Treatment as Usual In Early Intervention: Control Group or Legitimate Contender?

Background: Research on the effects of early intervention in autism is often complicated by a number of methodological factors, including the selection of an appropriate control group. With the increasing dissemination of empirically supported treatments (ESTs) and integration of practice guidelines into legislature, “treatment as usual” (TAU), may be becoming something quite different than once assumed. Currently, there is limited investigation into the nature of TAU, how these services compare with recommended standards, and outcomes observed in children receiving TAU (McLennan, 2008; Stahmer, 2007).

Objectives: This study describes the characteristics and outcomes of TAU services in the Tri-State area in a cohort of toddlers presenting for the first differential diagnosis from 2006-2010. We also compare the outcomes of children in our cohort to outcomes reported in recent clinical trials of early intervention services.

Methods: Forty-four consecutive referrals of children with an ASD received a comprehensive developmental evaluation between 14-26 months (T1) and were re-evaluated at approximately 36 months (T2) of age. Detailed parental report of intervention services was collected at T1 and 2, and cross-referenced with IFSPs and other documentation to promote accuracy. Evaluations at T1 and 2 included the Mullen Scales of Early Learning (Mullen), the Autism Diagnostic Observation Schedule (ADOS-G, Modules 1 & 2), the Vineland Adaptive Behavior Scales-II (VABS-II), and a best estimate clinical diagnosis from an experienced clinician.

Results: Analyses suggest that there is significant variability in the programming children received, although it was broadly consistent with practice parameters specified by local service agencies. On average, children received approximately 16.5 hrs/wk of total programming (SD=8.5), including 1.9 hrs/wk of speech therapy (SD=1.1), 0.9 hrs/wk of occupational therapy (SD=0.7), and 10.3 hrs/wk of early intervention services (SD=7.1). In general, children who presented with higher ADOS severity scores received more intensive programs (r=.32, p<.04), although there was still significant variability within this group. Despite this variability, results suggest that, as a group, children who received community-based services made substantial developmental gains on the Mullen between T1 and T2, with 1.4 SD (t=6.1; p<.001; d=.88) and 1.6 SD (t=7.8; p<.001; d=1.0) gain in Receptive and Expressive Language, respectively. Improvements in Visual Reception were more modest (0.5 SD; t=2.5; p<.02) while ADOS severity scores were stable across time. In addition, as a group, children evidenced significant improvements on the VABS in Communication (0.7 SD; t=4.7; p<.001) and there was no decline in standard scores in other domains.

Conclusions: Although there was significant variability in TAU services, children made significant gains in these programs, particularly with regard to language and communication. These gains were comparable to those reported in more specialized programs (e.g., Dawson et al., 2010). On average, children in TAU were enrolled in programs broadly consistent with local practice guidelines. These results suggest that there is potentially successful dissemination of ESTs and translation of research into practice in some community settings. As such, this may present significant challenges for researchers conducting intervention studies in areas in which TAU services may begin to bridge the research to practice gap.