Looking for safeguards in Assisted Suicide legislation – a 10 point checklist

Concerns about decriminalising assisted suicide, typically so that doctors can help people take their own lives, centre on one of four main themes:

• Religious or moral objections.  They are important for those who hold religious beliefs but not for those who don’t – apart from the (often neglected) need to respect the expression of sincere views, as far as possible, of others. Their force is undercut somewhat by the observation that senior members of religious groups are found on both sides of the assisted suicide debate.
• The likelihood of development from early restrictive use to wider application, the slackening of rules about who can assist or about cooling off periods – usually called the slippery slope argument. Campaigners for assisted suicide offer bland reassurance that no such thing is possible here – we are apparently unlike Canada, the Netherlands, Belgium, New Mexico, Vermont, or Washington State. Or like Oregon or Colorado, where slipperiness resides in re-interpreting the rules rather than rewriting them.
• Worries that for many people choices are so restricted that real autonomy does not exist even in the absence of overt coercion. These worries are most evident in the reservations of those who point to the unavailability of comprehensive palliative care and end-of-life services, which depend heavily on charitable funding to deliver what they can.
• A lack of immediate safeguards for the vulnerable. Supporters urge that proposed legislation is chock full of safeguards, while forwarding skeleton proposals that either have none of the relevant safeguards or defer their development to the executive, using so-called Henry VIII clauses. In Falconer’s House of Lords Bill many of the decisions were delegated to the Secretary of State, which creates an interesting dilemma if it is copied by Kim Leadbeater when she eventually lets us see her Bill: he has already said he will vote against it.

Here are my thoughts about what safeguards should look like, in a simple 10 point checklist – based upon the question – what would we expect in any other area of healthcare where we want to see competent and conscientios doctors working in a safe system? There’s a link at the bottom of the page.

