Predictors of Major Incident Fracture During Adulthood: Evidence from the Geelong Osteoporosis Study

As people age, preventing fractures becomes crucial for preserving independence and overall quality of life. This study aimed to identify factors associated with reduced hazard of major incident fractures during adulthood. Serial data from 1394 females and 1404 males (ages 20-96y) participating in the Geelong Osteoporosis Study were analysed. Fractures (hip, spine, proximal humerus, distal radius, pelvis) were identified through radiological reports. Sociodemographic and lifestyle factors, Anthropometry, dual-energy X-ray absorptiometry, comorbidities, medication use and biochemical markers were collected across the visits. Predictors of major incident fractures were examined using time-dependent Cox proportional hazards models with age as the primary time scale. Participants were followed from baseline until first fracture, death, loss to follow-up or 31/12/2020. Stepwise forward selection identified relevant predictors, and missing data were imputed using predictive mean matching. Over a median follow-up of 16.3y (IQR:10.5-23.5), fractures occurred in 430 participants (15.4%). Males had a lower fracture hazard than females (HR 0.71, 95% CI 0.52-0.97). Factors associated with a reduced hazard of fracture included never-smoking (0.70, 0.50-0.97), past smokers (0.67, 0.46-0.99), higher hip and spine T-scores (0.82, 0.72-0.93; 0.82, 0.76-0.89), lower plasma glucose (0.60, 0.42-0.87), lower HDL cholesterol (0.69, 0.52-0.90), absence of falls (0.74, 0.59-0.91), no muscle weakness (0.48, 0.23-0.98) and non-use of oral glucocorticoids (0.48, 0.31-0.75). Preserving bone mass, undertaking falls prevention strategies, avoiding smoking, monitoring oral glucocorticoid use and optimising metabolic health could be possible strategies for reducing fracture risk during ageing. The independent contribution of HDL might provide new perspectives into fracture risk assessment.