Use of the welfare-based model in the application of palliative sedation
There are many terminologies referring to the use of sedation at the end of life. Some may call it palliative sedation (PS) or terminal sedation (TS). In both situations, sedatives and sometimes opioids are used to induce unconsciousness as a means of circumnavigating awareness of suffering (Krishna 2013, 2). Both PS and TS involve continuous sedation and stand apart from lighter forms of sedation where patients can communicate with caregivers or intermittent sedation (Morita et al. 2001). In an estimated 1–3% of cases locally, continuous deep palliative sedation (CDPS) is applied (Krishna 2013, 2). CDPS is defined as the proportional and monitored induction of deep continuous sedation for the amelioration of all forms of intractable physical, psychological, and existential suffering in a cancer patient with a prognosis of less than 2 weeks following a holistic multi-professional assessment of the patient’s condition to affirm that suffering is in fact intractable and the application of such an intervention is in the patient’s best interests (Krishna 2013, 19). Despite this clear definition, CDPS remains clouded in controversy. One particularly worrying aspect is consent. Here we use a case study to highlight some of the issues underpinning this concern.
Mr P was a 54-year-old Chinese gentleman with metastatic non-small cell lung cancer. He underwent multiple lines of chemotherapy, immunotherapy, and palliative radiotherapy to his right lung mass. However, the cancer progressed to involve rib bones and pleura despite treatment and Mr P opted for best supportive care 2 years after diagnosis. Palliative care involvement began with his admission to the hospital for pain management.
Aside from pain as a result of the enlarging tumor mass in the right hemithorax, Mr P developed progressive breathlessness. Mr P declined invasive investigations including blood tests. At that time, he was on regular oxycodone for analgesia and breathlessness. He declined the use of morphine as pain control was suboptimal with morphine and he disliked the sedating effect of morphine. Attempts to alleviate breathlessness including increasing the dose of oral oxycodone and switching to intravenous oxycodone did not succeed. A repeat course of radiotherapy was not possible given the large tumor size and previous radiotherapy. Complementing the pharmacological treatment, non-pharmacological methods to reduce breathlessness were introduced as well. These include breathing and relaxation techniques, fan blowing, and supplemental oxygen. Referrals to the physiotherapist and occupational therapists were made to facilitate maximization of these non-pharmacological measures.
Contributing to the breathlessness was Mr P’s anxiety. He was only relaxed in the company of his family. When they were not by his side, he became anxious, worsening his breathlessness. The social worker was engaged to provide psycho-emotional and spiritual support. However, the social worker found it difficult to do so as Mr P was mostly lethargic and could not engage in long conversations. Alprazolam was started to reduce his anxiety but after trying the first dose, Mr P refused to continue with it, citing ineffectiveness.
Mr P continued to be breathless. At the multidisciplinary team (MDT) meeting, the option of PS was discussed given the intractability of his breathlessness and short prognosis. The option was discussed with Mr P and his family. Though Mr P did not commit to the use of sedatives at that time, he verbalized that comfort was most important to him. In light of worsening symptoms, poor control and exhaustion of options, the MDT opted for PS. As Mr P was still alert, we took verbal consent for PS again and he agreed.
He was started on light intermittent sedation with the use of as-needed subcutaneous midazolam injections but each dose alleviated his breathlessness for a short period only. Initially his family was worried about the drowsiness and tried to limit Mr P’s use of the midazolam injections. However, as Mr P was still symptomatic, we discussed with him and his family to use light continuous sedation. He was started on a midazolam infusion. The dose was increased according to Mr P’s report of symptom relief. Throughout the process of palliative sedation, Mr P’s vital signs were monitored regularly including the respiratory rate. The use of artificial nutrition and hydration was discussed with Mr P and his family prior to initiation of PS. As Mr P had edema over his upper limbs and he could still drink small amount of fluids, artificial hydration was not instituted.
