PCOS (Polycystic Ovary Syndrome) Global Prevalence and Trends

How Common Is PCOS Globally?

PCOS is the single most common endocrine disorder among reproductive-age women, affecting 8 to 13% of this population by World Health Organization estimates. That translates to somewhere between 116 and 200 million women depending on the population denominator used. The wide range exists because prevalence is inseparably tied to which diagnostic criteria a research team applies.

The WHO formally recognizes PCOS as a major public health concern affecting women globally, noting that the condition is "a common hormonal disorder" with substantial reproductive and metabolic consequences [1].

Why Prevalence Estimates Vary So Widely

Three major diagnostic systems produce materially different prevalence numbers for the same population.

The NIH (National Institutes of Health) 1990 criteria require both hyperandrogenism and oligo-anovulation. Studies using these criteria consistently return prevalence figures of around 6 to 8% [2].

The Rotterdam 2003 criteria require only two of three features: hyperandrogenism, oligo-anovulation, and polycystic ovarian morphology on ultrasound. Adding the third phenotype, polycystic morphology alone with irregular cycles, inflates population estimates to 15 to 20% in some cohorts [3].

The Androgen Excess and PCOS Society (AES) 2006 criteria mandate hyperandrogenism plus at least one other feature. Estimates under AES criteria fall between the NIH and Rotterdam figures.

A 2016 systematic review published in Human Reproduction Update (N=15 studies, combined sample exceeding 24,000 women) found that Rotterdam-based prevalence was roughly 50 to 80% higher than NIH-based estimates in the same populations [3]. That single methodological choice accounts for more variance in published prevalence than any true geographic or demographic difference.

The Underdiagnosis Problem

Underdiagnosis compounds every prevalence estimate. A cross-sectional study in the Journal of Clinical Endocrinology and Metabolism found that fewer than 30% of women who met Rotterdam criteria for PCOS had received a prior clinical diagnosis [4]. Barriers include irregular access to gynecologic care, clinician unfamiliarity with non-classic presentations (normal-weight women, mild cycle irregularity), and the absence of a single confirmatory biomarker.

Regional Prevalence: How Rates Differ by Geography

Prevalence is not uniform across populations. Studies from South Asia, East Asia, the Middle East, and Western countries reveal meaningful differences, though some of that variation reflects differing research methodology rather than true biological differences.

South and Southeast Asia

South Asian populations show some of the highest recorded PCOS rates. A 2019 community-based study from India using Rotterdam criteria found a prevalence of 22.5% among women aged 18 to 45 [5]. A meta-analysis of Southeast Asian populations published in PLOS ONE reported pooled prevalence of 10.7% (95% CI: 8.7 to 12.9%) using Rotterdam criteria, rising to 14.3% in urban subgroups [6].

Genetic predisposition, dietary patterns high in refined carbohydrates, and rapid urbanization are frequently cited contributors. Insulin resistance, a key metabolic driver of PCOS, appears more pronounced in South Asian women at lower BMI thresholds compared with European-ancestry populations [5].

East Asia

Chinese population studies report prevalence ranging from 5.6% to 10.5% depending on criteria and sampling frame. A landmark Chinese national study (N=15,924) published in 2013 found a prevalence of 5.6% using Chinese-adapted diagnostic criteria and 7.1% when Rotterdam criteria were applied directly [7]. Japanese and Korean cohort data cluster similarly in the 6 to 9% range.

Middle East and North Africa

Studies from Iran, Turkey, and Egypt report Rotterdam-based prevalence of 12 to 20%. A 2023 systematic review covering the MENA region found a pooled estimate of 13.8% (95% CI: 11.2 to 16.6%) across 18 studies with combined N exceeding 30,000 women [8]. Consanguineous marriage patterns and genetic founder effects in some subpopulations may contribute to clustering.

Europe and North America

European population-based cohorts generally report prevalence of 8 to 15% under Rotterdam criteria. A well-cited UK cohort study (N=230 unselected women) found prevalence of 26% under Rotterdam criteria but 8% under NIH criteria, the most extreme within-study divergence in the literature [2].

In North America, the CDC does not publish standalone PCOS prevalence data, but a nationally representative study using NHANES data estimated 6 to 12% prevalence depending on diagnostic threshold [9].

Trends Over Time: Is PCOS Becoming More Common?

Reported PCOS rates have increased over the past two decades. Whether that reflects a true rise in incidence or improved detection is debated, though evidence supports both explanations.

The Role of Obesity and Metabolic Change

Obesity amplifies PCOS phenotype expression. Fat tissue increases androgen conversion, worsens insulin resistance, and disrupts the hypothalamic-pituitary-ovarian axis. As global obesity prevalence has climbed, the WHO reports that worldwide obesity has nearly tripled since 1975 [10], more women cross diagnostic thresholds for PCOS even if their underlying hormonal milieu has not changed.

A prospective cohort analysis published in Human Reproduction found that weight gain of 5 to 10 kg over 5 years increased the odds of developing a PCOS phenotype by approximately 1.7-fold (OR 1.69, 95% CI: 1.2 to 2.4, P<0.01) [11].

Improved Diagnostic Awareness

High-resolution transvaginal ultrasound became widely available in the 1990s and 2000s, enabling detection of polycystic ovarian morphology that earlier equipment missed. The Rotterdam criteria's inclusion of morphology as a standalone qualifying feature in 2003 immediately expanded the diagnosable population. Publication of international guidelines by the Endocrine Society in 2013 and the international evidence-based PCOS guideline in 2018 trained more clinicians to recognize non-classic presentations [12].

The 2023 international evidence-based PCOS guideline states: "PCOS is underdiagnosed and undertreated, contributing to significant physical and psychological burden" [12].

Endocrine Disruptor Exposure

A growing body of research links prenatal and adult exposure to endocrine-disrupting chemicals (EDCs), bisphenol A, phthalates, and per- and polyfluoroalkyl substances (PFAS), with elevated androgen levels and PCOS-like phenotypes. A 2021 meta-analysis in Environmental Health Perspectives found that women with PCOS had serum bisphenol A concentrations approximately 56% higher than controls (standardized mean difference 0.61, 95% CI: 0.42 to 0.80) [13]. Whether EDC exposure causes PCOS or whether PCOS physiology alters EDC metabolism remains under investigation.

Metabolic and Reproductive Burden: Why Prevalence Numbers Matter

Prevalence data carries direct clinical weight. PCOS is not limited to menstrual irregularity and unwanted hair growth.

Metabolic Disease Risk

Between 50% and 80% of women with PCOS have insulin resistance regardless of BMI [14]. This drives substantially elevated lifetime risks for type 2 diabetes, metabolic syndrome, and non-alcoholic fatty liver disease. A Cochrane-reviewed meta-analysis found that women with PCOS have a 2.9-fold higher odds of developing type 2 diabetes compared with age- and BMI-matched controls (OR 2.87, 95% CI: 1.93 to 4.29) [15].

Cardiovascular risk markers, elevated LDL, low HDL, elevated triglycerides, appear at higher rates in PCOS cohorts even before diabetes onset. The American Heart Association flags PCOS as an independent cardiovascular risk factor in its 2021 scientific statement on cardiovascular disease in women [16].

Reproductive Impact

PCOS accounts for approximately 70 to 80% of anovulatory infertility cases globally [17]. The condition is the most common cause of ovulatory dysfunction evaluated in reproductive endocrinology clinics. A large European multicenter trial (N=1,508) found that 85% of women with WHO Group II anovulation (the category encompassing PCOS) achieved pregnancy within three ovulation induction cycles using letrozole or clomiphene [17].

Mental Health Comorbidity

Depression and anxiety prevalence is roughly threefold higher in women with PCOS than in age-matched controls. A 2018 systematic review (N=26 studies, 2,384 women with PCOS) published in Fertility and Sterility found pooled odds ratios of 3.78 for depression (95% CI: 2.40 to 5.95) and 5.62 for anxiety (95% CI: 2.28 to 13.88) [18]. Body image concerns, hirsutism, acne, and fertility uncertainty all contribute.

Diagnostic Criteria: How Definition Shapes the Prevalence Debate

The single most consequential variable in any PCOS prevalence study is the diagnostic criteria applied. Understanding the three main systems allows clinicians and patients to interpret reported rates accurately.

