OPINION AND ORDER

CRABB, Chief Judge.

This is a civil action for declaratory and injunctive relief brought pursuant to the Food, Drug, and Cosmetic Act, 21 U.S.C. § 301-394, the National Environmental Policy Act, 42 U.S.C. § 4321-4370d, and the Administrative Procedure Act, 5 U.S.C. § 500-706. Plaintiffs are American consumers of commercially sold dairy products. Defendants are Donna Shalala, Secretary of the Department of Health and Human Services, and Dr. David Kessler, Commissioner of the Food and Drug Administration. Plaintiffs challenge defendants' approval of intervenor-defendant Monsanto Company's new drug application for Posilac®, a milk production-enhancing, synthetic bovine growth hormone drug. Specifically, plaintiffs contend that: 1) the approval was arbitrary and capricious because the FDA failed to consider health and safety issues related to the use of Posilac; 2) defendants failed to require mandatory labeling of products from cows treated with Posilac; and 3) defendants failed to conduct an adequate environmental assessment or issue an environmental impact statement assessing the environmental effects of Posilac's approval. Plaintiffs request both a declaration under 28 U.S.C. § 2201 and Fed.R.Civ.P. 57 that defendants failed to perform their statutory duties in approving Posilac and a permanent injunction suspending the approval of Posilac until defendants comply with their statutory obligations.

The case is before the court on the parties' cross-motions for summary judgment. After a close review of the parties' submissions, I find that plaintiffs have offered no admissible, relevant evidence putting any material facts into dispute. Plaintiffs have not shown that defendants acted arbitrarily and capriciously in approving Posilac or in declining to require product labeling and they have not shown that defendants' environmental assessment was inadequate. Therefore, plaintiffs' claims must be dismissed.

From the proposed findings of fact of the parties, I find that the following material facts are undisputed.

UNDISPUTED FACTS

Bovine somatotrophin (bST), a bovine growth hormone, is a naturally occurring protein hormone produced in the pituitary gland of all cattle. In the 1930s, scientists discovered that injecting dairy cows with bovine growth hormone from other cattle could increase the cows' milk production but the discovery was not pursued on a wide scale because extraction of the hormone from cattle was not cost effective. In the 1980s, however, scientists developed a synthetic recombinant bovine growth hormone. Scientists can now isolate the gene responsible for natural bovine growth hormone, transfer that genetic material into bacterial cells called a "recombinant fermentation organism" and "program" the bacterial cells to produce a synthetic version of the hormone.

In the early 1980s, the FDA approved intervenor-defendant Monsanto Company's investigative new animal drug application for Posilac, a synthetic recombinant bovine somatotrophin. In 1987, Monsanto submitted a new animal drug application for Posilac to the FDA. Over the next several years, Monsanto supported the application with studies and reports documenting the safety and effectiveness of the drug. After reviewing those materials, the FDA approved Monsanto's application for the subcutaneous (injectable) use of Posilac on November 5, 1993. Posilac is the first genetically engineered animal drug to be approved for use in dairy cows and the first milk production enhancement drug to be approved for sale by the FDA.

The FDA approved Posilac despite criticism that the drug would have a significant negative effect on the health of dairy cows and despite concern about potential negative health effects on human consumers of dairy products derived from cows treated with rbST. Scientists, economists, farmers, and environmental and animal welfare organizations have questioned the safety and quality of rbST-derived products. In addition, the FDA received thousands of letters from consumers asking it either to deny approval of rbST or to require labeling of rbST-derived products. The General Accounting Office advised the FDA to withhold approval of Posilac until further research of rbST's potential negative impact on human health could be conducted. After the FDA approved Posilac for marketing, Congress delayed sale of the drug for 90 days while an inter-agency task force supervised by the Office of the President reviewed the data upon which the FDA based its decision. In January 1994, the task force concluded that the FDA's position was adequately supported. The FDA made available to the public a summary of the safety and effectiveness data submitted by Monsanto on which the agency relied in approving Posilac for mass marketing.

