Degree Name

Doctor of Philosophy (Clinical Psychology)


Department of Psychology


The aims of the study were to explore the prenatal predictors of postnatal distress and depression and to empirically test an attachment derived model of postnatal distress. This is a longitudinal prospective study that used a convenience sample of 133 pregnant women in the community. Postnatal distress was defined as a dysphoric state, measured by anxiety and depression following childbirth. A number of hypotheses and questions of interest were posited. The prenatal predictors of postnatal distress were found to be general symptomatology, mother idealization (at 6 weeks postpartum), and previous depression (at 6 months postpartum); 52% of clinical cases and 95% of non-clinical cases were predicted with an overall correct prediction of 84%. Postnatal distress (6 weeks postpartum) accounted for 22% of the variance of postnatal clinical disorders (6 months postpartum). The prevalence of postnatal minor and major depression combined was 17.5%. Highly anxious mothers tended to have a previous history of psychological problems, an unplanned pregnancy and were not breastfeeding.

The variables selected were consistent with attachment concepts and previous research findings. The relationships between a vulnerable sense of self (i.e. low self-esteem, high preoccupation with relationships, high mother idealization, low attachment, and low mother acceptance), birth experience and postnatal distress were explored using structural equation modeling. Attachment theory indicates that resilience or vulnerability can predict distress and mental illness when people are under conditions of stress. Several hypotheses were tested, employing a predictable stressful event (childbirth) and using measures of postnatal distress (at 6 weeks and 6 months) and clinical status (6 months). Vulnerability and resilience were defined as conditions that promote maladaptation while protective factors promote competent adaptation, even in difficult circumstances. Protective factors may provide a compensatory mechanism or influence at different levels of risk. Consequently, protective factors mediate the influence of risk factors in a dynamic interaction. Partial support was found for the resilience-vulnerability model that offered marginally acceptable levels of fit. A vulnerable sense of self predicted higher postnatal distress while a resilient sense of self predicted low postnatal distress. Birth experience was disconfirmed as a mediator between sense of self and postnatal distress. Unplanned pregnancy predicted higher levels of postnatal distress. Obstetric loss had an indirect negative impact on postnatal distress.