Delaware County Intermediate Unit v. MARTIN K., 831 F. Supp. 1206 (E.D. Pa. 1993)
September 15, 1993
DELAWARE COUNTY INTERMEDIATE UNIT # 25, Plaintiff,
MARTIN & Melinda K., individually and as Parents & Natural Guardians of Paul K., and Donald M. Carroll, Jr., Secretary of Education for the Commonwealth of Pennsylvania, Defendants.
United States District Court, E.D. Pennsylvania.
*1207 *1208 *1209 *1210 Joseph J. Mittleman, Asst. Dist. Atty., Michael F.X. Coll, Media, PA, for plaintiff.
Janet F. Stotland, Philadelphia, PA, E. John Wherry, Martin A. Kotler, Widener Univ. School of Law, Wilmington, DE, for defendants.
Claudia M. Tesoro, Office of Atty. Gen., Philadelphia, PA, for Donald M. Carroll, Jr.
ROBRENO, District Judge.
This matter involves a child's right to a free appropriate public education under the Individuals with Disabilities Education Act (the "IDEA"), 20 U.S.C. § 1400 et seq. After a bench trial, and pursuant to Fed. R.Civ.P. 52, the Court makes the following findings of fact and conclusions of law.
I. FINDINGS OF FACT
Paul K., born on July 24, 1988, is the minor son of defendants Martin and Melinda K. When Paul was two, he was diagnosed as suffering from "pervasive development disorder not otherwise specified." Pervasive development disorder ("PDD") is "characterized by qualitative impairment in the development of reciprocal social interaction, in the development of verbal and nonverbal communication skills, and in imaginative activity." American Psychiatric Ass'n, The Diagnostic and Statistical Manual of Mental Disorders 33 (3d ed. rev.1987). The most severe form of PDD is autism; all less severe forms are deemed "not otherwise specified." Id.
Paul was enrolled at a "special needs" center at the Ken-Crest facility ("Ken-Crest"), a private school, in August of 1990. Nine of Paul's fifteen and one-half hours per *1211 week at Ken-Crest were funded by the Delaware County Mental Health and Retardation Agency. During the summer of 1991, Paul's parents came to believe that the Ken-Crest program was not adequately meeting Paul's educational needs. Sometime that summer, Paul's parents began exploring the possibility of enrolling Paul in an educational program based upon the research of University of California at Los Angeles psychologist Dr. O. Ivar Lovaas (the "Lovaas program"). The Lovaas program is a method of educating pre-school age children with PDD that stresses behavioral modification techniques.
Paul turned three years old in July of 1991, at which point Paul's education became the responsibility of plaintiff Delaware County Intermediate Unit # 25 (the "IU"). See 20 U.S.C. § 1412(2) (B); 22 Pa.Code § 14.51. The IU is under contract with the Pennsylvania Department of Education to provide "early intervention services," i.e., education to children over age three but not yet eligible for first grade, in Delaware County, Pennsylvania. Pursuant to the procedural requirements of the IDEA, a series of meetings took place between Paul's parents, representatives of the IU, and other interested parties in September and October of 1991. The purpose of these meetings was to develop an individualized education program ("IEP") for Paul. See 20 U.S.C. § 1414(a) (5); 22 Pa. Code § 14.54.
The IU initially proposed, on an informal basis, continuation of Paul's education in the Ken-Crest program, again with nine hours of public funding per week. This proposal was not acceptable to Paul's parents. On October 4, 1991, the IU issued a Multidisciplinary Team Evaluation ("MDE") Report. See 22 Pa.Code § 14.53. The MDE, conducted by a team of professionals in cooperation with Paul's parents, is a comprehensive evaluation of Paul's educational development and needs. In mid-October of 1991, Paul's parents withdrew Paul from the Ken-Crest program and undertook to educate Paul through the Lovaas program. Supervision of Paul's education was conducted by Jacqueline Wynn, a Ph.D. candidate from UCLA who was studying under Dr. Lovaas. Ms. Wynn came to Paul's parents' home to instruct Paul's mother, and a group of college students hired by Paul's parents, in the proper methods of teaching Paul through the Lovaas method. Paul's education under the Lovaas program consisted of forty hours per week of one-to-one behavior modification therapy conducted at the K.'s home, plus a "mainstream" component. The mainstream component consisted of two and one-half hours per week of interaction with non-disabled children at the Ken-Crest facility.
During the September-October, 1991 meetings, the IU had promised that an IEP for Paul would be forthcoming on or before November 1, 1991. On January 25, 1992, having failed to receive an IEP, Paul's parents requested a due process hearing. See 20 U.S.C. § 1415(b) (2); 22 Pa.Code § 14.64. The IU issued an IEP on January 31, 1992. The IEP was accompanied by two Notices of Recommended Assignment (the "NORA"). See 22 Pa.Code § 14.61(a) (3). The first NORA proposed to place Paul in the Ken-Crest program for fifteen hours per week, plus weekly speech, group language, and occupational therapy. Although it is not specifically stated in the NORA, the parties agree that this placement was to run only through mid- to late-February of 1992. Thereafter, Paul was to be placed, pursuant to the second NORA, in a new program, scheduled to be implemented by the IU during that mid-to late-February period, at the IU's Ridley School facility. The new program was to be based on the principles espoused by Division TEACCH, a program developed at the University of North Carolina for educating children with PDD. The TEACCH program, as opposed to Lovaas, stresses a cognitive rather than a behavioral approach. The proposed TEACCH program at Ridley consisted of ten hours of instruction per week (four days at two and one-half hours per day), plus a modified form of the therapy described above. Paul's parents rejected these proposed placements on February 5, 1992, in part because they felt that the TEACCH methods were similar to those used unsuccessfully in the Ken-Crest program, and again requested a due process hearing.
The due process hearing was held on March 10, 1992. On March 29th, the Hearing *1212 Officer ruled, inter alia, that the TEACCH program offered by the IU was appropriate, but that Paul's parents were entitled to some degree of compensatory education because the IU had implemented it too late. Both sides appealed to the Special Education Due Process Appeals Panel (the "Panel"). See 20 U.S.C. § 1415(c); 22 Pa. Code 14.64(m). On June 2, 1992, the Panel overturned the Hearing Officer's decision. The Panel ruled that the TEACCH program proposed by the IU in the January 31, 1992 IEP was not appropriate, that the Lovaas program was appropriate, and that Paul's parents were entitled to reimbursement for costs incurred in educating Paul pursuant to the Lovaas method. The Panel also issued prospective relief requiring the IU to develop an IEP that implemented the Lovaas program within thirty days of the Panel decision.
The IU initiated this action on July 2, 1992, pursuant to 20 U.S.C. § 1415(e) (2). In September of 1992, the parties, after a conference with the Court, agreed to submit their dispute to mediation. Sometime during that same month, the IU expanded its Ridley program from ten hours per week to fourteen hours per week. On October 16, 1992, the parties reported that mediation proved unsuccessful, and discovery commenced. In January of 1993, the Ridley program expanded from fourteen hours per week to twenty-three hours per week. The bench trial in this matter was held during the week of April 26, 1993.
A. Statutory Background
The IDEA's purpose is to "assure that all handicapped children have available to them ... a free appropriate public education..." 20 U.S.C. § 1400(c). Toward that end, the IDEA allows states to receive federal funding to educate disabled students provided their educational programs comply with the requirements of the IDEA. See id. § 1412. Among these requirements is the rule that participating states must have in effect "a policy that assures all children with disabilities the right to a free appropriate public education." Id. § 1412(1).
Under the IDEA, a "free appropriate public education" is one that is "provided in conformity with the individualized education program required under section 1414(a) (5) of this title." Id. § 1401(a) (18) (D). The IEP is the "modus operandi" of the IDEA, School Comm. of Burlington v. Department of Educ., 471 U.S. 359, 368, 105 S. Ct. 1996, 2001, 85 L. Ed. 2d 385 (1985), in that it is the "centerpiece of the statute's education delivery system for disabled children," Honig v. Doe, 484 U.S. 305, 311, 108 S. Ct. 592, 597, 98 L. Ed. 2d 686 (1988). An IEP sets forth an individualized educational plan for a particular disabled student. The IEP must include, inter alia, a statement of the services to be provided to the child, an assessment of the child's current educational levels, and the annual goals set for that child. See 20 U.S.C. § 1401(a) (20). The IEP is developed jointly by parents and school officials. See id.
The IDEA provides disabled children with rights that are both procedural and substantive. In Board of Education v. Rowley, the Supreme Court's seminal IDEA case, the Supreme Court held that in actions brought pursuant to the IDEA, "a court's inquiry ... is twofold. First, has the State complied with the procedures set forth in the Act? And second, is the individualized educational program developed through the Act's procedures reasonably calculated to enable the child to receive educational benefits?" 458 U.S. 176, 207-08, 102 S. Ct. 3034, 3051, 73 L. Ed. 2d 690 (1982). The Supreme Court found that the IDEA's procedural requirements were of primary concern to Congress. See id. at 206, 102 S. Ct. at 3050 (stating that "adequate compliance with the procedures prescribed would in most cases assure much if not all of what Congress *1213 wished in the way of substantive content in an IEP").
Still, though Rowley clearly focused on the procedural aspects of the IDEA, the Supreme Court did not "espous[e] an entirely toothless standard of substantive review. Rather, the Rowley Court described the level of benefit conferred by the Act as `meaningful.'" Polk v. Central Susquehanna Intermediate Unit 16, 853 F.2d 171, 179 (3d Cir. 1988) (quoting Rowley, 458 U.S. at 192, 102 S.Ct. at 3043), cert. denied, 488 U.S. 1030, 109 S. Ct. 838, 102 L. Ed. 2d 970 (1989). In Polk, the Third Circuit rejected the argument that any educational benefit obtained by a child could be considered "meaningful." Rather, the Third Circuit found that the legislative history of the IDEA, as well as Rowley itself, required educational benefits to be more than trivial or de minimis. See id. 853 F.2d at 181-83; see also Board of Educ. v. Diamond, 808 F.2d 987, 991 (3d Cir. 1986) (holding that benefit must be more than trivial).
