Objective The aim of this study is to develop and validate a prediction model of delayed onset of lactogenesis among mothers with gestational diabetes mellitus. Methods This was a prospective study. A total of 511 mothers with GDM hospitalized at seven tertiary (grade 3A) hospitals in five cities of Guangdong Province, China, between October 10, 2023, and December 8, 2024, were enrolled in the study using convenience sampling. Univariate regression, LASSO regression and logistic regression were used to construct and validation risk prediction model. Logistic regression and R language was used to draw the nomogram. Model performance was evaluated using the area under the receiver operator characteristic curve (AUC), Hosmer-Lemeshow goodness-of fit test, and clinical decision curve analysis, assessing the discrimination ability, calibration, and clinical utility of the model. Results Based on logistic regression analysis, we identified several significant delayed onset of lactogenesis risk factors for mothers with gestational diabetes mellitus: pre-delivery BMI, Edinburgh Postnatal Depression Scale score, serum albumin levels, LATCH score, blood glucose control during pregnancy. The AUC was 0.828, (95% CI :0.779~0.877),with a sensitivity of 70.9%, a specificity of 83.7%,optimal cut-off point of 0.341, a maximum Youden index of 0.546. The Hosmer-Lemeshow test yielded a χ 2 value of 7.226 and p = 0.546.The internal verification showed the AUC was 0.806(95% CI :0.728, 0.884), indicating the model’s high discrimination ability. Calibration curves showed good agreement between predicted and observed values, confirming good calibration. The clinical decision curve analysis further supported the model’s clinical utility. Conclusion The prediction model constructed and verified in this study was to predict delayed onset of lactogenesis risk for mothers with gestational diabetes mellitus, providing an effective evaluation tool for healthcare professionals.
Luo et al. (Thu,) studied this question.