Abstract
BACKGROUND: Concerns have been raised around whether calcium supplements increase dementia risk. This post-hoc analysis of a five-year double-blind, placebo-controlled randomised trial of calcium supplements for primary fracture prevention evaluated the long-term risk for dementia in older women, randomised to either calcium supplements or placebo. METHODS: 1460 community-dwelling dementia-free Australian women (≥70 years) were randomised to 1200 mg/day calcium carbonate (n = 730) or placebo (n = 730) for five years and were observed for an additional 9.5 years afterwards. Over 14.5 years, all-cause dementia events (comprising dementia-related hospitalisations and/or deaths) were identified from linked health records. The influence of calcium supplementation on dementia outcomes were examined using Kaplan-Meier survival curves and Cox regression under intention-to-treat (ITT) and per-protocol (PP, ≥80% tablet compliance, n = 830; 50.6% calcium supplements) criteria. FINDINGS: Mean baseline age was 75.1 ± 2.7 years. Dementia events were recorded in 269 women (18.4%), comprising 243 hospitalisations (16.6%) or 114 deaths (7.8%). No differences in the cumulative dementia-free survival rates were observed between groups in ITT and PP analyses. Compared to placebo, calcium supplements did not increase risk of dementia-related events (unadjusted ITT hazard ratio [HR] 0.90, 95% confidence interval (CI) 0.71-1.15), hospitalisations (HR 0.89, 95% CI 0.69-1.15) or deaths (HR 0.78, 95% CI 0.54-1.13). Similar results were observed in PP analyses. INTERPRETATION: Calcium supplementation for five years did not increase the risk of all-cause dementia events over 14.5 years in community-dwelling older women. Findings do not support concerns that calcium supplementation increases long-term risk of dementia. FUNDING: National Health and Medical Research Council of Australia.