Introduction. Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders in women worldwide. Diagnostic criteria for this syndrome were first established in 1990 and have been repeatedly revised since then. The purpose of the work. Based on the literature review, to systematize the data and analyze the prevalence and diagnostic criteria of PCOS. Materials and methods. In this work, the literature from the PubMed, Oxford Academic, ResearchGate, CyberLeninka databases was studied. For the study, using keywords, studies were selected devoted to the analysis of diagnostic criteria and assessing the prevalence of PCOS. Results and discussion. The work considered changes in PCOS diagnostic criteria since 1990, as well as the frequency of occurrence of this pathology based on different approaches to its detection. The results of studies conducted in different years using the approved criteria showed that the use of more modern methods significantly increases the number of diagnosed cases of PCOS. The analysis demonstrated that the prevalence of PCOS varies from 6 % to 21 % depending on the criteria used. Conclusion. It was found that the presence of various methods for diagnosing PCOS leads to significant differences in prevalence rates. Methods based on the Rott and AE-PCOS criteria were recognized as the most effective due to the wide coverage of clinical manifestations, which is especially important for women who do not meet all the classic NIH parameters. Attention is also focused on the difference in the frequency of manifestation of individual symptoms, which emphasizes their heterogeneity and underestimation in diagnosis. Thus, the analysis confirmed the need to optimize diagnostics for the timely initiation of treatment, prevention of metabolic and reproductive disorders.