International Differences in End-Of-Life Attitudes in the Intensive Care Unit Results of a Survey

Home , Northern Europe

ORIGINAL INVESTIGATION International Differences in End-of-Life Attitudes in the Intensive Care Unit Results of a Survey

Arino Yaguchi, MD; Robert D. Truog, MD; J. Randall Curtis, MD; John M. Luce, MD; Mitchell M. Levy, MD; Christian Mélot, MD; Jean-Louis Vincent, MD, PhD

Background: Important international differences ex- tral Europe, whereas oral orders took preference in South- ist in attitudes toward end-of-life issues in the intensive ern Europe, Turkey, and Brazil. One third of Japanese care unit. physicians said that they would not apply do-not- resuscitate orders. Most participants from Japan, Tur- Methods: A simple questionnaire survey was sent by key, the United States, Southern Europe, and Brazil chose e-mail to participants at an international meeting on in- to treat the hypothetical patient with antibiotics if he/ tensive care medicine. Respondents were asked to choose she developed septic shock, whereas in Northern Eu- 1 of 3 to 5 possible answers for each of 4 questions re- rope, Central Europe, Canada, and Australia, terminal lated to the treatment of a hypothetical patient in a veg- withdrawal of mechanical ventilation and extubation were etative state due to anoxic encephalopathy after cardiac the more commonly chosen responses. arrest with no family and no advance directives. Conclusions: In countries where intensive care medi- Results: From 3494 valid addresses, 1961 complete ques- cine is relatively well developed, considerable differ- tionnaires (56%) were received from 21 countries. Sixty- ences remain in physicians’ attitudes toward end-of-life two percent of physicians from Northern and Central Eu- rope said they involved nurses in end-of-life discussions care in the intensive care unit. Substantial work re- compared with only 32% of physicians in Southern Eu- mains if an international consensus on these issues is to rope, 38% in Brazil, 39% in Japan, and 29% in the United be reached. States (PϽ.001 for all comparisons). Written do-not- resuscitate orders were preferred in Northern and Cen- Arch Intern Med. 2005;165:1970-1975

ISCUSSIONS ABOUT END- ing to the variability of these end-of-life of-life issues have be- decisions and the difficulty in achieving come common, espe- an international consensus on these is- Author Affiliations: cially in the Western sues. Numerous reports exist on this Department of Intensive Care medical literature. Ad- topic,4-13 and several surveys have been Medicine, Erasme Hospital, vances in medical technology and thera- conducted2,14-25; however, these surveys Free University of Brussels, D Brussels, Belgium (Drs Yaguchi, pies enable more lives to be saved but were limited to national or geographic re- Mélot, and Vincent); sometimes may merely prolong the dy- gions, and no large international survey has Department of Anesthesia and ing process. Recent studies1,2 have shown been performed. As previous studies have Medical Ethics, Harvard that the exact timing of death is often un- shown,14,17,18,20 substantial differences ex- Medical School, Children’s der the control of the physicians who care ist among European countries. Further- Hospital, Boston, Mass for the patient. This is particularly true in more, the guidelines in the United States26 (Dr Truog); Division of the intensive care unit (ICU),3 where death are also different from those within Eu- Pulmonary and Critical Care 27,28 Medicine, University of is commonly preceded by decisions not to rope. Although some would defend the Washington, Harborview start aggressive therapy (withholding) or need for the development of an interna- Medical Center, Seattle to discontinue life-sustaining therapy tional consensus or a global system of eth- (Dr Curtis); Department of (withdrawing). The question of who ics,3 this may not be achievable or even de- Medicine and Anesthesia, should be responsible for decision mak- sirable. In a recent survey of physicians University of California, San ing is a difficult issue; the patient is often involved in end-of-life care in 6 Euro- Francisco (Dr Luce); and unable to participate in this decision, and pean countries and Australia, Miccinesi et Department of Critical Care, 29 Rhode Island Hospital, Brown the roles of the medical team, the rela- al reported that country was the stron- University, Providence tives, and the legal system are highly vari- gest determinant of physician attitude to- (Dr Levy). able in different parts of the world, add- ward end-of-life decisions, although the in-