The IDEA allows for both administrative and judicial review of the placement proposed by a school in an IEP. Participating states must choose either a one- or two-tiered system of administrative review. See 20 U.S.C. § 1415(b) (2), (c). Pennsylvania has opted for a two-tiered system. See 22 Pa. Code § 14.64. Pursuant to Pennsylvania's system, parents who object to an IEP may first request an "impartial due process hearing" before a hearing officer. Id. § 14.64(a). The party aggrieved by the hearing officer's decision may then appeal it to the Special Education Due Process Appeals Panel. See id. § 14.64(m).
Under the terms of the IDEA, any party aggrieved by a final administrative resolution has the right to initiate a civil action in either an appropriate state court or a federal district court. See 20 U.S.C. § 1415(e) (2). In such a judicial action, "the court shall receive the records of the administrative proceedings, shall hear additional evidence at the request of a party, and, basing its decision on the preponderance of the evidence, shall grant such relief as the court determines is appropriate." Id. In Rowley, the Supreme Court found that, despite this "preponderance of the evidence" standard and a court's ability to hear additional evidence, a court must not approach the matter as if it were writing on a clean slate. Specifically, the Court held that:
The very importance which Congress has attached to compliance with certain procedures in the preparation of an IEP would be frustrated if a court were permitted simply to set state decisions at nought. The fact that § 1415(e) requires that the reviewing court "receive the records of the [state] administrative proceedings" carries with it the implied requirement that due weight shall be given to these proceedings.
Rowley, 458 U.S. at 206, 102 S. Ct. at 3050 (emphasis added) (alteration in original).
In his recent dissent in Fuhrmann v. East Hanover Board of Education, Judge Hutchinson noted that the Third Circuit "has not yet definitively stated what constitutes `due weight'" under the Rowley standard. 993 F.2d 1031, 1042 (3d Cir.1993) (Hutchinson, J., dissenting). Judge Hutchinson suggested that this Circuit follow the standard established by the First Circuit in Burlington v. Department of Education, 736 F.2d 773, 792 (1st Cir.1984), aff'd, 471 U.S. 359, 105 S. Ct. 1996, 85 L. Ed. 2d 385 (1985). After noting that the First Circuit "concluded that the weight due the administrative decision is best left to the discretion of the district court," Fuhrmann, 993 F.2d at 1042 (Hutchinson, J., dissenting), Judge Hutchinson quoted the following language from Burlington:
The traditional test of findings being binding on the court if supported by substantial evidence, or even a preponderance of the evidence, does not apply. This does not mean, however, that the findings can be ignored. The court, in recognition of the *1214 expertise of the administrative agency, must consider the findings carefully and endeavor to respond to the hearing officer's resolution of each material issue. After such consideration, the court is free to accept or reject the findings in part or in whole.
Id. (Hutchinson, J., dissenting) (quoting Burlington, 736 F.2d at 792). Judge Hutchinson further defined this standard by saying "if the evidence fairly and rationally supports the agency's findings, and those findings are not cast into doubt by other evidence the agency did not have before it, the district court is justified in deferring to the state education authorities' expertise in deciding what educational program is appropriate for an individual child." Id. 993 F.2d at 1043 (Hutchinson, J., dissenting) (citing Wexler v. Westfield Bd. of Educ., 784 F.2d 176, 181 (3d Cir.), cert. denied, 479 U.S. 825, 107 S. Ct. 99, 93 L. Ed. 2d 49 (1986)). Absent further guidance from the Third Circuit (the Fuhrmann majority having made no comment on the dissent's proposed definition of "due weight"), this Court will apply the standard endorsed by Judge Hutchinson.
The burden of proof in this case is on the IU. This result stems from either, or both, of two premises. First, since main-streaming is at issue in this case, the Third Circuit's recent decision in Oberti ex rel. Oberti v. Board of Education, 995 F.2d 1204 (3d Cir.1993), places the burden of proof on the IU at least as to the mainstreaming issue. See id. at 1219 ("[W]hen the IDEA's mainstreaming requirement is specifically at issue, it is appropriate to place the burden of proving compliance with the IDEA on the school."). To the extent that this Court must consider issues other than mainstreaming, it could be argued that Oberti also requires placing the burden of proof on the school for all purposes. See id. at 1218-19. Since the IU does not appear to contest the conclusion that it retains the burden of proof, the Court will read Oberti as maintaining the burden of proof on the IU in this proceeding.
Finally, a word about the "additional evidence" aspect of § 1415(e) (2). Although the IDEA clearly permits supplementation of the administrative record in a judicial appeal, courts have held that the nature and extent of such supplementation is limited by Rowley's command that courts give due weight to the administrative proceedings. In Burlington, the First Circuit fashioned this rule:
A trial court must make an independent ruling based on the preponderance of the evidence, but the Act contemplates that the source of the evidence generally will be the administrative hearing record, with some supplementation at trial. The reasons for supplementation will vary; they might include gaps in the administrative transcript owing to mechanical failure, unavailability of witnesses, an improper exclusion of evidence by the administrative agency, and evidence concerning relevant events occurring subsequent to the administrative hearing. The starting point for determining what additional evidence should be received, however, is the record of the administrative proceeding.
The determination of what is "additional" evidence must be left to the discretion of the trial court which must be careful not to allow such evidence to change the character of the hearing from one of review to a trial de novo.
Burlington, 736 F.2d at 790-91; see Fuhrmann, 993 F.2d at 1040 n. 1 (Mansmann, J., concurring) (citing Burlington in discussion of the additional evidence requirement). Consistent with this standard, the Court reviewed the entire record below and also heard additional evidence at trial which was neither cumulative nor inconsistent with the evidence presented at the due process hearing.
The Court now turns to the substantive issues of this case. First, the Court must determine whether the IU has offered, in accordance with the procedures established *1215 by the IDEA, an appropriate education for Paul K. The Court divides the analysis of this question into three separate time periods, and concludes, for the reasons set forth below, as follows: 1) as to October of 1991, to January of 1992, the IU failed to offer any IEP whatsoever; 2) as to January of 1992, to June of 1992, the IEP offered by the IU was not appropriate; and 3) as to June of 1992, to July of 1993, during which time the IU enhanced its TEACCH program by increasing its hours, Paul's parents were entitled to rely on the Panel's decision in their favor, thereby mooting the question of whether the expanded TEACCH programs constitute an appropriate education under the IDEA. Alternatively, the IU's failure to annually review and revise Paul's IEP during this period precludes the IU from arguing that its expanded TEACCH programs, which were never offered to Paul in conformity with the IDEA, are appropriate. Second, the Court must fashion appropriate relief. On this question, the Court concludes that Paul's parents are entitled to reimbursement from the IU for all of their Lovaas-related expenses, and that a prospective change in Paul's educational placement would not be appropriate at this juncture.
B. Has the IU Offered an Appropriate Placement for Paul K?
1. October, 1991, to January 31, 1992
The validity of the IU's conduct during this period warrants little discussion. It is not disputed that the IU did not issue an IEP for Paul until January 31, 1992. The IU, however, was required to implement Paul's IEP at the beginning of the 1991-1992 school year. See 34 C.F.R. § 300.342(a). Further, Pennsylvania regulations require creation of the IEP within thirty days of the MDE report, which in this case was issued on October 4, 1991.See 22 Pa.Code § 14.54(j) (1). As the Panel noted, "even if the IU's January 1992 IEP and NORAs were found to be appropriate, the IU's failure to offer a free, appropriate public education prior to that time would entitle Paul's parents to reimbursement for expenses incurred up to that point." Panel Decision at 13 n. 9. The Hearing Officer also found that the IU's delays in formulating an IEP were impermissible. See Hearing Officer Decision at 8. These determinations are entitled to deference. Because it would have been unreasonable for Paul's education to have been placed on hold while the parents waited for the IU's IEP, the IU's failure to develop an IEP in a timely fashion is a procedural violation of the IDEA that caused prejudice to Paul. See generally Andersen ex rel. Andersen v. District of Columbia, 877 F.2d 1018, 1021 (D.C.Cir.1989) (discussing whether a showing of prejudice is required).
2. January 31, 1992, to June 2, 1992
The only IEP ever prepared for Paul, and the one that forms the basis of this litigation, *1216 was issued by the IU on January 31, 1992. The IU's IEP proposed, principally, a new TEACCH program of ten hours per week, to be implemented within a month of the IEP. Not surprisingly, events have occurred since that time that affect the circumstances of this case. In particular: 1) the IU actually implemented its ten hour per week TEACCH program, and subsequently expanded it to fourteen and then twenty-three hours per week, and 2) Paul has been receiving continuous Lovaas training, so that Paul is not, developmentally speaking, the same child examined by the IU when the IU prepared the IEP.
Throughout this litigation, the parties have disputed whether and to what extent the Court can consider actual progress made by students in either the original or expanded TEACCH classes, and/or actual progress made by Paul through the Lovaas program, in evaluating the appropriateness of the January, 1992 IEP. Any questions as to this issue were laid to rest, albeit after the trial in this matter, by the recent Third Circuit decision in Fuhrmann. In Fuhrmann, a child's parents rejected an IEP proposed by the school for the 1989-90 school year, and placed the child in an alternative program. In evaluating whether the school's proposed placement for that year was appropriate, the Court stated that evidence of the child's progress in the alternative placement was not relevant. The Court based this conclusion on the requirement in Rowley that an IEP "need only be `reasonably calculated to enable the child to receive educational benefits.'" Fuhrmann, 993 F.2d at 1039-40 (quoting Rowley, 458 U.S. at 206-07, 102 S.Ct. at 3050-51) (emphasis in Fuhrmann). In a concurring opinion that received expressed approval from the majority, see id. 993 F.2d at 1040, it was explained that:
Rowley's requirement that a school district's program be "reasonably calculated" to enable a child to receive educational benefits is prospective; it is based on an evaluation done by a team of experts prior to the student's placement. At the time of the child's evaluation, the IEP must be reasonably calculated to enable the child to receive educational benefits. Thus I would not view Rowley's test of "appropriateness" as whether the child actually receives educational benefit as a result of his school placement. Instead, the appropriateness of a student's placement must be assessed in terms of its appropriateness at the time it is created and not at some later date when one has the benefit of the child's actual experience.
