311: Objectively-measured sleep parameters and antepartum depressive symptoms

Abstract

To determine if objectively-measured sleep disturbances in the second trimester are associated with high depression scores later in pregnancy In a large prospective cohort study of nullipara, participants completed the Edinburgh Postnatal Depression Scale (EPDS) during the first trimester and again at the third study visit (between 22 and 29 6/7 weeks). In the midtrimester, between these depression assessments, a subgroup underwent actigraphy and completed a sleep diary for 7 days. A high EPDS score was defined as an EPDS score > 10 or any thoughts of self harm. Sleep disturbances were defined as follows: short sleep duration < 7 hours, high fragmentation index >75%ile, high wake after sleep onset (WASO) >75%ile, elevated sleep onset latency >30 minutes, low time above light threshold 1000 lux (TALT) <25%ile, low sleep efficiency <80%. Univariable analyses were performed to investigate the association between maternal baseline characteristics, sleep disturbances and high depression scores later in pregnancy. All significant (P<0.05) covariates in univariable analysis were included in multivariable analysis. The study population for this analysis includes 778 women who had visit 3 EPDS scores and who underwent actigraphy assessment. This cohort represented a racially and ethnically diverse group of women with an average age of 27 years (Table). At visit 3, 108 (13.9%) women had high EPDS scores. Univariable analyses demonstrated that these women were significantly more likely to be younger, single, publicly insured, smokers, diagnosed with a mental health condition prior to pregnancy, and to have high EPDS scores at the first study visit as compared to women with low scores at visit 3. Women with elevated sleep onset latency and low-sleep efficiency were significantly more likely to have high EPDS scores at visit 3 (Table). However, in multivariable logistic regression, only preexisting mental health conditions (OR 3.7, confidence interval 2.2-12.0) and high EPDS scores at the first visit (OR 7.7, confidence interval 3.4-17.5) remained associated with high EPDS scores at visit 3. Objectively-measured sleep disturbances in the middle of pregnancy were not independently associated with high depression scores later in pregnancy.