Revealed preferences for longevity

Smoking takes 10 years off your life - but is this a sufficient reason to give up smoking? Why is a long life a better life?

The United Nations Human Development Index uses life expectancy as a measure of life quality because:

a long life is valuable in itself and... various indirect benefits (such as adequate nutrition and good health) are closely associated with higher life expectancy 

That longevity is a proxy for the things that make life enjoyable is a reasonable argument, especially in the context that the UNDP is making it. The elimination of polio, for example, not only increased life expectancy, but also improved life quality, by preventing people from experiencing years of disability. But to the extent that smoking shortens life by taking low quality years off at the end, is that such a bad thing?

And the statement that "a long life is valuable in itself" takes us back to the original question: why?

Moral philosopher John Broome's book Weighing Lives takes on some hard questions about longevity. Increasing one person's life has costs, either because it takes up resources that could have been used to improve another person's life, or because it decreases that person's life quality. Trade-offs have to be made. Broome provides a way of thinking about those choices that an economist can understand:

 

From "Weighing Lives" by John Broome

This diagram shows the evolution of well-being over time for a person with a terminal disease. Palliative care provides a higher quality of life for a shorter time; aggressive treatment extends life, but at the cost of life quality. Contemplating the problem, one can imagine that sometimes one might choose palliative care, and sometimes one might choose aggressive treatment, depending upon the precise well-being - longevity trade-off. 

It's a bit like economics. Once the problem is expressed in a diagrammatic form, the answer is fairly straightforward - it's drawing the diagram in the first place that's the challenge. Also, as in economics, the predictions of the model depend critically upon the underlying assumptions - in this case, the relative value given to longevity versus life quality. I didn't read Broome's book carefully, but I suspect he's too smart to even try to address that question.

What I like about Broome's framework is that it shows that trade-offs must be made, and it provides a way for planning, rational people to think about choices. Yet when life and death decisions have to be made, planning goes out the window. The primitive part of our brain, which has been shaped by millennia of fighting for survival, takes over, and votes to hang on at all costs. Death, yes - but not yet.

So how can we think about longevity? How can we abstract from the immediacies of life and death, and discover what our planning selves would want?

I've been thinking about this question lately, because my dog has just reached the grand old age of 14. As his puppyhood friends gradually disappear, and he himself increasingly suffers from bladder stones, cataracts, arthritis, deafness, flatulence, toothlessness, and (I suspect, though it's hard to tell) doggy dementia, I find myself wondering: is longevity all it's cracked up to be? 

When a person selects a pet, she reveals her preferences for life expectancy. Using Broome's framework, one can compare the life quality of different breeds. In this case, I've compared a standard poodle, which has an average life expectancy, and a miniature poodle, which has an exceptionally long life expectancy. I've drawn a diagram showing two dogs with identical patterns of life-quality over time, but with one life compressed, and the other extended.

 

Well-being of two hypothetical poodles over time

 

I don't think it's obvious that, in the diagram above, the miniature poodle's life is any objective sense better. Both dogs experience identical average life quality, it's just that one, by virtue of its genetic make-up, experiences that life quality over a longer period of time.

Indeed, when a person selects a pet, life expectancy is one of the last things considered (see, for example, this pet selection guide, or this one or this one). Instead, "experts" recommend choosing a pet who will be a good match for his or her owner in terms of activity level, sociability, and so on. Good health matters - sensible owners avoid breeds prone to health problems. But not life expectancy per se.

Actions speak louder than words, and reveal preferences more clearly. People who choose shorter-lived pets reveal that longevity per se does not matter much. 

With my dog, I try to let him live his life to his full potential - but the fact that his maximum possible life span is 17 or 20 years, and not 10 years or 40 years, is of little consequence.

In the same way, a good human life is one that is lived to its full potential.

But, as a practical matter, what does that mean?