Id. 993 F.2d at 1041 (Mansmann, J., concurring).
Since the IU's TEACCH program was not implemented until after the issuance of the IEP, Fuhrmann precludes this Court from considering the benefits actually received by children in the IU's TEACCH program in evaluating the validity of the IU's IEP. Instead, this Court must view the evidence from the perspective of a "snapshot" taken at the time of the preparation of the IEP. Id. (citing Roland M. v. Concord Sch. Comm., 910 F.2d 983, 992 (1st Cir.1990), cert. denied, 499 U.S. 912, 111 S. Ct. 1122, 113 L. Ed. 2d 230 (1991)). From that viewpoint, the Court cannot determine that the Panel erred in finding the IU's proposed IEP inappropriate.
*1217 Paul's parents claim the existence of many different deficiencies in the IU's originally offered TEACCH program. See Defs.' Conclusions of Law No. 47. To the Court, two are particularly significant. Either of these two deficiencies, considered without regard to the other, would render the program proposed in the IU's IEP inappropriate. First, the TEACCH program offered by the IU in January of 1992 provided for only ten hours per week of in-school instruction, despite the fact that TEACCH recommends thirty hours per week. At the due process hearing, and again at the trial before this Court, the IU's principal live witness was Dr. Gary Mesibov, co-director of Division TEACCH at the University of North Carolina, Chapel Hill. At trial, Dr. Mesibov acknowledged that the IU's TEACCH program was the only one in the country that he knew of that was only ten hours per week. See R.T. at 4-206; see also Defs.' Ex. 37 at 69. He also acknowledged that, at least to some extent, the benefit derived from a pre-school program was related to its "intensity," i.e., the number of hours offered. R.T. at 4-206. Indeed, evidence was presented at the due process hearing that a former colleague of Dr. Mesibov's, whom Dr. Mesibov admitted to be a "well respected researcher," felt that intensity was the most important factor in determining success. N.T. at 225-26;see also R.T. at 4-149 (expert testifying on behalf of Paul's parents testifies that "[i]n my opinion the critical issue is intensity of intervention"). Both at the hearing and at trial, Dr. Mesibov was tentative, at best, when evaluating whether a ten hour TEACCH program was sufficient to convey a meaningful educational benefit. At the due process hearing, Dr. Mesibov testified as follows:
Q: It's possible, isn't it, that ten hours a week is going to reach the point where simply no benefit is going to be returned?
A: It's possible.
Q: And that's certainly not inconsistent with your research?
Q: And you don't have any hard data which you can rely on saying a ten hour a week program is going to be sufficient?
A: We don't have hard data on how many hours is needed. That's correct.
N.T. at 226; see also id. at 225 (Dr. Mesibov states, "We recommend the full school day. That's what we think is best. I can't really say how much less you can have and make it work, though some of our school systems clearly do that with some of our children."). The following testimony, although somewhat disjointed, is also significant:
Q: If asked to counsel [Paul's parents] on whether they should put their child, newly discovered to be disabled, to put their child into a ten hour program that had never been tried before by a teacher who had never taught it before? Not being exactly certain of the importance of intensity on the success of the program, wouldn't you have been reluctant, sir, in good conscience to counsel them to put their child in that program, that there were other available.
A: Again, I'm I'm struggling with the question. Are you asking me based on Ron Faulkner's classroom? In that place
A: Would I counsel a parent to put their child in the program?
*1218 A: Or hypothetically
A: a classroom I didn't know, with a teacher I okay.
A: Hypothetically, I probably wouldn't. If it were
A: stated that way.
R.T. at 4-207 to -08. To be sure, other portions of Dr. Mesibov's testimony can be interpreted to indicate that he saw no significant problem with the IU's ten hour program. Taken as a whole, however, it appears to this Court that Dr. Mesibov's testimony concerning the ten hour program at the due process hearing at trial was ambiguous, at best. In evaluating Dr. Mesibov's testimony before the Hearing Officer, the Panel stated "[e]ven the expert brought from North Carolina to testify on behalf of the IU was most careful in his choice of words in what is best described as a qualified endorsement of the TEACCH-type program offered by the IU." Panel Decision at 6 n. 6 (citing N.T. at 208, 226-28). This Court agrees with the Panel's characterization of the testimony given by Dr. Mesibov at the due process hearing, and finds that it aptly describes Dr. Mesibov's testimony before this Court, as well. Juxtaposing Dr. Mesibov's equivocal endorsements with the testimony from Paul's parents' experts corroborating the position that ten hours was not appropriate, the Court finds that the IU's ten hour TEACCH program lacked sufficient intensity in terms of the number of hours offered.
Second, the Court finds that the IU's proposed IEP was deficient in that it lacked a *1219 satisfactory mainstreaming component. It is now axiomatic that, under the IDEA, public schools must mainstream disabled children to the extent possible. This requirement is codified in the IDEA at 20 U.S.C. § 1412(5) (B), which requires states receiving IDEA funds to establish "procedures to assure that, to the maximum extent appropriate, children with disabilities ... are educated with children who are not disabled." Id. In Oberti the Third Circuit adopted a two-part test for determining the appropriate level of mainstreaming: 1) "`whether education in the regular classroom, with the use of supplementary aids and services, can be achieved satisfactorily,'" and, if not, 2) "`whether the school has mainstreamed the child to the maximum extent appropriate,' i.e., whether the school has made efforts to include the child in school programs with nondisabled children whenever possible." Oberti, 995 F.2d at 1215 (quoting Daniel R.R. v. State Bd. of Educ., 874 F.2d 1036, 1048 (5th Cir. 1989)).
Neither party claims that Paul is ready for the regular classroom. The IU, moreover, does not challenge its statutory obligation to nonetheless mainstream Paul to the maximum extent appropriate. Indeed, the IU does not dispute that mainstreaming plays a significant role in the education of children with PDD Dr. Mesibov himself agreed that interaction between disabled and non-disabled students was "extremely important" within the TEACCH philosophy. N.T. at 216. Notwithstanding clear legislative directive regarding mainstreaming and its applicability to children suffering from PDD, the TEACCH program proposed by the IU contained no legitimate mainstreaming component. First, in a technical sense, the form on which the IU's IEPs are printed includes a section in which the IU is to identify the "Extent of Participation in Regular Education." On Paul's IEP, this section was left completely blank. See IU Ex. 15. The IU's failure to describe the mainstreaming component of Paul's education, or to set forth the reason for its absence, is a blatant violation of Pennsylvania law, which states: "Each ... eligible young child's IEP must contain the recommended service option placement and the rationale for why it represents the least restrictive environment." 11 P.S. § 875-103(7) (emphasis added). More substantively, the mainstreaming component of the IU's initial TEACCH program, as contemplated by the IU at the time of the issuance of the IEP, was practically non-existent. At a meeting between Paul's father and representatives of the IU that took place approximately two weeks before the IU implemented its TEACCH program, Paul's father inquired about the IU's proposed mainstreaming component. An IU representative responded by saying, "We haven't gotten that far. We are basically setting up the class. I think we need to take one step at a time." Defs.' Ex. 37 at 66; see also N.T. at 65 (IU representative states that IU intends to "eventually" incorporate mainstreaming); id. at 127 (different IU representative states "we have not established mainstreaming yet ... We will look at the situation and we can see if it can be achieved."). To the extent any mainstreaming component was included in the IU's initial proposal, it cannot be considered anything other than trivial. When asked about "informal mainstreaming," the first teacher of the IU's TEACCH program said at the due process hearing, "Well, just what I mentioned; when they're all outside and they're all having recess together and they see each other in the hallways. You know, just informal." N.T. at 178. Dr. Mesibov, however, suggested that this type of interaction was not sufficient to satisfy TEACCH's mainstreaming requirements:
*1220 Q: And when you say have experiences with [non-disabled children] you're not talking about simply finding yourself on the playground with an autistic child or finding a normal child and an autistic child on the playground at the same time?
A: No. That would certainly be one example. But when we're talking about it, we're talking about programmed interaction and mutual activities and more prolonged and continuous interactions as well as that.
N.T. at 216. Indeed, when asked at the due process hearing by counsel for the IU to identify the "additional activities" that are included in a full day TEACCH program but were not included in the IU's ten hour program, Dr. Mesibov stated, "[W]ell, based on my observation I would say some of the mainstreaming activities that have been talked about are inefficient in terms of time. But if you have more time, you can do them. They're inefficient, because the kids don't just walk right in and then start playing. There is some dead time and some training time." Id. at 228-29.
To this proof, the Court must now apply the teachings of Rowley, i.e., while public authorities are not required to "maximize" educational benefits, and courts are to use "independent judgment" in making a decision "based on a preponderance of the evidence," see Rowley, 458 U.S. at 205-06, 102 S. Ct. at 3050, judges must give the state administrative determinations under the IDEA "due weight," and must avoid substituting their own educational judgments for those of state administrative bodies. This is perhaps most especially true when the state administrative body rules against one of its own state's schools. See Burlington, 736 F.2d at 792. Here, the Panel ruled that the "watered down" TEACCH program was inadequate. Panel Decision at 6. In all candor, it appears to the Court that the question of whether the IEP now at issue was sufficient under the fairly lenient standard for educational appropriateness is a close one. Given that the IU has the burden of proof, however, and in light of the "due weight" standard, the Court finds that the IEP issued by the IU on January 31, 1992, did not offer an appropriate educational program.
*1221 3. June 2, 1992 to July 2, 1993
a. Paul's Parents' Reliance on the Panel Decision
As noted above, the Panel ruled, on June 2, 1992, that the TEACCH program proposed in the IU's January 31, 1992 IEP would not have provided Paul with a free appropriate public education, and required the IU to issue a new IEP implementing the TEACCH program. In September of 1992, the IU expanded its TEACCH program to fourteen hours per week. In January of 1993, the IU again expanded its program, this time to twenty-three hours per week. The IU argues that at least one of these expanded hour programs cured any deficiency that may have rendered its initially offered 10 hour TEACCH program inappropriate. See, e.g., R.T. at 1-22 to -23.