What to record	Rationale
How are the practicalities of assisting suicide specified?   There is a lot of evasive vagueness in campaigns for assisted suicide, including avoidance of the term itself. People can’t give an informed opinion if they don’t know in detail what is proposed (think for example of consenting for surgery).
1. How is safe, competent medical practice ensured?
For the doctors involved: what qualifications + experience are mandated how frequently can any one doctor prescribehow often should the same pair they work together, how the pairing is chosen	True independence is hard when only 1% of doctors are willing to participate, as in Oregon. For example that’s 6 GPs where I live in LeedsIn Canada one doctor can choose the second opinion, and seek more than one.The case work indicates a need for experience in elderly care, palliative care or psychiatry Idiosyncratic or “overenthusiastic” practice is a risk.
If expectation of imminent death (say within 6 months) is a criterion, does it specify whether that is with or without treatment?	There are medical conditions such as Type 1 diabetes or dialysis dependent renal failure where death will ensue if treatment stops. Diabetes is listed as the key underlying condition in cases in Oregon and Victoria.
2. How will safe, competent “assistance” be ensured?
How will the decision be made about what drugs can be used with the aim of causing death in each case?	There are no standardised regimes approved by regulatory authorities and it is acknowledged that none is perfect from experience in practice.   There is a need to monitor drugs used and compare to outcomes, otherwise there is no way to ensure acceptable practice.
What additional drugs can legally be used – after use of assisted suicide drugs and before death?	Unpleasant side effects like retching are common (more than 10%), delayed death and not dying are well described (again 10%+).   If a health professional has to stay with the person until death, then they will be under pressure to ensure extra drugs are given in these circumstances. Death by assisted suicide isn’t always dignified!
What exact mode of administration and “assistance” is allowed?	Most people imagine that assisted suicide involves swallowing an elixir or mixture of tablets. But use of a “medical device” and assistance to a very disabled person in using it (as for example in the Falconer Bill, and not excluded by the wording of the McArthur Bill) amounts to something much more like being given a lethal injection aka euthanasia. Will administration be allowed as in Victoria, Australia?   Public standing of the programme depends upon knowing the detail.
3. How will informed consent be ensured?
What form of informed consent is required?	Consent for any other procedure in health care requires informed understanding of exactly what is proposed, what its potential complications and side effects are, and what alternatives there are. The “declarations” in existing legislation do not cover all this ground.
4. Is there adequate ability for healthcare professionals to opt out?
How far does conscientious objection and opting out go?   Can a whole organisation, such as a care home, nursing home or managed group of such, raise conscientious objections?	“No duty to participate in anything authorised by the Act” does not seem to preclude eg requirement to inform patient it’s an option, or to refer if asked. Conscientious objection is not defined.   In Canada dissenting care or nursing homes risk losing their licence.
5. What aspects of assessment are mandated?
In competent clinical practice there is a great deal more to assessing somebody who is saying they want to kill themselves than simply checking mental capacity and lack of overt coercion.
Is there a requirement to obtain: …    Mental health and alcohol use life history Family/household circumstancesRecent life events and difficulties &. financial circumstances	These form an important part of assessing remediable factors in anybody at suicide risk, and certainly anybody who is saying they now actively want to kill themselves. Endorsed in suicide prevention strategies and in clinical guidelines and training.
Is there a requirement to interview a family member or close other?	Part of assessment as above, and can provide important information not revealed by patient eg that points to undue external influence   Close others are interested parties because they are directly affected by the death
Is there a requirement to interview GP and, as relevant, key specialist?	These people have a duty of care that the assisted suicide doctors don’t, and therefore have a right to be consulted.   They should be consulted because they are likely to have privileged information that may not be available otherwise, for example on prognosis or on external factors influencing the decision.
6. What requirements are there for integrated medical record keeping and review?
Is there a requirement for medical records review and if so which, and for what part of the person’s life?	Medical records may contain important information that is suppressed or forgotten by interviewees.
What detailed recording of the assessment is mandated?	If assisted suicide is to be a part of end of life care in the health service, then records need to be kept and form part of the patient record, as for any other contact. “No notes, no defence” is a well known legal caveat. An important element of auditing practice.
7. How will clinical practice be supervised or monitored?   For early career staff, direct supervision is needed. For more experienced staff peer contact and informal observation may suffice – but can’t easily happen in a standalone service.
What are the requirements for contemporaneous supervision or oversight of the clinical contacts?
8. What are the requirements to monitor and report outcomes of the programme, for audit and quality control?   Careful routine monitoring of performance against agreed outcomes (audit) is a standard part of good practice. Lack of it is a near uniform finding in investigations into healthcare failings.
What characteristics will be recorded of those who die by assisted suicide?	Reporting from other countries is often rudimentary – especially in relation to monitoring features of concern to those who worry about how the system protects vulnerable people. Will wider psychological and social features of individuals (as outlined in assessment) be included in monitoring reports?
Time to death/failure to die	An important outcome to determine whether the process is working at an acceptable standard on its own terms   Information that needs to be provided to the patient to obtain informed consent
Adverse events	An important outcome to determine whether the process is working at an acceptable standard on its own terms   Information that needs to be provided to the patient to obtain informed consent
Is there a formal mechanism for monitoring untoward incidents and responding to them promptly?
9. Monitoring impacts on others
Family outcomes   Impact on staff, including emotional impact, staff departures and absenteeism	An important outcome to determine whether the process is working at an acceptable standard. Especially negative psychological outcomes may be long-lasting. There may also be practical implications for families such as disputes over bequests. It is too narrow to regard this process as affecting only one person.   Some of this is information that needs to be provided to the patient to obtain informed consent
10. Is there a formal complaints and feedback process?	Should be routinely sought and collated, as in any healthcare setting.

A list like this could serve two purposes.  It is a template for designing a programme – if these elements aren’t present then the programme cannot even a priori  claim to have sufficient safeguards. And it is a template for what should be collected in a monitoring and reporting a programme, without which claims for safety rest on no evidence base.

In fact no checklist like this has been applied to assisted dying legislation and there are two reasons for that; reasons that explain why such legislation is inherently unsafe. Assisted dying practice everywhere except Switzerland requires medical involvement – it requires doctors to prescribe and oversee death rather than to exercise the sort of duty of care or professional obligation to prevent suicide that is reflected in these safeguards. And legislation is not about (despite the rhetoric) ensuring best care for everybody but about allowing one group of people to have an unconstrained personal choice – framed as a human right. It is the libertarian privileging of individual choice over community responsibility.

For further reading I recommend the excellent The Reality of Assisted Dying: Understanding the Issues eds Julian C. Hughes, Ilora G Finlay Open University Press 2024

There’s a link to a Word version here: