Maternal mental health impacting children

Our fourth task was to examine how the five identified maternal mental health trajectory groups predicted children’s mental health and cognitive and social development at the age of 7-8 years. Among a subsample (N = 102), we also analyzed differences in children’s physiological-hormonal stress regulation, measured by cortisol levels and diurnal patterns at the age of 10-11 years. This analysis included children from three maternal trajectory groups: stable low levels of mental health symptoms (n = 72) and prenatal (n = 15) and early postpartum (n = 15) mental health problems.

7.5.1 Child mental health and development at 7-8 years (Article I)

Maternal mental health trajectory group predicted child internalizing symptoms (partial ² = .03), but not externalizing symptoms. Pair-wise comparisons specified that the children of mothers with early postpartum (III) or heterogeneous high levels of (V) mental health problems showed more internalizing symptoms than the children of mothers with stable low levels of symptoms (I). Parity and current T4 maternal GHQ-psychological distress were significant covariates for both internalizing and externalizing symptoms. Maternal mental health trajectory group predicted also children’s cognitive developmental executive functions (partial ² = .02) and memory (partial ² = .03). Pair-wise comparisons specified that the children of mothers with heterogeneous high levels of problems (V) showed more problems in executive functions than the children of mothers with stable low levels of symptoms (I) or late postpartum mental health problems (IV). They also showed more memory problems than children of mothers in any other group. Both parity and current T4 maternal GHQ-psychological distress were non-significant covariates for child cognitive development. Children’s social development was not predicted by maternal mental health trajectory group. The current T4 maternal GHQ-psychological distress was a significant, but parity a non-significant covariate for child social development. Furthermore, neither child’s gender nor family’s fertility history moderated the effect between maternal mental health trajectory group and children’s mental health or social or cognitive development. In other words, mothers’ mental health similarly predicted child internalizing and externalizing symptoms and cognitive and social developmental problems a) among boys and girls, and b) in normative families and in those with former infertility.

In line with our hypotheses, children of mothers with chronic and severe course of symptoms displayed the highest level of mental health symptoms and developmental problems at the age of 7-8 years. However, the results did not support our second hypothesis. We had assumed that children of mothers with prenatal problems would display a higher level of mental health symptoms and developmental problems compared to children of mothers with postpartum problems or those without mental health problems. Our results showed instead that children of mothers with early postpartum mental health problems had an increased vulnerability to internalizing mental health problems.

7.5.2 Child stress regulation at 10-11 years (Article III)

Results showed no differences between the three groups of children in the average daily cortisol levels. Cortisol levels differed, however, in some single point measures:

at awakening (C1) and at 30 minutes after awakening (C2). Pair-wise comparisons specified that children of mothers with prenatal mental health problems (II) had at awakening (C1) lower cortisol levels than children of mothers with stable low levels of symptoms (I), whereas at 30 minutes after awakening (C2) they had higher cortisol levels than children of mothers with early postpartum mental health problems (III), suggesting a steeper increase of cortisol from awakening to 30 minutes after awakening among children of mothers with prenatal mental health problems (II) as compared to other children.

Figure 6 presents diurnal cortisol patterns in the three groups of children.

Results showed fluctuation across day within each group: children of mothers with stable low levels of symptoms (I), children of mothers with prenatal mental health problems (II) and children of mothers with early postpartum mental health problems (III). Pair-wise analyzes showed that among children of mothers with stable low levels of symptoms (I), cortisol increased from awakening (C1) to 30 minutes after awakening (C2) and decreased thereafter towards one hour after awakening (C3), late afternoon (C4) and evening (C5). A similar diurnal pattern was found among children of mothers with prenatal mental health problems (II). In contrast, children of mothers with early postpartum mental health problems (III) showed no change in cortisol from awakening (C1) to 30 minutes after awakening (C2) or from 30 minutes after awakening (C2) to one hour after awakening (C3). There was however a decrease of cortisol from one hour after awakening towards late afternoon (C4) and evening (C5) in the early postpartum trajectory group, suggesting a reduced CAR among the children of mothers with early postpartum mental health problems (III).

Cortisol difference scores of the pairs of consecutive time points (C2 – C1, C3 – C2, C4 – C3), including the DCD difference score (C5 – C4), did not differ between the three groups of children. The CAR difference score from awakening to one hour after awakening (C3 – C1) differed however between the groups. Pair-wise comparisons revealed that the difference occurred between the children of mothers with prenatal mental health problems (II) and the children of mothers with stable low levels of symptoms (I): the children of mothers with prenatal mental health problems showed increase in CAR from C1 to C3, whereas the children of mothers with stable low symptoms showed decrease in CAR from C1 to C3. This suggested

a steeper increase in CAR among children of mothers with prenatal problems compared to other children.

To sum it up, our hypothesis concerning a more adverse effect of maternal prenatal compared to postpartum problems on children’s cortisol was not supported. Instead, we found that both maternal prenatal and early postpartum mental health problems impacted children’s diurnal cortisol pattern at the age of 10-11 years, but in different directions. Whereas prenatal problems seemed to shape it towards an intensified CAR, early postpartum problems were associated with a reduced CAR.

Figure 6. Children’s cortisol value medians (microgram/l) according to early maternal mental health trajectory group. N = 102