Behavioral Interventions for Autism Spectrum Disorder: A Brief Review and Guidelines With a Specific Focus on Applied Behavior Analysis

Abstract

We conducted a comprehensive review of behavioral and educational interventions for individuals with autism spectrum disorder (ASD). The most prominent type of intervention, Comprehensive Early Intervention, often referred to as Early Intensive Behavioral Intervention (EIBI), has been found to be particularly effective in improving intelligence and adaptive behaviors. The naturalistic developmental behavioral intervention, designed to enhance social and communication abilities, showed effectiveness in improving language, cognitive function, and social initiation. However, more studies are needed to examine its effectiveness. Intensive individualized intervention, which provides a tailored intervention for a specific target behavior, was effective in improving social skills and communication, as well as reducing sleep, eating, and toileting problems. Cognitive behavioral therapy (CBT) is the most effective method for dealing with emotional difficulties, but it has not been widely used because of the shortage of trained experts. Parent-mediated intervention (PMI) involves parents acquiring knowledge and specific skills to improve their child’s functioning or reduce challenging behaviors. Speech and language therapy, sensory integration, Treatment and Education of Autistic and related Communications Handicapped Children, developmental approaches, and social stories are frequently used interventions. However, evidence of their effectiveness has yet to be well established. Based on these findings, intervention recommendations for autism include EIBI, Early Start Denver Model, intensive individualized intervention, CBT, and PMI. The choice of intervention should be tailored to the individual’s needs and delivered by qualified professionals with expertise in the specific intervention.

INTRODUCTION

Intervention approaches for individuals with autism spectrum disorder (ASD) encompass educational practices, developmental perspectives, and behavioral strategies [1]. Among these, applied behavior analysis (ABA) is recognized as the most extensively utilized and proven effective method for addressing the behavioral and educational needs of individuals with ASD. ABA is a scientific and systematic approach rooted in learning theory, best represented by operant conditioning [2] and classical conditioning [3] to induce behavioral changes in individuals diagnosed with ASD [4]. The term ABA came in the early 50s when learning theory was applied as a treatment strategy to change behaviors in persons with developmental disabilities [5]. New terms such as behavior intervention, behavior modification, behavior therapy, and behavior treatment were introduced in the early 70s, when learning theory was applied as a treatment method for changing behaviors of other clinical populations, including children with diverse behavior problems and persons with physical illness (e.g., diabetes, asthma, obesity, and so forth) [6]. Although these terms differ slightly, they are often used interchangeably, which can lead to confusion.

Numerous studies have established the efficacy of ABA in enhancing cognitive function, language skills, intelligence quotient (IQ), and social abilities in individuals with ASD [7,8]. Federal, state, and national organizations in the US and UK consistently recommend ABA-based interventions as the first choice of assessment and treatment guidelines for ASD. These organizations include the US Center for Disease Control [9], Australian Center for Disease Control [10], division 53 of the American Psychological Association (Society of Child and Adolescent Clinical Psychology) [11], American Academy of Child and Adolescent Psychiatry [1], as well as the states of NY [12], WA [13], and Maine [14]. In this study, our primary focus was on treatments with a strong recommendation, categorized according to the strictest evidencebased treatment criteria. Additionally, we introduce treatments that are widely used in clinical practice but need more accumulated evidence.

COMPREHENSIVE TREATMENT MODEL

The comprehensive treatment model (CTM) involves intensive training in various developmental areas, including language, cognitive function, social interaction, communication, and self-help skills. This training is based on behavioral principles and is typically started soon after the diagnosis of autism [7,8]. Early intervention, ideally before the age of 3 years, is recommended soon after the diagnosis. It is conducted in a one-on-one format for a few hours daily for an average of 2–4 years.

Early Intensive Behavioral Intervention (EIBI) [15] was designed to teach specific skills in language, cognitive function, self-help, social interaction, and motor skills using discrete trial training (DTT), which follows the principles of ABA. This service is provided by a team of professional therapists and implemented face-to-face with the child in a structured environment for a few hours per day for a few years, and active parental participation is required.

