Abstract

Autism spectrum disorder (ASD), described by impaired language acquisition, limited social relationships, and restricted and repetitive patterns affect 1 in 36 children in the United States. Applied Behavior Analysis (ABA) is a heavily backed practice to treat individuals with autism; specifically, early intensive behavioral intervention (EIBI) is a subcategory of ABA used to treat young children with ASD. This review aims to summarize and compare recent empirical studies examining the effectiveness of EIBI based on Dr. Ole Ivar Lovaas’s (a then UCLA professor who studied ABA) treatment model. An analysis of these studies reveals the importance of duration and intensity in EIBI treatment, with higher intensity and longer duration corresponding to increased IQ scores. Therefore, EIBI is a highly effective treatment for at least some children with ASD. However, future studies should also consider metrics other than IQ and look into the individual participant changes under EIBI rather than a group as a whole.

Keywords: Autism Spectrum Disorder (ASD), Applied Behavioral Analysis (ABA), Early Intensive Behavioral Intervention (EIBI)

I. Introduction

The term “autism” was first coined by the Swiss psychiatrist Eugen Bleuler in 1911 to describe individuals with schizophrenia who demonstrated a loss of contact with reality¹. In the 1940s, Leo Kanner (1943) and Hans Asperge (1944) independently described children with disorders in the areas of language acquisition, restricted and repetitive behaviors, and impaired social relationships^2’3. Kanner’s initial report described a group of 11 children with extreme autistic aloneness, whereas Asperge’s initial report portrayed a similar but less severely impaired group of children having autistic psychopathy. 

More than three decades later, autism was included in DSM-III (APA, 1980) under the category of Pervasive Developmental Disorders (PDDs). The definition of “infantile autism” in DSM-III emphasized characteristics of young children with autism⁴.

In DSM-IV (APA, 1994), the classification of PDDs includes the following subcategories: autistic disorder, Asperger’s disorder, pervasive developmental disorder not otherwise specified (PDD-NOS), Rett’s disorder, and childhood disintegrative disorder⁵. In DSM-V (APA, 2013), the conceptualization of autism reflects a shift from a multi-categorical diagnostic system to a single diagnostic system, autism spectrum disorders (ASD), based on multiple dimensions⁶. The change eliminated all other subcategories. 

About 1 in 36 children in the U.S. have autism, and 1 in 45 adults in the U.S. have autism⁷. In general, children with ASD demonstrate persistent deficits in the domains of social communication and social interaction, restricted, repetitive patterns of behavior, interests, or activities during the early developmental periods⁶. The deficits cause clinically significant impairment in social, occupational, or other important areas of current functioning and cannot be better explained by intellectual disabilities⁶.

A. What is Applied Behavior Analysis (ABA)?

Although a wide range of teaching programs, interventions, therapies, and dietary regimens are available to the public to help families of children with Autism, behavioral interventions based on Applied Behavior Analysis (ABA) principles have been documented to generate comprehensive, significant, and long-term improvements. It is one of the most cited and empirically validated evidence-based practices for treating individuals with autism. 

ABA is a type of intensive, structured, and systematic teaching program based on the behaviorism rooted in B.F. Skinner in 1938⁸. The first successful implementation of ABA approaches in children with autism was dated back in 1967. However, ABA did not gain its visibility and popularity until the 1990s. Most ABA lessons are taught in a way that breaks the teaching contents into simple elements. These elements are taught to a learner using 1:1 discrete trials in which the child is presented with a stimulus (e.g., “touch the cup”). Once children demonstrate correct responses and behaviors, positive reinforcement (e.g., preferred edible items, toys, activities) is provided to maintain and reinforce the correct responses. If a learner demonstrates incorrect responses and inappropriate behaviors, the instructor ignores such responses and continues to prompt and reinforce correct and appropriate responses. Over time, as the children master more skills, reinforcers will upgrade from primary reinforcers (e.g., drinks, snacks) to secondary reinforcers (e.g., verbal praises, thumps up). The ABA teaching approach also emphasizes skill generalization and skill maintenance. The teaching contents might start with simple tasks (e.g., body gestures, table manners, identifying colors, joint attention) to more complex tasks (e.g., language imitation, social skills). Children with autism often demonstrate very distinguished cognitive profiles and individual strengths and weaknesses. Thus, ABA programs are often individualized to meet each learner’s unique characteristics. 

