Chronological Age and Functional Reserve, Not Blood Pressure, Predict Loss of Independence in Ogimi, Okinawa

Background Optimal blood pressure targets in the oldest-old remain uncertain. Ogimi Village, Okinawa, an internationally recognised exceptional longevity community, offers a distinctive setting in which to examine determinants of functional decline. Methods We analysed annual municipal health examinations in Ogimi (2015–2024): 225 assessments from 154 community-dwelling residents aged ≥65 years. Outcomes were loss of independence (LOI, long-term care insurance database) and all-cause mortality. Handgrip strength, chair-stand time, body mass index, muscle mass ratio, and the blood-pressure-adjusted left pulse wave velocity coefficient (lPWVC) were summarised by principal component analysis; Cox models used cluster-robust variance and multiple imputation (M = 20). Results Mean age was 83.6 years; 33 LOI events and 37 deaths occurred. Antihypertensive use was 52.9%, similar below and above age 90 (51.8% vs 55.9%, p = 0.694). Age ≥90 years was the strongest predictor of LOI (hazard ratio 5.06–5.52) and mortality (3.07, 95% CI 1.54–6.11), whereas systolic blood pressure predicted neither. Handgrip strength and chair-stand time were not independently predictive, although the lowest functional reserve tertile showed poorer LOI-free survival (log-rank p = 0.036). Among those aged ≥85 years, lPWVC was positively but non-significantly associated with LOI (odds ratio 1.53, 95% CI 0.65–3.61), and adding functional reserve scores and lPWVC to age improved discrimination (AUC 0.604 to 0.722). Conclusions Chronological age dominated functional decline in this longevity population, whereas higher systolic blood pressure conferred no excess risk, consistent with reverse epidemiology in the oldest-old. These observational findings are hypothesis-generating and are not evidence against antihypertensive treatment.