REVIEW URRENT C OPINION
Uncertainty in end-of-life care Saxon Ridley a and Malcolm Fisher b
Purpose of review Uncertainty surrounding medical decision-making is particularly important during end-of-life decisionmaking. Doubts about the patient’s best interests and prognostic accuracy may lead to conflict. Recent findings Many authors have suggested recently that medical attitudes to uncertainty need review. It is inappropriate to avoid discussion of uncertainty during end-of-life care and American literature suggests that patients and families accept uncertainty in end-of-life discussions. Recently, authors have advocated the concept of ‘Practical Certainty’ accepting that absolute certainty is rarely possible in end-of-life decision-making and openly acknowledging that the physicians are as certain as they can be in the circumstances. Allowing time to provide acceptance of a palliative care pathway and using the collective wisdom of colleagues improves the accuracy of prediction and reduces conflict at the end of life. Summary The implications of this review are that doctors should not avoid discussing uncertainty in end-of-life conversations and the article provides some recommendations for minimizing conflict arising from end-oflife discussion. Keywords critical care, end of life, uncertainty
‘Hope is a powerful ally in fighting serious disease and terminal illness, but when false hope is tied to expectations for miracles, then hope is converted to an expectation of a positive outcome when none is possible’ Arthur L. Caplan, PhD
INTRODUCTION Assessing the appropriateness of life-supporting treatment on the intensive care unit (ICU) is difficult. Much of this difficulty stems from uncertainty concerning prognosis and the patient’s wishes. The patient and their family will be anxious about an uncertain future, possible complications and incomplete recovery. Managing such uncertainty properly is an essential skill for those delivering critical care.
UNCERTAINTY IN MEDICINE Uncertainty is relevant to all scientific disciplines. In medicine, uncertainty surrounds complex judgements about diagnosis, treatment and prognosis; it has many causes. Cassell [1] believes that as www.co-criticalcare.com
clinicians gain experience, the more they are beset by uncertainty. Experience may not reduce uncertainty; over time, the physician will remember those patients who have defied the best prediction and unexpectedly survived or died. Cassell concludes that although uncertainty is unavoidable, it is best managed by a well informed doctor who knows as much as possible about the patient and their illness. Katz [2] discusses doctors’ tendency to brush aside medical uncertainty because of simple denial, traditional ideas about ethical conduct, or from beliefs about the proper discharge of professional responsibilities. The quest for absolute certainty may be impossible as the extensive and varied body of present knowledge may now actually lead a Glan Clwyd Hospital, Denbighshire, UK and bRoyal North Shore Hospital of Sydney, University of Sydney, St Leonards, New South Wales, Australia
Correspondence to Professor Malcolm Fisher, AO, MBChB, MD, FCCM, FRCA, Intensive Care Unit, Royal North Shore Hospital, St Leonards, NSW 2065, Australia. Tel: +61 02 9926 7111; fax: +61 02 9926 7779; e-mail: [email protected] Curr Opin Crit Care 2013, 19:642–647 DOI:10.1097/MCC.0000000000000030 Volume 19 Number 6 December 2013
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Uncertainty in end-of-life care Ridley and Fisher
KEY POINTS Uncertainty during end-of-life decision-making arises from doubts about prognostic accuracy and difficulty in establishing the patient’s wishes or best interests. Uncertainty may handicap delivery of care and communication with the family. Key steps in managing uncertainty involve physicians recognizing their own uncertainty and minimizing that of the family primarily by making sure communication is flawless. Uncertainty about prognosis is best dealt with by employing practical certainty wherein after consultation and consensus, the senior physicians openly acknowledge that they are as certain as they can be in the circumstances.
to confusion. Fox [3] described three forms of uncertainty that junior doctors face. The first is incomplete or imperfect mastery of available knowledge. The second depends on the limitations of current medical knowledge. The third is a combination of the two, the difficulty in distinguishing between limitations of personal ignorance or ineptitude and lack of medical knowledge. Beresford [4] categorized clinical uncertainty into conceptual uncertainty (i.e. the inability to apply abstract knowledge to concrete situations), technical or procedural uncertainty (i.e. the absence of scientific data or practical skill) and personal uncertainty (i.e. the lack of previous relationship with a patient and knowledge of their wishes). Thus, in practice prognostic uncertainty at the beginning of a career is related to self-confidence (whether the physician knows sufficient) or whether the information exists to accurately make the decision. Later, the ability to prognosticate is weakened by initial prognoses that were ultimately incorrect and contradictory or controversial medical research. Whatever the cause of uncertainty, a hallmark of clinicians is the ability to tolerate it. In clinical practice, uncertainty is often minimized by de-individualizing the patients to make them more like a textbook case, pretending that uncertainty does not exist, redefining the problem to eliminate uncertainty, shrinking the problem to smaller dimensions and finally hoping that uncertainty will resolve with time. Doctors need to both understand and manage uncertainty, and so be aware of the limitations uncertainty imposes on the reliability and validity of their clinical judgements [5,6]. In ICU, this is an essential prerequisite for quality critical care at the end of life.
APPROPRIATENESS OF CONTINUING TREATMENT Treatment is inappropriate when it is not in keeping with the patient’s wishes or will not produce an outcome acceptable to the patient. Unfortunately, accurately assessing the patient’s wishes or estimating the burden and outcome of treatment is fraught with uncertainty.
