MON-187 Global PCOS Prevalence: A Meta-Analysis Using the Recently Developed, PCOS Epidemiology and Phenotype Tool

Abstract Disclosure: M. Amiri: None. S. Hatoum: None. A. Sheidaei: None. R.P. Buyalos: None. R. Azziz: None. Background: Polycystic ovary syndrome (PCOS) is a heterogeneous syndrome characterized by hyperandrogenism, ovulatory dysfunction, and/ or polycystic ovaries. Reported PCOS prevalence varies widely and is influenced by diagnostic criteria, age, and, most importantly, study quality. Objective: This systematic review and meta-analysis assessed the impact of study quality on reported PCOS prevalence, using our recently developed PCOS Epidemiology and Phenotype (PEP) tool. We compared prevalence estimates from high- and low-quality studies across different age groups (adults, adolescents, and mixed populations). Materials and Methods: We conducted a comprehensive search of PubMed and Embase for studies on PCOS prevalence in unselected populations from inception to February 2024. We included studies reporting PCOS prevalence using established diagnostic criteria (NIH, Rotterdam, AE-PCOS). Study quality was assessed using our PEP tool, which addresses methodological challenges inherent in PCOS prevalence research, such as recruitment methods, diagnostic criteria for PCOS, adjustment for incomplete assessments, exclusion of mimicking disorders, and evaluation of clinical and biochemical HA. We performed meta-analyses using fixed or random-effects models based on the significance of heterogeneity (>50%), and subgroup analyses based on study quality (high vs. low) and age group (adults, adolescents, and mixed populations). Results: From 7144 records, 88 studies (n=561,287) were included. Among studies on all age groups, high-quality studies reported PCOS prevalence of 6.63% (95% CI: 5.34-8.21) (NIH), 10.89% (95% CI: 8.41-13.99) (Rotterdam), and 10.61 % (95% CI: 6.70-16.40) (AE-PCOS). In contrast, low-quality studies reported substantially higher prevalence: 13.34% (95% CI: 8.01-21.38) (NIH) and 18.95% (95% CI: 11.33-29.96) (Rotterdam). Notably, all AE-PCOS studies were high-quality, precluding comparison. In adults, high-quality studies reported a prevalence of 7.25% (95% CI: 5.92-8.83) (NIH), 10.70% (95% CI: 7.87-14.38) (Rotterdam), and 11.57% (95% CI: 6.61-19.48) (AE-PCOS), whereas low-quality studies reported 13.07% (95% CI: 7.19-22.58) (NIH) and 17.50% (95% CI: 9.60-29.77) (Rotterdam). For adolescents, high-quality studies reported 4.44% (95% CI: 2.48-7.85) (NIH), but we found no high-quality studies using Rotterdam or AE-PCOS criteria, and no low-quality using any criteria. Conclusions: Study quality significantly influences reported PCOS prevalence; high-quality studies reported lower, more consistent prevalences compared to low-quality studies, the latter with wider confidence intervals, reflecting a lack of accuracy. Therefore, low-quality studies may overestimate PCOS prevalence, and using a tailored tool like the PEP is crucial for accurate global prevalence estimation. Presentation: Monday, July 14, 2025