LIVELY, Circuit Judge.
This petition for review of final action by the National Highway Traffic Safety Administration (NHTSA) of the Department of Transportation arises under the National Traffic and Motor Vehicle Safety Act of 1966, as amended (the Safety Act), 15 U.S.C. §§ 1381-1431 (1988). The ease concerns standards for the safe transportation on school buses of students in wheelchairs. The petitioners contend that the final rule relating to wheelchairs on school buses violates the Safety Act in two respects and, thus, constitutes arbitrary and capricious agency action.
First, the petitioners maintain that NHTSA acted unlawfully in promulgating standards for the seeurement of wheelchairs on school buses based only on “static” rather than “dynamic” testing. Static testing tests the strength of the individual components of a seeurement device. Dynamic testing tests the entire seeurement and restraint system using a test dummy under conditions emulating an actual crash. Second, they argue that the agency should have prescribed “crash-worthiness” standards for wheelchairs carried on school buses.
NHTSA announced the final rule on September 3, 1993, 58 Fed.Reg. 46873, after engaging in informal rulemaking procedures pursuant to section 553 of the Administrative Procedure Act (APA), 5 U.S.C. § 553 (1988).
The Safety Act authorizes judicial review under section 706 of the APA. See 15 U.S.C. § 1392(b); 1394(a)(1). The issues were well briefed and the court heard oral arguments before taking the matter under submission.
I.
The purpose of the Safety Act is to “reduce traffic accidents and deaths and injuries to persons resulting from traffic accidents.” 15 U.S.C. § 1381. To achieve this goal Congress required the Secretary of Transportation to establish appropriate safety standards for motor vehicles, which “shall be practicable, shall meet the need for motor vehicle safety, and shall be stated in objective terms.” 15 U.S.C. § 1392(a). The Safety Act, as originally adopted, did not refer specifically to school buses.
In 1974 Congress amended the Safety Act to require the promulgation of safety standards for school buses. Following this directive, NHTSA in 1976 promulgated Federal Motor Vehicle Safety Standard No. 222 (Standard 222) — School Bus Passenger Seating and Crash Protection, 49 C.F.R. § 571.222. Standard 222, however, established no safety standards for special seating used by handicapped passengers on school buses.
A.
On July 7, 1989, the petitioners filed a complaint against the Department of Transportation in the form of a letter, alleging that the Department’s exclusion of provisions for handicapped students from Standard 222 violated section 504 of the Rehabilitation Act of 1973, 29 U.S.C. § 794 (1988). Section 504 prohibits discrimination against handicapped [1002]*1002individuals under any program receiving federal financial assistance or any program or activity conducted by an Executive agency. At the time of this complaint, the Department of Transportation did not have a section 504 complaint procedure, so the Department referred the complaint to NHTSA. On February 14, 1990, NHTSA advised the petitioners that their complaint would be treated as a petition for rulemaking to amend Standard 222, and it published a notice of the granting of the petition on March 1, 1990. 55 Fed.Reg. 7346. The notice established rulemaking docket No. 90-05.
In the notice, NHTSA stated that a recent task force had reviewed the safety of school bus passengers confined to wheelchairs and, as a result, a “state-of-the-art” study of wheelchair securement and protection systems on school buses had been initiated in 1989 to support possible future rulemaking. Id. NHTSA also indicated in the notice that it had been examining efforts by various standards organizations and international agencies regarding safety standards for the transportation of persons in wheelchairs. Several of these standards were listed in the notice and later made available for the rule-making docket. Id.
In its second notice published May 30, 1990, NHTSA announced the completion of its state-of-the-art study and the study’s final report titled ‘Wheelchair and Occupant Restraint on School Buses.” 55 Fed.Reg. 21891. The significant conclusions of this report were that persons in wheelchairs should be transported in a forward-facing position, that the wheelchair and its occupant should be independently secured to the vehicle, that a combination torso and lap belt is one effective means of occupant restraint, and that a tiedown system attaching the frame of the wheelchair to the vehicle floor with four points is an effective securement system. The notice requested comments by July 30, 1990, on the report and other aspects of the pending rulemaking to establish requirements for school bus seating for handicapped students. Id.
B.
