Risk prediction of excessive gestational weight gain based on a nomogram model
Description
Background: Excessive Gestational Weight Gain is a global public health problem with serious and long-term effects on maternal and offspring health. Early identification of at-risk groups and interventions is crucial for controlling weight gain and reducing the incidence of excessive gestational weight gain. Currently, tools for predicting the risk of excessive gestational weight gain are lacking in China. This study aimed to develop a risk-prediction model and screening tool to identify high-risk groups in the early stages. Methods: A total of 306 pregnant women were randomly selected who underwent regular obstetric checkups at a tertiary-level hospital in China between January and March 2023.Logistic regression analysis was used to construct the risk-prediction model. The goodness of fit of the model was assessed using the Hosmer-Lemeshow test, and the predictive performance was evaluated using the area under the receiver operating characteristic (ROC) curve, calibration plots, and k-fold cross-validation. R4.3.1 software was used to create a nomogram. Results: The prevalence of excessive gestational weight gain was 50.32%. Logistic regression analysis revealed that pre-pregnancy overweight (OR=2.563, 95%CI:1.043- 6.299), obesity (OR=4.116, 95%CI:1.396-12.141), eating in front of a screen (OR=6.230, 95%CI:2.753-14.097); frequency of weekly consumption of sugar-sweetened beverages/ desserts/western fast food(OR=1.948,95%CI:1.363- 2.785); and pregnancy body image (OR=1.030, 95%CI:1.014 -1.047) were risk factors for excessive gestational weight gain.Parity (OR=0.453, 95%CI:0.275 -0.740),protective motivation to manage pregnancy body mass (OR=0.979, 95%CI:0.958-1) and the time of daily moderate- intensity physical activity (OR=0.228, 95%CI:0.113-0.461) were protective factors against excessive gestational weight gain. The area under the ROC curve of the model was 0.885, the mean value of ten-fold cross-validation was 0.857 for AUC. Conclusion: The risk-prediction model developed in this study proved to be effective, providing a valuable basis for early identification and precise intervention in individuals at risk of excessive gestational weight gain.
Files
Steps to reproduce
Materials Possible risk factors for EGWG were identified by reviewing the domestic and foreign literature using evidence-based methods and expert consultations. A self-developed questionnaire was administered to collect information. The questionnaire included questions on maternal age, pre-pregnancy body mass index (BMI), place of residence, education level, family monthly income, marital status, employment status, parity, smoking, drinking, dietary patterns, dietary behavior,protective motivation for weight management during pregnancy questionnaire[13], Body Image in Pregnancy Scale(BIPS)[14],Social Support Rating Scale (SSRS)[15], Edinburgh Postnatal Depression Scale (EPDS)[16],Perceived Stress Scale (PSS-10)[17],and Pregnancy Physical Activity Questionnaire (PPAQ)[18]. Definition of EGWG According to the recommendations for GWG, which represent the first industry standard for GWG in China [1], optimal GWG ranges have been established based on pre-pregnancy BMI categories. These categories include underweight (BMI < 18.5 kg/m2), normal weight (18.5 kg/m2 ≤ BMI < 24.0 kg/m2), overweight (24.0 kg/m2 ≤ BMI < 28.0 kg/m2), and obesity (BMI≥28.0 kg/m), with corresponding optimal GWG ranges of 11.0-16.0, 8.0-14.0, 7.0-11.0, and 5.0-9.0 kg, respectively [1]. EGWG was defined as the actual GWG surpassing the established optimal GWG values for the respective BMI categories. Procedures and ethical considerations The hospital’s ethics committee approved this study. Prior to participation, all participants provided written informed consent. The data collectors were uniformly trained to ensure accuracy and consistency of data collection. A face-to-face questionnaire survey was conducted from 14 to 27+6 weeks of gestation when pregnant women were admitted to the Department of Obstetrics and Gynecology. The investigator immediately collected the questionnaires, checked the authenticity of the data, and eliminated invalid questionnaires with inconsistent options or all the same options. A secondary check by another team member ensured the accuracy of the collected data and recalculated the scores. Weight beginning of pregnancy extracted from electronic medical records. Weight before delivery, measured a week before delivery. Statistical analysis Data entry was conducted using Excel, statistical analyses were performed using SPSS version 26.0, and the nomogram model was constructed using R software version 4.3.1.