Kokkinos · Journal of the American College of Cardiology 2022 · retrospective cohort study · n=750,302

Cardiorespiratory Fitness and Mortality Risk Across the Spectra of Age, Race, and Sex.

Cited 254 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized observational cohort study with long-term follow-up

PubMed 35926933 · doi:10.1016/j.jacc.2022.05.031 · record verified 2026-08-26

What was done

Researchers analyzed 750,302 U.S. veterans aged 30 to 95 years (mean age 61.3 ± 9.8 years), including 110,637 septuagenarians, 26,989 octogenarians, 142,798 African Americans, 35,197 Hispanics, 16,050 Native Americans, and 45,232 women. Cardiorespiratory fitness (CRF) was measured by peak metabolic equivalents (METs) achieved during standardized exercise treadmill tests and grouped into age- and sex-specific categories (quintiles and 98th percentile). Multivariable Cox proportional hazards models estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause mortality.

What was found

Over a median follow-up of 10.2 years (7,803,861 person-years), 174,807 participants died (22.4 per 1,000 person-years). CRF showed an inverse, graded association with mortality across age, sex, and race. The lowest mortality risk was observed at ~14.0 METs for men (HR: 0.24; 95% CI: 0.23–0.25) and women (HR: 0.23; 95% CI: 0.17–0.29), with no excess risk at extremely high fitness. Compared to the most fit (98th percentile), the least fit (lowest 20th percentile) had over 4-fold higher mortality risk (HR: 4.09; 95% CI: 3.90–4.20).

Why it matters

This study shows that the survival benefit of high cardiorespiratory fitness applies across diverse demographic groups, including elderly individuals and racial minorities, with no evidence of harm at extreme fitness levels.

Limits

The cohort consists entirely of U.S. military veterans who were referred for clinical exercise testing, which may limit generalizability to general populations. As an observational study, residual confounding from unmeasured health behaviors and baseline medical conditions cannot be fully excluded.

Cited by