Objective
The purpose of this study was to estimate the prevalence of polycystic ovary syndrome (PCOS) and its phenotypes as defined by the National Institutes of Health, Rotterdam criteria, and Androgen Society.
Study Design
Thomson Reuters MarketScan Commercial databases (Thomson Reuters Healthcare Inc, New York, NY) for 2003-2008 were used to calculate the prevalence of PCOS and to assess differences in demographic characteristics and comorbid conditions among women who were 18-45 years old with and without PCOS.
Results
The prevalence of PCOS was 1585.1 per 100,000; women with phenotype A or classic PCOS were most prevalent at 1031.5 per 100,000. Women with PCOS were more likely than those without PCOS to be 25-34 years old, be from the South, be infertile, have metabolic syndrome, have been seen by an endocrinologist, and have taken oral contraceptives.
Conclusion
This is the first study to use all available criteria to estimate the prevalence of PCOS. Providers should evaluate women with menstrual dysfunction for the presence of PCOS.
Polycystic ovarian syndrome (PCOS) is considered the most common endocrine disorder among women of reproductive age, but prevalence estimates have varied from a low of 4.6% found by Knochenhauer et al in 1998 to a high of 6.8% noted by Diamanti-Kandarakis et al in 1999. Other estimates have fallen between these ranges. The exact cause of PCOS still is unknown, but it is characterized by a heterogeneous presentation of hyperandrogenism, ovulatory dysfunction, and polycystic ovaries. The heterogeneity of PCOS has contributed to debates over its diagnostic criteria, which includes a combination of clinical, biologic, and imaging findings.
The most commonly used definition for PCOS came from the 1990 National Institutes of Health’s (NIH) international conference on PCOS. The NIH conference concluded that PCOS should be defined (in order of importance) by (1) hyperandrogenism or hyperandrogenemia or both, (2) ovulatory dysfunction, and (3) exclusion of related disorders such as hyperprolactinemia, thyroid disorders, and nonclassic adrenal hyperplasia. By 2003, physicians recognized the need to expand the PCOS definition; the European Society of Human Reproduction and Embryology and the American Society for Reproductive Medicine jointly sponsored a PCOS consensus workshop in 2003 and developed the Rotterdam criteria.
The Rotterdam criteria expanded the definition of PCOS to include a third criterion: the presence of polycystic ovaries (PCO). Taking the heterogeneity of the syndrome into consideration, the presence of any 2 of the 3 criteria was sufficient for a diagnosis. Essentially, the Rotterdam criteria added 2 phenotypes of women with PCOS: (1) women with normal menstrual periods and normal fertility but who have androgen excess and PCOs and (2) women with ovarian dysfunction and PCOs but with normal androgen levels.
The final criteria consensus occurred in 2005; the Androgen Excess Society commissioned a 7-member task force to look into the controversy surrounding PCOS and to recommend a definition to its members that was based on a review of published evidence. The task force concluded that androgen excess is a central feature of the disease; therefore, PCOS should be defined by the presence of hyperandrogenism (clinical or biochemical or both) in combination with ovarian dysfunction (oligomenorrhea or anovulation) or polycystic ovaries or both, with the exclusion of any other causes of these findings (eg, congenital hyperplasia). When all the phenotypes were examined, 4 main phenotypes emerged ( Table 1 ).
| Feature | Phenotype A (National Institutes of Health or classic) | Phenotype B (Rotterdam 1) | Phenotype C (Rotterdam 2) | Phenotype D (Androgen Excess Society) |
|---|---|---|---|---|
| Clinical hyperandrogenism (hirsutism, acne, or alopecia) | Features present | Features present | Features absent | Features present |
| Menstrual or ovulatory dysfunction or both | Features present | Features absent | Features present | Features present |
| Polycystic ovaries | Features absent | Features present | Features present | Features present |
Available prevalence studies have been limited by small sample sizes, (154-400 participants), by single geographic regions, or by narrow definitions of PCOS ( Appendix ). Studies that have used the NIH definitions have found PCOS to be associated with infertility, metabolic abnormalities, and coronary heart disease risk factors. A study that was based on a larger sample size of women in which all available definitions of PCOS are used would provide more insight into the true prevalence of this complex syndrome and would affect how providers care for women. The objective of this study was to estimate the prevalence of PCOS and its various phenotypes among women of reproductive age. We also sought to determine what comorbidities are associated with these phenotypes.
Materials and Methods
Data source
Data were derived from the 2003-2008 Thomson Reuters MarketScan Commercial databases (Thomson Reuters Healthcare Inc, New York, NY). These databases include longitudinal, deidentified health insurance claims data from large employers and health plans across the United States and contain information on inpatient admissions, outpatient visits, and pharmaceutical claims. The data are derived from multiple US states that are geographically diverse. In 2008, >35 million individuals were enrolled, including employees and their dependents who were <65 years old and geographically distributed throughout the United States.
This commercial database has met or exceeded requirements of the US Health Insurance Portability and Accountability Act of 1996 and, accordingly, does not require specific patient consent to participate in the study.
