Content uploaded by Sabrina Freeman
Author content
All content in this area was uploaded by Sabrina Freeman on Sep 21, 2015
Content may be subject to copyright.
Introduction
i
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
ii
Copyright © 2011 by SKF Books, Inc. All rights reserved.
No part of this book may be reproduced in any form or by
Any means without the prior written permission of the publisher.
Any errors and omissions are the responsibility of the author.
Library of Congress Control Number: 2010919311
Freeman, Sabrina Karen, 1958 –
The complete guide to autism treatments. A parent’s handbook:
Make sure your child gets what works! / Sabrina Freeman
Includes bibliographical references and index.
ISBN 978-0-9657565-7-0
Autism in children—Treatment. 2. Autism—Treatment. I. Title1.
RJ506.A9F725 2011 618.92’85882’06 C2011-900064-9
Published by:
SKF Books, Inc.
1050 Larrabee Avenue #104-357
Bellingham, WA 98225
Printed in the U.S.A.
Second Edition
Disclaimer: The cover photo is being used for illustrative purposes only and any person depicted in that
photo is a model.
Introduction
iii
Contents
Foreword ..................................................................................... ix
Introduction .............................................................................. xix
Section One: What Works and What Doesn’t? ............1
Cost of Autism Treatment .....................................................................3
Half-baked Research ............................................................................4
Behavioral Therapies ............................................................................5
What is Behaviorism as it Applies to Autism ...........................7
What is Applied Behavior Analysis? .............................7
What is Intensive Behavioral Treatment? .....................9
Is the IBT Program Home-based or Center-based? ......9
•Home-based Intensive Behavioral Treatment .......................11
•School-based Intensive Behavioral Treatment .....................25
•OffshootsofIntensive Behavioral Therapies........................39
-Pivotal Response Training and the Natural Learning
Paradigm ....................................................................39
-Positive Behavior Support .........................................53
-Verbal Behavior Therapy ...........................................61
-Fluency Training ........................................................67
OtherSchool-basedTherapies(non-behavioral) ................................81
•TEACCH(TreatmentandEducationofAutisticand
RelatedCommunicationHandicappedChildren) ................83
•ThePlayschool(ColoradoHealthSciencesCenter) .............95
•Giant Steps ..........................................................................105
•The Higashi School/Daily Life Therapy .............................113
•Walden Preschool ...............................................................121
1.1
1.2
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
iv
Child-lead/Parent-facilitated Therapies ............................................ 131
•Floor-Time(Greenspan’sDevelopmental,Individual
Difference,RelationshipModel-DIR)..............................133
•OptionsInstitute/Son-RiseProgram.......................................143
•RelationshipDevelopmentIntervention(RDI)..................151
•The Learning to Speak Program ............................................ 159
Biomedical Therapies ........................................................................ 167
•Diet/NutritionTherapy(GlutenandCasein-freeDiet)........169
•Chelation Therapy ................................................................. 183
•Intravenous Immunoglobulin .............................................. 191
•Secretin ............................................................................... 197
•Vitamin B6 and Magnesium ................................................ 205
Speech and Language Therapies ....................................................... 217
•Fast ForWord Program ....................................................... 219
•The Hanen Method ............................................................. 227
•Lindamood-Bell Learning Processes .................................. 235
•TheSCERTS Model ........................................................... 241
Miscellaneous Therapies .................................................................. 249
•Art Therapy ......................................................................... 251
•Auditory Integration Training ............................................. 255
•Cranio-sacral Therapy .......................................................... 267
•Dolphin-assisted Therapy ....................................................... 273
•Exercise Therapy ................................................................ 281
•Facilitated Communication Training .................................. 289
•Holding Therapy ................................................................. 309
•Music Therapy .................................................................... 317
•Pet-facilitated Therapy .......................................................... 327
•Sensory Integration Therapy .............................................. 335
•Vision Therapy ................................................................... 349
1.3
1.4
1.5
1.6
Introduction
v
Section Two: How Do We Know What Works and
What Doesn’t ............................................................................. 359
Why care about science? ................................................................... 362
Why we can’t always rely on experts? ............................................. 363
•Experts”donotalwaysknowaboutscience ............................ 363
•Theymaynotvaluescience ..................................................... 364
•Advancementtrumpsqualityconcerns.....................................364
•Someexperts’motivesarenotalwayspure.............................365
•Thescienticmethodversuspseudo-science ........................... 366
Using the scienticmethodtoprotectyourchild.................370
What if the treatment method is too new for
data collection? ..................................................................... 370
•Whataboutanecdotes? Can we use them at all? ..................... 371
•Weneedtocareabouttheory ................................................... 373
What is a theory? .................................................................. 374
Good research is motivated or driven by theory ................... 375
How do we generalize results in autism research? ............................. 378
Using science to move closer to the truth ......................................... 379
•Testingtheory ........................................................................... 379
•Experiments are not optional .................................................... 382
•Peerreview-necessarybutnotsufcient................................382
•Uncoverthefunding source for the study ................................ 383
•Findingthepeer-reviewed journal articles ............................... 386
• Isthejournal peer-reviewed?.................................................387
Analyzing a study ............................................................................. 389
•Howmanygroups are there in the study? ................................ 389
Is it a Between-Subjects Design? .......................................... 390
Is it a Within-Subjects Design? ............................................ 395
Is it a Between-Within Subjects Design? ............................. 398
Is it a Factorial Design? ........................................................ 401
Is it a Single-Subject Case Design? ...................................... 403
What’s the problem with Single-Subject Case Designs? ...... 406
2.1
2.2
2.3
2.4
2.5
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
vi
•Howmanychildrenshouldtherebeineachexperimental
group? ...................................................................................... 408
•Whatistheidealnumber?........................................................408
•Whatisacceptable?..................................................................409
•Howarechildreninastudyassignedtothe experimental
groups? .................................................................................... 410
•Whatisbeingmeasuredandhow?...........................................413
Autism(theDependentVariable)..............................................413
Treatment(theIndependentVariable)....................................415
•Whocollectsthedata? .............................................................. 418
•Whoadministersthetreatment? ............................................... 420
•Didresearchresultshappenbychance?....................................421
•Isthestudybiased and how can bias be avoided? ................... 422
Bias type 1: History ............................................................. 422
Bias type 2: Maturation ....................................................... 423
Bias type 3: Treatment contamination ................................. 423
Confounding variables in general ..........................................425
How researchers mistakenly ruin their own
well-designed studies ............................................................ 426
How can bias be avoided? ................................................................ 428
When is it time to apply the results to children? ................................ 429
•Istheresearch far enough along? ............................................. 430
•Testingonhumansubjects.......................................................431
2.6
2.7
2.8
Introduction
vii
Redagsforquackery......................................................................432
Conclusion ........................................................................................ 435
Afterword........................................................................................ 437
2.9
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
viii
Foreword
ix
Dr. Richard Foxx, Ph.D. BCBA-D1
Few conditions have been as fraught with fad, controversial, unsupported,
disproven, and unvalidated treatments as autism (Foxx, 2008). The
underlying reasoning for this relates directly to science being misunderstood
or ignored by parents, professionals, and paraprofessionals. Many people
simply do not understand how science works, and this includes any parents
andparaprofessionalswithnoscienticbackground.Manyprofessionals
also do not understand science, typically because their education featured
noscientictrainingoritwasgreatlydeemphasized.Unfortunately,this
is the case for many of the professionals who work directly with children
with autism. There are other professionals who understand science but
choose or have chosen to ignore it for professionals or monetary gain.
Sabrina Freeman (2007) recognized all of these factors and decided
to do something to help. The result is The Complete Guide to Autism
Treatments: A Parent’s Handbook: Make Sure Your Child Gets What
Works! Although the book is written for parents, professionals will greatly
benet either because they will now understand how to evaluate the
science behind treatments or have a reference to give to parents. The book
also would serve nicely in a graduate course on autism, ethics, or behavior
analysis. The piece de resistance is that Freeman is both the mother of a
child with autism and an accomplished social scientist with a PhD from
Stanford. Two of her other books, Teach Me Language(Freeman &
1 Reprinted by permission, The Behavior Analyst, Spring 2010
Foreword
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
x
Dake,1997)andScience for Sale in the Autism Wars(Freeman,2003),
are directly related to autism and attest to the depth of her scholarship.
Freeman’s gift is that she writes about what some view as complex
subjects in simple understandable language. Indeed, she points out that
thescienticmethodisnotdifculttounderstandandthatknowledgeof
it permits rational decision making when it comes to evaluating the next
treatment or purported cure. Freeman’s objective is to protect thousands
ofchildrenfromquackerywhileprovidingparentsandprofessionalswith
evaluative tools for judging the effectiveness of a treatment.
Thebookisorganizedintotwosections:“WhatWorksandWhatDoesn’t”
and“HowDoWeKnowWhatWorksandWhatDoesn’t.”Section1is
designed to produce informed consumers who will seek a treatment for
theirchildrenbecausetheyknowithasscienticvalidity.Anyonewitha
goodbackgroundinsciencewillndthissectionwelldoneandextremely
helpful. Those with little background in science are encouraged to read
Section2rstinorder to have the backgroundtofullyappreciatethe
in-depth evaluations available in Section 1.
In Section 1 every major treatment option is exposed to the following
questions:Whatisit?Whatevidencedopractitionershavethatthisreally
works? What does the therapy actually look like? Would I try it on my
child? What else do I think? What additional studies would I like to see
theresearchersdointhiseld?Whoelserecommendsfororagainstthe
treatment? So you are still on the horns of a dilemma? What’s the bottom
line? The answers are typically spot on, in this reviewer’s opinion, and
are consistent with Freeman’s reputation as a tireless advocate for the
rights of children with autism to receive science-based treatment. The
literature review is exhaustive.
Under behavioral therapies, Freeman examines applied behavior analysis,
intensivebehavioraltreatment(IBT)thatishomeorcenterbased,school-
based IBT and what she refers to as offshoots of IBT, including pivotal
Foreword
xi
responsetraining andthe natural learningparadigm (PRT/NLP),positive
behavior support (PBS), verbal behavior therapy, and uency training.
Freeman reports that she implemented an intensive home-based behavioral
treatmentprogrambasedon the pioneering workofIvarLovaas(Lovaas,
1987)andthatherdaughter,whoisnowan adult, made incrediblegains.
That said, she cautions readers that her anecdotal reporting of this outcome
should not sway them to use the method, even if it comes from someone
who respects science. Rather, she states that what should be convincing is
the “abundance of scienticevidencebehindthemethod”(p.18)anditwas
just such evidence that led her to choose to use IBT with her child.
High-qualityschool-basedIBTprogramsaresupportedbyscienticresearch.
Althoughallchildrenbenetedfromtheseprograms,themostsignicant
gains were made by children who began treatment before the age of 5 years.
Had she lived in New Jersey when her daughter was young, Freeman would
have seriously considered sending her to the Princeton Child Development
Institute. The offshoots of IBT receive a fair evaluation based on the
literature to date, and the kinds of studies suggested for researchers working
inthevariousareasareexcellent.Freeman’sreviewndsPRT/NLPtobe
promising, but it does not have enough research evidence to suggest that it
is globally effective in ameliorating the condition of autism. Based on the
scienticresearchtodate,verbalbehaviortherapyisdescribedasanemerging
treatment but not one that should be applied solely to ameliorate the symptoms
orconditionsofautism.Asimilarbottomlineisgivenforuencytraining,
in that there is limited evidence that points to its appropriate use for certain
deciencycharacteristicsofautism.
Freeman’s bottom line on PBS is that “there is no evidence to conclude that
PBSisanythingmorethanaphilosophyratherthanascience.Consequently,
there is no evidence to demonstrate that PBS ameliorates the condition of
autism” (p. 59). She urges PBS researchers to abandon the antiscience,
anti-intellectual discipline they have developed and return to the eld of
applied behavior analysis where they can compete with behavior-analytic
researchers and “have their PBS research properly scrutinized and evaluated
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
xii
bytheirABAacademicpeers”(p.58).HernalpointonPBSisthatits
literature makes autism appear to be an entirely different disability. In
the PBS autism world, children “seem to be very mild, and the behavior
problems are all easy to control, as long as the environment is ‘re-
engineered.’ Children with self-injurious behavior do not seem to be a
challengeforthisgroup”(p.58).Hercaveatisthatperhapsthechildren
inPBSstudiesare“notclassicallyautistic”(p.58).Afterreviewingthe
literature and based on living in a region where PBS is used extensively
by school districts and with governmental support, Freeman expresses
heropinionthat“positivebehavioralsupportisaverydangerouseld
forchildrenwithautism”(p.56).HerreasoningisthatPBSisakindof
religion of political correctness that “denies children with autism access
toproven, science-based treatmentmethods” (p. 56). She evennds
theterm“positive”attachedtobehaviorsupportasoffensivebecauseit
“carries a presumption that the PBS practitioner is different and apart
fromhis‘evil’ABAbehavioristcounterpart”(p.57).
In the “Other School-Based Therapies” section, Freeman reviewed
TEACCH, the Colorado Health Sciences Center playschool, Giant
Steps(Canada),Higashi/dailylife therapy,andtheWaldenpreschool.
Her bottom line is that there is not enough research evidence to date to
conclude that TEACCH is an effective treatment or that the Playschool
autism intervention substantively improves the condition of autism. She
ndsinsufcientevidencethattheHigashischoolorWaldenpreschool
have an effective curriculum for decreasing the symptoms associated
with autism or treating and educating children who have it. She found
no evidence in support of Giant Steps.
The“Child-Led/Parent-FacilitatedTherapies”sectionincludestheoor-
time (Greenspan/developmental, individual difference, relationship)
model (DIR), Options Institute/Son-Rise program, relationship
developmentinterventions(RDI)andtheLearningtoSpeakprogram.
Twotherapies,DIRandRDI,hadnotgeneratedenoughscienticevidence
to conclude that they were effective treatments for children with autism.
