Divergent associations of estimated cardiorespiratory fitness with cognitive trajectories in Chinese and U.S. middle-aged and older adults: A descriptive cross-national comparison from CHARLS and HRS.

INTRODUCTION: The association between cardiorespiratory fitness (CRF) and cognition may differ across populations, but cross-cultural evidence remains limited. We explored whether estimated CRF (eCRF) shows divergent associations with cognitive trajectories in Chinese and U.S. middle-aged and older adults. METHODS: We included 3458 participants from CHARLS and 6905 from HRS. Cognitive trajectories were identified using latent class growth modelling. eCRF was estimated using sex-specific non-exercise algorithms. The unit of eCRF is metabolic equivalents of task (MET). Multinomial logistic regression examined the association between eCRF (per 1-MET increase) and trajectory membership, with the persistently high trajectory as the reference. RESULTS: Three cognitive trajectories-persistently high, moderate, and low-were identified in both cohorts. In CHARLS, higher eCRF was unexpectedly associated with increased odds of the low (odds ratio [OR] = 1.28, 95% confidence interval [CI]: 1.16-1.43) and moderate (OR = 1.10, 95% CI: 1.03-1.18) trajectories. In HRS, higher eCRF was associated with lower odds of the moderate trajectory (OR = 0.94, 95% CI: 0.90-0.98), with no significant association for the low trajectory. DISCUSSION: This descriptive observation reveals divergent associations in the eCRF-cognition relationship between Chinese and U.S. adults, but should be interpreted cautiously. The finding persisted across multiple sensitivity analyses. However, given the lack of direct CRF measurement and objective assessment of activity type, we cannot distinguish whether this divergence reflects measurement differences and residual confounding, or true biological heterogeneity. Context-specific validation of eCRF equations and objective activity-specific measures are needed to clarify eCRF's role in cognitive aging.