Association of fractures with caffeine and alcohol in postmenopausal women: the Iowa Women's Health Study

Objective: To assess whether alcoholic and caffeinated beverages are associated with risk of fractures in women. Setting: Population-based sample surveyed by post. Subjects: A total of 34 703 postmenopausal Iowan women aged 55-69 years were surveyed. Design: A cohort of women reported alcoholic and caffeinated beverage intake and were followed for 6.5 years for fracture occurrence. Relative risks (RR) and 95% confidence intervals (CI) were computed using Cox proportional hazards regression. Covariates included age, tobacco use, physical activity, body mass index (BMI), waist to hip ratio (WHR), oestrogen use and calcium intake. Results: At least one fracture was reported by 4378 women (389 upper arm, 288 forearm, 1128 wrist, 275 hip, 416 vertebral and 2920 other fractures). The adjusted RR for highest versus lowest caffeine intake quintiles was 1.09 (95% CI 0.99-1.21) for combined fracture sites. Wrist fractures were associated positively (RR for extreme quintiles 1.37, 95% CI 1.11-1.69) and upper arm fractures were negatively associated (RR 0.67, 95% CI 0.48-0.94) with caffeine intake. The age-adjusted RR of total fractures for highest versus lowest frequency of beer usage was 1.55 (95% CI 1.25-1.92) and for liquor was 1.25 (95% CI 1.03-1.54). No other association was found between any specific fracture site and alcohol intake. Conclusions: We found a modest increase in fracture risk associated with highest caffeine intake, varying by site. Alcohol intake was low, but it also showed a weak positive association with fracture risk.