8582 THE PREVALENCE OF POLYCYSTIC OVARY SYNDROME IN ADULT WOMEN: THE IMPACT OF GEOGRAPHIC REGION

Abstract Disclosure: M. Amiri: None. S. Hatoum: None. R.P. Buyalos: None. A. Sheidaei: None. R. Azziz: Consulting Fee; Self; Spruce Biosciences, May Health, Core Access Surgical Technologies, Acacia Bio, Fortress Biotech, and Rani Therapeutics. Stock Owner; Self; Martin Imaging. Background: Polycystic ovary syndrome (PCOS) is a common hormonal disorder that affects women of reproductive age and is associated with various health complications. The reported prevalence of PCOS varies widely. However, variations in PCOS prevalence by geographic region are unclear. Objectives: This systematic review and meta-analysis aimed to compare PCOS prevalence in adults across World Health Organization (WHO) regions. Materials and methods: A search of PubMed and Embase was conducted for studies on PCOS prevalence in unselected populations published from inception to July 2023 using various diagnostic criteria. Studies were categorized according to: a) PCOS criteria (NIH 1990 [PCOS phenotypes A&B], AE-PCOS 2006 [PCOS phenotypes A-C], Rotterdam 2003 [PCOS phenotypes A-D], and ‘other/self-reported’ criteria); b) subject assessment type (direct assessment, survey/indirect assessment, or mixed assessment); and c) study quality, assessed using the PCOS Epidemiology and Phenotype (PEP) quality assessment tool. Meta-analyses were performed using fixed-effects or random-effects models depending on the heterogeneity among the studies. Publication bias was assessed using the Egger’s regression test and funnel plots. Subgroup analyses were conducted based on PCOS criteria, assessment types, study quality, and WHO region. Results: Of 6,644 records, 50 studies with including a total of 70,146 adult women (age ≥18 years) were included. The overall pooled prevalence of PCOS was 9.7% (95% CI: 8.25, 11.38) with a pooled PCOS prevalence by criteria of NIH 7.7% (95% CI: 6, 9.83), Rotterdam 11.83% (95% CI: 9.15, 15.17), AE-PCOS 11.48% (95% CI: 7.5, 17.18), and other/self-report 10.05% (95% CI: 6.63, 14.96). High-quality studies reported lower prevalences for Rotterdam and NIH criteria than low-qualities (6.99% vs 10.07% for NIH and 11.07% vs 20.97% for Rotterdam), with all studies using AE-PCOS being high-quality and all studies using other/self-report criteria being low-quality. Among high-quality studies using direct/mixed assessments, PCOS prevalence by NIH was highest in Western Pacific Region (8.19%) and lowest in Europe (6.47%), whereas by Rotterdam, it was highest in Europe (21.30%) and lowest in Western Pacific Region (7.67%). We also observed that highest PCOS prevalence based on AE-PCOS criteria was in Europe (16.09%) and lowest in Western Pacific Region (7.43%). Publication bias was not significant (Egger’s coefficient = 1.2, 95% CI: -1.49, 3.9, p = 0.38), and funnel plot showed a nearly symmetrical distribution of studies around the estimated effect size. Conclusions: This meta-analysis revealed that the reported PCOS prevalence in adults varies widely by WHO geographic region. Study quality, assessment type and diagnostic criteria for PCOS impact the assessment, with low-quality studies overestimating PCOS prevalence. Presentation: 6/2/2024