Past and recent physical activity and risk of hip fracture.

A population-based case-control study was conducted to examine the effects of past and recent physical activity on the risk of hip fracture in women. Cases included females aged 55-84 years with a first diagnosis of hip fracture in 1989 in metropolitan Toronto, Canada. Controls were a population-based random sample frequency matched by 5-year age groups. Data were collected on 381 cases and 1,138 controls by self-administered mailed questionnaires or telephone interviews. Past activity was calculated as a compilation of activity scores at ages 16, 30, and 50 years. Recent activity was defined as activity in the past year for controls and activity in the year before fracture for cases. Multiple logistic regression was used to control for age, previous fracture, obesity, smoking, osteoporosis, epilepsy, stroke or Parkinson's disease, daily intake of dietary calcium, and duration of use of supplemental calcium, fluoride, and estrogen. After recent activity was adjusted for, statistically significant effects were found for women who in the past had been active (odds ratio estimate (OR) = 0.66, 95% confidence interval (CI) 0.45-0.96) or very active (OR = 0.54, 95% CI 0.33-0.88). After past activity was adjusted for, a similar protective effect was found for women who were moderately active recently (OR = 0.61, 95% CI 0.41-0.90), but women who were very active recently were not protected from hip fracture (OR = 1.15, 95% CI 0.72-1.83). This study showed evidence of independent protective effects of past physical activity and of moderate levels of recent physical activity on the risk of hip fracture in postmenopausal women.