Communication of end-of-life decisions in European intensive care units

Transcription

1 Intensive Care Med (2005) 31: DOI /s x O R I G I N A L Simon Cohen Charles Sprung Peter Sjokvist Anne Lippert Bara Ricou Mario Baras Seppo Hovilehto Paulo Maia Dermot Phelan Konrad Reinhart Karl Werdan Hans-Henrik Bulow Tom Woodcock Communication of end-of-life decisions in European intensive care units Received: 21 December 2004 Accepted: 28 June 2005 Published online: 22 July 2005 Springer-Verlag 2005 P. Sjokvist died in December 2003 Funding was provided by the European Concerted Action project and by the European Commission (contract PL963733), the Chief Scientist s Office of the Ministry of Health, Israel (grant no. 4226), the European Society of Intensive Care Medicine (ESICM) and by OFES Switzerland (Biomed, no ) S. Cohen () ) Department of Medicine, University College London, London, UK simon.cohen@ucl.ac.uk C. Sprung Department of Anesthesiology and Critical Care Medicine, Hadassah Hebrew University Medical Center, Jerusalem, Israel P. Sjokvist Orebro University Hospital, Orebro, Sweden A. Lippert Herlev University Hospital, Herlev, Denmark B. Ricou Hôpital Cantonal Universitaire de Geneve, Geneva, Switzerland M. Baras Hadassah School of Public Health, Hebrew University, Jerusalem, Israel S. Hovilehto South Karelia Central Hospital, Lappeenranta, Finland P. Maia Department of Intensive Care, Hospital Geral Santo Antonio, Porto, Portugal D. Phelan Department of Intensive Care, Mater Hospital University College, Dublin, Ireland K. Reinhart Department of Anesthesiology and Intensive Care Medicine, University Hospital Jena, Jena, Germany K. Werdan Martin Luther University, Halle-Wittenberg, Halle, Germany H.-H. Bulow University Hospital of Glostrup, Glostrup, Denmark T. Woodcock Critical Care Directorate, Southampton University Hospitals NHS Trust, Southampton, UK Abstract Objective: To examine end-of-life (EOL) practices in European ICUs: who makes these decisions, how they are made, communication of these decisions and questions on communication between the physicians, nurses, patients and families. Design: Data collected prospectively on EOL decisions facilitated by a questionnaire including EOL decision categories, geographical regions, mental competency, information about patient wishes, and discussions with patients, families and health care professionals. Setting: 37 European ICUs in 17 countries Patients: ICU physicians collected data on 4,248 patients. Results: 95% of patients lacked decision making capacity at the time of EOL decision and patient s wishes were known in only 20% of cases. EOL decisions were discussed with the family in 68% of cases. Physicians reported having more information about patients wishes and discussions in the northern countries (31%, 88%) than central (16%, 70%) or southern (13%, 48%) countries. The family was more often told (88%) than asked (38%) about EOL decisions. Physicians reasons for not discussing EOL care with the family included the fact that the patient was unresponsive to therapy (39%), the family was unavailable (28%), and the family was thought not to understand (25%). Conclusions: ICU patients typically lack decision-making capacity, and physicians know patients wishes in only 20% of EOL decisions. There were regional differences in discussions of EOL decisions with families and other physicians. In European ICUs there seems to be a need to improve communication Keywords End-of-life decisions Europe Intensive care units Patient competency Patient families Regional differences

2 1216 Introduction Intensive care units (ICUs) are capable of sustaining life in patients beyond the point at which death becomes inevitable. It is therefore widely held to be humane to withdraw or withhold life-sustaining therapies in a timely fashion to minimise suffering and maximise dignity and comfort for dying patients [1]. Most ICU physicians believe that end-of-life (EOL) decisions should include discussions with patients and families, but attitudes vary between regions in Europe [2]. Some physicians believe that it is the physician alone who should make the decision following any discussion, while others believe such decisions should be shared [3]. Descriptive studies in Europe have reported that the involvement of patient and family in EOL decisions varies between 50% and 100% of cases [4, 5, 6, 7, 8, 9, 10]. Previous studies have evaluated isolated events in one or two countries. Although cultural backgrounds may have an impact on these decisions, there are no studies of EOL communication and decision making involving and comparing a substantial number of European countries. The purpose of this study was to examine the communication of EOL decisions in Europe, an important part of the Ethicus study [1] which showed that 73% patients dying in European ICUs have limitations of treatment. Methods This was a prospective