On September 24,1991, NHTSA published a Notice of Proposed Rulemaking (NPRM) proposing to amend Standard 222 to include certain requirements for wheelchair securement devices and occupant restraint systems on school buses. 56 Fed.Reg. 48140. The notice chscussed several aspects of the proposal including the agency’s tentative decisions: (1) to base the performance criteria for the required devices on static testing rather than on dynamic testing, and (2) not to establish crashworthiness standards for wheelchairs. Id. at 48141, 48144.
The agency acknowledged in the NPRM that dynamic testing is the “preferred” approach and that many parties active in the national and international efforts to establish standards for wheelchair securement and occupant restraint were working on standards based on dynamic tests. The agency concluded, however, that dynamic testing was not practicable because of numerous technical problems. NHTSA determined that these problems could not be resolved without unreasonable delay. Id. at 48141. In response to a comment to its May 30 notice, in which a commenter questioned the ability of some wheelchair designs to withstand crash forces, NHTSA rejected the suggestion that it establish standards for wheelchair crash-worthiness, reasoning that the development of objective tests would take “an extremely long time.” Id. at 48144. In the NPRM, NHTSA also requested comments on its proposed amendments to Standard 222 and comments on 17 specific questions. Id. at 48145-46. The deadline for comments was November 25, 1991. Id. at 48140.
In April 1992 the petitioners and their counsel met with a group of NHTSA officials in Washington to discuss the status of the pending rulemaking to amend Standard 222. The administrative record contains the minutes of this meeting.
NHTSA’s Final Regulatory Evaluation (FRE), completed in September of 1992, also recommended that the final rule be based on static testing standards since static testing was the most “expedient” means of resolving the safety problems related to wheelchairs used on school buses. Again NHTSA took the position that it would be “inappropriate” [1003]*1003to promulgate wheelchair crashworthiness standards due to the length of time required to resolve technical problems. NHTSA also noted that crashworthiness standards could undermine the concept of accessible transportation for the handicapped because such standards might increase the cost, weight and size of school bus-eligible wheelchairs, making them unavailable to some handicapped students. The record provided a basis for this concern.
C.
On January 15,1993, NHTSA published its final rule amending Standard 222. 58 Fed. Reg. 4586. In the rule, NHTSA formally rejected the position of commentators advocating dynamic testing, citing the desire to avoid extensive rulemaking delays due to technical deficiencies in dynamic testing as the reason. Id. at 4586-87. Additionally NHTSA again noted it was “inappropriate” to regulate wheelchair crashworthiness standards in light of extensive delays due to lack of existing standards on the topic. Id. at 4591. As an additional rationale, NHTSA claimed for the first time that it lacked authority to regulate the crashworthiness of wheelchairs because wheelchairs do not constitute “motor vehicle equipment” under the Safety Act. Id.
On February 1, 1993, Lyle Stephens submitted a petition for reconsideration of the final rule. Stephens argued there was sufficient information currently available to establish standards based on dynamic testing of a mobile seating device. Stephens also challenged NHTSA’s claim of lack of authority, stating that, although NHTSA may not have authority to regulate wheelchairs per se, they do have authority to regulate a mobile seating device. When a wheelchair is brought on a bus, he contended it becomes a mobile seat subject to regulation by NHTSA.
On September 3, 1993, NHTSA denied Stephens’s reconsideration petition and announced the final rule. 58 Fed.Reg. 46873. In the announcement, NHTSA reasoned that because wheelchairs do not constitute “motor vehicle equipment” or “accessories” under the agency’s internal definition, NHTSA lacked authority to regulate them. Id. at 46874. This was the first time NHTSA discussed the possibility of wheelchairs being “accessories” under the Safety Act, and it was the first time in the rulemaking process NHTSA published its internal definition of “accessory.” Following the denial of Stephens’s reconsideration petition, Simms and Stephens filed the instant petition for review of both the order issuing the final rule and the order denying the reconsideration petition. We consolidated this petition with an earlier one filed while Stephens’s petition for reconsideration was pending before NHTSA.
II.
