)
68 (19–95)
27 (6–127)
24 (6–89)
55 (15–127)
39 (12–105)
32 (20–60)
44 (12–27)
2 (1–42)
2 (1–42)
3 (1–36)
4 (1–25)
2 (1–31)
3 (1–36)
14 (1–215)
15 (1–215)
3 (1–37)
13 (3–155)
22 (4–151)
10 (1–155)
1,932
1,273
284
297
78
659
ICU, intensive care unit.
Table 2. Overview of Divergent Opinions in the Different Patient Groups
All patients
Survivors
ICU deaths
In-hospital deaths
Out-hospital deaths
All deaths
n
$%
n (%)
521
409
45
49
18
112
111 (21)
47 (12)a
30 (67)a
29 (59)b
5 (28)b
64 (57)a
$ %, patients for whom divergent judgment between nurses and doctors was found on at least 1
day.
a
p # .001 for the comparison of frequency of divergent judgment between survivors and dying
patients; b p # .001 for the comparison of frequency of divergent judgment between in-hospital deaths
and out-hospital deaths.
derwent 284 daily judgments. Divergent
opinions were noted in 30 of the 45 patients on at least 1 day with significantly
more pessimistic appreciation by nurses
than by doctors: Treatment was considered questionably or definitely futile with
respect to survival on 168 of the 284 days
by nurses and on 99 of 284 days by doctors and on 202 and 135 of the 284 days,
respectively, with respect to QOL (p #
.001 for survival and QOL). Withdrawal
was advocated affirmatively (“do you consider withdrawal?” answered by “yes”)
and questionably (“do you consider withdrawal?” answered by “maybe”) more of-
ten by nurses than by doctors (123 vs. 26
of the 284 days, p # .001). The affirmative readiness to withdraw treatment (answering the question, “do you think that
treatment should be withdrawn?” by
“yes”) when medical intervention was
judged to be futile with regard to survival
was significantly less present in nurses
(69%) than in doctors (100%; p # .001).
Some 297 daily judgments were available from the 49 patients who survived
the ICU but died in the hospital (Table 2).
Again, treatment was considered questionably or definitely futile with respect
to survival on more days by nurses than
by doctors (92 vs. 61 of 297 days) and on
119 vs. 70 of the 297 days, respectively,
with respect to QOL (p # .001 for survival and QOL). Withdrawal was advocated affirmatively (“do you consider
withdrawal?” answered by “yes”) and
questionably (“do you consider withdrawal?” answered by “maybe”) more often by
nurses than by doctors (61 vs. 26 of the
297 days; p # .001). For out-hospital
deaths, survival and QOL were only very
rarely called in question and withdrawal
was considered only exceptionally; but
again, the nurses’ judgment was more
pessimistic.
Survivors’ Own Assessment and Comparison With Nurses’ and Doctors’ Predictions. Six months after ICU admission,
94% of the former patients considered
their QOL as good (64%) or as fair (30%),
whereas full physical dependence was indicated in only 2.4% and partial physical
dependence in 19%. Compared with the
survivors’ statements, nurses’ (p # .001)
as well as doctors’ (p # .05) predictions
had been too pessimistic: of the 1,273
daily evaluations, nurses had considered
medical treatment to be questionably or
definitely futile with regard to QOL on
175 days (45 patients) compared with 105
days (26 patients) by doctors; with regard
to survival, nurses had considered medi-
Table 3. Characteristics of the Various Subgroups of Survivors and Non Survivors
Group
n
Age in Yrs,
Median (Range)
SAPS II,
Median (Range)
ICU Days,
Median (Range)
Hospital Days,
Median (Range)
Survivors, no %
Survivors, $ %
ICU deaths, no %
ICU deaths, $ %
In-hospital deaths, no %
In-hospital deaths, $ %
362
47
15
30
20
29
63 (18–91)
60 (21–86)
64 (30–81)
68 (19–90)
68 (28–77)
64 (24–95)
23 (6–73)a
40 (13–89)a
56 (24–92)
55 (15–127)
27 (12–94)a
49 (20–105)a
2 (1–26)a
7 (1–42)a
1 (1–4)a
5 (1–36)a
3 (1–5)a
6 (2–25)a
14 (1–182)a
31 (5–215)a
1 (1–13)a
5 (1–37)a
15 (4–155)a
13 (3–102)a
SAPS, Simplified Acute Physiology Score; ICU, intensive care unit; no %, no divergent judgment between nurses and doctors; $ %, divergent judgment
between nurses and doctors.