Addendum: some people have argued in the comments that smoking leads to a reduction in average life quality, as well as life length. On average Canadian smokers are less satisfied with life than non-smokers - around 40 percent of non-smokers report being very satisfied with life, as compared to around 30 percent of smokers. There are two big caveats to this graph however. First, it excludes institutional residents, including people in long-term care facilities - that is, the people who are experiencing low-quality late-life years. Second, it's not clear that smoking causes lower life satisfaction - people may smoke because they're unhappy, rather than the other way around.

 

Comments

[Lord]

The expected number of heart beats for all animals is about equal, so measured in such terms they have similar life spans, it is just some live faster and some slower. Time spent waiting must be worse for the former though.

[Squarely Rooted]

This post reminded me of this fascinating segment from Radiolab:

http://www.radiolab.org/blogs/radiolab-blog/2013/jan/15/bitter-end/

in which doctors, who have substantially greater-than-average information about the effects of treatments that extend life at the expense of quality-of-life, overwhelmingly report that they would not want almost any such procedure performed on them should the circumstance arise.

[Frances Woolley]

Squarely rooted - and in my very limited experience, if/when the time comes to make the decision, actually do choose life quality over quantity.

Lord - that's way too cool to actually be true....

[RPLong]

"But smoking shortens life by taking low quality years off at the end - is that such a bad thing?"

This is far from being a foregone conclusion. The research I've seen suggests that people seem to achieve greater levels of happiness as they age. This suggests that the final years are the highest quality, not the lowest quality.

And even if that isn't true for everyone, it's true for some. You never know which year is going to be the best year of your life. Maybe the first year of life is the best, but you never remember it once you get older. Maybe the best year occurs at age 93, and if you die before that age, you miss out.

Discussing longevity as though we know in advance which of our years are most preferable seems flawed to me.

[david]

http://en.wikipedia.org/wiki/Kleiber%27s_law

from http://skeptics.stackexchange.com/questions/5701/does-every-species-get-around-a-billion-heartbeats-on-average

So it seems that insofar as there is a pattern, humans are a bit of a odd data-point anyway.

[Geoff R]

Interesting piece!

I like the questions about longevity vs quality - but I object to characterizing the effect of smoking as "taking low quality years off at the end"
It's probably worth looking at the quality of life of smokers towards their end, vs non-smokers.
If quality of life drops because of disease and deterioration due to age, and smoking causes disease and certain types of deterioration, then smoking is not trimming off 'bad years', but compressing end years - like in the poodle example.

But why deal strictly with your "average life quality"?
If I will never again be as happy as I was at the age of six, that's not a reason to stop living at seven.

I think the discussion needs to concern itself with years of life that have a may have a negative rate of change in total quality of life.

[Matthew]

There is almost no disease or ailment known to medical science that isn't aggravated significantly by smoking.

[Bob Smith]

The heartbeat thing kinda makes sense. I mean, let's face it, mammals all have more or less the same basic technology (I see the other oddball in that chart are chickens, so not surprisng they're an outlier). Not surprising that we fail at more or less the same rate. If human's are the oddballs, it's because we're the only ones who bring ourselves to the mechanics on annual basis and replace parts that fail.

[Frances Woolley]

RPLong: "The research I've seen suggests that people seem to achieve greater levels of happiness as they age."

The studies that I've seen that obtain this result tend to be *all else being equal* studies of the effect of age on happiness. That is, controlling for marital status, income, employment, presence of disability, etc, age has a positive effect on happiness. The bottom line isn't "you get happier as you get older" it's "getting old does't suck quite so much as you think it will."

Geoff R: "I think the discussion needs to concern itself with years of life that have a may have a negative rate of change in total quality of life."

I'm not sure precisely what you mean. Is it that life is worth living as long as well-being is positive, whatever positive might mean in this case?

Matthew - "There is almost no disease or ailment known to medical science that isn't aggravated significantly by smoking."

