Actigraphic and self-reported sleep quality in women: Associations with ovarian hormones and mood

Background: Sleep and mood disturbances in women have often been linked to the menstrual cycle, implying an ovarian hormonal causation. However, most studies in this area have used self-reported menstrual cycle phase rather than direct measurement of ovarian hormone concentrations. Further, many studies have focused primarily on peri- and postmenopausal populations reporting clinical sleep difficulty. In this study, we examined the associations among sleep quality, mood, and ovarian hormone concentration in a random sample of community-dwelling, nonclinical women of reproductive age. Methods: Our sample consisted of 19 non-help-seeking women aged 18-43 years, each contributing an average of 39.5 nights of data. Over the 42 days of the study, we collected self-reported and actigraphic sleep-quality data, concentrations of urinary estrogen and progesterone metabolites (estrone-3-glucuronide (E1G) and pregnanediol-3-glucuronide [PdG], respectively), and daily mood ratings. Linear-mixed models were used to estimate associations, clustering longitudinal observations by the participant. Results: We found a significant positive association between Sleep Efficiency and E1G, and a significant negative association between Sleep Efficiency and PdG. Otherwise, the self-reported and actigraphic sleep measures were not associated with ovarian hormone concentrations. Self-reported sleep was strongly associated with mood, whereas actigraphic sleep was associated with only two of the 11 individual mood items, "Feeling on Top of Things" and "Difficulty Coping.". Conclusions: In this community sample of women of reproductive age, ovarian hormones play little, if any, role in day-to-day sleep quality. Our findings additionally highlight the different associations that self-reported and actigraphic sleep show with hormones and mood.