The study analyzed data from 409 women with infertility, defining high-risk infertility as lasting more than 3 years or having at least 2 documented causes. Demographic, reproductive, menstrual, and traditional Chinese medicine (TCM) variables were collected. They developed three predictive models: TCM model (model 1), modern medicine model (model 2), and integrated model (model 3) using LASSO regression and multivariable logistic regression. Among the participants, 163 women (39.9%) were classified as high-risk, and this group was older and varied in reproductive history. Model 3 had an initial AUC of 0.790 versus 0.771 for Model 2, but after correcting for optimism, Model 2 was more stable in discrimination, calibration, and precision. The TCM model (1) had limited ability (AUC 0.606). After internal validation, the modern medicine model showed the best prediction performance, while the integrated model serves more to explore the value of TCM variables.