By Prof, James Downar and Prof. Sam H Ahmedzai

Take Home Messages:

• We can never be 100% certain that an unconscious person is comfortable, but this is more of a concern for palliative sedation than for assisted dying.
• People may feel pressured into assisted death by family members, but we routinely allow the same family members to make end-of-life decisions for incapable patients without any oversight; we cannot be more concerned about autonomous decisions than about surrogate decisions.
• Social factors and mental illness can influence decision-making, but physicians routinely determine decisional capacity for analogous decisions (eg. stopping ventilators, dialysis and chemotherapy) without mandatory psychiatric or social work consultations.

As two doctors specializing in palliative medicine – one from Canada and one from Britain – we detect a recent shift in emphasis in the debate about assisted dying. Opponents have moved beyond the longstanding moral and ethical arguments to conjecture that assisted dying might not provide comfort, and may even cause suffering or complications at the end of life.

A recent Forum posting raised the possibility that people who receive an assisted death may be experiencing acute suffering despite appearing comfortable to an observer.¹ In truth, we can never be 100% certain about the comfort or awareness level of a person who appears to be unconscious and unresponsive. In an Australian single centre study, patients who became unresponsive while receiving analgesia in a palliative context (not intended  to cause deep sedation) could still show objective responses on tools such as bispectral index monitoring to breakthrough injections of opioids and benzodiazepines, indicating that unresponsiveness in non-assisted dying may hide continuing physical distress.² A Japanese multicentre study showed that in patients who were receiving palliative sedation, up to 17% of cases still showed visible signs of distress while sedated.³ So we respectfully point out that persistent suffering appears to be a routine part of normal end of life care, even with palliative sedation.

The risk of acute suffering in assisted dying is more difficult to estimate. Large studies suggest that patients are aware during 1 in 20,000 cases of general anesthesia, or 1 in 10,000 cases with neuromuscular blockade.⁴ No such evaluation has been done during assisted dying, but the incidence of awareness may vary more or less around these numbers if oral sedatives are used (as in the USA), or if very high doses of propofol are used (as in Europe and Canada). At any rate, with such estimates, it is hard to understand how we could question the comfort of assisted dying without questioning the comfort of palliative sedation to a far greater degree.

Moreover, recently published guidelines on palliative sedation from the European Association of Palliative Care recommend subcutaneous or intravenous benzodiazepines as a first line sedative, with propofol as a third-line option, titrated on the basis of distress assessed by the bedside clinicians.⁵ If this subjective approach to assessing comfort is considered adequate for a palliative sedation procedure that can last days or even weeks, then how can it be inadequate for a relatively brief closely-monitored procedure such as assisted dying?

Other opponents have suggested that assisted dying might even cause suffering through complications such as vomiting. Some US reports have described vomiting or other forms of distress in a small proportion of patients who receive oral medications.⁶ However a large retrospective Canadian study of 3557 cases of MAiD using intravenous propofol, raised no serious adverse events other than 1.2% who had problems with IV access or insertion of a second cannula, and only two reports of pain at the injection site.⁷ These numbers must be compared to the fairly common experience of injection site pain, and the need to resite the subcutaneous infusion cannulae commonly used in palliative care at end of life.

Opponents of assisted dying have also raised the possibility of pulmonary edema causing suffering during an assisted death, based on a non peer-reviewed post-mortem case series that reported pulmonary edema in the lungs of 33 out of 43 people judicially executed using benzodiazepines or barbiturates.⁸ Whether or not this study (or any subsequent report) ultimately survives peer review, it offered no argument to support the assertion that pulmonary oedema occurring in the minutes during or after death could have led to any cerebral experience of suffering. These post-mortem findings are also impossible to reconcile with the observation that lungs donated by people who receive assisted death by propofol appear to show no impairment of oxygenation immediately after they are retrieved, and function as well as lungs donated following neurological death (where donors do not receive any sedatives at all).⁹  Therefore, there is no rationale for applying this theoretical concern to the practice of assisted dying as a whole.

Other recent arguments against assisted dying include claims that people may request assisted dying under coercion from family members, from social distress, or untreated mental illness. Some have suggested that these concerns justify prohibiting assisted death entirely, while others have proposed mandatory consultations to psychiatry or social work for every applicant to rule out coercion or address social distress.