A. Cow Safety

Use of Posilac may affect cows adversely in several ways. Posilac increases the risks of reduced pregnancy rates, ovarian cysts and uterine disorders, decreased lengths of gestation periods and lower birth weight of calves. Posilac increases the risk of retained placentas and twinning rates in cows. It may cause increased bovine body temperatures, indigestion, bloating, diarrhea, enlarged hocks, enlarged lesions and injection site swellings. Additionally, use of Posilac increases the risk of clinical and subclinical mastitis, a bacterial infection of the udder. In absolute terms, rbST increases the risk of mastitis by about 0.1 case per cow per year. This risk is less than the risk of mastitis posed by seasonal change.

Before approving Posilac, the FDA reviewed the data submitted by Monsanto as part of its new drug application and considered Posilac's effects on animal health, including: 1) the acute and chronic toxicity of the drug; 2) its effects on reproduction; 3) calf birth traits, growth and health; 4) increased incidence of mastitis; 5) musculoskeletal effects; 6) digestive disorders including indigestion, bloating and diarrhea; 7) injection site reactions; 8) nutrient intake, body weight and body condition; 9) general cow health; and 10) miscellaneous health variables, including circulating anti-somatotrophin binding, blood variables, body temperature and urinalysis results. The agency determined that the risks to cows associated with the use of Posilac could be managed properly under a "manageable risk" criteria and that the risks to animal health were not significant enough to warrant denial of the drug application. (The FDA has never applied a zero risk standard when assessing the safety of new animal drugs.) Monsanto conducted a 14-day drug tolerance study that involved injecting a herd with dosages of rbST up to thirty times the normal dosage. Analysis of both cow and fetal blood and tissue revealed only one health side effect: slight swelling at the injection site.

B. Human Consumer Safety

Before approving Posilac, the FDA considered Posilac's possible negative effects on human health, including the possible effects of 1) increased levels of rbST in milk; 2) increased levels of insulin growth factor in milk; and 3) increased amounts of antibiotic drug residues in milk.

1. RbST

BST and its synthetic counterpart rbST are protein hormones that are orally inactive in humans. Upon ingestion, protein hormones (unlike steroid hormones) are broken down by enzymes in the intestines and digested. BST and rbST are orally inactive in cows as well. Like insulin, which must be injected to take effect, bST and rbST must be injected to stimulate milk production in cows. Even if injected into humans, however, rbST would not stimulate growth because human somatotrophin has a significantly different amino acid sequence from that of rbST and human receptors will not bond with rbST. Furthermore, heat destroys rbST, so that pasteurizing milk and cooking meat would tend to destroy at least 90% of the rbST in milk or beef. The FDA found that even at exaggerated doses the ingestion of rbST poses no significant risk to human safety.

2. Mastitis and antibiotics

Antibiotics are used to treat bovine mastitis. Trace residues of the antibiotics can appear in the milk of dairy cows. The level of antibiotic residue in milk is regulated, however. Forty-nine states have adopted the Grade A Pasteurized Milk Ordinance, established by the FDA and the Public Health Service in cooperation with state and local regulatory agencies and members of the milk industry. The ordinance defines standards for milk purity, establishes standards for production of pasteurized milk and milk products, details the method of inspections of farms and processing plants and provides that only producers, haulers, processors and distributors who meet the requirements of the ordinance shall be given permits. Under the milk ordinance, milk that is inspected and found to contain drug residues in excess of the standard must be discarded and the responsible producer is subject to regulatory sanctions.

Although states test milk routinely for drug residues in accordance with the milk ordinance, they generally test only for the presence of four of the antibiotics most commonly used to treat bovine infection, the beta-lactams, which are penicillin-like antibiotics. It has been estimated, however, that over fifty different drugs are used to treat bovine infections, some not approved for use on dairy cows, and the presence of residues from these other drugs would go undetected by most current testing procedures. The General Accounting Office has concluded that there is currently no means of assessing the degree to which current milk supplies are contaminated by these other drugs. One concern surrounding Posilac is that its use will increase the frequency of bovine infection, particularly mastitis, which will in turn lead to an increased need for antibiotics to treat dairy cow infections. Because the current regulatory scheme does not detect the presence of all drug residues in dairy products, there is a risk that greater amounts of antibiotic drug residue will be ingested by human dairy consumers.

