Moderate-level physical activity was associated with more favorable biological aging indicators relative to low-level PA, but high-level PA showed no significant associations in fully adjusted models, and associations varied across dimensions and subgroups.
Key Findings
Results
Moderate-level physical activity was independently associated with lower allostatic load, phenotypic age acceleration, and KDM biological age acceleration compared with low-level PA.
Allostatic load: β=-0.219, 95%CI: -0.308 to -0.130.
PhenoAgeAccel: β=-0.524, 95%CI: -0.780 to -0.268.
KDM Accel: β=-0.638, 95%CI: -1.073 to -0.202.
Analysis used weighted multivariable linear regression after covariate adjustment in 13,379 middle-aged and older adults from NHANES (2003-2010 and 2015-2018).
Results
High-level physical activity (>3000 MET-min/week) showed no significant associations with any of the biological aging indicators in the primary fully adjusted models.
PA was categorized as low (<600), moderate (600-3000), and high (>3000) MET-min/week based on self-report.
No significant association was found for AL, HD, PhenoAgeAccel, or KDM Accel at the high PA level.
Results
Dose-response analyses showed nonlinear relationships between PA and AL, PhenoAgeAccel, and KDM Accel, with the most favorable estimates at moderate PA levels.
Restricted cubic splines were used for dose-response analysis.
All P for nonlinearity were <0.001 for AL, PhenoAgeAccel, and KDM Accel.
Results
Subgroup analyses confirmed the main patterns and revealed a consistent positive association between high-level PA and KDM Accel across age groups.
A positive association implies higher KDM biological age acceleration with high-level PA.
This was observed consistently across age groups in subgroup analyses.
Results
Among cancer survivors, a reversed association was observed for homeostatic dysregulation.
This was identified in subgroup analyses and differed from the pattern in other participants.
The abstract describes it as a 'reversed association for HD among cancer survivors'.
Results
Estimates for allostatic load may be conservative because BMI was included as a covariate.
BMI is a component related to AL and may lie on the pathway between PA and AL, so adjustment may attenuate associations.
Conclusions
The relationship between PA and biological aging is multidimensional, varying across the four clinical indicators and subgroups.
The study integrated four indicators: allostatic load, homeostatic dysregulation, PhenoAgeAccel, and KDM Accel.
Cross-sectional design limits causal inference; the authors call for longitudinal and interventional research.
What This Means
This study used U.S. national survey data on 13,379 middle-aged and older adults to examine how self-reported physical activity relates to biological aging, which reflects how old the body appears based on routine clinical measures rather than calendar age. The researchers looked at four different measures of biological aging, rather than relying on DNA methylation clocks as most prior studies have.
People who reported moderate physical activity (600-3000 MET-min/week) had more favorable scores on three of the measures (allostatic load, phenotypic age acceleration, and KDM biological age acceleration) than those with low activity. Those with high activity (above 3000 MET-min/week) did not show significant differences in the main analyses. The relationship appeared nonlinear, with the best estimates at moderate levels. Some subgroup patterns differed, such as a link between high activity and higher KDM age acceleration across age groups, and a reversed pattern for one measure among cancer survivors.
This research suggests that moderate activity is associated with a younger biological profile, but because the data are a one-time snapshot and activity was self-reported, it cannot show that activity causes slower aging. The authors note that longitudinal and interventional studies are needed to confirm these patterns.
Du M, Zhang W, Xia L, Liu Y, Song Y, Li J, et al.. (2026). Multisystem assessment of the association between physical activity and biological aging: Evidence from a national cross-sectional survey.. Science progress. https://doi.org/10.1177/00368504261502626