High Cardiorespiratory Fitness Cuts Multimorbidity Risk by 21% Over 15 Years

Multimorbidity—the coexistence of two or more chronic conditions—now affects roughly one-third of adults globally, rising sharply with age and driving massive healthcare costs, disability, and mortality. Identifying modifiable factors that slow the accumulation of chronic disease is a central challenge for longevity medicine. This large-scale prospective study directly addresses that gap by examining whether cardiorespiratory fitness (CRF) influences not just individual disease risk but the broader trajectory of multi-disease accumulation over time.

Researchers enrolled 38,348 adults from the UK Biobank (mean age 55.2 years; 52% female) who were free of chronic disease at baseline. CRF was measured using a 6-minute submaximal cycling ergometer test, with estimated VO2max converted to metabolic equivalents (METs) and then standardized by age and sex before being tertiled into low, moderate, and high groups. Over a median follow-up of 11.6 years, participants were tracked via electronic health records for the incidence of 59 chronic diseases spanning metabolic, cardiovascular, and neuropsychiatric categories. Multimorbidity was defined as accrual of two or more conditions. Cox regression, Laplace regression, and linear mixed-effects models were applied to assess risk, timing, and trajectory of disease accumulation.

The results were striking. Compared to the low-CRF group, participants with high CRF had a 21% lower risk of developing multimorbidity (HR: 0.79; 95% CI: 0.76–0.83). The protective effect was also seen for moderate CRF (HR: 0.89; 95% CI: 0.85–0.92). Beyond risk reduction, high CRF delayed the median time to multimorbidity onset by 1.27 years (95% CI: 1.01–1.54). Crucially, linear mixed-effects models revealed that high CRF was associated with a significantly slower annual rate of chronic disease accumulation (β = −0.043; 95% CI: −0.050 to −0.036), meaning the protective effect compounds over decades. Benefits were observed across both middle-aged (<60 years) and older (≥60 years) subgroups, with no significant interaction by age group. Sankey plot visualizations illustrated clear divergence in disease trajectories between CRF categories over follow-up.

Sensitivity analyses reinforced these findings: results were consistent after excluding participants who developed multimorbidity within the first 3 years (reducing reverse causality concerns), using non-standardized CRF values, applying Fine-Gray competing risk models for death, and performing multiple imputation for missing covariate data. Adjustments for sociodemographic factors, smoking, alcohol, and physical activity did not materially change associations, suggesting CRF has an effect independent of general physical activity levels.

These findings carry meaningful implications for clinical practice and public health. CRF is a measurable, trainable physiological parameter—unlike many genetic risk factors. The data suggest that improving CRF could compress morbidity into a shorter window near end of life, extending the healthspan even if total lifespan is unchanged. However, as an observational study, causality cannot be confirmed. CRF was measured only at baseline, and changes in fitness over follow-up were not captured. The UK Biobank cohort is also predominantly White and healthier than the general population, limiting generalizability.