Prevalence of polycystic ovary syndrome: a global and regional systematic review and meta-analysis

BACKGROUND: Polycystic ovary syndrome (PCOS) affects women globally, but its prevalence across World Health Organization (WHO) regions has not previously been reported. OBJECTIVE AND RATIONALE: We aimed to synthesize evidence on the prevalence of PCOS by diagnostic criteria and by WHO geographic regions to inform the International Evidence-Based PCOS Guideline. SEARCH METHODS: A systematic search of OVID MEDLINE, All EBM, PsycInfo, EMBASE, and Cumulative Index to Nursing and Allied Health Literature was conducted from 1990 to November 2024. Studies assessing PCOS prevalence in an unselected population were included. Non-primary studies or those with unclear diagnostic criteria were excluded. The primary outcome was PCOS prevalence among adult women. The secondary outcome was PCOS prevalence among women of all ages. Random effects meta-analysis using the DerSimonian and Laird method was applied for estimating the overall effect size. Two reviewers independently assessed risk-of-bias (RoB) and evidence certainty. OUTCOMES: The search yielded 16 664 articles, of which 119 unique studies (in 137 articles) were eligible, and 92 (including 157 181 participants) were pooled in a meta-analysis. By diagnostic criteria, PCOS global prevalence among adult women only was 12.1% (95% CI: 9.8, 14.8; I2: 98.8%) using Rotterdam criteria, 7.9% (95% CI: 6.2, 9.9; I2: 96.2%) using the original National Institute of Health (NIH) criteria, 12.7% (95% CI: 8.2, 17.9; I2: 98.0%) using the Androgen Excess (AE)-PCOS criteria, and 7.8% (95% CI: 5.8, 10.0; I2: 99.4%) by self-report. By WHO regions, PCOS prevalence among adult women when using Rotterdam criteria was highest in the Eastern Mediterranean region (15.1%; 95% CI: 11.1, 19.7) and the South-East Asian region (14.3%; 95% CI: 5.8, 25.9), followed by the European region (11.7%; 95% CI: 5.1, 20.3), the region of the Americas (10.5%; 95% CI: 3.0, 21.7), and the Western Pacific region (9.1%; 95% CI: 6.2, 12.5), with no data from Africa. Subgroup analysis using Cochran's Q test indicated a statistically significant difference in prevalence by WHO region (P = 0.022). Subgroup analyses including adolescents yielded a lower prevalence globally, with a global prevalence of 11.4% (95% CI: 9.5, 13.5) by Rotterdam criteria, 7.1% (95% CI: 5.7, 8.7) by NIH criteria, 11.2% (95% CI: 7.4, 15.5) by AE-PCOS criteria, and 7.6% (95% CI: 5.8, 9.6) on self-report. Of the 119 studies, 30 had low, 49 had moderate, and 40 had high RoB. Certainty of evidence ranged from very low to low. WIDER IMPLICATIONS: This is the most comprehensive and contemporary review of PCOS prevalence and highlights past inconsistencies in diagnostic criteria and individual diagnostic features. Pooled PCOS prevalence was 12.1% by the Rotterdam criteria and was highest in the Eastern Mediterranean and the South-East Asian regions, with a potentially different health burden of PCOS across world regions. These findings directly inform International PCOS Guidelines, including updated guideline diagnostic criteria and refined individual features, emphasizing early, accurate diagnosis. REGISTRATION NUMBER: PROSPERO CRD42022372029.