The Court finds, however, that Paul's parents are entitled to reimbursement regardless of the validity of the expanded TEACCH programs. Specifically, the Court holds that when a state administrative body rules that parents may prospectively place their child in private training, the parents are entitled to rely on that ruling until such time as it is overturned, or, if no decision is made on the validity of the administrative ruling during the period of prospective relief established by the ruling, then until the expiration of that period.
This very issue was addressed by the First Circuit in Burlington. While the complete factual and procedural history of Burlington is exceedingly tortuous, the relevant aspects can be summarized as follows: In Burlington, parents of a disabled child refused to accept an IEP that a school had proposed for the 1979-1980 school year. Instead, in the summer and fall of 1979, the parents arranged for private placement for their child, and initiated administrative proceedings against the school. In January of 1980, the state administrative body determined that the school's IEP was, in fact, inappropriate, and ordered the school to pay for the expenses incurred by the parents in connection with their private placement for the entire 1979-1980 school year. The school appealed the administrative decision to the federal district court in February of 1980, and refused to reimburse the parents for any expenses incurred during the 1979-1980 school year.
The First Circuit held that the school was equitably estopped from challenging the parents' right to require the school to fund the private placement "for that portion of the school year subsequent to the [administrative] decision." Burlington, 736 F.2d at 801. Specifically, the court stated that "where the final state administrative decision rules a town's proposed IEP inappropriate and orders the town to fund placement, and the parents have complied with and implemented that decision, a town or local educational agency is estopped from obtaining reimbursement for the time period, usually one year, covered by the state agency decision and order."Id. at 800-01. The court accepted the argument that:
*1222 where a state agency orders a town to fund a private placement, after ruling against a town's IEP, parents will be placed in the difficult position of having to choose between the state directive to maintain the child in the private placement at the risk of ultimately using their own funds, or of moving the child to the town's placement which the state agency has determined to be inadequate. Choosing the latter option would contravene the express congressional policy of consistency and adequacy in a disabled child's education during an IEP contest; should the judicial decision uphold the State and parents, the child will have spent two and possibly three years in an inappropriate school. In sum, ... in order for the Act to provide appropriate education for disabled children from wealthy and poor families alike, parents must be entitled to rely on a state administrative decision in their child's favor, and not be put at risk of reimbursement by a judicial judgment which reverses that decision.
Id. at 800.
The operative facts in the case at hand are nearly identical to those at issue in Burlington. Here, the Panel ruled that the IU's IEP was inappropriate, and that Paul must receive Lovaas training at public expense for the 1992-1993 school year. As stated in Burlington, the law should not force Paul's parents to predict whether they will be able to financially afford Paul's private education in the event that the Panel decision is overturned by a court. Requiring parents to decide whether to take this financial risk would force parents to make the Hobbesian decision of whether to place their child in an educational setting that has already been found lacking by a state authority. It is true that Paul's parents assumed the initial financial risk that they would not be reimbursed when they decided to reject the IU's IEP and place Paul in Lovaas training. Burlington, 471 U.S. at 373-74, 105 S. Ct. at 2004. That does not mean, however, that they must endure this risk even after their initial decision is later validated by state authorities.
The Court finds that Paul's parents must be reimbursed for expenses incurred during all times that the Panel's decision was effective, regardless of whether the January 1992 IEP or either of the expanded TEACCH programs were actually appropriate. The IU apparently concedes that the IEP that should have been prepared pursuant to the Panel decision would have been effective for at least one year. Accordingly, Paul's parents may be reimbursed for educational expenses *1223 incurred for all periods between June 2, 1992 (the date of the Panel decision), and July 3, 1993 (the date of the expiration of the IEP that would have been prepared pursuant to the Panel decision, and therefore the date on which Paul's parents could no longer conclusively rely on the Panel decision for purposes of their financial expenditures).
b. The IU's Failure to Offer its Expanded TEACCH Programs
Alternatively, the Court determines that Paul's parents would be entitled to reimbursement for 1992-1993 even if the IU was not estopped from contesting Paul's parents' reliance on the Panel's decision. As noted above, the IU contends that its expansions of the TEACCH programs cured any deficiency in the original IEP. Paul's parents argue, however, that this Court may not consider whether these expanded programs are appropriate because neither was offered by the IU to Paul's parents in the form of a revised IEP. In response, the IU claims that it should be excused from the requirement of initiating IEP proceedings for the purpose of offering the expanded TEACCH programs. The IU appears to base its argument on two related theories. First, the IU notes that the Panel decided that any TEACCH program, no matter how time-in-tensive, would not be appropriate for Paul, at least for such time as the Panel Decision was effective. See Panel Decision at 6, 8-9. Therefore, reasons the IU, offering any enhanced TEACCH program during that time was precluded by operation of the Panel decision. Second, the IU argues that Paul's parents, since at least as early as the late summer or early fall of 1991, were adamant in insisting that only Lovaas was appropriate for their child, so that Paul's parents would have therefore rejected any IEP presented by the IU other than one incorporating Lovaas.
As an initial matter, the Court notes that schools are required by the IDEA to review IEP's at least annually. See, e.g., 20 U.S.C. §§ 1414(a) (5), 1413(a) (11); 34 C.F.R. §§ 300.342(a), 300.343(d). The Pennsylvania regulations go even further, requiring reevaluation of preschool children by a multidisciplinary evaluation team every year, see 22 Pa.Code § 14.53(j), and review of a preschooler's IEP by the "IEP team" within six months of every reevaluation, see id. § 14.54(j) (3). Nothing in the IDEA expressly exempts schools from this requirement during adversarial proceedings. To the *1224 contrary, it appears to the Court that the constantly changing educational development of children with disabilities compels periodic review of IEPs even during the pendency of litigation. The First Circuit, in Burlington, so held, stating:
[P]ending review of an earlier IEP, local educational agencies should continue to review and revise IEPs in accordance with applicable law, at least in the absence of a stipulation between the parties providing for how the outcome of the suit will affect later years. Without the IEP as a starting point, the court is faced with a mere hypothesis of what the Town would have proposed and effectuated during the subsequent years, an hypothesis which at the time of trial would have the unfair benefit of hindsight. Such a rear view proposal could not comply with the statutory requirements for drafting IEPs, could not have been submitted in a timely fashion to the parents, and cannot constitute evidence at trial.
Burlington, 736 F.2d at 794;see also McMullen v. McKenzie, No. 87-0055, 1988 WL 60356, at *1-*2 (D.D.C. June 2, 1988), aff'd sub nom. Andersen ex rel. Andersen, 877 F.2d 1018; cf. Edwards-White v. District of Columbia, 785 F. Supp. 1022, 1024 (D.D.C. 1992) (parent's request for a due process hearing for one school year constituted "reasonable notice" of the parents' dissatisfaction with the school's proposed placement, thereby requiring the school to issue an IEP for the following year despite pendency of proceedings).
The Court must reject both of the IU's contentions regarding its failure to comply with the annual review provisions of the IDEA. As to the IU's first argument, the Court simply cannot accept the IU's position that there was nothing it could have done to offer its enhanced TEACCH programs in compliance with the IDEA. In its post-trial memorandum, the IU claims that it was precluded from offering the TEACCH program because it was under the constraints of the Panel decision. Incredibly, the IU also argued, in a letter submitted to the Court after final argument, the completely disingenuous position that the appeal of the Panel decision to this Court operated as a stay of the effect of the Panel decision. See Letter from Pl.'s Counsel to Court (July 19, 1993). This latter argument is based on the theory that since an appeal to the Pennsylvania Commonwealth Court is permitted by the IDEA, see 20 U.S.C. § 1415(e), and such an appeal stays the Panel decision by operation of state law, see 42 Pa.Cons.Stat.Ann. § 5105(e); Pa. R.App.P. 1701(a), an appeal to the federal district court should have the same effect. If this argument is correct and the Panel's ruling was stayed by virtue of the appeal to the district court, how could the Panel's ruling possibly preclude the IU from offering the enhanced TEACCH programs? Certainly, acceptance of the IU's argument that the filing of this action stayed the effect of the Panel decision would require rejection of the position that the IU was legally barred from offering its enhanced TEACCH program.
Even if the Panel decision did, in fact, legally constrain the IU from offering a new IEP, the IU was free to move this Court for a stay of the operation of that decision in order to be permitted to offer Paul's parents its enhanced TEACCH programs. Indeed, if it is true that the IU would have been entitled to a stay of the Panel decision in a state court proceeding, this argument might well have been significant among the equitable considerations that would have gone into this *1225 Court's determination of whether a stay of the Panel decision would be warranted in this case. If Paul's parents contested the IU's motion for a stay and the motion was subsequently denied, it would be difficult for Paul's parents to subsequently argue that the enhanced TEACCH programs were not offered; obviously, if the motion for a stay were granted there would be no issue.
The IU's second argument, that it would have been futile for the IU to offer its expanded TEACCH programs because Paul's parents would not have accepted it, is also rejected. Futility is, indeed, a recognized concept in IDEA litigation. See generally Honig, 484 U.S. at 327, 108 S. Ct. at 606; Octavia P. ex rel. Lester H. v. Gilhool, 916 F.2d 865 (3d Cir.1990), cert. denied, 499 U.S. 923, 111 S. Ct. 1317, 113 L. Ed. 2d 250 (1991). The doctrine operates as a basis for relieving a party from its obligation to exhaust the administrative dispute resolution procedures that must generally precede a judicial action under the IDEA. The IU, however, has cited to no case in which "futility" has served as a basis for relieving a school from its statutory and regulatory obligations to review and revise IEPs according to applicable timetables. Even assuming that the analogy is apt, and that futility could relieve a school from its obligations, the Court cannot find that futility has been shown on the basis of the conjecture offered by the IU. Therefore, the Court cannot conclude that it is more likely than not that, if offered, Paul's parents would have rejected a fourteen or twenty-three hour TEACCH program.
