Gestational diabetes mellitus (GDM) and its consequences for the mother and child represent a serious not only medical, but also an economic problem. The urgency of studying this problem also lies in the fact that the number of pregnant women suffering from this disease is progressively increasing. The aim of the study was to study the risk factors and features of the course of pregnancy and childbirth in women with gestational diabetes mellitus, as well as to improve methods for diagnosing and predicting the development of diabetic fetopathy (DF). The work was carried out in 3 stages: 1 — prospective study of the course of pregnancy and childbirth in 104 women with GDM and in 50 pregnant women without GDM (control group); 2 — determination in the peripheral blood of the pregnant women of both groups of the level of glycemia, C-peptide, insulin, calculation of the insulin resistance index (HOMA-IR) with an assessment of the prognostic significance of these markers in the development of diabetic fetopathy; 3 — prediction of the development of diabetic fetopathy using a mathematical model that includes risk factors, the results of laboratory and instrumental research methods for this pathology. It was found that the violation of carbohydrate metabolism during pregnancy promotes the development of a large number of obstetric complications and in 36.5% of cases leads to the birth of children with diabetic fetopathy, which is manifested to a greater degree by macrosomia (30%), which increases the risk of perinatal complications and worsens the course of the period newborn in the future. An increase in the level of C-peptide was diagnosed in 87% of pregnant women with gestational diabetes, and an increase in the insulin resistance index in 93%, in contrast to the control group, where these indicators were 4 and 6%, respectively (p < 0.05). Using discriminant analysis, it was determined that the threshold for predicting the development of DF should be considered an increase in HOMA-IR above 7 with a confidence of 73%, but the level of C- peptide does not have a predictive value. The mathematical model, including risk factors, data of laboratory and instrumental methods for studying carbohydrate metabolism in the mother and fetus, created using regression analysis, reflects the probability of development of diabetic fetopathy in pregnant women with gestational diabetes with an accuracy of 91.4%, which will help to prevent this complication in at an earlier stage through insulin therapy.