(REPRINTED) ARCH INTERN MED/ VOL 165, SEP 26, 2005 WWW.ARCHINTERNMED.COM 1970

Downloaded From: https://jamanetwork.com/ on 09/25/2021 dividual physician characteristics of age, religious beliefs, sex, and previous experience with dying patients were Table 1. The Questionnaire also strong determinants. Other studies11,14,17,18,29,30 have shown the strong influence of religion, culture, race, le- Case Scenario gal background, social factors, and tradition on atti- A 55-year-old woman was resuscitated from cardiac pulmonary arrest tudes toward end-of-life issues. In addition, these issues due to cardiac ischemic disease and admitted to the intensive care 11,14,17 unit (ICU) with severe postanoxic lesions. Twenty-four hours later, are influenced by and change with temporal trends, she has decerebration movements and the evoked potentials are as illustrated by the Hippocratic Oath, which recog- absent. The consensus (including the senior neurologist) is that her nized the importance of the principles of beneficence, non- best possible outcome is a persistent vegetative state. She has no maleficience, and medical futility but did not discuss au- close relative or advance directive. tonomy or distributive justice. During the last 4 decades, Question 1. Which process do you follow to decide on the treatment autonomy and distributive justice have become increas- for this patient in the ICU? ingly predominant. 1. Decide by yourself. To better understand the international differences in 2. Decide after a consensus is reached with other physicians. 3. Decide after discussions involving other physicians and nurses. end-of-life decisions in the ICU, we conducted a simple 4. Refer to the ethical committee in your hospital. international survey by e-mail. The aim of the study was 5. Refer to court. to record current end-of-life attitudes of ICU physicians Question 2. Is this process likely to result in do-not-resuscitate (DNR) in those parts of the world where the practice of inten- orders being applied in the event of recurrent cardiac arrest? sive care medicine is fairly well developed. 1. Yes, written DNR orders. 2. Yes, verbal DNR orders. 3. No. METHODS Question 3. The patient remains absolutely stable for 5 days and, although still receiving mechanical ventilation, can breathe STUDY DESIGN spontaneously. What would be the usual strategy in your institution? 1. Keep the patient in the ICU (with or without tracheostomy) and A list of e-mail addresses of individuals who had requested in- start further interventions if a complication occurs. formation regarding the International Symposium on Inten- 2. Keep the patient in the ICU (with or without tracheostomy)— sive Care and Emergency Medicine was provided by the meet- “wait and see”—but withhold therapy if a complication occurs. ing’s secretariat. This symposium (www.intensive.org), held 3. Keep the patient in the ICU and start increasing doses of every year in Brussels, Belgium, is one of the largest interna- morphine or sedatives with the intent to decrease ventilatory tional meetings in this field, with close to 5000 participants at- conditions (“terminal weaning”). tending each year. We limited our list of addresses to physi- 4. Perform a tracheostomy and transfer the patient to the general cians and to countries with more than 100 physicians in the ward for continued care. 5. Perform a tracheostomy and transfer the patient to the general database and, at the end of 2003, invited physicians on this list ward, but with the intent to stop enteral feeding. to reply to a simple questionnaire sent by e-mail. Two remind- ers were sent to those who did not reply to the first e-mail. Question 4. While the possible options are being considered, let us The questionnaire consisted of a case scenario and 4 ques- imagine that the patient rapidly develops fever and septic shock, presumably due to lung infection. What would likely be done in your tions with 3 to 5 optional answers (Table 1). The question- institution? naire was given in English, which is the official language of this 1. Maintain mechanical ventilation and start antibiotics and international symposium. The participants were invited to an- vasopressors. swer by e-mail, and it was guaranteed that they would be anony- 2. Maintain mechanical ventilation and start antibiotics but no mous in the data analysis. We excluded answers received from vasopressors. nonphysicians from the analysis. European countries were di- 3. Give morphine and reduce ventilatory conditions (“terminal vided into 3 groups by geographic region: Northern Europe weaning”). (Denmark, Finland, the Netherlands, Sweden, and the United 4. Extubate and then give morphine (“terminal extubation”). Kingdom), Central Europe (Austria, Belgium, the Czech Re- public, France, Germany, and Switzerland), and Southern Eu- rope (Greece, Italy, Portugal, and Spain). are consistent with their associations in the table. The goal is STATISTICAL ANALYSIS to obtain a global view of the data that is useful for interpre- tation. The overall ␹2 of the contingency table reflects the de- Statistical analyses were performed using commercial soft- gree of departure from a purely random distribution between ware StatView 5.0 for Windows (SAS Institute Inc, Cary, NC). the responses and the countries. If no relationship exists be- A ␹2 test was used to compare groups. Multivariate statistical tween the countries and the responses, the ␹2 will equal zero. analysis was also performed, using multiple correspondence The correspondence analysis will partition the ␹2 in dimen- analyses. Briefly, the method consists of the analysis of the 2-way sions that correspond to different angles of view of the data set. contingency table in which the observed association between The first dimension is the best angle of view of the data, fol- the rows (the region of origin of the respondent) and the col- lowed by the second dimension, and so on. Each dimension umns (the responses to the different options) is summarized reflects a percentage of inertia, which corresponds to the pro- by the cell frequencies. The inference in correspondence analy- portion of the ␹2 statistic explored by this dimension (ie, the ses is whether certain levels of one characteristic (eg, region) information contained in the data). The axis of the graph, la- are associated with some levels of another characteristic (eg, beled dimension, cannot be interpreted from a clinical stand- responses to the different options). Correspondence analysis point. The advantage of the multivariate approach is to ana- is a geometric technique for displaying the rows and columns lyze the data globally and to take into account the correlation of this 2-way contingency table as vectors and points in low- among the variables, thus reducing bias in the results. PϽ.05 dimensional space, such that the row vectors and column points was considered statistically significant.

