The goal of this observational study is to examine the relationship between severe perinatal maternal mental illness and stress reactivity and general development of children aged 6 to 8 years. Mothers were previously recruited and assessed during pregnancy and their babies were assessed at 6 days, 8 weeks and 12 months post delivery. This follow-up will assess the occurrence of any psychiatric episodes (e.g., depression, mania, psychosis) in the intervening period. Assessments of the children will include evaluation of their response to a mildly stressful procedure, as well as their physical, cognitive and social-emotional development, and their mental health. Possible factors that may moderate the relationship between the mother's perinatal mental health and the child's stress reactivity and development, including the home environment, adversity, parent-child interaction and maternal and paternal (or co-parent) mental health will also be assessed.

The principle research question is "does maternal stress system in pregnant women at risk of postpartum psychosis modify stress reactivity in children aged 6 to 8"?.

The "stress system" the investigators are studying is the endocrine system known as the hypothalamic-pituitary-adrenal (HPA) axis. The investigators measured levels of hormones from this system in blood and saliva in a sample of women at risk and not at risk of postpartum psychosis during pregnancy and measured cortisol in the saliva of their babies, before and after a stressful experience at 6 days, 8 weeks and 12 months of age.

The investigators found that infants born to women at risk of postpartum psychosis (PP) who relapsed in the first 4 weeks post-delivery had altered HPA axis activity at 6 days post-birth, compared with infants born to women at risk who remained well in the postpartum period.

The investigators now plan to examine whether levels of mothers' stress hormones associated with risk of postpartum psychosis during pregnancy are also associated with longer-lasting children's cortisol stress reactivity and diurnal rhythm of cortisol during middle childhood (ages 6 to 8).

Genetic material will be extracted from the child's specimens to look at genes, epigenetics (DNA methylation) and changes in gene expression, specifically, changes in genes which might be relevant to the development of stress. Assessments of maternal endocrine and maternal and paternal / co-parent mood measures will be undertaken. Children's cortisol response to stress and cortisol diurnal rhythm will be carried out at ages 6 to 8 years, together with assessments of their growth, cognitive development, social-emotional functioning and mental health. The father or co-parent mental health, their relationship with the mother and their interaction/relationship with the child will also be measured to investigate whether these moderate any effects of maternal perinatal mental health on child development.

Prospective participants (n=72) were recruited in pregnancy, to take part in a study entitled "Risk factors of perinatal mental disorders: stress, electrophysiological and neuroimaging markers", also known as the Psychiatry Research and Motherhood - Psychosis (PRAM-P) study. Clinical assessments were undertaken in pregnancy and, following delivery, assessments of the infant and mother were undertaken at 6 days, 8 weeks and 12 months post delivery. At the final visit, 67 participants gave their written consent to be contacted at some time in the future to discuss the possibility of participating in a follow-up study.

Inclusion Criteria:

• Women who participated in the previous study (Risk factors of perinatal mental disorders: Stress, electrophysiological, and neuroimaging markers), who gave consent to be contacted again, together with their offspring, and the child's father/co-parent.

Exclusion Criteria:

• Women who participated in the previous study who refused consent to be contacted again.

The study comprised 2 groups: women at risk of postpartum psychosis during pregnancy (exposed group) and women with no current or past history of psychiatric disorders (unexposed group).

Cases were women at risk of postpartum psychosis because of a DSM-IV diagnosis of bipolar disorder, schizoaffective disorder or previous postpartum psychosis and controls were women who did not have a current or past history of psychiatric disorders. Clinical assessments were undertaken in the second and third trimesters of pregnancy and, following delivery, assessments of mother and infant were undertaken at 6 days, 8 weeks and 12 months post-delivery (n=72). At this stage, participants were asked to provide written consent to be contacted in the future about the possibility of taking part in a follow-up study.

Study data will be analysed and results will be submitted for publication; participant identity will not be revealed. Study data will be retained and may be used in future studies, if this happens, further Research Ethics Committee approval will be sought.

Authorised persons such as researchers, sponsors, regulatory authorities and Research and Development audit will have access to view identifiable data, for monitoring of the quality of the research.

Study data will be retained for 20 years after completion of the study; and will be disposed of securely.