Summary
A medRxiv preprint analysing 12,944 participants from nationally representative cohorts in China, England and the United States found that higher estimated cardiorespiratory fitness was associated with a lower risk of developing two or more cardiometabolic conditions. The observational study suggests that a non-exercise fitness estimate may help characterise risk, but it does not establish that increasing fitness prevents these conditions.
A secondary analysis of three nationally representative ageing cohorts found that people with higher estimated cardiorespiratory fitness had a lower risk of developing cardiometabolic multimorbidity, defined as the coexistence of at least two of hypertension, diabetes, heart disease and stroke.
The medRxiv preprint, posted on September 14, 2026, included 12,944 participants from the China Health and Retirement Longitudinal Study, the English Longitudinal Study of Ageing and the US Health and Retirement Study. During follow-up, 3,029 participants developed cardiometabolic multimorbidity.
In the pooled analysis, each one-standard-deviation increase in estimated cardiorespiratory fitness was associated with a hazard ratio of 0.61 for cardiometabolic multimorbidity, with a 95% confidence interval of 0.58 to 0.64. Compared with participants in the low-fitness category, those in the high-fitness category had a hazard ratio of 0.40, with a 95% confidence interval of 0.35 to 0.45. The reported trend across fitness categories was statistically significant, with P<0.001.
How the study estimated fitness
Cardiorespiratory fitness describes the body’s capacity to deliver and use oxygen during physical activity. Instead of measuring it with an exercise test, the researchers estimated it using a validated, sex-specific non-exercise equation. This makes the measure easier to derive from large health studies, although it remains an estimate rather than a direct exercise-test measurement.
The researchers used Cox proportional-hazards models to examine whether estimated fitness was associated with new cases of cardiometabolic multimorbidity. They also conducted supplementary analyses addressing proportional-hazards assumptions, competing risk from death, missing data, prediction performance and results across the individual cohorts.
The association remained directionally consistent in these sensitivity analyses. Adding estimated fitness to sociodemographic and behavioural predictors also improved the models’ ability to distinguish participants who did and did not later develop cardiometabolic multimorbidity, although the authors described the additional prognostic information as modest.
What the findings mean
The consistent direction of the association across cohorts in China, England and the United States suggests that estimated cardiorespiratory fitness may be useful for characterising cardiometabolic risk in ageing populations. A non-exercise measure could be incorporated into analyses or risk-assessment approaches that already collect demographic and behavioural information.
The study was observational and used previously collected, de-identified data. It therefore identifies an association rather than testing whether an exercise programme or another intervention changes the incidence of cardiometabolic multimorbidity. The authors also note heterogeneity between the cohorts, which is important when applying the pooled estimate to other populations.
The findings are presented in a medRxiv preprint and should be interpreted as preliminary research. They support further evaluation of how estimated fitness measures perform in clinical and public-health risk assessment, rather than establishing a treatment effect.