Association Between A Body Shape Index and Bone Mineral Density in Middle-aged and Elderly Adults: A Retrospective Analysis of NHANES 2005-2018

Zhi Zhuang Wang ¹ Guo Liang Ma ¹ Bo Xu ¹ Xin Chen ¹ Bo Wen Yang ¹ Xiao Kuan Qin ¹ Wei Li Duan ^2,3 Min Shan Feng ¹ He Yin ^1* Kai Sun ^1* Li Guo Zhu ^1,4*

• ¹ Department of Spine, Wangjing Hospital, China Academy of Chinese Medical Sciences, Beijing, China
• ² Nanyang Hospital, Wangjing Hospital, Chinese Academy of Traditional Chinese Medicine (Dushan Hospital District), Henan, China
• ³ Nanyang Key Laboratory of Orthopedic Biomechanics of Traditional Chinese Medicine, Henan, China
• ⁴ Beijing Key Laboratory of Bone Setting Technology of Traditional Chinese Medicine, Beijing, China

Introduction: Despite accumulating evidence on central obesity and osteoporosis, the role of a body shape index (ABSI), a nonlinear index quantifying body shape via body mass index (BMI), waist circumference (WC), and height, remains controversial and underexplored. Although recent meta-analyses suggest central obesity may modulate fracture risk bidirectionally, no research has comprehensively compared ABSI with traditional adiposity metrics, such as BMI, WC, and waist-to-height ratio (WHtR), to predict site-specific changes in bone mineral density (BMD) across anatomical regions. Methods: This study utilized National Health and Nutrition Examination Survey (NHANES) data from 2005 to 2018, involving 12,421 participants. ABSI was computed using the formula: ABSI = WC / (BMI²/³ × Height¹/²). BMD was assessed at four sites—the total femur (TF), femoral neck (FN), trochanter (TR), and intertrochanter (IN) regions—via dual-energy X-ray absorptiometry (DXA). The association between ABSI and BMD was analyzed via multiple regression models and a generalized additive model (GAM). To compare ABSI's predictive efficacy with conventional adiposity indices, regression analyses juxtaposed ABSI against BMI, WC, and WHtR in assessing correlations with site-specific BMD.Results: After full covariate adjustment, a significant negative association was observed between ABSI and BMD in four femoral regions (P < 0.01). Smoothed curve fitting revealed a significant nonlinear relationship and threshold effect between ABSI and BMD among middle-aged and older individuals. Additionally, an inverted J-shaped curve was observed between ABSI and BMD in all four femoral regions. Meanwhile, ABSI showed significant negative associations with BMD across all femoral sites (β = -0.27 to -0.31, p-trend < 0.000001), whereas BMI, WC, and WHtR exhibited positive correlations (WHtR showing the strongest effect: β = 0.41–0.69). This highlights ABSI’s ability to detect central adiposity-related bone loss obscured by conventional obesity metrics.Conclusion: ABSI’s robust inverse associations with femoral BMD (β = -0.27 to -0.31), persisting across nonlinear threshold analyses, establish it as a novel biomarker of central adiposity-related skeletal fragility. Unlike conventional indices reflecting mechanical loading benefits (BMI β = 0.008–0.012; WC β = 0.003–0.005; WHtR β = 0.41–0.69), ABSI specifically captures visceral fat-driven metabolic disorder—a critical pathway for osteoporosis risk stratification in normal-weight and obese populations.