Establishing and validating a nomogram to predict the risk of osteoporosis in patients with atopic dermatitis - a two-center study

To explore the risk factors for osteoporosis in patients with atopic dermatitis (AD) and develop a risk prediction model to visualize the risk. We retrospectively analyzed 499 AD patients admitted to Xuzhou Central Hospital from June 2016 to January 2023. Patients were categorized into osteoporosis, osteopenia, and healthy groups based on bone mineral density (BMD). Univariate and multivariate logistic regression analyses were performed to identify significant risk factors, and the data were randomly divided into a training set (75%) and a test set (25%). An additional 204 AD patients from the First Affiliated Hospital of Harbin Medical University (June 2020–January 2023) served as an external validation set. A nomogram was constructed based on multivariate analysis results, and its performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). In the training and test sets, 50 patients (7.9%) had osteoporosis, and 91 had osteopenia. Univariate analysis revealed significant differences in age, disease duration, autoimmune disease, allergic disease, AD severity, corticosteroid use, biologics, immunosuppressants, and 25-hydroxyvitamin D levels ( P < 0.05). Multivariate analysis identified age (OR = 1.042, 95% CI: 1.016–1.068, P = 0.001), disease duration (OR = 1.035, 95% CI: 1.003–1.069, P = 0.034), corticosteroid use (OR = 1.933, 95% CI: 1.382–2.712, P = 0.001), immunosuppressant use (OR = 1.970, 95% CI: 1.093–3.597, P = 0.025), and low 25-hydroxyvitamin D levels (OR = 0.838, 95% CI: 0.805–0.870, P = 0.001) as independent predictors of osteoporosis. The nomogram demonstrated good predictive accuracy, with AUC values of 0.761 (training set), 0.733 (testing set), and 0.686 (external validation set). Age, disease duration, corticosteroid use, immunosuppressant use, and 25-hydroxyvitamin D levels were significant predictors of osteoporosis in AD patients. The developed nomogram provides a reliable tool for individualized risk assessment.