Six review studies that tested the effectiveness of EIBI reported significant improvements in IQ and adaptive behaviors consistently 12 months after EIBI [16-21]. Some studies have also reported positive changes in other developmental areas. For instance, Reichow and Wolery [19] reported that the EIBI group showed additional improvements in receptive and expressive language compared to the treatment-asusual (TAU) group. Reichow et al. [20] analyzed the effect sizes (Hedge’s g) for each area. The results showed the average of medium to large effect sizes (language [g=0.50–0.57], daily communication [g=0.74], social interaction [g=0.42], and self-help skills [g=0.55]) in the EIBI group compared to TAU group.

Recently, a few studies have explored factors affecting the effectiveness of EIBI. Caron et al. [22] found that treatment dosage (the amount of time spent on therapy) and adherence (compliance with the treatment program’s contents and procedures) were positively correlated with treatment effectiveness. Based on these findings, they recommended administering EIBI for 25–40+ hours per week for 12–24 months. Zwaigenbaum et al. [23] emphasized the importance of early intervention, reporting that interventions started before the age of 3 had more positive effects than those started after the age of 5, especially when parental and family involvement was high.

However, two recent reviews reported no significant changes in symptom severity after intervention [24,25] and limited long-term effects [25], questioning the effectiveness of EIBI. Careful interpretation is needed because most studies included in the two reviews used quasi-experimental designs (e.g., clinical controlled trial) rather than randomized controlled trials (RCTs), and their sample sizes were small. Further studies are needed to establish the effectiveness of EIBI.

Recently, another line of intervention called naturalistic developmental behavioral intervention (NDBI) [26] was introduced, which combines behavioral principles with a developmental approach that emphasizes social ability and learning in a natural context. NDBI, along with EIBI, is effective in young children with ASD. The Early Start Denver Model (ESDM) [27] is a prominent example that has been extensively investigated. Compared with traditional EIBI, ESDM focuses on improving the social development and communication skills of children with ASD by facilitating their social interaction with their caregivers in daily living settings [28]. Although limited, a few review studies have reported the positive effects of ESDM on improving children’s social communication, language, and adaptive behaviors. For example, Waddington et al. [29] reviewed 15 ESDM studies, and the results showed improvements in language, imitation skills, and social communication after 3 to 12 months of ESDM intervention. Fuller et al. [30] conducted a meta-analysis of 12 ESDM outcome studies (6 RCTs and 6 quasi-experimental designs). The average effect size of ESDM (Hedge’s g) was 0.357 (p=0.024) across diverse developmental areas, which is a medium effect size, indicating moderate improvements. Among these, cognitive function (g=0.412) and language (g= 0.408) showed significant improvements [30]. In addition to the ESDM, other models have gained attention and demonstrated similar effectiveness. These include Joint Attention, Symbolic Play, Engagement and Regulation (JASPER) [31], Early Social Interaction [32], and Pivotal Response Treatment [33].

In Korea, treatment outcome studies on EIBI and NDBIs are rare. One study [34] examined the effectiveness of 6 months of EIBI in 19 children aged 2–5 years in an EIBI center using a one-group pre- and post-test design. The study showed significant improvements in the developmental scores of the children, along with improved parental quality of life. However, it is difficult to generalize the effectiveness of the EIBI in Korea because the study lacked a comparison group and had a small sample size.

FOCUSED INTERVENTIONS

In the focused intervention, therapists provide tailored interventions to individuals of all age groups in areas where deficits are reported, such as social skills, social communication, sleep, feeding, toilet training, and emotion regulation (e.g., anxiety, anger, and so forth). The main approach is behavioral intervention, which is typically conducted for a relatively short period with the goal of improving a specific target behavior. In this study, we reviewed effective interventions for the core symptoms of ASD, such as social skills and communication, as well as combined symptoms, including sleep, eating problems, toilet training, and emotional difficulties.

CORE SYMPTOMS

Social skills

Social skills training (SST) is the most widely researched intervention aimed at improving social skills, a core symptom of ASD. It is usually delivered face-to-face by experts in specialized institutions or teachers and peers in schools under the supervision of experts [35]. In SST, which is typically conducted in 8–13 sessions over 6–12 weeks, specific social skills such as greetings, initiating and responding to conversations, giving compliments, sharing, and matching facial expressions are taught through repetitive practice [36,37].

