Aging & Longevity

Dynamic frailty exposure and cognitive aging trajectories: A multicohort study.

TL;DR

Baseline frailty status, worsening from robust status, and higher mean frailty burden were associated with less-favorable cognitive aging across complementary rate-based and trajectory analyses, suggesting repeated frailty assessments may provide information for characterizing cognitive aging.

Key Findings

Baseline pre-frailty was associated with more negative annual global cognitive change compared to robust status.

  • Pooled β = -0.023 SD/year for pre-frailty versus robust status
  • 95% CI: -0.030 to -0.016
  • Estimates were derived from cohort-specific linear mixed-effects models pooled using random-effects meta-analysis
  • Total sample across three cohorts: n = 19,586

Baseline frailty was associated with more negative annual global cognitive change compared to robust status.

  • Pooled β = -0.023 SD/year for frailty versus robust status
  • 95% CI: -0.031 to -0.014
  • Association magnitude was similar to that observed for pre-frailty
  • Findings were consistent across domain-specific cognitive outcomes

Worsening frailty status from robust to pre-frail or frail over two waves was associated with more negative annual cognitive change.

  • Pooled β = -0.025 SD/year for worsening from robust to pre-frail or frail
  • 95% CI: -0.035 to -0.014
  • This finding was based on two-wave frailty change analysis
  • The association was identified using exposure-by-time interactions in linear mixed-effects models

Higher mean frailty burden across measurement waves was associated with more negative annual cognitive change.

  • Mean frailty burden was calculated as the average frailty level across available assessment waves
  • Association was observed for both global and domain-specific cognitive outcomes
  • Findings were pooled across the three nationally representative cohorts
  • This complemented the baseline and two-wave frailty change analyses

Transitions from initial pre-frailty or frailty showed no clear pooled associations with annual cognitive change.

  • Changes from initial pre-frailty or frailty did not yield statistically clear pooled estimates
  • This contrasts with findings for worsening from robust status
  • The asymmetry suggests frailty worsening from a robust baseline may be a particularly sensitive marker
  • Results were assessed using two-wave frailty change analysis

In secondary group-based trajectory modeling (GBTM), baseline frailty status, higher mean frailty burden, and frailty worsening were associated with membership in less-favorable cognitive trajectory groups.

  • GBTM identified distinct cognitive trajectory groups within the study sample
  • Logistic regression was used to assess associations between frailty exposures and trajectory group membership
  • Frailty improvement was directionally associated with more-favorable trajectory groups
  • GBTM findings were consistent with and complementary to the primary rate-based analyses

The study analyzed data from three nationally representative cohorts spanning China, England, and the United States.

  • Cohorts included the China Health and Retirement Longitudinal Study (CHARLS), English Longitudinal Study of Ageing (ELSA), and Health and Retirement Study (HRS)
  • Total pooled sample size was n = 19,586
  • Cohort-specific estimates were pooled using random-effects meta-analysis
  • Primary analyses used linear mixed-effects models with exposure-by-time interactions to estimate differences in annual global and domain-specific cognitive change

What This Means

This research suggests that physical frailty — a state of reduced strength, energy, and resilience that often develops with aging — is linked to faster cognitive decline over time. Using data from nearly 20,000 older adults across three large national studies in China, England, and the United States, the researchers found that people who were pre-frail or frail at the study's start experienced roughly 0.023 standard deviations more cognitive decline per year compared to those who were robust. People who became more frail over the course of the study also showed faster cognitive decline, while those who maintained a higher average frailty burden throughout the study period showed similar patterns of worsening cognition. Importantly, the direction of frailty change mattered: worsening from a robust starting point was clearly associated with worse cognitive trajectories, but transitions away from pre-frailty or frailty did not show consistent associations with cognitive change in either direction — though frailty improvement showed a directional trend toward more favorable cognitive trajectories. These findings held up across two different analytical approaches: one that measured the rate of annual cognitive change, and another that grouped participants into distinct long-term cognitive aging trajectory patterns. This research suggests that monitoring frailty status repeatedly over time — not just at a single point — could help identify older adults who are at greater risk for cognitive decline. Because frailty is potentially modifiable through interventions like exercise and nutrition, these findings point to frailty management as a possible avenue for supporting healthy cognitive aging, though the study's observational design means causality cannot be established.

Have a question about this study?

Citation

Wang X, Hao H, Jiang T, Yao X, Duan Y, Xia J, et al.. (2026). Dynamic frailty exposure and cognitive aging trajectories: A multicohort study.. Alzheimer's & dementia : the journal of the Alzheimer's Association. https://doi.org/10.1002/alz.71821