a
p # .001 for the comparison of SAPS II, ICU length of stay, and hospital length of stay between patients who got divergent ($ %) vs. those who did
not get divergent judgment (no %).
458
Crit Care Med 2003 Vol. 31, No. 2
Table 4. Divergent Judgment in Function of Admission Diagnosis
Diagnostic
Category
Cardiovascular
Pulmonary
Neurologic
GI/liver
Infectious
Various
Diagnostic Subgroup
Unstable angina/MI
Shock/CPR
Sepsis
Septic shock
n
% of Patients With
Divergent Opinion
Between Nurses and
Doctors
285
143
40
27
60
30
53
16
19
66
11
3
48
15
25
30
10
38
79
14
MI, myocardial infarction; CPR, cardiopulmonary resuscitation; GI, gastrointestinal.
cal treatment to be questionably or definitely futile on 63 days (24 patients) and
doctors on 24 days (7 patients). Only 15%
of survivors for whom nurses and 9% for
whom doctors had considered treatment
eventually futile with regard to future
QOL reported bad QOL 6 months later.
None of the survivors for whom withdrawal had been considered on one or
several days (25 days by nurses, 1 day by
doctors) reported bad QOL 6 months
later. Physical dependence was much
more frequent in patients for whom there
had been doubt for survival or QOL, more
so in the survivors in whom doctors had
had doubts than in the group where
nurses had had doubts (Figs. 1 and 2).
On the other hand, for the 24 patients
who qualified their QOL as “bad” 6
months after ICU, treatment had been
predicted to be futile with respect to QOL
in only 8% by the nurses and 4% by
doctors and as questionably futile in 25%
by the nurses and in 12.5% by doctors.
Among the survivors for whom neither nurses nor doctors had expressed
any doubt about survival or future QOL,
5% still considered their QOL as bad and
28% as fair, as did the surviving group as
a whole. The same was true for physical
handicap.
DISCUSSION
Nurses and doctors agreed in most
cases in their daily judgment about patient outcome. However, the sicker the
patients were and the longer they stayed
in the ICU, the more the judgments diverged. Differences concerned future
QOL more frequently than survival. Age
of the patients had no impact on divergent opinions. Nurses were more pessimistic than doctors with respect to all
Crit Care Med 2003 Vol. 31, No. 2
Figure 1. Answers of survivors 6 months after
intensive care unit discharge for whom treatment had been considered futile or questionable
in regard to survival by nurses or doctors. indep.,
independent; dep., dependent.
patients. In the very sick patients who
were going to die, nurses seemed to be
more realistic and were willing to withdraw treatment more often and earlier
than doctors. On the other hand, this
more pessimistic view led nurses occasionally to consider treatment withdrawal
in some very sick patients who finally
survived and ultimately described their
QOL as good as the rest of the survivors.
Neither nurses nor doctors could reliably
predict who would be satisfied with his or
her QOL 6 months after ICU admission.
The finding that outcome was appreciated differently in patients with severe
illnesses such as cardiogenic or septic
shock than in patients with more benign
disease like unstable angina seems to follow common sense. The longer a patient
stays in the ICU and the sicker he or she
is, the more questions may be raised.
