In healthcare, it’s ethically simpler if a person can make her own decisions. As I’ve written before, if someone can decide for herself, then consent and refusal are up to her. Decision-making capacity, as this is called, makes the ethical analysis much more straightforward. When a person might lack capacity, it gets more complicated. This is what drives a lot of the unease people have about assisted dying for people with mental disorders, since people assume that mental illness interferes with capacity. As I’ve written, sometimes mental illness affects capacity, but not always, so my view is that we should assess on a case-by-case basis instead of having a general prohibition on assisted dying for people with mental disorders.
The approach I take—reflected in this newsletter’s name—is that medicine is about promoting patient values. Determining if an intervention is worth it or safe enough or too risky is informed by science, but it’s really about value judgments. My view is that, since these decisions concern people’s lives, people should be allowed to make these judgments for themselves.
This approach doesn’t work for all cases, however. People lose or lack decision-making capacity, so we need another approach when people can’t make their own decisions. One is to have someone else give consent on the patient’s behalf, such as when a parent consents for a child. The other is for the patient to make a decision when he has capacity, then instruct his healthcare providers to respect his decision when he’s no longer able to consent. Decisions about code status—whether a patient will be resuscitated should his heart stop—are of this form.
These two approaches aren’t mutually exclusive. Think of it as two separate questions: Who is making decisions, and what are they basing those decisions on? For a patient with capacity, she should make her own decisions and she can decide basically however she wants. When a person loses capacity, the law usually states that her substitute decision-maker must make decisions in line with her wishes, if known. In some places, people can consent ahead of time, while in others, such as Ontario, advance consent isn’t permitted. You can state your wishes, but, when the time comes, someone still has to give consent on your behalf.
When assisted dying was first legalized in Canada, the law required consent immediately before the assisted death. This approach offered multiple opportunities to change one’s mind, but it has downsides. For one, pain medication can affect capacity, so people would sometimes have to forego pain medication to retain capacity. Or they would have an assisted death sooner to ensure that they wouldn’t lose capacity to consent.
In 2021, the law was changed to allow a ‘waiver of final consent’ for MAID. Once someone is assessed, approved for MAID, and a date has been set for the provision, if she’s at risk of losing capacity, she and her provider can make a plan for her to have an assisted death on the planned date if she no longer has capacity. With this change, concerns about pain management and other factors that affect capacity became less important, but the timeline is still fairly short, since the patient must already have a condition that qualifies her for MAID. (Additionally, this waiver only applies in cases where the person’s death is ‘reasonably foreseeable’.)
Many people, including me, believe the law should be changed to allow for a longer timeline by allowing people to make advance requests. These allow people to describe in writing the circumstances under which they would want an assisted death should they lose the capacity to consent. Something like this: “If I get dementia and lose capacity, I’m requesting that I have an assisted death once I’m no longer able to consistently recognize my loved ones.” The patient’s physician could then act on his wishes when the conditions are met.
The issue of advance requests was reviewed by The Special Joint Parliamentary Committee on Medical Assistance in Dying, which recommended changing the law to allow for advance requests “following a diagnosis of a serious and incurable medical condition, disease, or disorder leading to incapacity.” Since this requires a diagnosis before allowing the request, it’s a moderate change.
Dying with Dignity’s latest poll found that support among Canadians for advance requests is high: 82 percent of Canadians support them for people with a grievous and irremediable medical condition (i.e., the Special Joint Committee’s recommendation), while 72 percent of Canadians support advance requests for people without a qualifying medical condition (meaning they could say, for example, “if I get dementia, then I want MAID”). Other groups, such as the Alzheimer’s Society of Canada, also support advance requests.
The Ethical Case
Advance requests are widely used outside of MAID. People care about what happens to them even when they’re no longer capable of giving consent, and I suspect that advance requests for MAID are so popular because so many people have watched a loved one decline from dementia. For many, being unable to recognize their loved ones or participate in activities that give life meaning leads them to conclude they would rather die instead.