Establishing patients’ wishes Many ICU patients may be capable of understanding and retaining information so that they can express a clear choice. If ICU patients can communicate then they appear to make rational and sophisticated decisions despite being in such a hostile, stressful, and perhaps painful environment [7]. The commonest stressors described by ventilated patients (e.g. dyspnoea, pain, anxiety and fear) are important factors that influence cognition [8]. Sedative drugs will further impair thinking and communication. ICU staff have an obligation to respect patients’ wishes, but if these are unknown, the most appropriate treatment choice is uncertain. Advance Care Directives or other previously expressed views may help understand the incompetent patient’s wishes, but whether the views expressed in advance are relevant to the current situation may be uncertain. An advanced care plan relevant to the current situation, which has been discussed with a nominated surrogate in a structured session, is the most reliable record of the patient’s views [9]. Alas, few such care plans are available. Surrogate decision makers, either nominated as part of an advanced care plan or self-selected from family and friends, are normally called upon to assist with decision-making. Ideally, when trying to reflect the patient’s views, surrogates should apply a hierarchy of the patient’s expressed wishes, substituted judgment, and then patient’s best interest. If the patient’s views are unknown, then the surrogates should offer their substituted judgement of what they believe the patient would have wanted. The substituted judgement of families may introduce potential mismatch between the unknown wishes of the patient and the request to sustain life; disagreement within the family may lead to conflict. Surrogates frequently provide valuable insight into the patient’s wishes but they may occasionally have other agendas that may influence their perspective. Best interests encompass a more holistic approach than just best medical interests, but establishing best interest can be challenging. More social or emotional end points are encapsulated within best interests and could be regarded as equivalent (to medical best interests) in deciding how to treat patients. It is vital that ICU
1070-5295 ß 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-criticalcare.com
643
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Ethical, legal and organizational issues in the ICU
staff avoid substituting their own value judgements when making an assessment of best interests but fortunately physicians’ personal values appear not to be a major influence in prognostication [10]. Different religious and ethnic beliefs between the patient and the clinical staff may cause problems when treating minority groups. There may be apprehension about being culturally inappropriate, inadvertently causing offence or appearing discriminatory or racist. If this is not recognized and managed it may lead to uncertainty, disempowerment and inertia in practice [11]. Religious and community leaders may need to be involved if conflict arises.
dysfunction and use of inotropes or vasopressors. However, physicians are not always reliable or consistent even when presented with the same clinical information [16]. With hindsight, clinicians tend to remember data about the final outcome while de-emphasizing contradictory or ambiguous information. Knowing that a patient died or survived to discharge focuses attention on clinical features consistent with that outcome while other important features may be forgotten. Furthermore, a clinician’s perception of uncertainty is strongly influenced by judgements that later in the patient’s clinical course prove incorrect. Generally, ICU physicians tend to be pessimistic about outcome and have concerns about the inappropriate use of technology [17].
Prognostication in intensive care Prognostication in intensive care differs from other disciplines because of the patient’s severity of illness, dependence on life support and rapidly changing clinical situation. Even for senior doctors, prognostication in the critically ill is difficult [12–14], but surprisingly erroneous or biased prognostic estimates do not reduce intensive care treatment choices [10]. Scoring systems can improve survival prediction by quantifying and condensing influential variables into a single measure of illness severity. Scoring systems allow the patient to be placed in the high, medium or low mortality risk group. Unfortunately, they are less useful for predicting outcome for individuals. For an individual prognosis, the scoring system must generate a binary outcome (i.e. dead or alive) based upon the combination of a probability and a decision threshold. Conversion of a probability (ranging from 0 to 1) into a prediction (i.e. dead or alive) requires the decision threshold (e.g. 95% chance of correct prediction). Present systems do not achieve a sufficiently high sensitivity (i.e. correct predictions of the outcome event/total predictions) upon which to base individual treatment decisions. Furthermore, scoring systems are not comprehensive enough to evaluate the complex comorbidities of today’s ICU patient and may not capture the effect of influential higher order variables (such as the organization of ICU). Subsequent improvements in both delivery and application of critical care cannot be accommodated unless the scoring system is updated. As a general principle ICU physicians, especially when there is agreement between individuals, outperform scoring systems in outcome prediction. Rocker et al. [15] reported that physicians’ estimates of below 10% ICU survival were superior at predicting ICU mortality than illness severity, organ 644
www.co-criticalcare.com
Consensus and second opinion Minimizing uncertainty protects the patient’s interests from random and unjustifiable decisions. Clinicians need to collect and appraise evidence about prognosis, and then be honest with themselves and with patients and families about uncertainty. Uncertainty may be reduced by reviewing the literature on prognosis. The current clinical position can then be examined for additional clinical features that may alter the prognosis in one direction or the other. The patient’s response to treatment so far is indicative of overall clinical course. It is important to involve the referring team who will have greater understanding of the presenting pathology, its likely progression and local survival data. Previous local intensive care experience in similar circumstances should be considered. The accuracy of prognosis is improved when a number of physicians (blinded to each other’s prediction) agree on outcome [18]. Although final responsibility for withdrawing or withholding life support rests with the ICU medical staff, the views of other clinical staff are important. The nursing staff will have an established rapport with the patient’s family and so have a clearer idea of their perspective and wishes. Such consensus is reassuring to all who are involved in decision making.
MANAGING UNCERTAINTY WITH THE FAMILY CONFERENCE There are multiple sources of uncertainty for families of patients in intensive care. The fears of losing a loved one, concerns that their loved one is not getting the best possible treatment, or care being trammelled by complications and misinformation are all potent causes of anxiety and uncertainty. These concerns may be fuelled by the sensational Volume 19 Number 6 December 2013
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Uncertainty in end-of-life care Ridley and Fisher
anecdotes of poor treatment in the popular press, and particularly minor ethnic groups may fear that they are not getting a fair deal in their new society. The key to reducing such uncertainty is earning the trust of the family. The two major determinants of this are the intensive care environment and the ability and willingness of the staff to communicate. Families who become involved with intensive care, particularly for the first time, seek a cultural framework. If the environment in which they find themselves has a culture reflecting competence and good staff attitudes towards each other, ensuring the patient is treated respectfully, listening to their concerns and receiving regular explanations, then they and the carers will usually agree on what is best for the patient. It is important that the family see doctors, particularly senior doctors, regularly attending and reviewing the patient. In some units, families may be encouraged to stay at the bedside during ward rounds and procedures. It is important in the initial conversations to stress to the family that the current goal is to discharge the patient from hospital. However, if it becomes apparent over time that this is impossible, it is essential to discuss alternatives. Although families appreciate a single spokesman, rostering in intensive care may make this impossible and under these circumstances the specialist handing over care should personally handover the family and introduce the new specialist. Early open and honest discussion is important in relieving the family’s uncertainty as to the motives of the providers, and especially reassuring them that the decision is not being forced by costs or need for the bed. The basic tool for effectively communicating the appropriateness of continuing intensive care is the family conference in which shared decision-making is employed [19,20]. Such a conference helps to explore and manage the uncertainties faced by all parties. Formal end-of-life discussion should be carried out by experienced people who have been trained in running such discussions. Initially, the family’s understanding of the illness and what they have been previously told should be discussed, and then the prognosis and likely outcome explained. Subsequent to this, the values and wishes of the patient and family should be canvassed. It is important that the family appreciate that there will be time to consider a plan. The family should also be offered the opportunity to bring support persons to the next meeting. It is often helpful to explain to the family that if a palliation pathway is chosen the patient’s last days will be made as comfortable as possible, and that trying to make the patient better will lead to unnecessary suffering.