To date there have been no long term studies of the impact on human health of increased antibiotics in milk. However, the FDA determined that the increased risk of bovine mastitis caused by Posilac did not present a risk to human health because 1) the increased risk of mastitis from the use of Posilac is not great, and in fact is much lower than the increased risk caused by the many other factors that contribute to mastitis; 2) milk is tested for beta-lactam drugs, and any milk found to contain illegal residues is discarded; and 3) the beta-lactams are the most commonly used antibiotics to treat mastitis. The FDA and its Veterinary Medicine and Food Advisory Committees determined that state and local regulation of milk production and distribution ensures that milk is free from the most widely used antibiotics and that introduction of Posilac to the market would not cause a statistically significant increase in the amount of antibiotic residues found in milk.

3. Somatic cell count

Mastitis brings with it an increase in the amount of somatic cells (i.e., white blood cells) present in dairy milk. Studies by Monsanto showed a significant increase in the somatic cell count in milk from rbST-treated cows in some herds, although the increase was not found in all herds studied. When the various studies conducted by Monsanto on somatic cell count were pooled, the results demonstrated no significant increase in somatic cell counts attributable to rbST treatment. The FDA did not consider somatic cell count from the standpoint of human safety. However, the agency did conclude from Monsanto's studies that rbST did not increase somatic cell count any more than factors such as lactation stage and cow parity. (Cow "parity" is a term that refers to whether and how many times a cow has given birth.)

4. Insulin-like growth factor

Insulin-like growth factor (IGF-1) is protein hormone whose production is regulated at least in part by somatotrophin. It has the same biochemical composition in humans and cows and is present in all milk, human saliva and human digestive juices. RbST increases the amount of IGF-1 in milk. IGF-1 is denatured in the process of making baby formula, but it is not destroyed by pasteurization.

At the time Monsanto filed its new drug application for Posilac, not much was known about the hormone. Defendants have done no long term studies on the effects of increased levels of IGF-1 on human health. However, Monsanto did conduct a two week study of rats administered IGF-1 either by gavage (forced feeding) or systemically by implanted pump. The study indicated no negative health impact on the esophagus, stomach and intestines of the rats from increased levels of IGF-1 administered orally.

The FDA considered IGF-1 when it evaluated the safety and effectiveness of Posilac and concluded that any increase in IGF-1 levels posed no significant health risks to consumers. This conclusion was based on the FDA's determinations that: 1) IGF-1 levels in milk from rbST-treated cows were not significantly higher than milk from untreated cows; 2) IGF-1 was not orally active; 3) even if IGF-1 from milk was absorbed intact, levels of IGF-1 in the human blood-stream are so much higher (100 to 1000 times) than the levels found in cows' milk that the addition of small amounts of IGF-1 from cows' milk would be physiologically insignificant; and 4) the manufacturing process for infant formula denatures almost all of the IGF-1 in cows' milk. The agency concluded that IGF-1 was unlikely to affect either the gastrointestinal track of adults or infants in general.

C. Labeling

The FDA determined that the risks of Posilac to dairy cows could be addressed adequately by labeling the drug itself. The Posilac label contains a "precautions and side-effects" section that lists the potential adverse effects on cows, including reduced pregnancy rates, cystic ovaries, disorders of the uterus, decreases in length of gestation, increased twinning rates, decreased calf birth weight, an increased risk of clinical and subclinical mastitis, digestive disorders and infection site reactions. In bold letters, the warning section informs the dairy farmer that "[u]se of Posilac should be preceded by implementation of a comprehensive and ongoing herd reproductive health program," and "[m]astitis management practices should be thoroughly evaluated prior to initiating use of Posilac."

