Journal of Critical Care xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Journal of Critical Care

j ourna l homepage: www. jcc journa l .o rg

The Durban World Congress Ethics Round Table IV: Health careprofessional end-of-life decision making

Gavin M. Joynt, MBBCh, FFA(SA)(Crit Care), FHKCA (IC), FCICM a,⁎,Jeffrey Lipman, MBBCh, DA, FFA(SA)(Crit Care), FCICM, MD b, Christiane Hartog, MD c, Bertrand Guidet, MD d,Fathima Paruk, MBBCh e, Charles Feldman, MBBCh, DSc, PhD, FRCP, FCP (SA) f,Niranjan Kissoon, MD g, Charles L. Sprung, MD, MCCM, FCCP h

a Department of Anaesthesia and Intensive Care, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, NT, Hong Kongb Royal Brisbane and Women's Hospital, Herston, Australiac Center for Sepsis Control and Care, University of Jena, Jena, Germanyd Hôpital Saint-Antoine, Paris, Francee University of the Witwatersrand, Johannesburg, South Africaf Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africag BC Children's Hospital, University of British Columbia, Vancouver, Canadah Hadassah Hebrew University Medical Center, Jerusalem, Israel

a b s t r a c ta r t i c l e i n f o

⁎ Corresponding author. Tel.: +852 26322738; fax: +8E-mail address: gavinmjoynt@cuhk.edu.hk (G.M. Joyn

http://dx.doi.org/10.1016/j.jcrc.2014.10.0110883-9441/© 2014 Elsevier Inc. All rights reserved.

Please cite this article as: Joynt GM, et al, TheJ Crit Care (2014), http://dx.doi.org/10.1016/

Keywords:

SurrogateIntensive careEthicsFutilityTerminal illnessDefinitions

Introduction: When terminal illness exists, it is common clinical practice worldwide to withhold (WH) or with-draw (WD) life-sustaining treatments. Systematic documentation of professional opinion and perceived practicesimilarities and differences may allow recommendations to be developed.Materials and methods: Speakers from invited faculty of the World Federation of Societies of Intensive and CriticalCare Medicine Congress that took place in Durban (2013), with an interest in ethics, were approached to partici-pate in an ethics round table. Key domains of health care professional end-of-life decision making were defined,

explored by discussion, and then questions related to current practice and opinion developed and subsequentlyanswered by round-table participants to establish the presence or absence of agreement.Results: Agreement was established for the desirability for early goal-of-care discussions and discussions betweenhealth care professionals to establish health care provider consensus and confirmation of the grounds forWH/WD,before holding formal WH/WD discussions with patients/surrogates. Nurse and other health care professionalinvolvement were common in most but not all countries/regions. Principles and practical triggers for initiatingdiscussions on WH/WD, such as multiorgan failure, predicted short-term survival, and predicted poor neurologicoutcome, were identified.Conclusions: There was majority agreement for many but not all statements describing health care professionalend-of-life decision making.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

Approximately 10% to 30% of all emergency admissions to inten-sive care units (ICUs) will die in the ICU; [1-3] and, therefore, manage-ment of the dying process is a necessary skill for all ICU health careprofessionals. Life-support technology has advanced, so that it isnow possible to maintain vital organ function, despite the realizationthat a return of the patient to reasonable health and an acceptablequality of life are no longer possible. When a return to reasonablehealth is no longer possible, it has become common clinical practiceworldwide for ICU staff to limit life-sustaining treatments (LSTs) bywithholding (WH) or withdrawing (WD) LSTs [4-12]. There are,

52 26372422.t).

DurbanWorld Congress Ethicj.jcrc.2014.10.011

however large differences in the practice of WH/WD in differentparts of the world as evidenced by responses to several surveysthat have explored physicians' practice and ethical views on thesubject [13-20]. This self-reported variation in the pattern of practiceof health care practitioners has been confirmed by observationalstudies that also demonstrate regional variations in practice [4-12].Although guidelines for some jurisdictions do exist, [21-25] and areoften useful, they sometimes lack sufficient detail to guide dailypractice and seldom consider the variability inevitably introducedby cultural and regional differences. Although cultural and social dif-ferences are slow to evolve, people in the modern world migrate inlarge numbers, with ease and over great distances. It is thus impor-tant that ICU health care professionals are aware of similarities anddifferences in international end-of-life practice and expectations[26-28].

s Round Table IV: Health care professional end-of-life decisionmaking,

Table 1Participant views on key issues to be addressed before holding aWH/WD discussion withthe family, including early goal-of-care discussions

Agree Neutral Disagree All

1 Before WH/WD discussions with the family1A Initial discussion—explore patient WH/WD

preferences17 1 3 21

1B Initial discussion—explore patient values,goals on WH/WD

18 2 2 22

1C Hold discussions with the primary carephysician

19 3 0 22

1D Discussions with other intensivists 20 1 1 221E Discussions with nursing staff 19 2 1 221F Discussions with both physicians and

nurses in ICU19 2 1 22

2 Goal-of-care discussion2A Early goal-of-care discussion—general

medical issues17 3 2 22

2B Early goal-of-care discussion—info onWH/WD issues

8 8 6 22

WH/WD discussions should occur inresponse to:

3 Patient/family requests (including advancedirectives)

19 0 3 22

4 Physician requests 20 1 1 225 Nurse requests 16 3 3 22

Agree Neutral Disagree All8 Reported participation ICU nurses in

ICU rounds18 2 2 22

9 Routine formal discussions on WH/WDwith nurses

16 0 6 22

2 G.M. Joynt et al. / Journal of Critical Care xxx (2014) xxx–xxx

The decision to initiate the process of WH/WD for a particularpatient is a complex one and one that involves several stakeholders.To begin with, it requires recognition that WH/WD is potentiallyjustified, followed by a process to confirm the appropriateness ofWH/WD. Lastly, it must be demonstrated that the decision to pro-ceed is in the best interests of the patient himself/herself. Thisarticle will focus specifically on the initial part of the end-of-lifeprocess, that is, making the decision to proceed to WH/WD discus-sions with the family. With information drawn from the results ofan international meeting of experienced clinicians with an intereston end-of-life care, we have documented similarities and differencesin the way the above process is managed around the world, andwhere sufficient similarities were shown to exist, have developed state-ments of general agreement.

