How Many Assisted Deaths Should There Be?

The best answer to an uninteresting question

Source: Health Canada. “RFND” stands for ‘reasonably foreseeable natural death’.

I haven’t heard a new argument against assisted dying in a while, but I notice that, as with fashion trends, something is in vogue for a while, then people move on to a different point as surely as hemlines rise and fall. These days, almost everyone who emails me with an objection to assisted dying mentions the number of assisted deaths that have occurred in Canada as evidence that the system is broken. Some messages, as in a tweet I received after I had finished a draft of this article, are only about the number.

The official number of assisted deaths in Canada since 2016 is a bit over 30,000 people, but that doesn’t include 2022 or 2023, so the actual number is probably closer to 50,000.

People use this number as though it’s proof of something sinister, or, as in the tweet, in place of an argument. Take a forthcoming article in the American Journal of Bioethics, where Daryl Pullman says things like “The numbers for the years 2020 and 2021 are disturbing” and “the Canadian numbers are disturbingly high, and they are trending upward.” 

On their own, statistics aren’t arguments, so let’s do the quick Q&A version first. How many assisted deaths should there be? I don’t know because there’s no good answer to this question. What should we make of the ten-thousand-a-year number? Almost nothing. Isn’t it a sign of something disturbing or wrong with society? Not without more evidence. Isn’t its upward trend a bad sign, indicating contagion? No.

The longer version is that, sometimes, data on health outcomes and the number of people accessing services can be informative. Heart attacks are bad, so if the rate of heart attacks is increasing, we should be worried and try to do something about it. This is, of course, a value judgment, but it isn’t a controversial one. What about increased rates of coronary artery bypass surgery? More surgeries might be positive, since perhaps more people are getting access to the care they need, or it might indicate that surgeons are being too quick to pull out their scalpel or that coronary artery disease is on the rise. We have to look more.

Next, consider procedures such as vasectomy and tubal ligation (having your tubes tied). How many of those should there be? At around 55,000 procedures a year, Canada has one of the highest rates of vasectomy in the world, and it’s one of the few countries where there are more vasectomies than tubal ligations. Is that number too high or too low? What’s the ideal number?   

If you’re the Catholic Church, you’ll say that the number is too high and that the ideal number should be zero, since vasectomy and tubal ligation interfere with procreation. But if you don’t have a prior view about the morality of birth control, your answer should be “it depends” or “there’s no good answer to these questions”. People should have vasectomies if they want them, and the numbers tell us little. (Though the significant post-Dobbs vasectomy spike in the U.S. shows how other factors can influence these decisions.)

Suppose someone were to ask, “but doesn’t Canada’s high rate of vasectomies indicate a sign of vasectomy contagion? Isn’t 55,000 procedures a year disturbingly high?” Well, no. Not without more evidence or an argument. On the contagion point, it might be that more people are finding out about vasectomy and choosing to get it for themselves, but calling this ‘contagion’ is making a value judgment that needs defending. We need to know why it’s bad—and why we should oppose efforts to spread the word about the procedure—before ‘contagion’ becomes appropriate.

There are important differences between vasectomy and assisted dying, but the same points apply to both. Appealing to the number of assisted deaths as evidence of a problem only works if you’re already against assisted dying, in which case, you aren’t making an argument, you’re just assuming the conclusion you already believe. We should no more accept the raw numbers as evidence of a problem than we should conclude that 55,000 is too many vasectomies.   

‘How many assisted deaths should there be?’ isn’t the right question. The better ones are ‘how many people qualify?’ and ‘how many people want it?’