Exercise & VO2max — the strongest longevity lever (causal, too much)

Cardiorespiratory fitness is the strongest modifiable predictor of how long you live. Not diet, not sleep, not any supplement. Fitness, measured as VO2max, sits at the top of the longevity hierarchy, backed by decades of dose-response data and, critically, randomized trial evidence that improving it changes your biology [1] [2].

Fitness predicts mortality, and the relationship is dose-dependent

Each MET of cardiorespiratory capacity you gain translates into a quantifiable reduction in mortality risk [3]. The dose-response curve is steep at the low end. Moving from sedentary to merely "below average" fitness cuts risk dramatically, and it doesn't flatten out at high fitness levels [1] [2]. There is no identified ceiling where becoming fitter stops helping you.

The most common objection to this picture is reverse causation: maybe sick people are less able to exercise, so the association between low fitness and early death reflects illness driving inactivity, not inactivity driving illness. A randomized controlled trial by Howden and colleagues answers this directly. Two years of structured aerobic training in previously sedentary middle-aged adults reversed cardiac stiffening, a concrete, measurable change in cardiac structure that accumulates with age and sedentary behavior [4]. You can't fake that in an RCT with a placebo. Combined with the consistent dose-response pattern [1] [2] and the per-MET mortality quantification [3], the causal case is as strong as non-pharmaceutical lifestyle evidence gets. Some component of fitness as a marker of underlying health almost certainly remains, but "get fitter, live longer and function better" is well-supported.

How much is too much?

Cardiologist James O'Keefe has argued that extreme endurance exercise produces a U-shaped mortality curve, that decades of high-volume training raise atrial fibrillation (AFib) risk and coronary artery calcium (CAC) scores enough to erode the benefit. The AFib and CAC signals are real in lifelong extreme endurance athletes and deserve attention.

The mortality data, though, don't support a harmful upper limit for the vast majority of people. Arem and colleagues found no excess mortality even among individuals exercising at ten times the standard public-health guideline [2]. The AFib and coronary calcium signals appear at true physiological extremes, meaning decades of competitive high-volume endurance training, and even there, the overall cardiovascular benefit appears to hold. For the roughly 99% of the population who fall well short of those volumes, this concern is not relevant to their decision-making.

Daniel Lieberman's perspective is relevant here: the public-health problem with exercise is overwhelmingly too little, not too much [5]. Framing moderate-to-vigorous exercise as dangerous inverts the actual risk profile.

What this means practically

Sedentary adults who underwent two years of supervised aerobic training reversed age-related cardiac stiffening [4]. Mortality risk continued to fall at exercise volumes far above current guidelines, with no upturn in the data [2]. The per-MET mortality quantification gives a concrete sense of what each increment of fitness contributes [3].

The signal from this body of evidence is unusually clear. Cardiorespiratory fitness has a causal-grade relationship with longevity. The dose-response extends to high volumes without a mortality penalty, and raising VO2max through training produces measurable cardiac rejuvenation in randomized trials. For everyone, the error is exercising too little.

Where the evidence is debated

Reverse causation. Observational fitness-mortality studies cannot fully exclude the possibility that underlying illness reduces both exercise capacity and survival simultaneously. The Howden RCT [4] substantially weakens this objection by showing training causes cardiac changes, but critics note that hard mortality endpoints from long-term VO2max training RCTs remain scarce, since they are logistically difficult to run. The causal interpretation is well-supported, not proven beyond all doubt.

Extreme-endurance risk. The AFib and coronary calcium elevations seen in lifelong high-volume endurance athletes are a genuine, replicated signal. The unresolved question is threshold: how many years, at what weekly volume, does risk begin to accumulate meaningfully? Arem et al. found no mortality harm at ten times guideline volumes [2], but that study did not specifically enrich for multi-decade competitive athletes. Whether a mortality-relevant U-curve exists at true physiological extremes remains an open research question, one that does not change the practical conclusion for almost anyone reading this.

References

1. Mandsager et al. 2018, JAMA Netw Open — CRF & mortality · PubMed 30646252
2. Arem et al. 2015, JAMA IM — physical activity dose-response · PubMed 25844730
3. Kodama et al. 2009, JAMA — CRF & mortality meta-analysis · PubMed 19454641
4. Howden/Levine et al. 2018, Circulation — exercise reverses cardiac aging (RCT) · PubMed 29311053
5. ZOE — Daniel Lieberman: exercise myth-busting (evolutionary)