Exploratory Path Analysis of the Relationships between Maternal Mental Health, Parenting Stress, and Children's Functioning

Objective: Maternal mental health can impact many domains of child development and negatively impact the parent-child relationship, which is essential for children’s healthy functioning.  For example, empirical studies have documented that parental depression and stress levels can lead to higher levels of internalizing and externalizing behaviors.  However, few studies have explored the additional impact of parental trauma symptoms on young children.  A trauma history in the mother has the potential to create many serious symptoms, which can interact with depression and relate to negative outcomes in her children.  The objective of this paper is to examine the relationships between maternal trauma symptoms, depressive symptoms, and parenting stress and strain, and how those factors relate to their children’s functioning at intake for mental health treatment.

Methods: Data were collected as part of an evaluation a SAMHSA-funded project to establish a system of care for young children and families needing mental health services who were involved or at risk of involvement with child welfare.  At intake into services, an extensive packet of measures was administered to evaluate functioning on multiple levels for both child and parent.  For this analysis we included data on maternal functioning using the Beck Depression Inventory (BDI), the Harvard Trauma Questionnaire (HTQ), Parenting Stress Index, and the Caregiver Strain Questionnaire (PSI) and the child, the Child Behavior Checklist (CBCL). The final sample consisted of 91 biological mothers and their children seen at intake. Path analysis using SAS proc calis was used to assess the relationships between maternal mental health symptoms, stress levels, and children’s functioning.

Results: Results from the path analysis revealed that higher scores on the HTQ were directly associated with higher scores on the BDI) (p<.0001, which were then linked with higher scores on the PSI and the CGSQ (p<.001) and children’s higher externalizing behaviors on the Child Behavior Checklist (CBCL) (p<.05). However, the link between higher maternal HTQ scores and children’s externalizing behaviors on the CBCL appeared to be fully mediated by maternal BDI, PSI, and CGSQ scores, whereas maternal trauma symptoms on the HTQ appear to be directly related to children’s higher internalizing CBCL scores (p<.05), even after accounting for maternal depressive symptoms and stress/strain. We also found that several control variables, including maternal age, Hispanic ethnicity, and child welfare involvement were significantly related to maternal stress in the parent-child relationship.

Conclusions and Implications: This study shows that maternal trauma symptoms differ in how they relate to children’s internalizing and externalizing behaviors, and that targets for intervention must differ based on these two paths. Intervening to address parental stress and strain may address the relationship between maternal trauma and depressive symptoms and children’s externalizing problems, but maternal trauma symptoms continue to relate to children’s higher internalizing problems. Future studies should examine possible mediators by which maternal trauma symptoms impact children’s internalizing problems, and should examine the extent to which existing dyadic treatments for young children target maternal mental health symptoms and mothers’ experiences of parental stress and strain.