Review Summary Assignment Paper

THE JOURNAL OF SPECIAL EDUCATION VOL. 38/NO. 2/2004/PP. 95–103 95

It is indisputable that the identification of autism is signifi- cantly on the rise. Concomitantly, there has been a steady and steep increase in autism litigation (Zirkel, 2001). The most controversial segment of this litigation, which focuses on the appropriateness of applied behavior analysis (ABA) programs, has been subjected to insufficient systematic study. Review Summary Assignment Paper

The purpose of this study is to analyze the pertinent case law related to the two central issues of contention between parents and school districts—program selection (i.e., the choice between competing instructional approaches) and im- plementation of said program (e.g., its location, duration, or frequency)—in terms of winning parties (i.e., district or par- ent) and in terms of identifying the factors noted in the cases related to the outcome. This review of the literature addresses the current definitions of autism, ABA and its primary com- peting instructional approach, and the previous research on autism litigation.

The Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition (DSM-IV; American Psychiatric As- sociation, 1994) defines autism under the umbrella category of Pervasive Developmental Disorders (PDD), while medical pro- fessionals refer to both PDD and autism as autistic spectrum disorder (ASD; Filapek et al., 1999). Regardless of the nomen- clature used, ASD and PDD refer to the same continuum of behaviors with a cluster of unusual characteristics: lack of so- cial responsiveness, delays in speech or inadequate quality of speech, restricted or stereotypic interests, delays or abnormal- ities in social interaction, and lack of symbolic play (DSM-IV). On the other hand, the Individuals with Disabilities Education Act (IDEA) regulations (1999) define autism as follows:

The “Discrete Trials” of Applied Behavior Analysis for Children with Autism:

Outcome-Related Factors in the Case Law

Claire Maher Choutka, Patricia T. Doloughty, and Perry A. Zirkel, Lehigh University

This study provides an analysis of case law concerning applied behavior analysis (ABA) for students with autism to determine outcome-related factors. The authors classified the 68 pertinent hearing/ review officer and court decisions published in EHLR (Education for Handicapped Law Report) and IDELR (Individuals with Disabilities Education Law Report) into 2 groups representing the central is- sues of contention between parents and districts—program selection (e.g., instructional approach) and program implementation (e.g., its location, duration, or frequency). For both groups, the outcomes, in terms of who won, did not favor either parents or districts. The three factors predominantly associated with wins by either party for both groups of decisions were testimony of witnesses, documentation of progress, and Individualized Education Program elements. Review Summary Assignment Paper

A developmental disability significantly affecting verbal and nonverbal communication and social in- teraction, generally evident before age three, that adversely affects a child’s educational performance. Other characteristics often associated with autism are engagement in repetitive activities and stereo- typed movements, resistance to environmental change, or change in daily routines, and unusual re- sponses to sensory experiences.

The spectrum of autism and its unknown etiology (Na- tional Research Council, 2001) contribute to controversy, rather than consensus, in regards to selection of the appro- priate instructional approach. Professionals and parents agree that early intervention is vital (McGee, Morrier, & Daly, 1999), but the specific nature of the intervention is disputed. Al- though treatment programs for children with autism have shown positive results (e.g., Koegel, Koegel, Harrower, & Car- ter, 1999; Koegel, Koegel, Shoshan, & McNerney, 1999; Lo- vaas, 1987; McGee et al., 1999), methodological problems with some of the peer-reviewed, published studies haunt par- ents, educators, and the legal community. Review Summary Assignment Paper

The two most contested instructional approaches for children with autism are ABA and TEACCH (Treatment and Education of Autistic and Communication Handicapped Chil- dren; Gryzwacz & Lombardo, 1999). They share some basic components, such as predictable routines, supportive teaching arrangements, planned transitions, and family involvement. However, these approaches also have basic distinguishing characteristics and limited research.

