Effectiveness of behavioral interventions for autism spectrum disorder: A systematic review

Background: Autism Spectrum Disorder (ASD) is identied by a group of neurodevelopmental disorders. Given the importance and demand for behavioral interventions in autism disorder as well as the need for new intervention programs, a detailed evaluation of the outcomes of interventions for the ongoing impact of behavioral interventions for autism spectrum disorder is essential. In view of the importance of, and demand for, novel behavioural interventions in autism disorder, a comprehensive evaluation of the outcomes of interventions for the ongoing impact of behavioral interventions for autism spectrum disorder is necessary. Method: A systematic search of the electronic databases including PubMed, Embase, Scopus, Sciencedirect, ISI Web of Science, and Biomedcentral were conducted to nd evidences for the effectiveness of behavioral interventions for autism spectrum disorders. The search strategy was based on exploring studies published in different global languages from the earliest to 2019 databases. In addition to the databases mentioned above, relevant studies were searched using forward and backward citation tracing. Results:In the present review, 48 randomized controlled trials (RCTs),and 44 quasi-experimental articles matched the dened evaluation criteria. Eighty-three out of 93 studies were conducted in high-income countries, the results of which raised the challenges associated with acceptability of conducting those studies in low- and middle-income countries. The majority of those studies (93.5%) were classied as having either low or medium quality. From those studies, 32 studies used integrative programs, and 30 studies used social skills development interventions.The majority of the studies were based on integrative programs designed to foster social skills development interventions. The results of the study showed that social skills development interventions were effective in changing the desired outcomes. The results showed that,regardless of the study design and the type of intervention used, the possibility for success of interventions were greater in changing the participants' social skills. Conclusions :There were inconsistent evidence on the effectiveness of interventions in providing changes in targeted outcomes. There is a need for further investigation in behavioral interventions for autism spectrum disorders. Further research is needed to understand the effectiveness of other interventions and nding a quality interventions to achieve more signicant results.

autism spectrum disorders. Further research is needed to understand the effectiveness of other interventions and finding a quality interventions to achieve more significant results.

Background
Autism Spectrum Disorder is a neurodevelopmental disorder marked by social-communication impairment, restricted, repetitive and stereotyped patterns of behavior, and repetitive and stereotyped movements [1]. In addition to these main features, other behavioral problems such as anxiety, depression, sleep and nutrition disorders, attention disorder, and self-injurious and aggressive behaviors are detected in these children [2]. There has been a growing trend in the While there is no existing treatment for autism spectrum disorder, it is generally believed that early diagnosis and treatment seem to recover many people with autism over time [7]. Therefore, how different interventions could help to improve the functional ability of people with autism spectrum disorder is essential for families, health professionals, and policymakers [2]. Over the past 30 years, various treatments have been suggested to alleviate and recover symptoms associated with autism spectrum disorder. Current treatments include medications, diet changes, vitamin therapy, rehabilitation therapies, and behavioral, and developmental interventions [8]. Most interventions vary, depending on the theoretical framework, type of presentation, severity of intervention, the level of parent involvement, and comprehensive intervention used. In addition, interventions are very expensive and necessitate a large number of well-trained staff and technical infrastructure [9].
Based on the performance shown in the experimental studies, interventions included in the continuum of behavioral interventions are the dominant treatment approach to improve social, adaptive, and behavioral performance of people with autism spectrum disorder [10]. These interventions are guided by a therapist and are seeking to improve social and behavioral skills in children and their families. While behavioral interventions may be provided up to long hours per week, there is a debate about the intensity needed to achieve positive outcomes and the effectiveness of different approaches [11]. A comprehensive evaluation of review studies conducted on behavioral interventions in autism spectrum disorders reveals that most of these studies have methodological flaws, which is led to the weakness of their validity [2,9]. We identified one recent review on the effectiveness of behavioral interventions for Autism Spectrum Disorder among children's [2]. These reviews included various designs, such as randomized controlled trials (RCTs), controlled clinical trials (CCTs) or observational analytical studies (i.e., prospective or retrospective cohort studies with comparison groups) and reported data on the effects of a behavioral or developmental intervention in individuals with ASD until 2008 [2]. However, none of those reviews assessed the methodological quality of the included studies. There is also evidence of positive outcomes for many of the interventions reviewed in systematic review conducted on autism disorder. Therefore, further investigation is required to evaluate the effectiveness of behavioral interventions for autism spectrum disorder using rigorous scientific methods. Physicians, educators, and families of people with autism need to make informed decisions about treatment options. In this regard, clinical and research questions about the benefits of the related interventions need to be addressed and responded. Given the importance of current increasing trend in the novel intervention programs, a detailed evaluation of the effects of interventions on the continuing impact of behavioral interventions for autism spectrum disorder will provide the necessary information for policymakers, researchers, health care providers, and families. This systematic review aimed to identify, evaluate, and integrate evidence on the effects of behavioral interventions to improve the primary symptoms associated with autism spectrum disorders.

