The effect of physical fitness on mortality is overestimated
uu.se
uu.se
Blank slatism is the curse of the sciences. Of course the kind of people who die in car accidents and homicides are not identical to everyone else but for random luck. They're less educated[1], they make poor decisions, they have dangerous neighbors, frequently they're immigrants from countries with lots of car accidents and homicides. We've known that "everything is correlated" since 2014[2], when will science figure it out?
1. https://www.theverge.com/2015/10/2/9438197/traffic-accident-... 2. https://gwern.net/everything
It's 100% unsurprising that this is true of fitness too. This is what always happens. You look at something like a corporate health fitness plan and you find some correlate even after you 'control for' SES, prior health record etc etc; wonderful! Then you do a randomized experiment and it turns out that the residual confounding was still larger than any causal effect which might be there: https://gwern.net/doc/statistics/causality/2022-wallace.pdf Ah well. Maybe next time you'll manage to 'control for' the confounders...
> This suggests that people with high and low fitness levels may differ in other important ways, which is something that previous studies have not fully taken into account
Which further undermines the authors’ assumption that drownings are random accidents.
Fat is buoyant.
That's probably confounded. Anything over a BMI of 23 better be an increasing proportion of muscle, and even then there's a point that the stress on the heart isn't worth it.
Almost every physical and mental heath condition does bad things to nutrition and internal energy stores, even if only at a diet level.
It's hard to see since so many people are overweight or obese to start with, but the overall correlation goes that way enough to cause confounding.
It might be co-causal, for example: to take your list, maybe "making poor decisions" has a bigger impact on cardiovascular health than we thought.
Kinda like how wine after dinner improves health outcomes. The kind of people who can afford to regularly drink wine, but not a lot, also have more access to healhtcare.
Also riding horses makes you healthier! (it’s a proxy for wealth)
One could argue that high VO2max may just be another proxy for wealth. People who can afford to work out every day can also afford lots of other things that correlate with health and longevity.
Sweden had a low homicide rate as well (between 4x and 9x lower than the USA, depending on year)
(I will say that a morning workout still supercharges your day!)
And you'd be wrong. It does not supercharge my day. Maybe it superchargers your day. Just finished my morning workout and now I feel like I've been drugged and need a nap. I do not feel "supercharged"
I just noticed that in my 53 year existence, only by the past 10 years, a morning workout (which I didn't do prior to that) was figuratively and literally eye-opening, not just physically but mentally.
Specifically, it was the Orangetheory workout.
And I regretted not discovering it sooner, and mainly I blame people who didn't emphasize not just the physical but the mental boost
Nobody wanted to live a long time but be unhealthy in their older years.
It wasn’t until health and fitness podcasters started holding up the “healthspan” term as something novel that we had to hear about the supposed difference.
It’s also unfortunate that the “healthspan” influencers frequently get into health trends that have the opposite effect. The main example I can think of is Dr. Peter Attia, who started out praising ketogenic diets (recently shown to have very negative effects on cardiovascular health, not surprisingly) and later Rapamycin (briefly praised by lifespan influencers until they all decided it was hurting more than helping).
I think one of the best things people can do for long-term health is limit their consumption of health and fitness influencers who are pressured to constantly invent and push new things to keep themselves relevant.
The top metastudies on a simple google search both conclude positive effects on CVD.
The most recent example is the KETO-CTA paper, which was released in April to loud claims that high LDL in a keto diet was not really bad. It got picked up by various news outlets as vindicating ketogenic diets. The authors posted long Twitter threads full of meme animated gifs celebrating their victory over keto skeptics.
Yet when people started looming into the study they noticed that the study failed to report the data from the pre-registered trial endpoint. This is highly unusual because if you establish an entire clinical trial, you are expected to publish the primary results you pre-registered.
After a lot of prodding, people finally got the authors to release the percent change in non-calcified plaque volume (NCPV) values that they pre-registered for and they were extremely high. Even higher than seen in certain serious medical conditions. It became clear that they chose to hide the values because they were so incredibly bad.
> The top metastudies on a simple google search both conclude positive effects on CVD.
There are a lot of die-hard keto defenders who produce papers, blogs, and podcasts claiming it’s either neutral or good for CVD. You have to look past the SEO optimized content and go to actual CVD experts who do not have their internet presence tied to promoting keto diets.
I initially gave keto a shot for two weeks, and I found unexpectedly high benefits in increased energy, weight loss and an end to some long-standing stomach ulcer issues. So my two-week trial has ended up lasting 3 months.
It's been frustrating that trying to learn about keto is like drinking from a firehose of bullshit, because I'm really rather enjoying my extra energy, weight loss and lack of stomach pain and ulcers. I actually feel 20 years younger.
I'm quite skeptical of the long term elimination of whole grains though, so soon I'll be easing off keto with some vegetables and whole grains.
I suspect that something similar can be said of not doing enough exercise.
We are commenting on a press release from a university, i.e., written by professionals whose job is to increase the prestige of their employer. It is rich for you to criticize Peter Attia, M.D., and ignore that little gem. At least Attia is trying to understand some aspect of reality more fundamental than impression management and increasing the public's awareness of whatever their employers tasks them to target.
Low muscle mass is correlated with shortened lifespan, though. Very low muscle mass is known to be a high risk for early mortality.
> It is a rich for you to criticize Peter Attia, M.D., and ignore that little gem.
