Construction and validation of an early warning model for gestational diabetes mellitus based on baseline data, vitamin D, and thyroid function status
10.3760/cma.j.cn115807-20240919-00295
- VernacularTitle:基于基线资料、维生素D和甲状腺功能状态构建妊娠期糖尿病的早期列线图预警模型与检验
- Author:
Yan SUN
1
;
Shaowen SHI
;
Jiaying WANG
;
Qian REN
Author Information
1. 秦皇岛市第一医院生殖医学科，秦皇岛 066000
- Publication Type:Journal Article
- Keywords:
Thyroid;
Vitamin D;
Hypothyroidism;
Baseline data;
Gestational diabetes mellitus;
Warning model
- From:
Chinese Journal of Endocrine Surgery
2025;19(1):74-79
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct an early warning model for gestational diabetes mellitus (GDM) based on baseline data, vitamin D (VitD) , and thyroid function status.Methods:A prospective study was conducted to select 126 patients with GDM (GDM group) and 126 pregnant women without GDM (control group) admitted to the Obstetrics Department of Qinhuangdao First Hospital from Jan. 2022 to May. 2024. The single-factor and multi-factor LASSO Logistic regression analysis was used to analyze the influencing factors of the risk of GDM. Based on the results of the multi-factor analysis, an early warning model for GDM was constructed, evaluated, and validated.Results:Age, pre-pregnancy body mass index (BMI) , family history of diabetes, thyroid function, low density lipoprotein (LDL-C) , triglyceride (TG) , VitD, fasting plasma glucose (FPG) , glycated hemoglobin (HbA1c) and blood uric acid were compared, and the differences were statistically significant ( P < 0.05) . LASSO Logistic regression analysis showed that family history of diabetes, hypothyroidism, pre-pregnancy BMI, TG, VitD, FPG, HbA1c and blood uric acid were independently correlated with the risk of GDM ( P < 0.05) . A GDM early warning model was constructed based on the results of multiple factors, with a C-index of 0.876, indicating good predictive performance; The model evaluation and validation results show that the model has good internal and external calibration, high consistency between predicted values and actual observed values, and good predictive value and discrimination in external data sets. Conclusions:Baseline data such as hypothyroidism, VitD, and pre-pregnancy BMI are independent factors that affect the occurrence of GDM. The early warning model for GDM based on these indicators has good predictive performance and clinical applicability, and can be used as an effective model for early prediction of GDM in clinical practice, as well as guiding clinical prevention and treatment efforts.
