The aim of this thesis was to strengthen understanding of self-reported sleep in pregnancy by drawing together evidence from: previously published research; de novo analyses of existing and novel datasets; and the lived experience of pregnant women themselves. This mixed- methods approach found that: a lack of standardisation and potential flaws in the design of previous studies do not yet permit a formal meta-analysis to be performed; and previous findings remain vulnerable to error and publication bias.
The three de novo quantitative analyses of self-reported sleep conducted for this thesis sought to address many of the flaws in previous research. These analyses indicate that: several pre- existing/pre-pregnant sociodemographic and health characteristics contribute to the less favourable sleep commonly reported by pregnant women (as compared to age-matched non- pregnant women); and that variation in these and (un)related lifestyle and behavioural factors during pregnancy also contribute to variation in self-reported sleep amongst pregnant women. However, the last of these analyses provides evidence that variation in a commonly experienced phenomenon (glucose intolerance and, at its extreme, gestational diabetes) is associated with less favourable sleep in what appears to be a dose-response relationship.
Analysis of posts to web-based forums by women with first-hand experience of sleep in pregnancy confirm that pregnancy-specific somatic changes were experienced/understood to be the principal causes of less favourable sleep; although the advice offered to others facing similar problems tended to focus on behavioural and situational factors as suitable avenues for intervention.
On the basis of this evidence, it is clear that none of the self-administered sleep instruments/items available, and used, to-date are capable of comprehensively assessing the sleep of pregnant women. Future research must develop/use a dedicated sleep instrument to improve our understanding of the range, prevalence and likely determinants of the less favourable sleep more commonly reported by pregnant women.
