Intervention Effects of Intensity and Delivery Style for Toddlers With Autism (TADPOLE)

A multi-site randomized study of intensive treatment for toddlers with autism involving a three-site collaborative network plus a data coordinating center to evaluate the effects of intervention intensity and intervention style delivered for 12 months, on the progress of very young children with ASD ages 12-30 months old and their families, and the effect of children's developmental rates and autism severity on their response to intervention.

High quality, intensive early intervention is a powerful treatment for ASD, improving IQ and language markedly in randomized controlled clinical trials (RCTs), though little long term follow-up data exists. Few core characteristics that affect child change have been tested. Two potential core characteristics that invoke considerable debate among parents, professionals, and administrators are the delivery style of intervention: play-based versus discrete trial teaching, and the intensity (dosage) of intervention. This ACE treatment network conducted an RCTs to answer the following question: what are the effects of intensity and delivery style on developmental progress of toddlers with ASD? 87 young children with ASD, mean age 23.4 months who live within a specified radius near the university at each site were enrolled in one of three national sites and randomized into one of four cells varying on two dimensions: dosage - 15 or 25 hours per week of 1:1 treatment; and discrete trial teaching or naturalistic developmental-behavioral intervention. Other aspects of intervention held constant were: use of the principles of applied behavior analysis, 1:1 adult:child ratios, parent coaching in the assigned treatment, and treatment location. Developmental progress was measured frequently allowing for growth curve analysis to examine fine-grained differences in groups as well as interactions among major child and family initial variables and these two experimental variables.

autism, autism spectrum disorder, intensive intervention, early intervention, Early Start Denver Model

Children receive 15 hours a week of 1:1 intervention at home plus parent coaching using the Early Start Denver Model and following its manual

Children receive 25 hours a week of 1:1 intervention at home plus parent coaching using the Early Start Denver Model and following its manual

Children receive 15 hours per week of 1:1 intervention at home plus parent training using Early Intensive Behavioral Intervention (EIBI) and following the Manual "A Work in Progress"

Children receive 25 hours per week of 1:1 intervention at home plus parent training using EIBI and following the Manual "A Work in Progress"

The Early Start Denver Model (ESDM) is a comprehensive developmental and behavioral early intervention approach for children with autism, ages 12 to 48 months. The program encompasses a developmental curriculum that defines the skills to be taught at any given time and a set of teaching procedures based in applied behavioral analysis to deliver this content. The teaching format focuses on trained therapists delivering the intervention 1:1 by engaging reciprocally with children in typical toddler activities involving play or daily routines, adding additional structure for children who need it to progress well. Progress data is collected throughout each session and used to make decisions about teaching approaches.

EIBI is a one-to-one instructional approach based in applied behavior analysis used to teach skills in a planned, controlled, and systematic manner. Each trial or teaching opportunity has a definite beginning and end, thus the descriptor discrete trial. Within DTT, the use of antecedents and consequences is carefully planned and implemented. Positive praise and/or tangible rewards are used to reinforce desired skills or behaviors. Trial by trial data collection is an important part of DTT and supports decision making by providing teachers/practitioners with information about beginning skill level, progress and challenges, skill acquisition and maintenance, and generalization of learned skills or behaviors.

We used Time 4 means and standard deviations (SDs)s from several measures to compute a z score. The mean of a Z score is always 0 and z scores range from -3 to +3. For this composite a higher score reveals more mature developmental skills. Data from the following measures were used in this composite score: a weighted frequency of intentional communication measured within a 15 minute communication sample; number of different root words measured within the communication sample; the expressive language age equivalent score from he Mullen Scales of Early learning (MSEL); the expressive communication age equivalent scores from the Vineland 2 Adaptive Behavior Scale 2(VABS2); the expressive raw score from the MacArthur Bates Communicative Developmental Inventories; the expressive social communication abilities composite score from the PDD Behavior Inventory; and the expressive language raw score from the PDD Behavior Inventory

To build the composite scores for all the outcome measures, we first examined whether the proposed component variables were correlated (for 2 component variables) or factor loaded (for 3 or more component variables) at or above .3. We used Time 4 means and SD to compute a z score. The mean of a Z score is always 0 and z scores range from -3 to +3. For this composite a higher score reveals more severe symptoms of ASD. . from the following measures were used in this composite score: the calibrated severity score from the Autism Diagnostic Observation Schedule 2 and the expressive receptive social communication composite from the PDD Behavior Inventory

To build the composite scores for all the outcome measures, we first examined whether the proposed component variables were correlated (for 2 component variables) or factor loaded (for 3 or more component variables) at or above .3. We used Time 4 means and SD to compute a z score. The mean of a Z score is always 0 and z scores range from -3 to +3. For this composite a higher score reveals more mature developmental skills. from the following measures were used in this composite score: receptive language age equivalent score from the MSEL; receptive language age equivalent score from the VABS 2

To build the composite scores for all the outcome measures, we first examined whether the proposed component variables were correlated (for 2 component variables) or factor loaded (for 3 or more component variables) at or above .3. We used Time 4 means and SD to compute a z score. The mean of a Z score is always 0 and z scores range from -3 to +3. For this composite a higher score reveals more mature developmental skills. from the following measures were used in this composite score: 2 scores - the fine motor and the visual reception age equivalency scores - from the MSEL and 3 age equivalency scores from the VABS2 - daily living, motor skills, and socialization

Inclusion criteria: 12-30 months of age at time of assessment; ambulatory and without impairments affecting hand use; meets criteria for Autistic Spectrum Disorder on the APA Diagnostic and Statistical Manual, 5th revision criteria and on the Autism Diagnostic Observation Schedule for Toddlers: clinical consensus of ASD diagnosis by 2 independent staff (including a licensed psychologist) based on observation as well as record review; developmental quotient of >35 on Mullen Scales of Early Learning; normal hearing and vision screen; caregiver agreement to comply with all project requirements, including regular videotaping at home with provided equipment. Exclusion criteria: English not a primary language spoken at home; absence at 2 or more appointments without prior notice during the intake assessment; more than 10 hours per week of 1:1 ABA based treatment; other health or genetic conditions (i.e. fragile X syndrome, seizures, prematurity).

All de-identified data will be uploaded into NDAR as soon as primary papers are published. All demographic data are uploaded into NDAR as they are collected.

All de-identified data will be uploaded into NDAR as soon as primary papers are published. All demographic data are uploaded into NDAR as they are collected.

Rogers SJ, Yoder P, Estes A, Warren Z, McEachin J, Munson J, Rocha M, Greenson J, Wallace L, Gardner E, Dawson G, Sugar CA, Hellemann G, Whelan F. A Multisite Randomized Controlled Trial Comparing the Effects of Intervention Intensity and Intervention Style on Outcomes for Young Children With Autism. J Am Acad Child Adolesc Psychiatry. 2021 Jun;60(6):710-722. doi: 10.1016/j.jaac.2020.06.013. Epub 2020 Aug 24.