The guide

Why does VO2 max predict how long you live?

By GyroscopeUpdated November 14, 20268 sources

Because it measures how much work your heart, blood and muscles can do together. In 122,007 people tested at the Cleveland Clinic, being below average for your age and sex, against above average, carried 1.41 times the risk of death. Smoking carried 1.41 too.

Isn’t it just that healthy people can run faster?

People who are already sick can’t do well on a treadmill. Maybe the test is spotting illness, and fitness itself does nothing.

The researchers knew this, and the big studies try to take it out. In the Cleveland Clinic study, the results were adjusted for age, sex, body mass index, heart disease, diabetes, high blood pressure, high cholesterol, kidney failure, smoking and the medicines people took. The link held after all of that. Every one of the studies is still observational, though. They follow people. They don’t assign fitness at random, and no adjustment can fix that completely.

We still think the case is strong. The link shows up in study after study, it rises step by step with fitness, it holds in people with and without disease, and the one thing it measures, how much oxygen your body can use, is something training changes.

What did 122,007 treadmill tests show?

From 1991 to 2014, the Cleveland Clinic gave treadmill tests to 122,007 adults sent by their doctors, with an average age of 53. A Cleveland Clinic team (Mandsager and colleagues) sorted them by fitness for their age and sex and followed them for a median of 8.4 years, through 13,637 deaths.

Being unfit for your age carried as much risk as smoking and more than heart disease. The fittest 2.3 percent did better than the merely high group, so the benefit didn’t stop at “fit enough”. And in the patients aged 70 and older, elite still beat high (0.71 times the risk).

But these were people sent for a stress test, not a random sample of Americans. Fitness was estimated from how far they got on the treadmill, not measured with a mask. And it was one test, at one point in their lives.

Does each point count?

Yes, roughly. In 2009 researchers pooled 33 studies with 102,980 people. Each 1 MET higher, which is 3.5 ml/kg/min or about the difference between jogging at one speed and 1 km/h (0.6 mph) faster, went with 13 percent lower risk of death. A 2024 overview of 26 reviews, covering 20.9 million observations, put it at 11 to 17 percent per MET, and high against low fitness at less than half the risk (0.47).

Those figures don’t add up point by point. Going up 3 points doesn’t cut your risk by 30 percent or more. Risk ratios multiply, and the curve is steepest at the bottom. In the 2009 pooled studies, low fitness (under 7.9 METs, about 28 ml/kg/min) carried 1.40 times the risk of middling fitness and 1.70 times the risk of high fitness. Most of the gap was in the first step.

The same overview found the link for more than death. Each MET higher went with 18 percent lower risk of heart failure. The authors rated the certainty of the evidence across all the studies from very low to moderate, mostly because studies that follow people can’t rule out other causes.

A simple way to picture it

Think of the spare capacity in a building’s power supply. On a normal day it doesn’t matter whether the building could draw twice what it uses. On the day a heat wave hits and every air conditioner runs at once, the building with headroom keeps the lights on.

VO2 max is your body’s headroom. Sitting still uses 1 MET. Climbing stairs, carrying a child or hurrying for a bus all ask for more. The higher your ceiling, the smaller a share of it everyday life takes.

Why would fitness protect you?

VO2 max is mostly a measure of the heart. In healthy people its ceiling is how much oxygen-filled blood the heart can pump, and training raises it mainly by raising the heart’s maximum output (Bassett and Howley, 2000). A high number means a heart that can do a lot of work, and blood and vessels that can carry it.

Nobody has worked out how much of the link is the fitness itself and how much is everything that comes with being fit. Fit people tend to move more every day, weigh less and have lower blood pressure, and some of the adjustments above take part of that out. What we do know is that cardio lowers blood pressure, which bloodpressure.guide covers, and that low fitness late in life tracks loss of independence. Below about 18 ml/kg/min for men and 15 for women, it seems likely (Shephard 2009).

What did the trial find?

Only one long trial has assigned older people to training and counted deaths. In Generation 100, 1,567 people aged 70 to 77 in Trondheim, Norway, were assigned for five years to two sessions a week of 4-minute intervals, two sessions a week of steady 50-minute workouts, or the national activity guidelines.

Deaths came to 4.7 percent in the guidelines group. The interval group came out 1.7 percentage points lower and the steady group 1.2 points higher, and none of those differences was clear enough to rule out chance. All three groups held their fitness for five years, against an expected fall of 3.5 ml/kg/min, and the guidelines group chose to do more of its own exercise hard than the steady group did. Most participants were already healthy and active, with an average starting VO2 max of 28.6, near the top 10 percent for people their age.

The trial didn’t prove that training lowers deaths, and it didn’t rule it out either. In a fit, healthy group there were few deaths to count, and the comparison group trained too.

Where does that leave you?

We’re confident that low fitness goes with a much higher risk of death, at about the size of smoking, across hundreds of studies and millions of people. We’re fairly confident that raising your fitness lowers that risk, because the link rises step by step and fitness is something you change. Nobody has shown it in a trial big enough to be sure.

For the decision in front of you, that’s enough. Training costs little, it raises the number in nearly everyone who does enough of it, and it’s how people hold their fitness into their 70s. How to raise it has the plan.

What would one group up mean for you?

A relative risk like 1.41 or 5.04, measured in one group of patients, isn’t a number of years for you. lifespan.guide turns your habits into years on the life table. Try its life expectancy calculator or see what adds the most years.

To see where you stand and how far the next group is, use where do you stand?. Before you test yourself, keep in mind that a watch number can be several points off, and that the change linked to lower risk in the big reviews, 1.75 to 3.5 ml/kg/min, is about the size of a watch’s error for one person (INTERLIVE 2022). Use the same method each time.