(REPRINTED) ARCH INTERN MED/ VOL 165, SEP 26, 2005 WWW.ARCHINTERNMED.COM 1971

Downloaded From: https://jamanetwork.com/ on 09/25/2021 respectively). This was particularly true in the United Table 2. Numbers of Participants in the 21 Countries Kingdom, France, and Switzerland, where approximately 80% of physicians involved nurses (83%, 80%, and No. (%) of Response 78%, respectively). Of respondents from Southern Eu- Country or Region Participants Rate, % rope, only 32% replied that they would involve nurses United States 199 (10.1) 66.3 (PϽ.001 vs Northern or Central Europe); percentages for Canada 58 (3.0) 64.4 Turkey (41%), Brazil (38%), Japan (39%), and the United Australia 56 (2.9) 62.9 Northern Europe 517 (26.4) 56.6 States (29%) were similarly low. Denmark 37 (1.9) 44.6 Finland 32 (1.6) 62.7 DO-NOT-RESUSCITATE ORDERS The Netherlands 123 (6.3) 54.4 Sweden 84 (4.3) 68.3 The second question dealt with do-not-resuscitate (DNR) United Kingdom 241 (12.3) 56.0 orders, for which a striking difference was noted among Central Europe 734 (37.4) 54.9 Ͻ Austria 60 (3.1) 50.8 regions (P .001; Figure 1). More than 80% of physi- Belgium 233 (11.9) 55.7 cians in Northern and Central Europe, North America, and Czech Republic 34 (1.7) 53.1 Australia said that they would be likely to apply written France 121 (6.2) 48.6 DNR orders in the hypothetical case presented. In con- Germany 194 (9.9) 57.4 trast, in Southern Europe, Turkey, and Brazil, oral DNR Switzerland 92 (4.7) 60.9 orders were preferred. Japan was the country with the high- Southern Europe 257 (13.1) 46.0 Greece 42 (2.1) 36.8 est proportion (approximately one third of the respon- Italy 117 (6.0) 49.6 dents) who said that they would not apply DNR orders. Portugal 23 (1.2) 29.9 Spain 75 (3.8) 56.8 TREATMENT FOR A STABLE Turkey 27 (1.4) 46.6 VEGETATIVE PATIENT Brazil 39 (2.0) 69.6 Japan 74 (3.8) 81.3 Total 1961 (100) 56.1 Question 3 asked about the treatment of a patient in a vegetative state. Participants from Japan and Turkey were more likely to reply that they would continue full support than participants from the other countries (Figure 2). Australian and Canadian participants chose RESULTS terminal withdrawal of mechanical ventilation as their response more commonly than participants from other STUDY PARTICIPANTS countries. Northern Europe had a greater tendency to transfer patients to the general ward. From a total of 3673 valid addresses, 179 respondents declared that they had no ICU responsibility. From the TREATMENT FOR SEPTIC SHOCK remaining 3494 valid addresses, we received 1961 com- IN A VEGETATIVE PATIENT plete surveys (56.1%) from ICU physicians. The numbers and percentages of participants in each of the 21 Question 4 raised the hypothesis that the vegetative pa- Table 2 countries represented are presented in . The larg- tient developed septic shock (Figure 3). Choice of the est numbers of participants were from the United King- full-support option ranged from 2% of respondents in Aus- dom, Belgium, the United States, and Germany. The tralia to 65% in Greece. The option to give antibiotics, percentage of participants for each region ranged from with or without vasopressors (option 1 or 2), was se- 46.0% in Southern Europe to 81.3% in Japan. For all ques- lected by most physicians in Japan (96%), Turkey (81%), tions, there was a statistically significant difference among Ͻ the United States (65%), Southern Europe (62%), and the countries and regions in the answers given (P .001). Brazil (59%) (PϽ.001 vs other regions), whereas in North- ern Europe, Central Europe, Canada, and Australia, ter- DECISION MAKING minal withdrawal of mechanical ventilation and extubation were more commonly chosen. The scenario describes the management of a patient in a vegetative state due to anoxic encephalopathy after cardiac arrest with no family and no advance directives. The COMMENT first question dealt with the decision-making process. The United States was the country with the largest propor- This simple survey revealed striking international dif- tion (45%) of respondents saying that they would ask for ferences in the attitudes of ICU physicians toward end- an ethical consultant in the hospital or make a court re- of-life issues in adult patients. In our scenario-based ques- ferral. Fewer respondents in the United States and Canada tionnaire, we presented a patient with no family and no replied that they would make the decision alone or just advance directives so that the responsibility for decision with other physicians (27% and 29%, respectively) than making would be on the physician or health care team in the other regions. Respondents from Northern and Cen- rather than the family. Under these conditions, more phy- tral Europe replied more often than other areas that they sicians from the United States thought that it could be would involve nurses in such discussions (62% and 62%, helpful to refer outside the ICU than other countries; it

(REPRINTED) ARCH INTERN MED/ VOL 165, SEP 26, 2005 WWW.ARCHINTERNMED.COM 1972

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Would Apply Would Apply Would Not Apply Written Orders Verbal Orders DNR Orders

100

Respondents, % 40

0 United States Canada Australia Northern Europe Central Europe Southern Europe Turkey Brazil Japan Region

Figure 1. Percentage of respondents in the various regions for question 2 related to the application of do-not-resuscitate (DNR) orders.

is possible that the legal environment in the United States may have some influence on these decisions.31 In Eu- 0.30 Southern Europe rope, physicians seem to prefer to decide within the ICU Transfer team, rather than involving outside ethical or legal ad- General Ward