2. Methods

Speakers from the invited faculty list of the World Federation ofSocieties of Intensive and Critical Care Medicine Congress, held inDurban September 2013, were approached by the conference scien-tific convener to participate in the ethics round table. All round tableparticipants had a known interest in ethics. Round table participantswere asked to identify what in their opinion constituted the 3 mostpressing specific worldwide ethical issues that the group shouldaddress. Most responded that end-of-life issues, including WH andWD LSTs constituted the most important issue. Seventy questionsrelated to end-of-life care in ICU were sent to participants.

Before the round table discussion held in Durban, participants wereasked questions by e-mail circulation to determine local practice in theirhospitals and countries, in relation to the clinical practice of end of life.There were 20 responses with similarities and variations from differentcountries that seemed themost interesting (decision by author CS). Thesummary of these responses was sent to all participants by e-mail be-fore the face-to-face meeting. Respondents were asked to identify themost important and relevant topics from this list and to then preparefor face-to-face discussion at the Congress. The 5 topics with thegreatest clinical importance and differences between centers and coun-tries were chosen by agreement at the face-to-face round table discus-sion. They included questions related to WH and WD LSTs at the endof life for age, health care professional end-of-life decision making,patient/family end-of-life decision making, how to withdrawmechani-cal ventilation, and differences betweenWH andWD LSTs. The focus ofthis article is to describe the process of health care professional end-of-life decisionmaking. To further define key issues in this process, discus-sion in Durban identified the following key questions related to thefocus of this manuscript:

(1) Is there a need for consensus by any, some, or all stakeholderswhen making decisions to proceed to WH/WD?

(2) If consensus is required, is there or should there be a specificsequence by which ultimate consensus is achieved among allstakeholders?

(3) Should goal-of-care discussions routinely take place, and whenshould they be initiated?

(4) Which parties are or should be responsible for the initiation,discussion, and decision-making processes?

(5) What “general principles” justify the decision to considerWH/WDlife support?

(6) What “specific clinical factors” trigger decisions to proceed todiscussions and decisions on WH/WD?

After themeeting, questionswere developed to provide an overviewof current practice and opinion related to the questions. Questionsweretested for validity by giving all participants an opportunity to commenton, add, or change questions (using e-mail). This resulted in a final set ofquestions that were answered by round table participants. A total of 76questions were finalized, using a Likert scale format (strongly agree,

Please cite this article as: Joynt GM, et al, The DurbanWorld Congress EthicJ Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.10.011

agree, neutral, disagree, and strongly disagree), with a free-text optionsfor comment. The answers that specifically relate to the stated questionsthat are the focus of this report are section II, health care professionalend-of-life decision making. Questions from this section have beenrenumbered from 1 to 30 for clarity (Appendix 1).

3. Results

A list of round table participants can be found at the end of themanuscript. The final set of questions that was developed and thenanswered by e-mail circulation can be found in Appendix 1, andresults are presented in the order of the goals set out in the methods.The responses to the final questions developed by the round tablegroup are shown in Tables 1 to 3. To simplify the presentation ofresults, the Likert scale format (strongly agree, agree, neutral,disagree, and strongly disagree) was simplified by grouping stronglyagree and agree and strongly disagree and disagree. Results toquestions are presented in order, corresponding to the key questionsstated in the methods.

(1) At the original round table discussion, it was evident that therewere differences in practice relating to the need for consensusamong professional health care providers before engaging thefamily in discussions. Therefore, to characterize the processbefore a formal family/surrogate discussion, the followingconsiderations were explored: question 1 explored the possi-bility of discussing patient preferences regarding WH/WD ingeneral (statement A) or general values, goals, and preferenceswithout specific mention of WH/WD (statement B). Table 1documents participant responses on these key issues to beaddressed before holding a WH/WD discussion with the familyand favored the need to explore patient preferences and valuesbefore WH/WD discussions.

(2) Question 1 (statements C-F) explored the requirement, if any, forreaching a medical consensus before having a WH/WD discus-sion with the family. Participant responses are documented inTable 1 and generally favored the need to achieve medical con-sensus before family discussion.

s Round Table IV: Health care professional end-of-life decisionmaking,

Table 2Statements of general principles underlying the WD and WH of life support

Principles—justification for triggeringWH/WD

Agree Neutral Disagree All

Principles justifying WH/WD10A Patient's best interest not served by LST 19 0 3 2210B Insufficient net medical benefit gained

by LST21 1 0 22

The above conditions are met when:10C.1 With LST b3-month survival 12 3 7 2210C.2 With LST b1-month survival 14 5 3 2210C.3 With LST b2-week survival 17 2 3 2210C.4 With LST less than a few days' survival 21 1 0 2210C.5 ICU burden outweighs likely length of

time of survival18 3 1 22

10C.6 ICU burden outweighs likely futurequality of life

17 4 1 22

3G.M. Joynt et al. / Journal of Critical Care xxx (2014) xxx–xxx

(3) Question 2 explored the important option of implementing anearly goal-of-care discussion, andwhether this discussion shouldspecifically include information relating to WH/WD of lifesupport. Results in Table 1 generally supported the need forearly goal-of-care discussions.