NIH 1990 Criteria

Requires: (1) chronic anovulation AND (2) clinical or biochemical signs of hyperandrogenism. Polycystic ovarian morphology is not required. This is the most restrictive definition and produces the lowest prevalence estimates (approximately 6 to 8%) [2].

Rotterdam 2003 Criteria

Requires two of three: (1) oligo/anovulation, (2) clinical or biochemical hyperandrogenism, (3) polycystic ovarian morphology. The European Society of Human Reproduction and Embryology and the American Society for Reproductive Medicine jointly published these criteria and they remain the most widely used in current research [3].

AES 2006 Criteria

Requires: hyperandrogenism (clinical or biochemical) PLUS oligo-anovulation and/or polycystic ovarian morphology. This excludes the "normoandrogenic" Rotterdam phenotype and produces intermediate prevalence estimates.

The 2023 international PCOS guideline recommends Rotterdam criteria for clinical practice while acknowledging that the normoandrogenic phenotype (morphology plus anovulation without hyperandrogenism) carries a milder metabolic profile and may represent a distinct entity [12].

Populations at Highest Risk: Who Gets PCOS?

Certain characteristics substantially raise a woman's probability of meeting PCOS criteria.

Age and Reproductive Stage

PCOS prevalence peaks during the reproductive years, roughly ages 15 to 44. Adolescent diagnosis is challenging because irregular cycles and transient hyperandrogenism are normal within two years of menarche. The 2023 international guideline recommends a minimum of two years post-menarche before applying adult Rotterdam criteria to adolescents [12].

After menopause, hyperandrogenism and cycle irregularity resolve, but metabolic sequelae, elevated cardiovascular risk, diabetes, and dyslipidemia, persist. Some cohort data suggest that postmenopausal women with a history of PCOS carry a 2.2-fold higher risk of cardiovascular events compared with controls [16].

Genetic Predisposition

First-degree relatives of women with PCOS show a 20 to 40% prevalence of the condition [19]. Twin studies estimate heritability at approximately 70%. Candidate gene studies have implicated variants in DENND1A (a regulator of androgen biosynthesis), FSHR, LHCGR, and INSR, though no single genetic marker is diagnostically useful [19].

BMI and Body Composition

Obesity increases PCOS prevalence but does not cause it. Roughly 20 to 30% of women with PCOS are normal weight (BMI <25), and their metabolic profile is often underestimated in clinical encounters [14]. Lean PCOS phenotypes frequently present with elevated luteinizing hormone (LH) to follicle-stimulating hormone (FSH) ratios rather than the hyperinsulinemia prominent in obese phenotypes.

Implications for Healthcare Systems and Public Health Policy

Given that PCOS affects 8 to 13% of reproductive-age women, a period spanning roughly 30 years of life, the aggregate healthcare cost is substantial.

A 2021 Australian economic analysis estimated the annual direct healthcare cost of PCOS at AUD $400 million in Australia alone, with total economic burden (including productivity loss and quality-of-life impact) exceeding AUD $3.5 billion annually [20]. Extrapolated to the United States population using proportional scaling, comparable US figures likely exceed USD $8 billion annually.

The 2018 international evidence-based guideline for PCOS assessment and management explicitly calls for population-level screening strategies and improved clinician education as cost-effective interventions, noting that early lifestyle intervention in women with PCOS reduces downstream diabetes risk by approximately 50% [20].

Public health priorities include standardizing diagnostic criteria globally, improving access to diagnostic ultrasound and androgen assays in low- and middle-income countries, and integrating PCOS screening into routine well-woman visits starting in adolescence.

What Clinicians Should Do With These Numbers

Prevalence data translates into concrete clinical actions.

A 30-year-old woman presenting with irregular cycles and unwanted facial hair has a prior probability of PCOS exceeding 50% before any laboratory test. Clinicians should obtain serum total testosterone (or free androgen index), LH, FSH, and 17-hydroxyprogesterone to exclude congenital adrenal hyperplasia. Fasting glucose and a 2-hour 75 g oral glucose tolerance test are recommended by the Endocrine Society rather than fasting glucose alone, because women with PCOS have disproportionately elevated post-load glucose even with normal fasting values [12].

Clinicians practicing in South Asian, Middle Eastern, or Indigenous communities should apply lower clinical suspicion thresholds given the documented higher prevalence in these groups [5].