visers. Participants in Northern and Central Europe re- Northern Europe No Escalation plied that they were more likely to involve nurses in the 0 decision making than were respondents from Southern Central Europe European countries. Notably, participants from the United Terminal Brazil Withdrawal United States were the least likely to report involving nurses. States Turkey The application of DNR orders varied with geo- –0.30 Australia graphic region. Respondents from Australia, Canada, the Canada United States, and Northern and Central Europe gener- Dimension 2 (31% of Inertia) Japan ally replied that they would apply written DNR orders. However, Turkey, Southern Europe, and Brazil pre- 17 –0.60 ferred the oral DNR orders. A previous European study Do Everything also showed that DNR orders were less commonly written in Southern than in Northern and Central Europe. In Japan, approximately one third of respondents would –1.00 –0.50 0 0.50 1.00 not apply DNR orders at all. Dimension 1 (62% of Inertia) Participants from Turkey and Japan chose aggressive therapy with continued full support more often than par- Figure 2. Attitudes related to question 3 regarding a stable vegetative ticipants from other countries. In Northern Europe, Cen- patient. Vectors show the direction to the options. Percentages show the variance of the plots explained by the principal component. The proximity tral Europe, the United States, and Brazil, passive with- between points (regions) and the tip of a vector (responses) indicates drawal of therapy, such as transfer to the general ward, similarity between the end-of-life attitudes in these countries (see the was preferred over stopping feeding, extubating, or de- “Statistical Analysis” subsection in the “Methods” section for details). creasing ventilator conditions. When a serious septic com- Circles indicate answer by region; triangles, answer by option. plication occurred in this hypothetical patient, respondents from Southern Europe, Turkey, Canada, the United tral Europe, respondents were more likely to reply that they States, Brazil, and Japan usually treated the complication would actively withdraw therapy. In Australian respon- with antibiotics, and Japanese physicians were also likely dents, the occurrence of sepsis did not alter the treatment to give vasopressors. In contrast, in Northern and Cen- plan. These data confirm prior studies of geographic dif-

(REPRINTED) ARCH INTERN MED/ VOL 165, SEP 26, 2005 WWW.ARCHINTERNMED.COM 1973

Downloaded From: https://jamanetwork.com/ on 09/25/2021 80 A 80 B

60 60

40 40 Respondents, % Respondents, %

20 20

0 0 s e s e ope ope key zil ope ope key zil an nada stralia ur ur r Bra Japan nada stralia ur ur r Bra Jap d State Ca Au E E Tu d State Ca Au E E Tu nite ern tral ern Europ nite ern tral ern Europ U en uth U en uth North C So North C So Region Region

80 80 C D

60 60

40 40 Respondents, % Respondents, %

20 20

0 0 s e s e ope ope key zil an da ope ope key zil an stralia ur ur r Bra Jap na stralia ur ur r Bra Jap d State Canada Au E E Tu d State Ca Au E E Tu nite ern tral ern Europ nite ern tral ern Europ U en uth U en uth North C So North C So Region Region

Figure 3. Responses to question 4 related to treatment for septic shock in a patient in a vegetative state. A, Physicians who would recommend terminal extubation; B, physicians who would recommend terminal withdrawal of mechanical ventilation; C, physicians who would recommend only antibiotics; and D, physicians who would do everything.

ferences within Europe17,18 and studies of variability within ternative option would have been to use databases of re- the United States32 and Canada.15 In addition, our data show gional scientific societies, but logistically this would have that these geographic differences continue to persist and been much more difficult. Physicians attending such a that they extend around the world. Furthermore, the mul- conference may reflect a self-selected population with dif- tiple correspondence analyses show that the countries do ferent attitudes from other critical care physicians. How- not all lie on 1 axis, suggesting variability in the re- ever, we would expect that the bias would be toward phy- sponses within each country depending on individuals. sicians with a more global perspective. This would suggest Our study has several limitations. First, the respon- that true differences in different regions might actually dents were not a randomly selected sample of all critical be larger than found in this study. Second, to minimize care physicians because such a database does not exist. the length of the survey and maximize the response rate, Rather, we used a list of potential participants at a major we have no information on the respondents’ back- international critical care conference held annually in Eu- grounds, age, sex, position in their hospital, or religion, rope. This may have resulted in a distance factor; respon- and unidentified differences among countries in these sub- dents from Belgium and neighboring countries may be groups may potentially have influenced the results. Third, different from those coming from farther away. An al- the case scenario was by intent somewhat simplistic. Any

