Explicit health/safety-related non-completion of physical performance tests and subsequent disability and mortality: Evidence from CHARLS with functional outcome replication in ELSA.

OBJECTIVE

Physical performance tests are widely used to quantify functional capacity in older adults, yet participants who are unable to complete these tests because of health or safety concerns are often excluded from analyses or treated as having ordinary missing data. This study evaluated whether non-completion explicitly related to health/safety concerns could serve as prognostically informative process-level information.

METHODS

The study included 6267 participants from the China Health and Retirement Longitudinal Study (CHARLS), with functional outcome replication in 6175 participants from the English Longitudinal Study of Ageing (ELSA). In the CHARLS, the primary exposure was defined as non-completion for health or safety reasons explicitly recorded in the balance, gait-speed, or chair-stand domain. In the ELSA, replication analyses used the balance, chair-stand, and grip-strength domains for which explicit health/safety-related reasons were available. Outcomes included incident disability in activities of daily living, mortality, and secondary functional outcomes in the ELSA. Associations were evaluated using sequentially adjusted models and sensitivity analyses.

RESULTS

In the Chinese cohort, 721 participants (11.5%) had explicitly health/safety-related non-completion; 840 participants were identified in the English cohort. In health-adjusted models, explicitly health/safety-related non-completion was associated with incident disability in activities of daily living, and the association remained after further adjustment for baseline functional status. In mortality analyses, this status was associated with higher odds of death. Functional outcome findings in the English cohort were directionally consistent.

CONCLUSION

Explicitly health/safety-related non-completion of physical performance testing should not automatically be treated as ordinary 'missingness'. It may provide complementary process-level information alongside conventional physical performance scores and may help identify older adults at increased risk of subsequent functional decline and mortality.