Intrinsic capacity (IC) is a recent key concept proposed by the World Health Organization (WHO) based on aspects of functional ability (both physical and mental) rather than the presence or absence of diseases, with a potential to predict several health outcomes.

To explore associations between IC and cognitive function (prospectively), and between IC and plasma p-tau181 (cross-sectionally and prospectively) among community-dwelling older adults.

Observational study with 491 subjects ≥70 years (67.4 % female, mean 75.3 years, SD=4.4), participants from the Multidomain Alzheimer Preventive Trial (MAPT). IC domains (locomotion, cognition, psychological, vitality) were combined into a 0–100 score. Alternative classification was based on the number of domains’ abnormalities. Plasma p-tau181 was measured at baseline and 36 months of follow-up. A composite cognitive score (CCS) based on four tests was determined at baseline, 6, 12, 24, 36, 48 and 60 months.

Inverse cross-sectional associations were observed between baseline IC score and p-tau181 (unadjusted model: β=-0.08, 95 %CI -0.13 to -0.03; p = 0.0025). A significant mean difference in p-tau181 3-year changes was observed between participants with low and normal IC (based on IC score) (adjusted model: 1.71, 95 %CI 0.01 to 3.40; p = 0.0483). Prospective 5-year associations between IC and CCS were only observed in unadjusted analysis according to the alternative IC classification (-0.21, 95 %CI -0.38 to -0.04; p = 0.0156).

IC was associated with plasma p-tau181 and cognitive function, but findings varied according to the method of IC classification. Further research may help settle the role of IC as a predictor of neurodegenerative diseases such as AD. In this regard, multidomain interventions have potential to protect IC over the aging process and prevent cognitive impairment, and should also be encouraged.