To analyze the incidence of gestational hypertension in hospital delivery of obstetrics and gynecology from 2016 to 2022, understand its characteristics and changes, to explore the influencing factors of gestational hypertension, and to construct a predictive model of gestational hypertension, to provide basis for clinical prediction of gestational hypertension.

A total of 101 583 medical records of hospitalized parturient in a hospital of gynecology and obstetrics and children in Sichuan Province from 2016 to 2022 were collected. The χ² test was used for univariate analysis, logistic regression analysis was used to establish the prediction model of influencing factors, and ROC curve was used to evaluate the discrimination of the prediction model.

The incidence of gestational hypertension in hospital delivery of obstetrics and gynecology from 2016 to 2022 was 0.053. The results of logistic regression model showed that older age (OR=1.392, 95%CI: 1.243-1.559), local household registration (OR=0.760, 95%CI: 0.689-0.839), hypothyroidism (OR=1.443, 95%CI: 1.160-1.794), intrahepatic cholestasis of pregnancy (ICP) (OR=1.874, 95%CI: 1.570-2.237), and diabetes mellitus (OR=1.367, 95%CI: 1.230-1.519), multipara (OR=0.693, 95%CI: 0.623-0.770), multiple births (OR=2.443, 95%CI: 2.135-2.795), family history of hypertension (OR=3.347,95%CI: 2.666-4.203), and in vitro fertilization-embryo transfer (IVF-ET) assisted reproduction (OR=1.821, 95%CI: 1.594-2.080) were the influencing factors of gestational hypertension. The area under the ROC curve constructed by the predictive model of gestational hypertension was AUC=0.655 (95%CI: 0.645-0.665).

The risk factors of gestational hypertension are older age, hypothyroidism, ICP, diabetes, multiple births, family history of hypertension, and IVF-ET assisted reproduction. The protective factors of gestational hypertension are local household registration and multipara. The predictive model of gestationall hypertension established in this study has good discrimination ability and has certain predictive value.