B. Early Intensive Behavioral Intervention (EIBI): Lovaas Treatment

Is ABA equally effective in children with autism versus older learners with autism? The answer is probably No. Ivar Lovaas, then a professor at UCLA, published a paper in 1973 based on his intensive ABA treatment of 20 children with autism. The conclusion pinpointed the three most prominent predictors of treatment gains: intensity of treatment, family involvement, and child age⁹. After that, Lovaas devoted his efforts to early and intensive behavioral intervention (EIBI) that heavily involved families. In 1987, Lovaas published a groundbreaking paper based on his work with 19 children who participated in a minimum of two-year intensive behavioral treatment (an average of 40 hours of ABA treatment per week), with two control groups¹⁰.

The children in the experimental group had an average age of 32 months during the intake. At the follow-up, nine (47%) of the 19 children showed normal intellectual and educational functioning, eight (42%) obtained an average IQ score in the mild intellectual disabilities range, and two children were placed in classes for autism/intellectual disabilities, with an IQ in the profound range. Compared to the control group 1, the experimental group showed an overall IQ increase of 30 points, whereas the experimental group and the control group 1 showed comparable baseline data. Nevertheless, Lovaas’ study was not left unchallenged. Lovaas was asked by skeptical reviewers to collect more control group data and then was asked whether the findings could be replicated in other centers. In the next ten years, his efforts were reflected in several replication studies (e.g., Eikeseth et al. 2007; Sallows & Graupner, 2005)^11’12.

II. Research questions and methods

This paper aims to summarize recent empirical studies examining EIBI rooted in the Lovaas treatment model to compare and contrast the key elements of these studies. A few research questions are listed below. First, what were the characteristics of the participants involved in these EIBI studies? Second, what were the key elements of these EIBI studies? Third, what were the main outcomes of these EIBI studies?

In the following section, a literature review was conducted to summarize, compare, and contrast major empirical studies based on the 1987 Lovaas study. The search criteria included the following keywords: early, intensive, behavioral intervention or ABA, and children. Studies involving early intensive behavioral intervention but not following Lovaas’s treatment model were excluded. The databases included PsycINFO, Academic Search Complete, ERIC, MEDLINE, ProQuest, and Education Database. Eleven studies met the criteria and closely followed Lovaas’ model for intervention. Three of the studies had follow-up data collected years after the initial treatment. 

III. Review of Empirical Studies Utilizing Early Intensive Behavioral Intervention (EIBI) Based on Lovaas Model

A. Characteristics of Participants 

All of the studies recruited very young participants, ranging from 30 to 53 months old at the time of intake evaluation. Participants in the control groups had comparable ages. The diagnosis categories included autism spectrum disorders, children with autism, and pervasive developmental disorders. Most studies focused on children with mean IQs ranging from 50 to 80. Eldevik et al. (2006) focused on children with a baseline IQ in the 40s¹³. Magiati et al. (2007) focused on children with a baseline IQ in the 80s¹⁴. Not all studies reported the ages of diagnosis and the ages during which the intervention started. Most of the studies recruited 11 to 20+ participants for the treatment groups and a similar number of participants for the control groups. 

B. Characteristics of Intervention 

The key elements of the 11 studies were summarized in Table 1. Of the 11 studies, three provided EIBI in less than two years. The rest of them provided EIBI for at least two years. Most studies had an intensity of 25 to 40 hours of EIBI per week, whereas only one study provided 12.5 hours of EIBI per week¹⁵. The report of actual hours of intervention was variable, as some studies had very accurate documentation of therapy hours for their participants, and some studies only provided an estimation. Four studies reported that EIBI (i.e., Smith et al., 2000; Eikeseth et al., 2002, 2007; Sallows & Graupner, 2005) was more intensive during the first year, and the intensity decreased the following year^{11’12’16’17}. All children in the EIBI groups received similar treatment, which varied in intensity and duration; the children in the control groups received variable treatments, such as parent-directed invention, eclectic public-school service, and mixed interventions. Three studies involved two control groups, and the rest involved one control group. 

C. Outcomes of Intervention 

The most consistent data that can be compared across studies were IQ scores reported in these 11 studies. The increase in IQ score varies from 30 points to non-significant changes. Nine studies reported a significant mean increase in IQ scores in the EIBI group. Only two studies reported non-significant changes from intake to post-treatment evaluations (i.e., Eldevik et al., 2006; Magiati et al., 2007)^15’14. It is worth noticing that Eldevik’s study recruited participants with IQs in the relatively low range (the 40s), and the EIBI intensity was low (i.e., 12.5 hours/week), whereas Magiati’s study recruited participants with relatively high IQs (in the 80s). Many studies reported data on other measures, such as language, speech, symptom severity, and adaptive functioning, with the outcome data favoring the EIBI groups. Some studies reported outcome measures at time 1 and time 2. The data presented in Table 1 suggest that most gains from EIBI are made during the first year, and fewer gains are made during the subsequent year.

IV. Discussion

The present review examined eleven studies utilizing EIBI based on the Lovaas model among young children with ASD. All studies involved the treatment and control groups. Collectively, EIBI is highly effective for at least some children with ASD, evident in the significant IQ increases among nine out of the 11 studies. It is critical that EIBI starts when the children are very young, such as at the ages of 2 to 4 years old.  It remains clear that the intensity and duration of EIBI affect the treatment outcomes, with higher intensity and longer duration corresponding to better outcome measures as a general trend. EIBI enhances not only the cognitive abilities of children with ASD but also other domains, such as language, speech, adaptive functioning, and daily living skills. 

However, there are several limitations to this paper. First, although the IQ scores were relatively universal across all studies reviewed, there were substantial variations in terms of what instruments were used to measure IQ, such as some used Wechsler scales (Lovaas, 1987; Sheinkopf & Siegel, 1998) and others used Stanford-Binet and Bayley (e.g., Smith et al., 2000)^10’16’18. It remains unclear why IQ scores should be the most prominent in evaluating treatment outcomes. There were more variations in the measures of other skills, such as language, communication, and adaptive functioning. The scores derived from different testing instruments made the comparison difficult to conduct. 

Second, although there was relatively intact data about the hours of EIBI received by the treatment groups, some studies provided data based on actual documentation of therapy hours, and some provided estimations of such hours. Data were much less rigorous about the hours of other therapy or services received by the control groups. 

Third, although the treatment outcomes were generally positive for the EIBI groups, there were considerable individual variations among the participants. The present review only examined the group-based outcomes without further examinations of individual responses to intervention. Future researchers are encouraged to present individual developmental patterns under the EIBI condition and examine predictors that affect outcome measures (e.g., baseline IQ, language skills at intake, social relatedness at intake).

V. Conclusion

The data presented in this review suggest that EIBI shows promising outcomes in young participants. The intensity and duration of EIBI largely affects its effectiveness. Many studies did not utilize randomized controlled trials, which should be encouraged in future research. There are substantial variations of variables collected across different empirical studies, and more standardized assessments should be implemented. Individual responses to EIBI warrant further examination. Overall, this paper underscores the importance of implementing EIBI among young children with ASD.