Address: Claire Maher Choutka, 61 Eileen Lane, Limerick, PA 19468

96 THE JOURNAL OF SPECIAL EDUCATION VOL. 38/NO. 2/2004

ABA is the study of behavior and the manipulation of contingencies and setting events to increase or decrease specific behaviors. Behavior analysts use objective measures of the de- sired behaviors and monitor the results of instruction to en- sure skill acquisition. One small subset of this methodology is discrete trial therapy (DTT). A further modification of DTT is Lovaas therapy, after its namesake, Professor O. Ivar Lo- vaas, of the University of California at Los Angeles (UCLA). The “discrete trial” in DTT refers to the basic teaching unit delivered in one-to-one instruction. Although DTT and ABA are synonymous for most parents and school officials, discrete trial is but one aspect of ABA. Typically, parents request DTT for 40 hours a week, arriving at an Individualized Education Program (IEP) meeting armed with Lovaas’s original research study citing a 47% recovery rate (Lovaas, 1987) and the book Let Me Hear Your Voice (Maurice, 1993). The author of this book, a mother of two children diagnosed with autism, pro- vides a moving account of the recovery of her children from the wordless, noncommunicative world of autism through the use of intensive intervention, based on the pioneering work of Lovaas and provided by the author and some very talented grad- uate students. This programming is usually delivered in the home and requires a massive time commitment from families (Choutka, 1999). With this type of instruction, the outcomes of seemingly small increments of learning are quantified and measured to ensure progress in many preacademic skill areas (Anderson & Romanczyk, 1999).

In contrast to the one-to-one instruction characterizing ABA, TEACCH is a classroom-based model developed at the University of North Carolina at Chapel Hill in the 1970s. It stresses structured teaching, independent work time, and voca- tional skills. TEACCH also offers parent training to foster im- provement of parent skills in dealing with problem behaviors (Dawson & Osterling, 1997), highly predictable situations to promote appropriate behavior, and environmental structuring to facilitate independent functioning (Schopler & Reichler, 1971). The program principles include teaching skills and ac- cepting deficits in children and parents. Unlike DTT, TEACCH does not offer a research-based recovery-rate percentage (Schopler, Mesibov, & Baker, 1982) and thus is less attractive to parents as an instructional program for their children.

In light of the increased incidence of autism (National Research Council, 2001), the rising tide of litigation (Zirkel, 2001), and the dearth of conclusive research on instructional approach effectiveness (Gresham, Beebe-Frankenberger, & MacMillan, 1999), the present study offers a timely exami- nation of the factors associated with the outcomes of the published hearing/review and court decisions concerning in- structional approaches for students with autism. These case outcomes have significant implications for students, parents, and schools, yet there is even less research concerning such litigation than there is research on instructional approaches for students with autism.

The previous research concerning autism litigation is limited to three published sources. First, Heflin and Simpson

(1998) discussed four issues—instructional approach, support services, placement decisions, and service length—and cited 16 published cases as examples. Their recommendations in- cluded that districts provide evidence of the efficacy of the selected instructional approach, but they did not provide spe- cific, systematic case law support for their recommendations. Their case coverage was far from comprehensive, lacking var- ious pertinent published hearing/review officer decisions (e.g., Cobb County Sch. Sys., 1996) and court decisions (e.g., Union Sch. Dist. v. Smith, 1994).

Similarly, Gryzwacz and Lombardo (1999) did not an- alyze either the outcomes, in terms of which party won, or the factors associated with winning, but provided an overview of the debate concerning educational approaches appropriate for serving children with autism. They included a brief discus- sion of a limited sampling of approximately 10 pertinent cases. Their overall conclusion was that the courts generally defer to the “educational methodology” (Gryzwacz & Lomardo, 1999, p. 5), or instructional approach, selected by the school district unless it is blatantly inappropriate. Although Gryz- wacz and Lombardo provided the full text of 14 published hearing/review officer and court decisions concerning the use of Lovaas/ABA and other methodologies for children with ASD, they did not provide a systematic or complete analysis of these or other pertinent cases (e.g., T.H. v. Board of Educ. of Palatine Community Consol. Sch. Dist. 15, 1998; Dong v. Board of Educ. of Rochester Community Sch., 1998).

