Individual Prediction of Insulin Therapy in Gestational Diabetes: Development of a Risk Calculator Based on Real-World Data from the GestDiab Registry.

Introduction: The prevalence of gestational diabetes mellitus in Germany is approximately 10%. One third of affected women require insulin therapy when conservative measures such as dietary changes and physical activity are insufficient to achieve target glucose levels. Timely initiation of insulin therapy is crucial for optimising obstetric outcomes. Early identification of high-risk patients at the time of diagnosis would facilitate prompt and individualised treatment adjustments. Materials and Methods: A risk calculator was developed based on clinical parameters and medical history information to estimate the individual risk for insulin therapy. The models were derived from real-world data of the GestDiab registry, comprising 14157 pregnancies between 2018 and 2020, of which 4319 (30.5%) required insulin therapy. Results: Various models incorporating maternal age, gestational age at diagnosis, parity, gravidity, body mass index, 75 g oral glucose tolerance test values, HbA1c levels, history of gestational diabetes mellitus, and family history of diabetes were developed. Validation using the GestDiab cohort from 2021 demonstrated that the model including all variables exhibited the highest predictive power (AUC 0.740). Conclusions: The risk calculator is provided online to support both patients and physicians in making informed decisions. Individualised counselling based on personal risk assessments may enhance therapy adherence and potentially reduce the necessity for insulin therapy.