observational study of EOL decisions in 4,248 consecutive patients with any limitation of life-sustaining treatment or dying in 37 ICUs in 17 European countries between 1 January 1999 and 30 June The senior physician responsible for patient care and EOL decisions was responsible for completing a study form used in all centres. The development and testing of these questionnaires as well as the definition of EOL categories used [cardiopulmonary resuscitation (CPR), brain death, withdrawing, withholding and active shortening of the dying process], have been previously described [1]. Shortening of the dying process is a circumstance in which treatment is given with the intention of shortening the dying process and is distinct from withholding or withdrawing therapy [1]. No interventions, treatments, withholding or withdrawing therapies were given to patients as part of this observational study. Countries and centres were assigned anonymous codes. Ethics committee approval with a waiver of informed consent was obtained from each participating centre. Data collected included the type of EOL category, country, region, patient mental competency (defined as capacity to make decisions and assessed clinically by the ICU physicians), information about the patient s wishes regarding CPR or treatment limitations from patients or families, discussions with patients or families and discussions with ICU, primary and consulting physicians and nurses. Reasons why discussions did not occur with patients and families and who first brought up the topic of treatment limitation were noted. One reason, unresponsive to maximal therapy, was defined as a shock patient who did not increase blood pressure to vasopressors or a hypoxaemic patient who did not increase PaO 2 to increasing FIO 2 and positive end-expiratory pressure. The total population of 4,248 patients was assessed concerning questions of mental competency, information about the patient s wishes regarding CPR, withholding or withdrawing treatment or shortening of the dying process. Patients who died after unsuccessful CPR or brain death were excluded from analyses of discussions of EOL decisions (withholding, withdrawing or shortening of the dying process), which are available for 3,086 patients. Responses included yes, no and not applicable. The no and not applicable responses along with the few missing responses were all included in the no category. Countries were divided into three geographical regions prior to data analysis as numbers from individual countries were too small: northern (Denmark, Finland, Ireland, The Netherlands, Sweden, United Kingdom), central (Austria, Belgium, Czech Republic, Germany, Switzerland) and southern (Greece, Israel, Italy, Portugal, Spain, Turkey). Geographical comparisons were made because of previous questionnaire data suggesting regional differences [2]. The associations of information about patient wishes, discussions and agreements with the region and EOL decision were tested by the c 2 statistic appropriate for data clustered in ICUs. Statistical analyses were performed using SPSS version 11 and SUDAAN version Differences were considered significant at the level of p<0.05. Results Of the total 4,248 patients 195 (5%) were mentally competent at the time a decision was made to perform CPR, withhold or withdraw therapy or shortening of the dying process. Information about a patient s wishes concerning EOL decisions was available in 850 (20%) of the 4,248 patients: 113 (3%) from the patient, 795 (19%) from the family and 37 (1%) from others. Information concerning patient wishes was available more often in northern countries (461/1,505, 31%) than central (188/ 1,209, 16%) or southern (201/1,534 13%) countries (p<0.0001). In 3,086 patients excluding those with brain death or failed CPR, withholding, withdrawing or a shortening of the dying process decision was made. The number of discussions about withholding or withdrawing of therapy or shortening of the dying process with the patient and family are shown in Table 1. There were more discussions about shortening of the dying process and withdrawing (83%) than withholding therapy (57%, p<0.0001) In the 96 discussions with the patient the patient was told about EOL decision making 54 (56%) times and was asked 68 (71%) times. Thirty discussions involved asking and telling, 38 only asking, 24 only telling and 4 had no data. In the 2,107 discussions with the family the family was told 1,844 (88%) times and asked 797 (38%) times. There were 550 discussions (26%) that involved asking and telling, 247 only asking, 1,294 only telling and 16 had no data. Discussions with the patient and family occurred more often in northern (84%) and central (66%) countries than in southern (47%) countries (Table 2). Discussions did not occur with the patient because the patient was unconscious or incompetent in 94% (2,824/