The court’s function in reviewing final agency action following informal rule-making is prescribed by the APA. We review the administrative record, applying the standards set forth in section 706 of the APA, 5 U.S.C. § 706, and must set aside agency action that is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” Citizens To Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 414, 91 S.Ct. 814, 822, 28 L.Ed.2d 136 (1971) (Overton Park). Section 706 requires the reviewing court “to engage in a substantial inquiry.” Id. at 415, 91 S.Ct. at 823. While the agency head’s decision is entitled to a presumption of regularity, “that presumption is not to shield his action from a thorough, probing, in-depth review.” Id. In determining whether a decision was arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law the reviewing court must consider “whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment.” Id. at 416, 91 S.Ct. at 824. The ultimate standard of review, however, is a “narrow one. The court is not empowered to substitute its judgment for that of the agency.” Id.
The Supreme Court considered a rule promulgated by NHTSA under the Safety Act in Motor Vehicle Mfrs. Ass’n v. State Farm
[1004]*1004Mut. Auto. Ins. Co., 463 U.S. 29, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) (State Farm). The question in that case was whether NHTSA acted arbitrarily and capriciously in revoking an earlier requirement that automobiles be equipped with passive restraints (automatic seatbelts or air bags). Id. at 34, 103 S.Ct. at 2862. Before deciding the question the Court discussed the purpose of the Safety Act — to “reduce traffic accidents and deaths and injuries to persons resulting from traffic accidents.” Id. at 33, 103 S.Ct. at 2862 (quoting 15 U.S.C. § 1381). The Court then detailed the Safety Act’s mandate to the Secretary of Transportation “to issue motor vehicle safety standards that shall be practicable, shall meet the need for motor vehicle safety, and shall be stated in objective terms.” Id. (quoting 15 U.S.C. § 1392(a)). Further, the Court noted that the Safety Act directs the Secretary to consider “relevant available motor vehicle safety data” and to determine “whether the proposed standard is reasonable, practicable and appropriate for the particular type of motor vehicle, and the extent to which such standards will contribute to carrying out the purposes of the Act.” Id. at 33-34, 103 S.Ct. at 2862 (internal quotation marks and citations omitted).
In order for agency action to survive judicial review the agency must have “examine[d] the relevant data and articulate[d] a- satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made.’ ” Id. at 43, 103 S.Ct. at 2866 (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 245-46, 9 L.Ed.2d 207 (1962)). The agency’s explanation is subject to review under the standard set out in Over-ton Park, which requires, the court to “consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment.” Id. (quoting Overton Park, 401 U.S. at 416, 91 S.Ct. at 823-24). The Court continued:
Normally, an agency rule would be arbitrary and capricious if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. The reviewing court should not attempt itself to make up for such deficiencies; we may not supply a reasoned basis for the agency’s action that the agency itself has not given. SEC v. Chenery Corp., 332 U.S. 194, 196 [67 S.Ct. 1575, 1577, 91 L.Ed. 1995] (1947).
Id. Finally, an agency is not required “to consider all policy alternatives in reaching decision.” Id. at 51, 103 S.Ct. at 2871.
In Chrysler Corp. v. Department of Transportation, 472 F.2d 659 (6th Cir.1972), this court considered the requirements of APA § 706 as explained and applied in Overton Park. In Chrysler Corp., we emphasized the need for the agency to make a complete record and for the reviewing court to base its decision on a thorough and indepth examination of the record. “Clearly when factual issues are involved, including the issue of whether compliance is technologically feasible, the reviewing court must consider the record upon which the Agency based its order.” Id. at 668. “Any safety standards issued pursuant to the [Safety Act] must be tested for compliance with the statutory limitations of that Act, and this testing can only be done against the record.” Id. at 669.
Given the narrow scope of review permitted when a petitioner claims standards adopted by an agency are arbitrary and capricious, a reviewing court must take care not to merely rubber stamp agency decisions. Rather the court must “ensure that the agency took a ‘hard look’ at all relevant issues and considered, reasonable alternatives...:” Neighborhood TV Co., Inc. v. FCC, 742 F.2d 629, 639 (D.C.Cir.1984). The Neighborhood TV opinion summarized the role of the reviewing court as follows:
In short, the key to the arbitrary and capricious standard is its requirement of reasoned decisionmaking: we will uphold the Commission’s decision if, but only if, we can discern a reasoned path from the [1005]*1005facts and considerations before the Commission to the decision it reached.
Id.