Actually, I was reading in some behavioural economics book the other day - perhaps it was Predictably Irrational??? - that people tend to overestimate the negative health consequences of smoking. Will try to find the reference.

david - neat.

[Jacob Hoover]

"But smoking shortens life by taking low quality years off at the end - is that such a bad thing?"

Yes, unfortunately this is not a simple compression phenomenon. Smokers have a significantly higher percentage of lower well being levels. As a hospitalist physician, my clinical experience supports the CDC claim that "for every person who dies from a smoking-related disease, 20 more people suffer with at least one serious illness from smoking".

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5235a4.htm

[Patrick]

When it comes to a beloved pet (or human), despite our best intentions, when it comes down to the wire, we're just not as rational as we'd like to believe we are. I've learned that the hard way after inflicting unnecessary suffering by being too sentimental to euthanize sick and suffering dogs. It's *so easy* to convince yourself that the e.g new pain meds are really helping, that they're looking better today, that the malignant bone tumour protruding from her jaw really isn't as f*cking horrible as it looks. It's an easy trap to fall into, and believe me the pain of pulling the proverbial plug is bad enough, don't make my mistake and compounded it with being too sentimental, selfish, and weak to do the right thing at the right time.

[Ralph Musgrave]

Smokers pay taxes on tobacco while they’re alive, plus they die early, thus SAVING taxpayers’ money at the end of their life. Smoking is thus a brilliant way of getting deficits and public debts down.

[Frances Woolley]

Ralph, an old analysis of this is here: http://qed.econ.queensu.ca/pub/cpp/Sept1992/Raynauld.pdf - I don't think the basic facts have changed much since then however. Ian irvine at Concordia University has done a lot of good work on tobacco taxes and smoking policy.

[Frances Woolley]

Patrick - the problem is that they disguise their pain so well. E.g. my dog's limp instantly disappears when he's taken to the vet - I think he knows that any display of weakness and he'll end up in the comfort room. But he's really not in particularly bad shape - he's still likes chasing after puppies in the park.

[genauer]

http://www.statcan.gc.ca/pub/82-003-x/2013002/article/11769-eng.pdf

just some first, preliminary comments:
1. HUI3 does not include the pleasure, people might feel from smoking
2. the well known cross correlation of smoking, income, health is mentioned, but not shown
3. it looks the curve is just shifted a few years to the left, means smokers die a little earlier, but the path of deteroration is very similar

That would mean also, they consume less health care costs and pensions

[GSo]

The myth about people clinging to life is contradicted by the hundreds of thousands that have jumped out of their fox holes disregarding the machine gun spraying bullets over their heads.

How would the human longivity look absent antibiotics? My guess is that we have made ourselves outliers, we are not born outliers.

[genauer]

GSO,

I doubt very much, that antibiotics have a longevity impact of more than 0.5 years,
probably less than 0.2.

And the very most of you did never serve in the military, and would even much less be inclined to be stormtroopers or special ops.

Is there a single person in this blog, who had to contemplate, what 30 % death rate per day you would take before giving up? At what point would you refuse to carry orders to nuke your own family / hometown?

I worked with people, and discussed with them, who served in September 1989 in the GDR, less than a quarter century ago, and they had to ponder, at what point in time, in what chain of events, they will turn their loaded gun around, not targeting demonstrators, turning it half way, away from those alleged western imperialist "drones", or turning it full way, and pointing at the officer. And what happens then ?

"The Second Civil War" 1997, with James Coburn, gives you a glimpse of that, in a more familiar setting for you.

This has absolutely nothing to do with decision at old age,

but the very most of us will clinge to extend to the end of your years.

And despite all the tough talk, me, then, probably too.

[Frances Woolley]

genauer - "I doubt very much, that antibiotics have a longevity impact of more than 0.5 years,"

No, the effect of antibiotics is big. This news report says 8 years http://www.abc.net.au/science/slab/antibiotics/history.htm , but then there are all the other procedures like modern surgery that would be harder to do were antibiotics not available to treat post-surgical infections.