3 1217 Table 1 Discussions about withholding or withdrawing therapy or shortening of the dying process (SDP) with patients and families Discussed Not discussed n % 95% CI n % 95% CI Patients* Withholding , Withdrawing , SDP Total , Family** Withholding Withdrawing 1, SDP Total 2, *p<0.01, **p< discussed vs. not discussed Table 2 Number and proportion of end-of-life discussions with the patient and family region Discussed Not Discussed n % 95% CI n % 95% CI Patients* Northern Central Southern Total , Family** Northern 1, Central Southern Total 2, *p<0.0001, **p<0.001 discussed vs. not discussed Table 3 End-of-life discussions between ICU physicians and other health care providers by regions* (n=3086) n % 95% CI Northern Other ICU physicians 1, Primary physicians 1, Consulting physicians Nurses 1, Central n % 95% CI Other ICU physicians Primary physicians Consulting physicians Nurses Southern Other ICU physicians Primary physicians Consulting physicians Nurses Total Other ICU physicians 2, Primary physicians 1, Consulting physicians 1, Nurses 2, *p> between regions for each category 2,990) or would not understand in 3% (98/2,990). Discussions did not occur with the family because the patient was unresponsive to maximal therapy (385/979, 39%), the family was unavailable (275/979, 28%), or the family would not understand (241/979, 25%). Reasons for no family discussion were more commonly unresponsiveness to therapy and would not understand and less commonly unavailable family in southern countries (55%, 31% and 11% respectively) than in northern (28%, 18%, 47% respectively) and central countries (24%, 19%, 41% respectively; p<0.0001). The numbers of EOL discussions occurring between the primary ICU physicians and other physicians and nurses by region are demonstrated in Table 3. Agreement between the physicians and nurses and between the staff and patients and families (4 83%) are shown in Table 4. Lack of agreement was usually related to a not applicable response (17 96%). Disagreements were uncommon between ICU physicians and patients (0.6%), families (1.5%), nurses (0.6%) and other physicians (2%). In the 3,086 patients the EOL decision was first brought up by the ICU physician in 2,438 (79%), the primary care physician in 328 (11%), the family in 119 (4%), a consulting physician in 105 (3%), a nurse in 66 (2%) and the patient in 19 (<1%). Discussion The main findings in this study are that the majority of ICU patients who are dying or are the subject of EOL decision making in European ICUs lack personal decision-making capacity, and that patient wishes in this respect are known in 20% of cases, usually from the family. It must be noted that it is not known whether the family s knowledge of the patient s wishes came from direct conversations about preferences for EOL care or were merely general impressions. Therefore the figure of 20% may overstate the actual proportion of cases in which patient choices are known. In EOL decision-making instances (withholding, withdrawing or shortening of the dying process) the patient s family is consulted in only 68% cases, and there are significant variations across Europe in practices regarding communication relating to EOL decisions. This study examined these differences as there is a need to understand the differences and similarities and cultural diversity in order to develop consensus or provide a basis for appropriate legislation or regulation in the European Union. Previous studies have documented that most ICU patients (73 100%) are not competent to participate in EOL decisions [9, 10, 11, 12, 13]. Therefore some surrogate is found to represent the patient. Decisions by families may not correspond to the patient s actual wishes [14, 15, 16, 17]. Therefore it may be very difficult to respect patient autonomy. The present data provide a strong case for

4 1218 Table 4 Agreement on end-oflife decisions between interested parties (NA not applicable) (n=3,086) Yes No NA 95% n % n % n % CI, Yes Doctors and nurses 2,550 83% % % ICU physicians and other physicians 2,205 72% 64 2% % Staff and family 2,035 66% 45 1% 1,006 33% Staff and patient 116 4% % 2,949 96% 3 4 encouraging persons with serious chronic illnesses to express their wishes and values either in some form of an advance directive or to make them known to their family and friends who can subsequently report them to the ICU staff. Incompetent patients may well prefer decisions to be made by a relative who loves them even if the decision may differ from their own [18]. The extent to which serious illness can be held to impair decision-making capacity is an important consideration [19]. Patient preferences were not known because most of the involved patients were unconscious or mentally incompetent. In almost one-third of instances the EOL decision was made by a physician and was not discussed with the family. Despite the fact that family unavailability was given as a reason for the