Judith E. Nelson, MD, JD
Critical Care Workgroup Member
Mount Sinai School of Medicine
Curtis JR, Rubenfeld GD, eds. Managing death in the intensive care unit. Oxford:
Oxford University Press, 2001.
Death continues to be common in intensive care units, posing special clinical and
emotional challenges for professional caregivers. This landmark volume, the first
textbook devoted to care of patients dying in the ICU, is an indispensable resource
for clinicians and others. An impressive array of contributors are represented in
well-referenced chapters on diverse, important topics, which expert editors have
brought together in a comprehensive, coherent, organized, and accessible way. Together
with text that is uniformly clear, excellent tables and figures serve to summarize
the state of the art in this important area of practice and ongoing research.
Truog RD, Cist AFM, Brackett SE, et al. Recommendations for end-of-life care in
the intensive care unit: The Ethics Committee of the Society of Critical Care Medicine.
Crit Care Med 2001; 29:2332-2348.
As summarized in the abstracts of data-based articles set forth below, research
using quantitative and qualitative methods is solidifying the evidentiary foundation
for clinical care of patients dying with pulmonary disease and critical illness.
These recommendations are based on existing empirical data and consensus of expert
opinion. They are "written from the emerging perspective that palliative care and
intensive care are not mutually exclusive options but rather should be coexistent."
The recommendations are comprehensive, compassionate, and well-referenced to the
expanding literature in this field.
Claessens MT, Lynn J, Zhong Z et al. Dying with lung cancer or chronic obstructive
pulmonary disease: Insights from SUPPORT. J Am Geriatr Soc 2000; 48:S146-S153. Somogyi-Zalud
E, Zhong Z, Lynn J, Dawson NV, Hamel MB, Desbiens NA. Dying with acute respiratory
failure or multiple organ system failure with sepsis. J Am Geriatr Soc 2000; 48:S140-S145.
These are two reports of data prospectively collected during the Study to Understand
Prognoses and Preferences for Outcomes and Risks of Treatment (SUPPORT), a large
and important study addressing care provided to seriously ill, hospitalized adults
in nine disease categories at five academic medical centers. Overall, findings from
SUPPORT suggested that physician-patient/surrogate communication was limited and
that patient suffering at the end of life was significant. These reports focused
specifically on patients with Stage III or IV non-small cell lung cancer (n = 939),
acute exacerbation of severe chronic obstructive pulmonary disease (n = 1008), acute
respiratory failure (n = 1805), or multiple organ failure with sepsis (n = 1151).
Mortality during the index hospitalization was 12% for lung cancer patients, 11%
for those with COPD, and 44% for the ARF/MOSF cohort; 6-month mortality was 64%
and 32% for the cancer and COPD groups respectively, while 1-year mortality for
the ARF/MOSF group was 58%. As assessed during the second week of hospitalization
by interviews of patients or surrogates (with highest rates of surrogate substitution
for symptom reports among patients with ARF/MOSF), severe pain and dyspnea were
prevalent at the end of life in all cohorts. The majority of patients with lung
cancer or COPD expressed a preference for care focusing on comfort rather than life
prolongation. Among patients with ARF/MOSF, more than 80% who died during the hospitalization
rated their chance of survival as 75-90%, and over 50% expressed a preference to
die rather than survive with ventilator dependence. Nearly half of families of patients
with ARF/MOSF experienced adverse effects, with frequent reports of major financial
losses even though the vast majority of patients had medical insurance.
Gore JM, Brophy CJ, Greenstone MA. How well do we care for patients with end stage
chronic obstructive pulmonary disease (COPD)? A comparison of palliative care and
quality of life in COPD and lung cancer. Thorax 2000; 55:1000-1006.
This study explored differences in palliative care and quality of life between two
among the largest groups treated by pulmonary physicians, patients with chronic
obstructive pulmonary disease and those with lung cancer, using multiple methods
including administration of validated health-related quality-of-life assessment
tools, semi-structured qualitative interviews, and abstraction of medical records.
Study patients (n = 50 in each disease cohort) had advanced disease. Approximately
one-third of patients in each cohort reported their perception that diagnostic information
had been lacking or given insensitively, and approximately 80% thought that information
about prognosis or future management was inadequate. Whereas approximately one-third
of cancer patients received palliative care services and more were aware of their
availability, no COPD patient received or was offered comparable services, and assessment
of COPD patients for distressing symptoms was less frequent. Scores on anxiety and
depression scales and overall health-related quality of life were worse for patients
with COPD than those with lung cancer. This interesting study indicates that both
groups with advanced respiratory disease experienced distressing symptoms, deficient
communication, and impaired function, with compromise of quality of life, but that
patients with severe COPD fared worse, apparently because their needs were less
often recognized or met in the absence of expert palliative care.
Nelson JE, Meier DE, Oei EJ et al. Self-reported symptom experience of critically
ill cancer patients receiving intensive care. Crit Care Med 2001; 29:277-282.