Acknowledgment

Thank you for the support from Dr. Eric Rozenblat and Dr. Donna De Feo, who have provided me with opportunities to interact with trainers and learners at the Institute of Educational Achievement (IEA). IEA is a dissemination site of the Princeton Child Development Institute (PCDI), a world leader in research and autism treatment. IEA is also a member of the Alliance for Scientific Autism Intervention. At IEA, highly individualized instruction based on the science of applied behavior analysis is provided to each learner. 

Table I. Summary of Intervention and Participants Details

Study	Duration of EIBI Treatment	# for Treatment and Control Groupsa	Age at Intake (months)a	Hours of EIBI	Intake IQa	Follow-up IQa	Other Areas
Lovaas, 1987, McEachin et al., 1993	Minimum of 2 years	T=19, C1=19, C2=21	T=34.6, C1=40.9	40 hours/week	T=53 C=46	T=83.3 C1=52.2 C2=57.5	Higher adaptive behaviors and personality scores
Smith et al., 1997	2 or more years	T=11, C=10	T=36, C=38	30 hours or more/week	T=28 C=27	T=36 C=24	More expressive speech
Sheinkopf & Siegel, 1998	16 months	T=11, C=11	T=33.8 C=35.3	27.02 hours/week	T=62.8 C=61.7	T=89.7 C=64.3	Significant effects on symptom severity
Smith et al., 2000	2 or more years	T=15 C=13	T=36.07 C=35.77	24.52 hours/week for year 1, then reducing	T=50.53 C=50.69	T=66.49 C=49.67	Improved visual-spatial skills, language, and academics; no improvement in adaptive functioning or behavioral problems
Eikeseth et al., 2002, 2007	31.4 months	T=13 C=12	T=31.4 C=33.3	28 hours/week for year 1, 18 hours/week for year 2	T=61.92 C=65.17	T=79.08 at FU1, T=86.9 at FU2, C=69.50 at FU1, C=71.9 at FU2	Fewer aberrant behaviors and social problems
Sallows & Graupner, 2005	4 years	T=13 C=10	T=35 C=37	39 hours/week for year 1, 37 hours/week for year 2, then decreasing afterward	T=50.85 C=52.10	T=73.08 C=79.60	Outcomes were similar for both the treatment group and the control group.
Howard et al., 2005, 2014	14 months	T=29 C1=16 C2=16	T=30.86 C1=37.44 C2=34.56	25-40 hours/week	T=58.54 C1=53.69 C2=59.88	T=89.88 C1=62.13 C2=68.81	Differences were significant for cognitive, non-verbal, receptive, expressive, communication, self-help, and social skills.
Cohen et al., 2006, 2014	3 years	T=29 C1=16 C2=16	T=30.86 C1=37.44 C2=34.75	25-30 hours/week	T=60.57 C1=53.69 C2=61.00	T=89.43 at Yr3 C1=64.43 at Yr3 C2=71.77 at Yr3	Better on measures of cognitive, language, and adaptive functioning
Eldevik et al., 2006	20 months	T=13 C=15	T=53 C=49	12.5 hours/week	T=41.0 C=47.2	T=49.2 C=44.3	Did not make significant progress in adaptive behaviors, except in communication
Magiati et al., 2007	2 years	T=28 C=16	T=38.0 C=42.5	32 hours/week	T=83.0 C=65.2	T=78.4 C=65.3	No significant group differences
Remington et al., 2007	2 years	T=23 C=21	T=35.7 C=33.6	20-30 hours/week	T=61.43 C=62.33	T=68.78 at Yr1 T=73.48 at Yr2 C=58.90 at Yr1 C=60.14 at Yr2	Robust differences in intelligence, language, daily living skills, and positive social behaviors