The APA requires an agency engaged in rulemaking to give interested parties an opportunity to comment. 5 U.S.C. § 553(c). This APA section also provides that “[a]fter consideration of the relevant matter presented, the agency shall incorporate in the rules adopted a concise general statement of their basis and purpose.” 5 U.S.C. § 558(c). The court in Automotive Parts & Accessories Ass’n, Inc. v. Boyd, 407 F.2d 330 (D.C.Cir.1968), discussed this requirement. Relating this agency duty to judicial review, the court stated:
We do not expect the agency to discuss every item of fact or opinion included in the submissions made to it in informal rule making. We do expect that, if the judicial review which Congress has thought it important to provide is to be meaningful, the “concise general statement of * * * basis and purpose” mandated by [§ 553(c) ] will enable us to see what major issues of policy were ventilated by the informal proceedings and why the agency reacted to them as it did.
Id. at 338.
III.
With these principles in mind, we turn to the issues raised in this appeal.
The petitioners claim that NHTSA’s decision to use static rather than dynamic testing of the wheelchair securement and occupant restraining devices is not supported by the rulemaking record because a majority of the commentators and studies favor dynamic testing. Thus, they argue, NHTSA ignored significant facts in the record and failed to follow a reasoned path from the facts in the record to the final rule. In addition, the petitioners assert that NHTSA also ignored available alternatives to static testing, and thus acted arbitrarily and capriciously.
In evaluating NHTSA’s decision to utilize static rather than dynamic testing, the court must look to see if NHTSA explained the evidence available and offered a “rational connection between the facts found and the choice made.” Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207 (1962); State Farm, 463 U.S. at 52, 103 S.Ct. at 2871. In determining whether this rational connection exists, the reviewing court must remember that the Safety Act “does not require [NHTSA] to adopt the technological alternative providing the greatest degree of safety.” State Farm Mut. Auto. Ins. Co. v. Dole, 802 F.2d 474, 486-87 n. 23 (D.C.Cir.1986), cert. denied, 480 U.S. 951, 107 S.Ct. 1616, 94 L.Ed.2d 800 (1987).
NHTSA acknowledged in its NPRM that dynamic testing was the preferred approach among commentators. Nevertheless, the agency opted to utilize static rather than dynamic testing in its final rule amending Standard 222. The central rationale for NHTSA’s utilization of static rather than dynamic testing is that important technological issues remain unresolved with dynamic testing procedures. 58 Fed.Reg. at 4587; 56 Fed.Reg. at 48141. This determination must be upheld so long as there is a rational connection between the facts in the record and NHTSA’s final conclusion to use static testing.
In the preambles to its NPRM and final rule, NHTSA explained the benefits of using static testing and discussed its rationale for rejecting dynamic testing. Id. In its NPRM, the agency cited six reasons why dynamic testing was not suitable at the time: (1) need to develop an appropriate test dummy; (2) need to identify human tolerance levels for a handicapped child; (3) need to establish test conditions; (4) need to select a “standard” or surrogate wheelchair; (5) need to establish procedures for placing the wheelchair and test dummy in an effective test condition; and (6) need to develop an appropriate test buck to represent a portion of the school bus body for securement and anchorages. 56 Fed.Reg. at 48141. Because of these technical problems, NHTSA concluded that the delay involved in utilizing dynamic testing was unreasonable. 58 Fed.Reg. at 4587.
[1006]*1006The tests pointed to by the petitioners utilizing dynamic testing procedures do not demonstrate that dynamic testing was practicable at the time the final rule was promulgated, and none specifically resolves the impediments to dynamic testing identified by NHTSA. The Australian, International Standards Organization (ISO), and Canadian standards do include requirements based on dynamic testing, but none provides specific technical data describing research testing or the process by which the test procedure and performance requirements were decided upon. The State-of-the-Art Report, commenting on the Australian standard, noted that “[i]t is quite possible that no testing was done specifically to support or develop the standard.” A letter from the chairman of the Australian Standards Association, however, referred to testing that resulted in development of a test seat and a test protocol. The extent of such testing remains unclear and the doubt expressed in the State-of-the-Art Report appears justified. Further, the Canadian standard does not even apply to crash situations, but to “extreme driving maneuvers.” Other commentators recognized that dynamic testing was not possible at the time, but suggested NHTSA establish a schedule for completing the work necessary to develop dynamic test requirements. •
The University of Michigan Transportation Research Institute (UMTRI) provided the most specific comments to the six unresolved technical issues posed by NHTSA. Nevertheless, UMTRI’s comments did not provide answers to all the technical problems identified by the agency. The UMTRI concluded the first two problems noted by NHTSA were “irrelevant” and claimed that answers to all the remaining problems were “close at hand or already available.” Overall, the UM-TRI comments, although urging NHTSA to provide funding for further development of dynamic testing, clearly acknowledged that technological problems remained unresolved with respect to such testing. NHTSA thoroughly discussed UMTRI’s comments in its final rule and noted it did not believe it could resolve the existing impediments to dynamic testing before the national and international committees reached agreement.