I don’t feel bad at all for criticizing Peter Attia, especially after his ringing endorsements of the Oura Ring and claimed “investor” status were revealed to be a contractual deal where he got shares in the company in exchange for promoting the Oura Ring. He’s been revealed to be very oriented toward fame and self-promotion, as evidenced by his recent claims that Kevin Spacey’s accusers were all wrong and Kevin Spacey was totally innocent after Kevin Spacey’s team set up a promotional dinner between Attia and Spacey. I know a lot of listeners develop sorts of parasocial defensive relationships for their podcasters, but you can’t deny the irony of a CVD-focused podcaster starting his podcast career deep into Ketogenic stuff before he saw the writing on the wall and backed off.
https://www.theguardian.com/wellness/2025/mar/11/older-adult...
TL;DR: In the 1980s, Maria Fiatarone got 9 patients in their 80s and 90s to do 2 months of progressive resistance knee extension training. The minimum increase in 1RM strength was 61%.
Bryan Johnson quit using it, but he’s on 40 other things and there are bound to be drug interactions.
That said, limiting consumption of health and fitness influencers is bound to be good for lifespan: you’ll have more time to live.
And yes, it’s hard to make a living as an influencer saying “eat food, mostly green, move your body, and spend time with friends and loved ones.”
If nobody wanted that, then why have we spent so much effort keeping older people alive and in agony for a couple more days, often in direct opposition to their preferences and DNRs?
If nobody wanted a medical system that caused such suffering for such little reason, then how come the excellent How Doctors Die[0] needed to be written?
[0] https://www.saturdayeveningpost.com/2013/03/how-doctors-die/
He went into quite a bit of detail on exercise though, focusing on the predictable ways you'll lose capability as you age, if you don't work really hard on avoiding that.
Obviously, but they operated (and continue to) under a medical paradigm where preventative medicine means just avoiding a few major harm factors and getting just enough exercise to maintain the average observed health metrics for their age group.
People increasingly do not want to wind up like average 70 or 80 year olds, so it is absolutely appropriate that they start shooting for upper 10% or 5% -ile attributes now while they are younger so that in their old age the relative decline still keeps them at those same upper percentiles they will objvetively need to do the things in old age they want to do.
I thought semi-starvation was the one and only strongly evidenced way too extend lifespan? Which obviously doesn’t require, and in fact destroys, muscle mass.
Fit people:
1. more able to "get out of the way" / coordination / agility
2. faster reaction time
3. potentially suffer a reduced injury
4. more likely to recover than die from the injury
There's probably a dozen rational things we could think of to counter that hypothesis.
6. less likely to be nearby other types of random accidents
So in other words, light years better shape through out all age ranges compared to your average American.
This is clearly and obviously false though. Even the assumption is kind of ridiculous.
Of course a person with greater general fitness is going to, on average, suffer less severe consequences from a random accident. Including drowning is especially ridiculous, because that is a scenario where greater physical fitness directly contributes to your ability to survive such an accident.
The results can just as well be interpreted in a way that the original thesis is correct, but greater physical fitness also helps protect your body against the consequences of random accidents.
It’s not the exercise, it’s that people who recreationally work out are in a different socioeconomic group.
Also, this is consistent with many mouse studies.
> The researchers found that men with the highest fitness levels had a 53 per cent lower risk of dying in random accidents. Yet, it is unlikely that the men’s fitness would have such a big effect on their risk of dying in random accidents.
Wait, what?
https://aaafoundation.org/rates-motor-vehicle-crashes-injuri...
The risk of being in a motor vehicle crash increases as one gets older than ~70 years. The risk of dying increases even more dramatically.
Now maybe this effect is independent of physical fitness, but that’s quite an assumption. I would guess the contrary: that poor fitness quite dramatically increases one’s risk of a car crash, both due to reduced motor control and increased risk of various neurological issues.
Vascular dementia, for example, seems very likely to be correlated and often caused by poor fitness. Various sources seem to think that exercise can quite dramatically reduce Alzheimer’s risk. Alzheimer’s disease is related to PCA, and I suspect that PCA is very much under-diagnosed and that it causes a lot of crashes.
https://en.m.wikipedia.org/wiki/Posterior_cortical_atrophy
PCA can cause simultanagnosia and other problems with visual perception.
Have you ever seen an elderly person drive directly at you in a parking lot, at low speed, and appear to be completely unaware of your presence? This could be why.
For anyone from the US reading this just throw it out… we eat significantly less healthy and have significantly more obesity. Sweden as a country is significantly more healthy and active. To everyone reading this and thinking that this applies is sorely mistaken.
Fitness might not “extend your life many years” but being morbidly obese will most certainly end it significantly early.
This is obviously not military age swedes in the 90s. Still, it's not too far below the US's 70%-ish.
I haven't read the underlying paper so maybe they addressed this, but: couldn't this just be because sedentary/unfit people tend to drive everywhere? So they're involved in more car accidents than healthier people, who walk places more.
Also if you're fat and don't exercise, you've got a lower chance of surviving emergency surgery after a car crash, or swimming to safety after falling in water, or dodging a knife ... I don't think it's a given that physical fitness is unrelated to surviving those things.
It could be and that's the point - the exact causal effect of fitness might be overstated because some mortality reduction might have stem from - for example - driving less.
The causality was originally just assumed. Now with this new association, what changed? Logically nothing has changed right? There is no logical conclusion derivable from this.
If there are three options:
1. Sedentary lifestyle (<3 passive or active exercise weekly).
2. Moderate (~1 hour passive exercise daily, like walking a dog a few miles daily).
3. Active (>1 hr active, like running several miles or lifting).
I think there’s huge value in moving out of (1), but am not sure if it’s worth going from (2) to (3) w.r.t longevity
—Attributed to Neil Armstrong