The FDA considered whether special labeling should be required for milk and milk products derived from rbST-treated cows in addition to the drug's packaging label. In May 1993, the FDA's Veterinary Medicine and Food Advisory Committees held a meeting to address the issue. The committees received testimony from doctors, scientists, dietitians, dairy farmers, veterinarians, members of medical and scientific organizations and universities, dairy organizations and farm and agricultural groups. On the basis of testimony from the hearings and the data submitted by Monsanto, the FDA concluded that there is no significant difference between milk from cows treated with Posilac and milk from untreated cows. The agency concluded also that rbST had no significant effect on the organoleptic properties of milk or on the lactose, fat, protein, ash, phosphorous and vitamin content of milk. On February 3, 1994, the FDA issued interim guidance on voluntary labeling of milk from untreated cows, concluding that such milk cannot be labeled as "BST free" because BST occurs naturally in milk, but that farmers may label their milk "from cows not treated with rbST," if the statement is placed in proper context. 59 Fed.Reg. 6279-80 (February 10, 1994).

D. Environmental Issues

The FDA approved Monsanto's new drug application for Posilac without preparing an environmental impact statement, concluding that no statement was necessary because approval of Posilac would not have a significant impact on the environment within the meaning of the National Environmental Policy Act. The FDA based this conclusion on an environmental assessment prepared by Monsanto that addressed Posilac's potential impact on the environment.

The FDA assumes responsibility for the scope and content of the environmental assessment prepared by Monsanto. Agency officials have reviewed the information contained in that document. The environmental assessment refers to studies that evaluated the impact Posilac was likely to have on the dairy industry and concluded that its introduction to the market would not do fundamental damage to current structural trends. The environmental assessment addressed the possibility of shifts in land use patterns as a result of the use of Posilac, concluding that the drug would be unlikely to have any significant effect on land use. The assessment also addressed two alternatives to the current approval of Posilac: 1) additional controls on production; and 2) use of the drug, including labeling. Both alternatives were addressed only with regard to the environmental concerns surrounding the biotechnological aspects of the drug's preparation.

Neither the environmental assessment nor the finding of no significant impact addressed consumer health risks posed by increased IGF-1 levels and antibiotic residue levels in milk derived from rbST-treated cows, the impact of Posilac on the health of dairy cows or the potential risks to consumers caused by increases in IGF-1 and antibiotic residues. However, the agency's policy is not to reexamine issues under the National Environmental Policy Act already considered as part of its evaluation of a new animal drug application. The assessment failed to address the impact of Posilac on family dairy farms. Finally, the environmental assessment did not address possible alternatives of 1) approving the drug at lower doses; 2) approving Posilac only as a prescription drug; or 3) approving the drug for use only when accompanied by detailed instructions or educational seminars regarding methods to minimize health risks.

OPINION A. Standing

In their motion for summary judgment, defendants challenge plaintiffs' standing under both the Food, Drug, and Cosmetic Act and the National Environmental Policy Act. The constitutional requirements for standing derive from Article III's mandate that federal jurisdiction extends only to those situations in which plaintiffs can demonstrate a "case or controversy" between themselves and the defendant. Article III standing has three components: 1) an injury or threat of injury; 2) traceability of the injury to the defendant's conduct; and 3) the likelihood that a favorable decision would provide the plaintiff with a remedy. Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61, 112 S.Ct. 2130, 2136, 119 L.Ed.2d 351 (1992). Defendants dispute whether plaintiffs can satisfy the first element of the test. To do so, plaintiffs must allege injuries or threatened injuries that are "concrete and particularized" and show that they have "an actual stake in the outcome [of the case] that goes beyond intellectual or academic curiosity." Family Children's Ctr. v. School City, 13 F.3d 1052, 1058 (7th Cir. 1994) (quoting South East Lake View Neighbors v. Dept. of Housing and Urban Development, 685 F.2d 1027, 1033 (7th Cir. 1982)); see also Schmidling v. City of Chicago, 1 F.3d 494, 498 (7th Cir. 1993) (injury or threat "must be both real and immediate, not conjectural or `hypothetical'"), cert. denied, ___ U.S. ___, 114 S.Ct. 555, 126 L.Ed.2d 456 (1993). At the summary judgment stage, standing requires "a factual showing of perceptible harm." Lujan, 504 U.S. at 566, 112 S.Ct. at 2139. However, alleged injuries need not be particularly severe or costly: "even a minor or noneconomic injury will satisfy the strictures of Article III," Family Children's Ctr., 13 F.3d at 1058, and "even a small probability of injury is sufficient to create a case or controversy — to take a suit out of the category of the hypothetical. . . ." Village of Elk Grove Village v. Evans, 997 F.2d 328 (7th Cir. 1993).