Foreword
xiii
There was no evidence in support of Son-Rise and the Learning to Speak
programs as effective treatments. Freeman had personal experience with
DIR a number of years ago, because she chose it for her child when she was
rstdiagnosed.TheDIRphilosophy,whichturnseverythingthechilddoes
into a social interaction, was personally very appealing, and this treatment
was being offered by the psychiatrist who had diagnosed her daughter. Soon
after, Freeman abandoned DIR because, despite its personal appeal, there
were no data to support it. Her bottom line was “my child was wasting her
timeandIwaswastingmymoney”(p.139).
The biomedical therapies are familiar to many parents of children with autism.
They include the diet and nutrition therapies of gluten- and casein-free diets,
thecandidadiet,thenutritionaldeciencydiet,theketogenicdiet,chelation
therapy, intravenous immunoglobulin therapy, secretin therapy, and Vitamin
B6 and magnesium therapy. Although these theories have been around for 30
years,nonehaveanyindependentscienticsupport.Untilthereis,Freeman
regards their use as pure experimentation on a child.
Chelation therapy consists of removing harmful metal toxins from the
body by introducing chelating agents into the body. These bind with the
metal ions and then are expelled. Chelation is a recognized treatment
for children with lead poisoning but not for children with autism, who
do not have chronic heavy metal toxicity. Although many fad treatments
are costly in terms of money and time lost, chelation can lead to horrible
medical complications, including death. Freeman’s bottom line is chilling.
Chelating a child with no signs of heavy metal poisoning is engaging in
high-risk experimentation. There is no evidence to support chelation as
an effective therapy for children with autism.
There is not enough evidence to support any type of diet and nutrition therapy
as an effective treatment for improving the symptoms that characterize
autism. Unfortunately, parents are drawn to these approaches because they are
somethingtheparentcancontrolandfollow,andtheytwiththeparent’srole
of nurturer and provider of sustenance. Joining diet and nutrition interventions
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
xiv
as pure experimentation are intravenous immunoglobulin therapy, secretin
therapy, and Vitamin B6 and magnesium therapy. Although Freeman
is never shy about expressing her opinion based on her review of the
literature, she always provides a list of public and private agencies that
recommendagainstatreatmentorstatethatitfailedthescienticversion
of the sniff test.
The speech and language therapies include the Fast ForWord program,
the Hanen method, Lindamood-Bell learning processes, and the SCERTs
model. None of the four have any evidence to support their use as
effective treatments to improve the language impairment associated with
autism or ameliorate its symptoms. Freeman would especially like to see
the developers of the SCERTS model test their protocol against its main
competitor, intensive behavioral treatment. She makes this suggestion
for a number of therapies, especially those that are critical of intensive
behavioral treatment.
The miscellaneous therapies section is a veritable rogues’ gallery. All
of the classic fads are present, including auditory integration training,
craniosacral therapy, dolphin-assisted therapy, facilitated communication
training, holding therapy, sensory integration therapy, and vision therapy.
Othersonthelistincludearttherapy,musictherapy,andpet-facilitated
therapy. Because art and music therapies are regarded as relatively
harmless and not prohibitively expensive, most professionals tend to
give them a pass when harmful interventions are discussed. Art therapy
has no evidence of support, and there is not enough evidence for music
therapy to be considered an effective treatment for the symptoms of
autism. Freeman recommends removing the term therapy from music,
and I would add art. Some children with autism enjoy music and art,
and they can be used as reinforcers and for training in leisure activities.
In this limited role, both can have a place in a child’s program.
Although a dolphin ride may be reinforcing for a child with autism at
theDolphin Center ($2,000per week for1to 3weeksof treatment),
Foreword
xv
thereisnoscienticevidencethatitisaneffectivetreatment.Thistypeof
therapytsinthecategoryofthosethatareessentiallyignoredbyscientists,
because it is seen as not harmful but simply expensive. And, it is not the
type of day-to-day therapy that would replace an effective intervention like
applied behavior analysis. It is best used by parents who have money and
like salt-water vacations.
Freeman’s advice regarding pet-facilitated therapy is that “there is no
downsidetoowninganobedient,lovingdog”and“theexperiencemaybe
greatforyourchild; however,donotexpecttherapeuticresults”(p. 332).
Given that there is insufcient evidence to conclude that this therapy is
effective, a dog from the pound will serve as nicely as a costly, specially
trained therapy dog.
Vision therapy and craniosacral therapy are not commonly used for individuals
with autism, although they are recommended for other conditions, ailments,
diseases,anddisabilities.Thereisnoevidenceorinsufcientevidenceto
recommendeitherforautismoranyotherproblem.Quackwatch(aleading
Web site designed to expose harmful therapies) has craniosacral therapy
on its list of nonrecommended treatments. The use of vision therapy for
children with learning disabilities, in the form of eye exercises or specially
tinted glasses, is not supported by any pediatric or pediatric ophthalmology
professional academy.
Much as been written criticizing and condemning auditory integration
therapy, facilitated communication training, holding therapy, and sensory
integrationtherapy(Jacobson,Foxx,&Mulick,2005)andFreemancontinues
the practice. These therapies are particular insidious because they take
valuable time away from effective therapy and are highly seductive for
parents who desperately want to help their children with autism. Auditory
integration therapy is not only ineffective; it can be very costly. I know of
several situations in which parents who could ill afford to do so have spent
thousands of dollars on this treatment. Holding therapy is psychoanalytically
based and has followed the general course of psychoanalysis in the U.S.,
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
xvi
which is to say that it has fallen out of favor over time and is most likely
to be practiced or implemented in New York City. It has gone the way
of Bruno Bettelheim, refrigerator mothers, and the notion of a child with
autismhaving“attachmentissues.”
Sensory integration therapy has been a boon to occupational therapy,
because the underlying premise is that autism is a form of sensory
dysfunction.Althoughthereislittleornoevidenceregardingitsefcacy,
countless children with autism receive it in school as a legally mandated
part of their individualized educational program. Parents like it because
a sensory intervention that is designed to address a child’s neurological
needs has just the right amount of mind-body feel. Children seem to like
it because having a perky occupational therapist brush your arms, push
youonaswingormerry-go-round,orsqueezeyourarmsisaprettygood
waytospenda“treatmentsession.”
In the hall of shame of fads and autism, nothing ranks higher than
facilitated communication. This therapeutic intervention is proof positive
that H.L. Menken had it right when he said that “No one ever went broke
underestimatingtheintelligenceoftheAmericanpublic.”Itwouldbebad
enough if facilitated communication were simply worthless and costly,
but it also carries risks for parents and guardians, given the number of
them who have been falsely accused via facilitated communications of
sexually molesting their children and charges. Anyone whose child is
receiving facilitated communication is just a disturbed facilitator away
from being charged with a sex crime.
Onapositivenote,thereissomeevidencethatexercisetherapymayhave
some limited, short-term effect on the stereotypic behavior of individuals
with autism. And, no one can argue that engaging in exercise is not
good for all of us, especially those prone to obesity. Freeman endorses
incorporating exercise into a child’s daily life, but cautions that any
programmatic efforts should involve a behavior analyst, objectives, and
some data monitoring so that valuable therapeutic time is not wasted.
Foreword
xvii
Section 2 provides the basic rules and tools that enable one to evaluate the
autismtreatmentsdescribedinSection1.Freemangoesbeyondthescientic
method to look at how science is funded, how bias can slip in, the politics
of research, and what constitutes pseudoscience. Her goal in this section is
to inoculate parents from incompetent researchers or illegitimate purveyors
ofautismtreatmentinordertoprotectthechildrenfromthequackerythat
is pervasive in autism.
Freeman meets this goal admirably. She begins the section by asking “Why
careaboutscience?”followedbyadiscussionof“expertsandresearchers”
that the readers of this journal will recognize as a Brandisian lifting the rock
and letting the sunshine in. Although autism researchers and experts know
what Freeman is telling parents here, most have been reluctant to share it
with those outside our inner circle. For example, Freeman cautions that
“there are some very intelligent, talented researchers who produced biased
researchwhichtheyoftenhavepublishedinpeer-reviewedjournals”(p.384).
Her discussion of the world of academic publishing is revealing on multiple
levels. Consider these topical headings: “Advancement Trumps Quality
Concerns,”“Peer Review–NecessaryButNotSufcient,”and“Uncover
theFundingSourcefortheStudy.”
A mini course in experimental design is included that discusses the advantages
and disadvantages of between- and within-subject designs, factorial designs,
and single-subject case designs. Freeman’s discussion of how studies become
biased and how to avoid it is excellent, as is her treatment of the different
types of bias. Being a social scientist, Freeman knows her way around the
waterfront when she illustrates how researchers mistakenly ruin their own
well-designed autism treatment studies. She concludes the book with red
agsforquackery.
This book is a must read for any parent who has a child with autism, because
within its pages lies the unvarnished truth regarding what works and doesn’t
work and how to make the distinction. If you work with parents, encourage
them to buy a copy. If you work at a school or agency, make sure this book
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
xviii
is in the library. If you teach behavior analysis, use the book to expose
your students to an author whose writing is elegant, straightforward, and
brutally honest.
For a true understanding of the source of the passion that drove Freeman
to write this book, read Science for Sale in theAutism Wars (2003)
that describes a landmark legal battle between families with children
with autism and government and academic mercenaries. Although I
would never wish for anyone to go through what Sabrina Freeman has
experiencedinherlengthyghttohavesciencebeappliedtoherdaughter,
she took her disappointment and rage and turned them into two very
thought-provoking books for which parents with children with autism
and those of us who treat them should be very grateful.
Dr. Richard Foxx, Ph.D. BCBA-D
Professor of Psychology
Adjunct Professor of Pediatrics
Penn State University
References
Foxx, R.M. (2008). Applied behavior analysis (ABA) treatment of autism: The
state of the art. Child and Adolescent Psychiatric Clinics of North America, 17, 821-
834.
Freeman, S.K. (2003). Science for sale in the autism wars. Lynden, WA: SKF
Books.
Freeman, S.K. (2007). The complete guide to autism treatments. A parent’s hand-
book: Make sure your child gets what works. Lynden, WA: SKF Books.
Freeman, S.K., & Dake, L. (1997). Teach me language. Langley, BC: SKF Books.
Jacobson, J., Foxx, R.M., & Mulick, J. (Eds.). (2005). Controversial therapies for
developmental disabilities: Fads, fashion and science in professional practice. Lon-
don: Routledge.
Lovaas, O.I. (1987). Behavioral treatment and normal educational and intellec-
tual functioning in young autistic children. Journal of Consulting and Clinical Psy-
chology, 55, 2-9.
Introduction
xix
The Complete Guide to Autism Treatments was inspired by parents of
childrenafictedwith autism. Ihavespokento thousands ofparents
about various treatments and answered the same questions over and
over again. Many times I gave tutorials to individual parents. I realized
that parents need a clear way to understand how science works so they
can make appropriate treatment decisions for their children.
In addition, professionals and paraprofessionals need to have a better
understanding of the scientic method so they do not inadvertently
recommend a treatment with no science behind it to the parents of the
children they work with. It is crucial that professionals remember that they
hold considerable status and legitimacy in the eyes of parents, and with that
legitimacy comes responsibility — a responsibility to not inadvertently
send parents down the road of quackeryinautismtreatments.
Currently,manyparentsnditdifculttoevaluateautism treatments for
their child. They are forced to rely upon experts who may or may not know
enough about the science to provide accurate information. Therefore,
in a sense, parents need to become experts themselves. Fortunately, the
scienticmethodis not difcult tounderstand.It simply needs tobe
laid out in a form that is understandable. All parents, professionals and
paraprofessionals alike need to know how to make informed choices
about which therapies to use to treat the child’s autism. After reading this
book, my sincere hope is that everyone will be able to evaluate the next,
new purported treatment or cure that comes along. It is very important
Introduction
The Complete Guide to Autism Treatments: Make Sure Your Child Gets What Works!
xx
tobeabletoasktherightquestionsandtondtheawsinthescience
behind the purported treatment, or to nd the evidence that, in fact,
the treatment is effective. At a minimum, understanding the scientic
method will protect thousands of children from quackeryand,hopefully,
provideparentsandprofessionalswiththetoolstondtreatments that
are effective for autism.
I must apologize in advance to many deceased philosophers of science
insofar as I am going to make short shrift of most of their concepts;
however, parents of children with autism and the professionals they rely
upon only need to know enough about scientictheoriesandtheoretically
motivated researchtoprotecttheirchildrenfromquackeryandthevendors
of“snakeoil”treatmentsforautism.Parentsofchildrenwithautism
are better off when they understand statistics and how they are used to
reportstudyndings.Onlythenwillconsumersbeabletoevaluateclaims
about autism treatmentsthataresupportedthroughtheuse(ormisuse)
of statistics. In short, this book is designed to give those who care about
the futures of children with autism the information they need to make
suretheirchild“GetsWhatWorks!”
Thebookisorganizedintotwosections.InSectionOne,wescrutinizethe
range of treatment options offered to parents of children with autism and
use the tools of the scienticmethodtoevaluateeachtreatment to help
create informed consumers of autism treatment services. Section Two is
designed to provide a background in science for parents or professionals
who are newcomers to the scienticmethod.Thissectionisamust-read
for consumers who plan to independently scrutinize the next autism
treatment introduced into the marketplace. For those who may not have
a background in science,IsuggestthatSectionTwobereadrst as a
primer, prior to reading about specic autism treatments. Otherwise,
therstsectionisbestreadbytopic,asareference,orsequentiallyasa
comprehensive guide to autism treatments.
Section One: What Works and What Doesn’t?
1
Section One
What Works and
What Doesn’t?
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
2
Section One: What Works and What Doesn’t?
3
This section groups similar autism treatment approaches together. Although
the typology may not be perfect because some treatments do not easily lend
themselvestoaspecicorientation, this categorization is probably the easiest
way for readers to wade into the deep, murky world of available autism treatments
(Iusetheword“treatment”verylooselyforsomeofthesemethods).
When reading a section on a particular method, I have introduced the method
with no editorializing. In other words, I present the treatment method in the
clearest way available based upon what the treatment professional has said about
his or her method. No matter how wild or wacky an idea may sound, we must
look at the data rather than rely on our intuition to determine if the treatment
method is absurd or sensible.
