As of today, Quebec is allowing advance requests for medical assistance in dying. It works as follows. If you have decision-making capacity now and you’re diagnosed with a serious and incurable illness that will lead to incapacity, you can make an agreement with a physician or nurse practitioner that, at a future date when specific clinical manifestations chosen by you are met, the provider will give you a medically assisted death.
The primary purpose of this change is to allow people diagnosed with dementia to have MAID. The federal criminal code permits MAID only when a person is capable of consenting at the time of the provision, which often excludes dementia because, by the time the disease is “in an advanced state of irreversible decline in capability”, the person is no longer able to consent.
There are other details to Quebec’s approach. There is a standardized form the patient has to fill out and the request is then put in a provincial register. Before the provider administers MAID, she must check the register to ensure that she’s using the correct request. (This will prevent providers from acting on an outdated request.) There is also a specific form patients must use to rescind their requests, and doing so requires a physician or nurse practitioner to verify that they still have capacity.
To proceed with the provision, the provider must verify that the patient is exhibiting, on a recurring basis, the criteria stated in the patient’s advance request. The provider must also verify that the patient is “experiencing enduring and unbearable physical or psychological suffering that cannot be relieved under conditions considered tolerable,” which is the same criteria from the MAID legislation.
Although the terms are similar, there are important differences between advance requests for MAID, which is what Quebec is allowing, and advance directives, which Quebec isn’t allowing for MAID. An advance directive lets you state wishes about your treatment that will apply at any point in the future, whether or not you’ve been given a medical diagnosis. For example, I could write an advance directive now saying that I don’t want to be kept on a ventilator if there are low odds that I will ever regain capacity. Then, were I to end up in that kind of situation, my substitute decision maker would be legally required to make medical decisions for me based on that directive. The physician would be killing me by disconnecting the ventilator, but I can’t consent to that ahead of time. I can only direct treatment decisions.
Advance requests in Quebec are much more limited. They require a concrete medical diagnosis, not the hypotheticals allowed for advance directives. This means they will also be more temporally limited. Advance requests are also an agreement between the patient and a specific provider, whereas advance directives are more “to whom it may concern”.
Last year, I wrote a long post arguing for the ethical permissibility of advance requests, including an approach that is more permissive than Quebec’s. Although I’m in favour of the experiment, I’m against how Quebec went about it. Essentially, the Quebec government said “advance requests will now be legal here”, even though that violates the criminal code of Canada. Since it’s Quebec, they could be confident the federal government wasn’t going to intervene, and they were right. The Attorney General of Canada, along with the federal ministers of health and justice, put out an awkward statement which says, essentially, “To be clear, this is still illegal but we aren’t going to do anything about it.” This sets a bad precedent. The federal government is now launching a ‘national conversation’ on advance requests.
When it comes to arguments about assisted dying, the person I’ve learned the most from is Wayne Sumner, whose two books, Assisted Death: A Study in Ethics and Law and Physician-Assisted Death: What Everyone Needs to Know, are both excellent. In his honour, I hereby propose Sumner’s Law, which holds that most objections to assisted dying commit the person raising the objection to some type of inconsistency.
It was an example by Shelly Kagan, not Sumner, that convinced me that disconnecting a ventilator is killing, but it’s a clear instance of Sumner’s Law: Anyone who objects to assisted dying on the grounds that it’s killing and killing is always wrong must also be against disconnecting ventilators. Since they aren’t against disconnecting ventilators (I hope!), they’re being inconsistent.
Sumner’s Law comes up all over the advance request discussion. Most objections to advance requests to MAID commit the objector to some conclusions in other parts of medicine that are difficult to accept. For example, my Twitter timeline is full of people saying that advance requests are wrong because the person can’t consent at the time of the MAID provision. The problem is that, if you think consent is always required in the moment, consistency demands that you think a lot of other medical practices are unethical: Do-not-resuscitate orders, any procedure involving sedation, and advance directives are just a few examples. Since I don’t think these procedures are wrong, I think people are wrong about the consent point for MAID advance requests as well.
