Summary of Advances
In Autism Spectrum Disorder Research
Effectiveness of Community-Based Early Intervention for Children with Autism Spectrum Disorder: A Meta-Analysis
Nahmias AS, Pellecchia M, Stahmer AC, Mandell DS. J Child Psychol Psychiatry. 2019 Nov;60(11):1200-1209. [PMID: 31206690]
Research indicates that children with ASD often experience cognitive, social, and communicative improvements as a result of early interventions provided by university-based experts. However, most community-based early intervention programs—where most children with ASD receive services—often do not incorporate evidence-based approaches into their practice. Not much is known about whether community practice results in positive outcomes for autistic children, and how their outcomes compare to what is observed in university-based trials.
The purpose of this meta-analysis was to determine the effect of early intervention for ASD as it is implemented in community-based settings. Four outcome domains were considered: cognition, social skills, communication, and adaptive behavior. The researchers also aimed to identify moderating factors that may influence the effectiveness of community-based early intervention for children with ASD.
The researchers first conducted a systematic review of previous studies that investigated the relationship between early intervention in community settings and ASD outcomes. Studies were included that met the following criteria: written in English, pre-test/post-test group design, and age of children at the start of the study was less than 73 months. For a study to be included, it also had to provide information about children’s cognitive, social, communication, or behavioral outcomes based on a variety of common clinical measures. Ultimately, 33 studies from 9 countries were included in the analysis. Data from these studies were then coded for different categories of early intervention programs, including “model programs” (community care associated with universities or hospitals), “specified treatment as usual” (treatments at a school or agency with specified and defined procedures), and “unspecified treatment as usual” (treatment in which participants received an unspecified array of services).
The researchers next determined the effect size of each intervention on ASD outcomes. Effect sizes measure the magnitude of change from baseline to post-treatment, indicating the degree to which the intervention had a positive impact on ASD outcomes. They also explored whether effect sizes differed based on group and intervention characteristics, such as country of origin and study design.
The researchers found that the average effect sizes for intervention programs were small across all outcome domains. Children in model programs did show moderate improvement in cognitive outcomes and small improvements in outcomes for communication, social behavior, and adaptive behavior. When assessing for moderating factors, the researchers found that effect sizes differed by the country in which the study was conducted. For example, early interventions in Italy showed medium effects on communication and adaptive behavior outcomes, while early interventions in Israel showed medium effects only on communication outcomes. On average, intervention duration was negatively associated with communication and adaptive behavior outcomes, meaning that early intervention programs that lasted longer had smaller effect sizes than those that were shorter in duration. This was a surprising finding. The researchers suggested that it may relate to the use of standard scores; when used to evaluate changes in outcomes over time, these scores can highlight the developmental delay of children with ASD in comparison with their typically developing peers.
Overall, the effect sizes of early interventions in community settings were much smaller than those of university-based clinical research trials. These small effect sizes may have been the result of barriers to incorporating evidence-based research into practice, such as lack of high-quality training and resources to provide ongoing program supervision. Additionally, families who enroll their children in university-based research studies tend to be of higher socioeconomic status, which is associated with improved ASD outcomes, and therefore may not accurately represent the general population. Increased efforts to improve training and resources may improve the effectiveness of early interventions in community settings.
A Multisite Randomized Controlled Two-Phase Trial of the Early Start Denver Model Compared to Treatment as Usual
Rogers SJ, Estes A, Lord C, Munson J, Rocha M, Winter J, Greenson J, Colombi C, Dawson G, Vismara LA, Sugar CA, Hellemann G, Whelan F, Talbott M. J Am Acad Child Adolesc Psychiatry. 2019 Sep;58(9):853-865. [PMID: 30768394]
The Individuals with Disabilities Education Act requires that young children with ASD be provided with high quality research-based early intervention programs. However, there have been few studies conducted with scientific rigor that support the use of early intervention for children under 3 years of age. Of this limited research, promising results have been found by studies of the Early Start Denver Model (ESDM), an intensive intervention program for young children with ASD that incorporates play- and routine-based techniques built from methods of applied behavioral analysis. The results of these studies show significantly improved language skills in 18- to 30-month-old children with ASD.
This study aimed to replicate and expand on these promising findings using a randomized controlled trial (RCT) of intensive ESDM treatment. A larger and more diverse sample of toddlers was included, encompassing multiple treatment sites. The researchers designed the trial to follow all Consolidated Standards of Reporting Trials (CONSORT) guidelines to ensure scientific rigor and transparent reporting. The study enrolled 118 children with ASD from 14 to 24 months of age, who were randomly assigned to receive either ESDM treatment or “community” treatment (e.g., interventions their parents arranged, exclusive of ESDM). ESDM treatment was implemented by trained professional therapists in 2 phases across 3 university-based ASD specialty centers.
