Abstract / Summary
Abstract Atopic eruption of pregnancy (AEP) is a prevalent inflammatory disease in pregnant women and is frequently associated with considerable pruritus and impaired quality of life. There remains an unmet need for reliable tools to predict treatment effectiveness in order to inform and optimize clinical management. In this retrospective cohort study, clinical data from 653 patients diagnosed with AEP at Anhui Maternal and Child Health Hospital between 2009 and 2024 were analyzed. Multivariable regression analyses were performed to identify independent predictors of treatment effectiveness, and a predictive model was subsequently constructed based on these variables. The final model identified histopathological type, allergen screening status, and treatment modality as independent factors associated with treatment effectiveness. The model demonstrated moderate discriminatory capability, with a sensitivity of 81.5% and a specificity of 41.6%. The area under the receiver operating characteristic curve (AUC) was 0.639, reflecting limited predictive accuracy. This predictive model may provide adjunctive support for clinical decision-making in the management of AEP by integrating histopathological findings, allergen screening, and therapeutic approach. However, in light of its moderate predictive performance, the model should be interpreted with caution and applied as a supplementary aid rather than a definitive clinical decision-making tool.