Pediatric Services Pediatric Services: An intervention team serving children with developmental delays.

HomeParents' CornerParents' Corner ArchiveProfessional CornerProfessional Corner ArchiveCase in ProgressCase in Progress ArchiveInspirational MessagesInspirational Messages ArchiveDirect ServicesConsultingSeminars, Workshops, and MoreSpecial EventsRecommended ReadingRecommended Reading for ChildrenAsk the Experts News FlashCurrent Question and AnswerUnderstanding the LingoAbout the TeamTestimonialsFees, Location, and DetailsTypical Development: MakennaTypical Development: LaceyResourcesPrivacyStatementConfidentiality

Professional Corner

The ABA Controversy

Bernard Rimland, Ph.D. Wednesday, September 29, 1999

Dr. Rimland is a well-know parent advocate and his Autism Research Institute is home to the promotion of a variety of treatments of autism. These treatment have varying amounts of scientific support for their efficacy and safety. This is at the source of this controversy. This editorial is from the ARI website, http://www.autism.com/ari/editorials/aba.html.

I am a long-time and ardent supporter of what is now called the "ABA" (Applied Behavior Analysis) method of teaching autistic children. I remember very clearly the day in October 1964-35 years ago-that I first visited Ivar Lovaas in his clinic at UCLA. I met the autistic children Billie, Rickie, and Pam, who resided there.

Their speech was sparse and stilted, but the children were miles ahead of where they had been when they were filmed (this was before videotape) at intake. I spent the day with Ivar, and came away impressed. I returned home and started using "operant conditioning" with my then eight-year-old son Mark. He, too, began to improve.

A year later, in November 1965, I spoke to a group of parents in Teaneck, New Jersey, a suburb of New York City, and proposed that we start a national organization, the National Society for Autistic Children (now the Autism Society of America) dedicated to helping, not just baby-sitting, our children. The talk I gave was titled "Operant conditioning: breakthrough in the treatment of mentally ill children." (I said "mentally ill" because few were aware of the term "autism" then.) I traveled to city after city, giving the "Breakthrough" talk and starting a new chapter of the society in each city. My "Breakthrough" talk was translated into many languages in the '60s and '70s, and helped educate parents and educators around the world.

In 1987, when Lovaas' landmark study of ABA was published, we featured it in the ARI. I wrote dozens of letters of support for parents who wished to obtain ABA for their children, and in ARI Vol. 8, No. 3, in 1994, I published a generic letter of support (copies available on request) which was helpful, I'm told, to innumerable families fighting to get ABA for their children.

As a long-time advocate and supporter of ABA, I take a back seat to no one. Having said that, I must tell you that I am dismayed and appalled at the ludicrous position taking by many other supporters of ABA, who claim that ABA is the only scientifically validated treatment for autism. Not so! That position is not only false, it is absurd. Believe it or not, the Early Intervention Program of the New York State Department of Health has published a series of Clinical Practice Guidelines which makes that claim.

Considering the weight of scientific evidence, there are several treatment approaches which clearly meet the criterion of scientific validation, and of these, at least two surpass ABA in terms of scientific supportability. I will confine my comments here to the treatment modalities which most clearly exceed ABA in terms of level of scientific support (not necessarily in terms of percentage of children helped, nor in terms of the degree to which they are helped, but only in terms of weight of scientifically valid evidence that the treatment effect is real).

The New York State Guidelines, which recommend ABA as the only effective treatment, explicitly reject vitamin therapy, gluten- and casein-free diets, anti-fungal treatment, auditory integration training, sensory integration, and many other interventions. The Lovaas 1987 study, the centerpiece of the ABA early intervention movement, attracted a great deal of attention because it employed a control group of more-or-less equally impaired children who were given less intensive (fewer hours per week) treatment. This represented an important advance in methodology over no control group at all (the usual approach), but the study did not employ double-blind procedures. Those involved in the study knew how intensive the treatment was for each child. I am aware that it would have been exceedingly difficult, if not impossible, to develop a double-blind evaluation of intensive ABA treatment, but that does not change the fact that a double blind was not used. The results were thus to some extent contaminated by participant bias and expectancy. Further, all of the methods of measuring the effect of the treatments were to some degree subjective-and a source of error. No study of the effectiveness of ABA has used double-blind procedures or scientific laboratory analytical equipment.