lack of family discussions in four times as many cases in northern than in southern Europe (47% vs. 11%), there was more discussion in both northern and central regions than in southern countries, and there was more information about patient wishes available in northern countries than in central or southern countries. This study found that EOL decisions were discussed with families in 68% of cases. Previous studies have demonstrated that there is more communication of EOL decisions to ICU patients and families in North America (93 100%) than in Europe (44 88%) [2, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]. Conte et al. [21] in a French study on withholding and withdrawing life supporting therapy in an Emergency Department reported that the family was involved in 73% of cases, but that there was poor participation of the patient in the decisionmaking process. These findings are very similar to those in the present study. The need for good communication in EOL situations has long been recognised [16]. There is increasing public interest in medical and ethical matters with an expectation of more patient and family involvement in decision making. It is unclear whether this is true in southern Europe as it is in northern Europe and North America, where cultural or religious features may differ [1, 2]. ICU physicians initiated discussion of EOL decisions in most (70%) of the instances; the patient and family initiated EOL discussion in only 1 and 4%, respectively. Perhaps the public may take reassurance from data showing that ICU physicians have recognised the futility of life-sustaining treatments in individual patients, have been the first to initiate discussion about EOL decisions and have not supported persistence with a mechanistic prolongation of death when, as is sometimes feared, the inevitability of death might be considered to be evident to the patient and family. However, when discussions took place, the clinicians more often told the patients and families about the decision. Sjokvist et al. [3] reported that when the patient is incompetent, 61% of physicians thought they alone should make the EOL decision. In contrast, 73% of the general public and 70% of nurses prefer a joint decision by the family and physician [3]. A paternalistic pattern of medical staff communication is suggested by the twofold greater incidence (88% vs. 38%) of telling compared with asking in regard to EOL discussion with patient and family. However, telling families about EOL decisions when there is physiologically futile treatment is probably more appropriate than asking, as asking sends a mixed message, tends to confuse patients and families and undermines rather than enhances autonomy [22]. Both asking and telling were reported in 26% of instances, which suggests that the patient and/or family in these instances was told that death was inevitable (and that further life-sustaining treatment was futile) but were asked regarding patient EOL preferences. Discussions did not occur with families in many cases because the patient was unresponsive to maximal therapy, or because the physician believed that the family would not understand. While it may be reasonable in instances in which there was unresponsiveness to maximal therapy to forego discussion on the basis that there was no real option to discuss, nevertheless communication and informing the family about the situation is crucial to ensure that the futile condition is understood and can be accepted. The failure to enter discussion with family on the grounds that they would not understand does suggest a remarkably paternalistic approach in these instances. Heyland et al. [23] described the perspectives of family members to the care provided to critically ill patients who died in Canadian ICUs. Adequate communication was an important determinant of satisfaction with EOL care in the ICU. These discussions may nevertheless be very helpful for the relatives. Because of the legal uncertainty of some EOL practices in several European countries, the relative lack of communication may be secondary to physicians fear of charges of negligence or criminal actions [20]. There is more discussion by physicians and nurses in northern (84%) and central