For the foregoing reasons, the Court concludes that the IU has not, from the date Paul became eligible for special education services until the present, offered Paul a free appropriate education under the IDEA.
Under the IDEA, this Court is granted discretion to award "such relief as the court determines is appropriate." 20 U.S.C. § 1415(e) (2). Paul's parents ask the Court to Order the IU to reimburse them for amounts they have expended on Paul's Lovaas training. The seminal case on the issue of reimbursement under the IDEA is the Supreme Court's decision in Burlington. In Burlington, the Supreme Court noted that parents who question the appropriateness of a proposed IEP are forced to choose whether to:
go along with the IEP to the detriment of their child if it turns out to be inappropriate or pay for what they consider to be the appropriate placement. If they choose the latter course, ... it would be an empty victory to have a court tell them several years later that they were right but that these expenditures could not ... be reimbursed by the school officials.
Burlington, 471 U.S. at 370, 105 S. Ct. at 2002. Thus, the Supreme Court ruled that parents can obtain reimbursement if: 1) the placement proposed by a school in an IEP was not appropriate, and 2) the alternative placement into which the parents put their child was appropriate. See id. at 369-70, 105 S. Ct. at 2002.
For the reasons discussed above, Paul's parents have satisfied the first prong. As to the second prong, the IU claims that Lovaas does not provide an appropriate education because Paul's Lovaas trainers are not *1226 certified as teachers under Pennsylvania law. It is, in fact, true that none of the persons directly responsible for Paul's Lovaas education (a Ph.D. candidate at UCLA who is studying under Dr. Lovaas, Paul's mother, and a host of undergraduate psychology students) is a certified teacher. The Court, however, concludes that the absence of teacher certification is not a bar to reimbursement.
Although the Third Circuit has yet to address the question of whether parents must place their children in programs that meet state requirements in order to be reimbursed, other jurisdictions have reached varying results. To this Court, the Fourth Circuit's recent resolution of the issue is particularly persuasive. Briefly, in Carter ex rel. Carter v. Florence County School District Four, 950 F.2d 156 (4th Cir.1991), cert. granted, ___ U.S. ___, 113 S. Ct. 1249, 122 L. Ed. 2d 649 (1993), a disabled child's parents placed the child in a private school that had not been approved for education of disabled students by the state. The Carter panel rejected the school's contention that such approval was a prerequisite to reimbursement. The court found that the relevant federal authorities, 20 U.S.C. § 1413(a) (4) (B) and 34 C.F.R. §§ 300.400-300.403, only prohibited public schools from placing children into private settings that did not meet state educational requirements, without addressing whether parents had the right to make such placements. See Carter, 950 F.2d at 162. Relying on the Supreme Court's decisions in Burlington and Rowley, the Fourth Circuit concluded that the question of the validity of the parents' placement must be resolved simply by determining whether "the private school in which the child is enrolled succeeded in providing an appropriate education, i.e., an education that is reasonably calculated to enable the child to receive educational benefits." Id. 950 F.2d at 164. In the absence of guidance from the Third Circuit, this Court chooses to follow Carter and those other cases that have similarly held, see, e.g., Alamo Heights Indep. Sch. Dist. v. State Bd. of Educ., 790 F.2d 1153, 1161 (5th Cir.1986) (permitting reimbursement for summer program that "might not have been adequate under the [IDEA]," because it was better than "no summer program at all"); Shirk v. District of Columbia, 756 F. Supp. 31 (D.D.C.1991); Carrington v. Commissioner of Educ., 404 Mass. 290, 535 N.E.2d 212 (1989), and rejects those cases that reached contrary conclusions, see, e.g., Tucker v. Bay Shore Union Free Sch. Dist., 873 F.2d 563 (2d Cir.1989); Antkowiak ex rel. Antkowiak v. Ambach, 838 F.2d 635 (2d Cir.), cert. denied, 488 U.S. 850, 109 S. Ct. 133, 102 L. Ed. 2d 105 (1988); Taglianetti ex rel. Taglianetti v. Cronin, 143 Ill.App.3d 459, 97 Ill.Dec. 547, 493 N.E.2d 29 (1986). Further, this Court should give due weight to the Panel's determination that reimbursement is appropriate under these circumstances, see Panel Decision at 10-13, since the Panel is a state entity evaluating state educational standards.
The question, then, becomes whether Lovaas has provided Paul with a meaningful education. It is indisputably clear that it has. Paul's substantial educational gains over the past two years are well established in the record, and do not need to be set forth with great detail herein. Even the Hearing *1227 Officer acknowledged Paul's progress under Lovaas. See Hearing Officer Decision ¶ 15. Indeed, the IU itself has never seriously challenged the position that Paul has received meaningful educational benefits from his Lovaas training. See, e.g., Transcript of Closing Argument ("C.T.") at 12 (counsel for the IU states "I don't dispute the progress that the child has made.... Progress in the Lovaas program is not disputed by the IU."); R.T. at 1-26 (counsel for the IU states "I am not denying" that "the Lovaas Program conveys some benefit"). But see infra Section II(C) (2) (discussing the IU's argument that Paul's advancement in Lovaas was superficial). After reviewing the evidence, most particularly the videotapes of Paul submitted by Paul's parents, the Court must agree with the Panel that Paul's progress under Lovaas has, indeed, been "remarkable." Panel Decision at 9. Since the Court has determined that the IU failed to offer Paul's parents an appropriate education, and since Paul's parents' alternative placement was, in contrast, appropriate, the Court will award Paul's parents reimbursement for all amounts expended in connection with Paul's Lovaas training.
2. Prospective Relief
Finally, the Court must fashion prospective relief. The Court concludes that a change in placement at this point in Paul's educational development would not be appropriate, so that the IU must fund Paul's Lovaas education for the 1993-94 academic year (barring, of course, any changed circumstances during the year that might render Lovaas inappropriate).
Several courts, in the context of IDEA litigation, have recognized that changes in educational placements should be approached with restraint. In Visco ex rel. Visco v. School District, 684 F. Supp. 1310 (W.D.Pa. 1988), a school proposed placing two deaf sisters, aged eleven and thirteen, into a deaf education program that used a particular teaching methodology. The sisters had previously been receiving special education at a different school that used a different methodology. In upholding the sisters' mother's objection to the change, the Court stated that:
[a] change in schools for [the sisters] would ... mean a change in the educational method used to teach them. It would *1228 also have meant a change in the educational method at the time this complaint was filed. Once a student is making good progress in a particular program, a change must be viewed with a great degree of caution. In balancing the potential benefits against the potential harms, even if the benefits and the harms were to weigh equally, the child in question should not be moved from one program to another.
Id. at 1315. Particularly important to the court was the fact that only a relatively short time remained until that program would end. See id. at 1316. Similarly, in Holmes v. District of Columbia, 680 F. Supp. 40 (D.D.C. 1988), the court relied on the disruption inherent in changes in placement to bar a school's proposed change. The court stated that "[t]he appropriate place for this youngster is to permit him to finish the remaining seven months of his high school education in the environment that he has been accustomed to over the past three years. I conclude as a matter of law that it would be inappropriate to transfer this youngster at this time." Id. at 41-42. Later, the court stated that "[s]uch a transfer would have been risky and would have carried with it great risks of disruption." Id. at 43.
Also relevant for instant purposes is Block v. District of Columbia, 748 F. Supp. 891 (D.D.C.1990). In Block, a hearing officer determined that a school had failed to properly evaluate a disabled child and had failed to include certain necessary services in his IEP. The hearing officer, however, allowed the school to correct these deficiencies, and then held a second hearing. At the second hearing, convened in the middle of the school year, the hearing officer determined that, though the school had remedied the deficiencies, a mid-year change would have been too disruptive for the child. The district court upheld this determination. See id. at 891. The Block court stated "[o]bvious advantages inhere to any child who is permitted to learn in a stable environment. This advantage may have even more meaning to a handicapped child." Id. at 896 (quoting Burger v. Murray County Sch. Dist., 612 F. Supp. 434, 437 (N.D.Ga.1984)). Lastly, the Third Circuit has recognized that an appropriate placement is one that is designed, if possible, to avoid regression. See Diamond, 808 F.2d at 991 (stating the IDEA "requires a plan likely to produce progress, not regression or trivial educational advancement").
In light of the foregoing authority, and given the equities of this case, the Court finds that placing Paul in the IU's TEACCH program at this juncture would not be appropriate. There is much evidence in the record to support the conclusion that Paul will completely lose the ability to perform those skills which he has not yet "generalized" if he is now transferred from Lovaas to TEACCH. See, e.g., R.T. at 4-80 to -83; 4-112 to -113. This evidence is largely uncontradicted. In fact, at closing arguments, counsel for the IU appeared to acknowledge that a "substantial" loss would occur if Lovaas was now with-drawn. C.T. at 4-5. Also significant to the Court is the fact that Lovaas training is designed for pre-schoolers, and normally lasts three years. Paul's Lovaas training will therefore end one year from now. See id. at 32. Thus, the IU's prospective obligation to fund Paul's current Lovaas training will last only one more year. In the balance of relevant equitable considerations, Paul's completion of two-thirds of a three year program weighs in favor of continuing the Lovaas training. Further, the evidence submitted at trial suggests that the costs of providing TEACCH and Lovaas are surprisingly comparable. Compare R.T. at 4-271 with Direct Testimony of Martin K. at 20 ¶ 63. See generally Clevenger v. Oak Ridge Sch. Bd., 744 F.2d 514, 517 (6th Cir.1984) (cost considerations, though not relevant in determining whether a particular program is appropriate, can be considered when choosing between two appropriate programs). Additionally, the equities might be somewhat different if the IU had formally offered Paul's parents an appropriate educational program at any time during the course of dealings between the parties. Under the analysis set forth in Section II(B) (3) (b), supra, such an offer *1229 would have meant that Paul's progression to this point in Lovaas, i.e., to the point where regression precludes a change in programs, was caused by the parents' rejection of an appropriate program.