Some studies have reported limited effects of SST on improving social skills [35,36,38]. However, rigorous RCTs have recently shown promising results. Two meta-analyses found that the intervention group showed a medium (Hege’s g= 0.51) [39] to large effect size (Hege’s g=0.81) [40] in improvement in social skills and social responsiveness compared to the non-intervention group.

SST has been demonstrated to improve social skills effectively in individuals of all ages, including infants (0–5 years old), children (6–14 years old), and adolescents/adults (15–22 years old) [41]. Especially for children aged three years, a social skills intervention model named JASPER has been introduced, and its effectiveness has accumulated [42,43]. JASPER aims to improve joint attention and teach play skills. A review of 19 studies on the efficacy of JASPER by Waddington et al. [44] found that those who received JASPER training showed significant improvements in at least one area, such as joint attention, joint engagement, playing skills, and language, compared with the control group. However, despite its effectiveness, the JASPER is not easy to access in Korea because of the lack of manuals or guidelines.

Social communication

Augmentative and alternative communication (AAC) is often used to enhance communication skills in individuals with ASD. It involves teaching the use of alternative communication methods, such as visual cues (signs, symbols, pictures, and so forth), sign language, and communication aids [45].

The picture exchange communication system (PECS) is the most widely used AAC intervention, which teaches individuals to use pictures to communicate their needs (e.g., request items and name objects). The PECS follows six steps for each communication skill using various ABA strategies, such as shaping, differential reinforcement, and stimulus control [46]. Two review studies, which included 27 and 15 studies, respectively, found that PECS was effective in enhancing communication, such as initiating conversations and making demands [47,48]. Additionally, the PECS has shown particular effectiveness for children aged 0–5 years (toddlers to preschool) and 6–14 years (school-age) [41]. However, a metaanalysis of 11 studies found that the PECS had small to medium effect sizes in improving overall speech abilities [49]. Additionally, few studies have reported the spontaneous development of verbal speech during PECS training. However, further evidence is required [46].

COMBINED SYMPTOMS

Sleep

Individuals with autism and their parents frequently report sleep problems as a comorbid symptom. A recent review found that behavioral interventions increased sleep time and sleep efficiency, as well as a decrease in sleep latency, compared to the control group [50,51]. Another review that analyzed behavioral interventions for sleep problems found that different techniques were recommended depending on the specific sleep problem [52]. For instance, a faded bedtime¹⁾ is effective in initiating and maintaining sleep, whereas extinction²⁾ and scheduled awakening³⁾ are effective during nightwaking and night terror attacks, respectively. In terms of age, faded bedtime was applied to children aged 5–9 years, and the extinction technique was applied to children aged 2–12 years, both of which were effective [52]. Sleep hygiene and behavior interventions for improving sleep habits, such as bedtime routines and sleep schedules, are not sufficient but necessary components of most behavioral sleep interventions [52,53].

Eating problem

Approximately 50% of individuals diagnosed with autism report eating problems, such as food refusal or food selectivity. Similar to sleep, behavioral interventions based on ABA are effective in addressing these eating problems. A meta- analysis of 23 intervention studies [54] reported an increase in food intake when contingent reinforcement (CR)⁴⁾ and non-removal of the spoon (NRS)⁵⁾ were used in combination (improvement rate difference [IRD]=0.69) in children with ASD under 6 years. However, a small effect size was reported for the reduction of mealtime problem behaviors, such as tantrums (IRD=0.39). For severe food selectivity and related mealtime problem behaviors, a treatment package that includes both antecedent manipulation (such as stimulus fading) and consequence-based procedures (such as NRS and differential reinforcement) has been reported to be effective [55].

While experts typically lead interventions, there are reports that active participation by parents can be more effective in generalizing and maintaining the effects of eating interventions [55,56].