Methods
In the current systematic review, PRISMA checklist was used to report the findings of the study.

Search strategies
Database articles, including BioMed Central (BMC), PubMed, Sciencedirect, Embase, Web of Science, and Scopus were reviewed using the search strategy used in the related studies [2]. The search strategy in the scientific databases is set out in Appendix A, and according to the search requirements at each of the databases mentioned, necessary changes were made while searching websites.

Trial selection
All articles identified from various sources were first collected by a researcher using Endnote software. After integrating the articles from all the cited databases and deleting duplicate articles, the two researchers of the current study, independently reviewed all the articles and excluded the articles that were not relevant to the subject and the inclusion criteria. The abstracts of remaining articles were independently studied by two researchers. Then, the full text of the relevant articles was reviewed by two researchers and the articles that were fully consistent with the criteria were identified. Using forward citation and backward citation reviews, additional articles were added to the resource collection. Data were extracted by two researchers. At all stages, disagreements were resolved through consensus-based discussion and, finally, through the opinion of the third researcher.

Inclusion criteria
The PICO index (study population, type of study, type of intervention and type of outcome) was used to evaluate the inclusion and exclusion criteria [12].
-Type of Study: Types of RCTs and Quasi-Experimental Studies (cPPI and PPI).
Study population: Children, caregivers and families of children of all ages and both sexes who were involved with autism spectrum disorders and problems.
-Type of intervention: This included the evaluation of an intervention program at national, regional, organizational, community, or individual levels for the autism spectrum disorder.
Type of Outcomes: Studies with subjective outcomes (such as the use of questionnaires for reporting) and objective outcomes (such as the use of observation and surveys) for autism spectrum disorders.
-Study period: Studies from the first years of publication in the scientific database -Studies published in all languages of the world.
Exclusion criteria -Type of Study: Descriptive, Qualitative, Review, Structured Review, Meta-Analysis and Protocol.
-Study population: Studies conducted in other groups with developmental problems.
-Type of Intervention: Studies that have performed interventions for autism spectrum disorders along with other interventions for other developmental problems.
-Type of Outcome: Studies that their results are evaluated using qualitative data and the results of the evaluation cannot be compared.

Data extraction
The final articles after reviewing were summarized in predefined tables and finally the articles were analyzed according to goals and objectives. The information in the table included: -Full name of the first author of the study, year of publication of study and country of study -The study design consists of two general groups of controlled trial studies and quasi-experimental studies. Quasi-experimental studies were divided into two types controlled pretest/post-test interventions (cPPI) and pretest/post-test interventions (PPI) -Target group of the intervention programs: Studies were investigated based on conducting on children, family and child caregivers and a combination of child/family and caregivers.

Quality assessment
In order to determine the quality of the articles, two trained researchers reviewed the articles. To evaluate the quality of the studies, the EPHPP tool developed by the National Collaboration Center for Methods and Tools (NCCMT) for all types of studies was used [13]. By application of this tool, one of the qualities, i.e., strong, medium and poor were considered for each of the articles. The quality assessment based on this tool is based on an evaluation of 6 components, including sample selection bias, type of study, confounders, blinding, data collection methods, and sample dropout and exclusion. The quality assessment of the studies based on the tools was separately carried out by two researchers and finally, to determine the quality of the studies, the disagreement between the two researchers was resolved by consensus-based discussion. Kappa coefficient was used to evaluate the agreement between the two evaluators [14]. No studies were excluded because of poor quality.

Results
A total of 37200 references were identified and reviewed: 36990 references from the main sources, and 210 references from other sources. From these references, 1290 were selected for abstract review. After, In-depth abstract review of the abstracts, 365 references met the inclusion criteria and were selected for full review. Eventually, after ensuring that inclusion criteria were satisfied, 93 studies were included in this review (See Fig. 1).
A summary of the included articles is provided in Table 1. Of the two studies that had used for contemporary applied behavior analysis interventions approach, one study reported that some of the expected outcomes created significant change [41], and in one study, the intervention was not successful [88]. We addressed the quality of randomized trials and non-randomized interventions (i.e. cPPI and PPI)