5 1219 Table 5 Participating investigators of the Ethicus study Country Investigators Affiliations Austria W. Schobersberger a, D. Fries University Hospital Innsbruck, Innsbruck Belgium M. Lamy a, P. Damas, J-L Canivet, University of Liege, Liege D. Ledoux P. Lauwers University Hospital Gasthuisberg, Leuven Czech I. Novak a Charles University Hospital, Pilsen Republic D. Nalos Aro Masarykova Nemonice, Ustinad Labem Denmark A. Lippert a University Hospital, Herlev H-H Bulow, H. Christensen, L.Q. University Hospital, Glostrup Christensen M. Wanscher University Hospital, Arhus Finland S. Hovilehto a, S. Seittenranta, A. South Karelia Central Hospital, Lappeenranta Ahokainen Germany K. Werdan a, C. Kuhn Martin Luther University Halle-Wittenberg, Halle K. Reinhart; M. Brauer Friedrich Schiller University, Jena Greece A Armaganidis a Evangelismos Hoapital, Athens N. Maguina Agia Olga Hospital of Nea Ionia, Athens G Nakos, H. Tsagaris University Hospital, Ionnina Ireland D. Phelan a, B. Marsh, N. Collins Mater Hospital, University College, Dublin Israel C. L. Sprung a, Y. Weiss, P. Levin Hadassah Hebrew University Medical Center, Jerusalem A. Fisher, G. Gurman Soroka Hospital, Beer-Sheva M. Hersch Shaarei Zedek Medical Center, Jerusalem Italy G. Conti a University of Rome La Sapienza, Rome A. Braschi, A. Palo Policlinico San Matteo, Pavia Netherlands L.G.Thijs a, P. Scholten, P. Spronk, Free University Hospital, Amsterdam S. Simsek Portugal P. Maia a, F. Rua Hospital Geral Santo Antonio, Porto L. Ferreira, L. Telo Hospital Pulido Valente, Lisboa E. Oliveira, G. Gomes, J.E. Oliveira U.C.I.P. Hospital de S. Marcos, Braga R. Rodrigues, J. Salcher Hospital de S. Jose, Lisbon Spain F. Solsona a Hospital del Mar, Barcelona C. Bouza-Alvarez Hospital Gregorio Maradon, Madrid R. Abizanda, S. Mas Hospital General, Castello Sweden P. Sjokvist a, A. Nydahl, L. Berggren University Hospital, Orebro E. Wennberg Sahlgrenska University Hospital, Gothenburg Switzerland B. Ricou a, M. Diby Hôpital Cantonal Universitaire, Geneva R. Malacrida, A. Canonica, G. Vanini Ospedale Regionale, Lugano R. Ritz, M. Grob Kantosspital, Basel Turkey F. Esen a, S. Tugrul, O. Kutlay, University of Istanbul, Istanbul G. Korfali, F. Kahveci United S. Cohen a, C. Matejowsky University College London, London Kingdom C. Wallis Western General Hospital, Edinburgh T. Woodcock, K. de Courcy-Golder Southampton University Hospitals NHS Trust, Southampton a Country Coordinator (66%) countries than in southern (47%) regions. Further studies are needed to evaluate whether southern European families desire shared decision making or a paternalistic approach. In this study there was a relatively high level of discussions between nurses and physicians (78%), but this was reported by physicians, and nurses may have responded differently. The present study showed agreement on the decision between physicians and nurses (83%), between ICU physicians and other involved physicians (71%), and between the staff and family (66%) in the majority of instances. Agreement was not higher than recorded because discussions simply did not take place. Agreement between the staff and patient was rare, primarily due to patient mental incapacity. Disagreements were uncommon but may have been underestimated. Physicians decided about agreement and may not have been sensitive to the fact that nurses, patients, and families might not have agreed with their determination [24]. Agreement was similar in a study in Spain [10] and much higher in studies in North America and Australia [12, 13, 25]. It appears that the level of communication and agreement for EOL decisions in Europe could be improved. There are limitations to the present study. The patients studied may not be representative of what actually occurs in all European countries [1]. There were one to four centres in each country, and their policies may differ from the many other ICUs in the same country. In addition, participating investigators were interested in ethical is-

6 1220 sues, and their actions may not reflect those of other physicians. Also, the data reflect the opinions of senior ICU physicians, and other parties may have a different view of the events. A recent study by Ferrand et al. [24] found differences between what the physician believed and what the nurses said. The findings of the present study, however, do reflect what physicians from different countries state they do [2], and anonymity probably led to accurate reporting. The study highlights deficiencies and differences in communication practice within Europe. The frequency of these EOL decisions may also reflect the nature of the patient population, with increasing pressure to admit sicker patients who have little hope of recovery. The frequency of EOL decision making is an increasing fact of ICU life and the level recorded (10% of total patients or 73% of the study population [1]) corresponds with the experience in the United States of increasing active decision making [13, 26]. The present study demonstrates that EOL decisions in Europe are often not communicated to patients or families. Furthermore, when communication does take place, this mostly consists of the physician informing about the decision rather than shared decision making. This practice may be acceptable if it reflects the values of the persons being cared for. The desire of the patients and proxies to be involved either by receiving information or by shared decision making in EOL decisions may vary from region to region and country to country but also from patient to patient depending on his/her cultural