In contrast, of the 18 studies showing vitamin B6 and magnesium to be effective, 11 employed the double-blind procedure. Further, in addition to using soft behavioral and observational criteria such as were employed in the ABA evaluation studies, 10 of the studies of B6 and magnesium measured the presence of abnormal substances in the blood and urine of autistic children, and found the B6/magnesium to have improved the children's metabolism.

Five studies of B6/magnesium in autism have shown normalization of brainwave activity in the autistic children. These are hard, objective, measurable, scientifically replicable findings. Show me the equivalent of such solid scientific evidence in the ABA literature. It is absent.

The situation is similar with regard to the efficacy of the casein- and gluten-free diet. Study after study has documented the presence of abnormal substances in the urine of autistic children, with improvement in the children's urine and in the children's behavior when the special diets are implemented.

There are over 40 such studies, yet the New York State report claims the diets are ineffective. Very strange. Recently an organization has been formed with the title Association for Science in Autism Treatment (ASAT). Their literature espouses the same nonsensical, counterfactual position as the aforementioned New York State report: ABA is the only scientifically valid intervention for autism. Their position is indefensible: it requires a distorted view of what science is all about, as well as a willingness to ignore all relevant evidence. The "ABA is the only way" folks are wrong, not only because of their lack of information about research on the validity of other interventions, but because of their failure to recognize that parents have a right and an obligation to consider all possible forms of intervention, including those which may not yet have won the stamp of approval of whatever person or committee feels qualified to pass judgment on candidate interventions.

A case in point: ABA itself. I can't help but wonder how the ABA-only folks would view my efforts to have ABA accepted between 1964 and 1987. Though there were no control group studies, and certainly no double-blind studies to point to, the evidence was clear enough to compel me to fight for ABA.

Should I have abandoned my efforts for more than two decades while waiting for a control group study to appear? I'm glad I didn't. It is a major mistake to think of ABA as being competitive with, rather than complementary to, many other interventions, particularly such biologically based interventions as vitamin B6 and magnesium therapy.

For years, our publications have urged parents who are about to undertake megavitamin therapy to refrain from mentioning the new intervention to teachers, therapists, grandparents, sitters, etc., so they would be able to obtain objective input from these "blind" observers. Each such case is a mini double-blind single-subject experiment.

We have heard from hundreds of parents who have reported, "Our therapist, who did not know about the B6 we started last week, said our child has made more progress in one week than in the prior three months" (see letters sections in back issues of the ARI). If you really want to be scientific, do a mini double-blind trial on your child.

Last Wednesday the FEAT Newsletter published an editorial by Autism Research Institute's Bernard Rimland Ph.D., which mentions briefly the A.S.A.T., a newcomer ABA-Autism related organization who has challenged some of the ARI's recommendations for autism treatment. Joan Davin, of A.S.A.T. sends in this response by Bud Newsom. His remarks are below interspersed into excerpts of Rimland's editorial.]

Newsom: Bernie Rimland is, indeed, one of the best friends parents have ever had. His 1964 book, Infantile Autism, was the first serious blow to the hegemony of psychoanalysis in autism, and he has been a tireless advocate for better treatments ever since. He has focused on orthomolecular approaches for the past 30 years, and, not surprisingly, gets his feathers ruffled when his efforts seem to be slighted. However, I think his editorial is a little over the top and requires some comment:

Rimland: "The Lovaas 1987 study, the centerpiece of the ABA early intervention movement, attracted a great deal of attention because it employed a control group of more-or-less equally impaired children who were given less intensive (fewer hours per week) treatment." No. It attracted a great deal of attention because it showed that if treatment started early enough and was intensive enough, massive gains could be achieved for some children and very good gains for most.

The control group was necessary for scientific reasons, but if the experimental group children had shown only 4- or 5-point increases in IQ over the control group, the study would have been another yawner about statistically significant but clinically useless findings.

