How long do you want to live?

Some important white guy once quipped they only believe statistics they have manipulated themselves. There is no better way to demonstrate this problem than the question "how long do you want to live?", whose misuse has been haunting me to this day.

This question by and large measures two things, and neither of them is a desire for lifespan extension. The first one is the respondent's level of depression and the second one is their aversion to frailty and general level of optimism.

I want to live to 80
If you ask someone "how long do you want to live" there will be two common classes of pessimistic answers. One gives a number in the 30 to 40 range when you ask a young person and the other a number around 80. Here we will demonstrate why these answers are wrong per definition because they evade the question (1) and why it is foolish to even ask this question to begin with, without providing any further explanation.

Let us deal with the answer of around 80. How could it possibly be that so many people choose a number within 10 to 20% of the average life expectancy in the year 2020 if they could have chosen an infinite number of different life expectancies? The natural numbers range between 0 and infinity, so it is an extraordinary coincidence that answers would cluster around a certain value, even more extraordinary that this value is exactly the human average life expectancy. This suggests the obvious, people expect certain things to happen at certain ages. What happens around the age of 80? People grow old, frail and die. No one wants that, at least not the frail part.

However, this clearly does not address the question. We did not ask "how long do you want to live after you have grown old and frail?", we asked "how long do you want to live?"

Although there was no a priori reason to mistake the question we asked for the more specific and pessimistic one, it is somewhat - although not completely - understandable to mistake the two. It is still a bit baffling to me that people would answer in the most pessimistic way, because humans do have the ability to answer abstract questions using their positive imagination. When asked "do you want to have the ability to fly?" or "would you like to go to Mars?" people will rarely answer "a jetpack is too cumbersome and expensive, so no" or "the flight to Mars takes too long and the isolation would be horrendous for my mental health". Neither of these answers is wrong per se, but they all make the most pessimistic assumptions possible. If we asked these questions, the majority of answers would never cluster around the most pessimistic interpretation of such a question.

On the other hand, it makes sense to assume that people grow frail at the age of 80. This is our lived experience. However, as you may remember the goal of biogerontology is to extend the healthy human lifespan. So as biogerontologists, to determine public support for our research, we want to know how long people desire to live if they are healthy. If I wanted to portray aging research in a bad light, I would indeed conduct surveys just asking "how long do you want to live?"

Given our cultural baggage, currently, there is no safe way whatsoever of asking the how-long question. These how-long questions are fundamentally unable to uncouple the desire to live from assumptions about the future that are largely extrapolations of the present and status quo. Anyone conducting such a survey is either misinformed or intentionally trying to sabotage longevity research.

It is biologically impossible to extend human lifespan without extending healthspan. So if you answer 200 that means you want to live some 180 healthy years. If you so choose, you can off yourself at 180 giving you double the healthy years and zero suffering. Who would choose fewer options in favor of more options if they had that chance?

Better albeit imperfect ways of asking the how-long question:

  • "Do you want to be healthy for as long as possible?"
  • "Assuming you could maintain your current health and looks, how long would you want to live?"
  • "Given the option of taking a pill that allows you to stay young and live for as long as you desire, without any side-effects, would you take such a pill?"
The ultimate way of asking this question might be different:
  • which one of these two views do you find more reasonable? Which one do you agree with more?
  • A. I want my life to have a clear ending at a certain predetermined age. I do not wish to have the option of changing this age.
  • B. I want to live my life day-to-day as long as I am happy. I can choose to live for as long or as short as I want.

I want to live to 30 or 40
Obviously, saying "I want to kill myself" is not really an option in surveys or at a dinner party. Neither is "my life is going really badly and I do not want the pain to continue". However, it is pretty clear that a large number of people do not want to live longer because they are in pain or suffering from depression. Again, though, this fails to answer the question since their response can be rephrased as "I can only endure this pain for another X years and not more" and this is clearly not an answer to the how-long question. Alternative questions would be:

  • "assuming we could cure depression and keep you reasonably happy and you could maintain your current health and looks, how long would you want to live?"
  • "how long would you want to live if you were guaranteed to be as healthy and as happy as you have been during the best time of your life? (i.e. think about the best year in your life)"

What have we learned?

I am not saying there are no people who genuinely want to die at 30, 40 or 80. That would be arrogant. However, I do want to show that these people are rare and not actually identified by asking "how long do you want to live?". In fact, most people mistake this question (see ref 2). Which brings me to the big problem: why do we keep asking this dumb question in surveys (3) and claiming that it has any meaning?

As easy as it is to fool the public with statistics, it is equally easy to fool yourself, if you ask questions with a predetermined narrative. For a long time, biogerontologists have believed that people do not desire a long life, since they have been asking foolish how-long questions designed to give the conservative answers they masochistically crave. This narrative has to end. I think it is time for biogerontologists to stand up with pride and say: everyone wants to live longer and, regardless of that, our science is the only hope for humanity to avert a demographic catastrophe and the only hope to drastically increase healthspan. We are the science of the future.

References and notes

(1) We can argue whether the answers are fundamentally illogical because they evade the question or whether they just answer a special, extremely pessimistic version of the question. Either way the answers will be biased and useless because of the way the question is phrased. There is no right way of asking the question, but there are ways that are fundamentally more flawed than others.

(2) Above we have developed an intuition for why and how people mistake the how-long question. On top of that, we also have hard data showing this is exactly what happens. The answers get much more optimistic if you stipulate good healthspan which is a biologically reasonable assumption (also discussed here).

Asked “If doctors developed a pill that enabled you to live forever at your current age, would you take it?” a surprising number of people turned out to be hardcore life extensionists: "There were no differences by age...Among young adults, 40.0% indicated they would not take the pill, 34.2% indicated they would take the pill, and 25.8% indicated they were unsure."

The older people were community dwelling (n~310) and the younger participants (n= 593) were psychology students. Surprisingly no cohort effect was seen, as I would have expected the younger participants to be more positive towards life extension. However, it does look like there is a trend and the study may have been underpowered to detect differences (only 23-32% of older adults would take the pill).

At first glance, everything makes sense to me. In this 2011 article, the numbers, on a similar but different question, were 21% for the young and 12% for the older people. We see a suggestive cohort effect (older generations being less positive about lifespan extension)vand a normalization of lifespan extension research because the numbers are much higher in the more recent survey.

Barnett, Michael D., and Jessica H. Helphrey. "Who wants to live forever? Age cohort differences in attitudes toward life extension." Journal of Aging Studies 57 (2021): 100931.

(3) This survey at least tried having some sort of "DEPENDS ON LIFE QUALITY" question, but I speculate that even outright stating "how long would you want to live in good health" is not enough because people will naturally make assumptions about the unknowable "I will get bored and will want to die" or they will be unable to "suspend their disbelief" so to say. They will actually refuse to take the premise for granted, because it is so at odds with their experience. We need better ways of asking these questions.

Of course, the only rational answer if you fear that you will get bored, is "I want to live forever with the option of ending my life [when I get bored]" instead of "I want to live X years [making an a priori assumption about the time you will get bored, i.e. the unknowable future.]". While humans are not rational, I do feel with the right question design we could reveal that a relative or absolute majority of people is, indeed, able to think rationally about aging. (That is not to say that being purely rational is the only way to go about life, although I do think it is usually the best way.)