(4) Questions 3 to 5 explored responses on who should trigger aWH/WD discussion, and questions 6 to 9 explored the mecha-nisms by which nurses were included in the consensus process,if at all. A general but not unanimous agreement existed toinvolve other health care workers in WH/WD discussions(Table 1). The free-text answers to question 6 were collated. Inanswer to question 6, most participants stated that nurses'opinions were routinely solicited on ward rounds (n = 15).Three participants reported the use of formal multidisciplinarymeetings as the mechanism for inclusion of nurses in thedecision-making process. One participant (from Asia) indicatedthat nurses had no role in the WH/WD process. In answer toquestion 7, the communication process used to achieve consen-sus with various parties, including nurses and other health careprofessionals was varied. Mechanisms included informal

Table 3Participant's reported likelihood of usng specific practical triggers that may contribute tothe decision to discuss WH/WD LST

Practical triggers for WH/WDdecisions

Alwaystriggers

Usuallytriggers

Nevertriggers

All

11 Known advance directive 14 7 0 2112 Family request 16 4 2 2213 Irreversible condition 10 9 3 2214 Unsurvivable injury 18 4 0 2215 Brain injury—unclear outcome 8 11 3 2216 Brain injury—poor outcome

and dependent10 7 5 22

17 Brain injury—minimallyconscious state

17 3 2 22

18 MOF—1-2 organs 0 6 16 2219 MOF ≥3 organs for ≥3 d 4 9 9 2220 MOF ≥3 organs for ≥7 d 7 12 3 2221 Nonbeneficial therapy 14 7 0 2122 Poor ICU survival 8 14 0 2223 Terminal illness 15 3 4 2224 Poor quality of life 4 14 4 2225 Severe illness 1 6 15 2226 Age N80 y 3 7 11 21

Yes No All27 Are there universally

accepted triggers?14 8 22

Stronglyagree

Neutral Stronglydisagree

Total

28 Should we use triggersas above?

18 1 2 21

MOF, multiple organ failure.

Please cite this article as: Joynt GM, et al, The DurbanWorld Congress EthicJ Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.10.011

discussions on ward rounds, formal scheduled multidisciplinarymeetings, informal “huddles” of involved staff, telephone consul-tations with relevant individuals, and e-mail communications.

(5) Question 10 explored responses to statements describing thegeneral principles supportingWH/WD decisions and that justifytriggering a family conversation. Views on relevant general prin-ciples underlying WH/WD and justifying the reason for the dis-cussion of WH/WD with the family are reported in Table 2.

(6) Questions 11 to 28 explored responses for specific clinical factorsthat may potentially trigger the decision to discuss WH/WD.Table 3 shows participant views of their likely use of specificpractical triggers to inform the decision to discuss WH/WD LST.

All disagreements were examined to establish whether they wereaccompanied by a free-text explanation. Notable free-text commentsaccompanying responses that were considered outliers are summarizedbelow and grouped by region:

• Participants from China indicated strongly that nurses took no part inend-of-life decisions. Participants from Korea indicated that nurseopinions were very rarely expressed. Furthermore, that the elderlyseldom discussed end-of-life issues and as a result of advance direc-tives were almost never available.

• Regarding early goal-of-care discussions, most participants indicatedthat they were desirable; however, participants from China andKorea more strongly indicated that the issue of WH/WD should beavoided in the early stages of admission.

• Participants from the Middle East indicated that local laws, hospitalpolicy, and religious issues may restrict end-of-life care practice andshould be taken into consideration.

Other relevant free-text explanations that explained specific dis-agreements are provided in the discussion. In answer to question 29,all but one of the participants reported working in a “closed” ICU envi-ronment, where the resident specialist intensivists carried ultimate re-sponsibility for patient care in the ICU. There were no specificresponses to question 30 indicating the need for the addition of furtherquestions. A list of generally agreed statements, derived from thedeliberations, is provided in Table 4.

4. Discussion

Broad agreement existed for the principle that, before aWH/WDdis-cussion with the family/surrogate, discussions with relevant intensivecare physicians, primary care physicians, and nurses should take place.The intention of these discussions is to establish consensus that a WH/WD conversationwith the family is appropriate, and decisions to subse-quently initiateWH/WDwould be reasonable. This approachmay servesome important purposes. Firstly, prognostic certainty is difficult toachieve [29]; and, therefore, a broad agreement by competent andknowledgeable medical staff that prognosis justifies WH/WD is onemethod to enhance diagnostic accuracy, improve the sense of securityin the diagnosis, and reduce stress [30]. Secondly, it may increase theobjectiveness of the opinion when it is made by ICU physicians otherthan the attending physician alone [31]. Thirdly, it should ensure thatsurrogates receive a consistent message regarding WH/WD and avoidsthe potential risk of treatment team disagreements being communicat-ed to the family, potentially causing confusion and loss of trust [32].Obtaining treatment team consensus before the initiation of WH/WDdiscussion with the patient/surrogate was considered important. Inter-estingly, although somepublished guidelines imply the need for a broadmedical consensus during the process of WH/WD [30,33], others makeno reference to this aspect of decision making [21]. It has also beenshown that lack of agreement on end-of-life decisions between the pri-mary ICU medical team and other involved physicians may occur in upto 30% of cases and in up to 20% of cases with nurses [34]. It should benoted that 1 participant disagreed with the need for nursing consensus,

s Round Table IV: Health care professional end-of-life decisionmaking,

Table 4Durban (2013) Round Table—statements of general agreement

1. Knowledge of local, cultural, and religious practice and expectations as well aslocal legal restrictions should inform individual practice. These considerationsare applicable worldwide but may be particularly relevant in the Middle East andAsia and when treating patients originating from these regions.

2. Goal-of-care discussions should generally occur early (within 48 h) after thepatient's admission to ICU. Care should be taken to specifically address WH/WD inappropriate circumstances, for example, in patients with a very poor prognosis, orif the surrogate/patient requests such a discussion. Special sensitivity should beexercised when interviewing certain patients/surrogates, eg, in China and Korea,where early end-of-life discussions may be considered inappropriate.

3. Physicians are the most likely to initiate discussions that may lead to WH/WDdecision-making processes. It should, however, also be recognized that the wishesof patients themselves through advance directives and families through verbal orwritten requests have the right to trigger WH/WD discussions. Nurses and otherhealth care workers may trigger discussions among the medical health care teamthat may lead to WH/WD decision-making processes. In some parts of the world,eg, China and Korea, nurse involvement in end-of-life decisions is not expected.