The IU argues that the fact that Paul will regress if transferred to the TEACCH program demonstrates the "superficiality" of the gains obtained through Lovaas. C.T. at 5. The Court cannot agree. The fact that skills can be lost before they are generalized in no way renders them superficial; to the contrary, such loss only supports the need for continued treatment so that Paul can reach the point of generalization. The IU also appears to argue that while regression might occur as to those skills that Lovaas deems important, that does not mean that Paul will "generally regress." C.T. at 56-57. The Court also rejects this argument those skills that Lovaas deems important strike the Court as perfectly legitimate and broad based. Any regression in Paul's "Lovaas skills" would, in fact, represent a "general regression" of a magnitude sufficient to preclude Paul's transfer. Lastly, the IU argues that a "transition regression" argument was raised by the parents in Fuhrmann, and that it received "short shrift" by the district court and the Third Circuit. C.T. at 57. Of course, the effect of possible regression after a change in programs is a factual inquiry that varies from case to case. In Fuhrmann, the Court noted that there was credible expert testimony to support the conclusion that the child at issue would suffer no academic regression, and that, in any event, the child would not suffer "anything other than `normal regression' associated with any change of program." Fuhrmann, 993 F.2d at 1039. Here, in contrast, the likelihood of regression is practically conceded; further, the evidence supports the conclusion that such regression will be well beyond "normal." Accordingly, Fuhrmann has no bearing on the regression issue in the instant case. For these reasons, the Court rules that a change in placement from the Lovaas program would not be appropriate for Paul at this time.
One final point of clarification is in order. Since I have held that the IU's enhanced TEACCH programs cannot be considered for purposes of the 1991-1992 school year, see supra Section II(B) (2), that 1992-1993 is governed by procedural considerations, see supra Section II(B) (3), and that a change in placement is not warranted at this time in light of Paul's two years with the Lovaas program, see supra this section, I have had no occasion to address whether the twenty-three hour TEACCH program currently in place is appropriate under the IDEA. The fact that this issue has gone unaddressed might be troubling to the IU, since it claims that it brought this action partly because of its concern over the Panel's determination that TEACCH, even if "optimally implemented," would not be appropriate. See Panel Decision at 6. The IU has suggested, at least impliedly, that it is worried that this determination, if allowed to stand, will force the IU to provide Lovaas training for all PDD students within its jurisdiction.
Any ruling by this Court on the validity of the twenty-three hour program would be pure dicta and advisory only. Since federal courts are without jurisdiction to issue advisory opinions, the Court declines the invitation to make such a ruling. The Court does note, however, that the Panel, as required under the IDEA, clearly limited the scope of its decision to Paul's particular case, so that the IU's fears are perhaps unfounded. See Panel Decision at 8 ("We stress that our finding does not compel the conclusion that the Lovaas program is superior to the TEACH program in all respects or for all children. Other children may not be good candidates for the Lovaas program, or may not be as receptive to it over time."). More *1230 fundamentally, the Panel's determination as to the appropriateness of an "optimal" TEACCH might not be completely supported by the record. The Panel found TEACCH inappropriate for Paul because it felt that TEACCH "even under the best of circumstances, contemplates the student's involvement in special education throughout his educational career." Panel Decision at 6-7. It later stated that TEACCH "envisions a lifetime of special education for Paul." Id. at 8. In contrast, the Panel found that "[Lovaas] alone holds the promise of self sufficiency and integration into the regular curriculum." Id. at 7. To this Court, it seems that there is evidence in the record that can be read to directly refute these conclusions. Specifically, TEACCH appears to stress eventual independence, not a lifetime of special education. See, e.g., R.T. at 4-164 to -166. Further, there is some testimony in the record to support the contention that TEACCH students can become mainstreamed. See, e.g., id. at 4-174.
Thus, the Court's ruling today does not preclude the possibility that a significant increase in the IU's TEACCH program's intensity, coupled with an effective mainstreaming program, would render the battle between TEACCH and Lovaas a contest between two teaching methodologies, either of which would be appropriate under the IDEA. At that point, the Court will yield to the educational agency, since "courts lack the specialized knowledge and experience necessary to resolve persistent and difficult questions of educational policy.... Therefore, once a court determines that the requirements of the Act have been met, questions of methodology are for resolution by the States." Rowley, 458 U.S. at 208, 102 S. Ct. at 3051.
III. CONCLUSIONS OF LAW
In light of the foregoing, the Court makes the following conclusions of law:
1. There was no appropriate educational program in place for Paul from October, 1991, to January 31, 1992.
2. The educational programs proposed in the IU's January 31, 1992 IEP were not reasonably calculated to provide an appropriate education for Paul.
3. The IU cannot contest Paul's parents' right to reimbursement for the period between June 2, 1992, and July 2, 1993, because: a) Paul's parents are entitled to rely on the Panel decision in making educational and financial decisions regarding Paul's education, or, alternatively, b) the IU failed to take appropriate steps to revise Paul's IEP to implement the IU's enhanced TEACCH programs.
4. Paul's parents are entitled to reimbursement for all past expenses incurred in connection with educating Paul pursuant to the Lovaas method, despite the fact that the Lovaas instructors are not state-certified teachers.
5. Because Paul will suffer significant regression if removed from the Lovaas program at this point, and because Lovaas training will continue for no more than one additional academic year, the IU's enhanced TEACCH program is not appropriate for Paul, and the IU must reimburse Paul's parents for expenses incurred in connection with educating Paul pursuant to the Lovaas method from the date of this decision through the beginning of the 1994-1995 academic year. This reimbursement shall include funding for training during the summer of 1994.
An appropriate Order will be entered.
AND NOW, TO WIT, this 15th day of September, 1993, IT IS ORDERED that judgment is hereby entered in favor of defendants and against plaintiff, in accordance with the Opinion filed contemporaneously herewith, and further ORDERS as follows:
A. Plaintiff shall immediately initiate proceedings to prepare an Individualized Education Program for Paul K. that is consistent with applicable law and regulations and with the attached Opinion. This IEP shall include provisions for reimbursement of Paul K.'s parents for expenses they incur related to Paul's future Lovaas training.
B. Reimbursement shall be determined as follows: Within five days of the date of the filing of this Order, defendants Martin K. and Melinda K. shall submit to the Court, for purposes of determining a reimbursement award, a statement of all expenses incurred in connection with Paul K.'s Lovaas training to date (the statement may either replace or supplement Defendants' Exhibit 42, as submitted at trial). If plaintiff fails to object to the amounts set forth in the petition within five days of its submission, the defendants' right to recover those amounts will be deemed admitted. The Court will determine a procedure for adjudicating objections raised by plaintiff, if any, upon their submission.
C. Attorney's fees pursuant to 20 U.S.C. § 1415(e) (4) (B) shall be determined as follows: Defendants are to file a fee petition within ten days of the date of this Order. If plaintiff fails to object to the petition within five days of its submission, the defendants' right to recover the amounts set forth therein will be deemed admitted. The Court will determine a procedure for adjudicating objections raised by plaintiff, if any, upon their submission.
AND IT IS SO ORDERED.
 The Secretary of Education for the Commonwealth of Pennsylvania, originally named as a defendant in this action, was dismissed without prejudice by stipulation of the parties (Document No. 26).
 "K" is the first letter of the last name of Paul, Martin, and Melinda. While their full last name was frequently used during the trial of this matter, the Court will adhere to formality and refer to Paul, Martin, and Melinda as they appear in the caption of the case.
 The IDEA was preceded by the Education of All Handicapped Children Act ("EAHCA") and the Education of the Handicapped Act ("EHA"). Cases interpreting either of these two Acts apply to the issues in the case at hand; this Court will generically refer to all of these cases as "IDEA" cases.
 Pennsylvania regulations also require schools to provide "meaningful" educational benefits. See 22 Pa.Code § 14.1(v). The IDEA establishes only a floor of educational benefits, above which states are free to go. See F.G. ex rel. B.G. v. Cranford Bd. of Educ., 702 F. Supp. 1140, 1148 (D.N.J.1988) (citing Burlington v. Department of Educ., 736 F.2d 773, 792 (1st Cir.1984), aff'd, 471 U.S. 359, 105 S. Ct. 1996, 85 L. Ed. 2d 385 (1985)), aff'd without op., 882 F.2d 510 (3d Cir. 1989). Accordingly, it would seem that Pennsylvania schools would be required to provide more than "some benefit" even without the Third Circuit's interpretation of Rowley.
 At trial, direct testimony of the parties' witnesses was presented through written, verified statements. The Court required the parties to exchange written statements, and then to submit written objections to the admission of direct testimony of the opposing party's witnesses, before trial. The Court ruled on the objections during the trial. Though "live" direct examination at trial was not permitted, the parties were afforded the opportunity of "live" cross and re-direct examination of all witnesses whose direct testimony had been offered by way of written statement. The parties did not object to this procedure. See TPS Technologies, Inc. v. Rodin Enters. Inc., 816 F. Supp. 345, 347 n. 1 (E.D.Pa.1993) (requiring direct testimony to be offered by way of written statements subject to "live" cross examination); see also Manual for Complex Litigation, Second § 22.51 (1985); Federal Judicial Ctr., Litigation Management Manual 45 (1992).
 The IU contends that the October 4, 1991 MDE report was only a draft. If this document was a draft, the record does not include a final version, so that it would appear that the IU has still engaged in serious procedural violations. See 22 Pa.Code § 14.53(g), (i) (3) (requiring the creation of an MDE report).
 It appears that Martin K. stipulated to an extension of the deadline for the IU's preparation of an IEP and NORA until November 1, 1991. See Direct Testimony of Martin K. at 14. Martin K. knew that this IEP would have proposed the nine hour Ken-Crest program that he felt was inappropriate, and he allowed such an extension partly "in order to give the IU time to set out the contents of the nine hour Ken-Crest program in more detail so that the nature of the program would be clear to the Hearing Officer." Id. The IU's failure to develop an IEP for all periods after the extended deadline cannot be excused. Paul's parents may also recover their Lovaas-related expenses incurred during October of 1991 (subject to the availability of reimbursement, discussed infra Section II(C) (1)) because the IU had no appropriate education in place for Paul at that time. As discussed in Section II(B) (2), infra, the IU's 10 hour TEACCH program was not appropriate under the IDEA, because, inter alia, it was not sufficient in terms of the number of hours of programming. Nothing in the record could support the conclusion that the Ken-Crest program informally offered to Paul's parents in September of 1991, at one hour per week shorter than the deficient TEACCH program, was sufficiently better than the TEACCH program so as to render it appropriate despite its shorter weekly hours.
