Should patients and family be involved in "Do not resuscitate" decisions? Views of oncology and palliative care doctors and nurses
Background: "Do not resuscitate" (DNR) orders are put in place where cardiopulmonary resuscitation is inappropriate. However, it is unclear who should be involved in discussions and decisions around DNR orders.
Keywords: Asian Continental Ancestry Group, Decision making, Medical oncology, Palliative care, Resuscitation orders
Cardiopulmonary resuscitation (CPR) is an emergency procedure which is performed on a person in cardiac arrest in an effort to reverse the otherwise inevitable progression to death.  However, success is not always possible and not infrequently this procedure is associated with a high level of morbidity. It is for these conditions when CPR is not envisaged to be viable that the "do not resuscitate" (DNR) order was conceived. ,,,
This determination had almost exclusively been a physician-led decision until the 1980s when a shift in attitudes began to take shape. DNR orders began to veer toward patient-led determinations and were seen as a means for patients to reclaim the right to self-determination in the face of medical paternalism.  Even if the doctor was ultimately responsible for the decision, there was growing empirical evidence to suggest that most patients would like to play an active role in the determination of their resuscitation status. ,
Yet within Asian cultures, family-centric rather than patient-centric determinations prevail, sometimes limiting such assertions of autonomy. ,,, Family members, invested with the responsibility of safeguarding the patient from psychological distress which in turn was believed to hasten clinical deterioration, often influenced how much the patient knew and consequently how involved the patient was in deciding about treatments.  While in Western cultures, the disclosure of diagnosis and prognosis to patients has been associated with a better quality of life, this is not necessarily so in Asian cultures.  Mo et al. showed that in Korea, patients who were aware of their illness and who actively participated in the decision-making process did not score higher than others on outcome measures of quality of life and quality of death. In fact, those who were not fully involved in decision making scored better in some domains, some of which were statistically significant. 
Singapore is a society that holds fast to Asian values but is also increasingly influenced by Western biomedical ethical frameworks. For physicians and nurses, the increasing onus placed on patient autonomy and a growing tendency to limit the influence of the family in the decision-making process do seem at odds with traditional mores of the familial determination fed mainly by a beneficent wish to protect patients from distress and also to maintain hope. , As a result, local physicians have been found to maintain this status quo and prefer to discuss treatment options with the family rather than the patients themselves, particularly when DNR statuses are being determined. Nonetheless, there is increasing evidence that patient perception and involvement in the decision-making process might be changing. ,
It is from within this context that this study sets out to explore attitudes of oncology and palliative care doctors and nurses faced with sometimes conflicting cultural intuitions and professional opinions when deliberating DNR discussions. Additionally, this paper also attempts to elucidate the understanding of the implications of a DNR order among physicians and nurses of the various oncology specialities and palliative medicine as well as their perceptions of patient care following a DNR order being implemented.
The study took place in a tertiary specialist cancer center and an inpatient oncology ward located within the same campus. Participants were doctors and nurses who worked in oncology or palliative medicine. To minimize disruption to service provision, the questionnaire was explained and distributed at department meetings for doctors and at staff handover periods for nurses. Each participant completed a consent form and an anonymized once-off questionnaire comprising questions relating to discussing and deciding on DNR orders. The completed forms were placed within a box by each participant and collected by the investigators at the end of the day to ensure anonymity.
Over a 1-month period between March 1 and 31, 2011, 187 questionnaires were distributed and 146 questionnaires were returned (response rate 78.1%), out of which 37 (25.3%) were completed by doctors and 109 (74.7%) by nurses. The age of participants ranged from 19 to 68 years (mean 32.0, median 29) and years of experience in oncology or palliative care ranged from 0 to 40 years (mean 5.77 years, median 3.75). [Table 1] shows further demographic details of the participants. The responses to the questions are shown in [Table 2]. Where questions were left unanswered or more than one option was ticked, these are accounted for under "void."
There was a variation in what participants thought a DNR order involved. Although most agreed that therapies administered should be limited to those that can be provided in a general ward, negating any transfer to an intensive care unit, there was a significant discrepancy as to what this would in fact encompass. Nearly half the participants (48%) felt that this determination would not entail the utilization of aggressive fluid replacements in a crisis or the implementation of antibiotics should the need arise despite the fact that both could be carried out in a general medical ward. Conversely, 43% of respondents felt otherwise and would prevail upon the use of these measures should the need arise.
