Saturday, January 06, 2007

ABA Most Effective In Educating Autistic Children

In New Brunswick an interdepartmental committee was struck (Health, Education and Family Services) to study services available for autistic persons in New Brunswick and make recommendations. It sat from 1999 to 2001 and issued its report in November 2001. Its report then went unread by the lead Minister of the day, Health Minister Elvy Robichaud, for a further year. The committee studiously avoided making any recommendations about the effectiveness of any particular intervention.

The criteria for "Community" Autism Centres seeking funding from the government has required that these centres not promote any specific intervention and instead provide information on various interventions. Meanwhile our good neighbors in the State of Maine took a different approach. There the stakeholders group involved in education of autistic children examined the professional literature concerning efficacy of autism interventions and concluded in the MADSEC Autism Task Force Report 1999, revised Feb. 2000, that ONLY Applied Behavior Analysis (ABA) could be said to be evidence based effective intervention for educating autistic children. Since then specific studies have confirmed the superiority of ABA in educating autistic children. The Lovaas site has an excellent summary of the most significant of the hundreds of studies which substantiate ABA as the intervention of choice in educating autistic children:

Research Articles

Between 1985 and 2005, there were over 500 articles published concerning Applied Behavior Analysis and autism. Here is some of the key information from those sources.

1987 study by Dr. Lovaas

Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55, 3-9.

Original research in peer-reviewed journals indicating that 90% of children substantially improved when utilizing the Lovaas Model of Applied Behavior Analysis, compared to the control group. Close to half attained a normal IQ and tested within the normal range on adaptive and social skills.

1993 follow up study

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 (4), 359-372.

Follow-up research in early adolescence showed that children in the 1987 study maintained their skills and could succeed in life without costly special education and residential services.

1998 cost-benefit analysis of intensive, early behavioral intervention.

Jacobson, John W., Mulick, James A., & Green, Gina. (1998
). "Cost-Benefit Estimates for Early Intensive Behavioral Intervention for Young Children with AutismÑGeneral Model and Single State Case." Behavioral Interventions, 13, 201-226.

Demonstrates that providing behavioral treatment to all children with autism for three years, delivered between the ages of 2 to 6 years, would save approximately $200,000 per child for ages 3-22 years and up to $1,000,000 per child for ages 3-55 years. The savings per child even takes into account that some children will not benefit at all from behavioral treatment and some will only make modest gains.

1999 report from the New York State health department
Recommends a minimum of 20 hours per week of individualized behavioral interventions using ABA techniques, based on strong evidence in the research

2000 report from the Surgeon General
Acknowledges the efficacy of applied behavioral methods in the treatment of children with autism and cites the "well-designed study carried out by Lovaas and colleagues"

2002 study by Dr. Eikeseth directly comparing two treatments

Eikeseth, Svein, Smith, Tristram, & Eldevik, Erik Jahr Sigmund. (2002). Intensive Behavioral Treatment at School for 4- to 7-Year-Old Children with Autism. Behavior Modification, 26, 49-68.

Demonstrates that a focused behavioral treatment program is far superior than an eclectic special education approach that uses a variety of treatments. (Children in both groups received the same number of hours of treatment by qualified personnel.)

2005 replication study by Dr. Jane Howard directly comparing two treatments

Howard, Jane S. , Sparkman, Coleen R., Cohen, Howard G., Green, Gina, & Stanislaw, Harold. (2005). A Comparison of Intensive Behavior Analytic and Eclectic Treatments for Young Children with Autism. Research in Developmental Disabilities, 26 (4), 359-383.

Again demonstrates that a focused behavioral treatment program is far superior than an eclectic special education approach that uses a variety of treatments. (Children in both groups received the same number of hours of treatment by qualified personnel.)

2005 study by Dr. Sallows, replicating the 1987 Lovaas study results

Sallows, Glen O. & Graupner, Tamlynn D. (2005). Intensive Behavioral Treatment for Children with Autism: Four-Year Outcome and Predictors. American Journal on Mental Retardation, 110 (6), 417-438.

