Document Type


Publication Date



Maine Medical Center, Medical Education, Maine Medical Center Research Institute, Psychiatry

MeSH Headings

Child, Humans, Fathers, Mothers, Problem Behavior, Autism Spectrum Disorder, Male, Female, Mother-Child Relations


Background: Several studies have examined mother reports of problem behaviors in their children with autism spectrum disorder (ASD). However, few studies have examined the relative contributions of mother and father characteristics on child outcomes, largely due to limited father participation in research. Research testing dyadic models simultaneously examining mother and father effects is also lacking.

Objectives: Describe a sample of mothers and fathers and their children with ASD who were hospitalized in a specialized inpatient psychiatric unit due to problem behaviors. Test a dyadic model examining parenting efficacy as a mediator between parental stress and child problem behaviors for both mothers and fathers.

Methods: Fifty mother and father couples with a child hospitalized in a specialized inpatient psychiatry unit were prospectively enrolled in this study. ASD diagnosis of the child was confirmed by the ADOS-2, administered by a research-reliable examiner, and clinical evaluation with DSM-5 criteria. Both parents completed the Aberrant Behavior Checklist Irritability subscale, Parent Stress Index Short Form 4, and Difficult Behavior Self-efficacy Scale at hospital admission. Due to the high correlation (?=.76, p<001) between mother and father reports of child problem behaviors, a combined score was created with the average of the two. A variant of the actor-partner interdependence model (APIM) evaluated the effects of parent characteristics on child problem behaviors. Descriptive statistics were conducted using SPSS v.25, and the APIM was tested using AMOS v.25.

Results: Mothers were on average 42 years old (SD=7.3) and fathers were 44 years old (SD=8.2). The majority of parents were married (86%) biological mothers (82%) and biological fathers (76%). The average child age was 13 years (SD=3.7, range 5-19), 82% were male, 98% were Caucasian, and 94% were non-Hispanic/non-Latino. Average nonverbal IQ score was 71 (SD=28.8, range 30-145), and the mean Vineland Adaptive Behavior Scale II composite score and Vineland expressive communication subscale scores were 55 (SD=14.6) and 7 (SD=4.2) respectively. A slight majority completed ADOS-2 modules 1 or 2 (53%) indicating minimally verbal communication status. APIM results indicated mothers’ parental stress was a significant predictor of mothers’ parenting self-efficacy (?=-.58, p<0.001). Fathers’ parental stress predicted both their parenting self-efficacy (?=-.64, p<0.001) and child problem behaviors (?=.43, p<.01). Fathers’ parenting self-efficacy partially mediated the relationship between their parental stress and child problem behaviors (?=.32, p<.04).

Conclusions: Self-efficacy is the belief that one can produce a desired behavior change. For both mothers and fathers, parental stress reduced their parenting self-efficacy, which can reduce their persistence and effectiveness in managing their child’s behavior. Interventions that reduce stress in mothers and fathers (such as mindfulness training) may enhance their parenting self-efficacy and subsequently improve child behavior. Surprisingly, fathers’ parenting self-efficacy positively predicted child problem behaviors. The fact that only father characteristics predicted child problem behaviors suggests that greater effort should be devoted to collecting data from both parents and evaluating these data using the appropriate dyadic models.


2020 Costas T. Lambrew Research Retreat