After introducing the method, I then look at the evidence that supports the
claims made by those who teach or practice the method. In this subsection, I
highlight concerns about the studies and then give readers a chance to evaluate
mycomments.Finally,Iprovidea“BottomLine”regarding each treatment.
Scienceisdenedbydebate;therefore,Iwelcomereaderstodisagree with me
regarding my evaluationsbasedonthescienticevidence. The goal of this book
is to have consumers critically evaluate autism treatments so they are 100 percent
informed about a treatment before they attempt it on a vulnerable child.
Cost of Autism treatment
BecauseIamalsoaparent(andanethicalhumanbeing),Irefusetoevaluatea
treatment based on economics. There is a very serious political debate raging
among policy makers about the number of resources children with autism should
receive relative to other children with special needs. Aside from the draconian
and heartless nature of these debates, the arguments are also awed because
this group of children needs to be treated and educated rather than warehoused.
Unfortunately, highly bureaucratized systems concerned with short-term
budgeting(civilservantsforgettingtheirprimaryfunction),regularlyattemptto
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
4
provide children with treatments that are economical rather than effective, as
a means to ration resources.
In addition, there is a trend to ration treatment based upon an autistic child’s
functioning level. It is particularly disturbing to see a child with severe autism
not given the interventionrequiredduetothedegreeofseverity.Asaresultof
treatment rationing, much litigation takes place revolving around treatment for
children with autism. This book does not enter the turbulent treatment rationing
debate, although it is self-evident that as advocates for their children, parents
needtoghtforthemostappropriatetreatment available, regardless of cost to
the health care or educational systems.
Half-baked Research
Oneofthe primary shortcomingsofmostresearchintheautismeldisthat
researchers tend to apply their ndingsprematurelyonchildren.Itseemsas
though an autism treatment researcher or practitioner need only develop an
interesting idea and desperate parents are happy to volunteer their children to
receive the treatment. Unfortunately, much of this research is still very much in
the experimentalstage(andlacksevidence that it is effective).Well-meaning
parents and professionals who are uninformed how research must proceed
to determine a treatment’s effectiveness often recommend this experimental
treatment to parents of autistic children. This observation is particularly true
in the area of biomedical therapies for autism.
In this section, I have included every treatment offered to parents, irrespective
of whether it has been discredited, is still in the experimentalstage(half-baked),
or whether it is considered best practice. After evaluating the data, it is up to
consumers — the parents — to decide whether or not to experiment with their
child. However, it is very important to note that some of the unsubstantiated
treatments may actually be harmful for the child. Some caution is advisable.
AswillbecomeevidentthroughoutSectionOne,Istronglyrecommendagainst
pursuing these potentially harmful treatments.
Section One: What Works and What Doesn’t?
5
p Home-based Intensive Behavioral Treatment
p School-based Intensive Behavioral Treatment
p Offshoots of Intensive Behavioral Therapies
Behavioral Therapies Section 1.1
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
6
Section One: What Works and What Doesn’t?
7
Behavioral Therapies
What is Behaviorism as it Applies to Autism?
In the world of autism therapies, there is a considerable amount of research
conducted on various types of behaviorism. Behavioral methods or schools
of thought may be different in terms of their goalsfor the child(e.g., which
behaviors they would like to increase or decrease or which skills they would
likethechildtoacquire).However,keepinmindthattheactualmethod they
are relying upon has the same origin. It all ows from the work of B.F. Skinner,
the grandfather of behaviorism.
Behavioral intervention for individuals with autism involves behavior
modicationbased onB.F. Skinner’s principles of operant conditioning, used
to decrease undesirable behaviors and to teach and encourage new and desirable
behaviors. Behavioral practitioners and theorists analyze human functioning
based only on those behaviors that are overt and observable, as opposed to making
inferences about internal mental states.1 Behavioral theory proposes that the
use of reinforcement and punishment techniquestoeliminatenon-functionalor
destructive behavior,whilebuildingupthefrequencyandvarietyofalternative
behaviors, will provide a basis for aiding development.
What is Applied Behavior Analysis?
When it comes to autism,overthelastforty-veyearsbehaviorists have taken
lessons from research done on animals, and have signicantlymodied those
techniquesforusewithmanypeople,includingthosewithautism. The techniques
thatthiseldhasestablished are not simply to teach people with a wide variety of
problemshowto“behave;”rather,throughbehavioral techniquesthatoriginate
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
8
intheeldofApplied BehaviorAnalysis(ABA),treatmentprofessionalsare
able to change self-destructive or maladaptive behaviorssothatpersonsaficted
with autism can attain a large repertoire of important life skills, including
communication, academic, social, self-help, and foundation skills which promote
independence.
TheeldofApplied Behavior Analysis is very broad: the treatment of autism
isonlyasmallbutgrowingpartofthiseld.The certifying body, the Behavior
AnalystCerticationBoarddenesABAasfollows:“Applied behavior analysis
is a well-developed discipline among the helping professions, with a mature
bodyofscienticknowledge,established standards for evidence-based practice,
distinct methods of service, recognized experience and educationalrequirements
forpractice,andidentiedsourcesofrequisiteeducationinuniversities.”2 In
terms of autism, governmentagenciesoccasionallyattempttodenetheeld
of ABAas ayoung, emergent eld thathas insufcientdata on efcacy* or,
conversely, that there is not enough data on the application of ABA principles
for children with autism six years of age and older. This is categorically untrue,
as is evident by hundreds of studies conducted from 1980 to the present done in
thiseld,mostofwhichwereconductedwithadults,notchildren.3
Itis importantto understandthat notall those certied inApplied Behavior
Analysis necessarily have the expertise to design and implement an intensive
behavioral treatment program for children with autism. Before parents set up
an intensive behavioral treatment program with a behavior analyst, they need
tomakesurethatthisprofessionalhastherequisiteexperiencewitharangeof
autistic children.
*For more information on the way the governmentshavewarpedanddistortedtheeldofABA
to avoid paying for treatment for children with autism, I encourage you to read, Science for Sale
in the Autism Wars.4
Section One: What Works and What Doesn’t?
9
What is Intensive Behavioral Treatment?
Intensive Behavioral Treatment(IBT)forchildren withautism is centered on
the idea that the use of behavioralprinciplesinahighlyintensivemanner(e.g.,
fortyhoursperweekoftreatment)iseffective in ameliorating the symptoms of
autism. Researchers have found that the global development of children with
autismcanbeinuencedthroughtheuseof1)operantconditioning,2)techniques
researchedandappliedfromtheeldofbehavioranalysis,and3)ndingsfrom
the literature on child development. In other words, since autism is a Pervasive
Developmental Disorder, IBT can be used to intervene positively in the outcome
of autistic disorder by forcing development that is not occurring naturally. It
was hypothesized, and later supported, by research that the child’s delay or
disorder in language, social development, cognition, and overall functioning
can be mitigated or eliminated with early IBT. Although some describe IBT
as devoid of developmentalinuencesfromthetheories of child development,
this is, in fact, not the case.
Is the IBT Program Home-based or Center-based?
According to practitioners of IBT, when done competently, treatment should
take place during every waking hour of the child’s life in order to maximize the
child’s developmental window. Whether a child participates in a home-based
behavioral treatment program or a center-based treatment program is generally
a decision made by the child’s parents. There are differences in philosophy
regarding these two options when it comes to integration versus segregation.
The data generated by home-based programs is more plentiful and generally
stronger than that of center-based programs. This may have less to do with
comparative effectiveness of the two program approaches,butratherreects the
prolicnatureofthoseresearchers who conduct studies on home-based treatment
programs. In the next few pages, I will introduce the traditional home-based and
school-based intensive behavioral treatment programs, and then discuss autism
treatment offshootsfromthebehaviorismeld.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
10
Section One: What Works and What Doesn’t?
11
Behavioral Therapies: Home-based Intensive
Behavioral Treatment
What is Home-based Intensive Behavioral Treatment?
The pioneer in treating children with autism in a home-basedmilieuisDr.O.
Ivar LovaasoftheUniversityofCaliforniaatLosAngeles(UCLA),withthe
work he initiated in the 1960s and 70s at the Young Autism Project. Many
worldwide sites were originally established to replicate the ground-breaking
autism treatment work of Lovaas rst published in 1987. Today, intensive
home-based treatment programs for children with autismarenowquitepopular.
Although there are many reputable practitioners who never trained at either the
Young Autism Project or associated replication sites, as a result of the treatment
protocol developed and tested by Lovaas and colleagues, home-based IBT
programs have come to be referred to by parents as “The Lovaas Method”of
AppliedBehaviorAnalysis,or“Lovaas-typeABA.”* Unfortunately, there is at
this time no systematic way to differentiate those practitioners who are doing a
competent job of programming for an Intensive Behavioral Treatment program
fromthosewhoare unqualied, muchto thefrustrationofbothparents and
*This branding is disturbing to many reputable academics1 because they are concerned about a focus or
overrelianceonspecictechniquesratherthantheuseofdata-driven changes based on the principles of
ABA. They are also concerned that branding precludes new science-based advances. Although these
are legitimate concerns, consumers (the parents making crucial treatment decisions for their children)
ndthatbrandingprovides someprotectionfromeveryTom,DickorHarrywhowants tohang outhis
shingle and claim he knows how to create, maintain, and supervise a science-based behavioral treatment
program.ManycriticsofbrandingclaimthatcerticationinABAshouldbesufcienttoprotectparents
fromincompetentorunqualiedproviders.IdisagreeasthereareBoardCertiedBehavior Analysts who
create programs exclusively relying on certain techniquesthatarewithoutsufcientevidence of efcacy.
It is my view that the ethical guidelinesoftheself-policingboardcertifyingbodyisnotsufcientprotection
for consumers when it comes to efcacy.Hopefully,onedaythere willbea Board CertiedBehavior
Analyst specialization in autism and branding will fall out of favor. However, until that time, I predict
that parents will continue to brand and use this shortcut to refer to the home-based Intensive Behavioral
Treatment program that originally created the results from the landmark 1987 Lovaas study, even though
many of the techniqueshavebeenrenedandimprovedsincethe1970sand80s.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
12
ethical academicsinthiseld.Intheeldofintensive behavioral treatment in
general, but home-based treatment in particular, it is still very much a case of
caveat emptor.
Early behavioral treatment for children with autism applies behavior modication
principles to teach children with autism in their homes and communities under
the watchful eye of their parent or caregiver. This interventionidentiesskill
decits(areasofweakness)whichhaveresultedinthechild’slackofsuccess
in typical learning situations, and targets them for “manual” acquisition of
thenecessaryskills.Thedifcultythatchildrenwithautism typically have in
learning naturally from the environment is targeted by breaking down skills and
instructions into their smallest components.Thechildrstacquireseachstep
separately, then chains them together and eventually masters the entire skill. To
maketheskillacquisitionprocesseasier,severalmethods are used. Currently,
the common structures in competent IBT programs include: direct instruction
(thechildbeingdirectly taught the part of the skill by a therapist);1:1therapist
to student ratio(oneadulttoonechild);discretetrialtraining(atherapist-led,
highly structured teaching technique);discriminationtraining(anotherhighly
structured techniquethatteachesthroughdirectcomparison);prompting and
fading strategies(atechniquethathelpsthechildlearnbyprompting or giving
hints for the correct answer and then fading the prompts or hints once the learning
hastakenplace);shaping(atechniquewhichtakestheskilllevelofthechild
and, through well-planned reinforcement, teaches the child to improve his or
herskilllevel);andchaining(atechniquebywhichacomplexskillistaughtby
teaching a number of simple skills and connecting these simple skills to master
thecomplexskill);andusingavarietyofreinforcement strategies(atechnique
to reward the child for the correct response). Thebasiccurriculum includes
imitation skills, receptive language skills, toy play, and self-help skills.Once
these components have been mastered, the more advanced curriculum includes
Section One: What Works and What Doesn’t?
13
expressive language skills, abstract language and interactive play (withother
children). Further advancement has the child overcome decits in both the
home and school environments where the curriculum includes pre-academic
and academic abilities (such as w eather and calendar skills), socialization
skills, cause-effect learning and observational learning. The goal of this stage
istopreparethesechildrentolearn“naturally”fromtheschoolenvironment.
Intensive Behavioral Treatment programs follow a basic hierarchy of skills;
however,theyarehighlyindividualizedandexiblebasedontheskilllevelof
each child.
Importantly, in Intensive Behavioral Treatment programs, non-learning behaviors
(e.g.,self-destructiveormaladaptive)aretargetedforeliminationusingavariety
of behavioral techniques.Originally,thetreatment protocol employed extinction
(ignoringthebehavior),time-out(removingthechildfromthesituationfora
shortperiodoftime),physicalrestraint(holdingthechild’shandsifheorshe
werehurtinghimorherself),verbal reprimands(tellingthechild“no”or“stop”),
types of differential reinforcement(e.g.,rewarding the child for not engaging in
a particular behavior),andredirection(involvingthechildinanotheractivityto
interrupt a nonfunctional behavior).Manyofthesetechniquesareusedtoday;
however, time-outs and physical restraint have fallen out of favor with many
practitioners.
What evidence do practitioners have that this
really works?
Our wide literature search netted over 100 articles on comprehensive IBT
programs. Most of the publications were commentaries about the original
studies and the replications of Lovaas’ work. In terms of peer-reviewed articles
presenting data on IBT,therewerefourteen(14)articles. In each and every study
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
14
where the treatment delity was high,2,3,4,5,6,7,8,9,10 children in the experimental
groupsignicantlyimprovedoverchildreninthecontrol group. Even in some
of the parent-directed groups,9,11 the children in the experimental group fared
much better than the control group that did not receive the intensive treatment.
There are a few studies in which the parent-directed therapywasnotsufciently
rigorous, and therefore, the children did not make substantial gains.13,15,16 The
above articles will be now presented and discussed.
The original Lovaas study (1987), showed extremely promising results for
treatment efcacy.Theoutcomes indicated that 47% of the experimental group
(n=19) achieved normal functioning, 40% were assigned to classes for the
language delayed and 10% were assigned to classes for the autistic/retarded.
In contrast, only 2% of the controlgroup(n=40)achievednormalfunctioning.
