New numbers could help diagnose and treat a common, complex condition

Researchers gain better understanding of polycystic ovary syndrome's impact in U.S.

Polycystic ovary syndrome (PCOS) is the most common hormone disorder affecting people with ovaries who are of reproductive age. The causes aren’t fully understood, but PCOS can lead to infertility and increase the risk of chronic health conditions like diabetes, heart disease, and high blood pressure. People with PCOS often experience symptoms of higher testosterone levels like increased facial hair and irregular periods and have difficulty maintaining a healthy body weight. According to the Centers for Disease Control and Prevention, as many as 5 million people in the U.S. have the condition.

A new study from Kaiser Permanente Washington Health Research Institute (KPWHRI) researchers and collaborators provides the first estimates of the incidence of PCOS (the rate of new cases) in the U.S., as well as new estimates for prevalence (how many people currently have the condition). Onchee Yu, MS, a principal collaborative biostatistician at KPWHRI, and Susan Reed, MD, MPH, an affiliate researcher at KPWHRI and vice chair of obstetrics and gynecology at Harborview Medical Center, led the study. The research was published in the American Journal of Obstetrics and Gynecology.

“There were previously no estimates for incidence in the U.S., and prevalence estimates varied a lot,” Reed said. “Understanding how many people are impacted by this condition and how many new cases we are seeing can help guide health care organizations in directing resources for treatment and set priorities for further research.”

Findings from Kaiser Permanente data

The study used data from Kaiser Permanente Washington patients who identified as female in the health record and were between ages 16 and 40. The researchers found that incidence of PCOS was 42.5 per 10,000 person-years, meaning that during a year, there would be 42 new PCOS cases among every 10,000 people in this population. The prevalence of PCOS was 5.2%.

“Without any prior estimates for incidence in the U.S., we only had a study from the U.K. to compare these numbers to,” Yu said. “Our estimated incidence was between 4 and 5 times higher than those U.K. estimates. And the prevalence in our study was almost double previous estimates for the U.S.”

Prevalence was higher among people who identified as Hawaiian and Pacific Islander (7.6%), Native American (6.9%), and Hispanic (6.8%) in the health record. The researchers pointed out that since race and ethnicity are social constructs and do not impact biology, the variation suggests differences among doctors or care teams in screening for PCOS or differences in how people share their symptoms. In this study, the clinicians who were most likely to diagnose PCOS specialized in primary care, obstetrics and gynecology, endocrinology, or dermatology.

The rate of new cases was stable over the study period but shifted by age. New cases increased in the 16- to 20-year-old age group and decreased in the 26- to 30-year-old age group over the study period, which looked at data from 2006 to 2019.

“We know that many people are affected, and generally symptoms appear after puberty. The increasing incidence rate in a younger age group could indicate that people are getting a diagnosis earlier, or it could mean that they are getting the condition when they are younger,” Reed said. “If we can treat PCOS sooner and decrease the chronic health effects, that means we have an opportunity to reduce the impact on patient’s lives in the long term.”

There are no therapies targeting the causes of PCOS that are approved by the Food and Drug Administration, she said, so current treatment generally consists of lifestyle changes or medications to manage symptoms.

Identifying “true diagnoses”

Instead of simply using the diagnostic codes in the health record to identify new cases for their incidence estimates, the researchers looked at additional data in the medical record on 700 individuals — such as symptoms reported by a patient or their care team — to group cases as definite, probable, or possible new cases. They also identified prevalent cases that had been misclassified as new cases, where there was evidence of a diagnosis before the code appeared in the health record. Cases that couldn’t be classified were excluded from the final estimates.

“We were trying to get at whether these were true diagnoses,” said Yu. “This should be the gold standard for these kinds of studies, because when you use only the diagnosis codes or self-reporting, without validating a diagnosis using other data, there’s a higher likelihood of overestimating.”

In addition to guiding health care organizations to direct resources to diagnosing and treating PCOS, the researchers hope that having these new estimates will prompt further research to better understand the condition.

“PCOS is still kind of mysterious,” Reed said. “If we can unravel the mysteries of this condition, we can benefit the lives of people everywhere who have this diagnosis.”

KPWHRI coauthors on the study are Jennifer Covey, BS; Ann Kelley, MHA; Jane Grafton, BA; David Cronkite, MS; and Erika Holden, BA. This study was funded by Bayer AG.

By Amelia Apfel