4. When possible, treating ICU physicians and other specialty medical physiciansshould reach consensus that a WH/WD discussion is reasonable before formallystarting a discussion with the patient/surrogate. In most parts of the world,nurses and other health care staff who are meaningfully involved in the patient'slong-term care should also be consulted. In some parts of the world, eg, Chinaand Korea, nurse involvement in end-of-life decisions is not expected.

5. Consensus regarding the decision to trigger a WH/WD discussion is usuallyreached by discussion between relevant health caregivers on routine daily wardrounds. Additional formal discussion forums can be scheduled if necessary, forexample, in difficult or contested cases.

6. The decision to trigger a discussion of WH/WD LST should be made based on theprinciple that the patient's best interest is no longer served by LST, that occurswhen sufficient net medical benefit cannot be achieved by ongoing LST (ie,nonbeneficial therapy). This condition is met when the burden of ongoingtreatment outweighs the net health benefit that the patient is likely to derive interms of length of time of survival, or quality of life. Quality of life should bedetermined by appropriate patient/surrogate consultation.

7. Although no universally agreed triggers should be prescribed, practically, thefollowing may serve as triggers for initiating WH/WD discussions with thepatient/surrogate. A WH/WD discussion:a. May be considered when survival would be expected to be b1-3 mo.b. Should be considered when survival would be expected to be b2 wk.c. Should always be considered when survival would be expected to be less thana few days.d. Should usually be considered when multiple organ failure (three or moreorgans for 7 or more days, despite therapy) existse. Should usually be considered when very severe brain injury (with poor longterm outcome eg, minimally conscious state) existsf. Should always be considered when, in the health care team's opinion:

i. The patient is receiving nonbeneficial therapyii. The patient has a nonsurvivable injury

8. Neither age alone, nor severe illness should be considered a sufficient trigger toinitiate WH/WD discussions.

4 G.M. Joynt et al. / Journal of Critical Care xxx (2014) xxx–xxx

with the explanation that, in his country in Asia, nurses did not partici-pate in WH/WD decisions and would not expect to participate in suchdecisions. A member from Asia also expressed a physician consensusdecision would reduce the chance of error and legal challenge, aknown concern for ICU clinicians in certain Asian countries [20].

It was generally agreed that early goal-of-care discussions (within24-48 hours of admission) focused on prognosis, and potentialoutcomes were appropriate. The second statement explored thedesirability of specifically discussing WH/WD at this time. Commentsexplaining disagreement with this approach included the need toavoid an unnecessary discussion of WH/WD in patients with a goodprognosis. Early goal-of-care discussions have been increasingly investi-gated and recommended as a desirable practice [30,35,36]. Althoughthe timing and nature of goal-of-care discussions resulted in someinteresting and diverse responses, most believed that such conversa-tions should take place early, although several members would limitsuch meetings to only those with a predicted poor outcome. Themajor concern was the perceived potential to undermine patient orfamily confidence by discussing WH/WD too early. This may be particu-larly true inAsian countries,where discussionof deathmaybe consideredrude or even dangerous by inducing future “bad luck” [37] and

Please cite this article as: Joynt GM, et al, The DurbanWorld Congress EthicJ Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.10.011

undermine the filial duty of children to protect elders and parents [38].Others indicated a benefit to early meetings discussing goal-of-careexpectations, even if this is to confirm a hopeful prognosis. Subsequentgoal-of-care conversations may be then used to discuss WH/WD issuesandmay benefit subsequent communication by the earlier establishmentto good relationships with the patients/surrogates.

It was agreed that aWH/WDdiscussionwith the family should followa request from the patient, surrogate, or if indicated in an advance direc-tive. Few disagreements by round table members with the statementwere recorded, and it was noted in corresponding written explanationthat requests from family or nurses should only trigger discussions withthe surrogates, with an intention to proceed to WH/WD, if supported byphysician consensus. Thus, the appropriate response to all requests fromall stakeholders should be a health caregiver discussion, with the goal ofevaluating the merits of the request, before a formal WH/WD discussionwith the family.

The usual method for reaching health caregiver consensus wasface-to-face discussion, rather than phone calls, formal meetings,or written consultation. This discussion invariably took place onscheduled morning meetings or ward rounds and included physi-cians, nurses, and occasionally, other involved parties, such as physio-therapists and social workers. The exception was Mainland China,where the discussion was reported to only involve physicians. Someunits reported the use of informal discussions or ad hoc dedicatedmeetings to discuss goals of care, particularly in difficult cases. No partici-pant mentioned the need to consult an ethics committee at this stage.

There was general but not universal agreement for the principlesthat underlie the decision to consider WH/WD life support. Specifically,most participants agreed that, in “The treating physician's opinion, thepatient's best interest is no longer served by ongoing LST” was a basisfor initiating WH/WD discussions. Comments accompanying thedisagreeing responses stressed that subsequent decisions to take actionshould always follow discussions with surrogates and the wider healthcare team and not only result from the opinions of the treating physi-cian, whose religious or cultural values might be different than thepatient's. Disagreement in 1 case was explained by an expression of“legal concerns” relating to limiting LST in an Asian country. The state-ment “The patient's best interest is no longer served by LST, when suffi-cient net medical benefit cannot be achieved by ongoing LST” receivedstrong support, with only a single expression of disagreement being re-corded. This statement and the conditions under which it would bemetwere considered the basis for determining that treatment was“nonbeneficial” and could therefore trigger discussions toWH/WD LST.