Toilet training

Evidence has accumulated from diverse ABA-based toilet training programs for individuals with autism [57]. A review of 28 toilet training intervention studies found that techniques such as shaping⁶⁾, graduated guidance⁷⁾, scheduled toileting⁸⁾, and stimulus control⁹⁾ were effective. In clinical practice, these techniques are often used in combination and are tailored to the age and cognitive abilities of individuals [58,59].

The most typical toilet training for children with autism is a modification of rapid toilet training (RTT), originally devised by Azrin and Foxx [58,60,61]. The RTT is an intensive training method that combines positive reinforcement, positive punishment, and scheduled toileting, typically lasting 8 hours a day for 1–2 weeks [62,63]. This method has been reported to be particularly effective for individuals with autism and intellectual disabilities [64]. Specifically, it was initially designed for infants and toddlers aged 2–4 years, and a recent review showed promise for use in the age range of 3–12 years. However, further evidence is required [65]. Urine training should be continued until children reach 24 months of age, followed by bowel training when the urine training is completed.

Emotional difficulties

As children with autism grow into adolescence and adulthood, they often report emotional difficulties, such as depression, anxiety, and anger, in addition to ongoing behavioral problems [66]. Recent studies have shown that cognitive behavioral therapy (CBT), an evidence-based treatment for depression, anxiety, and anger, can be successfully applied to adolescents and adults with autism [67-69]. However, considering that CBT requires significant verbal and cognitive skills, it is important to assess the verbal, cognitive, and emotional abilities of adolescents and adults with autism before implementing CBT [70,71]. In Korea, access to CBT for individuals with autism is limited because very few experts provide CBT for adolescents and adults with autism.

PARENT EDUCATION

Parental education refers to the provision of either information or specific skills or both to parents to support their children appropriately. Psychoeducation provides knowledge and strategies to deal with autism issues. However, its effectiveness has yet to be investigated [72]. In contrast, parent- mediated interventions (PMI) involve training parents to implement various intervention techniques directly with their children. They have been frequently used for comprehensive early intervention and intervention of challenging behaviors [73].

PMI for comprehensive early intervention

PMI has been actively pursued as a means of providing interventions to young children with autism, typically in combination with EIBI [15]. However, PMIs may also serve as primary services in regions with limited accessibility to EIBI [74]. While the quality of intervention and service hours for PMIs may not be comparable to those of therapist-led interventions, their advantages include high accessibility to treatment and high generalizability, mainly because treatment is provided by parents almost every day in diverse real-life settings [75]. Numerous studies have reported improvements in children’s communication skills, expressive and receptive language, and adaptive behaviors upon the implementation of PMI [41]. Additionally, its collateral effects on parental adjustment and mental health, including maternal depression and parenting stress, are well-documented [76]. However, a review of 17 studies questioned the effectiveness of PMI due to inconsistencies in results across studies [77]. Another recent meta-analysis study, including 19 RCTs, also reported small effect sizes (Hedge’s g=0.18–0.27) [78]. However, these review studies included trials using diverse intervention techniques, such as floor time, massage, and ABA. Therefore, caution should be taken when interpreting the effectiveness of PMIs.

More recently, a parent-implemented Early Start Denver Model (P-ESDM) has gained attention as an alternative to therapist-led ESDM, whose effectiveness has been well-documented [79]. In P-ESDM, certified ESMD trainers teach parent-specific behavioral skills, such as joint attention and imitation, to facilitate social interaction with their child. Parents are then encouraged to apply these skills to their children’s everyday routines and activities in their homes and other daily living settings [80]. Although no systematic review of the effectiveness of P-ESDM is available, individual studies have reported improvements in children’s parental interaction skills [81], social communication skills [82], language development [83], and a reduction in autism symptoms [84]. Due to high parental preference for P-ESDM, more efforts have been made to disseminate it, including the development of internet-based P-ESDM, which has also demonstrated effectiveness [83,85].

In Korea, single-subject design studies have reported increased joint attention [86] and improvements in social interaction and communication behaviors [87,88] for PMI. However, further studies with better experimental designs are required to conclude their overall effectiveness.

