Association of Child and Family Attributes With Outcomes in Children With Autism

Importance: The prevalence and attributes of positive outcomes (or doing well) among children with autism spectrum disorder (ASD) in midchildhood are not well known.

Objective: To estimate the prevalence of doing well according to metrics of proficiency and growth and to investigate the extent to which significant associations exist between child- and family-level variables and doing well.

Design, Setting, And Participants: This longitudinal cohort study included children with ASD from regional clinics across Canada. Participants were sampled 3 times between ages 2 and 4.9 years (T1) and twice in follow-up into middle childhood (T2). Data were analyzed March 2018 through January 2020.

Exposures: Language and IQ assessments at first sample; household income, parent coping, and family functioning.

Main Outcomes And Measures: Key outcome domains of developmental health included measures of socialization, communication, independent living skills, and measures of internalizing and externalizing behaviors. Thresholds for doing well in these domains by either proficiency or growth were established. The extent to which language, IQ, household income, parent coping, and family functioning were associated with assessed outcomes was determined by logistic regression. The association between outcomes and concurrent Autism Diagnostic Observation Schedule (ADOS) classification scores was also estimated.

Results: In a total cohort of 272 children (234 [86.0%] boys; mean [SD] age, 10.76 [0.26] years), approximately 78.8% (95% CI, 73.2%-84.4%) of the sample were estimated to be doing well by either metric on at least 1 domain, and 23.6% (95% CI, 17.7%-29.4%) were doing well in 4 or 5 domains. It was possible to be doing well by either proficiency or growth and still meet ADOS criteria for ASD. For the growth metric, between 61.5% (95% CI, 40.7%-79.1%) and 79.6% (95% CI, 66.0%-88.9%) of participants had ADOS scores of 4 or greater; for the proficiency metric, between 63.8% (95% CI, 48.4%-76.9%) and 75.8% (95% CI, 63.0%-85.4%) had scores of 4 or greater. Doing well by either metric for all domains was associated with T1 scores on that outcome domain (eg, T1 daily living skills associated with doing well at T2 daily living by the proficiency metric as measured by the Vineland Adaptive Behavior Scales-Second Edition daily living skills scale [202 participants]: β = 0.07; OR, 1.07; 95% CI, 1.03-1.11; P < .001). Doing well in socialization by the growth metric was also associated with better T1 language skills scores (202 participants) (β = 0.04; OR, 1.04; 95% CI, 1.00-1.07, P = .04). Doing well in externalizing by the growth metric was also associated with higher household income at T1 (178 participants) (β = 0.10; OR, 1.10; 95% CI, 1.06-1.15; P < .001). Better family functioning at T1 was associated with doing well on both socialization and externalizing by proficiency metric and on internalizing by growth metric (socialization by proficiency [202 participants]: β = -1.01; OR, 0.36; 95% CI, 0.14-0.93; P = .04; externalizing by proficiency [178 participants]: β = 1.00; OR, 0.37; 95% CI, 0.16-0.82; P = .02; internalizing by growth [178 participants]: β = -1.03; OR, 0.36; 95% CI, 0.16-0.79; P = .01).

Conclusions And Relevance: This cohort study found that a substantial proportion of children with ASD were doing well by middle childhood in at least 1 key domain of developmental health, and that doing well was possible even in the context of continuing to meet criteria for ASD. These results support a strengths-based approach to treatment planning that should include robust support for families to increase the potential likelihood of doing well later in life.