To try to develop a practical expression of what the above statementof “sufficient netmedical benefit”maymean in terms of length or qualityof survival, members were asked to respond to some statements quanti-fying length of survival. Members were progressivelymore likely to con-sider length of time of survival as a trigger to initiate WH/WDdiscussions, as the patient's predicted length of survival decreased. Nomembers disagreed that a predicted length of survival less than a fewdays would meet the principle requirement to trigger a WH/WD discus-sion. This trend would seem to reflect the likely wishes of patients. Pre-vious surveys have indicated that a majority of patients would acceptICU care for survival times of as little as 1 month [39]. It is important torecognize that, in individual cases, “after” aWH/WDdiscussionwith sur-rogates/patients, continuing LST to reach agreed goals of care may re-main justified, despite only a few days of anticipated survival. This isbecause net medical benefit implies a cost and a benefit assessment;and in each case, the “cost” of ICU caremust beweighed against themed-ical benefit gained. Length of survival only quantifies potential benefit,the cost being the individual patient's perception of the burden of ICUcare. The patient's and surrogate's view of the burden of ICU caremay re-late tomany factors such as current and future physical pain and discom-fort, mental anguish, and evenmonetary cost in some instances. Lengthysurvival with a poor quality of life may be realistically not considered tobe a benefit by individual patients. Nevertheless, these factors can only

s Round Table IV: Health care professional end-of-life decisionmaking,

5G.M. Joynt et al. / Journal of Critical Care xxx (2014) xxx–xxx

be reliably ascertained by the initiation of an open and honest goal-of-care discussion. In free-text responses, the opinion was repeatedlyexpressed that this principle of balancing cost and burden should ulti-mately be confirmed by the patient's or surrogate's expression of theirperception of the “net balance of benefit.” There was strong agreementby participants with the statements that health burdens should beweighed against benefit in terms of both length of time of survival andexpectation of future quality of life.

In an attempt to provide some practical guidance, a number oftriggers that may provide grounds for initiating aWH/WD discussionwere proposed. To the authors' knowledge, this is the first time thatspecific clinical triggers have been suggested and their potential ac-ceptability to clinicians assessed. Multiorgan failure (involving ≥3organs and 1-week duration) and severe brain injury (especially pre-dicted minimally conscious state or neurologically dependent out-come) were considered reasonable triggers by the large majority.There is evidence for increased mortality in patients meeting theabove organ failure criteria [40-42]. Despite evidence that the inci-dence of WH/WD is independently associated with advancing age,[43,44] age alone even when relatively extreme (N80 years old)was not considered an appropriate trigger by most round tablemembers. This view by the round table group may be supported byrecent data indicating a good self-reported quality of life by elderlyICU survivors [45,46]. Advance directives and general assessments,such as unsurvivable injury, nonbeneficial therapy, and poor ICU sur-vival were universally accepted as reasonable triggers. Severe illnessalone was not considered an appropriate reason to trigger a WH/WDdiscussion, as potential illness reversibility was considered moreimportant. In summary, there was general agreement supportingthe use of several practical triggers; however, it was agreed that uni-versally acceptable triggers would be difficult to identify, and the listdeveloped at this round table meeting should serve as a guide only.The round table statements of general agreement identify commontriggers and provide different indication grades (may, should usually,and should always) for the use of common triggers, guided by thedegree of agreement achieved.

All participants indicated that they would usually or always imple-ment end-of-life discussions in response to advanced directives. Itshould, however, be recognized that, although advance directives arelegally regulated in some countries (eg, United States, United Kindom,Australia, Netherlands, Germany, Spain, Singapore, and India), theyare generally respected but not legally regulated in others (eg, Italy,Switzerland, Hong Kong, Japan, and South Korea); and, thus, levels ofimplementation may vary [47-50].

In the free-text responses, members fromMiddle Eastern and Asianregions expressed the need to be aware of local, religious, and legal con-straints when considering WH/WD discussions. Nevertheless, many ofthe processes described above were generally accepted as reasonableby all participants; and a table with agreed statements that some mayfind useful is provided (Table 4).

The report has several limitations. The major limitation of this studyis the composition of the round table group. Although participants wereinvited based on regional diversity, known interest in ethics, and per-ceived ability to represent their regions, some parts of the world andmany cultures were not represented. Most participants representedICUs of large, teaching hospitals. It is therefore likely that the practiceand views of some parts of the world and those in smaller, local, or re-gional units were not fully represented. Nevertheless, both similaritiesand variations in practice and opinion were identified. The final recom-mendations therefore acknowledge both outlying practice and opinionas well as emphasize agreed principles and practice as clearly as possi-ble. In addition, participant demographic data were not analyzed. Theround table discussionwas also not systematic a priori but concentratedon what were considered major issues. Thus, not all possible aspects ofdecision-making triggers were discussed at length, examples being theinfluence of "inability to continue to pay" by individuals in countries

Please cite this article as: Joynt GM, et al, The DurbanWorld Congress EthicJ Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.10.011

where intensive care is provided on a strictly pay for service model, orthe principle of distributive justice and its use as an ethical principleto justify decisions at end of life.

Although no systematic attemptwasmade to reach agreement on thislast issue, consideration of the use of the ethical principle of distributivejustice at the end of life raises several challenges. The moral principle ofdistributive justice implies that there is possible justification for prioritiza-tion or triage at the end of life. In ICU-related literature on end of life,although the resource cost of prolonged and “low-benefit” life supporthas been challenged as being inappropriately high, no practical advicehas been suggested as to how prioritization could be applied in this set-ting, nor have observational studies reporting its use been published. Ingeneral, most physicians tend to agree with Heaney et al [51], that clini-cians believe that the physician's first duty is to the patient whom he/she is currently treating and that we should avoid giving the family theimpression that we stop treatment to “free a bed for another patient.”Pronovost and Angus [52] make the valid point that, in the absence ofsuitable methods to calculate economical basis for resource allocation,such considerations will remain limited. Recently, however, as part of asystem of “triage” or prioritization during infectious outbreaks, it hasbeen suggested that patients not responding positively to ICU care andwith resultant poor prognosis should have life supportwithdrawn by dis-charge from the ICU [53]. Under these extreme circumstances of resourcestress, it appears that prioritization of resources to maximize overall re-source use at end of life may be considered acceptable. It could be arguedthat, in some poorly resourced countries, such a level of resource stressexists continuously and that routine triage could be justified and maybepracticed; however, no current literature exists documenting this prac-tice in resource-limited countries. Importantly, it should be rememberedthat if systematically practiced, under such circumstances, although over-all societal resource utilization ismaximized, personal choice at end of lifeis limited by the implementation of prioritization [54]. Early goal-of-carediscussions or use of advance directives, when personal choice of rejec-tion of unwanted burdens of prolonged ICU care may be expected to re-sult in earlier WH/WD, also results in conservation of scarce ICUresources [48]. In this case, however, it is an indirect benefit and not theprimary goal. Therefore, it is acknowledged, even in highly resourcedcountries, that medical resource consumption at the end of life cannotbe unlimited, [55], and the reduction of health care costs at the end oflife by early palliative care consultation has been shown to be associatedwith cost savings [56,57]. However, this approach would ultimately besuccessful at reducing costs in the ICU setting, and how it could be imple-mented, is currently unclear [58,59].

