(PS10-98) Prenatal Obsessive-compulsive Symptoms Predict Insomnia Severity During the Perinatal Period

The perinatal period is a time of significant risk for developing insomnia. As many as two-thirds of pregnant women report significant insomnia symptoms, which is linked with pre-term birth, gestational hypertension, and maternal and child psychopathology. Postpartum insomnia is also linked with poor mental health outcomes for new mothers. Thus, it is crucial to identify risk factors to facilitate early detection and provide targeted interventions. While previous research has implicated prenatal mental health symptoms such as anxiety and depression as precipitating factors for developing insomnia, other common and distressing prenatal mental health symptoms have yet to be studied. Thus, the present study aimed to examine whether prenatal obsessive-compulsive (OC) symptoms contribute to the prediction of subsequent insomnia severity across the prenatal and postpartum periods above and beyond anxiety and depression symptoms.  

As part of a 2-site longitudinal study on the course of perinatal obsessive-compulsive symptoms, 175 childbearing women (18 years or older) completed the Obsessive-Compulsive Inventory-12 item version (OCI-12), Perinatal Anxiety Screening Scale (PASS), Edinburgh Postnatal Depression Scale (EPDS), and Insomnia Severity Index (ISI) at 20 weeks and 34 weeks pregnant, as well as 6 months postpartum. We computed hierarchical linear regressions using ISI, PASS, EPDS, and OCI-12 scores at 20 weeks pregnant to predict ISI scores at 34 weeks pregnant and 6 months postpartum. In our models, ISI scores were entered in Step 1, followed by PASS and EPDS scores in Step 2, and OCI-12 total scores in Step 3. 

The overall regression model predicting ISI scores at 34 weeks pregnant was significant (R² = .37, p < .001). While adding the EPDS and PASS in step 2 did not significantly account for additional variance, adding OCI-12 scores did account for a small but significant increase (∆R² = .02, p = .008). In the final model, only ISI scores (β = .55, p < .001) and OCI-12 scores (β = .23, p = .008) emerged as significant individual predictors. The regression model predicting ISI scores at 6 months postpartum was also significant (R² = .37, p < .001). The EPDS and PASS accounted for significant variance in step 2 (∆R² = .04, p = .004), as did the OCI-12 in step 3 (∆R² = .02, p = .017). However, in the final model, only ISI scores (β = .42, p < .001) and OCI-12 scores (β = .22, p = .017) emerged as significant individual predictors.

When included together in the final models, OC symptoms, not anxiety and depression symptoms, significantly predicted insomnia in late pregnancy and at six months postpartum. The predictive value of perinatal anxiety and depression for postpartum insomnia might therefore be better explained by the presence of distressing obsessions and compulsions. These findings have potential implications for improving perinatal health outcomes, as assessing and addressing OC symptoms early in pregnancy may aid in the identification of women at risk of developing insomnia. Future research should continue to investigate the relationship between insomnia and OC symptoms, especially in the perinatal period.