The aim of this population-based, prospective cohort study was to investigate long-term associations between dietary calcium intake and fractures, non-fatal cardiovascular disease (CVD), and death from all causes. Participants were from the Melbourne Collaborative Cohort Study, which was established in 1990 to 1994. A total of 41,514 men and women (∼99% aged 40 to 69 years at baseline) were followed up for a mean (SD) of 12 (1.5) years. Primary outcome measures were time to death from all causes (n = 2855), CVD-related deaths (n = 557), cerebrovascular disease-related deaths (n = 139), incident non-fatal CVD (n = 1827), incident stroke events (n = 537), and incident fractures (n = 788). A total of 12,097 participants (aged ≥50 years) were eligible for fracture analysis and 34,468 for non-fatal CVD and mortality analyses. Mortality was ascertained by record linkage to registries. Fractures and CVD were ascertained from interview ∼13 years after baseline. Quartiles of baseline energy-adjusted calcium intake from food were estimated using a food-frequency questionnaire. Hazard ratios (HR) and odds ratios (OR) were calculated for quartiles of dietary calcium intake. Highest and lowest quartiles of energy-adjusted dietary calcium intakes represented unadjusted means (SD) of 1348 (316) mg/d and 473 (91) mg/d, respectively. Overall, there were 788 (10.3%) incident fractures, 1827 (9.0%) incident CVD, and 2855 people (8.6%) died. Comparing the highest with the lowest quartile of calcium intake, for all-cause mortality, the HR was 0.86 (95% confidence interval [CI] 0.76-0.98, p(trend) = 0.01); for non-fatal CVD and stroke, the OR was 0.84 (95% CI 0.70-0.99, p(trend) = 0.04) and 0.69 (95% CI 0.51-0.93, p(trend) = 0.02), respectively; and the OR for fracture was 0.70 (95% CI 0.54-0.92, p(trend) = 0.004). In summary, for older men and women, calcium intakes of up to 1348 (316) mg/d from food were associated with decreased risks for fracture, non-fatal CVD, stroke, and all-cause mortality.
Read full abstract