Background Studies examining physical activity and joint health are conflicting – with some showing a detrimental effect and others showing either no effect or a beneficial effect.
Objectives To explore the association between physical activity and knee or hip osteoarthritis leading to knee or hip joint replacement surgery.
Methods 536 community-dwelling adults aged 51–79 years were assessed at baseline, 2.7, 5 and 10 years. Physical activity was measured by pedometer (steps/day). The incidence of knee or hip replacement surgery was reported at each follow-up. Log-binomial regression was used to examine the association between steps/day with incident knee and hip replacement adjusted for age, sex, BMI, x-ray disease severity, and pain.
Results Over 10 years, 36 participants reported having a knee replacement and 27 reported having a hip replacement. For every 1000 steps/day increase at baseline, there was a 0.84 times (95% CI 0.72 to 0.99) lower risk of hip replacement (P=0.036). There was no significant association between steps/day and knee replacement (RR 1.00 95% CI 0.9999 to 1.00, P=0.778). Baseline BMI significantly predicted knee replacement (RR per SD increase in BMI 1.42 (95% CI 1.03 to 1.97, P=0.032)) but not hip replacement (RR 1.00 95% CI 0.91 to 1.10, P=0.984).
Conclusions Higher physical activity levels were protective against hip replacement while being overweight or obese increased the risk for knee replacement. This suggests different causal pathways for each site with regard to habitual activity and body composition.
Disclosure of Interest None declared