Coming closest to an empirical analysis of the pertinent case law,Yell and Drasgow (2000) analyzed 45 published cases tried between 1993 and 1998. They examined how the case law defined appropriate in terms of education for children with autism. However, as pointed out in Zirkel (2001), they omitted various relevant cases altogether (e.g., Capistrano Unified Sch. Dist., 1995; Dong v. Board of Educ. of Rochester Community Sch., 1998; Fairfax County Pub. Sch., 1995) and, in some in- stances (e.g., Burilovich v. Board of Educ. of Lincoln Consol. Sch. Dist., 1998; T.H. v. Board of Educ. of Palatine Community Consol. Sch. Dist., 1998), failed to provide the superceding pub- lished decision within the limited time period of their study. Moreover, Yell and Drasgow used a simplistic, dichotomous categorization of outcomes without defining the meaning of the terms won and lost. More specifically, it is not clear how they categorized published decisions that were either incon- clusive, such as when a court denied the motion for dismissal and thus preserved the matter for trial, or mixed, such as when the hearing officer or judge decided one issue in the parents’ favor and another in the favor of the district. Further, they found parents to be the “prevailing party” (p. 208) in 76% of the cases without recognizing the specific legal meaning of that term (e.g., G. v. Fort Bragg Dependent Sch., 2001). Review Summary Assignment Paper

This study expands the scope of the previous studies via an empirical analysis of a comprehensive sample of pertinent hearing/review officer and court decisions. More specifically, the sample for this study consists of all the ABA/DTT/Lovaas cases published in the Education for Handicapped Law Re-

port (EHLR) and Individuals with Disabilities Education Law Report (IDELR) through the end of Volume 34, divided into two groups—program selection and program implementation. The purposes of the study are (a) to determine the overall outcome of the two subsamples, both those cases focused on program selection and those cases focused on program im- plementation, and (b) to identify the outcome-related factors in both groups of cases.

Method

As listed in the Appendix, we identified 68 hearing/review of- ficer and court decisions that (a) were published in IDELR and EHLR through the end of Volume 34, which was in ap- proximately August 2001; (b) identified the child as having any of the autism spectrum disorders, including PDD, autis- tic disorder, Rett syndrome, and Asperger syndrome (Mauk, Reber, & Batshaw, 1997); and (c) referenced ABA, DTT, or Lovaas instructional approaches. Regarding the third criterion, TEACCH was not necessary as a separate selection factor, be- cause the only case in which TEACCH was at issue in the ab- sence of ABA DTT, or Lovaas limited its focus to teacher training (Sioux Falls School District v. Koupal, 1994). For the final criterion, in the rare instance where the reference to ABA, DTT, or Lovaas was indirect, we included the case where the connection was reasonably understood. For example, we in- cluded Calaveras Unified Sch. Dist. (1994), which mentioned the UCLA clinic, even though it did not mention Lovaas per se. Finally, as in the Zirkel (2001) study, the analysis included only the highest published decision in each case, thus avoid- ing double counting; however, for the sake of comprehensive clarity, the Appendix also lists, in brackets, any published prior decisions for the same case.

First, we classified the cases into two broad categories— program selection and program implementation. The cases concerning program selection were those in which the parents sought an instructional approach (e.g., Lovaas) other than that proposed by the district (e.g., TEACCH). In contrast, the cases concerning program implementation were those in which the parties agreed on the instructional approach of ABA/DTT/ Lovaas but the parent contested the location (e.g., home vs. school setting), duration (e.g., number of hours of instruction), or provider (e.g., particular individual or specific qualifications).

Second, we calculated the overall outcome of each case. As adapted from Lupini and Zirkel (2003), and as specified in Table 1, we used a nondirectional scale of 1 to 7 (1 = complete win for the parents, 7 = complete win for the school authori- ties; Lupini & Zirkel, 2003). This Likert-type scale reflects the multiple issues and varying dispositions, for example, the denial of a motion for summary judgment, thus improving the validity of previous research (e.g., Yell & Drasgow, 2000).