He subsequently developed terminal delirium and became very agitated. Use of anti-psychotics such as haloperidol did not provide the required symptom control. Intermittent deep sedation with use of a regular bolus dose of midazolam injection at night did not improve his agitation nor worsen it. Non-pharmacological measures such as regular orientation and having the same staff caring for him were not helpful. As Mr P was unable to give consent due to delirium, we spoke to the family regarding use of CDPS as he was imminently dying, symptoms were refractory and there was only a short period of time left to achieve symptom relief. He was started on a combined subcutaneous infusion of haloperidol and midazolam. Symptoms improved and he passed away peacefully a few hours later after the initiation of CDPS.
The view for informed consent for the application of PS is divided in the palliative care community. Informed consent is a prerequisite for the application of PS in some guidelines (Rousseau 2007; Cowan et al. 2007; Hospice and Palliative Nurse Association 2003; Claessens et al. 2008). Meanwhile, the American Medical Association Council on Ethical and Judicial Affairs (Levine 2008), the American Academy of Hospice and Palliative Medicine (2006), the Veterans Health Administration’s (2006) National Center for Ethics in Health Care, the National Hospice and Palliative Care Organization (Kirk and Mahon 2010), the European Association for Palliative Care (Cherny et al. 2009), the Hospice and Palliative Care Federation of Massachusetts (2004), the Alberta Health Services (2012) and the guidelines set out in the Oxford Textbook of Palliative Medicine (Krakauer and Quinn 2011), and the Textbook of Palliative Medicine (Cherny 2006) allow for the input of surrogates in the event the patient is not competent.
There are also a few who have revert to best interests determinations should the patient be unable to consent to this procedure or allow healthcare providers to overrule surrogate decisions if it is felt that they obstruct the best interests of the patient (Norwegian Medical Association 2014; Royal Dutch Medical Association 2009).
Validity of informed consent in the face of intractable symptoms
There is a growing body of professional opinion that holds that in most cases the ability of patients to give voluntary and informed consent is limited or the consent cannot be construed as legitimate if obtained (Krishna 2010; Krishna 2011a; Krishna 2012; Battin 2008). Beauchamp and Childress (2001) define informed consent as consisting of preconditions which contain elements of competence and voluntariness, informational elements which include disclosure, recommendations of a plan and understanding, and consent elements which are composed of decisions in favor of the plan and authorization of the chosen plan.
Patients with intractable symptoms are unlikely to fulfill the criteria for an informed consent. This is demonstrated in the case of Mr P for the following reasons:
The presence of intractable symptoms represents a coercive factor that could compromise voluntarism. It can be viewed that Mr P had no other option except to continue to suffer if he did not agree with sedation. A patient with intractable symptom may be pressured by significant others who are uncomfortable with the suffering witnessed by them. This ultimately affects his decision-making capabilities.
Mr P’s physical (breathlessness) and psychological (anxiety) suffering can affect his cognition. We were limited by his lethargy and rapid progression of cancer to properly address his anxiety with non-pharmacological methods as such cognitive-behavior therapy. Anxiolytics were prescribed but he rejected alprazolam. We also wonder how much information is retained in the face of great distress. He subsequently could not consent to continuous deep palliative sedation as he became delirious. Studies have shown that up to 80% of terminally ill patients suffer some cognitive impairment, affecting insight and hence their ability to give informed consent (Claessens et al. 2008; Breibart et al. 1995). These factors, coupled with these patients’ particular physical, psychological, spiritual, social, and economic situations, impede their ability to act in a manner that serve their best interest (Krishna 2010; Manson and O’neill 2007; Tai and Lin 2001).
Disclosure of information tends to be biased and incomplete in such circumstances (Krishna 2011a). Some have argued against the full disclosure of information under the principle of beneficence, whereby they insist that information should be conveyed piecemeal to protect the patient from unnecessary distress (Krishna 2010). Gillon has argued that the distress caused by conveying complete information violates the principle of non-maleficence (Gillon 2004). He states that certain types of information are merely “guesstimates” and the physician who imparts this information places an unreasonable burden on the family.
Relational Autonomy (RA)
Best Interests Principle (BIP)
Welfare-based Model (WM)
I will briefly discuss the first two models before we dwell further into the WM, which was applied for Mr P’s case.
