Early Intensive Behavioral Intervention for Children with Autism: What Does the Research Tell Us?
Adrienne M. Perry, Ph.D., C.Psych., BCBA, York University, Toronto
Background
Early Intensive Behavioral Intervention, often called Intensive Behavioural Intervention (IBI), is widely regarded as “best practice” for young children with autism (e.g., NYSDOH, 1999; Schreibman, 2000) because it is the strongest evidence-based treatment option among a myriad of treatments available for autism (Perry & Condillac, 2003). The purpose of this paper (and accompanying presentation) is to review briefly what the research tells us about this kind of intervention and, in particular, to describe some research on the Ontario IBI program, where this approach has been taken “to scale” in a large, community-based implementation.
Terminology is sometimes confusing for parents and professionals in this field. Applied Behaviour Analysis or ABA is the broader discipline or approach that includes many techniques of assessment and intervention, which can be helpful for many human problems. However, ABA is not limited to any age group or diagnosis and is not necessarily intensive. IBI, on the other hand, is defined as a specialized form of intervention designed for young children with autism, based on the principles of ABA. IBI involves very intensive teaching intended to alter children’s developmental trajectories in the hope that they will be able to function effectively in typical environments within two or three years.
IBI can be recommended based on research that has examined children’s response to intervention in terms of gains in cognitive functioning or IQ, adaptive behaviour, and language (and sometimes also diagnostic severity). The classic IBI study was reported by Lovaas (1987) who showed that very young children with autism receiving very intensive intervention for about two years made substantial gains compared to a control group of children receiving a minimal amount of intervention. In fact, 47% of the children demonstrated “best outcomes”, i.e., average cognitive functioning and the ability to participate in a regular education setting without support. Follow up with these children (McEachin, Smith, & Lovaas, 1993) indicated their gains were maintained over the longer term.
Since that classic study, other researchers have essentially replicated these findings of large gains in cognitive and adaptive functioning (e.g., Sallows & Graupner, 2005) in other small, model programs (see Handleman & Harris, 2008 for a description of such programs), sometimes in different settings, different countries, and with slightly older children. At this point, it seems reasonable to conclude that IBI has well-established efficacy for young children with autism or PDD-NOS compared to: (a) low intensity treatment (e.g., Smith, Groen, & Wynn, 2000); (b) “eclectic” community services (Cohen, Amerine-Dickens, & Smith, 2006; Howard, Sparkman, Cohen, Green, & Stanislaw, 2005); and c) equivalent amounts of high quality special education (Eikeseth, Smith, Jahr, & Eldevik, 2002; 2007; Howard et al., 2005), which tells us that it is not just the intensity but the behavioral technology underlying the intervention which seems to be crucial.
In treatment outcome research, there is an important distinction between “efficacy” (whether a treatment can be shown to work under ideal conditions) and “effectiveness” (whether a treatment works under routine clinical conditions). We know from the body of literature mentioned above that IBI has efficacy when evaluated in carefully controlled conditions (children may be selected, parents may be selected, therapists are carefully trained and supervised, treatment is implemented faithfully, etc.). However, we know much less about the “effectiveness” of IBI as it is typically implemented outside of these small, model programs, because such studies are rarely done. It might be expected that effectiveness will be more modest when children are less optimal candidates, training and supervision is less optimal, and so on (e.g., Bibby, Eikeseth, Martin, Mudford, & Reeves, 2002), although two small studies evaluating IBI as implemented in the community have showed very favorable results (Remington, et al., 2007; Sheinkopf & Siegel, 1998).
The Ontario IBI Program
IBI has been publicly funded in the province of Ontario, Canada, since the year 2000 through nine regional programs covering the entire province (415,000 square miles; population: 12 million). Service may be delivered via the public program or by approved private providers and may take place in a variety of settings (center-based, home-based, or in child care settings, etc.). Approximately 1,000 children are in service at any one time. For more information on the background to the Ontario program, please see Perry (2002) and Perry et al. (2008).