These arguments and proposals overlook the fact that people routinely make analogous end-of-life choices without requiring additional scrutiny for these decisions. Palliative care physicians obtain informed consent for palliative sedation, oncologists obtain informed consent for people to stop anti-cancer treatments, and nephrologists will do the same for people who wish to stop hemodialysis. In all cases, the decision may impact survival and could be influenced by mental illness or social distress, but physicians are left to assess decisional capacity and consult with psychiatrists and social workers at their discretion. Moreover, as many as 30% of older adults require important care decisions (including decisions to withdraw life-sustaining measures) to be made by surrogate decision-makers due to incapacity.¹⁰ If the potential for family coercion is so great that we cannot trust decisionally-competent people to determine whether they should receive an assisted death, how could we possibly be comfortable allowing the same family members to make life-and-death decisions for a decisionally-incapable patient, entirely on their own, without any oversight or accountability at all?

We respect that many palliative care providers are concerned about the practical aspects of assessing for, and providing assisted death. However, we also respectfully point out that many of the concerns being raised in the assisted dying debate apply to the same or greater degree to aspects of routine palliative care. We cannot apply the concerns we have described above selectively: all deaths deserve the same degree of oversight and due diligence in ensuring a peaceful dying.

References:

1. Thomas B, Barclay G. Do they go gently? BMJ Supp Pal Care, 2025. https://blogs.bmj.com/spcare/2025/03/20/do-they-go-gently/
2. Barbato M, Barclay G, Potter J, Yeo W. Breakthrough Medication in Unresponsive Palliative Care Patients: Indications, Practice, and Efficacy. J Pain Symptom Manage. 2018;56(6):944-50.
3. Morita T, Chinone Y, Ikenaga M, et al. Ethical Validity of Palliative Sedation Therapy: A Multicenter, Prospective, Observational Study Conducted on Specialized Palliative Care Units in Japan. J Pain Symptom Manage, 2005;30(4):308 – 319.
4. Pandit JJ, Andrade J, Bogod DG, et al. The 5th National Audit Project (NAP5) on accidental awareness during general anaesthesia: summary of main findings and risk factors. Anaesthesia. 2014 Oct;69(10):1089-101. doi: 10.1111/anae.12826. PMID: 25204236.
5. Surges SM, Brunsch H, Jaspers B, et al. Revised European Association for Palliative Care (EAPC) recommended framework on palliative sedation: An international Delphi study. Palliat Med. 2024 Feb;38(2):213-228. doi: 10.1177/02692163231220225
6. Oregon Death with Dignity Annual Report (2024). Accessed on May 4, 2025 at https://www.oregon.gov/oha/PH/PROVIDERPARTNERRESOURCES/EVALUATIONRESEARCH/DEATHWITHDIGNITYACT/Documents/year27.pdf
7. Stukalin I, Olaiya OR, Naik V, et al. Medications and dosages used in medical assistance in dying: a cross-sectional study. CMAJ Open. 2022;10(1):E19-26.
8. Zivot JB, Edgar MA. Lubarsky DA. Execution by lethal injection: Autopsy findings of pulmonary edema. Non-peer reviewed pre-print online https://www.medrxiv.org/content/10.1101/2022.08.24.22279183v1.full.pdf
9. Watanabe T, Kawashima M, Kohno M, et al. Outcomes of lung transplantation from organ donation after medical assistance in dying: First North American experience. Am J Transplant. 2022 Jun;22(6):1637-1645. doi: 10.1111/ajt.16971.
10. Silveira MJ, Kim SY, Langa KM. Advance directives and outcomes of surrogate decision making before death. N Engl J Med. 2010 Apr 1;362(13):1211-8. doi: 10.1056/NEJMsa0907901.

Authors

Prof. James Downar

Department of Medicine, University of Ottawa, Canada.

Declaration of interests

I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: Financial interest in Ampa Inc, which manufactures devices that deliver Transcranial Magnetic Stimulation.

Prof. Sam H Ahmedzai

Emeritus Professor, The University of Sheffield, UK

Declaration of interests

I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: Invited speaker at Jersey Citizens’ Jury on assisted dying, April 2021; work done online. Member of Medical Advisory Group to Liam McArthur MSP for his Scottish Parliament Bill on Assisted Dying, 2022-24; work done online. Provided advice on palliative and end of life and assisted dying issues, to Isle of Man House of Keys at request of Alex Allinson MHK. Work done was by email and a visit to Isle of Man in May 2024 with one night stay. Travelling costs and accommodation covered by Dignity in Dying. Invited to speak to SNP fringe meeting on assisted dying at SNP party conference, Edinburgh, September 2024. Cost of one night accommodation covered by Dignity in Dying. Invited to give oral evidence to the Terminally Ill Adults (End of Life) Bill committee in January 2025. Travelling costs paid by House of Commons.