The RA model adopts an inclusive view of family involvement in care determination and sees the patient and family as an intimately entwined unit (Sy et al. 2015). This model attempts to aid decision-making in the setting of a family-centric society by acknowledging the family’s role and position in keeping with local sociocultural beliefs. The family is conferred a significant role within the decision-making process (Sy et al. 2015). This is evident as the final decision-making lay with the patient but under the influence of his family or with the family. However, there is growing concern about the ability of the family to dissociate their interests from those of the family and to act in a way that advances the well-being of the patient (Sy et al. 2015). This leads to the discrediting of the RA approach. There is good empirical evidence that people are poor proxy decision-makers as regards to accurately representing other people’s desires and wishes (Seckler et al. 1991). A systematic analysis by Shalowitz et al. (2006) showed that overall, surrogates predicted patients’ treatment preferences with only 68% accuracy. Philosophical theory explaining how we represent other people’s thought processes indicates that we are also unlikely ever to achieve accurate simulations of others’ wishes in making a proxy decision (Wrigley 2007). The role of surrogate decision-maker itself is strenuous and frequently “emotionally draining” (Yang et al. 2012; Foo et al. 2012; Ho et al. 2013; Tan and Chin 2011). These emotional pressures along with financial, psychosocial, and caregiver strain may further hinder and potentially compromise the decision-making process (Krishna 2011a, b; Krishna and Chin 2011; Goh 2008; Meeker and Jezewski 2005). There is little empirical and clinical evidence to suggest that a RA framework could provide a viable compromise between principality and family-centric frameworks (Chan et al. 2006; Krishna et al. 2015a; Jafarey and Farooqui 2005; Moazam 2000). As such, the use of the RA model in decision-making may not be ideal in the application of PS.
In this case study, Mr P’s family was overwhelmed by his distress and rapid deterioration. They were struggling with keeping him comfortable but yet, concerned about his alertness. This could be seen when they attempted to limit the use of midazolam for symptom relief. If we were to apply the RA framework, it was likely that patient’s best interests would be compromised and great stress would have been placed on the family to make the call.
Best interests principle
It has been proposed that the physician takes on the role of decision-maker in such situation and that proxies or surrogates act as advisors to the medical team in helping make best-interests judgment in accordance with the Best Interest Principle (Krishna 2010; Shalowitz et al. 2006). The shortcomings of the RA framework are addressed with the use of the BIP. The limiting factor in the BIP model is that it solely relies on the primary physician and his or her team to accurately balance the benefits and harms of PS (Sy et al. 2015). This raises questions as to the accountability and the nature of the decisions.
It is then proposed that the primary physician is replaced by a multidisciplinary team (MDT) to confer greater accountability and transparency to the decision-making process (Krishna 2012; Krishna et al. 2015a). This model is the Welfare-based Model (WM), which I will describe below.
The welfare-based model
We applied the use of the WM in deciding on palliative sedation for Mr P. The Welfare-based Model utilizes best-interest decisions that are determined by a MDT review of each specific case. The MDT is seen as a “group of people of different healthcare disciplines, which meet together at a given time (whether physically in one place, or by video or teleconferencing) to discuss a given patient and who are each able to contribute independently to the diagnostic and treatment decisions about the patient” (Department of Health 2004). In the local setting, such a MDT would include physicians, psychiatrists, psychologists, social workers, and nurses (Zhuang 2016).
The Welfare-based Model is primarily beneficence driven and focuses on preservation of the best interests and welfare of the patient. Each individual is assessed by the MDT according to his or her own merits and particular circumstances in order to establish what the best interests of the patient ought to be with minimal compromise to his or her personal autonomy (Krishna et al. 2015b). Proxies, surrogates, and families are consulted to ensure that the patient’s opinions, values, and beliefs continue to be considered in determinations about their care when they themselves may be incapable of doing so (Krishna et al. 2015b). Input from other members of the MDT provides different perspectives to the case as patients and proxies may choose to confide in different members of the MDT as they see fit. For example, they may not share information about financial matters with the physician but rather with a social worker who may be of greater assistance. They may share concerns of mental health with a psychiatrist or psychologist. The nurses in the ward are more frequently in contact with the patients compared to the physician who may only see the patients once in a day. All these allow delineation of a particular patient’s contextual and person-specific definitions of harm within the confines of a case-specific setting (Krishna et al. 2015b). For objectivity, the opinion of a separate palliative team not involved in care of patient is sought as well.
