The factors predicting stress, anxiety and depression in the parents of children with autism

The parents of children with autism have been demonstrated to report significantly higher levels of stress, anxiety and depression than the parents of developmentally normal children. However, the factors predicting stress, anxiety and depression in this parental group remains poorly understood. The present study examined the variables predicting stress, anxiety and depression in the mothers and fathers of children with autism, and tested the validity of a path model describing the relationship between these variables. Three separate studies were carried out, the first focused on mothers of children with autism, the second focused on fathers of children with autism, and the third assessing model fit. Mothers (N=250) and fathers (N=229) of children with autism aged 4 to 17 years 11 months completed an on-line questionnaire measuring social and economic support, psychological distress, perceived parent-child attachment, parental locus of control, autism symptom severity and child externalizing behaviours. Stepwise regression analysis was used to examine the relationship between independent and dependent variables. The results of Study one, which focused on the mothers of children with autism, suggested a different pattern of predictive variables for stress, anxiety and depression in this maternal group. Aggressive Behaviour, Social Support and Parental Locus of Control significantly predicted maternal depression; whereas Mother's Age, Autism Symptom Severity and Perceived Limit Setting Ability significantly predicted maternal anxiety. The predictive model for maternal stress was a combination of the predictive models for maternal anxiety and depression. In contrast, the predictive model for fathers of children with autism, as investigated in Study two, was consistent across dependent variables. Social Support and Perceived Limit Setting Ability were the primary predictors for paternal stress, anxiety and depression. Conduct Problems, a variable identified in the existing literature as predictive of stress, anxiety and depression in the parents of children with autism, was not a significant predictor in any of the six stepwise regression models. The results indicated that the relationship between 'child-centric variables' (i.e. externalizing behaviours and autism symptom severity) and parental mental health problems may be mediated by other variables. The results of the stepwise regression analyses formed the rationale for a pathway model describing the relationship between the variables, which was assessed for statistical fit with the observed data in Study three. The model positioned parental cognitions and socio-economic support as a mediator of the relationship between 'child-centric variables' and parental distress. Confirmatory Factor Analysis was used to assess the fit of the model with the observed data. The model was shown to be a good fit with the data for both mothers and fathers. Invariance testing, using the Satorra-Bentler chi-square difference test, demonstrated support for metric invariance for the model across gender. The results of the study were used to propose changes to the existing support services offered to parents of children with autism, and the consideration of a more holistic approach, combining psychotherapeutic support for the parent with behavioural management programs related to the child.