Geriatrics

Five Years of HIIT Halves Sarcopenia Risk in Robust Septuagenarians

In the Generation 100 trial, 1,567 community-dwelling adults aged 70–75 were randomized to HIIT, MICT, or guideline-based usual-activity control. HIIT attenuated grip-strength decline at 1 and 3 years and modestly improved gait speed at 3 years versus control. Clinically defined sarcopenia rose in all groups but remained lowest with HIIT (56% vs 63% in controls at 5 years). Odds of incident clinically defined sarcopenia were roughly halved with HIIT versus control at 3 and 5 years, while MICT showed no benefit. 1

Physiotherapy-Led Integrated Care Improves Intrinsic Capacity and QoL

This RCT enrolled 71 community-dwelling older adults to a 12-week physiotherapy-led, person-centred integrated care program (PTICOPE) versus health-education control. PTICOPE combined centre-based physiotherapy with home multidomain activities targeting locomotion, vitality, cognition, and sensory and psychological domains. Compared with control, PTICOPE produced clinically meaningful gains in locomotion and smaller but significant improvements in vitality and cognition. Psychological, visual, and auditory domains showed no clinically important changes, but quality of life improved across all WHOQOL-BREF domains. 2

Two-Step ED Delirium–Cognition Screening Flags High-Risk Older Inpatients

In 830 adults aged ≥65 admitted from a Brazilian ED, clinicians applied the brief CAM and 10-point Cognitive Screener at baseline. Twenty-one percent had delirium and 28% had cognitive impairment without delirium; over half of the latter lacked documented dementia or memory complaints. Cognitive impairment without delirium predicted 90-day ADL decline and mortality, with risks similar to delirium after adjustment. A staged approach—bCAM first, then 10-CS if negative—outperformed delirium-only or cognition-only strategies for risk discrimination. 3

ELDER-ICU Mortality Model Needs Local Updating Before Use

This multicenter study externally validated the ELDER-ICU machine-learning mortality model for patients aged ≥65 across 12 ICUs in four countries. Discrimination remained strong in US and Austrian cohorts but dropped in South Korea and China, with AUROC as low as 0.698. Incremental training improved performance in most sites, while full retraining substantially boosted AUROC in the Asian centers. Isotonic regression and Platt scaling enhanced calibration, emphasizing that prognostic tools require local validation and recalibration before clinical deployment. 4