Currently, people can make requests or refusals for other types of care through advance directives, which are also called personal directives, living wills, and advance care planning. As I mentioned above, signing a Do Not Resuscitate order is a type of advance directive: “Should my heart stop, don’t try to resuscitate me.” I can also specify that, should I no longer be able to meaningfully interact with my family, I want my care to focus on keeping me comfortable, even if that means a shorter life. I could also specify that, if I require mechanical ventilation and a month goes by without improvement, I want the ventilator disconnected, even if this means I will die. My substitute decision-maker is legally obligated to make decisions based on my directive, though this doesn’t always happen.
We have control beyond treatment as well. You can make an advance request to donate your organs. You can also specify what will happen to your property through your will. The idea of extending control over what happens to our bodies when we’re no longer able to consent is commonplace—MAID is an exception—so the ethical basis for advance requests for MAID is a simple consistency argument: If there’s a case for following someone’s wishes about healthcare, this should include assisted dying. We can already request death through the removal of a ventilator, so we should be able to do the same for MAID.
What Kind of System Do You Want?
The reason this isn’t already allowed is that, if a patient lacks capacity, people might doubt that assisting her death is justified or what she really wants. There are a few different issues that advance requests raise and, depending on how serious you think those issues are, different responses.
In practice, the main issues are how far out people should be able to make a request and what the oversight process should be. The Special Joint Parliamentary Committee recommends a moderate approach with advance requests “following a diagnosis of a serious and incurable medical condition, disease, or disorder leading to incapacity.” In that case, you know what the medical issue is and can make plans specifically for it.
The downside of this approach is that other tragedies can befall us. Were I to end up in a coma or a persistent vegetative state, I would want an assisted death once it became unlikely that I was going to recover. I would feel better knowing that I could specify this in an advance directive, but I can’t if I have to wait for a diagnosis of a serious and incurable medical condition. On the other hand, as I’ll describe shortly, this kind of projection into the future comes with its own issues.
The other issue is oversight. Currently, the waiver of final consent is between the patient and the MAID provider. There’s a thorough, worked-out process, and, since the timeline is short, issues are less likely. Importantly, it isn’t a request, but opting out of final consent. Once the conditions are met, the provider who oversaw the waiver of consent, proceeds with the assisted death.
If longer timelines are allowed, such as in my coma example, there needs to be more oversight. Advance directives are made to one’s future providers, whoever they turn out to be. This is different from a waiver, which is made with a specific provider who can ensure the patient understands the implications of the process. In contrast, in most places, completing an advance directive requires only a witness, who doesn’t need to be a healthcare provider.
The role of substitute decision-makers was created because, before this, physicians had a lot of power and basically no oversight. But there are also problems with substitute decision making. Most clinical ethics consultations concern substitute decision-makers for the simple reason that, when the patient has capacity, we can defer to them. When it’s a substitute decision-maker, more problems can arise. Sometimes they make decisions that clearly go against the patient’s wishes. When this happens, replacing the SDM is difficult (going to court is a lot of work). In other cases, the SDM might not have spoken to the patient in decades, or the SDM will say “I know this is what she wants, but I can’t be the one who makes this decision.”
Advance Directives Have Problems
There are also problems with the directives themselves. In some cases, they’re too vague. They say things like “I don’t want to be hooked up to machines” or “I don’t want heroic measures”, which leaves the physicians, family members, and the clinical ethicist to wonder if she really meant no machines whatsoever—not even an IV drip?—and what constitutes a heroic measure.
In other cases, they’re too specific. If you see a loved one suffering due to a disease, then it’s understandable that you’ll develop strong preferences about how you want to be treated if you get that disease. This leads people to focus narrowly, such as a degenerative disease such as ALS, but ignore other situations. The result is that the advance directive doesn’t apply, or at least doesn’t seem to, for the situation the person ends up in.
The current approach of allowing a waiver of final consent changes the consent process. Typically, you need to give consent immediately before the MAID provision. The waiver eliminates that step, but you’ve still given informed consent. The provider goes over all the details, you make a plan, then you consent in writing.