In conclusion, although there was majority agreement for severalstatements when making decisions to proceed to WH/WD, there wassome disagreement, particularly by members representing Asia andthe Middle East. This was particularly evident with regard to nursingparticipation in decision making, concerns regarding the legal status ofWH/WD, and the effect of this on initiating discussions. Although allparticipants represented ICUs that practiced WH/WD, it was notedthat, in certain regions of the world that may have had no representa-tion, cultural or religious norms may directly influence the process ofWH/WD and may in some places prohibit it entirely. Nevertheless,where practiced, a number of generally agreed statements to guidehealth care professional end-of-life decision making are provided.

Participants in the Durban World Congress Ethics Round Table:

s

Rou

Country

nd Table IV: H

Name

ealth care pro

Institution

1

Australia Jeff Lipman Royal Brisbane and Women's Hospital,Herston

2

Belgium Jean-LouisVincent

Erasme University Hospital, Brussels

3

Canada NiranjanKissoon

BC Children Hospital, Vancouver

(continued on next page)

fessional end-of-life decisionmaking,

(

6 G.M. Joynt et al. / Journal of Critical Care xxx (2014) xxx–xxx

continued)

PleJ Cr

Country

ase cite this ait Care (2014

Name

rticle as: Joynt), http://dx.do

Institution

4

China Bin Du Peking Union Medical College Hospital, Beijing 5 France Bertrand

Guidet

Hôpital Saint-Antoine, Paris

6

France Elie Azoulay Hôpital Saint-Louis, Paris 7 Germany Christiane

Hartog

University Hospital Jena, Jena

8

Hong Kong Gavin Joynt Prince of Wales Hospital, Shatin 9 Israel Charles Sprung Hadassah Hebrew University Medical Center,

Jerusalem

10 Saudi Arabia Khalid Shukri Pan Arab Critical Care Medicine Society, Riyadh 11 South Africa Eric Hodgson Addington Hospital, Durban 12 South Africa Andrew Argent Memorial Children's Hospital, Cape Town 13 South Africa Fathima Paruk University of the Witwatersrand,

Johannesburg

14 South Africa Rudo Mathivha Chris Hani–Baragwanath Hospital, Soweto 15 South Africa Charles

Feldman

University of the Witwatersrand,Johannesburg

16

South Korea Younsuck Koh Asan Medical Center, Seoul 17 The

Netherlands

A.B.J.Groenveld

Vrije Univerditeit Mecical Centre, deBoelelaan

18

United States Chris Farmer Mayo Clinic, Rochester 19 United States Mitchell Levy Rhodes Island Hospital, Providence 20 United States Janice

Zimmerman

The Methodist Hospital, Houston

21

United States Edgar J.Jimenez

Orlando Regional Medical Center, Orlando

22

United States J. RandallCurtis

University of Washington, Seattle

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcrc.2014.10.011.

References

[1] Azoulay E, Alberti C, Legendre I, Buisson CB, Le Gall JR, European Sepsis Group. Post-ICU mortality in critically ill infected patients: an international study. Intensive CareMed 2005;31(1):56–63.

[2] Society of Critical Care Medicine. Critical Care Statistics. http://www.sccm.org/Com-munications/Pages/CriticalCareStats.aspx. [Accessed May 1, 2014].

[3] Moran JL, Bristow P, Solomon PJ, George C, Hart GK. Mortality and length-of-stayoutcomes, 1993-2003, in the binational Australian and New Zealand intensive careadult patient database. Crit Care Med 2008;36(1):46–61.

[4] Prendergast TJ, Luce JM. Increasing incidence of withholding and withdrawal of lifesupport from the critically ill. Am J Respir Crit Care Med 1997;155(1):15–20.

[5] Sprung CL, Cohen SL, Sjokvist P, Baras M, Bulow HH, Hovilehto S, et al. End of lifepractices in European intensive care units: the Ethicus Study. JAMA 2003;290(6):790–7.

[6] Hynninen M, Klepstad P, Petersson J, Skram U, Tallgren M. Process of foregoing life-sustaining treatment: a survey among Scandinavian intensivists. Acta AnaesthesiolScand 2008;52(8):1081–5. http://dx.doi.org/10.1111/j.1399-6576.2008.01636.x.

[7] Buckley TA, Joynt GM, Tan PY, Cheng CA, Yap FH. Limitation of life support: frequencyand practice in a Hong Kong intensive care unit. Crit Care Med 2004;32(2):415–20.

[8] Piva J, Lago P, Othero J, Garcia PC, Fiori R, Fiori H, et al. Evaluating end of life practicesin ten Brazilian paediatric and adult intensive care units. J Med Ethics 2010;36(6):344–8. http://dx.doi.org/10.1136/jme.2009.035113.

[9] Turner JS, Michell WL, Morgan CJ, Benatar SR. Limitation of life support: frequencyand practice in a London and a Cape Town intensive care unit. Intensive Care Med1996;22(10):1020–5.

[10] Hariharan S, Moseley HS, Kumar AY, Walrond ER, Jonnalagadda R. Futility-of-caredecisions in the treatment ofmoribund intensive care patients in a developing coun-try. Can J Anaesth 2003;50(8):847–52.