Forty-vepercentwereplacedinclassesforthelanguage-delayed with the other
53% placed in classes for the autistic and mentally retarded.15 The experimental
group made average IQ gains of over 30 points. These treatment gains were
assessedveyearslaterandfoundtobemaintained,withtheexceptionoftwo
children.5One of thesechildrenmovedback into alanguage delayed class;
however, another child joined a mainstream class and, therefore, outcome
percentages remained stable.
Smithandcolleagues(1997),undertookareplication of these results through
archival data; however, they used participants who were in the lowest functioning
range. Their resultsconrmedthattreatment gains were achieved, even with the
most challenging population of autistic children. Average IQ gains made were
tenpointsonaverage(+/-2)fortheexperimental group, versus an average three
point decrease in the control group. At intake, no child in either the experimental
or the control group had any speech; however, at follow-up, ninety-one percent
(10/11)oftheexperimental group used spoken words functionally, versus twenty
Section One: What Works and What Doesn’t?
15
percent(2/10)inthecontrol group. It is important to note that the Smith et
al.(1997)study selected children with diagnoses of autism and severe mental
retardation, making these results that much more impressive. As mentioned
earlier, there are several other designs replicating the results of Intensive
Behavioral Treatment.3,4,7,8,9,10.11,12,13,14 Although most of the above studies were
home-based, even those studies where the children were in a pre-school,7,8 showed
asignicantcomponent of home-based treatment.
The majority of studies that attempt to replicate Lovaas’ original work generally
use control groups, creating a between-subjects design. Anderson(1987)isan
exception: they used a within-subjects design with fourteen children receiving
treatment. Between-subjectsdesigns(usingacontrolgroup)areoftenusedto
control for confounding variableswhichcouldinuencetheoutcome or results
of the studies(seethenextsectionofthebookforadiscussionontheroleofa
controlgroup). Theonemethodological problem which exists in the studies
was the absence of random assignment to the experimental and control groups.
Due to parental protest at the time Lovaas conducted his original study, he was
unable to use randomization to assign children to groups. The National Institutes
ofHealth(thefundingsource of the study),gavetheirblessingtoLovaas to use
a different techniquetoassignchildrentogroups.Todiffuseparentalconcerns,
Lovaas assigned children to experimental condition based on funding and distance
from the UCLA clinic. In addition, he matched children in the control and
experimental groups to guarantee that the two groups were similar at intake.*
*Baer(1993)referredtothistechniqueasfunctionallyrandomassignmentandarguedthatitcouldbeequally
as convincing as random assignment providing the researchers did not control the way the children were
assigned. Baer explains that because assignment to control or experimental group was based on resources,
a variable out of the experimenter’s control, there is no reason why this procedure could not have created
true randomization. He states: “the child’s status as a best-potential case or a worst-potential case, even if
perceptible to the clinician, could not have affected the availability of those resources at the moment that the
childwasavailableforassignment,andso,inmyjudgment,theassignmentwasfunctionallyrandom.”17
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
16
In addition to matching, Lovaas(1987)variedtreatment intensity between the
control and experimentalgroupstodeterminewhetherhigh-quality,lowintensity
treatment would have positive effects. It did not.
None of the replications of the original study randomly assign children to groups
because it is ethically impossible to do so due to the original data which shows
the effectiveness of the treatment. In order to overcome this problem, Sheinkopf
and Siegel(1998)usedmatched pairs assignment. The Smithetal.(1997)article
examined archival records and attempted to match the groups based on age, IQ,
diagnosis, language and behavior. Each study used a variety of widely-accepted
measures of the dependent variable, autism.
Results indicate signicant improvement for the experimental group in all
home-based behavioral intervention studies. As mentioned above, the most
dramatic results came from the study by Lovaas (1987), which reported an
average of thirty point IQ gains in the experimental group. This program also
had the highest intensity of treatment at forty hours per week for two or more
years. The McEachin(1993) study is a follow-up on the children from the
Lovaas(1987)study, which shows that these children maintained their gains
andsubsequentschoolplacements.Inaddition,theSallowsetal.(2005)study
not only replicated the original Lovaas (1987) study demonstrating that the
experimentalgroupsignicantlyimprovedoverthecontrol group, but showed
the level of improvement of the children in the experimental group rivals that of
the Lovaas best outcomechildren.Anadditionalstudy,Cohenetal.(2006)used
aquasi-experimental design with twenty-one-age and IQ-matched children in a
community-based setting over a three-year period. They found that the children
whoreceivedIBTbasedontheUCLAprotocolfaredsignicantlybetterthanthe
matched children attending special education classes.10Onestudythatrequires
particular mention is the Howardetal.(2004)study in which IBT was contrasted
Section One: What Works and What Doesn’t?
17
withhighquality,intensiveeclecticprogramming. The Howard et al. study
clearly demonstrates that eclectic treatments for autism are not as effective as
IBT based on the principles of Applied Behavior Analysis.
What does the therapy actually look like?
Since the data demonstrate2 that an average of thirty to forty hours per week of
intensive intervention is crucial for best outcome to be achieved, the ideal therapy
program will have the child engage in therapy forty hours per week. The rationale
for this level of intensity comes from typically developing children. Children
without autism engage in at least forty hours per week of active learning;
however, for them it is a naturalistic, incidental type of learning. Since autistic
children do not generally learn useful skills or information naturally from their
environment during their free time, this learning needs to be facilitated, and
is best done through structured learning for approximately the same amount
of time as that which occurs for their typically developing peers. The UCLA
protocol starts therapy in an intensive one to one intervention in which skill
acquisition occurs using highly structured forms of learning. The therapy
rsttakesplaceinthehome,typicallywithyoungcollegestudentstrainedas
therapists, and eventually progresses into the preschool setting. As the child’s
abilities increase, the structure of the teaching decreases and learning begins
to happen more naturalistically. The eventual goal for children who have gone
through the program is to achieve independent learning, from their environment,
in the same manner as occurs with typically developing children. In good IBT
programs, naturalistic learning is programmed for children only once they are
ready, and not before. The ideal scenario occurs when the autistic child is able
to enter kindergarten independently and learn naturalistically in the same manner
as that child’s peers.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
18
Would I try it on my child?
Yes, I would and I did. My child began an intensive, home-based behavioral
treatment program based on the work of Dr. Lovaas when she was four years
old. I chose this methodbeforeitbecamepopular(approximately1992)because
it was the only treatmentthathadanyhighqualitybetween subject-designed
studies to evidence the effectiveness of the treatment. My decision was based
on science, which indeed bore fruit as my daughter did make incredible gains.
It is important to remember, though, that my anecdotal reporting regarding my
child’s gains should not sway you to use this method. Anecdotal reports are
unreliable to use when making the important decision about treatment methods
to use with your child, even if the anecdote comes from someone who respects
science. What should convince you, when choosing one method over another,
is the abundance of scientic evidence behind the method.Itwasscientic
evidence that led me to choose IBT for my child.
What else do I think?
Although it is very frustrating to parents, the lack of any known cause of autism
makes the behavioral treatment approach ideal because its effectiveness does not
depend on an underlying theory of cause. Based on the evidence provided by this
group of studies, it can be concluded that home-based behavioral intervention,
using the best practices models that can be found in the UCLA protocol and its
close approximations, is an effective method for the treatment for autism. As
is illustrated by the Smith(1997)study, this intervention is effective, even for
the autistic and severely mentally retarded population which, in my opinion,
presents the greatest challenge before us as parents and professionals.
Section One: What Works and What Doesn’t?
19
WhatIndparticularlyappealingabouthome-based IBT is that the parent is the
case manager. In other words, the child is under the watchful eye of those who
love him or her. In addition, the concept that my toddler or pre-schooler is able
to enjoy all the experiences of typically developing children, with a therapist
helping to facilitate this interaction, rather than the child being segregated from
the earliestage(toaccessmoreexpertise),isveryappealing.Afterspeakingto
thousands of parents, I found that the philosophy of integration and normalization
is a philosophy that is more comfortable for a parent of a newly-diagnosed child
to accept. Although this philosophy may or may not be a contributing factor
for the effectiveness of a behavioral treatment protocol, it is fortunate when
the treatment protocol naturally accommodates inclusion and integration, and
avoids stigmatization.
Oneissuethatparentsshouldbeawareofinrunninghome-based IBT programs
isthatifthedelityofthetreatmentisnotsufcientlyhigh(i.e.theprogram
is not “tight” or implemented correctly), some of the data indicate that the
child’s gains will suffer. That said, the parent-directed group which received
three hours of supervision every other week in the Sallows and Graupner
(2005)studywasofsuchhighqualitythattheirchildrenfaredaswellasthe
clinic-directed children.18 This ndingwasunexpectedandimportant,though,
because it demonstrated that parent-directed treatment programs with minimal
supervision(sixhoursamonth)canproduceexcellentoutcomes.
Critics of IBT(andtherearemany),claimthatthistreatmentisamechanistic
program which essentially turns children into robots. These programs are also
criticized because some claim that the children do not generalize their skills from
the therapy sessions into the natural environment. As I previously mentioned,
a good behavioral intervention program must be individually designed and
customized for each child. The program grows and is modied with the
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
20
child’s developing skills and, as a result, the program becomes increasingly less
structured as the child becomes better able to learn in that format. Generalization
of skills is programmedintoanyqualitybehavioral intervention program to ensure
that skillstaught“atthetable”willalsobetaughttobeusefulforthechildin
theireverydayliving.ThisiswidelyrecognizedintheeldofIBTasakeygoal
for programmers. In short, good IBT programs do incorporate generalization
anddo notcreate“robots.”Critics alsochargebehavioral therapists as being
abusive to children through the use of verbal and physical aversives. Fortunately,
the use of physical aversives* and other techniques,suchasphysical restraints,
are not part of the home-based treatment protocol(andhavenotbeenforover
twentyyears).
In the original Lovaas study(whichbeganintheearly1970s),amild physical
aversive (a slap on the thigh) was used with a small subset of the children.
This physical aversive was dropped from the protocol approximately twenty-
veyearsago.Currently,theLovaas Institute For Early Intervention(LIFE)
uses the techniques of extinction, redirection, differential reinforcement and
teaching alternate forms of behavior. The use of physical aversives no longer
occurs through practitioners from the institute or at any of the treatment sites
afliatedwiththeUCLAYoungAutism Clinic, or by any reputable independent
practitioners using the UCLA protocol.
What additional studies would I like to see the
researchers do in this field?
At this point, there are a large number of IBT research replication sites, both in
theUnitedStatesandthroughouttheworld(www.Lovaas.com lists the worldwide
*Today, if physical restraintsaretobeused(whichmaybenecessaryifthechildisseverelyself-injurious),
they are generally used only as needed, in highly controlled institutional settings such as hospitals, where
there is video monitoring, precise data collection and, depending upon jurisdiction, judicial surveillance.
Section One: What Works and What Doesn’t?
21
replicationsites).Thereplication sites are designed to do exactly what their name
implies: replicate the original study published by Lovaas and associates in 1987.
These replication sites use the original protocol from the Lovaas(1987)study with
a few exceptions. Replicating the treatment protocol utilizing rigorous scientic
method is crucially important for our children. Unfortunately, the randomization
to either a control or experimental condition increasingly becomes problematic
because the more evidence that is gained regarding the effectiveness of this
method, the more unethical it becomes to have a control group of children who
do not receive an intensive amount of this type of treatment. Due to the relentless
rationing of health care* and education for children with autism, continued
replication of the Lovaas’ initial landmark study(Lovaas,1987)byindependent
investigators is particularly important concerning the politics of autism policies
rather than the science of autism treatment.
Who else recommends for or against home-based
behavioral treatment as a method for the treatment
of autism?
There is a large number of reputable organizations that have conducted
independent reviews endorsing IBT as best practices. The New York State
Department of Health’s clinical practiceguidelines(1999)regarding the use of
IBT as a treatment for autism,wasbasedlargelyonvestudies, all conducted
by Lovaas and colleagues, or from partial replications of the protocol developed
by Lovaasandcolleagues.Largelybasedonthesevestudies, the New York
Report concludes: “It is recommended that principles of applied behavior analysis
*The incorporation of Intensive Behavioral Treatment will be fought by those in the autism industry
offering competing treatments and by governments and their policy analysts who do not want to pay for
this treatment. They are attacking the science behind IBT purely because they are self interested. The more
evidence that is published about the efcacyofIBT, the less likely it is that they will be able to continue
to deny children with autism best practices treatment.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
22
(ABA)andbehavior intervention strategies be included as an important element
of any intervention program for young children with autism.”19 In addition, a
U.S. Surgeon General, Dr. David Satcher, had the following to say about Lovaas’
work: “Thirty years of research demonstrated the efcacyofappliedbehavioral
methods in reducing inappropriate behavior and in increasing communication,
learning, and appropriate social behavior. A well-designed study of a psychosocial
intervention was carried out by Lovaasandcolleagues(Lovaas, 1987; McEachin
etal.,1993).Uptothispoint,anumberofotherresearch groups have provided
at least a partial replication of the Lovaas model.”20 It is important to remember
that this report was published in 1999 prior to the publication of additional
studies replicating these results. Additional organizations endorsing IBT include
the AmericanAcademy of Pediatrics (2001),21 the National Research Council
(2001),22 and the American Academy of Child and Adolescent Psychiatry.23
So you’re still on the horns of a dilemma?
If you are still thinking about whether or not to set up a home-based behavioral
treatment program for your child, you might want to read Lovaas (2003),24
Mauriceetal.(1996),25 and Leafetal.(1999)26 to gain an in-depth understanding
of how home-based treatment programs are administered. In addition, I encourage
you to view the videotape, “Behavioral Treatment for Children with Autism”
available in most university libraries or to be purchased on-line at the Cambridge
Center for Behavioral Studies.27 This videotape chronicles Lovaas’ research
from the late 1960s to the late 1980s and provides an overview of the original
treatment protocol.
Section One: What Works and What Doesn’t?
23
What’s the bottom line?
Based on the scienticresearch to date, there is substantial evidence that home-
based Intensive Behavioral Treatment is effective for children with autism.