[Jacob Hoover]

Professor Woolley: It would be interesting to see a more recent study on the societal costs of smoking. Certainly both associated health care costs and cigarette taxes have increased markedly above inflation for the last 20+ years.

Genauer: The "Dynamics of smoking cessation and health-related quality of life among Canadians" article uses "HUI3" functional status categories. Most of these functional metrics are known to not have a large correlation with smoking including vision, hearing, speech, dexterity, emotion, cognition, and pain. Perhaps the only significant item is ambulation, which would be secondarily affected by the curiously absent item from the list, breathing. It is uncommon for non-smokers to encounter breathing issues (shortness of breath) towards the last few years of their lives, but very common for smokers to do so, and this is perhaps the single greatest item that affects a smoker's morbidity.

[Shangwen]

For this type of discussion it helps to have a broad picture of the probable range of late-life outcomes. We typically imagine a scenario represented by Broome's aggressive intervention outcome--basically, a choice most people would regret. Certainly there are outcomes were there would be near-unanimity on the condition's severe disutility--locked-in syndrome, Huntington's disease, Kreuzfeld-Jakob, and so on. But many if not most situations are not like that. A lot of chronic or degenerative diseases begin gradually, progress unevenly, and don't represent uninterrupted suffering. These are the conditions most of us are likely to get. The Winnipeg woman who recently committed suicide in Switzerland had a very rare condition, yet her situation is the one we typically imagine when we think about the problem of prolonged life.

I'm glad you raised this because, like abortion, it's one of those topics that people talk about but few take seriously philosophically. We underestimate our ability to to adjust expectations and find incidental enjoyments in life even as our situation changes. And do we really talk about things like less aggressive treatment or assisted suicide because we think quality of life is that important? Or is it just unbearable to imagine diminishment from our current viewpoint? If that was such a universally held value, we might see a lot more abortions after genetic testing, and we'd take up JS Mill's argument that capital punishment should be maintained to spare people the horror of long imprisonment.

[Frances Woolley]

Shangwen: "We underestimate our ability to to adjust expectations and find incidental enjoyments in life even as our situation changes."

This is really important - but as we've discussed before, the big issue is dementia. There it's not clear that the ability to adjust expectations argument holds. Though it's not clear it doesn't, either. Peter Singer, another interesting moral philosopher who's a strong advocate of assisted dying, didn't help his mother along when she had Alzheimer's, on the grounds that her quality of life was actually o.k.

Jacob - pensions are a big issue too.

[J.V. Dubois]

Very interesting blog Frances. And it seems that I may be an odd dog owner. I actually looked at the longevity of the breed. For instance I would not pick Great Dane as a dog breed since they live only 7-10 years. Other than emotional distress that dying pet causes it is also about all the work you put into the dog, training him etc. Plus I always considered having an intelligent pet companion with large lifespan as an advantage (even if I do not have one). Some parrots like cockatoo may live up to 80 years.

David: I do not see humans as that far off - if we are talking about life expectancy of humans living in the wilderness where life expectancy of humans would be around 30 year. Plus just quickly eybealling the chart there seem to be some other odd data point. Like dogs and cats living around 30 years which is more like 15/10 for domesticated ones and 10/5 years for feral ones. Or horse living 55 years which is closer to 25 years for domesticated ones and 15-20 for wild ones.

[genauer]

Frances,

the news report says
"Between 1944 and 1972 human life expectancy jumped by eight years - an increase largely credited to the introduction of antibiotics"

It did certainly not "jump", that was a very continous process, and this yellow press like piece does not give the slightest hint to any scientific proof of any causation by antibiotics, just pure speculation.

Jacob,

the topic is about "revealed-preferences-for-longevity".

If you come with other Quality of life measures and how they develop under various conditions and decisions (e.g. smoking or not), then please show them.