[11] Mani RK, Mandal AK, Bal S, Javeri Y, Kumar R, Nama DK, et al. End of life decisions inan Indian intensive care unit. Intensive Care Med 2009;35(10):1713–9. http://dx.doi.org/10.1007/s00134-009-1561-x.

[12] Iyilikçi L, Erbayraktar S, Gökmen N, Ellidokuz H, Kara HC, Günerli A. Practices ofanaesthesiologists with regard to withholding and withdrawal of life support fromthe critically ill in Turkey. Acta Anaesthesiol Scand 2004;48(4):457–62.

[13] Vincent JL. Forgoing life support in western European intensive care units: the re-sults of an ethical questionnaire. Crit Care Med 1999;27(8):1626–33.

[14] Yap HY, Joynt GM, Gomersall CD. Ethical attitudes of intensive care physicians inHong Kong: questionnaire survey. Hong Kong Med J 2004;10(4):244–50.

[15] YaguchiA, TruogRD, Curtis JR, Luce JM, LevyMM,Mélot C, et al. International differencesin end of life attitudes in the intensive care unit: results of a survey. Arch Intern Med2005;165(17):1970–5.

GM, et al, The DurbanWorld Congress Ethici.org/10.1016/j.jcrc.2014.10.011

[16] Bito S, Asai A. Attitudes and behaviors of Japanese physicians concerningwithholdingand withdrawal of life-sustaining treatment for end of life patients: results from anInternet survey. BMCMed Ethics 2007;8:7. http://dx.doi.org/10.1186/1472-6939-8-7.

[17] Society of Critical Care Medicine. Attitudes of critical care medicine professionalsconcerning forgoing life-sustaining treatments. The Society of Critical Care MedicineEthics Committee. Crit Care Med 1992;20(3):320–6.

[18] Prendergast TJ, Claessens MT, Luce JM. A national survey of end of life care for criti-cally ill patients. Am J Respir Crit Care Med 1998;158(4):1163–7.

[19] Nelson JE, Angus DC, Weissfeld LA, Puntillo KA, Danis M, Deal D, et al. Critical CarePeer Workgroup of the Promoting Excellence in End of life Care Project. End of lifecare for the critically ill: a national intensive care unit survey. Crit Care Med 2006;34(10):2547–53.

[20] Weng L, Joynt GM, Lee A, Du B, Leung P, Peng J, et al. Attitudes towards ethical prob-lems in critical care medicine: the Chinese perspective. Intensive. Intensive CareMed 2011;37(4):655–64. http://dx.doi.org/10.1007/s00134-010-2124-x.

[21] Truog RD, Campbell ML, Curtis JR, Haas CE, Luce JM, Rubenfeld GD, et al. Recommen-dations for end of life care in the intensive care unit: a consensus statement by theAmerican College of Critical Care Medicine. Crit Care Med 2008;36(3):953–63.http://dx.doi.org/10.1097/CCM.0B013E3181659096.

[22] General Medical Council (UK). Treatment and care towards the end of life: goodpractice in decision making. http://www.gmc-uk.org/guidance/ethical_guidance/end_of_life_care.asp; 2010. [Accessed May 2, 2014].

[23] Health Professions Council of South Africa. Guidelines for the withholding andwithdrawing of treatment2nd ed. ; 2007 [http://www.hpcsa.co.za/downloads/conduct_ethics/rules/guidelines_withholding_withdrawing_treatment.pdf.Accessed May 2, 2014].

[24] Gruppo di studio ad hoc della Commissione di Bioetica della SIAARTI. SIAARTI guide-lines for admission to and discharge from intensive care units and for limitation oftreatment in intensive care. Minerva Anestesiol 2003;69(3):101–18.

[25] Mani RK, Amin P, Chawla R, Divatia JV, Kapadia F, Khilnani P, et al. Guidelines for endof life and palliative care in Indian intensive care units' ISCCM consensus ethicalposition statement. Indian J Crit Care Med 2012;16(3):166–81.

[26] Pochard F, Azoulay E, Chevret S, Vinsonneau C, Grassin M, Lemaire F, et al. Frenchintensivists do not apply American recommendations regarding decisions to forgolife-sustaining therapy. Crit Care Med 2001;29(10):1887–92.

[27] Engelhardt Jr HT. Critical care: why there is no global bioethics. Curr Opin Crit Care2005;11(6):605–9.

[28] Bülow HH, Sprung CL, Reinhart K, Prayag S, Du B, Armaganidis A, et al. The world'smajor religions' points of view on end of life decisions in the intensive care unit.Intensive Care Med 2008;34(3):423–30.

[29] Meadow W, Pohlman A, Frain L, Ren Y, Kress JP, Teuteberg W, et al. Power and lim-itations of daily prognostications of death in the medical intensive care unit. CritCare Med 2011;39(3):474–9. http://dx.doi.org/10.1097/CCM.0b013e318205df9b.

[30] Davidson JE, Powers K, Hedayat KM, Tieszen M, Kon AA, Shepard E, et al. AmericanCollege of Critical Care Medicine Task Force 2004-2005, Society of Critical CareMedicine. Clinical practice guidelines for support of the family in the patient-centeredintensive care unit: American College of Critical Care Medicine Task Force 2004-2005.Crit Care Med 2007;35(2):605–22.

[31] Thompson BT, Cox PN, Antonelli M, Carlet JM, Cassell J, Hill NS, et al. Challenges inend of life care in the ICU: statement of the 5th International Consensus Conferencein Critical Care: Brussels, Belgium, April 2003: executive summary. Crit Care Med2004;32(8):1781–4.

[32] Curtis JR, Patrick DL, Shannon SE, Treece PD, Engelberg RA, Rubenfeld GD. The familyconference as a focus to improve communication about end of life care in the inten-sive care unit: opportunities for improvement. Crit Care Med 2001;29(2 Suppl):N26–33.

[33] Carlet J, Thijs LG, Antonelli M, Cassell J, Cox P, Hill N, et al. Challenges in end of lifecare in the ICU. Statement of the 5th International Consensus Conference in CriticalCare: Brussels, Belgium, April 2003. Intensive Care Med 2004;30(5):770–84.