Rimland: "This represented an important advance in methodology over no control group at all (the usual approach) but the study did not employ double-blind procedures." Double-blind procedures are commonly used in drug studies and other kinds of studies where the effects are likely to be weak and/or the strong possibility of observer bias in measuring the results exists.

When ABA has effects, they are typically quite strong (by the usual statistical criteria) and are so obvious that observer bias is not a serious problem (and is handled with reliability checks). Interestingly, though, the 1987 study does have a kind of natural single-blind, in that the teachers who promoted the children to the second grade were purposely kept unaware of the child's earlier autism diagnosis and their intensive treatment. That they perceived the children as normal is an important control for the experimenters' bias. Further, the testing at age 13 for the follow-up children was carried out independently (i.e., by graduate students and psychologists not associated with the project and not informed of the children's histories; see the 1993 study). Finally, in the Sheinkopf and Siegel study, the IQ assessments were also done by psychologists who were not aware the children had been exposed to intensive treatment until after they did the testing.

Rimland: "Those involved in the study knew how intensive the treatment was for each child. I am aware that it would have been exceedingly difficult, if not impossible, to develop a double-blind evaluation of intensive ABA treatment, but that does not change the fact that a double blind was not used. The results were thus to some extent contaminated by participant bias and expectancy." I don't understand the retreat to Schoplerian, trivial variables here. As Dr. Rimland knows very well, if all it took to achieve great gains in autistic children were bias and expectancy, autism would have been solved long ago.

Rimland: "Further, all of the methods of measuring the effect of the treatments were to some degree subjective- and a source of error." Only to the extent that widely-used intelligence tests have these limitations (standard errors), which are well established and are not a factor when large differences are obtained.

Rimland: "No study of the effectiveness of ABA has used double-blind procedures or scientific laboratory analytical equipment." I've already commented on blindedness above. As for laboratory equipment, most people will take highly visible, clinically-significant behavior changes over blips in a blood sample any day.

Rimland: "In contrast, of the 18 studies showing vitamin B6 and magnesium to be effective, 11 employed the double-blind procedure. Further, in addition to using soft behavioral and observational criteria such as were employed in the ABA evaluation studies," Actually, scores from well-standardized intelligence scales are considered "hard" data in psychology. Average ratings from subjective rating scales completed by nurses, as used in B6 studies, are considered "soft."

Rimland: "10 of the studies of B6 and magnesium measured the presence of abnormal substances in the blood and urine of autistic children, and found the B6/magnesium to have improved the children's metabolism. Five studies of B6/magnesium in autism have shown normalization of brainwave activity in the autistic children. These are hard, objective, measurable, scientifically replicable findings." How do "abnormal substances in the blood and urine" and "brainwave activity" relate to clinically-significant improvements in the lives of the children? That these findings are valid is not the issue; the issue is, are they important?

Rimland: "Show me the equivalent of such solid scientific evidence inthe ABA literature. It is absent." First, ABA investigators are trying to show substantive changes in behaviors, not laboratory curiosities of who-knows-what-if-any significance. Second, ABA research relies on multiple observers and reliability checks in most applied settings, the standard in all areas of science where the phenomenon of interest can't be put on a switch. Thus, it is fully as solid as is needed for the questions it addresses.

Rimland: "The situation is similar with regard to the efficacy of the casein- and gluten-free diet. Study after study has documented the presence of abnormal substances in the urine of autistic children, with improvement in the children's urine and in the children's behavior when the special diets are implemented.

There are over 40 such studies, yet the New York State report claims the diets are ineffective. Very strange." Not at all strange, once you recall that most of those studies have no controls at all and show relatively minor effects. At best, in a few cases, they seem to help the child benefit from intensive teaching, so constitute "adjuncts," not "primary" treatments.

Rimland: "An organization has been formed with the title Association for Science in Autism Treatment (ASAT). Their literature espouses the same nonsensical, counterfactual position as the aforementioned New York State report: ABA is the only scientifically valid intervention for autism. Their position is indefensible: it requires a distorted view of what science is all about, as well as a willingness to ignore all relevant evidence." The evidence is not being ignored; it's being judged minor. Would you really counsel anyone to do B6/mag or cf/gf diets INSTEAD of ABA? No; at most, "along with."