The clinical picture, its likely outcome and exploring options Determining the likely outcome is within the competence of medical staff as they are best placed to balance the burden of disease against chances of survival. The patient’s response to intensive care support over time provides supplementary information regarding outcome, and gives the family time to understand the problem and the clinician time to gain their trust. A multinational study demonstrated that decisions to withdraw and withhold treatment were rarely made until the fourth day of admission to ICU [21], which also correlated with surrogate acceptance of prognosis [22].
Awareness of uncertainty It is important to appreciate the patients’ and their families’ perspectives of uncertainty. It is absolutely vital that the family receive consistent information from all staff with whom they may engage. This predicates regular discussion of appropriateness and prognosis on rounds and ward meetings. The receipt of differing information, particularly wherein the family have been given false information or hope, is a major cause of uncertainty. This may be compounded by anger when there have been significant complications or errors in care prior to ICU admission. Most US patients want physicians to discuss uncertainty because surrogates believe uncertainty is unavoidable, and that physicians are the best source of accurate prognostic information [23]. Discussing uncertainty leaves room for hope, increases trust in the physician and allows time to make difficult life-support decisions. Only a minority felt that discussions about uncertainty should be avoided as uncertain prognoses may be unnecessarily upsetting or that they should be told when the prognosis was more certain. Physicians need to be sensitive and deliver a consistent message. Both Evans et al. [23] and Woolf [24] suggest the ability to discuss and recognize uncertainty is a hallmark of best practice and that failure to disclose uncertainty does not meet surrogates’ needs. However, 80% of Australian physicians believe discussion of prognosis should be avoided in the face of uncertainty [25].
Employing practical certainty Gillis and Tobin [26] have recommended a change in thinking regarding uncertainty in end-of-life care. They suggest the certainty owed by physicians to families is different from scientific certainty, and
1070-5295 ß 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-criticalcare.com
645
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Ethical, legal and organizational issues in the ICU
practical certainty is the goal. Scientific certainty is not possible and it is wrong for physicians to pretend to have ‘scientific’ certainty, as medicine is largely based upon considerations of probabilities. Practical certainty is being as certain as it is reasonable to be in the circumstances and the ICU is certainly an uncertain environment. When explaining prognosis, the response to the question ‘Is there absolutely no hope?’ should explain that, as there is uncertainty regarding prognosis, the opinions of a number of people have been canvassed and that we are as certain as we can be that there is no further hope. It is important to recognize that practical certainty is the honest option for end-of-life decisions. Practical certainty applies to both the physician and family. The best estimate of prognosis relies heavily on time and the collective wisdom of experienced physicians whereas the best estimate of the patient’s wishes comes from the surrogates’ perspective. The family conference reassures both parties that decisions are based upon best knowledge of the patient’s views and the best estimates of likely outcomes.
Avoiding conflict Unfortunately, conflict between the family and healthcare staff occasionally arises; these tend to be worse case scenarios representing failure of communication and understanding by both parties. Conflict rarely arises spontaneously and usually results from a sequence of unfortunate circumstances or actions, but may relate to unrealistic expectations. Effective communication to improve family understanding needs consistent information delivery probably via a main person of contact, and earning trust and listening. The key constructs for discussion with the family [27] are as follows: (1) Recognize the importance of uncertainty. (2) Use appropriate language of the family conference – ‘withdraw care’ versus ‘nothing more we can do’. (3) Avoid discussion of odds. Families latch on to and concentrate on the given figure. When the risk of death is high and a treatment option is low risk, it may be perceived as acceptable in spite of potential suffering. (4) Provide familiar faces at family conferences and consistent information delivery. If rostering means the discussion leader is to disappear, the family should be formally handed over to the new discussion leader and other participants such as nurses encouraged to continue attending. 646
www.co-criticalcare.com
(5) Earn trust by the family seeing the way the staff interact with each other, and how they speak to and care for the patients. (6) Explain early that the goal of treatment is to return the patient to health and if this is not possible alternatives need discussion. (7) Give families written material regarding endof-life care and ICU management. (8) Present prognosis in terms of ‘Practical Certainty’ – ‘we are as certain as we can be’. (9) Encourage second opinions. (10) Involve known supporters, for example, church/family doctors. These are a major determinant of families deriving benefit from end-of-life discussions [28]. Furthermore, it is important to simultaneously avoid conflict by dealing with issues such as loss of trust (e.g. because of poor previous communication, failure to understand the patient’s perceived views, beliefs or wishes or unrealistic expectations), concern about care (e.g. symptom relief being ignored once a ‘Do Not Resuscitate’ order is agreed, lower nursing care once discharged from ICU), grief and guilt. In both the United Kingdom and Australia, neither the patient nor family can demand treatment that the doctor considers is contrary to the patient’s clinical needs. Several US states have embraced the principles of the Uniform Health-Care Decisions Act which allows physicians to refuse treatment that they deem medically ineffective, is against their conscience, or is contrary to generally accepted medical standards [27]. However, ‘clinical needs’ and ‘medical ineffectiveness’ may be more nebulous at the end of life and unilateral decisions may precipitate further conflict and entrenching of views.
CONCLUSION Families respect wisdom and leadership from those who have earned their trust through providing good treatment and communication early in the course of the illness. Structured negotiation ensures that all the relevant issues are discussed and will facilitate better understanding of all parties’ perspectives. However, end-of-life decision-making needs leadership to steer the discussions. Prior to implementing plans to introduce palliative care, withdraw or withhold artificial forms of life support, there should be consensus among the healthcare workers caring for the patient. Such consensus protects against incorrect decision-making. Practical certainty, wherein the participants are as certain as they can be, with both prognostication and knowing of patient Volume 19 Number 6 December 2013
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Uncertainty in end-of-life care Ridley and Fisher
wishes, may be an appropriate concept for physicians engaged in end-of-life decisions. The collective wisdom of experienced healthcare workers, including patient advocates, increases the likelihood of the most appropriate decisions being made. Recognizing uncertainty improves the families’ confidence in the clinical staff. However, the major challenge in providing better end-of-life care is changing the culture of individual establishments and clinicians. Results from trying to enact such changes have been disappointing [29]. Acknowledgements None. Conflicts of interest The authors have no conflicts of interest to declare.
REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Cassell EJ. The sorceror’s broom: medicine’s rampant technology. Hastings Cent Rep 1993; 23:32–39. 2. Katz J. Why doctors don’t disclose uncertainty. Hastings Cent Rep 1984; 145:35–44. 3. Fox RC. Training for uncertainty. In: Robert K, Merton ???, editors. The student physician: introductory studies in the sociology of medical education, 1st ed. Cambridge, Massachusetts: Harvard University Press; 1957. pp. 207–241. 4. Beresford EB. Uncertainty and the shaping of medical decisions. Hastings Cent Rep 1991; 21:6–11. 5. Zettel Watson L, Ditto P, Danks JH, Smucker WD. Accuracy and perceived gender differences in the accuracy of surrogate decisions about life-sustaining medical treatment among older spouses. Death Stud 2008; 32:273–290. 6. Christensen C, Cottrell JJ, Murakami J, et al. Forecasting survival in the medical intensive care unit: a comparison of clinical prognoses with formal estimates. Methods Inf Med 1993; 32:302–308. 7. Cassell EJ, Leon AC, Kaufman SG. Preliminary evidence of impaired thinking in sick patients. Ann Intern Med 2001; 134:1120–1123. 8. Thomas LA. Clinical management of stressors perceived by patients on mechanical ventilation. AACN Clin Issues 2003; 14:73–81. 9. Detering KM, Hancock AD, Reade MC, Silvester W. The impact of advance care plans on end of life care in elderly patients: randomised controlled trial. BMJ 2010; 340:1345–1354. Important study suggesting that end-of-life planning with a trained facilitator improves quality for patient, family and staff. 10. Edtein AS, Christensen C, Cottrel Coulson JJ, Ng M. Effective prognosis, perceived benefit, and decision style on decision-making in critical care. Crit Care Med 1999; 27:58–65.
11. Kai J, Beavan J, Faull C, et al. Professional uncertainty and disempowerment responding to ethnic diversity in healthcare: a qualitative study. PLoS Med 2007; 4:e323. 12. Meadow W, Pohlman A, Frain L, et al. Power and limitations of daily prognostications of death in the medical intensive care unit. Crit Care Med 2011; 39:474–479. The study shows consensus and time improve prognostic ability of staff but do not give absolute accuracy. 13. Christakis NA, Lamont E. Extent and determinants of error in physicians’ prognoses in terminally ill patients: prospective cohort study. West J Med 2000; 320:469–473. 14. Poses RM, Chaput De Saintonge M, McCLish DK, et al. An international comparison of physician’s judgements of outcome rates of cardiac procedures and attitudes towards risk, uncertainty, justifiability, and regret. Med Decis Making 1998; 18:131–140. 15. Rocker G, Cook D, Sjokvist P, Level of Care Study Investigators, Canadian Critical Care Trials Group. Clinician predictions of intensive care unit mortality. Crit Care Med 2004; 32:1149–1154. 16. Kostopoulou O, Wildman M. Sources of variability in uncertain medical decisions in the ICU: a process tracing study. Qual Saf Healthcare 2004; 13:272–280. 17. Wildman MJ, Sanderson C, Groves J, et al. Implications of prognostic pessimism in patients with chronic obstructive pulmonary disease (COPD) or asthma admitted to intensive care in the UK within the COPD and asthma outcome study (CAOS): multicentre observational cohort study. BMJ 2007; 335:1132–1134. 18. Meadow W, Pohlman A, Frain L, et al. Power and limitations of daily prognostications of death in the medical intensive care unit. Crit Care Med 2011; 39:474–479. 19. Heyland DK, Cook DJ, Rocker GM, et al. Decision-making in the ICU: perspectives of the substitute decision-maker. Intensive Care Med 2003; 29:75–82. 20. White DB, Braddock CH 3rd, Bereknyei S, Curtis JR. Toward shared decision making at the end of life in intensive care units: opportunities for improvement. Arch Intern Med 2007; 167:461–467. 21. Cook DJ, Guyatt G, Rocker G, et al. Cardiopulmonary resuscitation directives on admission to intensive care unit: an international observational study. Lancet 2001; 358:1941–1945. 22. Prendergast TJ. Resolving conflicts around the end of life care. New Horiz 1997; 5:62–71. 23. Evans LR, Boyd EA, Malvar G, et al. Surrogate decision-makers’ perspectives on discussing prognosis in the face of uncertainty. Am J Respir Crit Care Med 2009; 179:48–53. The study shows US families comfortable with discussing uncertainty. 24. Woolf SH. Do clinical practice guidelines define good medical care? The need for good science and the disclosure of uncertainty when defining ‘best practices’. Chest 1998; 113:166S–171S. 25. Brieva JL, Cooray P, Rowley M. Withdrawing and withholding of life-sustaining therapy used in intensive care: an Australian experience. Clinical Care Resusc 2009; 11:266–268. 26. Gillis J, Tobin B. How certain are you doctor? Pediatr Crit Care Med 2011; 12:71–72. The article describing the historical and ethical basis of ‘practical certainty’. 27. National Conference of Commissioners on Uniform State Laws. Uniform Health-Care Decisions Act. 22 Issues L. & Med 2006; 83:83– 97. 28. McDonagh JR, Elliot TB, Engleburg RA, et al. Family satisfaction with family conferences about end of life care in the intensive care unit: increased proportion of family speech is associated with increased satisfaction. Crit Care Med 2004; 32:1484–1488. 29. Curtis JR, Nielsen EL, Treece PD, et al. Effective equality-improvement intervention on end of life care in the intensive care unit. A randomised trial. Am J Respir Crit Care Med 2011; 183:138–155. The study shows a comprehensive intervention to improve end-of-life care in a US hospital did not lead to improvement.