(REPRINTED) ARCH INTERN MED/ VOL 165, SEP 26, 2005 WWW.ARCHINTERNMED.COM 1974

Downloaded From: https://jamanetwork.com/ on 09/25/2021 influence of relatives or the patients themselves on these statement of the 5th International Consensus Conference in Critical Care: Brus- decisions could not be taken into account. Moreover, as sels, Belgium, April 2003. Intensive Care Med. 2004;30:770-784. 4. Cook DJ. Health professional decision-making in the ICU: a review of the evidence. in all questionnaire studies, it is not always possible to New Horiz. 1997;5:15-19. provide an appropriate answer for all situations, and other 5. Teres D. Trends from the United States with end of life decisions in the intensive options may have been preferred if they had been of- care unit. Intensive Care Med. 1993;19:316-322. fered to the respondents. Fourth, the questionnaire was 6. Johnson N, Cook D, Giacomini M, Willms D. Towards a “good” death: end- developed by the authors and did not undergo reliabil- of-life narratives constructed in an intensive care unit. Cult Med Psychiatry. 2000; 24:275-295. ity or validity testing. Finally, although e-mail provides 7. DeVita MA, Groeger J, Truog R. Current controversies in critical care ethics: not a rapid means of communication, answers are not really just end of life. Crit Care Med. 2003;31:S343. anonymous, which may have prevented some physi- 8. Clarke EB, Curtis JR, Luce JM, et al. Quality indicators for end-of-life care in the cians from participating. intensive care unit. Crit Care Med. 2003;31:2255-2262. 9. Rubenfeld GD, Curtis JR. Improving care for patients dying in the intensive care Nevertheless, this international survey presents an up- unit. Clin Chest Med. 2003;24:763-773. to-date indication of physicians’ attitudes toward end- 10. Cook D, Rocker G, Marshall J, et al. Withdrawal of mechanical ventilation in an- of-life decision making in the ICU. Clearly, these atti- ticipation of death in the intensive care unit. N Engl J Med. 2003;349:1123- tudes differ not only in various regions of the globe but 1132. also within each country. Physicians’ attitudes surround- 11. Vincent JL. Ethical principles in end-of-life decisions in different European countries. Swiss Med Wkly. 2004;134:65-68. ing end-of-life decision making in the ICU remain highly 12. Cook D, Rocker G, Heyland D. Dying in the ICU: strategies that may improve end- variable despite widespread discussion and publication of-life care. Can J Anaesth. 2004;51:266-272. of guidelines and recommendations.26-28 This is one area 13. Angus DC, Barnato AE, Linde-Zwirble WT, et al. Use of intensive care at the end of intensive care medicine in which a global, uniform ap- of life in the United States: an epidemiologic study. Crit Care Med. 2004;32: proach is difficult if not impossible to apply at this time. 638-643. 14. Vincent JL. European attitudes towards ethical problems in intensive care medi- These issues are hard to measure objectively, and it is cine: results of an ethical questionnaire. Intensive Care Med. 1990;16:256- therefore difficult to determine best practice, unlike situ- 264. ations with clearer outcome measures such as the treat- 15. Cook DJ, Guyatt GH, Jaeschke R, et al. Determinants in Canadian health care ment of patients with acute myocardial infarction. Best workers of the decision to withdraw life support from the critically ill. JAMA. 1995; practice in end-of-life care could perhaps be defined in 273:703-708. 16. Cook DJ, Giacomini M, Johnson N, Willms D; Canadian Critical Care Trials Group. terms of the society (expressed through local laws and Life support in the intensive care unit: a qualitative investigation of technologi- social customs and beliefs), the profession (expressed cal purposes. CMAJ. 1999;161:1109-1113. through guidelines and consensus recommendations), or 17. Vincent JL. Forgoing life support in Western European intensive care units: the a more global view (where universal standards are iden- results of an ethical questionnaire. Crit Care Med. 1999;27:1626-1633. 