The Ontario Outcome Study
A study was recently published in the journal Research in Autism Spectrum Disorders (Perry et al., 2008) examining children’s outcomes in the Ontario IBI Program. It was a large, retrospective file review study. Available data from intake and exit assessments (diagnostic and developmental measures) were used together with program variables such as age and duration. It was a study of the effectiveness of IBI under “real-world” conditions in Ontario. These conditions were less optimal than those in the small, model programs cited above for several reasons: (a) the children were older and more severe; (b) parents were highly diverse socioeconomically and culturally; and (c) there were substantial capacity-building challenges (e.g., hiring and training) involved in mounting such a large program.
Measures. Measures included the Childhood Autism Rating Scale (CARS; Schopler, Reichler, & Renner, 1988), which is a standard observational measure often used in psychological/diagnostic assessments, measuring severity of autism symptoms; the Vineland Adaptive Behavior Scales (Sparrow, Balla, & Cicchetti, 1984), which is a parent interview measure regarding everyday skills in several domains of development (communication, self-help, social, and motor skills); and a cognitive or intellectual test administered to the child (various tests depending on the child’s age and ability level).
Participants. The data used in the study came from 332 files of children in the program (80% boys). Children were between 20 and 86 months old at intake, with an average age of about 4½ years. The duration of IBI received ranged from 4 months to 4 years with an average of 18 months. Many of the children had substantial developmental disabilities as well as autism. They were functioning, on average, at or below a 2-year level. However, there was a wide range of ability levels and so children were divided into 3 subgroups based on their initial level of functioning on the Vineland (Group A over 60, Group B in the 50s, and Group C below 50).
Results. (a) Autism Symptom Severity. Children showed significant reduction in autism symptom severity (CARS Total score). That is, children had less repetitive behaviour, related better to people, had better verbal and nonverbal communication skills, and improved imitation abilities, etc. at the time of exit compared to their initial scores. About half the children changed enough to fall into a milder category on this instrument. Of those who were in the mild/moderate autism range at intake, 41% improved so that they were in the non-autism range at exit. Of those in the severe autism range at intake, 59% improved to the mild/moderate range and 15% improved very substantially to the non-autism range.
(b) Cognitive and Adaptive Behaviour Level. Cognitive level (IQ and mental age) based on various cognitive tests, improved significantly for children, in some cases dramatically so (but these scores were unavailable for many children). Further, children gained significantly in developmental skills (increased age equivalents) in all areas of adaptive behaviour (communication skills, self-help skills, social skills, and motor skills). Standard scores, which are corrected for age, also increased significantly (though modestly) for communication and socialization, two key areas of difficulty for children with autism, but decreased slightly for daily living skills.
(c) Rate of Development. Children’s rate of development (based on the Vineland age equivalent scores) during IBI was approximately double their rate prior to IBI, and this was true for all three subgroups. This suggests that the developmental trajectory of children was altered during their participation in the IBI program.
(d) Range of Progress/Outcome. There was considerable heterogeneity in outcome, as would be expected given the population. Children were classified into seven categories of progress/outcome: average functioning, substantially improved, clinically significantly improved, less autistic, minimally improved, no change, and worse. The majority of children (75%) showed some measurable benefit or improvement during IBI. This included 11% of the children who achieved average functioning (similar to those described as “best outcome” in the efficacy literature). However, 25% did not seem to show improvement (the last two categories combined) at least on the available measures (some anecdotal evidence suggests that some of these children may have improved in problem behaviour).
Predictors of progress/outcome were also examined in this dataset and are included in a forthcoming paper (Perry et al., in preparation). In brief, children’s progress/outcome was clearly related to their initial functioning levels, on average, though not totally. Group A children showed outcomes relatively similar to those in studies from model programs. Children who started IBI before age four did better than those who started after age four on all scores at the exit assessment. Children with poor outcomes were not substantially different from other children in the sample, suggesting that other factors such as treatment quality or family variables may account for some of the variability.