The use of the WM confers greater accountability and transparency to the decision-making process (Krishna 2012; Krishna et al. 2015a). The application of a welfare-model that owes much to a beneficence-based approach is also congruent with prevailing sociocultural practices. In most family-centric societies, it is a beneficence-based model rather than an autonomy-based model that dominates local thinking (Tsai 1999).
The WM approach is demonstrated in Mr P’s case, whereby a multidisciplinary team met to discuss various options with the input from Mr P himself and his family. During the initial period, his family had some reservations towards the use of sedation but his continued suffering with the breathlessness and delirium was deemed not in alignment with his goal for comfort. It was then that they too agreed for continuous deep palliative sedation in order to protect his welfare.
The versatility of the WM enables its application in a variety of settings pertaining to PS. This is illustrated below.
Can PS be applied if the patient has previously declined such treatment?
The use of advance care plan (ACP), preferred care plan (PPC), or advance medical directives (AMD) may shed some insight to patient’s preferences but it is not commonly issued and can be vague, without specific guidance of what the patient would have wanted in various possible scenarios. However, it has been argued that a person’s values and goals change over time and may contradict previous arrangements, wishes, or goals that may be stipulated within said documents. The ACP and PPC are not legal documents unlike the AMD (Potter et al. 1994; Ryan 1996). Similarly, we can argue that when PS was brought up as part of advanced care planning, the patient may not have been able to envisage the impact of the situation to him or her. If the situation that has arisen is not consistent with what the patient may have envisaged whilst in a competent state, we can similarly use the WM to approach the case. The WM will aid the physician in deciding the need to override this position in the patient’s best interests.
Can PS be applied to a patient found to be incompetent but who persistently declines this treatment?
Competence is not an all or nothing finding (Chin 2009). Singapore’s Mental Capacity Act (MCA) recognizes that there are patients who retain competence to consent on lesser matters but unable to attain a level of competence to decide upon care decisions (Chin 2009). Physicians may find themselves in a situation where patients who have previously consented when competent but who now decline PS or those who have not voiced any preferences when competent but now decline PS. So what happens when we doubt the right of these patients to refuse medical intervention?
In such a case, the WM is also applicable for the decision-making process. The WM requires a complete review of the patient and the processes that had led up to this point by a MDT. Next is to assess a patient’s competence. If found to be incompetent and there are no reasons for refusal of PS, then the MDT can override the patient’s position. An independent palliative care specialist and/or a psychiatrist consult will also lend objectivity in such situations.
If patient is assessed to be competent and can provide reasons for declining PS, then the MDT can be justified in proceeding to the extent the patient allows.
In situations warranting the provision of PS, it may not be practicable to obtain informed consent for the purposes of decision-making. Furthermore, it is unlikely for informed consent to be considered legitimate or valid given the presence of intractable symptoms, which tend to impair the patient’s judgment and vitiate competence. Medical, psychosocial, and cultural factors lend bias to disclosure of information to patients, and there is a lack of reliable sources for obtaining consent such as proxies, surrogates, advance directives, and living wills. The use of the Welfare Model in decision-making for palliative sedation enables a holistic review of the patient’s best interests with minimal compromise to his or her autonomy.