In addition, treatment gains appear to be long-term and for a broad range of
functioning levels.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
24
Section One: What Works and What Doesn’t?
25
Behavioral Therapies: School-based Intensive
Behavioral Treatment
School-based Intensive Behavioral Treatment shares many of the components of
home-based Intensive Behavioral Treatment programs except that these programs
are based in preschool settings, which are often segregated or integrated with a
high ratio of autistic children to typically developing children. The main issue
with school-based IBT programs is treatmentdelity.Inotherwords,howmuch
ofthedayisthechildactuallyreceivingqualityautismtreatment, and how much
of the day is the child only receiving care-giving. Although this characterization
may appear somewhat blunt, this is indeed a concern with many school-based
programs. Below we will highlight three programs that produced data and discuss
each program separately.
What does school-based IBT look like?
Princeton Child Development Institute
The Princeton Child DevelopmentInstitute(PCDI)is a nonprot society that
runs a preschool, a school and two teaching homes. In addition, they offer
supported employment and career development for adults.1 The PCDI is not
afliatedwith a university;however,research is conducted with some of the
children enrolled in the PCDI and ndings are published in peer-reviewed
journals. Programs are individualized for each child based on that child’s skills
and decits. Each child’scurriculum is implemented using the principles of
applied behavior analysis. A curriculum for a student at the PCDI would typically
include nonverbal and verbal imitation, receptive instructions, toy play, receptive
and expressive language skills, reading and academic programs and social
initiations. These programs are delivered using a variety of techniques,which
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
26
include discrete trial training, incidental teaching, use of time delay, visual
schedules and video modeling, as appropriate. Direct instruction is used at the
PCDI, using a teacher-to-student ratio that ranges from one-to-two to one-to-
ve.Problematic(i.e.maladaptive)behaviors are targeted for elimination using
a variety of well-established behavioral strategies that originate from the rich
eldofapplied behavior analysis.
Douglass Developmental Disabilities Center
The Douglass Developmental Disabilities Center (DDDC) is a therapeutic,
experimental preschoolwhichisafliatedwiththeDepartmentofPsychology
at Rutgers University and is located on the university campus. The DDDC
is designed to research the treatment and education of children with autism;
accordingly, staff and doctoral students in psychology administer the center.
The DDDC has organized the preschool into three classrooms, each grouping
children based on ability.Oneofthegoals is to move the children from a small
group into a larger group setting once the child can function in that setting. A
typical curriculum at the DDDC includes the following: expressive and receptive
language skills(teachingthechildtocommunicateaswellasunderstandwhatis
beingsaidtohim);grossandnemotorskills(workingonthechild’scoordination
withhisentirebodyas well as using his ngers andhandsonsmallertasks);
affect(understandingandexpressingemotion);self-help(daily living skills to
promote independence); cognition (teaching concepts that are pre-academic
oracademicin nature); socialization(whichincludesinteracting with others,
promoting a concept of self, and controlling and promoting various behaviors).
In this broad curriculum, the various teaching programs or units are taught using
direct instruction with one teacher to one child, or in a group setting. Each child
receivesbetweenthirty-veandforty-vehoursofinstructionperweek,twelve
months per year.2
Section One: What Works and What Doesn’t?
27
LEAP Program for Preschoolers
The LEAP program(LearningExperienceAlternativeProgram for Preschoolers
andParents)forpreschoolers with autism promotes an integrated early childhood
education occurring across home, school and community settings. LEAP uses
behavioral practices and developmentally appropriate strategies to implement
the curriculum. A component of the LEAP philosophy is to teach a child with
autism to learn from his peers.3 Within the curriculum, children are taught to
transition from one activity to another, select play, and follow routines and
group activities. Independent playistaughtthrough:1)havingthechildmodel
peers;2)breakingthetaskdownintosmallermoremanageableparts;3)direct
teaching(instruction);4)cuingthechildtothecorrectanswer(prompting),and
5)rewards or reinforcements. Social interaction is taught by creating a structured
environment, using peers, teacher involvement, rewards and role-playing
scripts. Languageistaughtusing“milieuteaching”(whichincludesincidental
teaching)anddirect instruction at the beginning. Teaching style is both child
and teacher directed.Theclassroomhasthreeteacherstosixteenchildren,(ten
typical children and six children with autism).Problematic behaviors are dealt
with by using preventative and positive strategies. To prevent poor behaviors,
LEAP employs class rules, daily schedules, activities, instructional materials,
staff assignments and choice-making. They also use something they term
“Individualized Preventative Strategies,”suchasopportunitiesforadultorpeer
attention, waiting activities, choices and decreasing task demands.
What evidence do the practitioners have that school-
based IBT really works?
The literature on school-based autism programs is replete with descriptions of
programs that have very little data supporting them. The exception to this is
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
28
representedbyvearticlesthatreportsignicantgains with children who have
attended school-based IBT programs.4,5,6,7,8 Fenskeetal,(1985)publishedan
outcome study for eighteen subjects,nineunderageve,nineoverageveat
the PCDI.Oftheninestudentsunderageve,sixofthemindicatedapositive
outcome as a result of the intervention.Oftheninestudentsoverageve,only
one indicated a positive outcome. Positive outcome for this study was measured
by whether the child could live at home and attend a public school or whether
thechildcontinuedtorequiretreatment services.*Positiveoutcomeisdened
very strictly and does not include the gains of those who required ongoing
treatment(and,therefore,remainedintreatmentattheinstitute).Thisgroup’s
gains were not reported as they were in the negative outcome group based on
livingarrangements(whichisanindirect way to measure progress).Inaddition,
the study compared older children with younger children: there was no control
group for comparing results of no treatment or a different treatment, to the one
being offered to both groups of children in this study. Fortunately, these results
are similar to those of many of the home-based behavioral treatment studies
(which use very similar techniques). Therefore, we have some condence
that the outcome of the children from the study conducted at the PCDI was a
result of the curriculum and not a result of the children simply getting older
and maturing.
Three outcome studies published by the DDDC provide data on the individuals
enrolled in their programs.5,6,7 Methodological weaknesses do exist in all three
studies. The Harrisetal.(1990)study compares three groups of children assigned
todifferenttypesofclassrooms(tenchildrenwithautism-vepergroup,and
four typical peers).Classroomassignmentisbasedonseverityofbehavior,
which is problematic because it introduces a relevant variable to autism —
*These ndingsweresignicantatalevelofp<.02;pleaseseeSectionTwoforadiscussiononthemeaning
of signicancelevels.
Section One: What Works and What Doesn’t?
29
behavior. The authors themselves categorize this studyasa“quasi-experimental
design.”9 They found that the children did make gains in language development.
They also found that integration versus segregation didnot inuence rate of
development; however, due to the design awsofthisstudy, the data regarding
integrated versus segregated settings must be viewed as tentative.
Unlike the Harris et al. 1990 study, the other two studies — Handleman et al.
(1991)andHarrisetal.(2000)—donotstatethecriteria by which participants
were assigned to classrooms. Due to this missing information, we do not know
which classroom is responsible for the gains the children made. In other words,
is there an effect created by an integrated or segregated classroom or is this
variable irrelevant? Second, can the improvements seen be attributed to the
original functioning level of the children, or are they due to the techniquesused
in the classrooms themselves? Put another way, if the subjects assigned to the
integrated classroom have more skills that make them capable of learning in a
group setting, how representative are they of the autistic population or how similar
are they to the other group which is comprised of children with less skills? The
ability to learn in a group is an important goal for all autistic children; however,
a large amount of one-on-one teaching or intervention is veryoftenrequired
beforeachildcanactuallylearninagroupsetting.Specically,inorderfora
child with autismtolearnfromagroup,thatchildneedstobeabletorstlearn
through observation and then understand group instruction.
The Harrisetal(1990)study indicates that their subjects, as a group, could be
characterized as “high-functioning.”10 Unfortunately, the lack of representation of
the population of children with autismasawholemakesitdifculttogeneralize
the results, and the fact that the varied groups of children did not have different
forms of treatment(ornotreatment),makesitdifculttojudgewhetherornot
the treatment is responsible for the gains. Fortunately, in the later study the
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
30
researchers did publish the children’s pre-and post IQ scores, which indicate
thatsome ofthe childrendid improve signicantly (using awithin-subjects
designforthestudy).
The Harrisetal.(2000)study follows the original 27 children who spent time at
the DDDC between 1990 and 1992. Therefore, I will focus on this latest study
as it encompasses the long term results of children who purportedly made gains
at the DDDC using widely acceptable IQ measures. These researchers studied
the relationship between the age and IQ of the children when they entered the
DDDC program and their eventual school history. The researchers found that
those children who entered the program prior to their fourth birthday were more
likely to be in regular education than children who entered the program at a
later age. In addition, the intake IQofthesechildreninuencedtheireventual
educational outcomes. Fortunately, the researchers used a few different tests to
measure improvementinthechildren(includingwidelyacceptedIQ measures*),
which provides the reader with a good degree of condence that the gains
observed did, indeed, occur.
There is one study with outcome data from the LEAP preschool model. The data
showed that children made gains in eight out of eight areas measured.8 These
eightareas—nemotormanipulation and writing, language comprehension
and labeling, cognitive counting and matching, and gross motor object and
body movement12 — were measured using only one assessment of the dependent
variable: the LearningAccomplishmentProle(LAP).Unfortunately,theLAP
is not an assessment measure that has been widely proven to be reliable and
valid and is not widely used by psychologists in testing children with autism.
*The Harris et al study(2000)reports IQ testing using the Stanford-Binet test which is widely accepted.
They also use the CARS and the LAP which are less widely accepted measures. In previous studies, how-
ever, children were tested using the Peabody Picture Vocabulary Test-Revised(PPVT-R)6 and the Vineland
Adaptive Behavior Scales: Survey Form.11
Section One: What Works and What Doesn’t?
31
In addition, there was no blind, independent evaluator measuring these eight
dependent variables. The assessments were performed by the teacher, which
introduces rating bias; consequently, it is unwise to trust the results of the
LAP.
Unfortunately, there are several other methodological weaknesses which do
not allow the conclusion that the LEAP model is an effective intervention for
children with autism. The study design lacked a control group. The various
tests were not done independently prior to and after the study, and commonly
accepted psychometric measureswerenotused.Consequently,itisverydifcult
to know whether the gains made were a result of the intervention or simply due
to the child growing older.
Regarding the children in the study, only six children were involved, and
the diagnosis for these children was “autistic-like.” An additional concern
regarding the study was that the diagnosis of the children was not made by an
independent clinician.Thesechildrenwerelabeled“autistic-like”simplybased
on observations along the following criteria: self-stimulation; minimal or no
functional speech; prolonged tantrums; minimal or no positive interaction with
peers; mild to severe range of mental retardation based on McCarthy Scales of
Children’sAbilities(MSCA).13 It is not stated in the article, who it is that made
the observation or administered the MSCA. This lack of rigor in research could
resultinbiasintheclassicationoftheparticipantsasautisticorautistic-like
and could also result in errors in diagnosis, e.g., that these children may have
been PDD or PDD-NOSbutwereerroneouslylabeled“autistic-like.”Therefore,
the participants were not representative of the typical autistic population. The
Diagnostic StatisticalManual(DSM-IIIorIV)wouldhavebeenamorereliable
measure.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
32
What does the therapy actually look like?
IBT is often run in a preschool or school setting; however, the PCDI also runs two
teaching homes. Their preschool and school programsarelimitedtotwenty-ve
students at any given time and services are provided in-home and community
settings as well. Staff are initially trained by the PCDI and regularly evaluated.
The various school and teaching home programs are integrated to foster
consistency and resources are shared between programs. Progress is assessed
in areas of behavior, instructional procedures and family satisfaction.
Staff at the DDDC use a variety of teaching techniquesdependingonthelevelof
the child and the content being taught. They instruct using discrete trial training,
incidental teaching and communication training. Discrete trial training is a very
efcient,systematic, behavioral teaching techniquewhereachildworkswitha
teacher one-on-one, breaking down concepts to make it easier to teach. In contrast,
incidental teaching is a method which attempts to teach a child by focussing upon
the information the child needs to learn when he or she has the opportunity to
learn it naturally. The belief is that the concept will be more meaningful when it
isrelevanttothechild.Thethirdteachingtechnique,communicationtraining,
uses comprehensive speech and language instruction implemented by a teacher,
following the recommendation of the speech therapist.14
Problematic behaviorsaretargetedforeliminationusingthetypical,scientically-
substantiated behavioral techniquescustomizedtothechild.Examplesofthese
methods include a variety of techniquessuchastime-outs, verbally reminding the
child and overcorrection. Every two weeks a speech and language pathologist
assesses the progress of each child. In addition, children’s progress is measured
by using a variety of psychometric tests such as the Stanford-Binet IV, the Battelle
Developmental Inventory and the LearningAccomplishmentProle.
Section One: What Works and What Doesn’t?
33
The LEAP preschool modelisappliedforfteenhoursperweek,twelvemonths
per year. Students with autism are integrated with typical peers in a classroom
setting. In the study by Hoyson, Jamieson and Strain(1984),thereweresixteen
students in the classroom, ten typical students and six students with autism. The
curriculum is individualized for the student, and parents are viewed as partners in
the“educational”process.Theyuseamethod of individualized group instruction
termed TRIIC,theacronymfor“[Tri-I(Innovative,Integrative,Individualized)
Curriculum] for mainstreaming.”15 In this form of instruction, each child is given
individual objectives in three skill areas, and the teacher designs and implements
a group lesson plan that meets the needs of all the children in the group.
Would I send my child to a school-based Intensive
Behavioral Treatment program?
The decision to send one’s toddler off to a treatment facility, even if it is in a
preschool,isadifcultone.Iwouldbeverycarefultoestablish how much of
the day is treatment-based and how much is preschool. If I had lived in New
Jersey when my child was very young, I would have thought very seriously about
sending my child to the PCDI. However, I would have been vigilant to make
sure that the child received treatment every minute of every day. As a parent,
I’ve seen too many preschools that claim they are providing treatment, when
in fact they are providing childcare. This is a serious problem. Government or
universityafliationisnoguaranteethattheautismexpertiseissufcienttorun
a treatment program. Among the worst preschool programs that actually claim
to be ”therapeutic,”aregovernment-funded and staffed programs with a price
tagof$2.5milliondollarstotreattwenty-ve(25)children!