But to pick on one special, relatively small item towards the whole picture, and repeatedly not providing any evidence, how that develops relatively over time, like the link I gave, does simply not support your case.

The figure 1 ("Weighing Lives" by John Broome) above pretty clearly spells out the relevant question, and you do not address it with your postings.

[edeast]

Can someone explain, cumulative prospect theory. Pictures.

I think it would fit in with changing relative expectations, end of life, utility.
I'm just lazy.

[Frances Woolley]

genauer - try this reference then: http://cdi.mecon.gov.ar/biblio/doc/nber/w12269.pdf - it describes the changes associated with the large increase in life expectancy in the 1940s, 1950s, and 1960s. One could argue that DDT and vaccination programs also helped, but just take a look at the major killers in 1900. Just think about tuberculosis, for example. There's no doubt that antibiotics had a big impact on life expectancy. That they are losing their effectiveness is very very scary.

Re your comment to genauer - I think the bigger issue with the CCHS data is the exclusion of people who are experiencing really crummy quality of life, e.g. have dementia so can't answer the survey.

[Jacob Hoover]

genauer: You apparently did not read the cdc linked article I first posted, as all of your points had already been contradicted. It's unfortunate you believe I picked "one special, relatively small item towards the whole picture, and repeatedly not providing any evidence" when the opposite has already been shown.

Emphysema and chronic bronchitis take a heavy toll on quality of life. Most patients will tell you that shortness of breath is worse than pain. Since you did not read the very brief article, but automatically assumed your ideas were correct, let me summarize a couple key points. 49% of current smokers have chronic bronchitis and 24% have emphysema. As the report notes "chronic bronchitis and emphysema account for 59% of all smoking-attributable diseases".

Seems the only cherry picking was the post you listed which neglected the most important morbidity issues concerning smokers. So I will say again, the quality of life of smokers does not look like the compression shown by Professor Woolley with the "well-being of two hypothetical poodles over time". Whereas in her example both dogs experience identical average life quality, the smokers experience LOWER average life quality.

[Geoff R]

Hi Francis, thanks for responding to me above.

Geoff R: "I think the discussion needs to concern itself with years of life that have a may have a negative rate of change in total quality of life."

I'm not sure precisely what you mean. Is it that life is worth living as long as well-being is positive, whatever positive might mean in this case?

I just think that targeting high 'Average Quality of Life' is foolish for an individual - and that it makes more sense to target high 'total quality of life'.
As long as I'm generally more satisfied than suffering, why would I pull the plug?

Say I have been granted a free pass to ride roller coasters all day, an activity that brings me joy (higher QoL), but also makes me progressively more nauseous (lowers QoL). Assume for a moment, no other factors affect my Quality of Life. There's a point at which I'll want to stop riding, because the discomfort of nausea will outweigh the joy of riding.

If my level of nausea rises slowly I will presumably ride for some time, even though the rising discomfort of nausea will ensure that my highest average 'quality of life' is achieved after the first ride.

If I wanted to maximize average QoL, I'd ride once and leave the park. Doing so would mean missing out on a lot of joy.

[Shangwen]

I wholeheartedly agree with Jacob: there's no question that habitual smoking is globally detrimental to almost any person's health. Intuitively, since it is a habit to be sustained, rather than a hedonic experience, it's hard to see how it could add much to life.

Frances: quality of life and mental impairment are a very thorny issue. Were we all doomed to dementia, we'd all want to be that mythical happy old granny who sits in her favorite chair all day and chatters to her dead mother about the garden, and has no serious physical ailments. But of course the reality is otherwise. It's also very difficult to measure quality of life in the demented. Most notably, it's very hard to assess pain in the demented, which none of us would want undertreated. There's also the sad fact that many mental disorders that burn out in middle age (like PTSD) can be revived by dementia as the executive cognitive functions are destroyed. You see this a lot in veterans' hospitals, and it's extremely difficult to manage.