Rimland: "The 'ABA is the only way' folks are wrong, not only because of their lack of information about research on the validity of other interventions, but because of their failure to recognize that parents have a right and an obligation to consider all possible forms of intervention, including those which may not yet have won the stamp of approval of whatever person or committee feels qualified to pass judgment on candidate interventions."

Nothing is the only way for every child. Those groups' recommendations are about what generally works best to achieve the largest gains with the most kids given the current state of our scientific knowledge. The hundreds of studies supporting ABA procedures far outweigh all contenders combined.

Rimland: "A case in point: ABA itself. I can't help but wonder how the ABA-only folks would view my efforts to have ABA accepted between 1964 and 1987. Though there were no control group studies, and certainly no double-blind studies to point to, the evidence was clear enough to compel me to fight for ABA.

Should I have abandoned my efforts for more than two decades while waiting for a control group study to appear? I'm glad I didn't." You, like most people, supported it because year after year, dozens of studies with subjects as their own controls kept coming out showing impressive gains in one domain or another; why wait until it was all "packaged" and started early (the 1987 study) to realize that this was a useful approach?

Rimland: "It is a major mistake to think of ABA as being competitive with, rather than complementary to, many other interventions, particularly such biologically based interventions as vitamin B6 and magnesium therapy." I don't think most ABA practitioners are opposed to biologically-based therapies.

The problem is, the results remain highly variable and unpredictable and are based on soft data, when looked at scientifically, which is what the groups mentioned did. I don't think they were asked "Which combinations of treatments are best?" That would have been impossible to answer, since so very little work in that regard has been done.

Rimland: "For years, our publications have urged parents who are about to undertake megavitamin therapy to refrain from mentioning the new intervention to teachers, therapists, grandparents, sitters, etc., so they would be able to obtain objective input from these 'blind' observers. Each such case is a mini double-blind single-subject experiment." No, it's not double-blind; it's only single-blind, because the parents giving the B6 know they are giving it.

Rimland: "We have heard from hundreds of parents who have reported, "Our therapist, who did not know about the B6 we started last week, said our child has made more progress in one week than in the prior three months" (see letters sections in back issues of the ARI)." Bernie, you're a good enough psychologist to know that's anecdotal, not scientific, evidence. Is it real? Sure. But how often? If it's so powerful, why does it have so much trouble showing clinical effects in actual studies? See, e.g., Pfeiffer et al.'s review in JADD, 1995, 25, 481-493.

Rimland: "If you really want to be scientific, do a mini double-blind trial on your child." Fine. But that just shows that we're still at an exploratory stage with this. The groups you complain about had to produce valid, general recommendations, not suggestions of interesting things to try that may or may not be helpful.

  • Unable to find what you're looking for? Search the entire site to find information about any subject we have information on. Instructions:
    Type a word or words into the form below and press the Search button. You may use "quotation marks" to search for a phrase. Adding a plus sign (+) before a word or phrase will require its presence; adding a minus sign (-) before a word or phrase will require its absence.

HomeParents' CornerParents' Corner ArchiveProfessional CornerProfessional Corner ArchiveCase in ProgressCase in Progress ArchiveInspirational MessagesInspirational Messages ArchiveDirect ServicesConsultingSeminars, Workshops, and MoreSpecial EventsRecommended ReadingRecommended Reading for ChildrenAsk the Experts News FlashCurrent Question and AnswerUnderstanding the LingoAbout the TeamTestimonialsFees, Location, and DetailsTypical Development: MakennaTypical Development: LaceyResourcesPrivacyStatementConfidentiality

CONTENTS (except as noted) ©2003-8 by Pediatric Services

Corporate Office in Morro Bay, California (San Luis Obispo County)
Telephone: 805.550.8799 Fax: 805.772.8246

E-mail:
Click here to ask a question.

DESIGN ©2003 by William Blinn Communications

Worthington, Ohio 43085

Articles written by Pediatric Services staff are copyright by Pediatric Services.
All other articles are copyright by their respective owners.
Information provided is for educational use only
and is not intended to replace medical advice from your physician.

Last modified: January 26, 2013