1070-5295 ß 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-criticalcare.com
647
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Uncertainty in end-of-life care Saxon Ridley a and Malcolm Fisher b
Purpose of review Uncertainty surrounding medical decision-making is particularly important during end-of-life decisionmaking. Doubts about the patient’s best interests and prognostic accuracy may lead to conflict. Recent findings Many authors have suggested recently that medical attitudes to uncertainty need review. It is inappropriate to avoid discussion of uncertainty during end-of-life care and American literature suggests that patients and families accept uncertainty in end-of-life discussions. Recently, authors have advocated the concept of ‘Practical Certainty’ accepting that absolute certainty is rarely possible in end-of-life decision-making and openly acknowledging that the physicians are as certain as they can be in the circumstances. Allowing time to provide acceptance of a palliative care pathway and using the collective wisdom of colleagues improves the accuracy of prediction and reduces conflict at the end of life. Summary The implications of this review are that doctors should not avoid discussing uncertainty in end-of-life conversations and the article provides some recommendations for minimizing conflict arising from end-oflife discussion. Keywords critical care, end of life, uncertainty
‘Hope is a powerful ally in fighting serious disease and terminal illness, but when false hope is tied to expectations for miracles, then hope is converted to an expectation of a positive outcome when none is possible’ Arthur L. Caplan, PhD
INTRODUCTION Assessing the appropriateness of life-supporting treatment on the intensive care unit (ICU) is difficult. Much of this difficulty stems from uncertainty concerning prognosis and the patient’s wishes. The patient and their family will be anxious about an uncertain future, possible complications and incomplete recovery. Managing such uncertainty properly is an essential skill for those delivering critical care.
UNCERTAINTY IN MEDICINE Uncertainty is relevant to all scientific disciplines. In medicine, uncertainty surrounds complex judgements about diagnosis, treatment and prognosis; it has many causes. Cassell [1] believes that as www.co-criticalcare.com
clinicians gain experience, the more they are beset by uncertainty. Experience may not reduce uncertainty; over time, the physician will remember those patients who have defied the best prediction and unexpectedly survived or died. Cassell concludes that although uncertainty is unavoidable, it is best managed by a well informed doctor who knows as much as possible about the patient and their illness. Katz [2] discusses doctors’ tendency to brush aside medical uncertainty because of simple denial, traditional ideas about ethical conduct, or from beliefs about the proper discharge of professional responsibilities. The quest for absolute certainty may be impossible as the extensive and varied body of present knowledge may now actually lead a Glan Clwyd Hospital, Denbighshire, UK and bRoyal North Shore Hospital of Sydney, University of Sydney, St Leonards, New South Wales, Australia
Correspondence to Professor Malcolm Fisher, AO, MBChB, MD, FCCM, FRCA, Intensive Care Unit, Royal North Shore Hospital, St Leonards, NSW 2065, Australia. Tel: +61 02 9926 7111; fax: +61 02 9926 7779; e-mail: [email protected] Curr Opin Crit Care 2013, 19:642–647 DOI:10.1097/MCC.0000000000000030 Volume 19 Number 6 December 2013
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Uncertainty in end-of-life care Ridley and Fisher
KEY POINTS Uncertainty during end-of-life decision-making arises from doubts about prognostic accuracy and difficulty in establishing the patient’s wishes or best interests. Uncertainty may handicap delivery of care and communication with the family. Key steps in managing uncertainty involve physicians recognizing their own uncertainty and minimizing that of the family primarily by making sure communication is flawless. Uncertainty about prognosis is best dealt with by employing practical certainty wherein after consultation and consensus, the senior physicians openly acknowledge that they are as certain as they can be in the circumstances.
to confusion. Fox [3] described three forms of uncertainty that junior doctors face. The first is incomplete or imperfect mastery of available knowledge. The second depends on the limitations of current medical knowledge. The third is a combination of the two, the difficulty in distinguishing between limitations of personal ignorance or ineptitude and lack of medical knowledge. Beresford [4] categorized clinical uncertainty into conceptual uncertainty (i.e. the inability to apply abstract knowledge to concrete situations), technical or procedural uncertainty (i.e. the absence of scientific data or practical skill) and personal uncertainty (i.e. the lack of previous relationship with a patient and knowledge of their wishes). Thus, in practice prognostic uncertainty at the beginning of a career is related to self-confidence (whether the physician knows sufficient) or whether the information exists to accurately make the decision. Later, the ability to prognosticate is weakened by initial prognoses that were ultimately incorrect and contradictory or controversial medical research. Whatever the cause of uncertainty, a hallmark of clinicians is the ability to tolerate it. In clinical practice, uncertainty is often minimized by de-individualizing the patients to make them more like a textbook case, pretending that uncertainty does not exist, redefining the problem to eliminate uncertainty, shrinking the problem to smaller dimensions and finally hoping that uncertainty will resolve with time. Doctors need to both understand and manage uncertainty, and so be aware of the limitations uncertainty imposes on the reliability and validity of their clinical judgements [5,6]. In ICU, this is an essential prerequisite for quality critical care at the end of life.
APPROPRIATENESS OF CONTINUING TREATMENT Treatment is inappropriate when it is not in keeping with the patient’s wishes or will not produce an outcome acceptable to the patient. Unfortunately, accurately assessing the patient’s wishes or estimating the burden and outcome of treatment is fraught with uncertainty.