However, it is definitely alarming that in
two thirds of patients ultimately dying in
the ICU or on the ward, there is some
kind of disagreement with respect to
treatment strategy. Because optimal patient care needs a cohesive team ap-
Figure 2. Answers of survivors 6 months after
intensive care unit discharge for whom treatment had been considered futile or questionable
in regard to quality of life by nurses or doctors.
indep., independent; dep., dependent.
proach, reasons for this gap have to be
sought and eliminated. The divergent
judgment of healthcare providers may reflect their difficulty of assessing and
weighing different elements, such as
prognostic factors or supposed future disability and QOL. Ideally, the ICU team
should recognize the stage where survival
becomes impossible and where the prospect of enjoying an acceptable QOL after
ICU stay becomes so poor that even if the
patient does survive, further treatment
must be considered inappropriate and/or
futile. There are two main obstacles with
respect to this ideal situation: the ambiguous definition of futility and the difficulty of adequate prospective estimation
of future QOL.
As to futility, according to the Consensus Statement of the Society of Critical
Care Medicine, treatment is defined as
futile when it does not accomplish its
intended goal, that is, beneficial physiologic effect (34). Even if this definition is
much debated, the concept of futility cannot be ignored in real clinical life. It is
implicit in many decisions to start or to
stop treatment and preoccupies most ICU
physicians almost every day (35). Several
recent North American and European
studies have demonstrated that 70 –90%
of patients dying in an ICU did so during
the withholding and withdrawing of life
support, futility being the main reason
for withdrawal (14, 36 – 40).
We consider the fact that the term
“medical futility” may not have been interpreted in the same way by all healthcare workers to be one of the main reasons of divergent judgment in our study.
According to our everyday experience, it
was probably taken in the broader sense
by the nurses, comprising not only treatment that offers no physiologic benefit
but also treatment in which the benefit
was controversial, whereas it was rather
459
taken in the narrow sense of “close to
zero probability of success” by the physicians. Furthermore, the threshold for
calling a situation futile may be lower for
healthcare personnel who do not have the
responsibility for the final decision of
withdrawal or withholding. This hypothesis is supported by the finding that
nurses did not propose withdrawal in patients for whom they thought that treatment might be futile in as straightforward a way as doctors did. On the other
hand, nurses spend much more time with
the patients than physicians and are
asked to journey with them through
emotionally very burdensome circumstances. They may be more aware of the
suffering of the patients, may therefore
believe less in patient recovery, and may
admit treatment failure more easily than
doctors. Furthermore, a bias is probably
given by the fact that the nurses represented a much more heterogeneous
group than the physicians.
The findings of this study may be
somewhat limited by its open design, that
is, the fact that all providers had knowledge of all other previous and current
assessments. The ICU where this study
was performed has a closed format with a
relatively small team. Physicians and
nurses are used to openly discussing all
treatment strategies and major decisions.
In other words, there has always been a
“public dissemination” of physicians’ and
nurses’ attitudes. We therefore consider
that the open design did not induce a
major bias, even if physicians were often
first to complete the questionnaires. We
cannot formally exclude the possibility
that the second person completing the
questionnaire either deliberately expressed an alternative viewpoint or
tended to give the same answer as the
first person; for the previously mentioned
reasons, however, we do not believe that
this was the case. The open assessment
itself facilitated, if anything, the open discussion, since it took place before completing the questionnaire and not vice
versa; a divergent opinion, once noted on
the answering sheet, was not a matter of
discussion thereafter. Furthermore, the
completion of the assessment forms very
quickly became a routine part of the daily
ward round.
The observed differences between
nurses and doctors found in our study
were those that remained despite discussions about the optimal care of the patient. It is impossible to collect truly independent opinions from team members
460
caring for the same patient and interacting continuously. Independent opinions
and decisions between members of different professional groups can be assessed
and compared by the presentation of fictive case-vignettes to each member. Results from such studies are independent,
but interpretation of these results are
limited by the hypothetical character of
the presented samples and by the nonrepresentative nature of the samples under
study (41). Therefore, we considered the
open handling of the assessments as the
most natural and most relevant way to
detect persistent differences between
nurses and doctors.