And speaking of longevity, I think our debates on the issues often fail to take into consideration the simple fact of just how long one can live in a very impaired state. Highly salient diseases like Lou Gehrig's (which can kill you in about 2 years) are merciful compared to Alzheimer's. If you asked people how much time they'd like to spend in a state of continuous confusion and total vulnerability, punctuated by episodes of uncontrolled emotion, with accelerating physical decline, it wouldn't be long. The idea that dementia is just about being forgetful is a serious underrepresentation of its impact.

[Frances Woolley]

Geoff R - the discussion gets much more complicated as soon as one puts costs into the equation, and expands the discussion to include multiple individuals.

In a world with finite resources, only so many lives can be sustained. At some point one has to weigh lives; to make trade-offs. Maximizing the total quality of life over the entire population may mean not maximizing the total quality of of life of any one individual.

Shangwen - I was thinking of taking up smoking at one point, and got talked out of it by a smoker who made precisely that observation.

Have you seen some of this work that's being done with music therapy for dementia patients? This is a wonderful video: http://www.youtube.com/watch?v=fyZQf0p73QM .

[genauer]

Jacob,

I did read your little CDC thing, and adressed it, as far as it is relevant here.
Your personal fixation with smoking is just a very small part of the overall

Quality of life question, people ponder.

One more example for you:

"Like anybody, I would like to live a long life. Longevity has its place. But I'm not concerned about that now"

http://www.youtube.com/watch?v=wzG3VMTGRVA

for your "as all of your points had already been contradicted" I do ask you, politely, which single one, where : - )

[Matthew]

"Smoking is thus a brilliant way of getting deficits and public debts down."

Does anyone remember that episode of the Yes Prime Minister? Smokers courageously lay down their lives so that others can have access to quality medical care!

[Jacob Hoover]

genaur: Are you for real? [Unnecessary personal attack deleted. Watch your tone. SG] Did you also not read Professor Woolley's blog entry before these comments? If you did, you would know that the first of two main examples in this post was about smoking. Try the first sentence, first few paragraphs, at least one chart..."Smoking takes 10 years off your life - but is this a sufficient reason to give up smoking?" That was the very first sentence.

You don't need to ask politely, you just need to read. Again, it's all there for you. [Another personal attack deleted. Last warning. SG]

others: Another interesting thought about quality of life to ponder: were people any less happier 200 years ago. Will we be any happier 200 years from now if we live to be 150 years on average and spend the same percentage of years with chronic illnesses? How about the remaining primitive civilizations discovered over the last few decades with life spans similar to ours of many generations past. Are we any happier? Some of these questions have already been studied, but happiness is tough to precisely quantify and the answers aren't unambiguously clear. Fascinating topic, though!

[genauer]

Jacob,

please stop the insults, and just back up your claim "as all of your points had already been contradicted"

[genauer]

Jacob,

to provide some structure for you, please try to be specific:

what specific point did I make where, and how did you specifically contradict this where.

[Frances Woolley]

genauer, Jacob:

Arguing about the precise shape of the shape of the well-being curve for smokers v. non-smokers isn't all that interesting. Essentially, it's an empirical question, and it's not one that's easy to answer, because the decision to smoke is influenced by a whole bunch of other things (e.g. stress, education) that also influence happiness - Eric Crampton has written a lot about this in the context of evaluating the impact of drinking on well-being.

The more interesting question is how do we think philosophically about making these trade-offs? Even if we can think about them in some kind of abstract philosophical way, what problems do we encounter when we try to implement our plans?

Please try to keep your comments focussed on the underlying issues of moral philosophy and behavioural economics.

[The Keystone Garter]

A GAI would be efficient for retraining and education; basically a student loan for anyone who wants to learn some specified fields, that no one really cares if you payback. A lot of the stuff will be building and datalogging (uploading info to databases) biosensors and mind sensors. I can define what should be learned and built in a way that gets us to a stable plateau of technology; again, a lot of it is not building WMDs, enabling tyrants, and diffusing a predictably happy and good reasoning faculties high Q-of-L. We are going to need people and health care systems to figure out how to datamine if we get good human body/mind biosensors. Maybe a consultancy career here...