Third, through reviewing the published opinion of each case, in particular the Discussion or Legal Conclusions sec- tions, we were able to glean factors that were expressly related

to the decision, or outcome. For example, we grouped these factors into the two prongs that the Supreme Court used to de- fine appropriateness in the landmark decision of Board of Ed- ucation of the Hendrick Hudson Central School District v. Rowley (1982; hereafter called Rowley). More specifically, the Court required the IEP to (a) be developed in accordance with the procedural dictates of the IDEA and (b) be calculated to yield educational benefit.

As outlined in Table 2, the identified factors fit into two categories, each with two subcategories. The first category is compliance with IDEA requirements, which consists of (a) IEP elements (e.g., present educational level, measurable annual goals) and (b) other procedural requirements (e.g., timelines, notices). The second category is evidence of educational ben- efit, which consists of the two leading sources of evidence of substantive compliance: (a) documentation of educational prog- ress (e.g., progress charts, data sheets) and (b) effectiveness of witnesses. Some of the cases contained, and thus were coded for, more than one outcome-related factor. Review Summary Assignment Paper

For a factor to be recognized in Table 2, we established a minimum frequency criterion of 20%; specifically, the fac- tor must have appeared across groups or categories in at least 14 of the 68 cases. Based on this criterion, two other factors were eliminated in light of their relative infrequency. First, contention between parents and districts regarding the exper- tise of program implementers appeared in only seven cases. Second, “deference,” which refers to a court giving the benefit of the doubt to a lower level of decision-making (Newcomer & Zirkel, 1999), appeared in only eight cases, likely because only one third (23) of the 68 cases were court decisions, as compared to hearing/review officer decisions.

For the purpose of interrater reliability, the first two au- thors independently coded a random sample of 23 of the 68 cases. After receiving training from the third author, they ob- tained an agreement level of 97% for case category (i.e., pro- gram selection or program implementation), 94% for outcome (i.e., 1–7 scale), and 94% for factor identification (e.g., tes- timony of witnesses). Finally, the first two authors reread, reviewed, and discussed the disputed cases, to reach 100% agreement on all areas of coding.

Results

Of the 68 cases in the sample, 43 (63%) focused on program selection, that is, where the parent sought an instructional method other than the one proposed by the district. The aver- age outcome of these cases, based on the outcome scale of 1 (parents) to 7 (district), with a 4.0 as the neutral midpoint, was 3.9. Contributing to the slight skew in the parents’ favor, 20 of the decisions at the polar positions of predominantly or completely conclusive decisions were in favor of parents (i.e., outcome codes 1 or 2), whereas 18 of the conclusive decisions were in favor of districts (i.e., outcome codes of 6 or 7). The remaining five cases had outcomes of 3, 4, or 5 (i.e., incon-

THE JOURNAL OF SPECIAL EDUCATION VOL. 38/NO. 2/2004 97

98 THE JOURNAL OF SPECIAL EDUCATION VOL. 38/NO. 2/2004

clusive or split decisions), meaning they were not decidedly in favor of either the parent or the district (i.e., Adams v. State of Oregon, 1999; Asbury v. Missouri Dep’t of Elementary and Secondary Educ., 1999; CM v. Board of Educ. of Henderson County [consolidated case with M.E.], 2001; De Mora v. De- partment of Pub. Welfare, 2001; and Malkentzos v. DeBuono, 1996), and were thus not reported in the results.