Discussion
Although there are very limited number of review studies mentioned in the current review, there has been no systematic review that comprehensively examines the effectiveness of behavioral interventions to improve the primary symptoms associated with autism spectrum disorders in children. Thus, this study was conducted to eliminate the knowledge gap in this field.
Following article reviews conducted by the researchers, finally 93 studies were identified for evaluation in this systematic review. 49 studies were randomized trial, and the rest were quasiexperimental. The sample size of most studies was small, and the follow-up duration of interventions was largely short and unclear. Also, the outcomes measured in the studies were mainly based on observation. The intervention approach used in the 32 studies was integrative, and the majority could significantly provide changes in all outcomes. Twelve studies used models and related theories, and 43 studies were poor in terms of quality. For these reasons, we can conclude that these studies provide no convincing evidence about interventions conducted.
Most studies were randomized trials. According to a similar systematic review, most studies have been conducted on behavioral interventions in children with autism spectrum disorder [2]. If appropriately used, conducting these studies can provide sufficient information in this area. Most studies were based on integrative and behavioral interventions and social skills development. These studies used a variety of strategies, and its result is in line with a study on behavioral interventions among children with autism spectrum disorder [2].
This review study displays that there are limited studies in this field in middle and low-income countries. Despite the fact that, fewer people live in high-income countries compared to middle-and low-income countries, the majority of studies were conducted in high-income countries; this may be due to better identification of children with autism spectrum disorder in those countries. Moreover, our study showed that, from the eight RCT studies, seven of the articles were conducted in highincome countries, showing the importance of conducting such studies in low-and middle-income countries. However, since the majority of studies in high-income countries have been conducted by different types of research strategies, the evidence about studies of high-income countries is likely to be appropriate in low-and middle-income countries as well.
It is expected that the Integrated and combined interventions produce more positive outcomes than other interventions; though, the current review study revealed that such interventions, in changing the targeted goals, were less successful compared to social skills development intervention.
Interventions that work on only one outcome in children with autism show better and more important outcomes than interventions that evaluate and assess multiple outcomes. Moreover, the results of these studies showed that few participants took part in the study and the length of follow-up was short. Hence, studies with methodological weakness, few participants, and relatively short-term follow-up may not show the real effects of behavioral interventions on improving the primary symptoms associated with autism spectrum disorders.
Regardless of the study design and type of intervention used, our study showed that interventions have been successful in improving the skills of children with autism spectrum disorder, especially social skills. Therefore, this clarifies the need for effective interventions and follow-up in children with autism spectrum disorders. The results of this review study indicated that the most effective behavioral treatments for ASD include interventions that address behavioral, social, and communication deficits associated with the disorder.
The results of the current review revealed that the majority of studies did not explicitly use models and theories related to autism spectrum disorders. It is now clear that addressing social and behavioral science theories in designing a health plan could be related to the efficiency of the interventions. These frameworks help to recognize the different skills and conditions (such as the cultural, economic, and social conditions) in which the behavior occurs.
Some of the mentioned studies in the current review had poor and moderate design, and the majority of studies were classified as low quality. Almost half of the studies discussed in this review, were quasi-experimental, and other trial studies had some deficiencies in the method of work and presentation of results. This in turn had a negative effect on the quality of the mentioned studies.
Several factors contributed to the limitations of this systematic review including: Using different study designs, including randomized controlled trials and quasi-experimental studies resulting in a variety of outcomes. These limitations lead to the impossibility of conducting a meta-analysis. Another limitation of this study could be a diffusion bias due to overlooking gray sources to evaluate the effectiveness of the interventions.

Conclusion
To increase effectiveness of behavioral interventions for Autism Spectrum Disorders in Children, the following measures could be considered: application of randomized trial studies instead of quasiexperimental studies, increasing the duration of interventions and follow-ups, use of other intervention approaches, increasing sample size in studies for achieving the desired results, use of theory, models, and educational frameworks for creating novel pathways.

Consent for publication
Not applicable.

Availability of data and material
All data generated or analyzed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was supported by Hamadan University of Medical Sciences [reference number: 9804112817]. The funder had no role in the design of the study, data collection, analysis, interpretation of the data, writing of the manuscript, or the decision to publish.

Authors' contributions
All authors read and approved the final manuscript. SB, MA, EJ and AM conceived of the study and participated in the design, data collection and analysis as well as preparation. MA, EJ and AM participated in the data analysis and preparation. MA participated in data collection.

AND
Behavior Therapy or Social Skills Training or applied behavioral analy* or ABA or intensive behavioral intervent* or (IBI or IBT) or applied verbal behavior or verbal behavio* or (verbal NEAR (therap* or communicat*)) or lovaas or linwood or Douglass or CABAS or DTT or (Treatment NEAR Education NEAR Autistic NEAR communication NEAR Handicapped NEAR children) or teacch or floor time or "Social Communication Emotional Regulation Transactional Support" or scerts or (pivotal NEAR Figure 1 Flow diagram for the identification, screening, eligibility and inclusion of studies