ABSTRACT
Objective The Rothman-Keller model was used to establish a predictive model for the risk of birth defects on the basis of evidence-based medicine, which so as to provide the basis for pertinent interventions in China. Methods First, the odds ratio (OR) value of risk factors for birth defects was obtained by evaluating the literature of meta-analysis, and the risk score table of the Rothman-Keller model was constructed. Then the simulation data was used to build the model, the risk boundary value of risk prediction, and finally the actual data to was used for verification. Results The main risk factors for 20 birth defects were collected through 17 articles. In the actual data of Shanxi Province, the actual incidence rate of high-risk populations screened by Rothman-Keller model was 10.9%, and it was statistically different from other groups ( 2 =147.58，P<0.001). In addition, the rothman-keller model identified all patients with a family history of birth defects as high-risk. Conclusions Through the meta-analysis literature on birth defects in China, the study find the main risk factors and construct a risk prediction model. It can be used to predict the risk of birth defects and help screen high-risk groups. At the same time, it provides ideas for predicting the risk of other diseases.