The petitioners argue that NHTSA’s failure to promulgate dynamic testing standards was arbitrary and capricious because the agency ignored relevant facts in the record, failed to follow a “reasoned path” and failed to consider available alternatives. We disagree. As our previous discussion of this issue demonstrates, NHTSA consistently acknowledged the commentators’ preference for dynamic testing, but gave cogent reasons for not adopting that form of testing. We find that the agency thoroughly considered the relevant safety data while responding to the comments it received. The record fully supports the conclusion that a number of technological problems respecting dynamic testing remain unresolved. It was not an abuse of agency discretion to choose static testing, for which standards presently exist, rather than delaying “the entire process of providing crash protection to wheelchair occupants while major research programs are developed, initiated and completed.” 56 Fed. Reg. at 48141.
As a result of our “substantial inquiry,” Overton Park, 401 U.S. at 415, 91 S.Ct. at 823, we conclude that NHTSA’s final determination, basing its standards for seeurement and restraint devices for wheelchairs on school buses on static rather than dynamic testing, fulfilled its mandate under the Safety Act and was neither arbitrary, capricious or otherwise not in accordance with law. The agency decision was based on the “relevant factors” and did not embody “a clear error of judgment.” Id. at 416, 91 S.Ct. at 824. A careful reading of the agency’s published notices, from its original grant of the petition for rulemaking to its final rule, discloses a “reasoned path” in which comments were considered and appropriate responses given. Neighborhood TV, 742 F.2d at 639. Furthermore, NHTSA’s notices, discussions and responses complied with the APA’s requirement of a “concise general statement of their basis and purpose” in its treatment of the testing issue. 5 U.S.C. § 553(c). Finally, the agency was not required to consider every conceivable alternative. The Supreme Court has stated that rulemaking “cannot be found wanting simply because the agency failed to include every alternative device and thought conceivable by [1007]*1007the mind of man ... regardless of how uncommon or unknown the alternative may have been_” Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519, 551, 98 S.Ct. 1197, 1215, 55 L.Ed.2d 460 (1978). NHTSA did give thorough consideration to the one alternative put forward by the petitioners and various commentators — dynamic testing — and gave well-supported reasons for rejecting that method of testing.
The petitioners level both substantive and procedural attacks at NHTSA’s decision not to promulgate crashworthiness standards for wheelchairs carried on school buses. We consider these issues separately.
1.
Throughout the rulemaking process NHTSA concentrated on finding a method of safely securing wheelchairs of disabled students to the floor of school buses and providing safe restraint systems for the wheelchair occupants. The issue of the crashworthiness of wheelchairs was first raised in a comment to the May 30, 1990, notice announcing the availability of the State-of-the-Art Report. Although that report concerned wheelchair securement and occupant restraint only, the Minnesota School Bus Safety Committee filed a comment questioning the ability of various wheelchair designs to withstand crash forces. NHTSA responded to this comment in its NPRM, agreeing that it was a legitimate concern. NHTSA then noted that other organizations were working on safety standards for wheelchairs and that some manufacturers of wheelchairs were actively engaged, in concert with “standards” organizations, in developing “transport” wheelchairs that could “withstand a 30 mph/20 g crash test with an integral occupant restraint system.” 56 Fed.Reg. at 48144. NHTSA did not propose requirements for a “transport-certified” wheelchair, however, because it believed “that such concerns as development of appropriate levels of structural integrity for wheelchairs and other mobile seating devices, as well as the development of objective tests to ensure such integrity, will take an extremely long time.” Id.