Whereas previous symptom research relied on survivor recollections or surrogate
reports, which may be biased, this study systematically evaluated symptoms self-reported
in real time by critically ill cancer patients treated in a medical intensive care
unit. The study enrolled 100 consecutive patients, among whom 50 had capacity to
provide reports of a brief but diverse list of physical and psychological symptoms
modified from the Edmonton Symptom Assessment Scale; responses were given by all
patients able to respond, including 60% receiving mechanical ventilation and more
than one-third who died during the hospitalization. Between 55% and 75% of patients
reported pain, discomfort, anxiety, sleep disturbance, or unsatisfied hunger or
thirst they rated as moderate or severe. Depression and dyspnea at these levels
were reported by approximately 40% and 33% of responders, respectively. In addition,
patients reported significant pain, discomfort or both in association with common
ICU procedures, including mechanical ventilation, the presence of an endotracheal
tube or indwelling urethral catheter, endotracheal suctioning, and insertion of
central and peripheral venous catheters and nasogastric tubes. Among procedures
rated as most painful or uncomfortable were also noninvasive interventions, such
as turning and moving from bed to chair, a finding confirmed in the large-scale
Thunder Project II conducted by the American Association of Critical Care Nurses
(Puntillo KA et al., Am J Crit Care 2001;10:238-251). The study indicates that many
ICU patients, even those with grave illness, can self-report symptoms, and that
among patients with capacity to provide symptom reports, multiple distressing symptoms
are common at high levels of severity.
Pochard F, Azoulay E, Chevret S et al. Symptoms of anxiety and depression in family
members of intensive care unit patients: ethical hypothesis regarding decision-making
capacity. Crit Care Med 2001; 29:1893-1897.
This prospective, multicenter study established that anxiety and depression are
common among families of ICU patients. Over eight hundred family members of 637
patients in 43 French ICUs completed the self-administered Hospital Anxiety and
Depression Scale questionnaire. In multivariate analysis, several factors were found
to be independently associated with anxiety or depression, including acute disease
in the absence of chronic illness of the patient, absence of regular meetings between
medical and nursing staff to discuss patient and family needs, absence of a room
dedicated for family meetings, and contradictions in information given by clinicians
to families. The investigators suggested that anxiety and depression may impair
the ability of family members to make rational, appropriate, and ethically acceptable
end-of-life decisions for patients, who generally lack capacity to make their own
decisions during critical illness, but that factors related to caregivers and care
settings may be amenable to change that would reduce these distressing psychological
Azoulay E, Chevret S, Leleu G et al. Half the families of ICU patients experience
inadequate communication with physicians. Crit Care Med 2000; 8:3044-3049.
Inadequate communication with and comprehension by families of critically ill patients
are documented in this study conducted in the medical ICU of a French teaching hospital.
As revealed in interviews, more than 50% of relatives/other surrogates of 76 patients
(ICU mortality rate of 30%) failed to understand basic information about diagnosis
(the primary organ involved), prognosis (whether the illness was grave/expected
to result in death), or treatment of the patient (at least one from a list of major
treatments including mechanical ventilation, dialysis, and antibiotics). Various
factors were associated with poor comprehension, including caregivers' perceptions
that comprehension was poor, leaving open the issue of whether clinicians did not
or could not take steps to improve comprehension, or did so without success.
Teno JM, Fisher E, Hamel MB et al. Decision-making and outcomes of prolonged ICU
stays in seriously ill patients. J Am Geriatr Soc 2000; 48:S70-S74.
Further evidence of deficiencies in communication with critically ill patients and
their families is found in this report analyzing data from the SUPPORT study with
respect to patients who spent >14 days in an ICU. Even in this group with serious
illness and prolonged ICU stays, a substantial majority had not discussed their
prognosis or preferences for life-sustaining treatment with a physician. Almost
50% of patients who expressed a preference for palliative care (defined for this
purpose as "care that focuses on relieving pain and discomfort as much as possible,
even if that means not living as long") believed that the treatment they received
was contrary to this preferred approprach, and approximately 25% did not know what
approach to care was being taken by the medical team. In the absence of communication
with physicians, patients were more likely to receive care thought to conflict with
Cook DJ, Guyatt G, Rocker G et al. Cardiopulmonary resuscitation directives on admission
to intensive care unit: an international observational study. Lancet 2001; 358:1941-45.
In the absence of an explicit advance directive, the "default" implicit directive
is to perform cardiopulmonary resuscitation. This large-scale, observational study
conducted in 15 intensive care units in four countries demonstrated that explicit
advance directives continue to be uncommon during the first 24 hours in the ICU,
but that half of these directives expressed a preference against resuscitation.
The study enrolled almost 3000 patients consecutively admitted to closed ICUs in
university-affiliated teaching hospitals, among whom 44% were unable to participate
in decision-making during the first 24 hours in the ICU. Only 11% of patients across
the study sites had an explicit resuscitation directive during this period; almost
half of these directives were established by residents, who were more likely than
attending physicians to establish directives to resuscitate than not to resuscitate.