Replication study of the Lovaas Model of Applied Behavior Analysis by an independent author. Dr. Sallows states, "We found that 48% of all children showed rapid learning, achieved average posttreatment scores, and at age 7, were succeeding in regular education classrooms. These results are consistent with those reported by Lovaas and colleagues (Lovaas, 1987; McEachin, Smith, & Lovaas, 1993)."

2006 study by Dr. Cohen Howard, again replicating the 1987 Lovaas study results

Cohen, Howard, Amerine-Dickens, Mila, Smith, Tristram. (2006). Early Intensive Behavioral Treatment: Replication of the UCLA Model in a Community Setting. Journal of Developmental & Behavioral Pediatrics, 27 (2), 145-155.

Another replication study of the Lovaas Model of Applied Behavior Analysis by an independent author. Children in behavioral treatment scored significantly higher in IQ and adaptive behavior scores than the comparison group. Further, 29% (6 of 21) children were fully included in regular education without assistance and another 52% (11 of 21) were included with support. This compares to only 5% (1 of 21) children in the control group who were placed in regular education.

Despite this wealth of literature supporting the efficacy of ABA as an educational intervention for autistic children there remains opposition to its use for that purpose. There are many ulterior motives some cost based others emotion based. Arguments are also advanced on the basis that studies generally are not based on a random assignment to control groups. What these criticisms ignore are that as the trial judge in the Auton case concluded after reviewing the expert evidence in that case, whether done by design or not the assignments did not indicate any bias in group assignment - the assignments were not substantially different than they would have been under pure random assignment. Such criticisms also gloss over the ethical problems posed by random assignment to groups in which children would receive educational interventions which are generally believed not to be effective in educating autistic children.

The criticisms also ignore the reality that children have to be educated now. Parents will not accept waiting another generation for academics and anti-ABA die-hards to be convinced. Our children need to be educated now. Hundreds of studies support the efficacy of ABA in educating autistic children. It is long past time to give our children the opportunity of a real education and development to the best of their potential.


mcewen said...

I suspect [no evidence] that States vary greatly in their provision of services. Here [CA] there are any number of experts willing and able to 'diagnose / make recommendations' and the like [at a price] but finding the services thereafter is quite a different matter. The waiting lists are months long with no guarantees at the end [that they will take you child].
Sounds like we need Federal action - or is that a dirty word?
Best wishes

Autism Reality NB said...

Here in Canada the federal government has declined involvement in autism issues because health care & education are within provincial jurisdiction. This notwithstanding the development of co-operative federalism in any number of areas within provincial jurisdiction. That may change now with Andy Scott's private members motion calling for a National Autism Strategy endorsed by all parties except the separatist Bloc Quebecois.

mom in California said...

My son started full time ABA two months ago. I couldn't be happier. His language has skyrocketed, he's able to make choices and answer questions (instead of just repeating what we've asked). Oh, and did I mention that now he's letting us brush his teeth.

Kristina said...

And BTW my son has been exposed to ABA - to his great benefit. By use of ABA we have given him language and communication skills, reduced self injurious behaviors and permitted him to function in a school environment. And yes we have used ABA for such unmentionables as toilet training.

I am quoting this from your comment on the "Autism Crisis" blog. I think you left a comment on Autism Vox regarding our having some differences of opinion on some topics, but the benefits of ABA you describe for your son have been the same for mine. It would have been unethical for us, as Charlie's parents, to have not tried to teach him in the ways we have discerned to be most effective for him.

Alyric said...


Your list of references is well known. Since you do know of them, then perhaps you might care to explain the omission of the one and only randomised controlled trial of ABA. It has been said before but seems to bear repeating that truly randomising intake and carefully controlling the experiment is the only scientific way to assess what is really happening rather than what parents seem to want to believe is happening. On that scientific measure, it seems that ABA does not perform any miracles - far from it. See "Smith, T., Groen, A. D., Wynn, J. W. (2000). Randomized trial of intensive early intervention for children with pervasive developmental disorder. American Journal of Mental Retardation, 105, 269-85."

I also refer your readers to both Jane Howard's study - which is really not good science and Sallows and Graupner's study. Have a look at the poorly performing. A significant minority of these children didn't just do badly, they got worse. Governments should not be in the business of mandating treatments where a reasonable hypothesis could be made that this treatment is harmful for some children.

