Evaluating the Effectiveness of Early Interventions On Social Communication Outcomes for Young Children with Autism

Background: Deficits in social communication are a core feature of autism and are apparent in children with autism as early as their second year (Adamson, Bakeman, Deckner, & Romski, 2009; American Psychiatric Association, 1994; Weismer, Lord, & Esler, 2010). For over a decade, early intervention has been recommended as a way to address deficits (National Research Council, 2001; Lord & Bishop, 2010). A range of interventions targeting communication and other developmental outcomes have been developed and researched. However, the effectiveness of these early interventions in increasing expressive language has not been reviewed systematically. 

Objectives: The purpose of this study was to review the features of early interventions addressing communication skills and examine the magnitude and range of effects of these interventions on expressive language outcomes for young children with autism. 

Methods: Nine high quality group experimental studies addressing communication outcomes for young children with autism were included in this study. Study inclusion criteria included: (1) group design experimental studies, (2) included at least one intervention that addressed social communication skills, (3) measured one or more expressive language outcomes, (4) participants diagnosed with or at risk for autism, (5) participants with a mean age less than or equal to 42 months, (6) published in peer reviewed journals, (7) written in English, and (8) published between 1985 and 2012. Studies were selected through a keyword search of online databases and archival searches; then, abstracts were screened for key components. Studies meeting criteria were coded for: child characteristics (IQ, severity, age), the type of intervention approach (developmental, behavioral, or hybrid), agent of intervention, extent of parent involvement in intervention, the intensity and duration of treatment, and child social communication outcomes. Effect sizes were calculated for expressive language outcome measures. 

Results: Although all nine studies assessed communication outcomes, studies varied widely in the extent to which they explicitly targeted social communication and the measures of social communication. A total of 480 children with autism participated in these studies, ranging in age from 12 to 54 months at the start of the studies. Fifty-six percent of these interventions were developmental in approach, 11% were behavioral, and 33% were hybrid. Length of interventions ranged from 3 to 24 months. All studies included a parent-training component. Effect sizes for expressive language varied across studies, ranging from .06 to .58. Four studies used the MacArthur-Bates Communicative Development Inventory (MCDI; d = 0.38-0.58), three used the Mullen Scales of Early Learning-Expressive Language subscale (MSEL; d = 0.06-0.56), one used the Early One-Word Picture Vocabulary Test (EOWPVT; d = 0.18), and one used the Communication and Symbolic Behavior Scales (CSBS; d = 0.43). 

Conclusions: Although all interventions targeted social communication, expressive language outcomes varied by intervention and measure. Across studies there was evidence that early intervention did improve expressive language. Additional analyses are needed to examine the intervention features associated with differential outcomes and child characteristics that predict expressive language outcomes.