Impact of a Diagnosis of Polycystic Ovary Syndrome on Diet, Physical Activity and Contraceptive Use in Young Women

Findings From the Australian Longitudinal Study of Women's Health

Tessa Copp; Erin Cvejic; Kirsten McCaffery; Jolyn Hersch; Jenny Doust; Ben W. Mol; Anuja Dokras; Gita Mishra; Jesse Jansen


Hum Reprod. 2020;35(2):394-403. 

In This Article

Abstract and Introduction


Study Question: Do diet, physical activity and contraceptive use change after receiving a diagnosis of polycystic ovary syndrome (PCOS)?

Summary Answer: Using longitudinal data 12 months apart, young women newly diagnosed with PCOS were more likely to stop using contraception but did not change their physical activity or vegetable intake.

What is Known Already: Diagnostic criteria for PCOS have widened to capture more women, despite limited evidence of the benefits and harms. Possible benefits of a PCOS diagnosis are that it may help women with family planning and motivate them to implement healthy lifestyle changes to reduce the reproductive, metabolic and cardiovascular risks associated with PCOS. However, there are no empirical studies investigating how women respond to a diagnosis of PCOS with respect to their health behaviour, and longitudinal population-based studies are lacking.

Study Design, Size, Duration: This is a longitudinal analysis of two waves of data collected 12 months apart from the cohort born 1989–1995 in the Australian Longitudinal Survey on Women's Health, a population-based cohort study. Women in this cohort were first surveyed in 2012–2013, aged 18–23 years.

Participants/Materials, Setting, Methods: Women who responded to the 2014 survey (aged 19–24, n = 11 344) and 2015 survey (aged 20–25, n = 8961) were included. Using logistic regression, multinomial logistic regression and linear regression, change in vegetable intake, physical activity and contraceptive use were compared for women newly diagnosed with PCOS to women not reporting a diagnosis of PCOS. Changes in psychological distress and BMI were also examined.

Main Results and the Role of Chance: Young women reporting a new diagnosis of PCOS were no more likely to increase their vegetable intake or physical activity than women not reporting a PCOS diagnosis. Women newly diagnosed with PCOS were 3.4 times more likely to stop using contraception during the 12-month study period than women without PCOS (14% versus 4%, 95% CI = 2.3 to 5.1, P < 0.001). This difference remained significant after controlling for demographics, chronic conditions associated with PCOS, endometriosis, BMI and psychological distress (P < 0.001).

Limitations, Reasons for Caution: All data was self-reported including PCOS diagnosis, assessment of diet quality was limited to vegetable intake only. The exact timing of diagnosis within the 12-month period and whether the women intended to conceive are unknown. The number of women reporting a new diagnosis of PCOS was also relatively small.

Wider Implications of the Findings: These findings suggest that a diagnosis of PCOS may not produce short-term benefits by way of improving health behaviour. The observed reduction in contraception use suggests some women may be at increased risk of unplanned pregnancies, highlighting the importance of counselling about contraceptive needs. Both potential benefits and harms must be considered when determining the appropriateness of a PCOS diagnosis.

Study Funding/Competing interest(s): The Australian Longitudinal Study on Women's Health is funded by the Australian Government Department of Health. BWM reports consultancy for ObsEva, Merck, Merck KGaA and Guerbet. No further competing interests exist.

Trial Registration Number: N/A


Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in reproductive-aged women and is associated with adverse reproductive, metabolic and psychological outcomes (Dumesic et al., 2015). The Rotterdam diagnostic criteria (2003) expanded the initial National Institute of Health criteria for PCOS (1991) by including polycystic ovarian morphology (The Rotterdam ESHRE/ASRM PCOS Workshop, 2004), including women with milder phenotypes and increasing the number of women diagnosed (Skiba et al., 2018). This expansion aroused considerable debate, as polycystic ovaries using the 2003 Rotterdam definition are also commonly found in women without PCOS (Duijkers and Klipping, 2010; Lauritsen et al., 2014), raising concerns about overdiagnosis (Copp et al., 2017). The rationale for the expanded criteria was to capture the broad clinical expression of PCOS (Dumesic et al., 2015). It is posited that knowledge of a PCOS diagnosis might motivate women to engage in recommended lifestyle changes (Banting et al., 2014), reducing the potential risks of long-term consequences, such as insulin resistance, obesity and type II diabetes (Cooney and Dokras, 2018; Gibson-Helm et al., 2017). Additionally, a diagnosis of PCOS may improve access to treatment, such as the oral contraceptive pill to help manage hormonal symptoms (e.g. irregular menstruation, acne, hirsutism) and reduce the risk of endometrial carcinoma (Teede et al., 2018). However, the criteria have expanded despite a lack of clear evidence of meaningful benefits in terms of reduced risk of adverse outcomes for those newly diagnosed. Additionally, neither the potential benefits nor harms of this expanded definition have been rigorously investigated (Copp et al., 2017).

Whether a PCOS diagnosis actually results in healthier behaviour remains unclear. The limited research available suggests that a PCOS diagnosis could positively impact diet but also negatively impact perceived weight control and potentially even lead to disordered eating (Lin et al., 2018; Moran et al., 2017). A cross-sectional study of women aged 31–36 years in the community found that women with PCOS were more likely to use both healthy (e.g. reducing meal size, fat or sugar intake) and maladaptive (e.g. use of laxatives, fasting or diuretics) weight loss methods than women without PCOS (Moran et al., 2017). Studies examining weight-related beliefs have also found that women with PCOS are aware that they are at higher risk of obesity and associated diseases (Lin et al., 2018; Moran et al., 2010) but perceive poorer control over their weight and fewer benefits of healthy behaviours on weight gain compared to women without PCOS (Lin et al., 2018). Conversely, studies in other conditions suggest that communicating personalised risk information or giving a chronic disease diagnosis has no impact on health-related behaviour (Dontje et al., 2016; French et al., 2017). Longitudinal research is needed to examine the impact of a PCOS diagnosis on health behaviour.

Despite hormonal contraceptives being a first-line treatment for regulating hormonal aspects associated with PCOS (Legro et al., 2013; Teede et al., 2018), previous cross-sectional studies of women aged 18–23 and aged 28–33 have found that women with PCOS are less likely to report using contraception than their peers (Joham et al., 2014; Rowlands et al., 2016). However, the extent to which this is due to diagnosis of the condition is unclear, as contraceptive use prior to diagnosis was not assessed. Further investigation of change in contraceptive use after a diagnosis of PCOS is warranted.

Given the limited knowledge of the impact of a PCOS diagnosis on health behaviour, the purpose of this study was to examine whether there were significant changes in young women's behaviour after receiving a diagnosis of PCOS, compared to women without PCOS. This paper uses a national, longitudinal cohort study of young Australian women to examine changes in vegetable intake, physical activity and contraceptive use over a 12-month period comparing women newly diagnosed with PCOS to women without PCOS. Changes in BMI and psychological distress were also examined as secondary outcomes.