In interviewing the school staff, I would need to know how the school program
and home program are coordinated and monitored. In addition, I would need to
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
34
see evidence that the ratio of one teacher to more than one student is effective.
This is particularly important at the beginning of treatment when most children
with autism do not have the skills to pay attention, understand instructions
orsitatatable. Iwouldalsohavequestions about how the skills learned at
school are going to be generalized across settings(i.e., school tohome)for
children who are not receiving any therapy outside the classroom. If these
questions were answered to my satisfaction, then I might have enrolled my
child. Most importantly, I would need to know when the child is destined to
leave the therapeutic program and is slated to be integrated with his typically
developing peers.
Mychildrequiredone-on-one treatment from the outset, as she did not have
any skills that would have allowed her to learn in a group setting. Therefore,
aslongasshecouldbeplacedinitiallyinthepreparatoryclassroom(withone
teachertoonechild),IwouldhaveconsideredtheDDDC program. I would be
quitenervousaboutmychildmovingtolearninagroup-settingtotheexclusion
of one-on-oneteachingwhichis,inmyview,muchmoreefcientthangroup
teaching. Eventually, we want and need our children to learn in a group setting;
however, it may take the preschool some years to achieve that goal. I would
requireassurancesthatitismydecisionwhensheisreadyforthegroupsetting
and that the decision was only motivated by what is best for my child, rather
than some budgetary constraints requiring more“efciency” and, therefore,
moving her away from a one-on-one treatment setting.
Although the LEAP curriculum sounds like an acceptable curriculum for children
with autism, there is not enough evidence that the children make substantial
gains; therefore, I would not enroll my child in a LEAP program. The integrated
classroom is an interesting idea since it makes sense that a child with autism
should be with typically developing children. However, I would like to see
Section One: What Works and What Doesn’t?
35
a child with autism learn a number of skills prior to integration, in order for
integration to be worthwhile. If my child needed to be mainstreamed, I would
prefer to mainstream the child into a setting that was not therapeutic in nature
but rather was the kind of setting where parents would send their typically
developingchildren.Oncemychildweretaughttolearnthroughmodelling
the behaviors of others, the last thing I would desire is for her to be exposed to
(andpossiblymodel)other autistic children who might engage in repetitive,
self-stimulatory or other maladaptive behavior.
What else do I think?
What is particularly compelling about the PCDI is that although they concentrate
on early intervention, if the child has not graduated from the preschool into a
kindergarten for typically developing children, then that child continues in a
treatment program. The parents are not suddenly left with the impression that
they are on their own because their child is already too old to be in an intensive,
behavioral treatment program. The aging out issue is a criticism of many IBT
programs that tend to concentrate on the younger children and wash their hands
oftheolderchildreneventhoughthesechildrenmayrequiremoretreatment.
It’s a particularly common occurrence among government-funded programs
worldwide, where the sooner the children can graduate from IBT, the less money
the government has to spend.
With respect to the measure used by LEAP – the Learning Accomplishment
Prole (LAP) – this measure does not give a comprehensive assessment of
all relevant areas of developmentand,therefore,doesnot adequately assess
its own intervention strategies.Specically,issuessuchasIQ and behavioral
change are not assessed, and language assessment is limited to naming and
comprehension. In addition, LEAP claims to target social interaction, independent
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
36
play, functional skills and peer teaching; however, none of these content areas
are assessed in the outcome study. The only areas assessed by the LAP are the
eightareasthattheyidentify(nemotorwritingandmanipulation, gross motor
object and body movement, cognitive matching and counting, and language
naming and comprehension). This measure excludes some very important
skilldecitandbehavioral excess areas. Furthermore, there is no evidence that
self-stimulatory or other maladaptive behaviors are targeted for elimination.
While LEAP uses strategies to prevent such behaviors, there is no mention of
how behaviorsaretargetedwhentheyoccur.Duetothenatureandfrequency
of these non-learning behaviors, it is essential that behaviors which obstruct
learning are controlled and, ideally, eliminated.
What kind of study would I like to see the school-based
IBT researchers do?
There are several important factors I would like to see in future outcome
studies on school-based IBT programs. More dataisrequiredtocomparethe
progress of subjects in experimental versus control groups, creating groups
with varied types and intensity of intervention. The dependent measure of
positive versus no positive outcomeneedstobedenedandoperationalized
moreexplicitly.Specically,theadditionaluseofIQ and language assessment
indicators would be helpful in further examining outcome, particularly in
groups of different ages. A measure which indicates the amount of progress
beingmade,evenbythoseindividualswhorequirecontinuedtreatment, must
be incorporated in future research. These variables would ideally be measured
by at least one independent evaluator who is blind to the assignment of subjects
to groups. In addition, factors such as treatmentintensity(homeandschool),
student-to-teacher ratio, dual diagnoses and age at treatment initiation need to
be controlled more stringently in order to determine the many factors which
inuencetreatment outcome.
Section One: What Works and What Doesn’t?
37
In terms of LEAP specically, new outcome data is required before any
conclusions can be made about the LEAP model and its efcacy.TheScientic
Review of Mental Health Practice had the following to say about LEAP:
“Although certain aspects of the LEAP program appear promising, the paucity
of the available research, and especially the absence of controlled research,
precludejudgmentsaboutitsusefulness.”16Oftheutmostimportanceis the
need for a control group in any further investigations.Thisisrequiredinorder
to determine the source of the changes found in the results. Also, it would be
necessary to provide an experimental design which includes a larger subject
pool of children diagnosed with autism by an independent source. Ideally,
these children would be assessed for baseline levels of ability using various
measures including IQ, behavior and more extensive language measurement.
These assessments should also be administered by independent evaluators who
have no knowledge of the experiment, rather than teachers or other individuals
directly involved with the experiment.
Who else recommends for or against the School-Based
IBT for the treatment of autism?
There are many organizations that recommend ABA throughout the child’s life
and in every setting. The Association for Science in Autism treatment describes
ABA as being effective across a variety of settings including school and home.17
In addition, the BehaviorAnalystCerticationBoarduseschildreninschool
settings as an example of the application of ABA.18 After examining the data, no
bonadescientistwoulddisagree with the delivery of school-based behavioral
treatment to treat the condition of autism.Thequestion, however, is whether
treatment should be designed by the school-based or home-based professionals
consulting with the parent.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
38
So you’re still on the horns of a dilemma?
Ifthe question is where to enroll the child — in an ABA school or a home-
based program — the answer depends upon where the child can receive the best
program. The parent needs to determine whether the ABA school in the area
(ifitexists)isofahighquality.Ifnot,thenahome-based program may be the
only option, bringing in competent professionals from the community or, if that
isimpossible,then yinginprofessionals from a different region, or in some
cases, another country.
What’s the bottom line?
The scienticresearch to date collected on children who attend high quality,
school-based, Intensive Behavioral Treatment programs provides evidence to
conclude that their condition improved in school-based treatment settings and
thatalthoughthemostsignicantgains were made by those children who began
treatmentbeforetheageofve,olderchildrenmadesignicantgains as well.
Section One: What Works and What Doesn’t?
39
Offshoots of Intensive Behavioral Therapies:
Pivotal Response Training and the Natural
Learning Paradigm
What is Pivotal Response Training/the Natural Learning
Paradigm?
Pivotal Response Training/the Natural Learning Paradigm (PRT/NLP) is a
techniquetomotivateindividualswithautism to respond to multiple cues. PRT/
NLP targets an autistic person’s lack of motivation and tendency to concentrate
on one stimulusatthe expense of otherstimulior“thebigpicture”(termed
stimulus overselectivity) by targeting these two areas which are considered
pivotal. These behaviors are considered to be pivotal because the theory is
that changing them results in a change in many other behaviors.1 The goal of
the intervention is to provide an easy-to-implement strategy which can also be
used in the community.2 PRT uses some principles of behavior modication
to teach the person with autism. The components of the interventionare: 1)
ensuringattention;2)interspersingmaintenancetasks(tasks that have already
been mastered); 3) allowing the child to lead; 4) giving the child multiple
cue instructions; 5) providing reinforcement immediately; 6) providing
reinforcementcontingently(rewarding the personbasedontheiranswer);7)
providing reinforcement that is directly related to the behaviorortask,and8)
providing reinforcement for any goal-directed attempt at responding. PRT is
designed to discourage the individual from engaging in aggressive, self-injurious,
self-stimulatory and ritualistic behaviors; however, how these behaviors should
bedealtwithisnotspecied.TheNatural Learning Paradigm encompasses the
philosophy of Pivotal ResponseTraining,whichdenesthelearningaschild-led
in a non-demanding setting where Pivotal Response Training occurs.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
40
What evidence do the practitioners have that this
technique works?
PRT/NLP relies upon literature which studies the lack of joint attention behaviors
characteristic of children with autism3andtheramicationsofthelackofjoint
attention on the development of speech and language.4 Although there has not
been a single study comparing children in a comprehensive Pivotal Response
Training program to a well-settled behavioral treatment program(suchasLovaas
and colleagues created), Pivotal Response researchers have found a positive
relationship between very targeted interventions and an increase in speech.5
Although there are dozens of articles which relate to naturalistic teaching
and, by extension Pivotal Response Training, there are currently twelve peer-
reviewed journal articles providing outcome data on individuals with autism
who have been treated using PRT/NLP.4,6,7,8,9,10,11,12,13,14,15, 16 These twelve studies
concentrate on encouraging language through play, encouraging social behavior
and/or sociodramatic play or comparing the natural language paradigm versus a
more structured adult led approach, which they refer to as analog teaching. In
almost all the studies, the sessionswerevideotapedandsubsequentlycodedby
different researchers whose coding was compared to ensure consistency. This
safeguard was important because in some of the studies, peers or parents, not
professionals, were involved in sessions with the autistic child. In most of the
studies, the interobserver agreement(theagreementbetweenthoseresearchers
whocodedthesessions)wasrelativelyhigh,i.e.,inLaski et al. the interobserver
agreementdidnotdropbelowseventy-sevenpercent(77%)andattimeswasas
highasninety-eightpercent(98%).17
Eleven of the twelve studies were single-subjectcasedesigns(seethenextsection
for an in-depth discussion on SSCD)utilizingasmallnumberofchildren(with
the largest studyinvolvingtenchildren),mostofwhomhadadiagnosis based
Section One: What Works and What Doesn’t?
41
on a version of the Diagnostic StatisticalManual(DSM).Overthelasteighteen
years, proponents of PRT/NLP have published dataonatotaloffty-onechildren
who were involved in very short term and/or low intensity experiments(often
lasting no more than thirty minutes a week over three months and often less than
that).Themajority of childreninthesestudies were over three years of age,
withmanybetweentheagesofveandtenyears,someofwhomwereveryhigh
functioning.15,16 Based on these children, results have been reported that children
with autism utilizing PRT/NLP have more prosocial behavior, improvements in
social skills(andplay),andanimprovement in speech and language. Although
these results sound encouraging, the studies as a whole have several serious
drawbacks. Due to the complexity of the studies and the various claims made,
each claim will be discussed separately.
Is Pivotal Response Training/Naturalistic Learning
Paradigm more effective for language acquisition?
PRT/NLP researchers and proponents claim that naturalistic teaching, when
used for speech and language, is more effective than the traditional research
supporting discrete-trial training.16,17,18,19,20 This may have occurred with the
subjects in their experiments; however, this claim cannot be generalized to the
population of children with autismfortworeasons:1)thesmallnumberofchildren
per study(usuallytwoor threechildrenineachdesign)istoofewchildrento
make generalizations about the effectiveness of PRT/NLP for the overall autistic
population and none have any follow-up after the study to see whether the
observed gains were permanent;*2)thedata from PRT/NLP is based on children
with varying degrees of language impairment who have had discrete trial training
*There is one study where the researchers retrospectively studied intervention data from children who
did well or poorly in prior treatment. Unfortunately, they used retrospective pre-intervention archival
data and compared it to the postintervention data rather than following the children from their study
longitudinally.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
42
learning histories prior to being part of these studies. Naturalistic or incidental
learningispredictedtobemoreefcientforchildrenwhohadachievedsome
competency in language and/or who had extensive amounts of past treatment
using discrete trial training. The reasoning here is the child has already achieved
the skills needed to learn in a more natural setting.15 This is an important point
because the suggestion that comes out of the PRT/NLP literature is to abandon
one of the most important techniquesforsomechildrenintheABA toolbox
-- discrete trial training. This would be a severe mistake with a child who
appears to be completely unteachable (which is common for children with
autistic disorder),andforwhichdiscrete trial training may be the only option
at the beginning of a treatment program.
Onestudy4 compared naturalistic teaching with more structured teaching to
determine which was more efcient. They found that naturalistic teaching
was much more motivating than structured teaching. However, this study has
a fundamental awwhichseriouslyunderminestheresults of the study.* In the
naturalistic condition, the clinician used highly-motivating three-dimensional
items to teach the target sounds; whereas, the analog condition used picture
cards with the items on the cards to teach the same sounds, and then praised the
child and reinforced the child’s correct response with food or a desired object.
This research demonstrates that using a desired object to teach a sound(orany
conceptforthat matter)willbemorepowerfulbecausewhatisbeing taught
is intrinsically rewarding. However, it does not demonstrate that naturalistic
teachingismoreefcient.Thisstudy needs to be done with the clinician in both
conditions using the highly-reinforcing three-dimensional items to teach, in both
*The way analog vs. naturalisticteachingisdenedconfusesthefundamentaldifferences.InKoegel, Koegel
andCarter(1999),theydenethedifferencebetweennaturalistic and analog teaching very strictly, making
the point that in analog teaching the child has no choice.14 There is, however, no contradiction between
analogteachingandgivingachildachoiceoftheactivityheorshewouldliketodorst.Thedifference
has more to do with the child-led versus adult-led nature of the actual teaching trial.
Section One: What Works and What Doesn’t?