Reference

1. Bleuler E. In: Dementia Praecox or the Group of Schizophrenias. (Trans. J. Zinkin) Zinkin J, editor. New York, NY: International Universities Press; 1950. [↩]
2. Kanner, L. (1943). Autistic disturbances of affective contact. Nervous Child, 2, 217-250 [↩]
3. Asperger, H. (1944). Die “autistichen Psychopathen” im Kindersalter. Archive fur psychiatrie und Nervenkrankheiten, 117, 76–136. doi: 10.1007/BF01837709. [↩]
4. American Psychiatric Association (1980) Diagnostic and statistical manual of mental disorders (3rd Edition) (DSM-III). American Psychiatric Publishing, Inc. [↩]
5. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th Edition) (DSM-IV). [↩]
6. American Psychiatric Association, DSM-5 Task Force. (2013). Diagnostic and statistical manual of mental disorders: DSM-5™ (5th ed.). American Psychiatric Publishing,Inc.. https://doi.org/10.1176/appi.books.9780890425596 [↩] [↩] [↩]
7. Centers for Disease Control and Prevention. (2023). Data and statistics on autism spectrum disorder. Centers for Disease Control and Prevention. https://www.cdc.gov/autism/data-research/index.html  [↩]
8. Skinner, B. F. (1938). The behavior of organisms: an experimental analysis. Appleton-Century. [↩]
9. Lovaas, O. I., Koegel, R., Simmons, J. Q., & Long, J. S. (1973). Some generalization and follow-up measures on autistic children in behavior therapy. Journal of Applied Behavior Analysis, 6(1), 131–166. https://doi.org/10.1901/jaba.1973.6-131 [↩]
10. Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55(1), 3–9. https://doi.org/10.1037/0022-006X.55.1.3 [↩] [↩]
11. Eikeseth, S., Smith, T., Jahr, E., & Eldevik, S. (2007).Outcome for children with autism who began intensive behavioral treatment between ages 4 and 7: A comparison controlled study. Behavior Modification, 31(3), 264–278. https://doi.org/10.1177/0145445506291396 [↩] [↩]
12. Sallows, G. O., & Graupner, T. D. (2005).Intensive behavioral treatment for children with autism: four-year outcome and predictors. American journal of mental retardation : AJMR, 110(6), 417–438. https://doi.org/10.1352/0895-8017(2005)110[417:IBTFCW]2.0.CO;2 [↩] [↩]
13. Eldevik, S., Eikeseth, S., Jahr, E., & Smith, T. (2006). Effects of low-intensity behavioral treatment for children with autism and mental retardation. Journal of autism and developmental disorders, 36(2), 211–224. [↩]
14. Magiati, I., Charman, T., & Howlin, P. (2007).A two-year prospective follow-up study of community-based early intensive behavioural intervention and specialist nursery provision for children with autism spectrum disorders. Journal of Child Psychology and Psychiatry, 48(8), 803–812. https://doi.org/10.1111/j.1469-7610.2007.01756.x [↩] [↩]
15. Eldevik, S., Eikeseth, S., Jahr, E., & Smith, T. (2006). Effects of low-intensity behavioral treatment for children with autism and mental retardation. Journal of autism and developmental disorders, 36(2), 211–224. https://doi.org/10.1007/s10803-005-0058-x [↩] [↩]
16. Smith, T., Groen, A. D., & Wynn, J. W. (2000). Randomized trial of intensive early intervention for children with pervasive developmental disorder. American Journal on Mental Retardation, 105(4), 269–285. https://doi.org/10.1352/0895-8017(2000)105<0269:RTOIEI>2.0.CO;2 [↩] [↩]
17. Eikeseth, S., Smith, T., Jahr, E., & Eldevik, S. (2002). Intensive behavioral treatment at school for 4- to 7-year-old children with autism: A 1-year comparison controlled study. Behavior Modification, 26(1), 49–68. https://doi.org/10.1177/0145445502026001004 [↩]
18. Sheinkopf, S. J., & Siegel, B. (1998). Home-based behavioral treatment of young children with autism. Journal of autism and developmental disorders, 28(1), 15–23. https://doi.org/10.1023/a:1026054701472 [↩]