When asked who a DNR decision should be discussed with, 78.8% of all responders felt that such determinations ought to include the patients themselves while 78.1% of all responders felt that the next of kin must play a part in such a process. Thirty-six (24.7%) responders thought that a wider involvement of the family beyond simply the next of kin was called for in such deliberations; 21 (14.3%) participants felt conversely that neither the patient nor the next of kin should be involved in DNR discussions.
Participants were then asked as to who should ultimately be responsible for determining a DNR order. Responses appeared spread across the various options offered ranging from "always the doctor" to "always the patient/family." There was a statistically significant difference (P<0.001 for Pearson's Chi-square test) between the responses from nurses and those from the doctors. Physician respondents tended to favor the physician being given the final say while nurses, the patient/family [Figure 1].
On the question of primacy between patient and family, responses were varied with 23.3% believing that priority was situation dependent. A small majority of 81 (55.4%) participants were found in favor of the patient rather than the family. Twenty-seven (18.5%) participants however felt that the family ought to be afforded precedence.
A striking finding of this study is the presence of variance in the understanding of the "do not resuscitate" label among healthcare professionals despite them working within the same campus. While there was an acceptance that a DNR order signified a change of tact from treatments focused upon life extension to one of maximizing the quality of life and comfort measures, there were differing views on how patients would be managed when they deteriorated. The use of antibiotics and fluids in sepsis may extend life expectancy, but may also facilitate symptom management without being detrimental to the patients, which may in turn improve their quality of life. Even if the goal of treatment has shifted away from life extension, the small amount of time granted by such interventions may prove to be providential where there are goals to be met or families awaited. Hence, in patients with a DNR order, the clarification of how "aggressive" medical management should be is helpful particularly to the out-of-hours on-call team. ,
It is remarkable that within such diversity in the comprehension of a DNR status, nearly half the participants (49.3%) had no concerns that a DNR order would lead to the provision of "substandard" care. Perhaps this is related to an increasing tendency for decisions to be regularly reviewed and a feeling that these patients would be referred on to the palliative care teams for better symptom control and holistic management. Likewise patients at the end of life who upon declaration of a DNR order would become eligible for the Care for the Dying Pathway, which is increasingly being employed on campus and evidenced to be an effective tool in the early identification and therefore treatment of symptoms. ,,
With regard to the question of who should be involved in the determination of a DNR order, this study shows that patients and family should play a more important role than other healthcare professionals, reflecting the increasing importance of patient autonomy within the medical decision-making process. Although a large majority (78.8%) of participants thought that DNR orders should be discussed with patients, this does not seem to be consistent with local practice.  In many cases, patients tend to be too ill to discuss such plans and frequently it is the family that occupies the main decision-making position. It is possible that while healthcare professionals think that they should have DNR discussions with patients, in reality they are reluctant to broach this subject due to lack of time, fear of patient's response, a sense of inadequacy in ability to handle such conversations, or as a result of the local perception that such discussions about death are in fact taboo.
It is thus not altogether surprising that 78.1% of healthcare professionals felt that a DNR decision should be discussed with the next of kin. While helpful in trying to ascertain goals, values, and preferences of the patient, the next of kin may harbor vested interests and may not act in the patient's best interests, in which case a discussion with the wider family unit may clear the air. It is perhaps for this reason that 24.7% respondents suggested such a wider familial involvement in decision making.
Why is there a great discrepancy between the involvement of next of kin and as many family members as possible? It is possible that doctors involve only the next of kin mainly for pragmatic purposes, as a decision-making process involving many family members can be difficult and inefficient. Communicating to the next of kin as a surrogate decision maker may also be seen as easier than discussing a DNR order directly with the patient. Qualitative studies should be done to explore this further.
In any case, evidence exists that would suggest that patients may have a better quality of life if they are not actively involved in decision making.  It appears that within the regnant family-centric "Asian ethical framework", acting in patients' best interest may mean denying them of their right to self-determination. , The acceptability and viability of such a posit however still requires further elucidation. ,, Questions regarding whether familial determinations without patient involvement are really made in the patient's best interests need to be balanced against the necessity of family input to ascertain the many psychosocial, financial, physical, clinical, and spiritual issues to be considered so as to formulate a care plan that best provides for the needs of the patient. ,,
Despite the importance of patient and family involvement, a significant minority of 14.3% of participants thought that DNR decisions should not be discussed with the patient or family. This may reflect the view that the DNR order is a medical decision regarding a treatment with medical indications that do not have to be discussed with the patient or family if the treatment is thought to be inappropriate for the patient. ,,, In keeping with local guidelines, the Scottish integrated policy on decision making and communication of DNR orders states that where CPR is unlikely to have a medically successful outcome, it should not be attempted and it is "an unnecessary and cruel burden" to ask patients and relevant others to decide about CPR when it is not a treatment option. ,,
DNR orders are highly emotive and may be associated with negative connotations of failure; hence, patients and their families may struggle to make a decision that is truly in patient's best interest. Simultaneously, discussions on a futile treatment in the vain hope of family acquiescence for this venture may in fact be in breach of the ethical principle of nonmaleficience. Furthermore, if the patient or family decides against a DNR order, the provision of a futile treatment may in fact be in breach of the physician's overriding duty of care to the patient and their family. Exchanging patient autonomy for a treatment that will be more harmful than beneficial to the patient would also be a contravention of a physician's primary obligation to act in the patient's best interests.
