A study on the distribution of high-risk pregnancies and the influencing factors of pregnancy outcomes in Changde City, Hunan Province, China.
Source: PubMed, NCBI / U.S. National Library of Medicine
To analyze the distribution of risk severity and the influencing factors of pregnancy outcomes among high-risk pregnant women in Changde City, Hunan Province, China. A total of 19,301 pregnant women from the jurisdiction of Changde City over the past year (July 2022 to July 2023) were selected as observational subjects. Their demographic information (including age, education level, occupation type, place of permanent residence) and prenatal care data [including gestational week at first check-up, number of prenatal visits, gravida, parity, high-risk assessment (dynamic evaluation), gestational week at delivery, mode of delivery, and maternal and neonatal outcomes] were analyzed and summarized. Univariate and multivariate logistic regression analyses were used to screen variables influencing pregnancy outcomes in high-risk pregnant women. The XGBoost algorithm was employed to construct a model for predicting pregnancy outcomes. 14,196 (73.55%) were initially assessed as high-risk pregnancies. Furthermore, 567 high-risk pregnant women subsequently experienced an escalation in their risk level. Among the 14,196 high-risk pregnant women, the incidences of severe preeclampsia and postpartum hemorrhage were relatively high among adverse maternal outcomes, while the incidence of preterm birth and low birth weight infants was higher among adverse fetal and neonatal outcomes. Univariate and multivariate logistic regression analyses revealed that age, permanent residence, having fewer
Abstract
To analyze the distribution of risk severity and the influencing factors of pregnancy outcomes among high-risk pregnant women in Changde City, Hunan Province, China. A total of 19,301 pregnant women from the jurisdiction of Changde City over the past year (July 2022 to July 2023) were selected as observational subjects. Their demographic information (including age, education level, occupation type, place of permanent residence) and prenatal care data [including gestational week at first check-up, number of prenatal visits, gravida, parity, high-risk assessment (dynamic evaluation), gestational week at delivery, mode of delivery, and maternal and neonatal outcomes] were analyzed and summarized. Univariate and multivariate logistic regression analyses were used to screen variables influencing pregnancy outcomes in high-risk pregnant women. The XGBoost algorithm was employed to construct a model for predicting pregnancy outcomes. 14,196 (73.55%) were initially assessed as high-risk pregnancies. Furthermore, 567 high-risk pregnant women subsequently experienced an escalation in their risk level. Among the 14,196 high-risk pregnant women, the incidences of severe preeclampsia and postpartum hemorrhage were relatively high among adverse maternal outcomes, while the incidence of preterm birth and low birth weight infants was higher among adverse fetal and neonatal outcomes. Univariate and multivariate logistic regression analyses revealed that age, permanent residence, having fewer than 5 prenatal visits, BMI ≥28 kg/mand pregnancy achieved by assisted reproductive technology significantly influenced pregnancy outcomes ( < 0.05). The pregnancy outcome prediction model based on the XGBoost algorithm achieved an area under the curve (AUC) of 0.742 (95% CI: 0.727-0.757) in the training set and an AUC of 0.762 (95% CI: 0.744-0.780) in the test set. The initial assessment rate of high-risk pregnancy among pregnant women in Changde City is high, and numerous factors influence adverse pregnancy outcomes. Clinically, it is essential to enhance precise screening and tiered management of high-risk pregnancies, implement targeted interventions for core risk factors, and strengthen health education and social support to minimize adverse maternal and infant outcomes and reduce the associated socioeconomic burden.