Screening for high hip fracture risk does not impact on falls risk: a post hoc analysis from the SCOOP study

Summary: A reduction in hip fracture incidence following population screening might reflect the effectiveness of anti-osteoporosis therapy, behaviour change to reduce falls, or both. This post hoc analysis demonstrates that identifying high hip fracture risk by FRAX was not associated with any alteration in falls risk. Introduction: To investigate whether effectiveness of an osteoporosis screening programme to reduce hip fractures was mediated by modification of falls risk in the screening arm. Methods: The SCOOP study recruited 12,483 women aged 70–85 years, individually randomised to a control (n = 6250) or screening (n = 6233) arm; in the latter, osteoporosis treatment was recommended to women at high risk of hip fracture, while the control arm received usual care. Falls were captured by self-reported questionnaire. We determined the influence of baseline risk factors on future falls, and then examined for differences in falls risk between the randomisation groups, particularly in those at high fracture risk. Results: Women sustaining one or more falls were slightly older at baseline than those remaining falls free during follow-up (mean difference 0.70 years, 95%CI 0.55–0.85, p < 0.001). A higher FRAX 10-year probability of hip fracture was associated with increased likelihood of falling, with fall risk increasing by 1–2% for every 1% increase in hip fracture probability. However, falls risk factors were well balanced between the study arms and, importantly, there was no evidence of a difference in falls occurrence. In particular, there was no evidence of interaction (p = 0.18) between baseline FRAX hip fracture probabilities and falls risk in the two arms, consistent with no impact of screening on falls in women informed to be at high risk of hip fracture. Conclusion: Effectiveness of screening for high FRAX hip fracture probability to reduce hip fracture risk was not mediated by a reduction in falls.