Claimant, Dr. Henry J. Niemeyer, M.D., appeals from an order of the district court affirming the Food and Drug Administration's (FDA) refusal to accept claimant's proposed relabeling of two previously condemned "Diapulse" devices.1 We affirm.

This matter arose from two civil seizures under Section 304(a) of the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. Sec. 334(a), to condemn two identical Diapulse devices. The first device was seized from claimant's medical offices on February 26, 1973; the second on February 3, 1975. The two cases were consolidated for trial. The Government, on behalf of the FDA, alleged that the devices were misbranded within the meaning of 21 U.S.C. Secs. 352(a) and 352(f)(1) because they were ineffective for the claims contained in their labeling and failed to bear adequate directions for their use.2 Claimant elected not to defend the devices' labeling and instead sought to have the devices brought into compliance with the law through relabeling.

The district court, Judge Will, ruled that the Government was entitled to partial summary judgment, but, over the Government's objection, declined to condemn the devices and instead conducted a bench trial to determine the validity of the claims in the relabeling proposal. Judge Will entered a judgment in favor of claimant, directing that the devices be returned to claimant provided that they be relabeled solely for use as an adjunct treatment in healing soft tissue inflammation, sprains and strains. The Government appealed, contending that the district court erred in conducting a trial de novo on the relabeling proposal rather than condemning the devices and requiring claimant to submit a compliance proposal to the FDA.

This court reversed, holding that a trial de novo was improper and that claimant's only recourse was to seek, after any unsuccessful remand for consideration of relabeling, judicial review of the agency action "under an appropriate standard of judicial review of administrative action." U.S. v. An Article of Device ... Diapulse, 650 F.2d 908, 911 (7th Cir.1981).

Upon remand the district court, Judge Moran, entered an order of condemnation and remanded the matter to the FDA for consideration of the claimant's proposed relabeling. The FDA denied claimant's request for approval of proposed new labeling in a letter from Harry E. Butts, Director, to Michael G. Berkman, claimant's attorney, dated November 3, 1982. In August 1983 the Government sought the entry of a judgment and claimant responded by seeking review of the FDA's decision. Judge Moran, by order dated March 21, 1984, upheld the FDA's decision.

* The first issue raised by claimant's appeal is the scope of judicial review applicable to the FDA's decision. Claimant argues that because the entire evidentiary record of the trial before Judge Will is now a part of the administrative record of the FDA in the present case3 this is an "evidentiary" rather than a "non-evidentiary" record reviewable under the substantial evidence standard of Sec. 706(2)(E) of the Administrative Procedure Act, 5 U.S.C. Sec. 706(2)(E). The Government argues that the FDA's decision was the product of informal agency adjudication on a non-evidentiary record reviewable under the arbitrary and capricious standard set forth in Sec. 706(2)(A) of the Act. Each party contends he can succeed under either standard. Judge Moran did not reach the question of which standard applied, finding that under either standard the appeal would fail.

In all cases agency action must be set aside if the action was "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. Sec. 706(2)(A). "Review under the substantial-evidence test is authorized only when the agency action is taken pursuant to a rulemaking provision of the Administrative Procedure Act itself, 5 U.S.C. Sec. 533 (1964 ed., Supp. V), or when the agency action is based on a public adjudicatory hearing. See 5 U.S.C. Secs. 556, 557 (1964 ed., Supp. V)." Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 414, 91 S. Ct. 814, 822, 28 L. Ed. 2d 136 (1971). Neither situation exists here. The FDA's decision to deny relabeling is the result of informal adjudication4 subject to review under the arbitrary and capricious standard.

A condemned device may be brought into compliance with the provisions of the Food, Drug, and Cosmetic Act (FDCA) "under the supervision of an officer or employee duly designated by the Secretary." 21 U.S.C. Sec. 334(d)(1). Claimant's devices were seized before 1976; at that time the Act did not require preclearance of medical devices, nor did it set forth scientific standards for demonstrating a device's effectiveness. The FDA took the position that in general the requirements were the same as those established statutorily for drugs, i.e., claimant has the burden of producing "adequate and well-controlled investigations, including clinical investigations, by experts qualified by scientific training and experience ... on the basis of which it could fairly and responsibly be concluded by such experts that the drug [or device] will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, or suggested in the labeling or proposed labeling thereof." 21 U.S.C. Sec. 355(d).5 The FDA applied this standard in evaluating claimant's proposed relabeling of Diapulse. The agency's consideration of claimant's compliance proposal was an informal adjudicatory process. The agency was not required to conduct an "on the record" hearing designed to produce a record that is to be the basis of agency action--"the basic requirement for substantial-evidence review." Citizens to Preserve Overton Park v. Volpe, 401 U.S. at 415, 91 S.Ct. at 823.

Claimant argues that because the trial before Judge Will occurred and the transcript of that trial is part of the administrative record the FDA's action is based upon a public adjudicatory hearing and therefore subject to substantial evidence review. It is clear, however, from U.S. v. An Article of Device ... Diapulse, supra, that the trial should not have occurred, and its record is part of the administrative record in this case solely for the information it contains and as a matter of administrative convenience. The Agency is not bound by the findings in that case.6 The fact that trial procedure took place does not transform the Agency's decision making process into an adjudicatory hearing.