- Alberta Health Services - Calgary Zone. 2012. Clinical practice guidelines for palliative sedation.Google Scholar
- American Academy of Hospice and Palliative Medicine. 2006. Statement on palliative sedation.Google Scholar
- Beauchamp, Tom L., and James F. Childress. 2001. Principles of biomedical ethics. New York: Oxford University Press.Google Scholar
- Cherny, Nathan I. 2006. Palliative sedation. In Textbook of palliative medicine, ed. Eduardo Bruera, Irene J. Higginson, Carla Ripamonti, and Charles F. von Gunten, 976–987. London: Hodder Arnold.Google Scholar
- Cherny, Nathan I., Lukas Radbruch, and Board of the European Association for Palliative Care. 2009. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care. Palliative Medicine 23 (7): 581–593. https://doi.org/10.1177/0269216309107024.
- Chin, Jing Jih. 2009. Ethical issues related to the mental capacity act. Singapore Family Physician 35 (3): 22–25.Google Scholar
- Cowan, J.D., T. Palmer, and L. Clemens. 2007. Palliative Sedation in Palliative Medicine. In Palliative Medicine, ed. T. Declan Walsh, Augusto T. Caraceni, Robin Fainsinger, Kathleen M. Foley, Paul Glare, Cynthia Goh, Mari Lloyd-Williams, Juan Nunez Olarte, and Lukas Radbruch, 982–988. Philadelphia: Saunders/Elsevier.Google Scholar
- Department of Health. 2004. Manual for Cancer Services. London: Department of Health. http://webarchive.nationalarchives.gov.uk/20091105213859/http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/documents/digitalasset/dh_4135597.pdf, accessed 30 July 2017.
- Foo, Wei Ting, Yiliang Zheng, Grace Meijuan Yang, Ann K. Kwee, and Lalit Kumar Radha Krishna. 2012. Factors considered in end-of-life decision-making of healthcare professionals. BMJ Supportive & Palliative Care 2 (1): A45–A46. https://doi.org/10.1136/bmjspcare-2012-000196.131.
- Gillon, Raanan. 2004. Telling the truth, confidentiality, consent, and respect for autonomy. In Bioethics, ed. Harris Jones, 507–528. Oxford: Oxford University Press.Google Scholar
- Ho, Zheng Jie Marc, Lalit Kumar Radha Krishna, Cynthia Goh, and Chung Pheng Alethea Yee. 2013. The physician–patient relationship in treatment decision making at the end of life: A pilot study of cancer patients in a Southeast Asian society. Palliative & Supportive Care 11 (1): 13–19. https://doi.org/10.1017/S1478951512000429.
- Hospice and Palliative Care Federation of Massachusetts. 2004. Palliative Sedation Protocol. A report of the Standards and Best Practices Committee. http://c.ymcdn.com/sites/www.hospicefed.org/resource/resmgr/hpcfm_pdf_doc/pal_sed_protocol_2004.pdf, accessed on 1 July 2017.
- Hospice and Palliative Nurse Association (World Health Organization). 2003. Palliative Sedation at the End of Life.Google Scholar
- Kirk, Timothy W., and Margaret M. Mahon. 2010. National Hospice and palliative care organization (NHPCO) position statement and commentary on the use of palliative sedation in imminently dying terminally ill patients. Journal of Pain and Symptom Management 39 (5): 914–923. https://doi.org/10.1016/j.jpainsymman.2010.01.009.
- Krakauer, E.L., and T.E. Quinn. 2011. Sedation in palliative medicine. In Oxford textbook of palliative medicine, ed. Derek Doyle, Geoffrey W.C. Hanks, and Neil MacDonald, 1560–1567. New York: Oxford University Press.Google Scholar
- Krishna, Lalit Kumar Radha. 2010. Consent in terminal sedation. Indian Journal of Ethics 7 (3): 161–165. https://doi.org/10.20529/IJME.2010.054.
- Krishna, Lalit Kumar Radha. 2011a. Decision making at the end of life: A Singaporean perspective. Asian Bioethics Review 3 (2): 118–126.Google Scholar
- Krishna, Lalit Kumar Radha. 2011b. The position of the family of palliative care patients within the decision making process at the end of life in Singapore. Ethics and Medicine 27 (3): 183–190.Google Scholar
- Krishna, Lalit Kumar Radha. 2012. Best interests determination within the Singapore context. Nursing Ethics 19 (6): 787–799. https://doi.org/10.1177/0969733011433316.
