A simple nomogram prediction model for childbearing intention among female cancer survivors

Aim The research aims to identify the characteristics of patients with the childbearing intention after tumor diagnosis and construct a clinical prediction nomogram. Methods Data were obtained from the National Health and Nutrition Examination Survey (NHANES). Women diagnosed with cancer with ages ≤50 and with complete information about delivery were included. Univariate and multivariate logistic regression models were developed to identify associated factors. Predictive accuracy of the clinical prediction model was evaluated by the area under the curve (AUC). Results 224 and 749 participants with and without live birth after cancer diagnosis were included, respectively. In the univariate logistic regression analysis, age at tumor diagnosis, live birth before cancer diagnosis, comorbidity before cancer, family poverty income ratio, and tumor type were associated with the childbearing intention after cancer ( p <0.05). In the multivariate logistic regression analysis, age at tumor diagnosis and live birth before cancer diagnosis were correlated with fertility intention (OR 0.811, 95%CI 0.780 to 0.840; 0.155, 95%CI 0.080–0.294, respectively), others were without statistical significance probably due to the false association or indirect association ( p >0.05). A nomogram model including age at cancer diagnosis and live birth before cancer diagnosis was constructed, the AUC of which was 0.9312. Conclusions The reproductive intentions of young female cancer patients were associated with age at tumor diagnosis and live birth before cancer diagnosis. A nomogram model, including these two items with AUC 0.9312, could promptly identify patients to have children, which assists in formulating individualized treatment plans for tumors and enables timely fertility preservation.