When reviewing agency action under the arbitrary and capricious standard,

the court must consider whether the decision was based on consideration of the relevant factors and whether there has been a clear error of judgment. Although this inquiry into the facts is to be searching and careful, the ultimate standard of review is a narrow one. The court is not empowered to substitute its judgment for that of the agency.

Citizens to Preserve Overton Park v. Volpe, 401 U.S. at 416, 91 S.Ct. at 824 (citations omitted). The decision which must be reviewed here is contained in a letter from Harry E. Butts, Director, to Michael G. Berkman, claimant's attorney, dated November 3, 1982.

Claimant proposed that the Diapulse device be relabeled for adjunctive use in the treatment of inflammation in soft tissues, sprains or strains. His submission to the FDA included over 50 published scientific studies concerning Diapulse or other allegedly similar devices as well as the transcript of the testimony from the trial before Judge Will. Claimant asserts that the studies he relies upon are adequate and well-controlled and establish Diapulse's efficacy for the adjunctive uses described.7

The FDA reviewed all of the materials submitted by claimant and determined that they failed to demonstrate the efficacy of the Diapulse for the uses described. Director Butts' letter summarized:

Generally speaking, the papers that you have submitted or referenced do not provide adequate data from well-controlled studies to support a finding that the Diapulse device is effective for the purported uses. For the most part, the studies were not designed well, were not quantifiable, and were not otherwise conducted under applicable and essential principles of adequate and well-controlled clinical investigations.8

The Butts letter begins with some general comments on the inadequacy of claimant's evidence and then specifically reviews twenty-two of the documents submitted by claimant.

Claimant proposed to label the Diapulse device for adjunctive use in the treatment of soft tissue inflammation, sprains and strains. Director Butts noted that a key principle of a well designed study is that the indications to be studied must be well defined in terms of specific medical conditions recognized by the medical community. He found the terms "soft tissue inflammation," "sprains," and "strains" too broad and vague to be uniformly accepted by qualified scientific investigators in the medical community. For example, Director Butts indicated that "soft tissue" includes all body components other than bone. "Inflammation" is associated with a large number of medical conditions and in some respects the presence of inflammation is desirable.

Claimant answers this charge by noting that while the studies he relies upon do not recite "treatment of soft tissue inflammation, sprains or strains," each study does deal with a medical condition in which inflammatory response (inflammation) is present. However, in his testimony at the trial before Judge Will claimant indicated that everything that goes wrong with the human body is an inflammatory process which he believes is amenable to treatment with the Diapulse Machine. The FDA's charges of vagueness appear well taken.

The term "adjunctive use" must similarly be carefully defined, and must be considered when the studies are designed. Director Butts indicated that ordinarily adjunctive use studies must include four groups of patients, including groups treated with the primary treatment, the adjunct treatment, the primary and adjunct treatment combined, and an untreated control group. At the trial before Judge Will, Dr. Michaelson, a Government witness, explained the basis of the four-group experimental design in studying an adjunct treatment. He noted that in designing an experiment the experimenter wants to control all variables to the greatest extent possible. The four-group design allows for the greatest control and allows the experimenter to ascertain the interaction of the primary and adjunct treatments as compared to either treatment alone or no treatment at all. Dr. Michaelson noted that while either treatment might be effective alone the combination could have a synergistic effect, producing greater healing, or an additive negative effect, producing less healing than either treatment alone or no treatment at all. None of the studies submitted by claimant utilized the four-group design and Director Butts concluded all were improperly designed to evaluate the Diapulse devices in an adjunctive role.

Claimant contends that the FDA's designated four-group protocol for study of the Diapulse device in an adjunctive role is "flagrantly improper." He argues that the designation of an injured "untreated control group" would deprive injured patients of any treatment for a recognized medical condition and be both professionally incompetent and unethical. Claimant refers to no scientific evidence or opinion in the record that supports his view that this FDA requirement is arbitrary and capricious. Both the drug approval regulations, 21 C.F.R. Sec. 314.111(a)(5)(iii)(4)(i), and the medical device approval regulations, 21 C.F.R. Sec. 860.7(f)(1)(iv)(a), provide for untreated control groups. The regulations also provide for active treatment controls where no treatment or placebo treatment would be contrary to the patient's best interest. 21 C.F.R. Sec. 314.111(a)(5)(iii)(4)(iii) and 21 C.F.R. Sec. 860.7(f)(1)(iv)(c). Furthermore, the FDA regulations provide that such research should be undertaken with the patient's informed, written consent. See 21 C.F.R. Sec. 312.1(a)(12) and (13), forms FD-1572 and FD-1573 6g; 21 C.F.R. Sec. 312.20 (Recommendations Guiding Medical Doctors in Biomedical Research Involving Human Subjects ), I.9. Although claimant's challenge to the four group procedure has at least surface appeal to lay persons, we have no basis in the absence of expert opinion to hold that it is a clearly excessive demand and therefore arbitrary and capricious.9

