Background: There has an increasing number of published trials on psychosocial intervention programmes for pre-school children with autism spectrum disorder (ASD). To achieve better quality of unbiased evidence for the effectiveness of ASD interventions, it is necessary to conduct a comprehensive review that covers studies with adequate quality standards, such as randomised controlled trials (RCTs), and different types of intervention In this study, we categorize interventions for ASD as behavioural, social-communication focused, and multimodal developmental based on Howlin's classification of early interventions for children with ASD. The aim of this study was to compare these three models and investigate the strengths and weaknesses of each type of intervention and to identify the approaches that contribute to a successful outcome for children with autism.
Methods: We performed a systematic review and meta-analysis. We included RCTs targeting children with ASD 6 years old or younger. A random effects model was used to present the effect estimate for the outcomes. This study also performed combined meta-analyses of all the three models to investigate the overall effectiveness of the intervention programmes.
Results: 32 randomized controlled studies were found to be eligible for inclusion. The synthesized data included 594 children from 14 RCTs. There was no statistically significant difference in the effects on autism general symptoms between the social-communication-focused model and the multimodal developmental model (p = 0.83). The results suggest that there is evidence of an effect on 'reciprocity of social interaction towards others' (standard mean difference [95% confidential interval] = 0.53[0.29,0.78], p<0.01) and 'parental synchrony' (SMD = 0.99[0.70,1.29], p<0.01).
Conclusion: The small number of studies included in the present study limited the ability to make inferences when comparing the three models and investigating the strengths and weaknesses of each type of intervention with respect to important outcomes. Since the outcome of 'reciprocity of social interaction towards others' could be a dependent variable that might be context-bound to interactions with the child's parent, we cannot conclude the interventions for pre-school children with ASD have significant effects on a generalized skill to engage in reciprocal interactions with others. However, the outcomes of 'reciprocity of social interaction towards others' and 'parental synchrony' may be promising targets for interventions involving pre-school children with ASD.
Trial registration: Prospero CRD42011001349.