The Philosopher’s Brief
More philosophically, you might oppose advance requests for MAID on the grounds that the person who made the advance request has, in a literal sense, ceased to exist. This might seem weird—and it is!—but it’s a popular view among philosophers and aligns with many non-philosophers’ assessments of what happens to people who get dementia and other conditions. This is why Nancy Cruzan’s gravestone reads “Born July 20, 1957. Departed Jan. 11, 1983. At peace Dec. 26, 1990.”
Many people believe that what it means for us as persons to exist is distinct from what it means for our bodies to be alive. One popular view, the psychological continuity view, says that existing across time is about being psychologically connected by our memories and forward-looking desires. According to this view, someone with dementia will gradually cease to exist even as his body remains alive. Another theory is that personal identity is about being capable of consciousness. According to this theory, people with dementia don’t cease to exist in the same way, but there’s still a split between their wishes pre-dementia and who they are once dementia advances.
Suppose that the individual with dementia has no memory of ever making a request for MAID. The skeptic wonders why we should honour the advance request, given that the person who made it minimally doesn’t remember doing so and maximally doesn’t exist. The problem becomes even more pressing if we imagine the patient with dementia shaking her head or saying “no” when the MAID provider arrives. Even more difficult would be a case where, contrary to the prediction made by the capable person who made the advance request, the individual with advanced dementia is content rather than suffering. (Though, importantly, MAID wouldn’t be permissible under Quebec law in this latter situation, since the individual has to be experiencing “enduring and unbearable physical or psychological suffering.”)
Here, I’ll follow Jeff McMahan’s argument. McMahan’s book, The Ethics of Killing, has been tremendously influential, not only because of the stuff about killing but also his theory of personal identity. (He also raises serious problems with the whole-brain definition of death.) According to McMahan, our capacity for consciousness is what makes us us.
McMahan argues that, if the individual with dementia gives no indication that she’s against MAID, then we should honour the advance request because respecting people means honouring their autonomous preferences. Per Sumner’s Law, if we deny this in the case of advance requests, then we should also deny that we have a duty to honour a person’s preferences about her property or body as stated in her will.
Here’s McMahan:
If we regard it as a failure of respect for the person to disregard these preferences—and most of us do—then we should also accept that respect for a person can require that we honor the autonomous preferences she previously expressed concerning how she is to be treated after she ceases to be competent (p. 497).
In the case where the individual with advanced dementia now has a conflicting preference with the person who made the advance request, McMahan argues that we should honour the advance request because, by definition, the individual with dementia is making a non-autonomous, and therefore uninformed, decision. Respecting an individual requires us only to honour her autonomously formed preferences. If the conflict is between an autonomously formed advance request and a non-autonomously, uninformed refusal, the former should get priority. This is consistent with other medical situations: although refusals by people without capacity can sometimes have best-interest effects that justify respecting the refusal, in those cases, we respect the refusal because doing otherwise is bad for the patient’s well-being, not because the refusal itself should be respected on autonomy grounds. After all, the patient without capacity doesn’t have autonomy to promote.
This is a brief discussion of an extremely complex point, but it’s worth raising because I think some people attracted to McMahan’s embodied mind account or the psychological continuity account might not realize that neither justify the rejection of advance directives or advance requests. It’s true that things get more complicated when the prior autonomous wish of the individual who made the advance request conflicts with the current best interest of the patient with advanced dementia, but, even then, McMahan believes that the person’s prior autonomous wishes take priority. (If you’re interested, check out his discussion in The Ethics of Killing around p. 502.)
Even if these points turn out to be mistaken, the skeptics still face the consistency point that, if they’re against advance requests for MAID, then they should also be against advance directives. I think advance directives are good and appropriate, despite the shortcomings with them I discussed in my other advance request post. So, I conclude that we’re stuck with allowing advance requests.
Assisted Dying in Spain and The Netherlands is Working Great
As it happens, Quebec isn’t the most permissive jurisdiction on this issue. Spain legalized assisted dying in 2021, and the law treats a request for an assisted death in an advance directive the same as anything else the person includes. The Netherlands also allows people to request assisted dying in an advance directive without discussion with a physician.
