Contraception use and pregnancy outcomes in women with polycystic ovary syndrome: data from the Australian Longitudinal Study on Women's Health

STUDY QUESTION Do contraception use, pregnancy outcome and number of children differ in women with and without polycystic ovary syndrome (PCOS)? SUMMARY ANSWER Women with PCOS were less likely to report use of contraception and more likely to report a miscarriage, whilst number of children was similar between groups. WHAT IS KNOWN ALREADY The oral contraceptive pill is used in the management of PCOS, but the patterns of contraception use in women with PCOS is not known. In women with PCOS who undergo assisted reproduction, the risk of pregnancy loss appears higher, yet pregnancy loss and family size among community-based women with PCOS is not known. STUDY DESIGN, SIZE AND DURATION This is a cross-sectional analysis of a longitudinal cohort study. Mailed survey data were collected at five time points (years 1996, 2000, 2003, 2006 and 2009). Data from respondents to Survey 4 (2006), aged 28–33 (n = 9145, 62% of the original cohort aged 18–23 years) were analysed. PARTICIPANTS/MATERIALS, SETTING, METHODS This study was conducted in a general community setting. Data from participants who responded to the questions on PCOS, contraception and pregnancy outcome were analysed. The main outcome measures were self-reported PCOS, body mass index (BMI), contraception use, pregnancy loss and number of children. MAIN RESULTS AND THE ROLE OF CHANCE In women aged 28–33 years, women with PCOS were less likely to be using contraception (61 versus 79%, P < 0.001) and more likely to be trying to conceive (56 versus 45%, P < 0.001), compared with women not reporting PCOS. A greater proportion of women with PCOS reported pregnancy loss (20 versus 15%, P = 0.003). PCOS was not independently associated with pregnancy loss; however, BMI was independently associated with pregnancy loss in the overweight and obese groups (OR 1.2, 95% CI 1.04–1.4, P = 0.02 and OR 1.4, 95% CI 1.1–1.6, P = 0.001, respectively). Fertility treatment use was also independently associated with pregnancy loss (adjusted OR 3.2, 95% CI 2.4–4.2, P < 0.001). There was no significant difference in number of children between women with and without PCOS. LIMITATIONS, REASON FOR CAUTION PCOS, contraception use and pregnancy outcome data were self-reported. Attrition occurred, but is reasonable compared with similar longitudinal cohort studies. WIDER IMPLICATIONS OF THE FINDINGS This community-based cohort aged 28–33 years provides insights into the contraceptive use, pregnancy loss and family size of a large cohor