Establishing patients’ wishes Many ICU patients may be capable of understanding and retaining information so that they can express a clear choice. If ICU patients can communicate then they appear to make rational and sophisticated decisions despite being in such a hostile, stressful, and perhaps painful environment [7]. The commonest stressors described by ventilated patients (e.g. dyspnoea, pain, anxiety and fear) are important factors that influence cognition [8]. Sedative drugs will further impair thinking and communication. ICU staff have an obligation to respect patients’ wishes, but if these are unknown, the most appropriate treatment choice is uncertain. Advance Care Directives or other previously expressed views may help understand the incompetent patient’s wishes, but whether the views expressed in advance are relevant to the current situation may be uncertain. An advanced care plan relevant to the current situation, which has been discussed with a nominated surrogate in a structured session, is the most reliable record of the patient’s views [9]. Alas, few such care plans are available. Surrogate decision makers, either nominated as part of an advanced care plan or self-selected from family and friends, are normally called upon to assist with decision-making. Ideally, when trying to reflect the patient’s views, surrogates should apply a hierarchy of the patient’s expressed wishes, substituted judgment, and then patient’s best interest. If the patient’s views are unknown, then the surrogates should offer their substituted judgement of what they believe the patient would have wanted. The substituted judgement of families may introduce potential mismatch between the unknown wishes of the patient and the request to sustain life; disagreement within the family may lead to conflict. Surrogates frequently provide valuable insight into the patient’s wishes but they may occasionally have other agendas that may influence their perspective. Best interests encompass a more holistic approach than just best medical interests, but establishing best interest can be challenging. More social or emotional end points are encapsulated within best interests and could be regarded as equivalent (to medical best interests) in deciding how to treat patients. It is vital that ICU
1070-5295 ß 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-criticalcare.com
643
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Ethical, legal and organizational issues in the ICU
staff avoid substituting their own value judgements when making an assessment of best interests but fortunately physicians’ personal values appear not to be a major influence in prognostication [10]. Different religious and ethnic beliefs between the patient and the clinical staff may cause problems when treating minority groups. There may be apprehension about being culturally inappropriate, inadvertently causing offence or appearing discriminatory or racist. If this is not recognized and managed it may lead to uncertainty, disempowerment and inertia in practice [11]. Religious and community leaders may need to be involved if conflict arises.
dysfunction and use of inotropes or vasopressors. However, physicians are not always reliable or consistent even when presented with the same clinical information [16]. With hindsight, clinicians tend to remember data about the final outcome while de-emphasizing contradictory or ambiguous information. Knowing that a patient died or survived to discharge focuses attention on clinical features consistent with that outcome while other important features may be forgotten. Furthermore, a clinician’s perception of uncertainty is strongly influenced by judgements that later in the patient’s clinical course prove incorrect. Generally, ICU physicians tend to be pessimistic about outcome and have concerns about the inappropriate use of technology [17].
Prognostication in intensive care Prognostication in intensive care differs from other disciplines because of the patient’s severity of illness, dependence on life support and rapidly changing clinical situation. Even for senior doctors, prognostication in the critically ill is difficult [12–14], but surprisingly erroneous or biased prognostic estimates do not reduce intensive care treatment choices [10]. Scoring systems can improve survival prediction by quantifying and condensing influential variables into a single measure of illness severity. Scoring systems allow the patient to be placed in the high, medium or low mortality risk group. Unfortunately, they are less useful for predicting outcome for individuals. For an individual prognosis, the scoring system must generate a binary outcome (i.e. dead or alive) based upon the combination of a probability and a decision threshold. Conversion of a probability (ranging from 0 to 1) into a prediction (i.e. dead or alive) requires the decision threshold (e.g. 95% chance of correct prediction). Present systems do not achieve a sufficiently high sensitivity (i.e. correct predictions of the outcome event/total predictions) upon which to base individual treatment decisions. Furthermore, scoring systems are not comprehensive enough to evaluate the complex comorbidities of today’s ICU patient and may not capture the effect of influential higher order variables (such as the organization of ICU). Subsequent improvements in both delivery and application of critical care cannot be accommodated unless the scoring system is updated. As a general principle ICU physicians, especially when there is agreement between individuals, outperform scoring systems in outcome prediction. Rocker et al. [15] reported that physicians’ estimates of below 10% ICU survival were superior at predicting ICU mortality than illness severity, organ 644
www.co-criticalcare.com
Consensus and second opinion Minimizing uncertainty protects the patient’s interests from random and unjustifiable decisions. Clinicians need to collect and appraise evidence about prognosis, and then be honest with themselves and with patients and families about uncertainty. Uncertainty may be reduced by reviewing the literature on prognosis. The current clinical position can then be examined for additional clinical features that may alter the prognosis in one direction or the other. The patient’s response to treatment so far is indicative of overall clinical course. It is important to involve the referring team who will have greater understanding of the presenting pathology, its likely progression and local survival data. Previous local intensive care experience in similar circumstances should be considered. The accuracy of prognosis is improved when a number of physicians (blinded to each other’s prediction) agree on outcome [18]. Although final responsibility for withdrawing or withholding life support rests with the ICU medical staff, the views of other clinical staff are important. The nursing staff will have an established rapport with the patient’s family and so have a clearer idea of their perspective and wishes. Such consensus is reassuring to all who are involved in decision making.
MANAGING UNCERTAINTY WITH THE FAMILY CONFERENCE There are multiple sources of uncertainty for families of patients in intensive care. The fears of losing a loved one, concerns that their loved one is not getting the best possible treatment, or care being trammelled by complications and misinformation are all potent causes of anxiety and uncertainty. These concerns may be fuelled by the sensational Volume 19 Number 6 December 2013
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Uncertainty in end-of-life care Ridley and Fisher
anecdotes of poor treatment in the popular press, and particularly minor ethnic groups may fear that they are not getting a fair deal in their new society. The key to reducing such uncertainty is earning the trust of the family. The two major determinants of this are the intensive care environment and the ability and willingness of the staff to communicate. Families who become involved with intensive care, particularly for the first time, seek a cultural framework. If the environment in which they find themselves has a culture reflecting competence and good staff attitudes towards each other, ensuring the patient is treated respectfully, listening to their concerns and receiving regular explanations, then they and the carers will usually agree on what is best for the patient. It is important that the family see doctors, particularly senior doctors, regularly attending and reviewing the patient. In some units, families may be encouraged to stay at the bedside during ward rounds and procedures. It is important in the initial conversations to stress to the family that the current goal is to discharge the patient from hospital. However, if it becomes apparent over time that this is impossible, it is essential to discuss alternatives. Although families appreciate a single spokesman, rostering in intensive care may make this impossible and under these circumstances the specialist handing over care should personally handover the family and introduce the new specialist. Early open and honest discussion is important in relieving the family’s uncertainty as to the motives of the providers, and especially reassuring them that the decision is not being forced by costs or need for the bed. The basic tool for effectively communicating the appropriateness of continuing intensive care is the family conference in which shared decision-making is employed [19,20]. Such a conference helps to explore and manage the uncertainties faced by all parties. Formal end-of-life discussion should be carried out by experienced people who have been trained in running such discussions. Initially, the family’s understanding of the illness and what they have been previously told should be discussed, and then the prognosis and likely outcome explained. Subsequent to this, the values and wishes of the patient and family should be canvassed. It is important that the family appreciate that there will be time to consider a plan. The family should also be offered the opportunity to bring support persons to the next meeting. It is often helpful to explain to the family that if a palliation pathway is chosen the patient’s last days will be made as comfortable as possible, and that trying to make the patient better will lead to unnecessary suffering.