1. Standing under the Food, Drug, and Cosmetic Act

Defendants raised their standing argument once before in this litigation in a motion to dismiss. In an opinion and order entered September 1, 1994, I concluded that standing could be premised on plaintiffs' allegations that they were unable to consume dairy products known to be free of milk from rbST-treated cows. Defendants do not challenge plaintiffs' alleged inability to consume milk products known to be free of milk from rbST-treated cows. Rather, they attack the minor premise of plaintiffs' allegation, that consuming dairy products derived from rbST treated cows is harmful. This argument misconstrues plaintiffs' rights under the Food, Drug, and Cosmetic Act.

The federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 301-394, is intended to protect American consumers from the risks of consuming unsafe or ineffective food and drugs and thus requires new drug applicants to demonstrate the safety and effectiveness of new drugs before they may be marketed. One aspect of plaintiffs' challenge to the FDA's approval of rbST is that the agency did not hold Monsanto to its burden under the act of demonstrating the safety of Posilac. If this allegation is true, plaintiffs' injury is their exposure to a potentially dangerous drug whose safety has not been demonstrated in accordance with the act. Cutler v. Kennedy, 475 F. Supp. 838, 848-49 (D.C. Cir. 1979), overruled on other grounds in Chaney v. Heckler, 718 F.2d 1174, 1188 n. 35 (D.C. Cir. 1983).

"[Defendants'] suggestion that [plaintiffs] lack standing because the injury they allege is speculative confuses the anticipated . . . consequences of the [agency action] with the . . . injury [plaintiffs] actually claim." Florida Audubon Society v. Bentsen, 54 F.3d 873, 879 (D.C. Cir. 1995). Contrary to defendants' theory, plaintiffs need not show that rbST is actually harmful in order to satisfy the requirements of standing. The Food, Drug, and Cosmetic Act creates legal rights or interests for consumers, "`the invasion of which creates standing, even though no injury would exist without the statute.'" Cutler, 475 F. Supp. at 848 (quoting Linda R.S. v. Richard D., 410 U.S. 614, 617 n. 3, 93 S.Ct. 1146, 1148 n. 3, 35 L.Ed.2d 536 (1973)). Because the act places the responsibility on the sponsor of the drug to demonstrate the drug's safety and directs the FDA to approve for marketing only those drugs whose safety has been demonstrated, any significant uncertainty regarding the drug's safety is a burden to be borne by the sponsor, not the consumer. If the FDA has failed to follow the dictates of the act, as plaintiffs allege, it has shifted the costs of uncertainty from the sponsor of the drug to the American consumer. This increased risk of potential harm that the consumer must bear is an injury in fact for standing purposes.

2. Standing under the National Environmental Policy Act

To have standing under the National Environmental Policy Act, 42 U.S.C. § 4321-4370d, plaintiffs must show that an agency's alleged noncompliance with the act has adversely affected them and that they are within the zone of interests protected by the act. Under the act, individuals are aggrieved by an agency's failure to prepare an environmental impact statement if they can show that the failure creates a reasonable risk that "`environmental impacts will be overlooked'" and that they have a "sufficient geographical nexus to the site of the challenged project that [they] may be expected to suffer whatever environmental consequences the project may have." Florida Audubon Society v. Bentsen, 54 F.3d at 877 (quoting City of Davis v. Coleman, 521 F.2d 661, 671 (9th Cir. 1975)).

