While most studies of early behavioral interventions for children with ASD have looked only at immediate outcomes at the end of early interventions, this long-term study is the first to search for evidence of sustained effects of early interventions. In this study, the researchers conducted a follow-up analysis on the results from a previous randomized clinical trial, which compared groups of children with ASD aged 18 to 30 months receiving either the Early Start Denver Model (ESDM) intervention or the more usual “Community treatment” approach in the greater Seattle area over a 2-year period. ESDM includes integrating applied behavioral analysis methods with developmental approaches and parent coaching to promote learning, social reciprocity (the back-and-forth flow of social interaction), and affective (emotional) engagement. Results reported in a previous study showed that at the end of the treatment period, ESDM had an overall positive impact compared with treatment as usual, including significant improvement in intellectual ability (particularly in how the children spoke and what they understood when others were talking) and daily living skills, and a less severe autism diagnosis than at the start of the trial. There were also significant differences in social behavior between the two groups, although there were no significant differences in autism symptom levels.

The current study examined whether long-lasting outcomes could be observed 2 years after the interventions in the first study ended. To accomplish this, the 39 children remaining available from the original study of 48, now age 6, were again assessed. The researchers found that while the two groups were not significantly different in terms of the observed improvements in intellectual functioning, the adaptive behavioral improvements seen in the ESDM group were maintained even though these children had received fewer services than the “Community treatment” group after the end of the original intervention trial. The ESDM group also showed a trend toward improved peer relations, a measure which had not been previously measured. Surprisingly, the ESDM group had less severe overall ASD symptoms than the “Community” group, which was unexpected given that this had not been observed immediately following the intervention. This is the first study to examine the longer-term effects of early ESDM behavioral intervention on the course of ASD. It suggests that early intensive interventions can not only produce immediate improvements in intellectual ability, language, and social behavior, but that these improvements can be sustained over time and be generalized to core ASD symptoms.

Research has suggested that when children receive interventions during the prodromal phase of ASD—which occurs during the first year of life when very early signs and symptoms might indicate the start of ASD, but before more specific symptoms are observed and before the disorder is diagnosed—the severity of the ASD may be reduced. This study tested the effect of a very early intervention during this prodromal period. This study used an adapted Video Interaction to Promote Positive Parenting (iBASIS-VIPP), in which parents are guided by therapists to improve social interaction in their infants, as it has been shown to improve parental interaction with their infants in other developmental contexts. Families were randomly assigned either to receive the intervention (28 families) or no intervention (control, 26 families). The members of the study team who assessed the level of attentiveness displayed by the infants were blinded, that is they did not participate in the intervention and were not aware of which infants were assigned to which group.

There is some evidence that following intervention, infants showed improvements across multiple outcomes and risk-markers for developing ASD, including behavior, social interaction, and brain function. The primary outcome measured in this study was the infants’ attentiveness to their parents, and there was a positive improvement in this measure following intervention, although given the degree of statistical uncertainty there is the possibility that there was no effect or a negative effect of the intervention. The secondary outcomes overall provide more evidence for a positive treatment effect, primarily due to the strong impact that intervention had on parent directiveness, or the degree to which parents control or direct the interaction with their child during play. Parent non-directiveness encourages children to engage and interact with their caregivers, and may be reduced in parents with children with neurodevelopmental abnormalities. This intervention acted directly on improving parent non-directiveness, so it is important that there was a strong effect in improving this measure. Although the results need to be confirmed in a larger study before more definite conclusions can be made, the trial did show that it is possible to deliver and test a very early intervention of this kind, in particular because all of the families who began the intervention completed it successfully.

Establishing positive peer relationships can be challenging for children with ASD. This study looked at children with ASD, in kindergarten through fifth grade, in their regular school settings rather than in clinical settings to compare two social skills interventions. One group consisted of a mix of children with ASD who were from different grade levels and classrooms, and the other group had a mix of typical children and children with ASD from the same grade level/classroom. The teaching approach was also varied—comparing direct instruction on social skills (didactic SKILLS that make information logical and accessible) to shared, activity-based teaching of social skills in a natural setting (activity-based ENGAGE groups). The purpose of the study was to assess improvement of social network connections (such as increase in social interactions and mutual feelings of friendship) for children with ASD, and this was measured using a confidential survey administered to both the children with ASD and neurotypical peers who agreed to participate. The survey assessed peer engagement in the playgroup, the number of peers nominated as friends by the child with ASD, and the number of peers who nominated the child with ASD as a friend. Based on a previous study, the researchers expected the ENGAGE groups (the inclusive groups with both neurotypical children and children with ASD) to establish better relationships with other children.

The 137 study participants were school-aged children with ASD located at four different school sites across the U.S. and observed over the course of 2 years. The children’s interactions were assessed using the peer surveys prior to the start of the experimental interventions, immediately after the 8 weeks of intervention (two times per week), and again 8 weeks later. The SKILLS intervention included lessons that described social interactions, body language, teasing, and friendship tips. The ENGAGE intervention, which included not only children with ASD but their school peers (in a ratio of about 1:3), had the children meet in social groups targeting peer engagement and acceptance, and shared interests of each group were used to generate a list of possible activities—such as conversational exercises, structured games, free play, improved storytelling, and music. A Friendship Survey evaluated the friendships children made and their networks of peer relationships. In addition, observers who did not know which children had ASD conducted behavioral observations at time-points before, during, and after interventions that assessed a child’s engagement with his/her peers.