Population
We restricted the analysis to women 18-45 years old. The upper limit of age was set at 45 years to keep within the age limit that had been used by other prevalence studies. PCOS was defined by the presence of hyperandrogenism, the presence of ovulatory dysfunction, and/or the presence of polycystic ovaries. Combinations of these conditions were used to create 4 mutually exclusive PCOS phenotypes that were based on the 3 available criterion recommendations (NIH, Rotterdam, and Androgen Society; Appendix ). International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes were used to identify these conditions in the database. Each condition was defined in the following manner:
Clinical hyperandrogenism
The presence of ICD-9-CM codes for acne (706.0 or 706.1), alopecia (704.0x), or hirsutism (704.1) defined clinical hyperandrogenism . Elevated testosterone, which is another characteristic of clinical hyperandrogenism, was not included because no ICD- 9-CM codes were available for this laboratory value in the database.
Ovulatory dysfunction
The presence of any 1 of the ICD-9-CM codes for the 626 series defined ovulatory dysfunction . These codes deal with disorders of menstruation and other abnormal bleeding from the female genital tract: 626.0, 626.1, 626.2, 626.3, 626.4, 626.5, 626.6, 626.7, 626.8, or 626.9.
Polycystic ovaries
The presence of ICD-9-CM code 256.4 defined polycystic ovaries .
Exclusion of other conditions
We excluded women who met the criteria for any of the PCOS phenotypes and had ≥1 of the following conditions: adrenal hyperplasia (code 255.0 or 255.9), hyperprolactinemia (code 253.1), or thyroid disorder (codes 244.0, 244.1, 244.2, 244.3, 244.8, 244.9, 242.90, or 242.91).
For example, to have phenotype A (classic PCOS), hyperandrogenism (code 706.0 or 706.1, 704.1, or 704.0x) and ovulatory dysfunction (codes 626.0, 626.1, 626.2, 626.3, 626.4, 626.5, 626.6, 626.7, 626.8, or 626.9) must be present, but polycystic ovaries (code 256.4), adrenal hyperplasia (code 255.0 or 255.9), hyperprolactinemia (code 253.1), and thyroid disorder (codes 244.0, 244.1, 244.2, 244.3, 244.8, 244.9, 242.90, or 242.91) all must be absent. More information on the other phenotypes is available in the Appendix .
Because of the heterogenic nature of PCOS and its complex presentation and definitions, we also assessed the prevalence of associated conditions seen among women with PCOS. The additional conditions that were studied were obesity (codes 278.0, 278.00-278.02, 783.1, or V77.8), infertility (code 628.0 or 628.9), syndrome X or metabolic syndrome (code 277.7), and diabetes mellitus (codes 250.00, 250.02, 250.90, or 250.92). For all conditions that are related to PCOS, diagnoses that were reported on inpatient claims were considered valid; diagnoses that were reported solely on outpatient claims were included only if the same diagnosis was reported on ≥2 outpatient claims that were made >30 days apart.
Unique identification numbers were used to track all women longitudinally over the study period. Women were included if they were enrolled in any health plan, even if no health claims were submitted during the study period.
Covariates
Pregnancy was defined as any inpatient hospital admission for a pregnancy-related diagnosis or procedure (codes V27.x, 650, 72.0, 72.1, 72.21, 72.29, 72.31, 72.39, 72.4, 72.51-72.54, 72.6. 72.71, 72.79, 72.8, 72.9, 73.22, 73.59, 73.6, 74.0-74.2, 74.4, or 74.99), excluding records with codes or procedures for hydatidiform mole, ectopic pregnancy, other abnormal products of conception, or abortion (codes 630, 631, 632, 633.x, 634.x-638.x, 639.x, 69.01, 69.51, 74.91, or 75.0). The variable “seen by endocrinologist” comprised any inpatient or outpatient visit to a provider who specialized in endocrinology and metabolism during the study period. Oral contraceptive use was identified by ≥1 filled oral contraceptive pill (OCP) prescriptions during the study period; no other types of contraception (eg, patch, ring, or injections) were included.
Statistics
The period prevalence and associated 95% confidence intervals (CI)s were estimated for women with PCOS in 3 age groups (18-24, 25-34, and 35-45 years old). The demographic characteristics and clinical characteristics of women with and without PCOS were compared with the use of 2-tailed χ 2 tests. We also assessed the prevalence of infertility, metabolic syndrome, obesity, pregnancy, diabetes mellitus, and OCP use among the PCOS phenotypes. We used multivariable logistic regression analyses to assess the likelihood of having these clinical conditions among women with and without PCOS. SAS statistical software (version 9.2; SAS Institute, Cary, NC) was used for all analyses. All analyses were conducted at an a priori significance level ( P < .05).
