Construction and Validation of Predictive Model for Decreased Bone Density in Premature Ovarian Insufficiency

Abstract

Premature ovarian insufficiency (POI) is an endocrine disorder that affects women of childbearing age. It is characterized by premature ovarian failure, leading to decreased estrogen levels and increased gonadotropin levels. POI not only impacts reproductive health but is also associated with various long-term health issues, particularly osteoporosis due to decreased estrogen protection, making individuals more susceptible to developing osteoporosis. This study utilized a retrospective cohort study design, including patients diagnosed with POI at the First Affiliated Hospital of Henan Polytechnic University during a certain period. Through detailed medical history collection, bone density measurements, and assessment of other relevant factors, we analyzed potential risk factors associated with decreased bone density caused by POI. Multivariable logistic regression analysis was used to identify independent risk factors, upon which a predictive model was constructed. The results showed that a total of 198 POI patients were included in the study, among whom 54 (34.8%) had decreased bone density. Multivariable analysis revealed that age, parity, family history of osteoporosis, BMI, and regular physical exercise were significantly associated with decreased bone density. Based on these factors, we developed a predictive model and evaluated its predictive performance using the receiver operating characteristic (ROC) curve, with an area under the ROC curve of 0.953 (95% CI: 0.924-0.983), indicating good predictive accuracy.