There are both strengths and limitations of this study (as with any study). The primary limitation is that the study has no Comparison group, which means gains cannot be conclusively attributed to the IBI program, per se, as opposed to maturation, other treatments, or unknown factors. Other limitations are that there is no measure of treatment quantity or quality, family involvement, or problem behaviour; assessments were only at intake and exit rather than at regular intervals; assessments were not independent of treatment providers; and we have no follow-up information on whether gains were maintained. The principal strength of the study is that it is the largest (and one of the only) studies which demonstrates the effectiveness of IBI in a large and diverse community sample.
Research in Progress Extending these Findings
We have recently completed a waitlist comparison group controlled study in one region of the Ontario program (Flanagan & Perry, in progress), involving 67 matched pairs of children (equal on age and developmental and diagnostic severity at Time 1). Results indicate superior outcomes for children in the IBI group versus the waitlist group at Time 2 (correcting for the different duration and age at Time 2). Results for a subset of children (19 matched pairs) who are similar to children in the model programs were very similar to results from the efficacy studies with about half of these children classified as “best outcomes” (Flanagan, Perry, & Freeman, 2008).
Another study is examining the specific developmental trajectories of children during IBI, using the ABLLS data collected by program staff at multiple points during intervention (Sullivan & Perry, in progress). Preliminary results show that some children progress quickly initially and that initial mastery of imitation skills predicts skill level on subsequent assessments (Sullivan, Perry, Freeman, & Bebko, 2007).
We are also following up with children after they have been discharged from IBI (from one to five years) to see whether their gains are maintained (Prichard & Perry, in progress). Preliminary results suggest that children who progressed very well in IBI do seem to maintain their cognitive and adaptive gains as well as maintain their lack of autism symptomatology (Prichard & Perry, 2008).
Finally, we are beginning a new prospective study using a waitlist comparison group (Dunn Geier, Freeman, Perry, Barrowman, & Gaines, in progress). This study will address a number of the limitations of the previous research. It will include 60 children in IBI and 60 waitlist children in a stratified age-cohort design, all of whom will be assessed at Time 1 and again after 12 months of either IBI or “treatment as usual” while waiting for IBI. Measures will consist of standardized developmental and diagnostic measures, including cognitive, adaptive, language, and diagnostic severity. In the IBI group, there will also be measures of the quantity and quality of treatment received (using the York Measure of Quality of IBI [YMQI; Perry, Flanagan, & Prichard, 2008]) to rate monthly videotapes of IBI sessions) and a measure of parent involvement (Solish & Perry, 2008; Solish & Perry, in progress). These latter measures will help us address the issue of heterogeneity of outcomes more fully by examining the proportional contribution of child factors, family factors, and treatment factors.
References
Bibby, P., Eikeseth, S., Martin, N. T., Mudford, O. C., & Reeves, D. (2002). Erratum to “Progress and outcomes for children with autism receiving parent-managed intensive interventions”. Research in Developmental Disabilities, 23, 79-104.
Cohen, H., Amerine-Dickens, M., & Smith, T. (2006). Early intensive behavioral treatment: Replication of the UCLA model in a community setting. Developmental and Behavioral Pediatrics, 27, S145- S155.
Dunn Geier, J., Freeman, N. F., Perry, A., Barrowman, N., & Gaines, R. Prospective study of the effectiveness of Intensive Behavioral Intervention. Study in progress.
Eikeseth, S., Smith, T., Jahr, E., & Eldevik, S. (2002). Intensive behavioral treatment at school for 4-7-year-old children with autism: A 1-year comparison controlled study. Behavior Modification, 26(1), 49-68.
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.
Flanagan, H. E., Perry, A., & Freeman, N. L. (2008, November). Effectiveness of Intensive Behavioral Intervention: A waitlist comparison group design. Presentation at the Ontario Association for Behaviour Analysis annual conference, Toronto, ON.
Flanagan, H. E., & Perry, A. Effectiveness of Intensive Behavioral Intervention: A waitlist comparison group design. Doctoral dissertation in progress, York University, Toronto, ON.