PMI for challenging behaviors

Studies investigating the effectiveness of PMI for challenging behaviors, such as aggression and self-injurious behaviors, have been actively conducted. In this type of PMI, a certified ABA therapist (e.g., BCBA or BCaBA¹⁰⁾) runs individual or small-group sessions for 4–16 weeks to teach parent-specific behavioral skills to deal with challenging behaviors, which include both didactic lectures and hands-on training [89]. A meta-analysis reported that PMI reduced challenging behaviors (standardized mean difference [SMD]=0.67) and hyperactivity (SMD=0.31) in children, as well as parenting stress (SMD=0.37) [90]. Recently, a study reported that telehealth- based PMI could be as effective as in-person PMI in reducing challenging behaviors [91]. Social distancing policies implemented during the COVID-19 pandemic have facilitated the adoption of long-distance service delivery methods, including telehealth and internet-based services. These are expected to continue even after the pandemic ends because of their cost-effectiveness and accessibility.

PMIs are also recommended for managing sleep-related problems, which are reported in 40%–80% of children with autism [92]. PMI for sleep typically includes psychoeducation and hands-on training in recording a daily sleep diary, setting up a sleep hygiene routine, and applying faded bedtimes to reduce sleep problems [93]. A review of 11 PMI studies found that PMI is effective in reducing sleep-related maladaptive behaviors, such as sleep resistance and waking at night, and in improving overall sleep time in children with autism [92].

In Korea, there have been a few PMI studies focused on reducing challenging behaviors. These studies have reported a decrease in challenging behaviors in children [94,95], as well as a reduction in parenting stress and an improvement in parental self-efficacy [94,96].

OTHER INTERVENTIONS

Speech-language therapy

Speech and language therapy (SLT) is the most common special education program for children with autism [97]. SLT is not a specific intervention but rather a general term encompassing a wide range of therapies implemented by speechlanguage pathologists (SLPs). Therefore, the targeted skill development areas of SLT are highly diverse, and SLPs may employ various heterogeneous intervention models depending on their trained practice area. A recent review reported that the developmental-naturalistic model is the most extensively studied approach, followed by ABA, in SLT for individuals with autism [97]. Therefore, it is recommended that parents seeking SLT services first evaluate whether the intervention goals of a prospective SLP are appropriate for their child’s needs and whether the intervention model employed is evidence-based [98].

Sensory integration therapy

Sensory-based interventions aim to enhance children’s levels of arousal and behavioral control through activities such as wearing weighted vests or swinging [99]. Although it is often used as an intervention to improve core symptoms in children with autism, there is currently limited evidence supporting its efficacy [99,100].

Developmental approaches

Developmental approaches prioritize child-led spontaneous play and joint attention with adults to enhance the communication skills of young children with autism. In Korea, the DIR/FloorTime [101] and Hanen’s More than Words [102] programs have been introduced, but scientific evidence for their effectiveness needs to be more consistent or sufficient [75,103].

Treatment and Education of Autistic and Related Communication-handicapped Children

According to Wong et al. [41], the CTM category includes the Treatment and Education of Autistic and Related Communication-handicapped Children (TEACCH). Along with ABA, it is one of the most widely recognized comprehensive intervention models [104]. The TEACCH emphasizes the importance of creating a highly structured environment and actively using visual support to facilitate communication and learning based on the theorized profile of the unique strengths and needs of individuals with ASD [100]. While TEACCH has been widely used in clinical settings, such as special education or clinics, evidence for its effectiveness has been mixed. An earlier meta-analysis reported limited effectiveness [105]. However, more recent research has reported that it can improve cognitive and language skills, as well as reduce stress in caregivers [106]. Further research is required to establish the effectiveness of TEACCH in autism interventions.

Social stories

Social stories teach social cognition and appropriate social behaviors for children with autism by using stories that describe specific social situations in simple sentences for children with autism [107]. Some single-subject design studies have found that social stories are effective in reducing inappropriate behaviors. However, recent systematic reviews have consistently concluded that there still needs to be more evidence to consider social stories as evidence-based practice [107,108].

