The Impact of Hormone-Related Factors on Wedge Deformity in Middle-Aged Populations

Vertebral deformity, particularly among adult over 50, poses significant public health challenges. This study explores the role of hormones and their metabolites in vertebral wedge deformities, using data from 1,317 individuals aged 40 and older from the NHANES 2013-2014 dataset. Key features included testosterone, estradiol, insulin, and related metabolites. Advanced machine learning models, notably XGBoost, were employed to predict deformity risk. A comprehensive cross-analysis of multiple machine learning interpreters identified testosterone, RBC folate, and insulin as significant predictors, with elevated levels indicating a higher deformity risk. These findings reinforce the link between hormone levels and vertebral deformities, suggesting that regular biomarker monitoring could aid early detection and intervention. The study offers new insights for improving clinical outcomes and guiding future research in vertebral health.