Background: Impaired intrinsic capacity (IC), which affects approximately 90% of older adults, is associated with a significantly heightened risk of frailty and cognitive decline. Existing evidence suggests that multidomain interventions have the potential to enhance cognitive performance and yield positive effects on physical frailty.

Objective: To examine roles of baseline IC and its subdomains on the efficacy of multidomain interventions in promoting healthy aging in older adults.

Design: a cluster-randomized controlled trial.

Setting and participants: 1,054 community-dwelling older adults from 40 community-based clusters across Taiwan.

Intervention: A 12-month pragmatic multidomain intervention of exercise, cognitive training, nutritional counseling and chronic condition management.

Measurements: Baseline IC was measured by 5 subdomains, including cognition (Montreal Cognitive Assessment, MoCA), sensory (visual and hearing impairment), vitality (handgrip strength or Mini-Nutritional Assessment-short form), psychological well-being (Geriatric Depression Scale-5), and locomotion (6m gait speed). Outcomes of interest were cognitive performance (MoCA scores) and physical frailty (CHS frailty score) over a follow-up period of 6 and 12 months.

Results: Of all participants (mean age:75.1±6.4 years, 68.6% female), about 90% participants had IC impairment at baseline (2.0±1.2 subdomains). After covariate adjustment using a generalized linear mixed model (GLMM), the multidomain intervention significantly prevented cognitive declines and physical frailty, particularly in those with IC impairment ≥ 3 subdomains (MoCA: coefficient: 1.909, 95% CI: 0.736 ~ 3.083; CHS frailty scores: coefficient = -0.405, 95% CI: -0.715 ~ -0.095). To assess the associations between baseline poor capacity in each IC subdomain and MoCA/CHS frailty scores over follow-up, a 3-way interaction terms (time*intervention*each poorer IC subdomains) were added to GLMM models. Significant improvements in MoCA scores were shown for participants with poorer baseline cognition (coefficient= 1.138, 95% CI: 0.080 ~ 2.195) and vitality domains (coefficient= 1.651, 95% CI: 0.541 ~ 2.760). The poor vitality domain also had a significant modulating effect on the reduction of CHS frailty score after the 6- and 12-month intervention period (6 months: coefficient= -0.311, 95% CI: -0.554 ~ -0.068; 12 months: coefficient= -0.257, 95% CI: -0.513 ~ -0.001). CONCLUSION AND

Implications: A multidomain intervention in community-dwelling older adults improves cognitive decline and physical frailty, with its effectiveness influenced by baseline IC, highlighting the importance of personalized strategies for healthy aging.