In the first phase of the study, the ESDM participants received 1 hour of weekly ESDM parent coaching over 12 weeks. In the second phase, they received at least 10 hours weekly of one-on-one ESDM intervention in a home or childcare setting and 4 hours monthly of ESDM parent coaching for 24 months. Using the Early Screening of Autistic Traits Questionnaire, the Infant Toddler Checklist, the Modified Checklist for Autism in Toddlers, the Vineland Adaptive Behavior Scales, and the Autism Diagnostic Observation Schedule (ADOS), the researchers assessed the children for language improvement, adaptive behavior, autism severity, and developmental quotient (DQ, which provides a general measure of cognitive ability). Assessments were conducted by examiners who were naïve with respect to treatment group at 4 time periods: at enrollment, end of phase 1, end of the first year of phase 2, and end of the study period. The primary outcome measure was improvement in children’s language abilities, based on the Mullen Scales for Early Learning.
Eighty-one participants completed the study through phase 2. At the end of phase 2, the researchers found that, at 2 of the study sites, the ESDM group showed significant improvements in their language skills as compared to the community group. Overall, the ESDM group was found to be approximately five months more advanced in their language skills than children in the community group. However, the ESDM group and the community group showed no significant differences in adaptive behavior, ASD symptom severity, or DQ. Both the ESDM and community group made comparable improvements in these domains.
The researchers also aimed to explore whether children’s initial DQ would moderate effects of treatment on group outcomes. For language outcomes, initial DQ did not moderate the effect of ESDM in comparison to community treatment. However, for ADOS severity, initial DQ was found to moderate the effect of ESDM in comparison to community treatment. For children in the ESDM group and not the community group, those with higher initial DQ showed greater reductions in ADOS symptom severity than children with lower initial DQ.
The present study partially replicated findings from previous studies of ESDM, concluding that children who receive ESDM treatment show greater language advances than children who receive community interventions, but no difference in adaptive behavior and DQ outcomes. These results support implementation of ESDM as a high-quality early intervention for young children with ASD. Importantly, the researchers suggest that other early interventions may also result in significant improvements for children with ASD, but these approaches will require similar community-based efficacy studies conducted with scientific rigor and implemented with fidelity to help ensure that families receive high-quality services.
Effect of Wearable Digital Intervention for Improving Socialization in Children with Autism Spectrum Disorder: A Randomized Clinical Trial
Voss C, Schwartz J, Daniels J, Kline A, Haber N, Washington P, Tariq Q, Robinson TN, Desai M, Phillips JM, Feinstein C, Winograd T, Wall DP. JAMA Pediatr.. 2019 May 1;173(5):446-454. [PMID: 30907929]
Most children with ASD face challenges with socialization, often struggling to maintain eye contact, engage in joint attention, and interpret facial expressions. Applied behavioral analysis (ABA), a therapeutic technique that teaches adaptive behaviors, is considered by experts to be the gold-standard intervention for children with ASD. However, many families encounter barriers to accessing ABA, including high costs and long waitlists. Mobile technology-based social learning aids are an increasingly popular option for families who are unable to access traditional ABA services or are seeking to supplement ABA with strategies that are more generalizable to real-world contexts. Despite the potential benefits of these technologies, few clinical studies have investigated their efficacy as interventions for ASD.
This study tested the effectiveness of a wearable social learning aid called Superpower Glass (SG) for children with ASD. Powered by Google Glass, SG is a digital glasses aid that is designed to augment children’s at-home social interactions by providing facial engagement and feedback. The technology uses facial recognition software to analyze emotions and provide the child with a simple emoticon to indicate another person’s emotional state. SG also provides engagement activities that help the child understand how facial expression relates to emotion. The glasses collect data that sync to a smartphone app, enabling caregivers to track the child’s progress.
Researchers hypothesized that children who used SG in conjunction with traditional ABA would show greater improvements in social skills as compared to children who only received ABA treatment as usual. They enrolled 71 children with ASD between 6 and 12 years of age who were already receiving ABA twice a week at home. Forty children received the SG intervention in addition to ABA, while the 31 control participants continued ABA without the intervention. Participants who received the SG intervention were instructed to use the glasses and practice the engagement activities for four 20 minute sessions at home 4 times per week for 6 weeks.
The researchers assessed the children before and after the 6 week intervention, using the Vineland Adaptive Behavioral Scale (VABS-II) and the Affect Recognition Domain of the Developmental Neuropsychological Assessment (NEPSY-II). They also designed and administered the Emotion Guessing Game (EGG), which evaluated the child’s ability to recognize emotional expressions produced by a live actor.
The study results revealed that children who received the SG intervention showed significant improvement in the socialization domain of the VABS-II. Slight positive improvements were found in the other outcome measures, but these were not statistically significant. Additional follow-up evaluations found that children did not maintain social improvements on the VABS-II 6 weeks after discontinuing use of the SG technology. The researchers noted that on average, children used SG for only half of the recommended sessions, suggesting that social skills may have improved more significantly if participants had used the device as initially intended. Future research can work to improve consistent engagement on behalf of both parents and children.
These findings support the use of technology-based social learning aids as a supplement to enhance traditional ABA interventions. Digital devices like the SG represent important progress towards accessible, home-based care for children with ASD.