For cases regarding program selection, Table 3 summa- rizes the outcome-related factors relative to the conclusive outcomes in favor of parents and districts. The most frequent outcome-related factors were, in order, testimony of witnesses (n = 30), documentation of educational benefit (n = 28), and IEP elements (n = 25). For the parent-won cases of program

selection, the most frequent factors were documentation of educational benefit (n = 20), testimony of witnesses (n = 16), and IEP elements (n = 14). Similarly, for the district-won cases of program selection, the most frequent factors were testimony of witnesses (n = 14), IEP elements (n = 11), and documenta- tion of progress as substantive evidence of the educational benefit of the district’s program (n = 8). Review Summary Assignment Paper

The remaining 25 (37%) cases focused on program im- plementation; that is, the parents did not dispute the instruc- tional approach but contested the proposed location, duration, or provider. The average outcome of these cases, based on the previously mentioned 7-point scale, was 4.0. Establishing an overall tie position, 13 were at the polar positions of pre-

TABLE 1. Outcome Code Descriptions

Outcome code Description

1—Parent complete win This category consists of summary judgments in favor of the parent (i.e., decisions without a trial), as well as other conclusive wins on all major issues of the case in favor of the par- ent, including summary judgments.

2—Decision largely, but not completely, This category represents conclusive decisions in the parent’s favor for the majority of the for the parent issues or the awarding of relief (e.g., compensatory education, tuition reimbursement) of

more than 50% and less than 100% of what the parent originally sought. In the rare instance when these two criteria are conflicted, relief criteria are the controlling factor. Further, in review officer and court decisions where the published opinion does not specify the amount of relief sought by the parent, the frame of reference was the amount of relief awarded by the preceding level.

3—Inconclusive decision favoring parent This category includes the granting of a preliminary injunction (an interim decision after a short proceeding as well as the reversal of a dismissal of a case by a lower court), which means that the case will return to the lower court for a trial. Additionally, this category in- cludes the denial of a summary judgment motion sought by school authorities (because this preliminary ruling will result in a trial to determine the ultimate, conclusive winner).

4—Split decision This category includes the awarding of relief (e.g., compensatory education, tuition reim- bursement) of approximately 50% of that originally sought by the parent. Further, in situa- tions where the original amount of relief sought is unknown, this category includes the awarding of relief approximating 50% of that originally awarded by a lower court to the parent. In addition, this category includes cases in which petitions by both parties for re- hearing are denied, as well as the denial of cross motions for summary judgment (because the effect in such situations does not favor either party).

5—Inconclusive win for the school This category includes the denial of a preliminary injunction or summary judgment sought authorities by the parent (in that the parent still has the opportunity for a trial). In addition, it includes

cases dismissed for failure to exhaust administrative remedies (i.e., cases where the parent did not resort first to a due process hearing) and cases dismissed without prejudice (be- cause, after correcting the specified technical defects, the parents may still have their day in court). Review Summary Assignment Paper

6—Decision largely, but not completely, This category includes the awarding of relief (e.g., compensatory education, tuition reim for school authorities bursement) of clearly less than 50% of that originally sought by the parent. Further, in

situations where the original relief sought is not known, this category includes the awarding of relief approximating 50% of that originally awarded by a lower court to the parent.

7—Complete win for school authorities This category includes granting of a summary judgment in favor of school authorities (be- cause in both cases, the school authorities have won decisively at this preliminary step, end- ing the proceedings against them).

dominant or completely conclusive decisions in favor of par- ents and 12 were conclusively in favor of districts.

In addition, Table 3 summarizes the factors related to the conclusive outcomes in favor of parents and districts in cases regarding program implementation. The most frequent outcome-related factors were, in order, testimony of witnesses (n = 24), documentation of progress (n = 16), and procedural compliance of the IEP with the requirements of IDEA (n = 16). For the parent-won cases of program implementation,

the most frequent factors were, in order, testimony of parent- retained expert witnesses (n = 13), procedural compliance of the IEP with the requirements of IDEA (n = 8), and docu- mentation of progress as evidence of the substantive standard of educational benefit (n = 7). For the district-won cases of program implementation, the most frequently noted factors were testimony of district witnesses (n = 11), documentation of progress (n = 9), and procedural compliance of the IEP with the requirements of IDEA (n = 8). Review Summary Assignment Paper