Physical resilience—the ability to maintain or recover function after a health challenge—is an important concept in geriatric medicine, but there's no consensus on how to measure it. This study addresses a critical gap: when researchers talk about 'resilience' in older adults, are they measuring the same thing? The authors tested three competing frameworks: the 'phenotype' approach (measuring functional change after illness), the 'expected recovery' approach (comparing actual recovery to what's predicted), and the 'adapted ageing' approach (assessing baseline function relative to age-matched expectations).
They enrolled 330 older adults receiving chemotherapy (TENT cohort) and 2,111 older adults presenting to emergency departments with acute illness (APOP cohort), median ages 75 and 78 respectively. Physical function was assessed using standardized Activities of Daily Living (ADL) scales at baseline and 12 months. The key outcomes were mortality and a composite endpoint combining functional decline, quality of life decline, or death.
The results were sobering: agreement between the three approaches was poor to moderate (Cohen's kappa -0.10 to 0.64), meaning they identified different patients as 'resilient.' More importantly, all three approaches showed only modest ability to predict 12-month mortality (AUC 0.55–0.69; where 0.5 = coin flip and 1.0 = perfect prediction) and composite outcomes (AUC 0.52–0.66). Simple baseline measures—like functional score at follow-up or age alone—predicted outcomes just as well.
Several limitations deserve emphasis: this is observational data from specific cohorts (chemotherapy and acute illness patients), not a randomized trial, so causality cannot be inferred. The physical resilience measures were applied retrospectively rather than prospectively designed. Notably, the paper acknowledges a fundamental problem: all three approaches rely on widely-spaced assessments (baseline and 12 months), potentially missing the dynamic, rapid fluctuations that characterize resilience. The authors suggest that higher-frequency longitudinal monitoring might better capture resilience as a real-time phenomenon.
For longevity research, this paper delivers an important cautionary message: published 'resilience' metrics in older adult studies may not measure a unified construct, and current approaches lack predictive power for clinical outcomes. This suggests that resilience frameworks need theoretical refinement and validation before they're useful in clinical practice or as endpoints in longevity trials. The finding also highlights a broader challenge in gerontology: many intuitively appealing concepts (frailty, resilience, biological age) remain poorly operationalized and may not capture the complexity of aging biology.
The methodological contribution is valuable—showing what doesn't work—but the practical impact is limited since the paper offers solutions rather than a validated measurement framework ready for deployment.
0 Comments
Log in to join the discussion.