Defendants contend that plaintiffs have failed to allege an injury cognizable under the National Environmental Policy Act and have made no showing of a sufficient nexus between the agency's decision not to prepare an environmental impact statement and the alleged harm. Defendants do not challenge plaintiffs' allegations that they are consumers of milk. Rather, defendants' challenge assumes that plaintiffs must prove that rbST is actually harmful. This assumption is flawed, however, because it collapses the standing inquiry into the merits of plaintiffs' claims.

Plaintiffs have introduced evidence tending to show that rbST creates an increased risk of mastitis that will increase dairy farmers' reliance on antibiotics in the treatment of their dairy cows. Plaintiffs offer additional evidence that over 50 drugs are used to treat infections in cows, only 4 of which are seriously regulated. (I note that in analyzing standing, I am able to consider evidence that is outside the administrative record and for that reason may not be considered when assessing plaintiffs' claims of violations of the Food, Drug, and Cosmetic. See infra, section C.) For standing purposes, I conclude that this evidence demonstrates a reasonable risk that 1) farmers will rely increasingly on illegal or otherwise unregulated drugs to treat the higher incidence of mastitis, which in turn will 2) increase the levels of drug residues that might show up in consumers' milk. Plaintiffs argue that there is a risk that increased exposure to drug residues may affect human bioimmune systems and the bacteria living in our digestive systems. Whether plaintiffs are likely to succeed on this claim does not determine whether they have standing under the National Environmental Policy Act. The evidence they have proffered demonstrates a reasonable risk of serious environmental harm caused by the proposed agency action that would clearly affect them as dairy product consumers, were it ever to come to fruition. Therefore, I conclude that the consumer plaintiffs satisfy the requirements for Article III standing under the National Environmental Policy Act.

B. Exhaustion

Defendants contend that plaintiffs have failed to exhaust their administrative remedies with respect to their labeling claim. Under FDA regulations, the secretary's decisions on new drug applications constitute final agency action for purposes of exhaustion. 21 C.F.R. §§ 10.25 and 10.45. Defendants do not argue that the agency's decision to approve Posilac was not final; rather, it appears that they consider the decision on labeling outside the scope of the agency's decision to approve Posilac, at least for exhaustion purposes. They assert that plaintiffs could have requested the agency to review its decision that labeling rbST-derived dairy products was not necessary by filing a citizen petition pursuant to 21 C.F.R. § 10.25 and they argue that an administrative ruling on such a petition would provide a reviewing court with a "better" record on the need for labeling, because the agency considered the issue only "informally" when it approved Posilac for marketing. Defendants note that none of the plaintiffs have filed a petition under 21 C.F.R. §§ 10.25 or 10.30 requesting the FDA to require labeling of food derived from rbST-treated cows.

Defendants' argument has some merit. Ordinarily, FDA review of new animal drug applications does not entail consideration whether food products derived from animals treated with the new drug should be labeled. It does not appear that any statute or regulation requires a determination of the question. Where new animal drug applications are concerned, usually the relevant focus is on proposed tolerance levels and withdrawal periods to assure that no residues from the animal drug make their way into animal products intended for human consumption. See 21 U.S.C. § 360b. Here, however, the FDA addressed the issue whether food products derived from cows treated with rbST should be labeled as such. After publishing notice in the federal register, 58 Fed.Reg. 21314 (April 20, 1993), the agency held a two-day meeting in May 1993 to receive oral and written testimony on the issue. Defts' Mem. Ex. 13 Vol. XVII. Andrew Kimbrell testified at the hearing on behalf of the Coalition for Responsible Technology, a Wisconsin group founded by plaintiff Stauber. On May 7, 1993, the agency's Food Advisory Committee and the Veterinary Medicine Advisory Committee deliberated the labeling issue. Defts' Mem.Ex. 13, Vol. XVIII. A final conclusion from the agency on the need to label food products does not appear in the conclusions section of the freedom of information summary, Defts' Exh. 1 Vol. I at 121-22. However, the agency's decision not to require labeling flowed a fortiori from its conclusion that it was unnecessary to establish a tolerance level or withdrawal time because there is no essential difference between milk derived from rbST-treated animals and regular milk. From a review of the administrative record, I conclude that the agency reached a final decision on the issue whether to require labeling of food derived from rbST-treated cows that is currently ripe for judicial review.

