Children Born to Older Fathers and Grandfathers May Live Longer | Hacker News Reader

Children Born to Older Fathers and Grandfathers May Live Longer

extremelongevity.net

52 pointsSlipperySlope16 comments

(edit:) As quite rightly pointed out by stiff, this summary article is actually not lying to us, as its final sentence does explicitly and correctly pointed out that this paper's goal was actually not to attempt to establish any kind of causation here. I will not delete or edit the rest of my comment here, but I would recommend disregarding it.

People who live longer are more likely than people who die young to have children while they are old. People who are children of people who genetically live long are more likely to themselves genetically live long. Thereby it would be surprising--downright shocking (in the scientifically-exciting "time to reevaluate what kinds of traits can be passed down by genes" sense)--if we did not find the statistically significant correlation that people born to older people tend to be genetically able to live longer than people born to young people.

FTR, The paper even admits the possibility of this selection bias, but then seems to just disregard it with a "regardless" as, despite this headline and summary, the paper seemed more excited by the increased evidence that TL was a good marker of longevity. I hate the way people describe science. :(

It's all well and good to question statistics, but:

> People who live longer are more likely than people who die young to have children while they are old.

That's a statement with pretty poor support. Certainly the median childbearing age is well within the lifespan of even short-lived people. The relative population of fathers who have children at 40 is not meaningfully increased by the number of people who died before 40 of correlating/age-related illness (staying with the hypothesis from the article: people basically don't die before 40 because of the length of there telomeres). Frankly my intuition says that the effect you posit is going to be near-zero. Do you have any numbers?

If you're going to criticize, do it right.

Actually, no: I will contend I am "doing it right". I am not arguing that my explanation is true (if I were, I'd be just as bad as these people I'm ranting about); I only will claim that my explanation is obvious and reasonable and thereby has to be considered and explicitly countered or controlled for before you can just assume that it is due to something else.

I totally accept that it /might/ be true that if you are older when you have children, the children will live longer, but this paper does not demonstrate that. That said, I don't even think the researchers of this paper were /really/ attempting to demonstrate that, and so I will put the blame squarly on this "Extreme Longevity" site for not paying enough attention.

(edit:) Actually, while I maintain that ajross' commentary was unhelpful (as I stand by my first paragraph in this comment), as pointed out by stiff this article does conclude with an explicit note discounting the notion that this article is establishing causation.

Except that, as I point out, your explanation is neither obvious nor reasonable (frankly, I'd say it's flat wrong). Yet you introduced it without evidence anyway as a way of saying "What about this?". That's the same technique creationists use to "challenge" evolution. You can't make that argument, logically, with a flawed counter hypothesis.

The only reason I posted is that you got uppity in your final sentence, complaining about journalists lack of statistical sense. Frankly, yours is no better.

As I explicitly stated, the authors of the paper actually agreed that this was a sufficiently interesting selection bias to note (although not to bother refuting, as far as I can see; they just say that that if that were true then sperm TL would be an interesting marker, and then move on with a "regardless").

> It is possible that older men who volunteer to donate sperm have better reproductive health and longer sperm TL and that a similar selection bias results in men with longer sperm TL tending to reproduce at later ages.

For more context on my general crusade against poor science reporting, I encourage you to read my comment history. (In which case, I will claim your snark about believing in evolution is ironic. I contend and argue that it is publishing summary headlines like this that makes people not believe in other science, such as evolution.)

http://news.ycombinator.com/item?id=4082886

It's just the headline that is stupid (altough note it says may), the article itself states in the last paragraph:

This study was unique in showing that telomere length determination by paternal age was generationally additive. It was not designed to determine if older paternal age translates into longer lifespan.

The bibliography in the actual paper cites lots of articles concerning the impact of telomere length on various health-related issues, so it seems that this really was not the point of this study.

I actually had just noticed this, and was about to add an "edit: actually, the journalists did note this". Thanks for pointing it out, however. It is then quite silly of me to have spent time reading the paper but not adequately reading the journalistic summary ;P. (I will even then further defend Extreme Longevity by saying that "May" is maybe pushing it, but not unreasonable.)

In the Heinlein (SF, eg. "Time enough for love", and related books) universe, long life is achieved by paying children of long lived parents too marry each other.

Ultimately the program is so successful that they are persecuted by the rest of the earth of being so long lived that they have to flee the earth!

Or in Larry Niven - Ringworld , breeding people for luckiness

I think that statement only makes sense if people who live longer also have higher fertility rates.

You might look on this in the context of work on selective breeding of flies at older ages by Michael Rose:

http://www.fightaging.org/archives/2011/04/the-work-of-micha...

Variability of longevity in response to changing circumstances persists across generations for calorie restriction and age of childbirth in a range of laboratory species, so we should not be completely surprised to see some signs of that in human populations. The trick has always been to be able to peel that away from the confounding factors. (And you can argue over whether this paper manages to do that sufficiently well).

From an evolutionary perspective it makes perfect sense for organisms to adjust for longevity or a fast burn in response to environmental changes. Species that are able to do this should be more successful, and this is an old, old set of machinery - as shown by the fact that is shared with flies, yeast, and so forth.

This story was submitted two days ago: http://news.ycombinator.com/item?id=4099940

I'd like to see an explanation of how this matters in plain English. I'm not sure what 0-1 in telomere length is, in relation to years, but the difference the chart reveals is less than 1/10th of a percent. The bias in the chart alone makes it seem huge.

Then I'd like to know if the telomere "consumption" rate is constant (I doubt it). Because unless its constant, this just leads to our potential maximum lifespan—not the expressed lifespan (due to stress, diet, etc). Like genotype/phenotype differences.

I once read somewhere that the after a certain age, like 50 or something, the less quality you sperm is and hence your kids. So would this be the trade off?

"people in the Philippines"

So no chance that older father = more education = better job = more money = better healthcare, compared to teenage parents?

I just quickly skimmed the published paper, but it looks like they controlled for standard demographic variables.

Reply on news.ycombinator.com