Liao · Journal of the American Medical Directors Association 2026 · prospective cohort study · n=1194

Risk of All-Cause Mortality in Different Muscle Health States Among Community-Dwelling Older Adults.

Cited 0 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective cohort study evaluating mortality risk across muscle health categories with registry linkage

PubMed 41734884 · doi:10.1016/j.jamda.2026.106140 · record verified 2026-08-31

What was done

A prospective cohort study evaluated 1,194 community-dwelling adults aged 65 years and older undergoing routine health checkups in northern Taiwan between 2015 and 2021 (median follow-up 54.8 months). Muscle mass, handgrip strength, and walking speed were measured to classify participants into four categories: robust, dynapenia, presarcopenia, and sarcopenia. All-cause mortality was determined via linkage to the national death registry through December 2021. Sex-stratified Cox proportional hazards models with sequential covariate adjustment were used to assess mortality risk across phenotypes.

What was found

Of the 1,194 participants (56.78% women, n=678), 27.9% had dynapenia and 17.7% had sarcopenia. Over follow-up, 52 participants (4.35%) died. Dynapenia was associated with the highest comorbidity burden and adverse metabolic profile. Higher gait speed and handgrip strength inversely correlated with mortality in both sexes, whereas low muscle mass was predictive only in women. In multivariable models, sarcopenia was associated with higher mortality risk in women (adjusted HR 4.89, 95% CI 1.08–22.21), whereas dynapenia was associated with increased mortality in men (adjusted HR 3.57, 95% CI 1.10–11.63).

Why it matters

The findings indicate that muscle health phenotypes confer distinct, sex-specific prognostic risks in older age. Screening and interventions may need to prioritize muscle strength loss in older men and combined mass/functional loss in older women.

Limits

The overall event rate was low, with only 52 deaths across the entire cohort, producing wide confidence intervals for the subgroup hazard ratios. The cohort was restricted to individuals attending routine health checkups in northern Taiwan, which may introduce healthy-user selection bias and limit generalizability to other populations. Cause-specific mortality was not detailed in the abstract.

Cited by