Most patients were satisfied with their
QOL and their functional status 6 months
after ICU discharge. Compared with the
patients’ own assessment, nurses as well
as doctors had been too pessimistic but in
some cases also too optimistic concerning future QOL. From those patients in
whom treatment had been considered potentially futile with respect to survival or
future QOL, only few complained about
bad QOL; they indicated, however, more
physical handicap than patients in whom
futility had never been considered.
Nurses and doctors had obviously
equated presumed future physical handicap with bad future QOL and had not
considered bad QOL in patients with no
foreseeable physical handicap. It is well
known, however, that QOL does not depend on physical well-being alone (18,
42). It includes different aspects of daily
life, such as physical, mental, and social
well-being, but also employment, economic status, spirituality, and more (43).
Perceptions of health and its meaning
vary between individuals and within an
individual over time (43– 46). Functional
impairment translating into QOL is experienced differently by young versus elderly patients, physical handicaps being
more easily accepted by elderly than by
younger patients (47). Such reasons
probably explain why patients’ subjective
judgment of QOL did often not correlate
with their physical handicap.
Despite these difficulties, QOL is considered an important determinant for
treatment decision. In a survey by the
Ethics Committee of the Society of Critical Care Medicine (48), four factors were
considered very important in making decisions to withdraw treatment: patients’
QOL, the likelihood of surviving hospitalization, chronic disorders, and reversibility of the acute illness. In another study,
faculty members placed their highest val-
U
tmost caution has
to
be
applied
when future qual-
ity of life as presumed by
nurses and doctors is used as
an argument for withholding
or
withdrawing
further
treatment.
ues on disease-based information
whereas students and house-staff preferred QOL factors (3). Our study, however, tells us that we must not use future
QOL—as presumed by nurses or doctors—as an argument for treatment withholding or withdrawal. Admitting that we
cannot reliably predict our patients’ future QOL, we should not include this
aspect in terminal decision making, except in situations where the patients
themselves describe their QOL as miserable or intolerable. By doing so, we could
considerably lessen the frequency of divergent judgment and thereby lower tension and frustration in the healthcare
team. The question of how many additional days of prolonged suffering we are
allowed to impose on very sick dying patients in order not to lose lives of very
sick but potentially surviving patients remains one of our major challenges.
CONCLUSION
In the present investigation, we assessed for the first time nurses’ and doctors’ prospective judgment of the futility
of medical treatment with respect to presumed survival and future QOL of ICU
patients. In addition, the prediction of
QOL by nurses and doctors was compared
with the QOL reported by the surviving
patients 6 months after ICU admission.
Nurses and doctors agreed in their appreciation of eventual futility of medical
interventions in the vast majority of patients but only in half of their daily judgments. The sicker the patients were and
the longer they stayed in the ICU, the
more the judgments diverged. Overall,
nurses were more pessimistic than doctors. Nurses’ and doctors’ appreciation of
their patients’ future QOL proved to be
Crit Care Med 2003 Vol. 31, No. 2
unreliable. Utmost caution has to be applied when future QOL as presumed by
nurses and doctors is used as an argument for withholding or withdrawing further treatment.
ACKNOWLEDGMENTS
We thank the staff of the Medical ICU of
the University Hospital of Bern, Switzerland, for their active participation in the
study. We also thank Martin Sauter for his
help in data collection and analysis.
REFERENCES
1. Kruse JA, Thill-Baharozian MC, Carlson RW:
Comparison of clinical assessment with
APACHE II for predicting mortality risk in
patients admitted to a medical intensive care
unit. JAMA 1988; 260:1739 –1742
2. Task Force on Guidelines SCCM: Recommendations for intensive care unit admission
and discharge criteria. Crit Care Med 1988;
16:807– 808
3. Caralis PV, Hammon JS: Attitudes of medical
students, house staff and faculty physicians
toward euthanasia and life-sustaining treatment. Crit Care Med 1992; 20:683– 690
4. Elstein AS, Christensen C, Cottrell JJ, et al:
Effects of prognosis, perceived benefit, and
decision style on decision making and critical care on decision making in critical care.