C. Standard of Review and Evidentiary Issues

The decision of the FDA to approve a drug is an informal agency action. Scott v. Food and Drug Administration, 728 F.2d 322, 324 (6th Cir. 1984); Upjohn Mfg. Co. v. Schweiker, 681 F.2d 480, 483 (6th Cir. 1982). Federal court review of informal agency action is limited in scope: it may be set aside only if the agency's determination is arbitrary and capricious, an abuse of discretion or otherwise not in accordance with the law. Upjohn Mfg. Co. v. Schweiker, 681 F.2d at 483; 5 U.S.C. § 706(2)(A). The arbitrary and capricious standard is highly deferential: even if a reviewing court disagrees with an agency's action, the court must uphold the action if the agency considered all relevant factors and the court can discern a rational basis for the agency's choice. Camp v. Pitts, 411 U.S. 138, 143, 93 S.Ct. 1241, 1244, 36 L.Ed.2d 106 (1973); Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 415-16, 91 S.Ct. 814, 823-24, 28 L.Ed.2d 136 (1971), overruled on other grounds in Califano v. Sanders, 430 U.S. 99, 97 S.Ct. 980, 51 L.Ed.2d 192 (1977). Moreover, when a decision goes to the core of an agency's expertise, generally the court must defer to the agency's more informed judgment. International Fabricare Institute v. U.S.E.P.A., 972 F.2d 384, 389 (D.C. Cir. 1992) ("The rationale for deference is particularly strong when the [agency] is evaluating scientific data within its technical expertise."); Tri-Bio Laboratories, Inc. v. United States, 836 F.2d 135, 142 (3d Cir. 1987) ("[I]n evaluating scientific evidence in the drug field, the FDA possesses an expertise entitled to respectful consideration by th[e] court") (citing Weinberger v. Bentex Pharmaceuticals, Inc., 412 U.S. 645, 653-54, 93 S.Ct. 2488, 2494, 37 L.Ed.2d 235 (1973)), cert. denied, 488 U.S. 818, 109 S.Ct. 57, 102 L.Ed.2d 35 (1988).

The Supreme Court has made clear that courts are precluded both from considering evidence that the agency never had a chance to review and from creating a record different from that considered by the agency. Camp v. Pitts, 411 U.S. at 142, 93 S.Ct. at 1244; see also Edwards v. United States Dept. of Justice, 43 F.3d 312, 314 (7th Cir. 1994); Cronin v. United States Dept. of Agriculture, 919 F.2d 439, 443-44 (7th Cir. 1990). The district court may not undertake a de novo review of the agency's decision and its factual underpinnings. Cronin, 919 F.2d at 444 ("only in an emergency should a reviewing court . . . conduct its own evidentiary hearing" in a review of agency action). This rule of evidence is derived from the arbitrary and capricious standard itself, which requires the court to ask not what it would have done if it had been in the agency's place, but rather whether the agency approached the problem in the right way, properly considered the evidence before it and issued a decision supported by the record in front of it. Bradley v. Weinberger, 483 F.2d 410, 414-15 (1st Cir. 1973) (district court erred in considering affidavits submitted by the parties in a case challenging an FDA determination on labeling; review was limited to the administrative record). This rule is fully applicable to federal claims challenging the FDA's approval of new drug applications. See Upjohn Mfg. Co. v. Schweiker, 681 F.2d at 483.

