Research objective: The primary objective of this research is to show the correlation of HOMA-IR index with total fat percentage and levels of visceral body fat in PCOS patients. The secondary goal of this research is to show the correlation of HOMA-IR with waist circumference, hips circumference and body mass index. Subjects and methods: In this study, 30 subjects with a diagnosis of PCOS were included who came for regular check-ups to the Diabetes Center of the Department of Endocrinology, Diabetology and Metabolic Disorders of the KBC Split during the period from November 1, 2022 to May 31, 2023. Inclusion criteria is: women aged 18 to menopause with a diagnosis of PCOS. Exclusion criteria are: pregnant women, girls under the age of 18, women in menopause, women on OHK therapy or metformin. The test subjects were subjected to the measurement of anthropometric parameters and parameters that determine body composition with the help of a calibrated scale with a height meter, a tailor's tape measure and a professionally calibrated scale that works on the principle of bioelectrical impedance. The respondents also filled out a general questionnaire that contained general demographic questions, as well as questions related to the disease. All other specialist and laboratory findings submitted by the patients during this examination were taken into account. Results: The total fat percentage was strongly positively correlated with the HOMA index (ρ = 0.668, p < 0.001). The level of visceral fat was strongly positively correlated with the HOMA index (ρ = 0.698, p < 0.001). The percentage of muscle mass was strongly negatively correlated with the HOMA index (ρ = -0.699, p < 0.001). A strong positive correlation between body mass index and HOMA index was shown (ρ = 0.652, p = 0.001). Waist circumference had a very strong positive correlation with the HOMA index (ρ = 0.708, p < 0.001). Conclusion: A positive statistically significant correlation of all anthropometric parameters and parameters that determine body composition with the HOMA index was shown, except for the percentage of muscle mass, which showed a negative statistically significant correlation with the HOMA index.