People with elite cardiorespiratory fitness died at a fifth the rate of the least fit

Among 122,007 patients who took a treadmill stress test at the Cleveland Clinic, those in the top 2.3 percent for fitness for their age and sex died at roughly a fifth the rate of those in the bottom quarter. The adjusted hazard ratio was 0.20, with a 95 percent confidence interval of 0.16 to 0.24. Read the other way, the least fit group carried about five times the mortality hazard of the elite group, a ratio of 5.04.

The researchers followed the patients for a median of 8.4 years, and 13,637 died in that time.

Metabolic equivalents and the five fitness tiers

The study, led by Kyle Mandsager, then an electrophysiology fellow, with senior author Wael Jaber, was published in JAMA Network Open in October 2018. Fitness was not estimated from a questionnaire. Patients underwent symptom-limited exercise treadmill testing between 1991 and 2014, and their peak performance was converted into estimated metabolic equivalents, or METs, a unit in which one MET is the body’s resting energy use and higher numbers mean the body is sustaining more work.

Raw METs fall with age and differ between men and women, so the authors ranked each patient against others of the same age and sex. The bottom quarter was labelled low. The groups above that were below average (25th to 49th percentile), above average (50th to 74th), high (75th to 97.6th) and elite, which began at the 97.7th percentile.

Cardiorespiratory fitness is a summary of the whole oxygen supply chain: lungs that load the blood, a heart that pumps it, vessels that deliver it and muscle that uses it. A treadmill test that raises the workload in stages until the patient must stop measures the weakest link in that chain, and a low ceiling can reflect trouble in any of them, including trouble that has not yet produced symptoms. For that reason fitness works as a readout of underlying health as well as a possible cause of it.

Hazard ratios from low fitness to elite

A hazard ratio compares how often an event occurs in one group against another over the same time, after adjusting for other differences. In the Cleveland Clinic data the adjustment covered age, sex, height, weight, body mass index, medications and existing conditions. A value of 1.0 means no difference, and 0.20 means one-fifth the rate. Elite patients had that 0.20 against the low group, and 0.77 against the high group, so the gradient continued even at the top end.

The gradient ran the whole way. Patients just below average had a hazard ratio of 1.41 relative to those just above average. In the authors’ comparison, low fitness carried a risk comparable to or exceeding that of established clinical risk factors, with reported hazard ratios of 1.41 for smoking, 1.40 for diabetes and 1.29 for coronary artery disease.

That is a statement about association in a group of patients who were sent for stress testing, not about a randomised intervention. People who are fit tend to differ from the unfit in diet, income and prior health in ways that statistical adjustment cannot fully erase, and some of the least fit patients may have been unwell at the time of testing.

No ceiling on the benefit

The paper’s central conclusion, in the authors’ words, is that cardiorespiratory fitness is inversely associated with long-term mortality “with no observed upper limit of benefit.” The finding matters because concern had grown that extreme endurance exercise might eventually cause harm. In the Cleveland Clinic’s release, Jaber put the practical reading plainly, saying that aerobic fitness is “something that most patients can control” and that the study found no limit to how much exercise is too much.

Mandsager, quoted in the same release, described the study’s novelty: “This relationship has never been looked at using objectively measured fitness, and on such a large scale.” The EurekAlert version of the release adds that the benefit was especially visible among patients aged 70 and over, for whom elite fitness was associated with about 30 percent lower mortality than high fitness, and among those with hypertension. Elite fitness in the study corresponded to the performance of trained endurance athletes, so the top tier is a demanding standard, reached by a small share of any age group. The practical message in the data is the slope rather than the summit: each step up the percentile ladder was accompanied by a lower hazard, including the steps from below average to above average that most adults can plausibly climb.

The full text is open at PubMed Central and under the DOI 10.1001/jamanetworkopen.2018.3605. What the data cannot say is how much a sedentary person gains by moving up one tier. The paper reports differences between groups of different people, not the change within one person who trains, and that within-person effect is the number the cohort leaves open.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview