The obesity paradox in osteoporosis risk among older adults is mostly driven by women: a population-based prospective study

Objectives The obesity paradox is widespread among older adults at risk for various diseases, including the unclear association between obesity and osteoporosis (OP) risk. This study aimed to investigate the association between obesity and OP risk in an older population. Methods A cross-sectional and prospective study was conducted based on the UK Biobank cohort. The association of body mass index (BMI), waist circumference (WC), and fat percentage with BMD was examined using Spearman correlation analysis with baseline BMD data. Cox proportional hazards regression analysis was used to investigate the association between obesity and OP risk. Restricted cubic spline (RCS) were used to assess the nonlinear associations of BMI, WC, and fat percentage with OP. Results Baseline cross-sectional analyses revealed a positive association between BMI, WC, and fat percentage with BMD in women, while no significant correlation was found in men. Cox analyses showed that obesity as defined by BMI, WC, and fat percentage was associated with a 33%, 23%, and 31% reduction in the risk of OP in women but not in men, respectively. Conjoint analysis showed that lower BMI was associated with increased risk of OP in older adults, whereas the lowest risk was observed in women with higher BMI and higher body fat. RCS revealed an inverse J-shaped nonlinear association between obesity metrics and OP risk in women. Conclusion Lower BMI is an independent risk factor for OP in older adults, and the obesity paradox for OP risk exists only in women.