Results
PCOS prevalence
A total of 12,171,830 women who were 18-45 years old comprised the study group. Of these, 192,936 women (1.6%) met at least 1 of the definitions for PCOS. The overall prevalence of PCOS among the MarketScan database population of women 18-45 years old was 1585 per 100,000 ( Figure ). Among the phenotypes, women with phenotype A (classic PCOS) were the most prevalent, at 1032 per 100,000 ( Figure ). The estimated period prevalence (per 100,000) of PCOS from 2003-2008 among women 18-24 years old was 1300 (95% CI, 1287.2–1312.9); among women 25-34 years old, the estimated period prevalence as 2187 (95% CI, 2173.2–2201.9); and among women 35-45 years of age, the estimated period prevalence was 1286 (95% CI, 1276.0–1295.4).
Demographic and clinical characteristics
The mean age of the 12 million women in the study was 32.2 years (SD, ± 8.38). For the cohort of women with PCOS, the mean age was 31.4 years (SD, ± 7.19); for women without PCOS, the mean age was 32.2 years (SD, ± 8.40).
Women with PCOS were more likely than those without PCOS to be 25-34 years old, reside in the southern United States, be infertile, have metabolic syndrome, have received care from an endocrinologist, or have taken OCPs ( Table 2 ). On the other hand, women without PCOS were more likely to be obese, to have had a pregnancy, or have had diabetes mellitus ( Table 2 ).
| Variable | Polycystic ovary syndrome (n = 192,936) | No polycystic ovary syndrome (n = 11,978,894) | P value a |
|---|---|---|---|
| % b | % b | ||
| Age, y | < .0001 | ||
| 18-24 | 20.6 | 24.5 | |
| 25-34 | 45.3 | 32.6 | |
| 35-45 | 34.7 | 42.9 | |
| Geographic region c | < .0001 | ||
| Northeast | 10.3 | 9.8 | |
| North Central | 23.0 | 21.6 | |
| South | 47.5 | 45.4 | |
| West | 18.7 | 22.7 | |
| Unknown | 0.4 | 0.6 | |
| Comorbidity | |||
| Infertility | 2.5 | 0.7 | < .0001 |
| Metabolic syndrome | 0.11 | 0.09 | .0002 |
| Obesity | 0.6 | 1.3 | < .0001 |
| Pregnancy | 0.4 | 7.4 | < .0001 |
| Diabetes mellitus | 0.3 | 1.4 | < .0001 |
| Seen by an endocrinologist | 5.7 | 1.9 | < .0001 |
| Oral contraceptive pills use | 49.7 | 24.3 | < .0001 |
a χ 2 test that compared the proportion between women with and without polycystic ovary syndrome;
b Might not add up to 100% because of rounding;
c Final number differed because of missing data from each region.
Characteristics most prevalent among the 4 PCOS phenotypes were OCP use and having received care from an endocrinologist ( Table 3 ). Infertility was reported more frequently among women with phenotypes C and D than among women with phenotypes A and B. Table 4 shows the results of the multivariate logistic modeling that assessed the likelihood of having these clinical conditions among women with and without PCOS. We found that, although women with any PCOS phenotype had a higher probability of being infertile (odds ratio [OR], 3.28; 95% CI, 3.19–3.39) or having taken OCPs (OR, 3.06; 95% CI, 3.03–3.09) compared with women without PCOS, they were less likely to have had a pregnancy (OR, 0.04; 95% CI, 0.035–0.040; Table 4 ). Women with phenotype C or D were more likely to be infertile than women without PCOS (OR, 7.70; 95% CI, 7.40–8.02 and OR, 4.28; 95% CI, 3.92–4.68, respectively). Although women with phenotype D were more likely to be obese compared with women without PCOS (OR, 1.26; 95% CI, 1.09–1.46), women with phenotype A were less likely to be obese (OR, 0.19; 95% CI, 0.17–0.21) compared with women without PCOS. Finally, women with phenotypes B, C, or D were more likely to have metabolic syndrome than women without PCOS ( Table 4 ).
| Condition | Phenotype A a (n = 125,558) | Phenotype B b (n = 9906) | Phenotype C c (n = 47,450) | Phenotype D d (n = 10,022) | ||||
|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | n | % | |
| Infertility | 991 | 0.79 | 141 | 1.42 | 3163 | 6.67 | 617 | 6.16 |
| Metabolic syndrome | 14 | 0.01 | 36 | 0.36 | 121 | 0.26 | 46 | 0.46 |
| Obesity | 285 | 0.23 | 135 | 1.36 | 631 | 1.33 | 180 | 1.80 |
| Pregnancy | 302 | 0.24 | 34 | 0.34 | 438 | 0.92 | 70 | 0.70 |
| Diabetes mellitus | 126 | 0.10 | 67 | 0.68 | 296 | 0.62 | 74 | 0.74 |
| Saw endocrinologist | 1,994 | 1.59 | 2069 | 20.89 | 4,892 | 10.31 | 1985 | 19.81 |
| Oral contraceptive pills use | 63,219 | 50.35 | 5442 | 54.94 | 21,490 | 45.29 | 5816 | 58.03 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