Overall, both systems are working well. International media covered a case from The Netherlands in 2019 of a physician who was investigated for administering euthanasia on a patient with advanced dementia. However, that provider was cleared of all wrongdoing. There’s a good summary of the case here.
This isn’t to say that there are never problems, but advance directives also cause problems. Most people correctly believe they’re acceptable. Advance requests for MAID are also nearly as popular as MAID itself: the best polling we have shows support above eighty percent. Given this, I’m optimistic that Quebec’s experiment will prove successful. The rest of Canada will be watching closely.
I do think someone with advanced dementia is a different person than they once were. What, then, do we do with a patient who made an advance request but now that they've lost the capacity to decide otherwise (but meets the criteria) appears to be enjoying listening to music and eating their meals?
What I worry about here is what others have already discussed: the elevation of critical interests over experiential interests. This advance request to be killed is a critical interest. Other critical interests are someone's hope to be a good father, brother, son, mechanic, citizen, etc., to build a table, to run a marathon, etc. Not everyone can form critical interests. People with dementia could have done so at one time, but others, like those born with severe intellectual disabilities, never had the capacity to do so.
Experiential interests are focused on sensations. Most of us have an experiential interest in slaking thirst, satisfying hunger, avoiding pain, etc. Even people with severe dementia can continue to have experiential interests.
So, when you have someone who has forgotten they made an advance decision based on a critical decision, but are living their life oriented around experiential interests, if you honor that previously expressed wish you're going to elevate critical interests above experiential interests, suggesting that one is more valuable than the other. This is step closer to, if not an embrace of, the belief that some lives are not worth living because of their rational capacities. It is the judgment that builds a culture that points at someone with dementia and says, "We never want to live like that," when, in fact, the person with dementia might not have any qualms about continuing to live now.
This point doesn't address the situation in which someone with dementia appears to have trouble meeting their experiential interests (e.g., they're in unremitting pain or afflicted with persistent bouts of distressing psychosis with behavioral disturbances), but it does raise concerns about the validity of living wills in many circumstances for people with dementia. It doesn't necessarily invalidate surrogate decision-making for people with dementia, though. Rather, it should provoke surrogates (and clinicians) to consider whether they're biasing themselves against *all* people with cognitive impairments when they value critical interests over experiential interests.
I find the curious the anthropology you are using very curious and interesting.
You seem to be operating based on the idea that there is such a construct exists as a "continuous, stable, conscious psychological self" - are you sure you agree this exists? Because as someone with a Master's in counselling, I'm not sure such a thing truly exists. When someone has an eating disorder, what is your approach? Are you sure a consistent autonomous self exists? Because, from my knowledge and experience, it doesn't. Persons are often complex, inconsistent and ambivalence. What do you do as a counsellor? You partner with the healthy side of the patient - the side who wants to live - the side who wants to become healthy, and you form an alliance with that part of their self in order to overcome the parts of themselves driving the eating disorder.
1. I do not think such a continuous 'psychological self' exists - people are very inconsistent
2. When does such a psychological self begin? I don't imagine it would be there in, let's say, young children.
3. It also seems quite arbitrary that this "self" all-of-a-sudden becomes discontinuous when someone gets something like "dimension" - but not, lets say, any number of other neurological, psychological or other disorders. Why is it not the case that persons undergo change and transformation in their psychological self through out their lives (childhood self, adolescent self, parent self, career self, retirement self, physically sick self, dementia self). Your argument seems inconsistent to me. Of course our psychological selves undergo change when we get dementia, but do they not also change in light of any number of other life circumstances, physical, mental or other brain disorders? Or does this reveal our culture's own "ideal" self - the autonomous, independent, un-relational, 'well-intact', 'in control' conscious self, who charts his own path and is dependent on no one. This does not seem like most people to me - but only fits a certain stereotype of person - which is very exclusive and leaves out many other human beings.