The increase in the frequency of pregnancy loss and the negative prognostic trend with the probable development of the demographic crisis in connection with the martial law in Ukraine and the mass migration of the female population provoked by it, requires the search for all possible ways of prevention and prophylactic of this pathology.
 Aim. Development and scientific substantiation of a risk-oriented model of prevention of pregnancy loss.
 Material and Methods. In the course of creating a model for determining the probability of the appearance of symptoms of pregnancy loss, we used a step-by-step method of logistic regression with the gradual exclusion of unreliable signs according to the Wald method.
 Results and Discussion. In the process of sociological research, the importance of 30 risk factors was analyzed, and 14 factors were singled out from among them using the logistic regression method, which, when combined, have a probable impact on the development of pregnancy loss. Namely: emotional strain/stress in everyday life, first ultrasound and biochemical screening, history of herpes type 1, chlamydial infection, COVID-19, tonsillitis, allergic rhinitis/sinusitis, autoimmune thyroiditis, presence of type I diabetes, use of artificial insemination, wine consumption, additional use of folic acid and drugs, marital status, and computer work. Among these 14 factors, three have a preventive effect, while the other 11 increase the risk of developing the pathology of pregnancy loss. This preventive model is reliable (p<0.001) and was transformed into the application "Individual prediction of the occurrence of pregnancy loss pathology" created for the Android platform, which provides 95.5% accuracy of the prediction of the occurrence of pregnancy loss pathology.
 Conclusions. The personalized preventive model can be applied by general practitioners/family doctors and obstetrician-gynecologists in order to form risk groups among pregnant women and prevent the occurrence of pregnancy loss pathology.