In the final rule NHTSA acknowledged comments suggesting the agency should do more to provide acceptable crash protection. The rule discussed specific comments, both pro and con on the questions, and adhered to its earlier decision, stating:
The agency agrees that some wheelchairs may not perform as well as others in a crash situation. However, after reviewing the comments, the agency remains convinced that it is inappropriate to include requirements applicable to wheelchairs and other mobile seating devices that may be used on school buses. Any attempt to do so would lead to extensive delays in this rulemaking, since there are no recognized performance requirements for such devices. Moreover, the agency does not have authority to regulate wheelchairs as such, since they are not “motor vehicle equipment” within the meaning of section 102(4) of the Safety Act. The agency has the authority to impose performance requirements on seating devices that are designed specifically for use on vehicles, as it has done with respect to certain child restraint systems (see, Standard No. 213, Child Restraint Systems). However, most wheelchairs are not designed for use in motor vehicles. Their use in vehicles is incidental to their primary function of providing mobility.
58 Fed.Reg. at 4591.
Thus, NHTSA concluded it would be “inappropriate” to establish crashworthiness standards due to technological problems that would result in significant delays in the rule-making process. This determination can be set aside as arbitrary and capricious only if the agency
has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.
State Farm, 463 U.S. at 43, 103 S.Ct. at 2867.
NHTSA’s assertion that no expert consensus exists on a variety of technical problems [1008]*1008related to wheelchair crashworthiness standards is supported by the record in that none of the existing or proposed standards placed in the record includes any design specifications or performance criteria for a wheelchair that will withstand the forces of a crash. The Australian standard uses a “test seat” rather than an actual wheelchair, which “shall be capable of resisting significant deformation during a test.” This hardly tells manufacturers or consumers what constitutes a crashworthy wheelchair, and it does not establish an “objective” test standard “capable of producing identical results when test conditions are exactly duplicated” as this court has interpreted 15 U.S.C. § 1392(a) to require. Chrysler Corp., 472 F.2d at 676. Likewise, neither the Canadian Standard, the British Standard, the Swedish Standard, nor the ISO Standard says anything about the strength or performance requirements of wheelchairs.
It is settled law that “[t]he [Safety] Act expressly permits [NHTSA] to consider such factors as reasonableness and practicality in addition to safety features.” State Farm Mut. Auto. Ins. Co. v. Dole, 802 F.2d 474, 486-87 n. 23 (D.C.Cir.1986), cert. denied, 480 U.S. 951, 107 S.Ct. 1616, 94 L.Ed.2d 800 (1987). Practical considerations also support NHTSA’s conclusion not to establish wheelchair crashworthiness standards. In NHTSA’s FRE, it noted that requiring only “transport-certified” mobility aids be used on school buses would undermine the concept of accessible transportation, the fundamental precept of the Americans With Disabilities Act, 42 U.S.C. § 12131 et seq. The agency noted this would directly discriminate against those who could not afford what may be a more expensive mobility aid. In making its determination, NHTSA also relied on comments from Transportation Alternatives, apparently an “interested person,” which noted that “improve[d] crashworthiness is not necessarily a fair trade-off for increased wheelchair weight, which would be a serious burden to wheelchair occupants during the rest of their day when they must move about in the wheelchair itself.” 58 Fed.Reg. at 4591.
The agency’s determination that it was inappropriate to prescribe crashworthiness standards for wheelchairs was neither arbitrary, capricious, or otherwise not in accordance with law.
2.
NHTSA’s second reason for refusing to promulgate crashworthiness standards— its purported lack of authority — is attacked by the petitioners both as substantively incorrect and as involving procedural error. An agency may only act within the scope of its authority as conferred by statute. Here the agency questioned its authority to prescribe safety standards for wheelchairs as such, and, upon determining that wheelchairs are not “motor vehicle equipment,” concluded that it lacked the requisite authority to regulate wheelchairs. Substantively, the petitioners contend that NHTSA possesses the authority to regulate wheelchair crashworthiness because wheelchairs are motor vehicle equipment when used to transport their occupants in motor vehicles. At the very least, say the petitioners, wheelchairs when so used are “accessories” under the “motor vehicle equipment” definition. The parties discuss this contention at length in their briefs, but for reasons hereinafter explained we find it unnecessary to decide the issue of the agency’s authority. Rather, we conclude that NHTSA’s determination that it was “inappropriate” to issue crashworthiness standards was not arbitrary or capricious.
