VO2 Max and Longevity: What 122,007 Patients Taught Us

VO2 max is one of the strongest predictors of how long you'll live. In a study of 122,007 patients, the least fit group had roughly five times the mortality risk of the fittest — a gap larger than smoking, diabetes, or heart disease. And the benefit never plateaued: more fitness kept meaning less risk.

That study is Mandsager et al., published in JAMA Network Open in 2018, and it's the reason I stopped thinking of cardio as a weight-loss tool and started thinking of it as the closest thing we have to a longevity drug. This article walks through what the study actually found, why the numbers are so striking, and what they imply for training.

What did the Mandsager 2018 study actually do?

Between 1991 and 2014, the Cleveland Clinic ran treadmill exercise tests on 122,007 patients. Not surveys. Not self-reported activity. Actual maximal treadmill tests, which are the clinical gold standard for measuring cardiorespiratory fitness.

The researchers then followed those patients for a median of 8.4 years — about 1.1 million person-years of observation in total — and recorded 13,637 deaths. Because they had a real measured fitness value for every patient, they could sort the entire cohort into performance groups matched for age and sex:

• Low: below the 25th percentile
• Below average: 25th to 49th percentile
• Above average: 50th to 74th percentile
• High: 75th to 97.6th percentile
• Elite: 97.7th percentile and above

Then they asked a simple question: how does fitness group relate to dying, of anything, from any cause?

How big is the mortality gap between low and elite fitness?

Enormous. After adjusting for the usual confounders, elite performers had an adjusted hazard ratio of 0.20 compared with the low-fitness group. Flip that around: the least fit patients died at roughly five times the rate of the fittest ones over the follow-up period.

Here are the key comparisons the paper reported, all statistically significant:

Two things in that table deserve a second look.

First, the jump from below average to above average. That's not elite athlete territory — that's moving from the 40th percentile to the 60th. It was still associated with a 41% difference in mortality risk. The biggest gains go to people climbing out of the bottom half.

Second, the elite vs. high comparison. Even people in the 75th to 97.6th percentile — already fit by any reasonable standard — had measurably higher mortality than the elite group. Which brings us to the study's most surprising finding.

Is there an upper limit to the benefit of fitness?

No — and this is what made the study famous. Before 2018, there was a live debate about whether extreme fitness might backfire. Marathoners with coronary calcium. "Too much exercise" think pieces. Mandsager's cohort was big enough to actually test it.

The answer: benefit kept accruing at every level, all the way to the top. The elite group — people above the 97.7th percentile for their age and sex — had the lowest mortality of anyone, including a significant edge over the merely "high" group. The authors found no upper threshold beyond which fitness stopped helping or started hurting.

That doesn't mean everyone needs an elite VO2 max. It means the dose-response curve doesn't flatten where most people assume it does.

How does low fitness compare to smoking and diabetes?

This is the part that changed how I train. Within the same cohort, the researchers computed adjusted hazard ratios for the traditional risk factors medicine spends most of its energy on:

• Smoking: HR 1.41
• Diabetes: HR 1.40
• Coronary artery disease: HR 1.29

Now compare those to the fitness gaps above. Being below average instead of above average (HR 1.41) carried the same statistical weight as being a smoker. Being in the low group instead of elite carried several times the weight of any single traditional risk factor.

Peter Attia has made this point repeatedly in his public writing at peterattiamd.com: in his reading of this data, a low VO2 max predicts death more strongly than smoking, diabetes, or hypertension, which is why he ranks cardiorespiratory fitness at or near the top of the modifiable variables for lifespan. I'm paraphrasing his framing rather than quoting him, but the underlying numbers are right there in the paper — and they're hard to argue with.

Yet no physician has ever asked me for my VO2 max at an annual physical. We screen relentlessly for the 1.3–1.4x risk factors and mostly ignore the 5x one.

How much life expectancy does each point of VO2 max buy?

Rhonda Patrick has popularized a useful way to make this concrete. In her FoundMyFitness content at foundmyfitness.com, she cites research estimating that each increase of roughly 1 ml/kg/min in VO2 max is associated with about 45 extra days of life expectancy. That's an epidemiological association, not a guarantee — but it turns an abstract lab value into something you can do arithmetic with.

Run the arithmetic. In Helgerud's 2007 trial, eight weeks of 4x4 interval training improved VO2 max by about 7.2% — call it 3 to 4 ml/kg/min for a typical adult starting in the 45–55 range. By Patrick's cited figure, that's on the order of five to six months of associated life expectancy, from two months of training. I cover that study in detail in the 2007 study that made the 4x4 famous.

Can you actually move your VO2 max percentile?

Yes, and this matters because Mandsager is observational. It shows fit people die less; it can't prove that becoming fit transfers the benefit. But the interventional evidence is strong that VO2 max itself is trainable at any age:

• Helgerud 2007: +7.2% in 8 weeks with 4x4 intervals in already-active men.
• Howden 2018: ~18% over two years in sedentary middle-aged adults, alongside a greater than 25% improvement in left-ventricular compliance. I break that one down in the 2-year study that reversed cardiac aging.

Because the Mandsager percentiles are age- and sex-matched, a mid-single-digit ml/kg/min improvement is often enough to move an average person up a full fitness group — say, from below average to above average. Per the table above, that's the 41% gap.

The most studied tool for making that climb efficiently is the Norwegian 4x4 protocol: four rounds of 4 minutes at 85–95% of max heart rate with 3 minutes of easy recovery between them. About 38 minutes including warm-up and cooldown. If you're wondering how often to run it, I've written up the evidence on frequency in how often should you do the Norwegian 4x4, and the recovery metric worth tracking alongside it in heart rate recovery.

What's the honest takeaway?

Mandsager 2018 doesn't prove causation, and no single observational study should run your life. But when a 122,007-person cohort shows a 5x mortality gradient, no upper limit to benefit, and effect sizes that dwarf the risk factors we actually screen for — and when randomized trials confirm the underlying variable is trainable — the rational move is obvious. Train your VO2 max like your life depends on it, because the data suggests it does.

---

This study is why I built Viking 4x4 around one number. The app runs the 4x4 protocol against your live heart rate, then tracks the signals of a rising VO2 max — per-interval heart-rate reports and 60-second heart-rate-recovery trends — week over week. Other apps time your workout. Viking 4x4 shows your heart adapting.