[34] Cohen S, Sprung C, Sjokvist P, Lippert A, Ricou B, Baras M, et al. Communication ofend of life decisions in European intensive care units. Intensive Care Med 2005;31(9):1215–21 [Epub 2005 Jul 22].

[35] Seaman JB. Improving care at end of life in the ICU: a proposal for early discussionof goals of care. J Gerontol Nurs 2013;39(8):52–8. http://dx.doi.org/10.3928/00989134-20130530-01.

[36] Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. SurvivingSepsis Campaign Guidelines Committee including the Pediatric Subgroup. Survivingsepsis campaign: international guidelines for management of severe sepsis and sep-tic shock: 2012. Crit Care Med 2013;41(2):580–637. http://dx.doi.org/10.1097/CCM.0b013e31827e83af.

[37] Ip M, Gilligan T, Koenig B, Raffin TA. Ethical decision-making in critical care in HongKong. Crit Care Med 1998;26(3):447–51.

[38] Cong Y. Ethical challenges in critical care medicine: a Chinese perspective. J MedPhilos 1998;23(6):581–600.

[39] Danis M, Patrick DL, Southerland LI, Green ML. Patients' and families' preferences formedical intensive care. JAMA 1988;260(6):797–802.

[40] Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, SibbaldWJ.Multiple organdysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med1995;23(10):1638–52.

[41] Cook R, Cook D, Tilley J, Lee K, Marshall J, Canadian Critical Care Trials Group. Multi-ple organ dysfunction: baseline and serial component scores. Crit Care Med 2001;29(11):2046–50.

[42] Ferreira FL, Bota DP, Bross A, Mélot C, Vincent JL. Serial evaluation of the SOFA scoreto predict outcome in critically ill patients. JAMA 2001;286(14):1754–8.

[43] Frost DW, Cook DJ, Heyland DK, Fowler RA. Patient and healthcare professional fac-tors influencing end of life decision-making during critical illness: a systematic

s Round Table IV: Health care professional end-of-life decisionmaking,

7G.M. Joynt et al. / Journal of Critical Care xxx (2014) xxx–xxx

review. Crit Care Med 2011;39(5):1174–89. http://dx.doi.org/10.1097/CCM.0b013e31820eacf2.

[44] Brandberg C, Blomqvist H, Jirwe M. What is the importance of age on treatment ofthe elderly in the intensive care unit? Acta Anaesthesiol Scand 2013;57(6):698–703. http://dx.doi.org/10.1111/aas.12073.

[45] Hofhuis JG, van Stel HF, Schrijvers AJ, Rommes JH, Spronk PE. Changes of health-related quality of life in critically ill octogenarians: a follow-up study. Chest 2011;140(6):1473–83. http://dx.doi.org/10.1378/chest. 10-0803.

[46] Kaarlola A, Tallgren M, Pettilä V. Long-term survival, quality of life, and quality-adjusted life-years among critically ill elderly patients. Crit Care Med 2006;34(8):2120–6.

[47] Andorno R, Biller-Andorno N, Brauer S. Advance health care directives: towards acoordinated European policy? Eur J Health Law 2009;16:207–27. http://dx.doi.org/10.1163/157180909X453053.

[48] Nicholas LH, Langa KM, Iwashyna TJ, Weir DR. Regional variation in the as-sociation between advance directives and end-of-life Medicare expendi-tures. JAMA 2011;306(13):1447–53. http://dx.doi.org/10.1001/jama.2011.1410.

[49] Chu LW. One step forward for advance directives in Hong Kong. Hong Kong Med J2012;18(3):176–7.

[50] Makino J, Fujitani S, Twohig B, Krasnica S, Oropello J. End-of-life considerations inthe ICU in Japan: ethical and legal perspectives. J Intensive Care 2014;2:9. http://dx.doi.org/10.1186/2052-0492-2-9.

Please cite this article as: Joynt GM, et al, The DurbanWorld Congress EthicJ Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.10.011

[51] Heaney M, Foot C, Freeman WD, Fraser J. Ethical issues in withholding and with-drawing life-prolonging medical treatment in the ICU. Curr Anaesth Crit Care2007;18(5–6):277–83.

[52] Pronovost P, Angus DC. Economics of end-of-life care in the intensive care unit. CritCare Med 2001;29:N46–51 [Suppl.].

[53] ChristianMD, Hawryluck L,Wax RS, Cook T, Lazar NM, HerridgeMS, et al. Developmentof a triage protocol for critical care during an influenza pandemic. CMAJ 2006;175(11):1377–81.

[54] Moskop JC, Iserson KV. Triage in medicine, part II: underlying values and principles.Ann Emerg Med 2007;49(3):282–7.

[55] Fowler R, Hammer M. End-of-life care in Canada. Clin Invest Med 2013;36(3):E127–32.

[56] Morrison RS, Penrod JD, Cassel JB, Caust-EllenbogenM, Litke A, Spragens L, et al. Costsavings associated with US hospital palliative care consultation programs. Arch In-tern Med 2008;168(16):1783–90. http://dx.doi.org/10.1001/archinte.168.16.1783.

[57] Huynh TN, Kleerup EC,Wiley JF, Savitsky TD, GuseD, Garber BJ, et al. The frequency andcost of treatment perceived to be futile in critical care. JAMA InternMed 2013;173(20):1887–94. http://dx.doi.org/10.1001/jamainternmed.2013.10261.

[58] Curtis JR, Engelberg RA, Bensink ME, Ramsey SD. End-of-life care in the intensivecare unit: can we simultaneously increase quality and reduce costs? Am J RespirCrit Care Med 2012;186(7):587–92.

[59] Luce JM, Rubenfeld GD. Can health care costs be reduced by limiting intensive care atthe end of life? Am J Respir Crit Care Med 2002;165(6):750–4.

s Round Table IV: Health care professional end-of-life decisionmaking,