 While Fuhrmann was decided after the trial of this case, neither party disputes its applicability.
 The question of whether the Court can consider Paul's progress in Lovaas is somewhat more difficult. The Panel ruled, in part, that TEACCH was inappropriate for Paul in light of the progress he made in Lovaas during the four months prior to the issuance of the January 31, 1992 IEP. See Panel Decision at 8-9. Arguably, the time that the Fuhrmann "snapshot" is to be taken, however, is at the time of the MDE, since the school relies upon that evaluation in preparing the IEP. See Fuhrmann, 993 F.2d at 1041 (Mansmann, J., concurring) ("At the time of the child's evaluation, the IEP must be reasonably calculated to enable the child to receive educational benefits.") (emphasis added). Under such a reading of Fuhrmann, the Panel may have been precluded from considering Paul's progress in Lovaas, since Paul was not enrolled in Lovaas until after the MDE. The Court need not resolve this issue, however, since it determines that the January 31, 1992 IEP was inappropriate regardless of whatever progress Paul had made in Lovaas before that time.
 To the extent the record contains evidence to support the conclusion that TEACCH makes no recommendation as to the minimum number of hours required, the Court rejects it as against the greater weight of contrary evidence.
 The following citation conventions are used in the Opinion: R.T. (volume page) trial transcript; N.T. due process hearing transcript; IU Ex. / Parents' Ex. exhibits presented at the due process hearing; Pl.'s Ex. / Defs.' Ex. exhibits presented at trial.
 Dr. Mesibov did not completely agree, saying "[t]aken in isolation I think the intensity is one consideration, certainly up to a minimal point. But I don't think it's true that more is necessarily better all the time." N.T. at 226.
 Ron Faulkner was the teacher of the IU's original 10 hour TEACCH program.
 The testimony referred to by the Panel was as follows:
THE HEARING OFFICER: They're deriving some education benefit as is anticipated in an appropriate program.
THE WITNESS [Dr. Mesibov]: Yes. I think they would continue to receive an educational benefit in that program.
BY MR. COLL [counsel for the IU]: And is the benefit, by your judgment, something meaningful?
A: I think its a very meaningful start.
Q: Would you characterize it in any way as a minimal kind of benefit for these children?
A: What do you mean?
Q: Is it just lip service to these goals rather than something
A: No. I think its a substantial program.
N.T. at 207-08. The first portion of the text on page 226 of the hearing transcript is quoted above in the text of this opinion. The remainder is as follows:
Q: Dr. Mesibov, you were asked the question: Is it possible that ten hours of programming per week would not be enough, meaning that would not be sufficient to convey an educational benefit. What's your opinion about ten hours a week, particularly the kind that the IU is providing.
A: It's a hard question to answer. As I say, I think my preference in recommendation would be for more, because you're squeezing a lot of things into that ten hours, which, I think having a little more time you could spread out better. So, my initial reaction is, I would probably want more than that. But then the other part of it, it really depends partly on and what we try to do is consider the school in the context of the family and the community.
It partly depends on what's happening where the child is when he's not in school, the level of involvement the family is providing. So, with a child with autism the school situation is really a small party of the programming, not a large part.
The reason why I stumble with this question and have trouble is, at least in our program, the school program is also there's a clinic component which involves parent training. So one can get a better feel of what the total picture is.
So, if the child had ten hours of school and five good hours with the parents, that might be enough in certain situations, and then if you had a good child care situation.
So, I struggle with the question as both of you ask it. I just it's only ten hours of his week, but he's still doing something else. And if that's productive and useful, yeah, then the ten hours might be enough. If not, then maybe you would want more.
Id. at 226.
 The expert testifying on behalf of the IU at the due process hearing stated as follows:
Q: In your opinion would Paul respond well in general to a Division TEACCH type of program?
A: ... I think it's an open question whether Paul would respond well to a Division TEACCH model that's implemented according to the recommendations of Division TEACCH. If Paul was given thirty hours a week of Division TEACCH, as it is recommended to be given, I think it's an open question what kind of progress he would make.
Q: ... And you don't see any way that he would succeed in a ten hour TEACCH program?
N.T. at 298-99.
 It is true that Dr. Mesibov disagreed with the Lovaas philosophy that complete mainstreaming of children into non-disabled first grade classes was always a desirable objective. See R.T. 4-173, -218. At no time, however, did Dr. Mesibov deny that interaction with non-disabled students was an important aspect of the TEACCH program (any such denial would be of little relevance, in any event, since it would run counter to the clear statutory policy of the IDEA). The fact that all involved agree Paul should now be mainstreamed to the extent possible makes the mainstreaming issue in this case different from that of the typical mainstreaming case, where the parties dispute whether mainstreaming is appropriate at all. That no one contests the appropriateness of a mainstreaming component for Paul K. makes the IU even more culpable for its failure to incorporate a proper mainstreaming component into the original TEACCH program.
 The IU argues that this Court should defer to the Hearing Officer's determination that the ten hour TEACCH program was appropriate, and not to the Panel's determination that it was not. In support, the IU cites Doyle v. Arlington County School Board, 953 F.2d 100 (4th Cir.1991). It is true that Doyle, like the case at hand, involved a two-tiered system of administrative review in which no supplementation of the record occurred at the second tier. The IU claims that Doyle held that courts, when faced with conflicting administrative rulings, should defer to first-tier determinations since live testimony is presented only at that level. Contrary to the IU's assertions, however, Doyle did not create a per se rule to that effect. Rather, the Doyle court specifically limited its holding to the facts of that case. In Doyle, the second-tier hearing officer discredited a witness, credited by the first-tier officer, solely because of the title of a report written by the witness and the fact that the witness testified favorably vis-a-vis the parents. The Fourth Circuit held that due weight should be afforded to the first-tier determination because the second-tier hearing officer's reason for discrediting the expert deviated "far from the accepted norm of a fact-finding process designed to discover truth." Id. at 104; see also id. at 105 ("When a state administrative appeals authority has departed from the fact-finding norm to such an extent as here, we think the facts so found as a result of that departure are entitled to no weight....") (emphasis added).
No such drastic departure from the norm is present in the case at hand. While it is difficult to discern the precise reasons relied upon by the Hearing Officer with regard to the 10 hour program, it is nonetheless clear to this Court that the Hearing Officer did not place sufficient emphasis on the issues of intensity and mainstreaming. Given the importance of these concerns, it is not impermissible for this Court, in the exercise of its discretion, to give due weight to the Panel's treatment of these considerations. See Panel Decision at 9 n. 8 (Panel's discussion of the deficiencies of the IU's proposed mainstreaming component). Further, at least in the absence of egregious circumstances, it is appropriate for this Court to afford respect to the state's election of a two-tiered system by giving due weight to the higher of the two tiers.
 The "interim" IEP that would have placed Paul in the Ken-Crest program for approximately three weeks, at 15 hours per week, pending implementation of the TEACCH program, warrants little discussion. The Court has determined that the 10 hour TEACCH program was not appropriate. Even assuming, arguendo, that the 15 hour Ken-Crest program could have provided meaningful education, Paul's parents cannot be faulted for refusing to take Paul out of Lovaas in order to put him in a program for three weeks, after which no appropriate program would be in place. Accordingly, the Court concludes that the 15 hour Ken-Crest program, when offered for only three weeks to be followed by an inappropriate program, was not itself appropriate.
 The school eventually agreed to pay the parents prospectively, pendente lite, for costs incurred in connection with the 1980-81 school year and beyond, but at all times refused to reimburse them for amounts already expended in connection with 1979-1980. The district court initially granted the parents' request for retroactive reimbursement pendente lite, but that decision was overturned by the First Circuit in an earlier appeal on grounds that the parents had failed to show irreparable harm. See Burlington v. Department of Educ., 655 F.2d 428, 433-34 (1st Cir.1981).
 The Burlington Court's reference to the school being estopped from obtaining reimbursement is somewhat confusing. In Burlington, the school had agreed, pendente lite, to prospectively fund the child's education for the years 1980-81 and beyond. The school's suit was, in large part, an effort to obtain the return of funds given to the parents for that purpose. It is clear from the Court's recitation of the facts, however, that at no time, during the litigation or otherwise, did the school provide the parents with reimbursement for the parents' funds expended in connection with the 1979-80 school year. Accordingly, the school was not, vis-a-vis 1979-80, seeking reimbursement, but, rather, was challenging the parents' right to recover amounts they had already expended. Since the estoppel question related only to the issue of the parents' right to rely on the administrative decision, and since that decision governed only 1979-80, the First Circuit in fact ruled that the school was estopped from contesting the parents' right to reimbursement from the school for the period of time between the issuance of the administrative decision and its expiration. The Court made this holding clear later in the opinion, stating "where parents have paid the bill for the period [during which they relied on an administrative decision in their favor], they must be reimbursed." Burlington, 736 F.2d at 801.
 The Supreme Court ruling affirming the First Circuit decision in Burlington addressed only two of the many issues raised by the First Circuit. The Supreme Court ruled that: 1) reimbursement is available as an equitable remedy under the IDEA, and 2) notwithstanding the IDEA's "stay put" provision, 20 U.S.C. § 1415(e) (3), parents may be reimbursed if they place their child in private placement after rejecting a proposed IEP which is subsequently determined to be inappropriate. The Supreme Court expressly stated "we do not consider the estoppel ruling below," a reference to the "reliance" question addressed herein. Burlington, 471 U.S. at 374, 105 S. Ct. at 2004.
