To explore the clinical epidemiological characteristics of hypertensive disorders of pregnancy (HDP) of pregnant women, and to construct the related predictive model for HDP, and to study the predictive value of the model for HDP. Methods: 102 pregnant women with HDP who had admitted in hospital during the second trimester of pregnancy from February 2019 to May 2023 (in study group) were selected in this study retrospectively, and 100 healthy pregnant women were selected in control group. The clinical data of the women in the two groups were collected. The clinical epidemiology characteristics of the women with HDP were analyzed. The risk factors affecting the HDP occurrence of the women were explored, and the prediction model for HDP occurrence of the women based on these risk factors was established, and the predictive value of the model was explored. Results: There were significant differences in the age, the body mass index (BMI) value before pregnancy, the family hypertension history, the history of diabetes before pregnancy, the conception way, the place of residence and the gravidity of the women between the two groups (P<0.05). There were no significant differences in the gestational weeks, the parity, the occupation, the family history of diabetes, the hepatitis B virus infection and the hypothyroidism of the women between the two groups (P>0.05). Multivariate logistic regression analysis showed that the advanced age, the high value of pre-pregnancy BMI, the history of HDP, the low education level, the pregnancy by assisted technology, the residence in rural area, and the high gravidity of the women were all the risk factors of their HDP occurrence (P<0.05). ROC curve was used to
analyze the predictive values of these risk factors with differences of the women for their HDP occurrence, which showed that the area under the curve (AUC) of the age, the pre-pregnancy BMI value, the HDP history, the education level, the conception way, the residence and the gravidity of the women for predicting their HDP occurrence were 0.707, 0.925, 0.606, 0.569, 0.601, 0.640, and 0.735, respectively. Based on the results of multivariate logistic regression analysis and ROC curve analysis, the prediction model for HDP was established, and finally including the age, the pre-pregnancy BMI value and the gravidity. ROC curve was used to test the predictive value of this model, and the AUC, the sensitivity and the specificity of this model for predicting the HDP occurrence of the women were 0.948, 92.2% and 85.0%, respectively. Conclusion: Among the pregnant women with HDP, their most manifestations are the advanced age, the high value of pre-pregnancy BMI, the previous history of HDP, the low education level, the pregnancy by assisted technology, the residence in rural areas, and the more gravidity, and so on. The Age, the pre-pregnancy BMI value and the gravidity of the women are included to establish a model for predicting their HDP, and the predictive model has certain clinical application value.