Sleep disturbances in women with and without polycystic ovary syndrome in an Australian National Cohort

• Published in: Clinical endocrinology (Oxford) 2019-04, Vol.90 (4), p.570-578
• Main Authors: Mo, Lin, Mansfield, Darren R., Joham, Anju, Cain, Sean W., Bennett, Christie, Blumfield, Michelle, Teede, Helena, Moran, Lisa J.
• Format: Article
• Language: English
• Subjects: Body mass index; Comorbidity; Mental depression; Polycystic ovary syndrome; Sleep; Womens health
• ISSN: 0300-0664; 1365-2265
• DOI: 10.1111/cen.13922

Summary Background Polycystic ovary syndrome (PCOS) is a common condition in reproductive‐aged women. Sleep disturbances may be more prevalent in PCOS. It is not known if this relationship is independent of other factors. Aim To examine the prevalence of sleep disturbances in a large community‐based cohort study in women with and without PCOS and its relationship to clinical, demographic and comorbid factors. Methods We examined data from survey 5 (2009) of the Australian Longitudinal Study on Women's Health (n = 6578, n = 484 PCOS and n = 6094 non‐PCOS). Sleep duration and disturbances were self‐reported. Three classes of sleep pattern were derived during latent class analysis (normal sleep duration with average sleep, normal sleep duration with sleep symptoms and short sleep duration with sleep symptoms) and compared between women with and without PCOS using multivariate regression, adjusting for body mass index (BMI), depressive symptoms, demographic and comorbid factors. Results Women with PCOS had similar sleep duration but were more likely to experience difficulty sleeping often (RRR 1.67, 1.20‐2.33, P = 0.003) and sometimes (RRR 1.39, 1.07‐1.80, P = 0.015), with restless sleep reported occasionally (RRR, 1.35 1.00‐1.83, P = 0.049). They reported severe tiredness often (RRR 1.48, 95% CI 1.08‐2.04, P = 0.016) and described more sleep difficulties within the last 12 months (OR 1.29, 1.04‐1.60, P = 0.018) on adjusted analyses. Compared to the class of average sleep duration with no sleep disturbances, PCOS was associated with increased relative risk of having average sleep duration with sleep symptoms (RRR 1.40, 95%CI 1.11‐1.77, P = 0.004) and short sleep duration with sleep symptoms (RRR 1.46, 95%CI 1.07‐1.99, P = 0.016) on adjusted analyses. Conclusion Sleep disturbances are more prevalent amongst women with PCOS after adjusting for BMI, depressive symptoms, demographic and comorbid factors. Targeted screening and management of sleep disturbances is warranted in PCOS.