and religious ties [20]. Further studies are needed to examine whether such agreement between practice and attitudes exists. Acknowledgements The Ethicus Steering Committee: C. Sprung, S. Cohen, L. Epstein, D. Miranda, F. Lemaire, and G. Van Steendam. Participanting investigators and their affiliation are listed in Table 5. References 1. Sprung CL, Cohen SL, Sjokvist P, Baras M, Bulow HH, Hovilehto S, Ledoux D, Lippert A, Maia P, Phelan D, Schobersberger W, Wennberg E, Woodcock T (2003) End of life practices in European intensive care units the Ethicus study. JAMA 290: Vincent JL (1999) Forgoing life support in western European intensive care units: the results of an ethical questionnaire. Crit Care Med 27: Sjokvist P, Nilstun T, Svantesson M, Berggren L (1999) Withdrawal of life support who should decide? Differences in attitudes among the general public, nurses and physicians. Intensive Care Med 25: Pijnenborg L, van der Maas PJ, Kardun JW, Glerum JJ, van Delden JJ, Looman CW (1995) Withdrawal or withholding of treatment at the end of life. Results of a nationwide study. Arch Intern Med 155: Deliens L, Mortier F, Bilsen J, Cosyns M, Vander Stichele R, Vanoverloop J, Koen I (2000) End-of Life decisions in medical practice in Flanders, Belgium: a nationwide survey. Lancet 356: Melltrop G, Nilstun T (1996) Decisions to forego life-sustaining treatment and the duty of documentation. Intensive Care Med 22: Turner JS, Michell WL, Morgan CJ, Benatar SR (1996) Limitation of life support: frequency and practice in a London and Cape Town intensive care unit. Intensive Care Med 22: Sjokvist P, Sundin PO, Berggren L (1998) Limiting life support. Experiences with a special protocol. Acta Anaesthesiol Scand 42: Ferrand E, Robert R, Ingrand P, Lemaire F, French LG (2001) Withholding and withdrawal of life support in intensive care units in France: a prospective survey. French LATAREA Group. Lancet 357: Esteban A, Gordo F, Solsona F, Alia I, Caballero J, Bouza C, Alcala-Zamora J, Cook DJ, Sanchez JM, Abizanda R, Miro G, Fernandez del Cabo MJ, de Miguel E, Santos JM, Balerdi B (2001) Withdrawing and withholding life support in the intensive care unit: a Spanish prospective multi-centre observational study. Intensive Care Med 27: Smedira NG, Evans BH, Grais LS, Lo B, Cooke M, Schecter WP, Fink C, Epstein-Jaffe E, May C, Luce JM (1990) Withholding and withdrawal of life support from the critically ill. N Engl J Med 322:309Ÿ Lee DKP, Swinburn AJ, Fedullo AJ, Wahl GW (1994) Withdrawing care. Experience in a medical intensive care unit. JAMA 271: Prendergast TJ, Luce JM (1997) Increasing incidence of withholding and withdrawal of life support from the critically ill. Am J Respir Crit Care Med 155: Cook DJ, Guyatt G, Rocker G, Sjokvist P, Weaver B, Dodek P, Marshall J, Leasa D, Levy M, Varon J, Fisher M, Cook R (2001) Cardiopulmonary resuscitation directives on admission to intensive care unit. An international observational study. Lancet 358: Layde PM, Beam CA, Broste SK, Connors AF Jr, Desbiens N, Lynn J, Phillips RS, Reading D, Teno J, Vidaillet H (1995) Surrogates predictions of seriously ill patients resuscitation preferences. Arch Fam Med 4: Uhlmann RF, Pearlman RA, Cain KC (1998) Physicians and spouses predictions of elderly patients resuscitation preferences. J Gerontol 43:M115 M Suhl J, Simons P, ReedyT, Garrick T (1994) Myth of substituted judgement. Surrogate decision making regarding life support is unreliable. Arch Intern Med 154: High DM (1998) All in the family: extended autonomy and expectations in the surrogate health care decisionmaking Gerontologist 28: Cassell EJ, Leon AC, Kaufman SG (2001) Preliminary evidence of impaired thinking in sick patients. Ann Intern Med 134: Pochard F, Azoulay E, Chevret C, Vinsonneau C, Grassin M, Lemaire F, Herve C, Schlemmer B, Zittoun R, Dhainaut J-F (2001) French Intensivists do not apply American recommendations regarding decisions to forgo lifesustaining therapy. Crit Care Med 29: Le Conte P, Baron D, Trewick D, Touze MD, Longo C, Vial I, Yaltin D, Potel G (2004) Withholding and withdrawing life support therapy in an emergency department: prospective study. Intensive Care Med 30:

7 Tomlinson T, Brody H (1990) Futility and the ethics of resuscitation. JAMA 264: Heyland DK, Rocker GM, O Callaghan CJ, Dodek PM, Cook DJ (2003) Dying in the ICU: perspectives of family members. Chest 124: Ferrand E, Lemaire F, Regnio B, Kuteisan K, Badet M, Asfar P, Jaber S, Chagon JL, Renault A Robert R, Pochard F, Herve C, Brun-Buisson C, Duvaldestin P (2003) Discrepancies between perceptions by physicians and nursing staff of intensive care, end-oflife decisions. Am J Respir Crit Care Med 167: Daffurn K, Kerridge R, Hillman KM (1992) Active Management of the dying patient. Med J Aust 157: Prendergast TJ, Claessens MT, Luce JM (1998) A National Survey of End of Life Care for critically ill patients. Am J Respir Crit Care Med 158: