The association between physical activity in different domains of life and risk of osteoporotic fractures

A large body of epidemiological evidence suggests an inverse relationship between physical activity and risk of fractures. However, it is unclear how this association varies according to the domain of life in which the activity is undertaken. In this analysis of the European Prospective Investigation of Cancer-Norfolk study, we assessed total and domain-specific physical activity using a validated questionnaire (EPAQ2) in 14,903 participants (6514 men, mean age 62 year) who also underwent quantitative ultrasound of the heel. After a median follow-up of 7.5 years, there were 504 fractures of which 164 were hip fractures. In multivariable linear regression analysis, broadband ultrasound attenuation (BUA) was positively associated with total and leisure-time activities while showing no association with transportation and work activities. Home activities were associated with a lower BUA among younger participants. In multivariable Cox proportional-hazards models, moderate activities at home and in leisure time were associated with lower hip fracture risk among women (hazard ratios [HR] 0.51 and 0.55, p value 0.02 and 0.03, respectively). Among men, leisure-time activities were associated with lower risk of hip fracture (HR = 0.58; p for trend < 0.001) whereas activities at home were associated with higher risk of any fracture (HR = 1.25; p for trend = 0.008). Walking for leisure or transport was associated with lower risk of fracture in both men and women. Multivariable fractional polynomial modelling showed a U-shaped association between home activities and fracture risk especially among women. This study suggests that different domains of physical activity may relate differently to fracture risk and these relationships may vary by sex.