Geriatrics

Frailty nearly doubles mortality risk in community-dwelling older adults

Across 59 prospective cohorts including 185,355 community-dwelling older adults, non-robust status significantly increased all-cause mortality versus robust peers. Overall, non-robust individuals had an 83% higher mortality risk (HR 1.83, 95% CI 1.70–1.97). Physical, multidimensional, and phenotype-based frailty models all showed similarly elevated mortality, underscoring robust prognostic value regardless of instrument. Within phenotype models, frail adults had higher mortality than prefrail adults, demonstrating clear dose–response by frailty severity. 1

Cardiovascular multimorbidity configurations drive frailty transitions and death

In 14,511 adults aged ≥60 years, multimorbidity configurations shaped frailty transitions and mortality over median 3.8-year frailty and 5.6-year mortality follow-up. Multimorbidity by simple counts rose with frailty, but latent multimorbidity classes better discriminated risk profiles and trajectories. “Major CVD & Vascular” and “Heart & Vascular” groups had the steepest frailty increases and higher transition probabilities from pre-frailty or frailty to death. “Vascular” and “Metabolic” groups showed lower frailty levels and greater likelihood of frailty improvement. 2

“Weekend warrior” MVPA pattern protects against sarcopenia and falls

In this wrist-accelerometer cohort with median 7.8-year follow-up, older adults meeting ≥150 minutes/week MVPA had lower sarcopenia and fall risks. Compared with inactive peers, active weekend warriors had reduced probable sarcopenia, sarcopenia, falls, and recurrent falls (HRs approximately 0.74–0.82). Risk reductions in (probable) sarcopenia were at least as large for weekend warriors as for regularly active participants. Both activity patterns consistently lowered fall risk across different weekend warrior definitions. 3

Non-pharmacologic dementia caregiver interventions in LMICs improve psychosocial outcomes

This review of 32 RCTs from 13 LMICs found non-pharmacologic interventions significantly reduced caregiver burden, distress, and depression versus controls. Meta-analyses showed meaningful post-intervention improvements in burden, distress, and depressive symptoms, despite high between-study heterogeneity. In-person, remote, and hybrid delivery models all generally improved burden, with only two studies not showing benefit. Risk of bias and design variability limit certainty, but overall direction of effect remained favorable across settings. 4