Efficacy of Adapted Responsive Teaching for Infants At-Risk for ASD

Background:  Studies of early behavioral interventions for toddlers with ASD are just beginning to emerge (Carter et al., 2011; Dawson et al., 2010); however, empirical data are lacking on the efficacy of interventions with infants at-risk for a later diagnosis of ASD in a community (non-clinical) sample. Theoretically, intervening with infants at-risk for ASD prior to the emergence of all the diagnostic symptoms could be more efficacious than interventions initiated after diagnosis, and may have implications for prevention (Dawson, 2008). Thus, empirically-validated interventions appropriate to very young infants and toddlers at-risk for ASD are needed.

Objectives:  To evaluate the potential of a parent-mediated intervention (Adapted Responsive Teaching [ART]; Adapted from Mahoney & MacDonald, 2007) designed for one-year-olds at-risk for ASD to improve developmental functioning and ameliorate the severity of core symptoms of autism.

Methods:  This randomized controlled trial tested effects of an experimental treatment (ART) versus a control condition (Community Services [CS]) with 16 infants at-risk for ASD, using an intent-to-treat (ITT) analysis. Infants were identified through birth records and screened with the First Year Inventory (FYI) at 12 months of age. Those with FYI risk scores >95^th percentile were invited for a comprehensive developmental assessment (Time 1). If risk indicators were confirmed, families were invited to enroll in the intervention. Eligible families who consented were randomly assigned to either ART or CS, using a 2:1 randomization ratio. Sixteen families enrolled; 11 were assigned to ART and 5 to CS. ART families received a 6-month home-based intervention designed to enhance parent responsiveness and promote social-communication and sensory-regulatory functions. Children participated in Time 2 assessments immediately after completing the intervention phase (~22-24 months of age) and a Time 3 outcome assessment ~6 months after completing the intervention phase (~30-34 months of age). Hierarchical linear growth models where used to model growth in the assessments over the three measurement occasions.  Estimation was conducted under a Bayesian framework, allowing for exact inference in finite samples (Yuan & MacKinnon, 2009).

Results: The ITT estimate is based on the expected difference between ART and CS groups at 20 months after the first assessment. The ART group significantly outperformed the CS group on the Mullen Scales of Early Learning with the largest effects for receptive language (effect size, mean of posterior distribution = 1.271) and visual reception (.525).  Likewise, the ART group significantly outperformed the CS group on the Vineland Adaptive Behavior Scales with large significant effects for all subscales: socialization (1.274), daily living (.795), motor (=.882), and communication (.702).  Although effect sizes were large on several measures of sensory processing and parental responsiveness, group differences failed to reach statistical significance in this small sample.

Conclusions: Although both groups made gains, there was broad supportive evidence for greater benefits of the ART treatment relative to the CS condition on overall development and adaptive behavior, particularly in the social-communication domain.  These findings support the promise of parent-mediated early intervention with infants at-risk for ASD in improving developmental outcomes on standardized tests. Further analyses of parent responsiveness and sensory processing functions are warranted with a larger sample.