Handleman, J. S., & Harris, S. L. (Eds.). (2008). Preschool education programs for children with autism (3rd ed.). Austin, TX: PRO-ED.
Howard, J. S., Sparkman, C. R., Cohen, H. G., Green, G., & Stanislaw, H. (2005). A comparison of intensive behavior analytic and eclectic treatments for young children with autism. Research in Developmental Disabilities, 26, 359-383.
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.
McEachin, J. J., Smith, T., & Lovaas, O. I. (1993). Long-term outcome for children with autism who received early intensive behavioral treatment. American Journal on Mental Retardation, 97, 359-372.
New York State Department of Health (1999). Autism/Pervasive Developmental Disorders. Clinical practice guidelines technical report. New York: Author.
Perry, A. (2002). Intensive early intervention program for children with autism: Background and design of the Ontario preschool autism initiative. Journal on Developmental Disabilities, 9(2), 121-128.
Perry, A., & Condillac, R. A. (2003). Evidence-based Practices for Children with Autism Spectrum Disorders. Toronto, ON: Children’s Mental Health Ontario.
Perry, A., Cummings, A., Dunn Geier, J., Freeman, N. L., Hughes, S., LaRose, L., Managhan, T., Reitzel, J., & Williams, J. (2008). Effectiveness of Intensive Behavioral Intervention in a large, community-based program. Research in Autism Spectrum Disorders, 2, 621-642.
Perry, A., Cummings, A., Dunn Geier, J., Freeman, N. L., Hughes, S., LaRose, L., Managhan, T., Reitzel J., & Williams, J. (in preparation). Predictors of outcome of Intensive Behavioral Intervention in a large, community-based program.
Perry, A., Flanagan, H. E., & Prichard, A. (2008). The York Measure of Quality of IBI (YMQI). Unpublished Manual, York University, Toronto, ON.
Prichard, A. E., & Perry, A. (2008, November). Short-term follow-up of children who received Intensive Behavioral Intervention. Presentation at the Ontario Association for Behaviour Analysis annual conference, Toronto, ON.
Prichard, A. E., & Perry, A. Short-term follow-up of children who received Intensive Behavioral Intervention. Doctoral dissertation in progress, York University, Toronto, ON.
Remington, B., Hastings, R. P., Kovshoff, H., degli Espinosa, F., Jahr, E., Brown, T, et al. (2007). Early intensive behavioral intervention: Outcomes for children with autism and their parents after two years. American Journal on Mental Retardation, 112, 418-438.
Sallows, G. O., & Graupner, T. D. (2005). Intensive behavioral treatment for children with autism: Four-year outcome and predictors. American Journal on Mental Retardation, 110, 417-438.
Schopler, E. Reichler, R. J., & Renner, B. R. The Childhood Autism Rating Scale (CARS). Los Angeles: Western Psychological Services.
Schreibman, L. (2000). Intensive behavioral/psychoeducational treatments for autism: Research needs and future directions. Journal of Autism and Developmental Disorders, 30, 373-378.
Sheinkopf, S. J., & Siegel, B. (1998). Home-based behavioral treatment of young children with autism. Journal of Autism and Developmental Disorders, 28, 15-23.
Sparrow, S. S., Balla, D. A., & Cicchetti, D. V. (1984). Vineland Adaptive Behavior Scales. Circle Pines, MN: American Guidance Service.
Solish, A. & Perry, A. (2008). The role of parent involvement in child outcome in IBI. Research in Autism Spectrum Disorders, 2, 728-738.
Solish, A. & Perry, A. Modeling predictors and correlates of parent involvement in IBI. Doctoral dissertation in progress, York University, Toronto, ON.
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, 269-285.
Sullivan, A., Perry, A., Freeman, N. l., & Bebko, J. (2007, May). Developmental trajectories of young children with autism enrolled in an IBI program: What the ABLLS can tell us about their progress. Presentation at the Association for Behavior Analysis, San Diego, CA.