18. Sprung CL, Cohen SL, Sjokvist P, et al. End-of-life practices in European inten- tified and then applied to individual situations). How- sive care units: the Ethicus Study. JAMA. 2003;290:790-797. ever, although it is important to understand why differ- 19. Cardoso T, Fonseca T, Pereira S, Lencastre L. Life-sustaining treatment deci- ences exist, it may not be necessary or even appropriate sions in Portuguese intensive care units: a national survey of intensive care to have an international consensus. Whether the inter- physicians. Crit Care. 2003;7:R167-R175. national critical care community can, or should, work 20. van der Heide A, Deliens L, Faisst K, et al. EURELD consortium. End-of-life decision- making in six European countries: descriptive study. Lancet. 2003;362:345- toward a global consensus on these issues regarding end- 350. of-life care in the ICU remains unclear. Although we be- 21. Azoulay E, Pochard F, Chevret S, et al. Opinions about surrogate designation: a lieve that there needs to be some global consensus on ba- population survey in France. Crit Care Med. 2003;31:1711-1714. sic ethical principles, determining which aspects should 22. Guyatt G, Cook D, Weaver B, et al. Influence of perceived functional and employ- be treated as globally definable and which must be left ment status on cardiopulmonary resuscitation directives. J Crit Care. 2003; 18:133-141. to individual countries, regions, or cultures requires fur- 23. Kavic SM, Atweh N, Possenti PP, Ivy ME. The role of advance directives and fam- ther work. Continuing discussion among critical care phy- ily in end-of-life decisions in critical care units. Conn Med. 2003;67:531-534. sicians, based on results of surveys such as this, is needed 24. Bryce CL, Loewenstein G, Arnold RM, Schooler J, Wax RS, Angus DC. Quality of to better understand the importance of international dif- death: assessing the importance placed on end-of-life treatment in the intensive- ferences in end-of-life decision making in the ICU. care unit. Med Care. 2004;42:423-431. 25. Teno JM, Clarridge BR, Casey V, et al. Family perspectives on end-of-life care at the last place of care. JAMA. 2004;291:88-93. Accepted for Publication: May 17, 2005. 26. Truog RD, Cist AF, Brackett SE, et al. Recommendations for end-of-life care in Correspondence: Jean-Louis Vincent, MD, PhD, Depart- the intensive care unit: the Ethics Committee of the Society of Critical Care Medicine. ment of Intensive Care Medicine, Erasme Hospital, Free Crit Care Med. 2001;29:2332-2348. University of Brussels, Route de Lennik 808, B-1070 Brus- 27. Cohen SL, Bewley JS, Ridley S, Goldhill D. Guidelines for limitation of treatment for adults requiring intensive care. Available at: http://www.ics.ac.uk/downloads sels, Belgium ([email protected]). /LimitTreatGuidelines2003.pdf. Accessed July 25, 2005. Financial Disclosure: None. 28. Swiss Academy of Medical Sciences. Medical-ethical guidelines on borderline questions in intensive-care medicine. Available at: http://www.samw.ch/content /Richtlinien/e_Intensivmedizin.pdf. Accessed July 25, 2005. REFERENCES 29. Miccinesi G, Fischer S, Paci E, et al. Physicians’ attitudes towards end-of-life decisions: a comparison between seven countries. Soc Sci Med. 2005;60:1961- 1. Deliens L, Mortier F, Bilsen J, et al. End-of-life decisions in medical practice in 1974. Flanders, Belgium: a nationwide survey. Lancet. 2000;356:1806-1811. 30. Young RJ, King A. Legal aspects of withdrawal of therapy. Anaesth Intensive Care. 2. Gajewska K, Schroeder M, de Marre F, Vincent JL. Analysis of terminal events in 2003;31:501-508. 109 successive deaths in a Belgian intensive care unit. Intensive Care Med. 2004; 31. Batlle JC. Legal status of physician-assisted suicide. JAMA. 2003;289:2279-2281. 30:1224-1227. 32. Prendergast TJ, Claessens MT, Luce JM. A national survey of end-of-life care 3. Carlet J, Thijs LG, Antonelli M, et al. Challenges in end-of-life care in the ICU: for critically ill patients. Am J Respir Crit Care Med. 1998;158:1163-1167.

(REPRINTED) ARCH INTERN MED/ VOL 165, SEP 26, 2005 WWW.ARCHINTERNMED.COM 1975

Downloaded From: https://jamanetwork.com/ on 09/25/2021