Comparing Biological Age Estimates Using Domain-Specific Measures From the Canadian Longitudinal Study on Aging

Area of Science:

• Gerontology
• Biomedical Science
• Aging Research

Background:

• Biological age (BA) estimates are increasingly used to predict health outcomes in older adults.
• Current methods often assess BA validity by its correlation with chronological age (CA).
• The optimal approach for validating BA estimates and identifying informative measure panels remains unclear.

Purpose of the Study:

• To compare different panels of biological measures for estimating BA.
• To evaluate how well these BA estimates approximate CA and predict health-related outcomes.
• To determine if CA approximation or health outcome association is a better metric for BA validity.

Main Methods:

• Utilized data from the Canadian Longitudinal Study on Aging.
• Composed BA measure panels based on physiological systems (blood, organ function, physical/cognitive performance) and an age-optimized panel.
• Compared BA estimates from these panels based on their association with CA and health outcomes (frailty, multimorbidity, disability, healthcare utilization).

Main Results:

• BA panels using all measures or age-optimized measures most closely approximated CA (r=0.74, r=0.77).
• However, a panel of physical performance measures (CA r=0.59) showed comparable or stronger associations with health-related outcomes.
• All BA estimates were significantly associated with most health outcomes, with minor sex variations.

Conclusions:

• While approximating CA is a consideration for BA estimation, its association with prospective health outcomes may be more critical.
• The choice of BA measures should be guided by specific research objectives.
• Physical performance measures may offer robust predictions of health trajectories in aging.