Trajectories of grip strength decline and risk of new-onset cardiovascular disease: evidence from the HRS and ELSA cohorts.

Year of Publication
2026
Author
Journal
Frontiers in public health
Volume
14
Number of Pages
1836439
ISSN Number
2296-2565
Abstract

BACKGROUND: Handgrip strength is an important indicator of overall health, yet most studies rely on single measurements. Evidence on its long-term trajectories and new-onset cardiovascular disease (CVD) risk remains limited. This study examined associations of baseline handgrip strength and its trajectories with new-onset CVD in two aging cohorts.

METHODS: Data were from the Health and Retirement Study (HRS) in the United States and the English Longitudinal Study of Ageing (ELSA). Participants aged ≥50 years without baseline CVD were included. Handgrip strength across three waves was used to construct trajectories via group-based trajectory modeling. Cox regression assessed associations with new-onset CVD. Fine-Gray models, discrete-time logistic regression, restricted cubic splines, and mediation analysis (depression) were conducted.

RESULTS: A total of 4,984 (HRS) and 4,982 (ELSA) participants were included. Over median follow-ups of 10 and 12 years, 1,438 and 1,184 CVD events occurred, respectively. Higher baseline handgrip strength was associated with lower CVD risk (HRS: HR = 0.79, 95% CI: 0.73-0.86; ELSA: HR = 0.83, 95% CI: 0.76-0.91). Two trajectories were identified: accelerated decline and steady decline. Compared with steady decline, accelerated decline was linked to higher CVD risk (HRS: HR = 1.44, 95% CI: 1.22-1.70; ELSA: HR = 1.31, 95% CI: 1.08-1.58). Results were consistent across sensitivity analyses. A linear inverse association was observed, and depression partially mediated this relationship.

CONCLUSION: Lower baseline handgrip strength and accelerated decline are associated with increased CVD risk. Longitudinal monitoring may aid early risk identification and prevention.

DOI
10.3389/fpubh.2026.1836439
PMID
42433396
PMCID
PMC13352478
Download citation