The short version
- In 750,302 US veterans followed a median of 10.2 years, the least fit fifth had an all-cause mortality hazard ratio of 4.09 compared with the extremely fit, and the authors concluded that being unfit carried a greater risk than any cardiac risk factor they examined.
- Mortality risk kept falling up to roughly 14 METs with no evidence of harm at extreme fitness, giving hazard ratios of 0.24 in men and 0.23 in women.
- The 1989 JAMA study of 13,344 adults found age-adjusted mortality fell from 64.0 to 18.6 deaths per 10,000 person-years across fitness quintiles in men and from 39.5 to 8.5 in women.
- Cardiorespiratory fitness can be estimated without exercise testing from age, sex, BMI, resting heart rate and reported activity, and that estimate still predicted mortality in 32,319 adults at a hazard ratio of 0.85 per standard deviation in men.
- A meta-analysis of 14 trials in 429 middle-aged and older adults found interval training raised VO2 max by 2.26 mL/kg/min versus 1.34 for moderate continuous training.
Ask a preventive cardiologist which single measurement tells them most about the person sitting in front of them, and a surprising number will not say cholesterol. They will say how long that person can stay on a treadmill.
This article is the anchor for everything else we publish about living longer, because it is the one number on the list with both a huge evidence base and a training response. Almost nothing else in the category has both.
The largest look at this so far
Kokkinos and colleagues followed 750,302 US veterans aged 30 to 95 who had completed a standardised treadmill exercise test. Over a median 10.2 years and 7.8 million person-years, 174,807 of them died.
Sorted into age- and sex-specific fitness categories, the association was inverse and graded the whole way up. The least fit fifth carried a hazard ratio of 4.09 against the extremely fit. The lowest mortality risk sat at roughly 14 METs, with hazard ratios of 0.24 in men and 0.23 in women, and there was no sign of risk turning back upward at the very top — no penalty for being unusually fit. The authors' own summary of the whole dataset is the sentence worth keeping: being unfit carried a greater risk than any of the cardiac risk factors they examined.
What a MET actually is
One MET is the oxygen cost of sitting quietly, about 3.5 millilitres of oxygen per kilogram of body weight per minute. Your VO2 max is the ceiling on that — the most oxygen your body can take in, move and use at full effort. Peak METs achieved on a graded treadmill test is the practical stand-in, and it is what most of the mortality literature is actually built on.
What makes it a good marker is that it is not a marker of one organ. It sums up how much blood your heart moves per beat, how much blood you have, how densely your muscles are capillarised, and how many mitochondria are in them. That breadth is exactly why no capsule shifts it and why training does.
This is not a new result
Blair and colleagues published the founding version in JAMA in 1989: 10,224 men and 3,120 women given a maximal treadmill test and followed for just over eight years. Age-adjusted all-cause mortality fell across fitness quintiles from 64.0 to 18.6 deaths per 10,000 person-years in men, and from 39.5 to 8.5 in women. Nearly four decades later, in a cohort seventy times larger, the shape of the curve has not changed.
The caveat we would rather you heard from us
Every study above is observational, and no one will ever randomise people to be unfit for twenty years. Two problems follow. Illness that has not yet declared itself lowers your treadmill result and raises your death risk, so some of the association is disease running backwards through the data. And clinic-referred cohorts are not the general population.
So treat the hazard ratios as generous. What survives the criticism is the direction, the consistency across four decades and several continents, and the more measured claim the American Heart Association made in its 2016 scientific statement: fitness deserves to be treated as a clinical vital sign, because adding it to conventional risk factors meaningfully improves how people get classified.
Getting your number without a lab
A graded exercise test ordered through your doctor is the real measurement. If that is not available, there is a validated shortcut: Stamatakis and colleagues estimated fitness from age, sex, BMI, resting heart rate and self-reported activity in 32,319 adults, and that estimate still tracked mortality — a hazard ratio of 0.85 per standard deviation in men and 0.88 in women over a mean nine years of follow-up. It also discriminated better than any of its own component parts.
The estimate is coarse. It is also free, and a coarse number you retest every six months beats a precise number you never get.
How much of it can you actually move?
A meta-analysis of 14 trials in 429 middle-aged and older adults found interval training raised VO2 max by 2.26 mL/kg/min and moderate continuous training by 1.34, with interval work ahead by 1.10 mL/kg/min in the direct comparison. In MET terms that is a few tenths of a MET from a short structured programme, and more with more time. This is the part of the story that is genuinely under your control, and it is slower and less dramatic than the internet suggests.