The petitioners argue that NHTSA violated the APA’s notice and comment procedural requirements. Specifically, they contend that because the agency relied on its lack of authority for the first time in the final rule, they were deprived of an opportunity to comment and contest the accuracy of the determination. According to the petitioners NHTSA compounded this procedural error in its order denying Stephens’s motion for reconsideration of the final rule by listing yet another reason for not prescribing crashwor-thiness standards. In that order NHTSA stated that a wheelchair is neither motor vehicle equipment nor an “accessory” under NHTSA criteria. NHTSA conceded that it had the authority to regulate wheelchairs if they qualified as accessories.
[1009]*1009The APA commands agencies to give interested persons an opportunity to comment on proposed rules. 5 U.S.C. § 553(c). The petitioners argue that they were not given the required opportunity to comment and add facts to the record to support NHTSA’s authority to regulate because prior to publication of the final rule NHTSA gave no indication that it lacked authority, and the record was closed by the time NHTSA introduced the question. Even though Stephens did address the question of whether wheelchairs are motor vehicle equipment when used by school bus passengers in his motion for reconsideration, the agency introduced the “accessories” argument at a time when no response was permitted.
NHTSA consistently sought to make it safer for students in wheelchairs to travel on school buses by focusing on seeurement and restraint. In choosing static testing, which was available, over dynamic testing, which was not available, the agency elected to push ahead with early promulgation of standards based on available technology rather than delay the creation of standards indefinitely while awaiting the completion of technological tasks in the distant future. The agency’s treatment of the issue of crashworthiness followed the same line of reasoning. NHTSA wrote in the NPRM:
At the present, almost any type of wheelchair or mobile seating device can be used on school buses. The agency believes this proposed rule would improve on these situations by mandating adequate securement and restraint devices, even though requirements for a “transport certified” wheelchair are not proposed.
56 Fed.Reg. at 48144.
In its FRE published in September 1992, NHTSA stated it was inappropriate to regulate wheelchair crashworthiness because of the delay referred to above, and because such standards could decrease the availability of wheelchairs for those who could not afford standardized wheelchairs. We believe the findings in the NPRM and FRE at least foreshadowed the ultimate conclusion that the agency lacked authority to promulgate such standards. It is not difficult to understand why, when faced with evidence that requiring only standardized, crash-proof wheelchairs be used on school buses might result in fewer wheelchair-bound students having the opportunity to travel to school on buses, NHTSA determined that such an undertaking was beyond its authority.
Nevertheless, NHTSA’s decision must be upheld if either of the independent grounds relied upon — inappropriateness and lack of authority — validly supports the result. Carnegie Natural Gas Co. v. FERC, 968 F.2d 1291, 1294 (D.C.Cir.1992). Regardless of whether the petitioners had an opportunity to dispute the agency’s lack of authority reasoning, we think it abundantly clear that the conclusion that it would be inappropriate to set standards for wheelchair crashworthiness was not arbitrary or capricious. The record fully supports this determination and NHTSA certainly fashioned a reasoned path from its first consideration of the issue to its final determination.
IV.
The petitioners also argue that NHTSA’s refusal to promulgate crashworthiness standards is “otherwise contrary to law” because that refusal violates section 504 of the Rehabilitation Act of 1978, 29 U.S.C. § 794 (1988). That statute prohibits discrimination against otherwise qualified individuals with disabilities in “any program or activity conducted by any Executive agency.” 29 U.S.C. § 794(a).
As previously noted, the petitioners commenced this action by filing a complaint under section 504. Because the Department of Transportation did not have in place a procedure for dealing with Rehabilitation Act claims, it treated the petitioners’ complaint as a petition for rulemaking to amend Standard 222.
The petitioners contend that NHTSA has failed to provide the full protection to wheelchair-bound students that it provides to non-handicapped students, thus violating section 504. They argue that the amendment to Standard 222 in the final rule is deficient in failing to provide “static standards for the chair itself or dynamic testing standards for [1010]*1010the entire seating system.” According to the petitioners, NHTSA did not discharge its duty to disabled students by merely promulgating regulations for secuTement of wheelchairs and restraints for their occupants.
The agency replies to the petitioners’ arguments by reiterating its contention that it considered comments suggesting the need for dynamic testing of securement and restraint systems and the establishment of crashworthiness standards for wheelchairs and found that no adequate technical foundation existed for such standards.