After the intervention, the study authors found there was no significant change in social network connections over time based on intervention group, but there were other modest effects of improvement. Overall, the results unexpectedly showed more benefits from a SKILLS-based, social skills group consisting of all ASD children than for a mixed/inclusive (typical plus ASD children) group. Children in the SKILLS group had more peer engagement and less isolation during recess. Another observation concerned the quality of the child-teacher relationship; children with low teacher-child closeness or high conflict improved more in their social connections through the SKILLS intervention while children with higher teacher-child closeness improved more through the ENGAGE intervention. This study shows the importance of inclusion and social participation for students with ASD and suggests that the quality of teacher-student relationship may be an important factor in choosing a social intervention approach; however, research is still needed to understand how we can best support opportunities for these interactions inside and outside the classroom.

There is a critical need for more effective treatments to address the core symptoms of ASD, including impairment in social interaction and communication, during early childhood. Some psychotropic drugs (drugs that change chemical levels in the brain, affecting mood and behavior) seem to be effective for behavior problems, but they can have unwanted and sometimes harmful side effects and do not sufficiently address the core social impairments of ASD. Behavioral interventions, on the other hand, can significantly improve social impairments, but they are often time-consuming and costly. The hormone oxytocin has been identified as playing an important role in social thinking and behavior, and it has been hypothesized by researchers that it could be used to treat psychiatric disorders and conditions that are characterized by social difficulties. Initial studies with oxytocin in adults with autism and in youths with autism (ages 12 to 19) have showed some promise, but research has also shown that earlier interventions for autism offer the best chance for improvement over the long term. This study is the first to test the use of the synthetic hormone oxytocin as an early intervention for young children with autism to improve their ability to interact socially.

This clinical trial, conducted by researchers at the Brain and Mind Centre at the University of Sydney (Australia) between October 2010 and October 2012, assessed whether oxytocin nasal spray improved the social interaction of 31 children with ASD ages 3 to 8, as observed by caregivers, and found that among these children, a 5-week course of oxytocin nasal spray significantly improved their social responsiveness, as reported by caregivers, compared with no treatment, or placebo. In addition, the study showed that after oxytocin administration, there was no evidence of decline on any of the social interaction or behavioral scales that were used, and the treatment was well tolerated, with no significant differences in reports of adverse events between oxytocin and placebo. Thirst, urination, and constipation were the most common, relatively mild side effects. If these findings can be confirmed in studies with larger numbers of children, it is possible that oxytocin could be the first medical treatment developed for the core symptom of impaired social interaction in childhood autism.

The second year of life can be a particularly critical developmental period for children with ASD: there is dynamic brain growth overall, as well as changes specifically in the nerves and their connections in the brain. Interventions during this period may therefore be able to alter the abnormal developmental trajectories associated with autism and prevent some ASD impairments. However, interventions that have been developed for older children may not be appropriate for infants or toddlers, as younger children interact and learn differently from older children. It is within this context that a multidisciplinary working group of experts met to perform a review of the literature evaluating early interventions (designed for children <2 to 3 years of age) and to develop recommendations for policy makers and clinical practitioners. The recommendations listed below represent a mix of guidelines for clinical practice as well as directions for future research. The statements focused on clinical practice emphasize that intervention should start as early as possible and use multiple approaches to target a broad range of ASD symptoms. The recommendations for both clinical practice and future research include sociocultural context as a moderatingfactor that should be considered and incorporated into studies and practice. The authors recommend that future research should focus on improving the rigor and consistency within and across studies, to improve applicability and utility for developing personalized intervention strategies. Lastly they recommend that researchers and clinicians consider medical comorbidities (co-existing medical conditions) that may affect the response to intervention. The recommendations of the panel were as follows:

Statement 1: Current best practices for interventions for children aged <3 years with suspected or confirmed ASD should include a combination of developmental and behavioral approaches and begin as early as possible.

Statement 2: Current best practices for children aged <3 years with suspected or confirmed ASD should have active involvement of families and/or caregivers as part of the intervention.

Statement 3: Interventions should enhance developmental progress and improve functioning related to both the core and associated features of ASD, including social communication, emotional/behavioral regulation, and adaptive behaviors.

Statement 4: Intervention services should consider the sociocultural beliefs of the family and family dynamics and supports, as well as economic capability, in terms of both the delivery and assessment of factors that moderate outcomes.

Statement 5: Intervention research should include socially and culturally diverse populations of participants and evaluate familial factors that may affect participation, acceptability, and outcomes of therapeutic approaches as well as willingness to participate in investigative studies.

Statement 6: Future research should prioritize well-defined sampling strategies, rigorous investigative design, fidelity of implementation, and meaningful outcome measurements.

Statement 7: Research is needed to determine the specific active components of effective interventions, including but not limited to the type of treatment provided, the agent implementing the intervention(s) (parent, therapist, teacher, or combination), consistency of service provision across environments and between providers, and duration of treatment and hours per week.

Statement 8: Adopting a common set of research-validated core measures of ASD symptoms (including but not limited to cognitive function, communication, and adaptive behavior) that can be used across multiple sites will facilitate comparisons across studies of children with ASD aged <3 years.

Statement 9: Future research should examine biological and behavioral heterogeneity as moderators of individual responses to interventions.

Statement 10: Intervention providers should consider medical disorders that may affect a child’s clinical presentation (especially behavior) and response to an intervention and should refer to appropriate health care providers as indicated.