 The Panel did not give its ruling an expressed duration, i.e., the Panel made no determination as to how long Paul would be required to remain in Lovaas by virtue of the Panel decision. In arguing that it has been precluded from offering TEACCH by virtue of the Panel decision, see infra Section II(B) (3) (b), the IU apparently acknowledges that the Panel decision was effective for at least one year. Further, as noted above, the Panel ordered the IU to create an IEP for Paul implementing Lovaas within 30 days of the Panel's decision. See Panel Decision at 16. It seems that Pennsylvania regulations would require a new IEP for Paul no later than one year after the creation of the IEP prepared pursuant to the Panel's ruling. See 22 Pa.Code §§ 14.53(j), 14.54(j) (1). It is also true that Pennsylvania regulations require an "IEP team meeting" within six months of an evaluation, although there is no requirement that a new IEP be devised at this meeting. See id. § 14.54(j) (3). Even assuming that Paul's parents, under section 14.54(j) (3), could rely on the Panel decision for only six months, they would still be entitled to reimbursement for the remainder of the school year under the analysis set forth in Section II(B) (3) (b), infra.
 A possible argument to counter the Court's analysis, not raised by the IU, would be the claim that the IU was without authority to issue prospective injunctive relief beyond the school year at issue in the IEP that was being contested before it. It is at least arguable that Paul's parents should not be permitted to rely on a Panel award that is beyond the Panel's authority. As noted, however, the IU does not raise this argument, possibly because it runs counter to its claim, discussed in Section II(b) (3) (b), infra, that the IU has been precluded from offering the expanded TEACCH programs since the time of the issuance of the Panel's award. Moreover, no such impediment on the Panel's authority is immediately apparent. In any event, even if the Panel is precluded from issuing injunctive relief for 1992-1993, reimbursement to Paul's parents for that year would still be appropriate for the reasons discussed in Section II(b) (3) (b), infra.
 During the trial, the IU seemed to indicate at one point that it was Paul's parents' responsibility to initiate the preparation of a new IEP. This claim appears to have been abandoned (in that it was not briefed by the IU), and rightly so federal regulations expressly state that "[e]ach public agency shall initiate and conduct meetings to periodically review each child's [IEP] and if appropriate revise its provisions." 34 C.F.R. § 300.343(d) (emphasis added).
 Although the IU does not raise the point to this Court, it is true that 22 Pa.Code § 14.61(b) states that "No change in the identification, evaluation, educational placement or IEP of an ... eligible young child may be made during the pendency of an administrative or judicial proceeding unless agreed to by the parties to the proceeding." The fact that reevaluation under this section is permitted by consent of the parties would seem to allow the IU to at least offer its expanded TEACCH programs to Paul's parents. Significantly, moreover, this provision is much more prohibitive than its federal counterpart, the "stay put" provision of 20 U.S.C. § 1415(e) (3). The federal provision only precludes a change in "educational placement" during review proceedings, and does not expressly prohibit reevaluation (the IU does not raise the federal stay put provision in response to the claim that it retained a continuing obligation to reevaluate). While the Court need not rule on the issue since it was not raised by the IU, to the extent § 14.61(b) restricts a child's rights under the IDEA, i.e., allows schools to go below the "floor" established by the First Circuit's interpretation in Burlington by precluding reevaluation during the pendency of adversary proceedings, it would be invalid.
 In Burlington, the First Circuit held that the effect of the school's failure to review the initial IEP was that the parents could not automatically be assigned the burden of proof as to subsequent years. Burlington, 736 F.2d at 794-95. Instead, the Court held that, on remand, the losing party as to the initial year would have the burden of proving "changed circumstances that render the district court's determination as to the initial year inappropriate for guiding its order of relief for subsequent years." Id. at 795 (footnote omitted). The only "changed circumstances" at issue in the case at hand are the enhanced TEACCH programs that the IU failed to offer. While the enhancements may have been sufficient to render the TEACCH program appropriate, see infra Section II(C) (2), the fact that they were not offered to Paul in compliance with the IDEA results in a procedural violation that prejudiced Paul's parents to such an extent as to preclude the IU from relying upon them in this case.
 The cases cited by the IU in support of its argument for waiver of the exhaustion requirement due to futility are inapposite. In each of those cases, the plaintiff's claim did not significantly differ from the one raised during the administrative process. See, e.g., DeBlaay ex rel. DeVries v. Spillane, 853 F.2d 264, 267 (4th Cir. 1988) ("[R]eexhaustion [is] inconsistent with the statutory scheme when the complaint remains the same though the IEPs change."); Johnson v. Lancaster-Lebanon Intermediate Unit 13, 757 F. Supp. 606, 614 n. 6 (E.D.Pa.1991) (stating that exhaustion of IEP created during proceedings not required "[b]ecause of the importance of speedily resolving EHA cases and the similarity of the 1990-91 IEP to the 1989-90 IEP"). The case sub judice is in marked contrast. As noted above, the IU claims that the 14 and 23 hour TEACCH programs it failed to offer to Paul's parents differed from the 10 hour program that was before the Appeals Panel to such an extent as to cure any deficiencies in the original program. Accordingly, this is not a case where the issue not "exhausted" by the plaintiff had no effect on the nature of the dispute. To the contrary, it is the IU itself that asserts that its enhanced programs had a profound effect on its case.
 The Supreme Court granted certiorari in Carter during its most recent term. That term expired without the Court reaching a decision. Given the equities of the instant case, and the uncertainty over when a decision from the Supreme Court in Carter might be forthcoming, this Court concludes that waiting to resolve this matter until the Supreme Court decides Carter would not be in the best interests of justice.
 The IU claims that Carter, even if affirmed by the Supreme Court, can be distinguished on the ground that the private placement in Carter involved an unapproved private school, while in the case at hand Paul received his education primarily at his home. This is a distinction without a difference, however, since Pennsylvania regulations specifically permit home education of pre-school children, and, indeed, recognize that home education for preschoolers can often be most appropriate. See 22 Pa.Code § 14.55(c) (exempting home education from the hierarchy of program deemed to be "least restrictive" for preschoolers). Accordingly, Paul's Lovaas program can be deemed appropriate despite the fact that it is administered in Paul's home.
 See generally R.T. at 4-50 to -52; 4-64 to -68; 4-111 to -112; Direct Testimony of J. Wynn at 24 ¶ 67.
 In Section II(B) (3), supra, the Court discussed two alternative theories concerning Paul's 1992-1993 education. The issue of the appropriateness of Lovaas is not relevant to Paul's parents' right to reimbursement for that year under the first of these two theories, discussed in Section II(B) (3) (a), supra. This is so because, under this "estoppel" theory, Paul's parents were entitled to rely on the Panel decision allowing them to expend funds on Lovaas training. In contrast, under the alternative "failure to offer a revised IEP" theory discussed in Section II(B) (3) (b), supra, Paul's parents' right to reimbursement does depend on the appropriateness of Lovaas, since the IU's failure to offer an appropriate program has no bearing on whether Paul's alternative placement was appropriate.
Additionally, although the IU does not raise the point, it seems that it could be argued that Fuhrmann requires the Court to view the validity of the parents' alternative placement without resorting to hindsight, i.e., without consideration of how Paul actually progressed during his Lovaas training. See supra Section II(B) (2). Fuhrmann, however, based its requirement for prospective evaluation on the fact that Rowley held that the IDEA requires schools to develop IEPs that are "reasonably calculated" to provide educational benefit; Rowley says nothing about parents' responsibilities under the IDEA. Cf. Carter, 950 F.2d at 162-63 (holding that parent-placement in a non-state-approved private school did not preclude reimbursement). If Fuhrmann does apply to the parents' decision, whether Paul's progress during his four months of pre-IEP Lovaas training could be considered would become an issue, as discussed in note 9, supra. The Court need not resolve these issues, however, because it concludes that even if Paul's progress during Lovaas could not be considered at all, Lovaas remained an appropriate placement for Paul. Even when viewed without resort to hindsight, Paul's parents' decision to enroll Paul in Lovaas was appropriate since Lovaas had a proven success record, as verified by a 15-year study conducted by the Lovaas program. Lovaas, moreover, had been used as a method for educating children suffering from PDD in many other programs across the country. There is no doubt that Lovaas, when implemented according to the program's recommendations (as contrasted to the weakened TEACCH program offered in the IU's IEP), is designed to provide Paul with meaningful educational benefits.
 For purposes of this opinion, "prospective" relief includes the approximately two-month period between July 2, 1993, i.e., the end of the period discussed in Section II(B) (3), supra, and the date of this opinion.
 Roughly speaking, a "generalized" skill is one that can be performed in differing environments or under varying circumstances, as opposed to one that can only be performed in the structured setting in which it is learned. See generally R.T. 3-78 to -84.
 The Court will Order the IU to initiate proceedings to create an IEP for Paul implementing the Lovaas method. In its order on prospective relief, the Panel ruled that the IU had the option of reimbursing Paul's parents for their future Lovaas expenses, or of creating a Lovaas program of its own. Given the time that has elapsed since the Panel decision, and that Paul has only one year of Lovaas training remaining, it seems to this Court that the appropriate program for Paul would be a continuation of the training he receives from his current trainers, rather than a change to new trainers hired by the IU. Accordingly, the IU will be required to reimburse Paul's parents for their future Lovaas expenses.
 The only citation to the record supplied by the Panel in support of its position on the appropriateness of the "optimal" TEACCH program was a reference to N.T. at 198. See Panel Decision at 7. In this portion of the Due Process hearing, Dr. Mesibov states that "we see [TEACCH] as being more of a building block or cumulative program that's going to continue throughout the child's education system and life." Immediately after this phrase, however, Dr. Mesibov states "I think there's much more of an emphasis on independence training in our system," and then states that TEACCH stresses generalization to a greater extent than Lovaas. In light of these and other comments made by Dr. Mesibov regarding the goals of TEACCH, it seems that it might be unfair to characterize TEACCH as a program that anticipates nothing better than "a lifetime of special education for Paul."
 The issue of whether Paul is entitled to an "extended school year," i.e., publicly funded education over the summer, was disputed during the administrative proceedings. The Panel concluded that Paul was entitled to summer programming. See Panel Decision at 13-15. The IU did little, if anything, to address this issue before this Court. In light of the "due weight" standard of Rowley, and since the Lovaas programming is designed to be administered year-round, the Court rules that the IU must fund Paul's summer Lovaas programming.