The clinical picture, its likely outcome and exploring options Determining the likely outcome is within the competence of medical staff as they are best placed to balance the burden of disease against chances of survival. The patient’s response to intensive care support over time provides supplementary information regarding outcome, and gives the family time to understand the problem and the clinician time to gain their trust. A multinational study demonstrated that decisions to withdraw and withhold treatment were rarely made until the fourth day of admission to ICU [21], which also correlated with surrogate acceptance of prognosis [22].
Awareness of uncertainty It is important to appreciate the patients’ and their families’ perspectives of uncertainty. It is absolutely vital that the family receive consistent information from all staff with whom they may engage. This predicates regular discussion of appropriateness and prognosis on rounds and ward meetings. The receipt of differing information, particularly wherein the family have been given false information or hope, is a major cause of uncertainty. This may be compounded by anger when there have been significant complications or errors in care prior to ICU admission. Most US patients want physicians to discuss uncertainty because surrogates believe uncertainty is unavoidable, and that physicians are the best source of accurate prognostic information [23]. Discussing uncertainty leaves room for hope, increases trust in the physician and allows time to make difficult life-support decisions. Only a minority felt that discussions about uncertainty should be avoided as uncertain prognoses may be unnecessarily upsetting or that they should be told when the prognosis was more certain. Physicians need to be sensitive and deliver a consistent message. Both Evans et al. [23] and Woolf [24] suggest the ability to discuss and recognize uncertainty is a hallmark of best practice and that failure to disclose uncertainty does not meet surrogates’ needs. However, 80% of Australian physicians believe discussion of prognosis should be avoided in the face of uncertainty [25].
Employing practical certainty Gillis and Tobin [26] have recommended a change in thinking regarding uncertainty in end-of-life care. They suggest the certainty owed by physicians to families is different from scientific certainty, and
1070-5295 ß 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-criticalcare.com
645
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Ethical, legal and organizational issues in the ICU
practical certainty is the goal. Scientific certainty is not possible and it is wrong for physicians to pretend to have ‘scientific’ certainty, as medicine is largely based upon considerations of probabilities. Practical certainty is being as certain as it is reasonable to be in the circumstances and the ICU is certainly an uncertain environment. When explaining prognosis, the response to the question ‘Is there absolutely no hope?’ should explain that, as there is uncertainty regarding prognosis, the opinions of a number of people have been canvassed and that we are as certain as we can be that there is no further hope. It is important to recognize that practical certainty is the honest option for end-of-life decisions. Practical certainty applies to both the physician and family. The best estimate of prognosis relies heavily on time and the collective wisdom of experienced physicians whereas the best estimate of the patient’s wishes comes from the surrogates’ perspective. The family conference reassures both parties that decisions are based upon best knowledge of the patient’s views and the best estimates of likely outcomes.
Avoiding conflict Unfortunately, conflict between the family and healthcare staff occasionally arises; these tend to be worse case scenarios representing failure of communication and understanding by both parties. Conflict rarely arises spontaneously and usually results from a sequence of unfortunate circumstances or actions, but may relate to unrealistic expectations. Effective communication to improve family understanding needs consistent information delivery probably via a main person of contact, and earning trust and listening. The key constructs for discussion with the family [27] are as follows: (1) Recognize the importance of uncertainty. (2) Use appropriate language of the family conference – ‘withdraw care’ versus ‘nothing more we can do’. (3) Avoid discussion of odds. Families latch on to and concentrate on the given figure. When the risk of death is high and a treatment option is low risk, it may be perceived as acceptable in spite of potential suffering. (4) Provide familiar faces at family conferences and consistent information delivery. If rostering means the discussion leader is to disappear, the family should be formally handed over to the new discussion leader and other participants such as nurses encouraged to continue attending. 646
www.co-criticalcare.com
(5) Earn trust by the family seeing the way the staff interact with each other, and how they speak to and care for the patients. (6) Explain early that the goal of treatment is to return the patient to health and if this is not possible alternatives need discussion. (7) Give families written material regarding endof-life care and ICU management. (8) Present prognosis in terms of ‘Practical Certainty’ – ‘we are as certain as we can be’. (9) Encourage second opinions. (10) Involve known supporters, for example, church/family doctors. These are a major determinant of families deriving benefit from end-of-life discussions [28]. Furthermore, it is important to simultaneously avoid conflict by dealing with issues such as loss of trust (e.g. because of poor previous communication, failure to understand the patient’s perceived views, beliefs or wishes or unrealistic expectations), concern about care (e.g. symptom relief being ignored once a ‘Do Not Resuscitate’ order is agreed, lower nursing care once discharged from ICU), grief and guilt. In both the United Kingdom and Australia, neither the patient nor family can demand treatment that the doctor considers is contrary to the patient’s clinical needs. Several US states have embraced the principles of the Uniform Health-Care Decisions Act which allows physicians to refuse treatment that they deem medically ineffective, is against their conscience, or is contrary to generally accepted medical standards [27]. However, ‘clinical needs’ and ‘medical ineffectiveness’ may be more nebulous at the end of life and unilateral decisions may precipitate further conflict and entrenching of views.
CONCLUSION Families respect wisdom and leadership from those who have earned their trust through providing good treatment and communication early in the course of the illness. Structured negotiation ensures that all the relevant issues are discussed and will facilitate better understanding of all parties’ perspectives. However, end-of-life decision-making needs leadership to steer the discussions. Prior to implementing plans to introduce palliative care, withdraw or withhold artificial forms of life support, there should be consensus among the healthcare workers caring for the patient. Such consensus protects against incorrect decision-making. Practical certainty, wherein the participants are as certain as they can be, with both prognostication and knowing of patient Volume 19 Number 6 December 2013
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Uncertainty in end-of-life care Ridley and Fisher
wishes, may be an appropriate concept for physicians engaged in end-of-life decisions. The collective wisdom of experienced healthcare workers, including patient advocates, increases the likelihood of the most appropriate decisions being made. Recognizing uncertainty improves the families’ confidence in the clinical staff. However, the major challenge in providing better end-of-life care is changing the culture of individual establishments and clinicians. Results from trying to enact such changes have been disappointing [29]. Acknowledgements None. Conflicts of interest The authors have no conflicts of interest to declare.
REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Cassell EJ. The sorceror’s broom: medicine’s rampant technology. Hastings Cent Rep 1993; 23:32–39. 2. Katz J. Why doctors don’t disclose uncertainty. Hastings Cent Rep 1984; 145:35–44. 3. Fox RC. Training for uncertainty. In: Robert K, Merton ???, editors. The student physician: introductory studies in the sociology of medical education, 1st ed. Cambridge, Massachusetts: Harvard University Press; 1957. pp. 207–241. 4. Beresford EB. Uncertainty and the shaping of medical decisions. Hastings Cent Rep 1991; 21:6–11. 5. Zettel Watson L, Ditto P, Danks JH, Smucker WD. Accuracy and perceived gender differences in the accuracy of surrogate decisions about life-sustaining medical treatment among older spouses. Death Stud 2008; 32:273–290. 6. Christensen C, Cottrell JJ, Murakami J, et al. Forecasting survival in the medical intensive care unit: a comparison of clinical prognoses with formal estimates. Methods Inf Med 1993; 32:302–308. 7. Cassell EJ, Leon AC, Kaufman SG. Preliminary evidence of impaired thinking in sick patients. Ann Intern Med 2001; 134:1120–1123. 8. Thomas LA. Clinical management of stressors perceived by patients on mechanical ventilation. AACN Clin Issues 2003; 14:73–81. 9. Detering KM, Hancock AD, Reade MC, Silvester W. The impact of advance care plans on end of life care in elderly patients: randomised controlled trial. BMJ 2010; 340:1345–1354. Important study suggesting that end-of-life planning with a trained facilitator improves quality for patient, family and staff. 10. Edtein AS, Christensen C, Cottrel Coulson JJ, Ng M. Effective prognosis, perceived benefit, and decision style on decision-making in critical care. Crit Care Med 1999; 27:58–65.
11. Kai J, Beavan J, Faull C, et al. Professional uncertainty and disempowerment responding to ethnic diversity in healthcare: a qualitative study. PLoS Med 2007; 4:e323. 12. Meadow W, Pohlman A, Frain L, et al. Power and limitations of daily prognostications of death in the medical intensive care unit. Crit Care Med 2011; 39:474–479. The study shows consensus and time improve prognostic ability of staff but do not give absolute accuracy. 13. Christakis NA, Lamont E. Extent and determinants of error in physicians’ prognoses in terminally ill patients: prospective cohort study. West J Med 2000; 320:469–473. 14. Poses RM, Chaput De Saintonge M, McCLish DK, et al. An international comparison of physician’s judgements of outcome rates of cardiac procedures and attitudes towards risk, uncertainty, justifiability, and regret. Med Decis Making 1998; 18:131–140. 15. Rocker G, Cook D, Sjokvist P, Level of Care Study Investigators, Canadian Critical Care Trials Group. Clinician predictions of intensive care unit mortality. Crit Care Med 2004; 32:1149–1154. 16. Kostopoulou O, Wildman M. Sources of variability in uncertain medical decisions in the ICU: a process tracing study. Qual Saf Healthcare 2004; 13:272–280. 17. Wildman MJ, Sanderson C, Groves J, et al. Implications of prognostic pessimism in patients with chronic obstructive pulmonary disease (COPD) or asthma admitted to intensive care in the UK within the COPD and asthma outcome study (CAOS): multicentre observational cohort study. BMJ 2007; 335:1132–1134. 18. Meadow W, Pohlman A, Frain L, et al. Power and limitations of daily prognostications of death in the medical intensive care unit. Crit Care Med 2011; 39:474–479. 19. Heyland DK, Cook DJ, Rocker GM, et al. Decision-making in the ICU: perspectives of the substitute decision-maker. Intensive Care Med 2003; 29:75–82. 20. White DB, Braddock CH 3rd, Bereknyei S, Curtis JR. Toward shared decision making at the end of life in intensive care units: opportunities for improvement. Arch Intern Med 2007; 167:461–467. 21. Cook DJ, Guyatt G, Rocker G, et al. Cardiopulmonary resuscitation directives on admission to intensive care unit: an international observational study. Lancet 2001; 358:1941–1945. 22. Prendergast TJ. Resolving conflicts around the end of life care. New Horiz 1997; 5:62–71. 23. Evans LR, Boyd EA, Malvar G, et al. Surrogate decision-makers’ perspectives on discussing prognosis in the face of uncertainty. Am J Respir Crit Care Med 2009; 179:48–53. The study shows US families comfortable with discussing uncertainty. 24. Woolf SH. Do clinical practice guidelines define good medical care? The need for good science and the disclosure of uncertainty when defining ‘best practices’. Chest 1998; 113:166S–171S. 25. Brieva JL, Cooray P, Rowley M. Withdrawing and withholding of life-sustaining therapy used in intensive care: an Australian experience. Clinical Care Resusc 2009; 11:266–268. 26. Gillis J, Tobin B. How certain are you doctor? Pediatr Crit Care Med 2011; 12:71–72. The article describing the historical and ethical basis of ‘practical certainty’. 27. National Conference of Commissioners on Uniform State Laws. Uniform Health-Care Decisions Act. 22 Issues L. & Med 2006; 83:83– 97. 28. McDonagh JR, Elliot TB, Engleburg RA, et al. Family satisfaction with family conferences about end of life care in the intensive care unit: increased proportion of family speech is associated with increased satisfaction. Crit Care Med 2004; 32:1484–1488. 29. Curtis JR, Nielsen EL, Treece PD, et al. Effective equality-improvement intervention on end of life care in the intensive care unit. A randomised trial. Am J Respir Crit Care Med 2011; 183:138–155. The study shows a comprehensive intervention to improve end-of-life care in a US hospital did not lead to improvement.
1070-5295 ß 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-criticalcare.com
647
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