43
the structured(analog)andnaturalisticconditions.Otherwise,thedifferences
in teaching techniques are being confused with differences in: 1) degree of
reinforcement,and2)therelevanceofthereinforcer to what is being taught.*
Do children with autism emit less disruptive
behaviors with PRT/NLP?
The second claim made by proponents of PRT/NLP is that children emit less
disruptive behaviors using the naturalistic teaching paradigm.21 It is plausible
that initially there would be a difference between adult-led and child-led therapy
in terms of disruptive behavior. It makes perfect sense that behavior will not
be a concern if no demandsaremadeofachild.However,therealquestion is
whether these children will progress to the point where they can cope in situations
where their ideas or way of doing things is not adopted, and be able to learn
todo whatothersrequireofthemwithout emittingdisruptive behavior, as all
typically developing children are expected to learn from an early age. Another
point worth emphasizing is that in good analog teaching, disruptive behaviors
should not occur on a regular basis even when demands are placed on the child.
I suggest that the researchers have inadvertently compared a naturalistic learning
environment to a poor analog teaching environment where the reinforcement
levelsareinsufcient.Inotherwords,thetwodifferenttypesofprogramswere
ofadifferentquality.Theycomparedahigh-qualitynaturalistic-teachingprogram
withaverypoor-qualityadult-ledanalogprogram.
Aspreviouslymentioned,thelearninghistories(previoustypesoftreatments)
ofthesechildrenneedtobetakenintoaccount.Onesetofresearchers describe
*For this studytobevaluable,thereshouldbefourconditions:1)analog condition, relevant reinforcer;2)
analog condition, irrelevant reinforcer;3)NLP condition, relevant reinforcer,and4)NLP condition, irrelevant
reinforcer. Although we can predict that condition four will be the least successful, it is not clear whether
conditiononeorthreewillbemoreefcient.Inshort,thevariableofreinforcer needs to be controlled.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
44
achildwhoactuallysays“Nocards.”22 This indicates to me that this child has
experienced poorly delivered therapy which has created an aversion to learning.
A poor therapy experience prior to the current study may have seriously biased
the results of that studyaswouldanychild’spriorlearninghistory(oneofthe
issues researchers using single-subject case designs studiesmustaddress).
Do children with autism increase their social and play
skills with PRT/NLP?
There have been a few studies8,9,10,11,12,16 which attempt to use PRT/NLP to increase
the ability of autisticchildrenintheseareas.Onestudy has been published in
two separate articles, one concentrating on language and toy play,11 the other
observing social behavior.12 Both of these articles appear to be a replication of the
earlier publishedndings.8 This study, conducted on two children with autism,
and two typically developing peers, reports positive changes in social language
and play skills. The children were ten years old and had language abilities over
three years of age prior to entering the study. Although their language ability
is not at age level, it is at a degree much higher than many young children with
autism, so their skills might not be representative of children with classic autism.
This was also the case of a more recent study which used two children who were
eight and nine years of age16 (treatmentprogramsforchildrenwithautismtypically
begininthetoddleryears).
Anadditionalquestion concerns the validity of ndings regarding play. These
children with autism may have memorized repertoires that they learned from
typically developing children when in therapy, and then use when playing with
another group of children. In other words, creative pretend play does not occur.
Although this criticism may seem hypercritical, and I think that children with
autism may benet by memorizing a number of scripts to use while playing
Section One: What Works and What Doesn’t?
45
with peers, it is important to differentiate whether the child is reproducing play
repertoires or whether he or she is truly engaged in pretend play(thetwocanbe
differentiatedbytheuniquenessofeachsession without peer prompting).Two
alternative hypothesestoexplaintheresultsarethat:1)anautistic child may be
incorporating a peer into rigid, role playing, which is not about joint attention or
truesocialengagement,butsimplytheuseofapeerasa“tool”forahigherlevel
of self-stimulatory orrepetitivebehavior,or2)thechildmaybeusingmemorized
scripts which are activated when particular toys are present. In addition, one
study counted the number of play date invitations made after the intervention.
This measure may have more to do with the parent’s ability to be reinforcing to
the typical peers than any actual increase in friendship.16
Another article published by Stahmer10 was far superior to the above studies
and allows us to unravel the complexity of the ndingsonplay. This study
used a control group who provided language training, had more extensive
dependent measures,madesuretheobserverswere“blind”totheconditionof
the participants and reported the statisticalsignicanceof theresults(a “p”
value).Animportantcontributionofthisstudy is the researcher’s honesty when
she suggests that for individuals without a certain level of language ability, the
intervention may not be developmentally appropriate.23 Stahmer discusses one
child whose stereotyped play interfered with his learning and noted that the
“children with the best language skills were the most creative and spontaneous
during play.”24 This is an illuminating point because this data demonstrates that
if Pivotal Response Training does ameliorate autism, it is only so for a high-
functioning subset of children or a subset of children who have reached a certain
level with well-settled IBT programs. Stahmer’s research is important because
it introduces the concept that PRT/NLP may be useful for a certain subset of
children with autism but not effective(orpremature)foranothergroupofchildren
whodonotpossesstheprerequisiteskills.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
46
Do skills learned through PRT/NLP generalize across
settings and people?
The ability of children with autism to generalize play skills based on this
treatment methodremainsdebatable.Itis still an open questionwhether the
childrenwhobenettedfromPivotal Response training in the above-mentioned
studiesalreadyhad(priortoPRT/NLPtreatment)manyoftheskills needed to
learn and generalize symbolic, complex and creative play. We are still uncertain
regarding whether it was the method of intervention or rather the children’s
readiness to generalize that made the difference, if indeed, these children did
actually generalize play and social skills at all.
The claim that PRT/NLP skills will generalize across settings and people is
more convincing when the consequence ofusing language is reinforcing. To
illustrate, if a child learns to ask for juice and receives juice every time he asks,
the data suggest that this skill will generalize across settings. Whether or not
alessreinforcingrequestwillgeneralizeisstillanopenquestion.Onestudy25
attempts to address this problem by gradually changing hidden reinforcing items
in a bag to less preferred ones, without affecting the spontaneity of the child
asking,“What’sthat?”referringtothehiddeniteminthebag.Whetherornot
thisquestion will be a permanent part of the child’s asking repertoire down the
road is unknown, although it is plausible that this skill may be maintained as
long as the reward is unpredictable. Nowhere is there any compelling evidence,
however,thatthisquestion-asking skill will generalize more or less successfully
if it were taught using an analog method(aslongastheskillweregeneralized
afterbeingtaught).
Section One: What Works and What Doesn’t?
47
What does the therapy actually look like?
PRT/NLP is described as the use of “loosely controlled environments [e.g.
a playground] and that utilizes shared control [e.g. turn taking] and multiple
exemplars [e.g. many toy materials].”26 Techniques such as turn-taking
opportunities, working on mastered skills and gaining the child’s attention are
used to set the child up for success. Parents are trained to use these techniqueswith
their children to encourage language development and use. The three variables
that structurethelearningsituationaredescribedinthePRT/NLPmanualas:1)
thechildisgivenaninstruction,question or spontaneous opportunity to respond;
2)thechildresponds,and3)thechildisgivenaconsequence.27 Although this
sounds very similar to the traditional one-on-one behavioral treatment procedure,
thenaturalconsequenceoftheinstructionoropportunityisofimportance.To
use their illustration: a child is cold while playing outside; the mother tells the
childtoputonacoat,andthechilddoes.Thenaturalconsequenceisthatthe
child plays outside again, but this time he feels warm.28Naturalconsequences
such as those illustrated above can be highly motivating and, therefore, useful
whenteachingandmaintainingaskillastherewardisalwayspresent(inthis
case,warmth).
Would I try it on my child?
If my child were recently diagnosed, I would not rely upon PRT/NLP to ameliorate
her autism due to the lack of data reporting efcacyforyoungchildren who
are not yet speaking. This is an example of a promising area of research that is
leaving the laboratory too early and being incorporated prematurely by parents
and educational systems prematurely. Because I wanted to provide my child
with the most evidence-based treatment, PRT/NLP would not have been my
choice. That said, in established, well-settled behavioral treatment programs,
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
48
natural consequencesthat arereinforcingshouldbeusedwheneverpossible.
This principle is a foundation in the PRT/NLP literature.
What else do I think?
Most of these studies do not sufciently rely upon standardized outcome
measures. The behavioral outcomes are generally measured by researchers who
are part of the study as are the emotional and social outcomes(withtheexception
of the occasional study where they use the Vineland Adaptive Behavior Scales —
atestwhichlooksatavarietyofbehaviors). In addition, the social and language
behaviors(theoperationalization of the dependentmeasure)occasionallyuse
researcher observationalone.Thisisnotanadequateorunbiased measure of
treatment outcome and there is no indication within some of the studies about
who is rating the observations. In addition, one study includes a measure of
teacher reported social behaviors within the classroom.29 In my view, this is
a biased measure of change because it is not clear how much information the
teacher has about the study(asshewasinstrumentalinchoosingpeers for the
study).30
The authors of the PRT manual claim that it is designed for any child, including
those who are nonverbal. This conclusion is premature because subjects in all of
the studies had baseline language abilities that were higher than those typically
found in the population of young autistic children. In contrast, Stahmer(1995)
suggeststhatforindividualswithoutsufcientlanguage ability, the intervention
might not be developmentally appropriate.10 In other words, if a child is not
speaking yet, it is too soon for PRT.
This intervention approach emphasizes that the task must be child-led. The
manual states that the child must be able to choose the topic of an activity and
Section One: What Works and What Doesn’t?
49
when to stop the activity. Concurrently, the authors state that disruptive behavior
is not acceptable and parents must take control until the child is capable of non-
disruptive behavior. It is not clear how this philosophy meshes with the child-led
philosophy and how parents are to “take control”inthisframework.Itislikely
that a child who has had no interventionatall,andsubsequentlyhasfewskills,
will be very resistant to initiating or remaining involved in an interaction.
Another issue of concern for the child-led approach is the lack of motivation
whenthematerialisdifcultorintrinsicallynon-reinforcing.Itisunlikelythat
achildwillinitiatelearningdifcultconcepts, as he or she has no understanding
of them and the concepts might not be relevant. In addition, when a child is in
school, that child will be expected to participate in classroom learning, as do his
peers. This will be an additional challenge unless the child has learned how to
sit and learn material that is perhaps not intrinsically motivating. In addition,
the application of this method in a mainstreamed classroom is problematic. The
requirementthatpeers undergo extensive training in order to learn the strategies
proposed by PRT/NLP is highly unrealistic.
Another problem with this approachis“wheretobegin.”Theauthorsrecommend
that instructions given to the child should be multi-cued instructions. Children
with autism do not typically understand multi-cued instructions. Indeed, they
need to be taught how to understand multi-cued instructions and there is no
techniqueofferedtoguideparentsinhowtoteachthistotheirchildren.Prior
to using Pivotal Response Training, it can be argued that children need to be
taught a variety of single instructions before they are expected to understand
multi-cued instructions.
ThelastpointIneedtomakeregardsefciency.Ifwearerequiredtowaitfor
children with autism to initiate everything they need to learn, I am concerned
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
50
that we will lose precious time that early interventionrequires,andthefuture
ofthesechildrenmaybecompromised.Inaddition,itisextremelyinefcient
tondadirect,naturalconsequenceforeverythingthatthechildmustlearn.
Anotherdifcultywith this offshoot of behaviorism is that it smacks of the
“parent as therapist” ideology of how autism should be addressed, wherein
parents,asopposedtoprofessionals,delivermostorallofthe“treatment.”6,31,32
Laski et al. actually report on a parent-training as a positive outcome when
they state:
This study presented a promising new parent-training program designed
to increase autistic children’s Verbal Behavior. Post treatment increases in
parents’ requests for vocalizations from their autistic children were observed
in the generalization settings. Additionally, parents showed evidence of
generalizing these behaviors with the siblings of their autistic children. These
generalization effects are encouraging in that they may provide additional
supportofthe motivating qualities of natural language programs for both
parent and child...33(emphasisadded).
It goes without saying that it is not the parents with the neurological disorder,
it’s their child. The assumption that increased parent vocalization will cause
increasedchildvocalizationisnotsufcientlysupportedbydata. In addition,
the expectation that the parents must be responsible for the therapeutic treatment
of their child plays into the rationing of health care for children with autism,
which governments will happily entertain if they receive academic justication
for it.*
*The concept of parent as therapist is particularly offensive when one considers the fact that parents of
children with autism also have to make a living like everyone, at the some time as being responsible for
their autistic child’s progress. I’d like to suggestthatthefree-wheeling1950s“LeaveittoBeaver”family
unit is rare in 21st century modern society and would like to see any of these researchers be productive
in their academic careers undertaking the role of therapist for their autisticchildren.Onlyintheeldof
autism are our children considered so unworthy that the responsibility of treatment falls on the family
instead of on professionalsintheeld.
Section One: What Works and What Doesn’t?
51
What kind of study would I like to see the researchers do
on Pivotal Response training and the Natural Learning
Paradigm?
There may be a place for the use of PRT/NLP in the education of individuals with
autism; however, exactly where and/or if it can be used effectively needs to be
established. For individuals who have some language ability and are able to learn
somewhatincidentally,i.e.,theydonotrequiremasstrialsofrepetition in order
to retain certain pieces of information, Pivotal Response behaviors may be a good
method of prompting generalization of desired behaviors. Research to determine
exactly who, and how much, this approachcanhelpisdesperatelyrequired.It
also needs to be determined how effective this intervention is when compared to
other treatments. I would like to see a between-within subject design utilizing
a comprehensive protocol based on the PRT/NLP paradigm as compared to an
intensive, well-settled behavioral treatment program. In addition, it is crucial
that the PRT/NLP researchers use standardized language and IQ measures prior
to and after the study to determine to what extent autism has been ameliorated
using their protocol.TheauthorsclaimthatoneofthebenetsofPRTisthatthe
behavior will occur in natural environments; however, this has yet to be supported
byhighquality,long-termevidence. Clearly, a well designed, longitudinal study
with large numbers needs to be conducted prior to recommending this method
to anyone responsible for the treatment of children with autism.