Director Butts further noted that the proposed labeling for the devices must specify conditions of use that are similar to those followed in the studies. Only two of the studies submitted appear to study conditions of use for which claimant seeks approval, and they involved only one of the proposed conditions, sprains. Wilson, D.H., "Treatment of Soft-tissue injuries by pulsed electrical energy." British Medical Journal 2 (April, 1972); Wilson, D.H., "Comparison of short-wave diathermy and pulsed electromagnetic energy in treatment of soft-tissue injuries." Physiotherapy, 60 (10):309-10 (October 1974). Moreover, the treatment time in the two Wilson studies was considerably longer than that on claimant's proposed relabeling for the seized device. The device was applied one hour each day in the Wilson studies, two to six times longer than the time indicated on the proposed labeling for the Diapulse device. Director Butts noted that this difference is most likely significant and, at the very least, repetition of the studies using the times proposed by claimant is needed.10

The Director characterized the remainder of claimant's studies as falling into four categories: (1) reports of studies concerning basic biological phenomena which offered little more than encouragement for follow-up studies directed toward the medical indications to be treated; (2) studies with animals which again were only indicative; (3) studies in humans concerning medical conditions sufficiently different from those proposed by claimant (e.g., surgical wound healing, burns and bone healing) so that the results--even if positive and from well-controlled studies--could not be extrapolated; and (4) studies conducted with devices that are substantially different from Diapulse. Director Butts indicated that the material submitted by claimant provides, at best, only a theoretical basis for speculation that Diapulse might be effective for various medical indications. He concluded that such evidence was insufficient to support a finding by the FDA that the Diapulse device is effective for the uses proposed by claimant.

In addition to these generalized criticisms of the evidence in support of claimant's relabeling proposal, Director Butts specifically addressed twenty-two of claimant's supporting documents and found them insufficient to determine the effectiveness of the Diapulse device. This court has carefully reviewed those documents, Director Butts' criticisms of them and claimant's responses. Despite claimant's vigorous contention to the contrary, we find that the FDA's analysis of the evidence and its decision to deny claimant's relabeling proposal were not arbitrary and capricious. The FDA thoroughly examined the claimant's supporting documents and found they fell far short of the standards for "adequate and well-controlled investigations." 21 U.S.C. Sec. 355(d); 21 U.S.C. Sec. 360c(a)(3)(B). Claimant has failed to demonstrate that the FDA committed a clear error of judgment or acted arbitrarily and capriciously in making its decision to deny relabeling.11

Claimant's final argument is that the FDA acted arbitrarily, capriciously and in bad faith in denying relabeling for claimant's Diapulse devices while approving labeling for an allegedly similar device, the Bi-Osteogen device. He contends that the Agency applied more stringent approval requirements to Diapulse than to the Bi-Osteogen device and that this arbitrary singling out reflects a pattern of bias against Diapulse.

Claimant asserts a "remarkable functional similarity" between the pulsed electromagnetic output of his Diapulse machines and the electrical output of the Bi-Osteogen machine. Both the Diapulse and Bi-Osteogen devices produce pulsed electromagnetic energy. However, the Bi-Osteogen device operates on a different frequency, with a different wave form and different radiation patterns from Diapulse. Director Butts found that these differences result in very dissimilar depositions of energy to the body which preclude generalizing from the effects of the Bi-Osteogen device to the Diapulse device. The FDA approved the use of the Bi-Osteogen device for the treatment of ununited bone fractures, not for adjunctive treatment of inflammation of soft tissue, sprains or strains. In addition the Bi-Osteogen device is applied 10 to 12 hours per day over a period of several months, compared to the 10 to 30 minutes per day proposed by claimant for the use of the Diapulse device. See Bassett, C. Andre L., "Pulsed Electromagnetic Field Treatment in Ununited Fractures and Failed Arthrodeses," Journal of the American Medical Association, Vol. 247, No. 5 (February 5, 1982) 623-28. Claimant's contention that the Bi-Osteogen and Diapulse devices are so similar that approval of one and rejection of the other is arbitrary and capricious is unfounded.

Claimant's argument that more stringent requirements were applied to Diapulse than to the Bi-Osteogen device apparently relates to the requirement of untreated controls. The studies in support of the Bi-Osteogen did not utilize untreated control groups. In those studies the patients themselves served as their own controls because they had been treated for bone fractures without success prior to becoming investigational subjects for the Bi-Osteogen device. The FDA found the use of patients as "historical controls" in such circumstances scientifically acceptable since such patients generally do not heal over time as is the case with patients that have strains, sprains and soft tissue inflammation. The drug and medical device regulations recognize historical controls as appropriate in certain circumstances. See 21 C.F.R. Sec. 314.111(a)(5)(iii)(4)(iv) and 21 C.F.R. Sec. 860.7(f)(1)(iv)(d). Claimant failed to demonstrate that the FDA discriminated against the Diapulse device by applying more stringent standards or acted arbitrarily, capriciously or in bad faith in denying its relabeling while approving labeling for the Bi-Osteogen device12.

The judgment of the district court is AFFIRMED.