Crit Care Med 1999; 27:58 – 65
5. Cook DJ, Guyatt GH, Jaeschke R, et al: Determinants in Canadian health care workers
of the decision to withdraw life support from
the critically ill. Canadian Critical Care Trials
Group. JAMA 1995; 273:703–708
6. Le Gall JR, Lemeshow S: Development of the
SAPS II system from a European/North
American multicenter study. JAMA 1993;
270:2957–2963
7. Knaus WA, Wagner DP, Draper EA, et al: The
APACHE III prognostic system: Risk prediction
of hospital mortality for critically ill hospitalized patients. Chest 1991; 100:1619–1636
8. Knaus WA, Draper EA, Wagner DP, et al:
APACHE II: A severity of disease classification
system. Crit Care Med 1985; 13:818– 829
9. Lemeshow S, Teres D, Klar J, et al: Mortality
Probability Models (MPM II) based on international cohort of intensive care unit patients. JAMA 1993; 270:2478 –2486
10. Consensus conference ESICM and SRLF:
Predicting outcome in ICU patients. Intensive Care Med 1994; 20:390 –397
11. Lynn J, Harrell F, Cohn F, et al: Prognoses of
seriously ill hospitalized patients on the days
before death: Implications for patient care
and public policy. New Horiz 1997; 5:56 – 61
12. SUPPORT Principal Investigators: A controlled trial to improve care for seriously ill
hospitalized patients: The study to understand prognoses and preferences for outcome
and risks of treatment (SUPPORT). JAMA
1995; 274:1591–1598
Crit Care Med 2003 Vol. 31, No. 2
13. Knaus WA, Harrell FE Jr, Lynn J, et al: The
SUPPORT prognostic model. Objective estimates of survival for seriously ill hospitalized
adults. Study to understand prognoses and
preferences for outcomes and risks of treatments. Ann Intern Med 1995; 122:191–203
14. Prendergast T, Luce JM: A national survey of
withdrawal of life support from critically ill
patients. Am J Respir Crit Care Med 1996; 153:
A360
15. Atkinson S, Bihari D, Smithies M, et al: Identification of futility in intensive care. Lancet
1994; 344:1203–1206
16. Kollef MH: Private attending physician status
and withdrawal of life-sustaining interventions in a medical intensive care unit population. Crit Care Med 1996; 24:968 –975
17. Kollef MH, Ward S: The influence of access to a
private attending physician on the withdrawal
of life-sustaining therapies in the intensive care
unit. Crit Care Med 1999; 27:2125–2132
18. Carr AJ, Higginson IJ: Measuring quality of
life. Are quality of life measures patient centred? BMJ 2001; 322:1357–1360
19. Albrecht GL, Devlieger PJ: The disability paradox: High quality of life against all odds. Soc
Sci Med 1999; 48:977–988
20. Skevington S: Measuring quality of life in
Britain. Introducing the WHOQOL-100.
J Psychosom Res 1999; 47:449 – 459
21. Brooks R, Baumann A, Daffurn K, et al: Posthospital outcome following intensive care.
Clin Int Care 1995:127–135
22. Heyland DK, Guyatt G, Cook DJ, et al: Frequency and methodologic rigor of quality-oflife assessments in the critical care literature. Crit Care Med 1998; 26:591–598
23. Kerridge R, Brooks R, Hillman K: Quality of
life after intensive care. Yearbook of Intensive Care 1994:827– 838
24. Fernandez RR, Cruz JJ, Mata GV: Validation
of a quality of life questionnaire for critically
ill patients. Intensive Care Med 1996; 22:
1034 –1042
25. Gill TM, Feinstein AR: A critical appraisal of
the quality of quality-of-life measurements.
