The research team, led by Billy Fitzgerald with colleagues Christopher McCusker, Maria Dempsey and Audrey Dunn Galvin, recruited 117 caregivers of autistic children aged 4 to 12 years who were enrolled in mainstream education. Roughly two-thirds of the children were male, reflecting the well-documented gender skew in autism diagnoses. Caregivers completed a battery of validated questionnaires covering five distinct family processes: parenting stress, parental mental health problems, marital satisfaction, family cohesion and family conflict. They also reported on child characteristics including age, gender, autistic traits, comorbid diagnostic status and sleep quality, provided family structural information such as household income, caregiver education and family composition, and rated their satisfaction with the family’s relationships with healthcare and educational professionals.

Child outcomes were measured using the Strengths and Difficulties Questionnaire, a widely used screening instrument that captures emotional symptoms, conduct problems, hyperactivity, peer problems and prosocial behavior, alongside caregiver ratings of academic performance. The analytical strategy was deliberately staged. The researchers first ran exploratory bivariate analyses to identify which individual variables were associated with outcomes, and then used hierarchical regression models to test the unique and combined contributions of family processes after accounting for child and systemic factors. This approach allowed them to partition the variance in outcomes across the different levels of the child’s ecology, which is precisely the kind of multilevel accounting the bioecological model demands.

The headline finding concerns behavior problems. Two variables emerged as significant individual predictors: caregiver and parenting-related stress, and child sleep quality. Poorer sleep and higher parenting stress both tracked with more behavioral difficulties. But when the researchers looked at the collective contribution of the family process variables as a block, family processes accounted for more of the variance in behavior problems than any other set of factors. In other words, the emotional temperature of the household, the mental health of the parents, the quality of the couple relationship and the degree of cohesion or conflict in the family together formed the strongest statistical signal in the data.

The picture changed sharply for competence-related outcomes. Prosocial behavior, the capacity to engage positively with peers and adults, was predicted by autistic traits alone: children with higher trait severity showed lower prosocial scores. Academic performance, meanwhile, was predicted only by the presence of a comorbid diagnosis, meaning children with an additional condition such as ADHD or a language disorder tended to perform worse academically. Family processes, family structure and satisfaction with professionals did not add predictive power for these outcomes once child characteristics were taken into account. The study thus draws a clean statistical dividing line: the family shapes how autistic children behave, but child-level characteristics dominate how they perform and connect.

That division has a compelling logic rooted in developmental science. Behavior problems in childhood are highly stress-reactive; they fluctuate with the emotional climate a child inhabits day to day. Sleep is a particularly interesting variable here because it sits at the intersection of child and family. Sleep difficulties are among the most common co-occurring conditions in autistic children, and disrupted sleep is known to amplify emotional and behavioral dysregulation in neurotypical children as well. At the same time, a child’s chronic sleep problems feed back into parental stress, creating a bidirectional loop in which exhausted parents and sleep-deprived children each worsen the other’s state. The bioecological framework explicitly models these reciprocal processes rather than treating the child as a fixed set of traits acting on a passive environment.

The null findings for systemic factors deserve equally careful attention. Family satisfaction with relationships with healthcare and educational professionals did not predict behavioral or academic outcomes in the final models, and neither did household income, caregiver education or family composition. This is not to say that professional partnerships and socioeconomic resources are irrelevant; a substantial literature links family–professional partnership quality to family quality of life and service adequacy. But in this sample, once family processes and child characteristics were modeled, the systemic layer added little explanatory power for the specific outcomes measured. For clinicians and educators, the implication is that interventions targeting only the professional interface or only material resources may miss the more proximal levers of change.

The practical implications are concrete. Parenting stress emerged as a central, modifiable target, and there is an established evidence base for addressing it: randomized trials of structured parent training have shown meaningful reductions in behavioral problems in autistic children, and mindfulness-based stress reduction programs for caregivers of children with developmental disabilities have demonstrated reductions in parental distress. Sleep interventions, including behavioral sleep programs tailored to autistic children, represent another tractable avenue, given that sleep quality independently predicted behavior problems in this study. The findings suggest that family-level supports should not be treated as peripheral add-ons to child-focused services but as core components of any strategy aimed at improving behavioral outcomes for autistic children in mainstream schools.

The study also carries implications for how schools interpret difficult behavior. If a significant share of the variance in behavioral difficulties is attributable to family processes, then classroom behavior may often function as a barometer of family stress and child sleep rather than as a straightforward reflection of classroom management or autism severity. This reframing does not absolve schools of responsibility for creating supportive environments, but it argues against locating the entire problem, and the entire solution, within the school gates. A coordinated response that includes screening for parental stress and sleep problems, and offering family-based support where indicated, may be more effective than behavior plans alone.

As the first wave of a longitudinal study, this research captures a snapshot, and the authors frame it accordingly. Cross-sectional regressions identify associations and relative contributions at one point in time; they cannot establish that family processes cause behavior problems or that improving sleep will definitively reduce them. The longitudinal design will allow the team to test whether family processes measured now predict change in outcomes over time, which would substantially strengthen the causal case. Sample size is another consideration, as 117 caregiver–child dyads limit the number of predictors that can be modeled simultaneously and the precision of the estimates. Even so, the study offers one of the clearest head-to-head comparisons to date of family, child and systemic influences on autistic children’s outcomes in mainstream education, and its central message is hard to ignore: to understand and support autistic children’s behavior, look first to the family system that surrounds them.

Subject of Research: Multilevel influences of family processes, child characteristics and systemic factors on behavioral and academic outcomes of autistic children in mainstream education

Article Title: Relative Impact of Family Processes and Child and Systemic Factors on Behavioral and Academic Outcomes for Autistic Children in Mainstream Education

Article References: Fitzgerald, B., McCusker, C., Dempsey, M., & Galvin, A. D. (2026). Relative Impact of Family Processes and Child and Systemic Factors on Behavioral and Academic Outcomes for Autistic Children in Mainstream Education. Journal of Autism and Developmental Disorders. https://doi.org/10.1007/s10803-026-07362-9

Image Credits: AI Generated

DOI: 10.1007/s10803-026-07362-9

Keywords: autism, autistic children, family processes, parenting stress, sleep quality, behavior problems, academic performance, mainstream education, bioecological model, Strengths and Difficulties Questionnaire, family-professional relationships, child mental health