Abstract
Insomnia, short/long sleep duration, and evening-preference were associated with adverse pregnancy/perinatal outcomes in multivariable regression (MVreg). However, observational studies were vulnerable to residual confounding. Randomised controlled trials of interventions to improve sleep were rare in pregnancy; those that had been conducted were small. The aim of this thesis is to explore effects of sleep traits on stillbirth, miscarriage, gestational diabetes, hypertensive disorders of pregnancy, perinatal depression, preterm birth, low/high offspring birthweight (LBW/HBW).The main approach was Mendelian randomization (MR), and the main cohorts were UK Biobank (UKB), Avon Longitudinal Study of Parents and Children, Norwegian Mother, Father and Child Cohort, Born in Bradford, and FinnGen (N=356,069). Two-sample MR used inverse variance weighted as main analysis, and nonlinear MR was used in UKB to explore the nonlinearity of effects of sleep duration. Additionally, MVreg was used in main chapters of insomnia and sleep duration, which were measured in pregnancy in birth cohorts.
For insomnia, two-sample MR supported its effects on increased risk of miscarriage, perinatal depression and LBW, with MVreg also supported unfavourable associations with miscarriage and perinatal depression. For sleep duration, MR provided evidence of nonlinear effects on stillbirth, perinatal depression and LBW, with MVreg also supported nonlinear associations with perinatal depression and LBW. For chronotype, there was two-sample MR evidence of a weak protective effect of morning-preference on miscarriage and detrimental effect on HBW. In stratified analyses, I found the effects of sleep duration on gestational diabetes differed between genetically predicted morning- and evening-preference groups.
This thesis highlights the potential for sleep traits to influence adverse pregnancy/perinatal outcomes. Further MR studies with larger number of cases for pregnancy and perinatal outcomes, and participants of non-European ethnic background are needed to validate my findings, which could then lead to well-designed randomised controlled trials of interventions to improve sleep in pregnancy.
| Date of Award | 27 Sept 2022 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Sponsors | China Scholarship Council |
| Supervisor | Debbie A Lawlor (Supervisor), Kate M Tilling (Supervisor), Eleanor C M Sanderson (Supervisor) & Maria C Borges (Supervisor) |