Cumulative Evidence and Consensus of Pivotal Response Treatment as a Naturalistic Developmental Behavior Intervention for Young Children with Autis

In contrast to Early Intensive Behavioral Intervention (EIBI), there are far fewer meta-analyses evaluating Naturalistic Developmental Behavioral Interventions (NDBIs) (Schreibman et al., 2015). Indeed, Pivotal Response Treatment (PRT)---a prominent NDBI---lacks meta-analytic studies. In the context of developing a line of research on the naturalistic practice construct, I meta-analyzed both single-case (SCD) and group design (GD) PRT studies that included children with autism under 6 years old, employing Hierachical Linear Modeling (HLM), Tau-U non-overlap index, the Shadish d-statistic analog, and Cohen's d. Thirty-six studies were included (30 SCD, 6 GD). Accounting for study quality, I found medium to very large SCD effect sizes (ES) across social-communication and language, play, and affect outcomes (omnibus HLM ES: gamma100 = 0.87), with larger estimates for social-communication and language. PRT produced favorable play and child affect outcomes; however, with slightly smaller and far fewer ES, claims of positive PRT effects beyond social-communication and language are tenuous. Effects were low to medium for GDs, again with larger estimates for communication and language. Evidence for positive PRT effects on receptive language, adaptive behavior, and cognitive functioning is also weak, making it difficult to draw comparisons with EIBI meta-analyses, or to indicate PRT as a comprehensive early intervention. PRT effects significantly varied between participants. For SCD studies, outcome area, study quality, research group, and setting significantly moderated PRT effects. PRT procedures were inconsistent across studies, but key components (i.e., child choice and natural reinforcement) were ubiquitous. Based on the broader literature and present findings, I outline a programmatic line of PRT and NDBI research.