Where this leaves the rest of the shelf
Hold the four-fold hazard ratio in your head while you read anything else in this category. Senolytics are a promising mechanism with no proven human lifespan effect. Biological age tests disagree with each other badly enough that a single reading tells you very little. The omega-3 index is a real biomarker whose supplement trials keep coming back flat. Sarcopenia matters enormously for whether you stay independent, and it too is trained rather than bought.
None of those are worthless. They are just all smaller than this, which is the entire logic behind how we would rank a longevity budget: the interventions with the largest effects are the ones nobody can sell you.
What we would actually tell you
Get a number. Any number — a clinic test, an estimate, a timed hill you can repeat. Write it down with the date.
Then build a week that plausibly raises it: mostly easy aerobic work you can sustain conversationally, plus one or two harder intervals sessions if your joints and your schedule allow. Retest in six months and compare against yourself, not against a chart.
We sell supplements and we are telling you that none of them will do this. That is not modesty. It is that the trials do not exist, and we are not going to imply they do.
Good questions
What is a good VO2 max for my age?
Ask instead where you sit in your own age and sex band, because that is how the mortality data is actually built. In the 750,302-veteran cohort, risk kept dropping up to about 14 METs, roughly 49 mL/kg/min, with no penalty above that. Your first reading matters far less than whether the second one, six months later, is higher.
Can any supplement raise my VO2 max?
Nothing on our shelf will raise it on its own, and we would not sell you one on that promise. VO2 max reflects cardiac output, blood volume, capillary density and muscle mitochondria all at once, which is why it responds to weeks of training and not to a capsule. If a product implies otherwise, ask which trial measured it.
Is a smartwatch VO2 max estimate accurate?
Treat it as a trend line, not a measurement. Wrist estimates are built from heart rate and pace models, not from analysed expired gas, so the absolute figure can be well off. The useful signal is the direction over months on the same device with the same kind of runs. A validated non-exercise estimate is comparable in spirit and equally approximate.
Does this apply if I am already over 70?
Yes, and that is one of the clearer parts of the veteran data, which included septuagenarians and octogenarians and found the same graded relationship across the age spectrum. Improvements are slower and the intensity has to be set against your joints and any cardiac history. Talk to your doctor before starting a hard interval programme, not after.
Is high-intensity interval training necessary, or is walking enough?
Walking counts and consistency beats intensity you will not repeat. In middle-aged and older adults, interval training produced a larger VO2 max gain than moderate continuous training by 1.10 mL/kg/min, which is real but modest. Most people get further by adding easy aerobic minutes they enjoy than by adopting a hard protocol they abandon in a month.
Should I pay for a lab VO2 max test?
Only if the number will change what you do. A clinical graded exercise test through your doctor gives you a diagnostic-grade result and screens your heart at the same time, which is the version worth paying for. A commercial performance-lab test is a nice-to-have. If the money is coming out of a budget that could fund a gym membership instead, buy the gym membership.
Sources
- Kokkinos P, Faselis C, Samuel IBH, et al. Cardiorespiratory Fitness and Mortality Risk Across the Spectra of Age, Race, and Sex. J Am Coll Cardiol, 2022. View study
- Blair SN, Kohl HW 3rd, Paffenbarger RS Jr, et al. Physical fitness and all-cause mortality. A prospective study of healthy men and women. JAMA, 1989. View study
- Ross R, Blair SN, Arena R, et al. Importance of Assessing Cardiorespiratory Fitness in Clinical Practice: A Case for Fitness as a Clinical Vital Sign: A Scientific Statement From the American Heart Association. Circulation, 2016. View study
- Stamatakis E, Hamer M, O'Donovan G, Batty GD, Kivimaki M. A non-exercise testing method for estimating cardiorespiratory fitness: associations with all-cause and cardiovascular mortality in a pooled analysis of eight population-based cohorts. Eur Heart J, 2013. View study
- Poon ET, Wongpipit W, Ho RS, Wong SH. Interval training versus moderate-intensity continuous training for cardiorespiratory fitness improvements in middle-aged and older adults: a systematic review and meta-analysis. J Sports Sci, 2021. View study
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.