In Davidson v. U.S. Department of Energy, 838 F.2d 850, 854 (6th Cir.), cert. denied, 487 U.S. 1207, 108 S.Ct. 2849, 101 L.Ed.2d 886 (1988), we stated, “Any alleged adverse discriminatory impact of the DOE regulations is not a matter properly before a court reviewing agency action under the APA.” We added that “artful pleading” cannot be employed to transform a claim for violation of the Rehabilitation Act into a suit under the APA. Id. The Davidson panel reached this conclusion by adopting the language and reasoning of the court in Cleghorn v. Herrington, 813 F.2d 992 (9th Cir.1987).
Davidson represents the minority view on this issue. Other courts of appeals have considered Davidson and specifically refused to follow it. See Cousins v. Secretary of the U.S. Department of Transportation, 880 F.2d 603, 609 (1st Cir.1989); Clark v. Skinner, 937 F.2d 123, 125-26 (4th Cir.1991); Ward v. Skinner, 943 F.2d 157, 160 (1st Cir.1991), cert. denied, — U.S. -, 112 S.Ct. 1558, 118 L.Ed.2d 207 (1992). More recently the Court of Appeals for the Ninth Circuit criticized Davidson for failing to distinguish Cleghom, and held that an alleged agency violation of a substantive statute may be the basis of a claim brought pursuant to the APA. J.L. v. Social Security Administration, 971 F.2d 260, 266 (9th Cir.1992).
Despite the unequivocal language in Davidson, we find important factual differences that distinguish it from the present case. Davidson was a declaratory judgment action brought in district court to have regulations of the Department of Energy held unlawful. The Davidson plaintiffs, who were interested persons, had notice of the rule-making that led to adoption of the regulations and an opportunity to offer comments and objections. Nevertheless, they did not participate in the process and voice the concerns on which they based their action in ‘’district court. Thus, there was an element of waiver in the case, and the district court case constituted a collateral attack on the rule. Davidson’s pronouncements are, therefore, limited by its facts.
The present case reached this court by an entirely different route. Here, the petitioners, seeking a change in an Executive agency regulation, presented a claim under the Rehabilitation Act to the agency at the initiation of the rulemaking process and participated in this process every step of the way until a final rule issued. They then followed the prescribed procedure by seeking review in the court of appeals rather than mounting a collateral attack on the regulations in district court. Under these circumstances, we agree with the courts which have held that the Rehabilitation Act claim is properly before the court of appeals pursuant to a petition for review.
In the NPRM, NHTSA stated its evaluation of the proposed tiedown and restraint standards as follows:
This proposal would provide a level of occupant protection for students in wheelchairs comparable to that currently provided to persons able to use standard bench seats. In addition, this proposal is intended to prevent potential injuries to all occupants caused by a wheelchair which is not adequately secured.
56 Fed.Reg. at 48140. NHTSA adhered to this view in the final rule. As the agency further found, any effort to include the requested requirements in its rule would delay the amendments to Standard 222 with their resulting immediate improvements to the safety of disabled students on school buses. The Safety Act directs the agency to promul[1011]*1011gate requirements only if they are practicable.” To attempt to fashion rules in an area where many technical problems have been identified and no consensus exists for their resolution would not be faithful to the agency’s mandate, and would be unreasonable if the attempt caused delay in going forward with a rule that would benefit disabled students.
We find no merit in the petitioners’ contention that the final rule violates the Rehabilitation Act. To the contrary, we find that the rule mates a substantial contribution to the increased safety of disabled students confined to wheelchairs while being transported on school buses. Rather than excluding students from, or denying them the benefit of participation in activities, the rule enhances the opportunities for disabled students to attend school with other students. Wheelchair bound students are clearly “otherwise qualified” to attend school , and therefore are within the protection of the Rehabilitation Act. Southeastern Community College v. Davis, 442 U.S. 397, 99 S.Ct. 2361, 60 L.Ed.2d 980 (1979). This case is singular, however, in that the agency has actually expanded the availability of transportation for these disabled students rather than “excluded [them] from the participation in ... any program or activity by [NHTSA].” 29 U.S.C. § 794(a). The fact that NHTSA determined, after careful study of comments, not to adopt all suggestions and demands of the petitioners does not render the final rule unlawful under the standards of the APA, or demonstrate a violation of the Rehabilitation Act.
The petition for review is DENIED.