JAMA 1994; 272:619 – 626
26. Niskanen M: Quality of life after intensive
care. Schweiz Med Wochenschr 1998; 128:
1474 –1477
27. Short TG, Buckley TA, Rowbottom MY, et al:
Long-term outcome and functional health
status following intensive care in Hong
Kong. Crit Care Med 1999; 27:51–57
28. Gillick M, Hesse K, Mazzapica N: Medical
technology at the end of life—What would
physicians and nurses want for themselves?
Arch Int Med 1993; 153:2542–2547
29. Davies K, King D, Silas J: Professional attitudes
to cardiopulmonary resuscitation in departments of geriatric and general medicine. J R
Coll Physicians Lond 1993; 2:127–130
30. Abizanda R, Almendros C: Ethical aspects of
intensive medicine: Results of an opinion
survey. Med Clin (Barc) 1994; 102:521–526
31. Eliasson AH, Howard RS, Torrington KG, et
al: Do-not-resuscitate decisions in the medi-
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
cal ICU: Comparing physician and nurse
opinions. Chest 1997; 111:1106 –1111
Uhlmann RF, Pearlman RA, Cain KC: Understanding of elderly patients’ resuscitation
preferences by physicians and nurses. West
J Med 1989; 150:705–707
Bergner M, Bobbitt RA, Carter WB, et al: The
Sickness Impact Profile: Development and
final revision of a health status measure. Med
Care 1981; 19:787– 805
Ethics Committee of the Society of CCM. Consensus statement of the Society of Critical Care
Medicine’s Ethics Committee regarding futile
and other possibly inadvisable treatments. Crit
Care Med 1997; 25:887– 891
Pellegrino ED: Decisions to withdraw lifesustaining treatment. A moral algorithm.
JAMA 2000; 283:1065–1067
Ferrand E, Robert R, Ingrand P, et al: for the
French LATERA group. Withholding and
withdrawal of life support in intensive-care
units in France: A prospective survey. Lancet
2001; 357:9 –14
Manara AR, Pittman JAL, Braddon FEM: Reasons for withdrawing treatment in patients
receiving intensive care. Anaesthesia 1998;
53:523–528
Keenan SP, Busche KD, Chen LM, et al:
Withdrawal and withholding of life support
in the intensive care unit: A comparison of
teaching and community hospitals. Crit Care
Med 1998; 26:245–251
Prendergast TJ, Luce JM: Increasing incidence of withholding and withdrawal of life
support from the critically ill. Am J Respir
Crit Care Med 1997; 155:15–20
Vincent JL: Ethical issues in critical care medicine: United States and European views and differences. Intensive Care World 1996; 13:142–144
Richter J, Eisenmann MR: Attitudinal patterns
determining decision-making in the treatment
of the elderly: A comparison between physicians and nurses in Germany and Sweden. Intensive Care Med 2000; 26:1326–1333
Higginson IJ, Carr AJ: Measuring quality of
life. Using quality of life measures in the
clinical setting. BMJ 2001; 322:1297–1300
Carr AJ, Gibson B, Robinson PG: Measuring quality of life. Is quality of life determined by expectations or experience? BMJ 2001; 322:1357–1360
Allison PJ, Locker D, Feine JS: Quality of life:
A dynamic construct. Soc Sci Med 1997; 45:
221–230
Calman KC: The ethics of allocation of scarce
health care resources: A view from the center. J Med Ethics 1994; 20:71–74
Headey B, Holstrom E, Wearing A: The impact of
life events and changes in domain satisfaction of
well-being. Social Indic Res 1984; 15:203–227
MacEntee MI, Hole R, Stolar E: The significance of the mouth in old age. Soc Sci Med
1997; 45:1449 –1458
The Society of Critical Care Medicine Ethics
Committee: Attitudes of critical care medicine professionals concerning forgoing lifesustaining treatments. Crit Care Med 1992;
20:320 –326
461