Among patients with explicit directives, the directive was do-not-resuscitate for
50%. Substantial variation was observed among countries, cities within countries,
and individual centers within cities. Increasing age was not associated with greater
likelihood of having an explicit directive, but did predict a directive against
resuscitation among those with directives. Explicit directives were more likely
for patients admitted to ICUs at night and on weekends. This rigorous observational
study suggests that implicit directives may not reflect true preferences of patients,
that earlier and more effective communication may help to promote concordance between
patients' preferences and their treatment, and that practices in end-of-life care
of critically ill patients remain variable and influenced by factors other than
the patient's clinical status.
Lilly CM, De Meo DL, Sonna LA et al. An intensive communication intervention for
the critically ill. Am J Med 2000; 109:469-475.
This was a before-and-after, interventional study in the medical ICU of a university
tertiary care hospital comparing "usual care" with a "proactive process of communication"
in which a formal, multidisciplinary meeting with the patient/family was held within
72 hours of ICU admission. Objectives of the meeting were to review treatment options,
discuss the patient's perspectives on end-of-life care, agree on a care plan, and
establish "clinical milestones" to establish the effectiveness of that plan. Follow-up
meetings were held for patients failing treatment to explore palliative care options.
In addition, a weekly case review was conducted with representation from multiple
disciplines, including critical care attending physicians. The study involved 134
and 396 consecutive patients in the pre-intervention and intervention groups respectively.
In the pre-intervention period, family meetings involving attending physicians were
held for less than half (43%) of patients. This proportion increased significantly
to 68% during the period of the intensive communication intervention. The intervention
was also found to reduce median length of ICU stay and "nonconsensus" with respect
to dispositional goals.
Perspectives on Care at the End-of-Life
Puntillo KA, Benner P, Drought T, Drew B, Stotts N, Stannard D, Rushton C, Scanlon
C, White C. End-of-life issues in intensive care units: A national random survey
of nurses' knowledge and beliefs. Am J Crit Care 2001; 10:216-229.
This study involved a mail survey of a stratified random sample of members of the
American Association of Critical-Care Nurses to investigate nurses' knowledge, beliefs,
and ethical concerns with respect to care of patient dying in intensive care units.
The questionnaire used hypothetical clinical scenarios to explore the areas of pain
management, withholding or withdrawing life support, assisted suicide, and euthanasia.
Among 3000 nurses in the sample, 906 (30%) returned completed questionnaires. Most
respondents thought that dying patients received inadequate analgesia. More than
80% disagreed with assisted suicide, and 95% opposed euthanasia, while almost all
approved of forgoing life-supporting treatment in appropriate circumstances. Approximately
one-third of the nurses indicated they had acted against their conscience in caring
for dying patients, apparently in situations where treatments they considered futile
were continued. Communication between physicians and nurses was generally thought
to be effective, but many respondents identified an unmet need for grief counseling
and staff debriefing activities. Because nurses provide most bedside care at the
end-of-life, it is especially important for physician and other members of the critical
care team to understand nursing perspectives, as revealed in studies such as this.
Steinhauser KE, Clipp EC, McNeilly M, Christakis NA, McIntyre LM, Tulsky JA. In
search of a good death: Observations of patients, families, and providers. Ann Intern
Med 2000; 132:825-832.
This is an important, qualitative study that used focus groups and in-depth interviews
of patients, families, and health and pastoral care professionals involved in end-of-life
care to identify attributes of a "good death." Participants discussed their experiences
with deaths of family members, friends, or patients, reflecting on what made those
deaths good or bad. Data analysis used a grounded theory approach, with open and
axial coding of transcripts of group meetings and individual interviews. Six themes
emerged as attributes of a good death in discussions involving a wide spectrum of
participants: symptom control, clear decision making, preparation for death, completion,
contributing to others, and affirmation of the whole person. Participants particularly
feared dying in pain, but stated that apprehensions could be reduced through communication
and clear, shared decision making, with the expectation of aggressive symptom management.
Both patients and families expressed a need for information about the expected course
of the illness, including both physical and psychosocial changes, allowing better
preparation for the dying process. Participants affirmed the importance of empathic
health care providers and of achieving a sense of completion, including resolving
conflicts and saying good-bye, and contributing to the well-being of others as possible
in the context of the illness. Professional roles influenced views more strongly
than gender or ethnic differences. Physicians' discussions differed from those of
other groups, reflecting greatest attention to "biomedical"/ physiologic aspects
of death, whereas other groups had a broader view, according equal importance to
psychosocial and spiritual issues. The investigators "heard many stories of health
care providers' discomfort with death and dying," thought to be attributable to
factors including feelings of failure, a desire for professional distance, and inexperience.
Rigorous qualitative research, as reflected in this study, is valuable for hypothesis
generation and detailed description of important phenomena, such as perceptions
of the quality of death.
Return to Top
Promoting Excellence in End-of-Life Care was a national program of the Robert
Wood Johnson Foundation dedicated to long-term changes in health care institutions
to substantially improve care for dying people and their families.