[Jacob Hoover]

My apologies to the referee and others, I will try to not mix ad hominem, playful or otherwise, with discourse and try to stay away from the former altogether.

Genauer: I will again summarize and concede all points to you:

1) Even though the CDC shows "Quality of life measures and how they develop under various conditions and decisions (e.g. smoking or not)", you are right.

2) Emphysema and chronic bronchitis, even though they represent the lion's share of morbidity to smokers is just a "pick on one special, relatively small item towards the whole picture, and repeatedly not providing any evidence, how that develops relatively over time, like the link [you] gave, does simply not support [my] case."

3) Because "Figure 1 ("Weighing Lives" by John Broome) above pretty clearly spells out the relevant question, and [I] do not address it with [my] postings", therefore other questions listed by Professor Woolley and discussed by me and others are irrelevant.

4) Even though the opening and half of the blog post has to do with smoking, I have a "personal fixation with smoking" and although several people here are pondering it, it "is just a very small part of the overall Quality of life question, people ponder".

[Jacob Hoover]

Professor Woolley: I had hoped not to suggest a precise shape of the well-being curve, only to note that it is generally lower than non-smokers.

[Shangwen]

Matthew: Yes, (Prime) Minister--there's no better inside story about public policy!

Frances: Yes, seen that video several times, wonderful stuff.

At work the other day we were having a similar quasi-philosophical discussion, looking at all the awful conditions with which people are more likely to commit suicide, and those where, surprisingly, suicide rates don't seem that high. Notably, schizophrenia, depression, and especially trigeminal neuralgia have relatively high suicide rates; but there are conditions where most of us assume lots of people would want to kill themselves yet don't: Huntington's, dementias, amputations. (This was a freewheeling, only half data-based conversation.) If that's accurate, there is much more suffering in a life where episodes of normalcy are punctuated by periods of severe impairment, than in a life of gradual decline or major loss. The adjustment factor again.

[Geoff R]

Thanks Professor,
This topic continues to interesting me, and you've certainly caused me to ponder.

I agree that the situation needs to be assessed differently at the societal than the individual level, but even in the strictly hedonistic view of the world (blech), I still think the target should remain a high average of the 'total' (lifelong) quality of life of the individuals of that population. Policy built on targeting just high average quality of life in the poodle respect would actually incentivize terminating people at peak life quality.
[Perhaps by promoting the practice of gorging ourselves on sweets (assuming that is more fun than eating healthy), aiming for fatal diabetes, and tapping out at 40?]

There is some cost-benefit analysis that needs to be done in deciding when to spend societal resources to extend the lifespan of an individual - I'm not proposing we ignore that, just that we aim for high totals in that analysis.

Ad absurdum:
"It's not worth operating on this baby, because she's really happy right now, and we'll probably never get a higher average QoL than right now"

(Also, I apologize for the misspelling of 'Frances' above, I know how frustrating mispelled names can be

[Name99]

@J V DuBois
" I do not see humans as that far off - if we are talking about life expectancy of humans living in the wilderness where life expectancy of humans would be around 30 year. "

When you talk of "humans in the wilderness" you are referring to life expectancy at birth. That is not the relevant metric here. The metric that matters is something more like "maximum realistic lifespan" which you could operationally define as something like the lifespan of the 95th percentile of the cohort that make it to age 21. (The assumption is we can't make assumptions about the absolute longest lived, who may be unusual genetically, but we can assume that if most people died from accidents, infectious disease, starvation, that sort of thing, then by eliminating those we can move most people to that 95th percentile --- and indeed we have, in developed societies.)

By this sort of metric, humans have always had their sixty and seventy year old members of society. They were rare, yes, but they weren't freakish myths like say Methuselah.