Newly introduced interventions

Newly introduced interventions are considered promising, but further research is required to establish their evidence base. These interventions include PEERS, Pediatric Autism Communication Therapy (PACT), Research Units in Behavioral Intervention (RUBI), and Unstuck and Ontarget interventions. The PEERS program targets teenage adolescents to improve their social competence and social skills, leading to enhancements in social skills, knowledge, social responsiveness, and overall social skills, encompassing social communication and social cognition [109]. PACT uses specific video feedback techniques to help recognize, respond to, and enhance parent-child communication [110]. RUBI is a parent training program developed to decrease challenging behaviors in children with ASD [111]. Unstuck and On-target is a classroom-based intervention approach for high-functioning students that provides lessons promoting cognitive and behavioral flexibility in everyday situations [112].

FUTURE DIRECTION

The traditional model of evidence-based practice has primarily relied on comparing averages from group comparison studies, such as RCTs. Although this approach provides valuable information about the effectiveness of treatments, it does not reveal individual effects and differences. The mixed results often observed in group-comparison studies may be due to the heterogeneity of ASD symptoms. Hence, there is a growing need for personalized assessment and treatment approaches.

In recent years, advancements in information and communication technology and the Research Domain Criteria model have facilitated the collection of digital phenotyping, behavioral, and neurobiological data, and these datasets are rapidly accumulating in the ASD population. These developments promise to enable individualized assessment, treatment customization, and real-time monitoring. Consequently, as tailored assessments and interventions become more accessible in the near future, guidelines should reflect on the implications of this transformative shift.

Additionally, there have been recent developments in research models, such as the Autism Europe Project and the Autism Innovative Medicine Studies-2 (AIMS-2) Trials Project, which are currently underway. It is essential to incorporate the relevant results from these new models into future guidelines, as they demonstrate promising outcomes and provide solid research results that support their efficacy.

SUMMARY

CTM

• EIBI provides comprehensive intervention in most developmental areas, including language, cognition, self-help skills, social skills, and adaptive behaviors. It typically begins before the age of 3 and involves receiving intervention for at least two years. Previous studies have reported significant improvements in the outcomes of children receiving EIBI.

• In addition to the basic principles of EIBI, NDBIs, which emphasize interacting with children in natural settings and developing communication skills, have also been shown to improve language, imitation, and social initiation. However, more rigorous research is required to confirm the effectiveness of NDBIs.

Focused intervention

• Social skills: SST, which teaches specific social skills based on behavioral principles, has consistently been shown to be effective in improving social skills in individuals of all ages, including infants, children, adolescents, and adults.

• Social communication: PECS, which uses picture cards to teach communication skills, is a well-established intervention that is effective in teaching individuals how to make requests or initiate conversations. However, further research is needed to establish its effectiveness in speech development.

• Sleep: While the most effective techniques may vary depending on the specific sleep problem, behavioral interventions such as delaying sleep time, gradual extinction, and scheduled awakening have been reported to be effective in reducing sleep problems in individuals with autism.

• Eating problems: As with sleep problems, behavioral interventions have been shown to reduce eating problems among individuals with autism. Contingent rewards and NRSs are the most commonly used and reported to be effective.

• Toilet training: Behavioral interventions such as shaping, gradual guidance, scheduling, toileting, and stimulus control have been reported to be effective in toilet training in individuals with autism.

• Emotional difficulties: CBT is the most effective treatment for emotional problems such as depression, anxiety, and anger in individuals with autism. Assessment of verbal and cognitive abilities is required before CBT can be applied.

Parent education and other non-behavioral interventions

• PMI, for comprehensive early intervention, which trains parents to implement various intervention techniques with their children directly, has been widely used, either in combination with EIBI or as a standalone approach. In recent years, P-ESDM has emerged as an effective intervention method for improving functions in children, such as parent-child interactions and language.

• PMI for challenging behaviors, which teaches parentspecific behavioral skills to deal with the behaviors of their children, has consistently been shown to be effective in reducing challenging behaviors, hyperactivity, and parenting stress.

• When considering an intervention method with uncertain evidence, it is important to have thorough discussions between parents and experts using the most up-to-date scientific knowledge available.