Even so, should not patients' right to self-determination be valued above all, even if their choice is perceived by the physician to cause more harm than good? Who should ultimately decide on a DNR order? Should it be the patient and/or family because it is about the patient's life in the end, with input from healthcare professionals so that they can make an informed decision? Or should it be the doctor because it is ultimately a medical intervention with medical indications, with input from the patient and/or family so that the decision can be in line with the patient's values and preferences?
Results from this study appeared to veer toward a decisive response by either of the key personnal within the deliberations rather than the "safety" of a situation-dependent response. There was almost an equal preference for the doctor (41.8%) as there was for the patient and/or family (41.1%) to fulfill this fundamental role. Indeed, only 16.4% opted for a response that would depend on the situation. In keeping with legal and practice guidelines, it was the physicians who opted for the physician holding the decisive position while in keeping with other studies, most nurses felt that it ought to be the patient who ought to wield the final say. ,,
Existing studies in the Asian context suggest that between the patient and family, it is the family that presides over key decisions, particularly in the end-of-life issues. ,, The rationale it would seem is so that the patient is relieved of the burden of knowledge and weight of such a major decision. ,,,,, Certainly, received knowledge and current practice suggest that the involvement of the patient in the process is seen as unimportant and possibly even detrimental to general well-being and health outcomes. ,,,, However, this norm is challenged by the results of this study. When asked whether the patient or the family should be more involved in DNR discussions and decisions, a majority of 55.4% of participants thought that patients should be more involved than the family, suggesting that even in end-of-life decision making, the Asian culture does not preclude the involvement of patients.
Indeed the right approach to DNR discussions and decisions varies even within a particular culture, and needs to be individualized. Yet, even when it has been determined how and with whom to have a DNR discussion with, it is difficult to know when is the most appropriate time to have that discussion, as shown by the wide range of responses to the final question. A discussion too early may be unnecessarily distressing for the patient or family and may be misinterpreted as the doctor giving up on the patient. On the other hand, a discussion that occurs too late may mean that the patient is not physically able to participate and the family may be too preoccupied with the patient's imminent death to be able to engage in it.
This study is limited by the small sample size which prevented further statistical analysis of factors associated with various responses to the questions. Further research employing qualitative methodology will also be valuable in helping to better understand the results obtained from this questionnaire study. Additionally, the study could be expanded to include nurses, physicians, and allied heathcare from other specialities to facilitate a wider grasp of the situation.
In summary, the determination of DNR statuses is difficult and practices vary depending on country, culture, and even healthcare professionals working in different specialties. In this study, even though all participants were oncology or palliative care healthcare professionals working within a single institution, there was a wide range of responses to questions on various aspects of DNR discussions and decisions, indicating that views differed even within this narrow profile. This was also present across all questions in the subgroup of palliative care professionals, who may be expected to deal with DNR discussions and decisions in a more consistent manner. An institutional guideline may facilitate a standardized approach to this complex issue, but given the myriad factors involved would serve more as a guide to a thought process rather than a protocol to be followed.
From the viewpoint of local oncology and palliative care healthcare professionals, patients should be involved in discussions pertaining to their own resuscitation status even though when and to what extent this should occur remains unclear. Moreover, questions persist as to whether such practical and perceptual changes can and will be translated to clinical practice. Efforts to confront issues of collusion and improve direct patient involvement remain unstudied as do efforts to change perceptions; however, these efforts do continue as further work is still required in this field. It is hoped that more research on this matter will further illuminate this process toward a patient-centered care framework balanced upon the specifics of an individual case scenario.
We thank all participants for their time taken to complete the survey forms and the reviewers whose comments greatly improved this manuscript.
[Table 1], [Table 2]