Construction of a Risk Prediction Model for Pre-eclampsia in Northern China: A Study of 798 Hospitalized Pregnant Women

Abstract

Objective To screen the risk factors for pre-eclampsia in Northern China and construct a risk prediction model. Methods Clinical data of 798 hospitalized pregnant women from March 2023 to August 2023 at the First Hospital of Jilin University were collected and analyzed. The patients were divided into a pre-eclampsia group (N = 176, case group) and a non-pre-eclampsia group (622 cases, control group) based on clinical diagnosis. Patient medical history, family history, early pregnancy diagnostic data from prenatal care handbooks, and lifestyle information during pregnancy were collected, resulting in a total of 106 parameters for analysis. Univariate analysis and logistic regression analysis were used to identify independent risk factors associated with pre-eclampsia. Subsequently, ROC curve analysis was conducted to evaluate the predictive performance of the model, and a disease prediction model was constructed. Results (1) Independent risk factors for pre-eclampsia during early pregnancy included a history of pre-eclampsia, chronic hypertension, family history of hypertension, systolic blood pressure ≥ 120mmHg, diastolic blood pressure ≥ 80mmHg, education level, daily intake of dairy products > 100mL, education level of college or above, and white blood cell count > 1.3×10⁹/L. (2) The probability (P) of prediction was calculated based on these factors using the formula P = 1/[1 + EXP(-1.670 + 3.326×history of pre-eclampsia + 3.151×history of chronic hypertension + 1.276×family history of hypertension + 0.786×systolic blood pressure ≥ 120mmHg + 3.205×diastolic blood pressure ≥ 80mmHg + 0.625×daily intake of dairy products > 100mL-0.792×education level of college or above + 1.000×white blood cell count > 1.3×10⁹/L)]. (3) The area under the ROC curve based on this data was 0.804 [95% CI (0.756, 0.852)], P<0.05. (4) Validation of the model among 107 pregnant women, yielded an accuracy rate of 82.24%. Conclusion The risk prediction model, developed using identified risk factors, effectively predicts pre-eclampsia risk in high-risk individuals, offering valuable guidance for clinicians' decision-making.