I also take issue with your statement that Sallows and Graupner is a replication of Lovaas. This is untrue and something much worse than that. Sallows results showed very clearly that the intervention - in this case a version of ABA (not strictly Lovaas by the way)was likely to be irrelevant in their outcomes. It made no difference what the therapists were doing. Reason being - Sallows found that less ABA and less professionally supervised ABA gave far and away the best outcomes. if ABA was the thing that was crucial, you would get the opposite result. So what Sallows did then was what was really important. He tried to match up the characteristics of the autistic children with the outcomes. Gee wow and what a concept!! Imagine - they started to look at the kids rather than the environment and found some really interesting trends.

There's another institution also looking at the characterstics of autistics - Dr Mottron's outfit at the University of Montreal. Dr Gernsbacher at the University of Wisconsin is doing this as well.

Question for Kristina and Harold? Has your ABA therapist shown any interest in keeping up with these developments and incorporating them into what they do? I have not seen that ABA therapists have any great motivation to improve their practice - mostly because according to their precepts (they don't actually have any theories) what they do now is all that is necessary. One of the autistics testifying before the Canadian Senate said that ABA needed adjustment because their methods didn't seem to follow how autistics think. She did, however, think that adjustment was possible. That would be nice, if it indeed happened. The obstacle would be ABA itself.

Autism Reality NB said...

Hello Alyric

The comments about the various studies listed are from the Lovaas site, they are not my original remarks, and I also provided thel ink to the Lovaas site. With respect to the Smith, Groen and Wynn remarks about truly randomising control groups I believe that position was discussed in the Auton decision. Even the experts who testified for the government in that case stated that failure to follow that approach did not negate all conclusions that could reliably be drawn from studies which were not based on "true" randomisation. Your reference to the Smith, Groen, Wynn study is interesting since it actually confirmed gains in autistic children who received ABA intervention.

You have offered a shotgun list of criticisms of the various studies listed on the Lovaas site without providing any published references to support them. I will offer no response to your characterization other than to say that I have not seen the "bad science" characterizations that you use to describe these studies.

You have raised other issues including the unnamed autistic person who testified that ABA methods didn't seem to follow how autistics think. I wonder who that unnamed autistic person could have been? How does she know how ALL autistic persons think? And how does a person, autistic or not, know how non verbal, low intellect, low communication autistic persons ALL think especially since she obviously does not share their intellectual or communication limitations?

As for my son's therapists, his ABA therapy involves more than straight DTT all the time if that is the intent of your question.

I appreciate you visiting this site and offering your insights and I sincerely ask you - what OTHER methodologies for educating autistic children enjoy as much support in the professional literature as ABA. I have no vested interest in ABA except that which it has offered our family in teaching our son. If you know of another empirically validated methodology I would appreciate being informed of such and reviewing the literature you offer to support your claim. As you know, the NY, California, Maine, US Surgeon General and Association for Science in Autism Treatment authorities have endorsed ONLY ABA as meeting an evidence based standard for educating autistic children. But if those learned authorities missed something I would be happy to know and use it to assist my son's education.

Best wishes.

Harold Doherty
Conor's Dad

Anonymous said...

Have you looked at Pivotal Response Treatment( Koegel& Koegel)which uses ABA procedures and a developmental approach?In Nova Scotia this is being used as a preschool intervention with great results.

Andrew said...

Harold, We're enjoying all aspects of your blog, particularly your tact. We have our own new site and blog w/ autism and behavioral leanings at

The erroneous, misguided assumption is that ABA is a methodology and that behavior analysts are some homogenous group with a predetermined agenda. Being a behavior analyst means you embrace science and allow the data you have to determine the course of intervention. We use single subject designs while our colleagues in psychology largely use inferrential statistics. Each has its own merits and its own weaknesses. If you are studying "ABA" you would never be studying just one thing.

I always wonder why people argue so vehemently against a scientific endeavor which has provided decades of effective interventions and glimmers of hope for individuals and their loved ones.