- Krishna, Lalit Kumar Radha. 2013. Continuous Deep Palliative Sedation: An Ethical Analysis. Doctoral Dissertation, National University of Singapore. http://scholarbank.nus.edu.sg/bitstream/handle/10635/43527/final%20submission%20.pdf?sequence=1, accessed 1 August 2017.
- Krishna, Lalit Kumar Radha, and Jacqueline Joon Lin Chin. 2011. Palliative sedation within the duty of palliative care within the Singaporean clinical context. Asian Bioethics Review 3 (3): 201–215.Google Scholar
- Krishna, Lalit Kumar Radha, Deborah S. Watkinson, and Ng Lee Beng. 2015a. Limits to relational autonomy—The Singaporean experience. Nursing Ethics 22 (3): 331–340. https://doi.org/10.1177/0969733014533239.
- Krishna, Lalit Kumar Radha, Jason Te Tay, Deborah S. Watkinson, and Alethea Chung Pheng Yee. 2015b. Advancing a welfare-based model in medical decision. Asian Bioethics Review 7 (3): 306–320. https://doi.org/10.1353/asb.2015.0020.
- Levine, M.A. 2008. Sedation to Unconsciousness in End-of-Life Care (CEJA Report 5-A-08). American Medical Association Council on Ethical and Judicial Affairs (CEJA). https://www.ama-assn.org/sites/default/files/media-browser/public/about-ama/councils/Council%20Reports/council-on-ethics-and-judicial-affairs/a08-ceja-palliative-sedation.pdf, accessed 1 July 2017.
- Manson, Neil C. and Onora O'Neill. 2007. Rethinking informed consent in bioethics. Cambridge University Press.Google Scholar
- Norwegian Medical Association. 2014. Guidelines for palliative sedation at the end of life. http://legeforeningen.no/Emner/Andre-emner/Publikasjoner/Retningslinjer/Guidelines-for-palliative-sedation-at-the-end-of-life-2014/, accessed 30 July 2017.
- Rousseau, Paul. 2007. Palliative Sedation. In Principles and practice of palliative care and supportive oncology, ed. Ann Berger, John L. Shuster Jr., and Jamie H. Von Roenn, 623–632. Philadelphia: Lippincott Williams & Wilkins.Google Scholar
- Royal Dutch Medical Association (Committee on the National Guideline for Palliative Sedation). 2009. Guideline for Palliative Sedation. https://www.knmg.nl/web/file?uuid=f63bc1b5-ba8d-405b-b901-d3ccc37b6123&owner=5c945405-d6ca-4deb-aa16-7af2088aa173&contentid=550&elementid=1890448, accessed 30 July 2017.
- Seckler, Allison B., Diane E. Meier, Michael Mulvihill, Barbara E. Cammer, and Paris. 1991. Substituted judgement: How accurate are proxy predictions? Archives of Internal Medicine 115 (2): 92–98.Google Scholar
- Sy, Jann Adriel, Mark M.J. Tan, and Lalit Kumar Radha Krishna. 2015. A review of decision-making models in end-of-life care in Singapore. Clinical Case Reports and Reviews 1 (8): 169–172. https://doi.org/10.15761/CCRR.1000157.
- Tan, Jacinta O.A., and Jacqueline Joon Lin Chin. 2011. What doctors say about care of the dying. Singapore: Lien Foundation.Google Scholar
- Veterans Health Administration (National Ethics Committee). 2006. The Ethics of Palliative Sedation.Google Scholar
- Yang, Grace M., Ann K. Kwee, and Lalit Kumar Radha Krishna. 2012. Should patients and family be involved in “do not resuscitate” decisions? Views of oncology and palliative care doctors and nurses. Indian Journal of Palliative Care 18 (1): 52–58. https://doi.org/10.4103/0973-1075.97474.
- Zhuang, Qingyuan. 2016. Applying the welfare model to an ethically challenging terminal discharge. Asian Bioethics Review 8 (1): 66–76. https://doi.org/10.1353/asb.2016.0005.