Who else recommends for or against Pivotal Response
Training as a method for the treatment of autism?
The Developmental BehavioralPediatricsOn-line(asitecloselyconnectedwith
the AmericanAcademyofPediatrics)hasreviewedPRTandstates:“Although
each of the components of the Pivotal Response intervention model has been
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
52
extensively tested, there are no randomized trials comparing PRT to any
other intervention model. The only published follow-up study was done
retrospectively.”32,34 Other than that, PRT/NLP has been protected from
criticism as it falls under the general rubric of ABA which is a well-established
discipline.
So you’re still on the horns of a dilemma?
I would recommend that prior to embarking on a program reliant solely on PRT/
NLP, that you have your child in a well-settled behavioral treatment program and
await more data which demonstrates that a comprehensive behavioral treatment
program using PRT/NLP is more effective than a traditional behavioral treatment
program. That said, the incorporation of natural consequenceswhenpossible
into a well-settled behavioral treatment program is certainly a powerful way to
reinforce skills or positive behavior.
What’s the bottom line?
Based on the scienticresearch to date, there is not enough evidence that Pivotal
Response Training/The Natural Learning Paradigm is globally effective in
ameliorating the condition of autism. There is some very preliminary evidence
to suggest that this method can be used to target symbolic play skills in some
individuals with autism who possess above average language abilities. However,
further studies with larger sample sizes and standardized testingarerequiredto
appropriately evaluate the method.
Section One: What Works and What Doesn’t?
53
Offshoots of Intensive Behavioral Therapies:
Positive Behavioral Support
What is Positive Behavioral Support?
Positive Behavioral Support (PBS) is behaviorism guided by philosophy.
Practitioners of this methodclaimthatitisaneweldthathasitsrootsinapplied
behavior analysis, the inclusion movement and person-centered values.1 The
philosophy promotes the inclusion of people with disabilities in mainstream
society. By re-engineering the environment, it is claimed by PBS proponents that
theindividual’squalityoflifeisenhancedandthereby,behavior problems can be
minimized. What appears to differentiate PBS from other forms of behaviorism is
the promotion of educational systems to take responsibility in the re-engineering
of environment and the practice of PBS. Proponents claim that the elimination
of problem behavior is not the direct focus of PBS, but rather, a fortunate by-
product. They state: “the primary intervention strategy involves rearranging the
environmenttoenhancelife-styleandimprovequalityofliferatherthanoperating
directly on reducing problem behaviorperse.”2 They differentiate PBS from
other forms of behaviorismbytheir“Life-span Perspective” and suggest that
meaningful change may be slow and, in fact, may take decades.
Positive Behavioral Support is differentiated from traditional behaviorism by the
emphasis on “ecologicalvalidity,”whichproponentsdeneastheapplicability
of the science to real-life settings. In other words, their vision is for parents,
teachers and job coaches, rather than professionals, to practice PBS. An additional
component of the PBS philosophy is “stakeholder participation”whichtheydene
as a consumer-driven, rather than an expert-driven, applied science. In other
words, the consumer is supposed to become an active participant in delivering the
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
54
PBS treatment. The third concern of PBS practitioners is that the interventions
be measured not by their “objective effectiveness”butratherbytheimpacton
theperson’squalityoflife(e.g.,isthepersonhappyandinanacceptableliving
arrangement).
What evidence do the practitioners have that this
really works?
Here is where Positive Behavioral Support becomes somewhat tricky to evaluate.
Since PBS practitionershavethemselvesoftencomefromtheeldofapplied
behavior analysis (ABA), some spending most of their academic career in
thiseld,itisverydifculttoseparatemuchoftheresearch they cite that has
been done on ABA from the research that has been done on PBSspecically.*
A comprehensive database search netted sixty-ve (65) articles on Positive
Behavioral Support (attempting to differentiate PBS from ABA). Of these
sixty-vearticles, there were only six articles presenting experimental data of
any kind on children with autism.Ofthosesixarticles, one study concerned
parent perceptions of an early intervention program,3 seven case studies were
presented in four articles,3,4,5,6,7,8 one single-subject case design demonstrated a
decrease in disruptive behavior,7 and one article reported on parent-professional
collaboration.8 Aside from these articles, all other articles to which PBS lays
claimactuallyowfromtheeldofABA. Another concern regards the time-
line for progress; PBS proponents evaluate changes made along the life span.
Understandably,itisverydifculttoevaluatea treatment’s value if its effect is
observable only over decades.
*Proponents seem to have expropriated decades of research in applied behavior analysis as their
ownwhentheyaredemonstratingtheefcacyofanintervention.12
Section One: What Works and What Doesn’t?
55
What does the therapy actually look like?
The difference between PBS and ABA is not in what you see, but rather, in the
design of the interventions. The process that the PBS practitioner goes through
is much the same as that of a behavior analyst, although the terminology is
different. The PBS practitioner does a Functional Behavioral Assessment(FBA)
to determine the function of the child’s behavior and then designs an intervention
to reduce or, ideally, prevent the behavior from occurring again. However,
the behavioral intervention that is chosen may or may not differ from that of a
traditional behavior analyst, depending less upon what might actually be the ideal
intervention, and more upon the PBS view of the feasibility of the intervention in
the“realworld.”Inotherwords,thebehavior to replace the problematic behavior
must be: “acceptable to caregivers; appropriate to the setting; within a person’s
skillsetoreasytolearn,”9 and appears to be more concerned with philosophy
rather than science-based measures of treatment outcomes.
Would I try it on my child?
I have refused, and am vigilant, to protect my daughter from anyone with this
treatmentperspectivewhethertheybeanautism“professional”orateacher.In
my view, PBS is a case of political correctness interfering with science. I want
to ensure that my child’s treatmentisnotinuencedby“resources”inthesystem
atanyonetime(orlackthereof—mostoftenthecase).Specically,Idonot
desire that the educational systemberesponsibleforher“support”becausethis
rigid,calciedsystem has shown itself time and time again to be a receptacle for
incompetence when it comes to children with autism.10,11 School districts often
fund programsbaseduponthePBSphilosophybecauseitis:a)politicallycorrect
and in line with the prevailing educational philosophy for typically developing
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
56
students,andb)isinexpensive,asuntrained(orinsufcientlytrained)employees
are expected to implement the interventions.
It is crucial that all the tools in the ABA toolbox are at the disposal of the
professionals who design and implement my child’s treatment program, and
notonlytheonesthathavebeendeemedas“acceptable”or “philosophically
pure”accordingtotheschooldistrict,butalsotheonesthatmaynothaveany
relevance in my child’s life or her autism treatment needs.
When philosophy, rather than data,inuencesdecisionsitis harmful. When
my child was very young, prior to her being mainstreamed in school, she
requiredanintensiveone-on-one ABA treatment regime to reach the point where
mainstreaming was desirable and possible. Without that work, her mere physical
proximity to typically developing children would have been of no use. PBS
might make everyone in the system feel as if they are good people; however,
children with autism need to progresstothepointwhere:1)mainstreaming is
actuallyofbenettothem,and2)theyaretreatedwithdignityinamainstream
setting and not treated like the token disabled person whose disruptive nature is
simply tolerated due to political correctness. This is all too often the case when
philosophy, rather than science, guides decisions.
What else do I think?
In my opinion, Positive Behavioral Support is a very dangerouseldforchildren
with autism. The reason this philosophy is dangerous(asidefromtheobvious
which is research being subjugated by a form of religion – and I think PBS is
atypeofreligionofpoliticalcorrectness),isthatitdenieschildrenwithautism
access to proven, science-based treatment methods.Frankly,Indeventheterm
PositiveBehavioralSupport offensive.The factthatitis“positive”behavior
Section One: What Works and What Doesn’t?
57
support carries a presumption that the PBS practitioner is different and apart from
his“evil”ABA behaviorist counterpart. In fact, traditional behavior analysts
have very stringent ethical guidelinesthatpractitionersare requiredtofollow
inordertobecertied.12 All academic research done on human subjects must
pass university ethics boards and any clinical treatment that may be considered
ethicallyquestionable cannot be conducted by reputable practitioners without
judicial oversight. It is no longer the freewheeling 1950s where many ethically
questionableactivitiescantakeplacebehindthewallsofgovernment institutions
in the name of therapy. There are laws now in place which protect disabled
people from direct harm.
I also nd it interesting that PBS practitioners see no contradiction between
inclusion and mainstreaming, and redesigning the environment to accommodate
children with autism. To illustrate, if we do not teach children with autism to be
able to cope with the general chaos of life, how are we going to have them go
into a shopping mall and function properly? Asking the mall administration to
turn off the music in the elevators prior to a child entering is not practical!
I live in a region where Positive Behavioral Support is used extensively by
school districts and is wholeheartedly supported by government. This region
happens to be an area where autism policy is functionally in the 1950s in terms
of efcacy.Consequently,thereisalargeandsteadyexodusofparentsoutof
the public school system into the private system, or home-schooling, due to this
globally ineffective and harmful philosophywhichmasqueradesasanapplied
behavioral science.
OnenalpointonPBS:whenonereadstheliterature from Positive Behavioral
Support, autism appears to be an entirely different disability. These children
seem to be very mild, and the behavior problems are all easy to control, as long
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
58
astheenvironmentis“re-engineered.”Childrenwithself-injurious behavior do
not seem to be a challenge for this group. Perhaps children who participate in
PBS studies are not classically autistic. My caveat regarding the type of children
participating in PBS studies is also supported by Durand and Rost13 who truthfully
caution those reading the literature on PBS that there may be a selection bias in
the subjects for the studies that they do conduct.
What kind of study would I like PBS researchers to do?
I think that this group of researchers should abandon the anti-science, anti-
intellectual discipline they have developed. They should return to the eld
of applied behavior analysis, compete with researchersinthateldandhave
their PBS research properly scrutinized and evaluated by their ABA academic
peers.
Who else recommends for or against Positive Behavior
Support as a method for the treatment of autism?
MulickandButter(2005)14 provide a very useful, in-depth critical analysis of
Positive Behavior Support that I highly recommend prior to even thinking about
using this so-called autism treatment method. Mulick and Butter lay out the
complete history of Positive Behavior Support and expose the pseudo-science
of PBS in detail.
So you’re still on the horns of a dilemma?
If you are still not sure whether your child should be in a behavior management
program based on Positive Behavior Support, I would encourage you to ask
the purveyors of PBS how they intend to measure short-term outcomes. You
need to make sure that these outcomes objectively measure the child’s progress
Section One: What Works and What Doesn’t?
59
(measuringbehavior and IQ)andnottheapprovalratingoropinionsofothers
about how the child is progressing. In addition, it is crucial to ensure that the
goalsareshort-termandsubstantive,ratherthanfuzzy,long-termquality-of-life
goals which can be easily manipulated to appear rosy.
What’s the bottom line?
Based on the scienticresearch to date, there is no evidence to conclude that PBS
is anything more than a philosophyratherthanascience.Consequently,thereis
no evidence to demonstrate that PBS ameliorates the condition of autism.
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
60
Section One: What Works and What Doesn’t?
61
Offshoots of Intensive Behavioral Therapies:
Verbal Behavior
What is Verbal Behavior?
In 1957, B.F. Skinner(thegrandfatherofbehaviorism)publishedabookcalled
Verbal Behavior.1 In this book, Skinner applied his ideas about learning to
Verbal Behavior.Specically,SkinnerdenedvarioustypesofVerbal Behavior
that humans exhibit. This typology improved behaviorist understanding of
how different parts of language are developed and enabled them to teach the
various functions of language and set up an environment to promote the use of
these parts of language. Sundberg and Partington(1998)takeSkinner’s ideas
and apply them to teaching language to children with autism in a user-friendly
formatdesignedforthenonacademicaudience.Theydenethevarioustypes
of verbal behavior and provide a discussion of each language repertoire in their
book, Teaching Language to Children with Autism or Other Developmental
Disorders. They describe these language repertoires as: receptive(complying
with or following directions);echoic(repeatingwhatothershavesaid);imitation
(copyingactions);tact(labelling);mand(asking);RFFC- Receptive by Function,
Feature,and Class(identifying items basedon theirdescription); intraverbal
(answeringquestions on a more conceptuallevel);textual(readingwords),and
written(scribingwordsthatareheard).2
Put simply, Verbal Behavior Therapy is an attempt to utilize the principles of
behaviorism to teach children with autism to communicate. The concept of
verbal behaviorhasexistedforapproximatelyftyyears;theprimaryimpetus
of verbal behavior is the application of the ideas of B.F. Skinner to children with
autism. Although Sundberg and Partington have been working on these ideas
since approximately 1978, this area has become increasingly popular in the last
The Complete Guide to Autism treatments: Make Sure Your Child Gets What Works!
62
ten years because ideas regarding how to teach language to children with autism
are presented in a more accessible format than was previously available. In
addition, there has always been a need to target languageacquisitionforchildren
with autismasthesechildrenhavesignicantlanguage decits.
What evidence do the practitioners have that this
really works?
Here’swherethedifcultybegins.AlthoughtheideasofSkinner regarding verbal
behavior are compelling and theoretically rich, the testing of these ideas has lagged
far behind. Unfortunately, what little data does exist is not necessarily on children
with autism.3 A comprehensive literature search using all the major academic
databases did not net even one study to provide evidence that a comprehensive
Verbal Behavior program would signicantlyimprove the language ability of
children with autism, and/or facilitate more comprehensive or global improvement
in their condition. However, Verbal Behavior researchers have done studies
concentrating on evoking manding (asking)4,5 and increasing vocal behavior.6
This area will hopefully bear more fruit with additional studies.
What does the therapy actually look like?
The teaching sessions are initially one-on-one(onetherapisttoonestudent)and
they look very similar to traditional, well-settled behavioral treatment programs.
Prior to working with the child to teach the various parts of language, there
is1) a language assessment using a Behavioral Language Assessment Form7
that determines which skills the child has mastered and which skills need to
bedeveloped, and 2)an emphasis onmakingthe therapeutic setting fun and
reinforcing.TherstskilltheygenerallyteachisManding(whichisteaching
thechildtorequest).Typically, the therapist has an item the child wants but
needs h