Advance directives aren’t a substitute for consent, since you can make one without being informed (they aren’t decision-specific). You could write one now that will influence your care years from now, and likely won’t be specific to the situation you’ll end up in. Or you could write one based on false information. People tend to have mistaken beliefs about cardiopulmonary resuscitation, for example. They think the survival rate is much higher than it is, that it’s quick and painless, and that, once it’s over, you’ll wake up and go on about your business. In most cases, none of this is true. This applies to MAID, since people might make decisions based on false information, but it’s a more general problem.
These are genuine issues, but the solution is to improve advance care planning instead of doing away with advance directives. There are lots of good resources for improving communication about end-of-life wishes, such as The Conversation Project. The solution is to encourage people to discuss their wishes more, not do away with advance directives.
What’s It Like to Have Dementia?
A more philosophical issue is that various states of lacking capacity will be quite different to how we experience the world now. Dementia is such a disease. Some people experience distress, while others seem to be content. It’s hard to know what it’s actually like, and it’s easy to project how we think it will be. Since we don’t really know, it makes making informed decisions more difficult.
For these reasons, some philosophers conclude that we shouldn’t use advance directives. Instead, we might use a best-interest standard: if a person seems content, we should disregard an advance directive that says he doesn’t want medication that will prolong his life. Since we don’t really know what it’s like to have dementia, we can’t make decisions in advance about it.
I appreciate this argument, but I don’t find it convincing. Sometimes we make decisions about our lives without knowing what the result will be or how it’s going to feel. Consider what are sometimes called transformative experiences. Having a child or falling in love are examples. Before you fall in love, you can read all the sonnets in the world and listen to every love song, but you don’t really know what it’s like to fall in love.
When my wife was pregnant, someone told me that, once my daughter was born, I wouldn’t be able to remember what life was like before her. And it’s true! It’s hard to remember the experience. I think a liberal arts education is also transformative: you won’t come out the same as you went in—maybe you literally won’t be the same person—but it’s impossible to anticipate the differences ahead of time.
Given the nature of transformative experiences, we undertake them without complete information. Maybe having a child is the right decision, or maybe not. We can’t know with certainty beforehand. But stopping people from having children on these grounds isn’t justified, so I’m inclined to say that advance directives are okay too.
The Problem of Suffering
The standard ethical case for assisted dying is that it promotes well-being by preventing suffering, and it promotes autonomy by letting people die on their own terms. In U.S. states where assisted dying is legal, instead of a suffering condition, they require only that natural death is likely within six months. In contrast, Canada’s approach explicitly requires a person to be experiencing intolerable suffering to qualify.
If Canada makes advance requests for MAID legal, this might have to change. Consider a person with end-stage dementia who has an advance directive stating that, once she’s in this state, she wants an assisted death. She might meet the criteria that she set for herself when she had capacity, but she might not be suffering. The same is true of someone in a persistent vegetative state, who might be incapable of feeling pain.
So the law might need to change to allow for advance requests for people without suffering. This is true even if The Special Joint Committee’s moderate approach is adopted. It recommends allowing requests “following a diagnosis of a serious and incurable medical condition, disease, or disorder leading to incapacity.” It doesn’t say anything about suffering.
This suits me fine. Jeremy Davis and I argue in our paper “The Case for an Autonomy-Centred View of Physician-Assisted Death” that the suffering condition is already unnecessary. Requiring suffering is too restrictive; we should focus on value judgments instead.
The ethics of advance requests for MAID are complex because the ethics of advance requests in general are complex. Since we allow requests for other types of care, including at the end of life, there’s a good case for allowing it for MAID as well.
Quick Hits
Here are some other stories that have caught my attention.
Give homeless people cash and they spend it on housing, UBC study finds.
Men in England to be offered blood pressure checks in barbershops. Since the relative risk of heart attacks is twice as high for men, it’s good to see some gender-specific outreach.
When Trucks Fly. I wasn’t looking for an article on monster trucks, but this New Yorker story is one of the best I’ve read in a while.