The agency review is like a trial in a district court: it is supposed to be the main event. Plaintiffs challenging an agency action do not advance their cause by submitting to the reviewing court evidence that could have been presented to the agency but never was. Just as in an appeal from a district court trial, the rule is that if a piece of evidence was never introduced, for purposes of appeal, it does not exist. The one exception is that a reviewing court may consider evidence that is relevant to determining whether the agency decision was arbitrary and capricious. See Overton Park, 401 U.S. at 420-21, 91 S.Ct. at 825-26 (exception to rule limiting judicial review of agency action to administrative record made where supplemental evidence would explain process by which agency made its determination).

Plaintiffs have submitted an assortment of evidence, including affidavits and other evidence tending to show that rbST may have an adverse impact on human health. Because none of this evidence was ever submitted to the FDA during its review of the Posilac application, none of it is admissible in this proceeding. Plaintiffs cannot ask this court to rely on opinions within the medical community regarding health risks posed by rbST without first establishing that those opinions were presented to the FDA before it granted approval of Posilac.

Plaintiffs did submit an affidavit from Dr. Richard Burroughs, who was employed by the FDA as a veterinary medical officer and was involved in the review process of Monsanto's application for Posilac. In his affidavit, Dr. Burroughs points out what he considers to be key flaws in the agency's review of the drug application and errors in Monsanto's safety testing of the drug. The information in Burroughs's affidavit might have been admissible as evidence tending to explain the record actually before the agency or the agency's decision-making process. Overton Park, 401 U.S. at 420-21, 91 S.Ct. at 825-26. However, in proposing findings of fact in support of their motion for summary judgment, plaintiffs rely on the Burroughs affidavit only once, in support of the proposition that "[o]ne of the major concerns is that rbST causes many potentially serious health problems in cows." I have accepted this fact as undisputed.

With the preliminary issues resolved, I turn now to the merits of the case.

D. The FDA's Approval of Posilac

The FDA is the federal agency responsible for assuring that new animal drugs are safe and effective for their intended use. Before approving a new animal drug application, the FDA requires a new drug sponsor to demonstrate that the food products from the subject animals will be safe for human consumption, that the drug is safe and effective for the animals in question and that the manufacture and use of the drug will not harm the environment. Approval proceeds in two steps. First, the drug sponsor seeks approval of its investigative new animal drug application, which outlines how the sponsor will conduct the necessary research. This research covers human food safety, animal safety and drug efficacy and includes clinical and field trials, laboratory studies, on-site inspections and other investigative methods. Second, upon completion of the studies, the sponsor submits a new animal drug application that employs investigational studies to establish the safety and efficacy of the drug.

Throughout the application process, the burden is on the applicant or drug sponsor to show that the new animal drug is both effective and safe under the proposed conditions of use. 21 U.S.C. § 360b; 21 C.F.R. § 514.1(b)(8). Effectiveness must be demonstrated on the basis of "substantial evidence" that the drug "will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, or suggested in the proposed labeling thereof." 21 U.S.C. § 360b(d)(3). The Food, Drug, and Cosmetic Act does not indicate the standard an applicant must meet to demonstrate a new drug's safety or the evidence upon which the FDA must base its safety determination. However, the act requires the FDA to consider four specific factors when assessing safety: 1) the likelihood that the drug or a substance formed in food because of the drug will be consumed; 2) the cumulative effect that the drug will be likely to have on man or other animals; 3) safety factors that experts consider appropriate for extrapolating from animal experimentation data; and 4) whether it is likely that the conditions of use proposed or suggested in the labeling will be followed. 21 U.S.C. § 360b(d)(2); 21 C.F.R. § 514.111(a)(4). The FDA is required to take into account the drug's effect on the health of the target animal. 21 U.S.C. § 321(u). A new drug application must contain "full reports of adequate tests by all methods reasonably applicable to show whether or not the drug is safe for use under the conditions prescribed, recommended, or suggested in the proposed labeling." 21 U.S.C. § 360b(d)(1)(B). Such testing includes preclinical acute, subacute and chronic toxicity testing, clinical studies and filed investigations. 21 C.F.R. § 514.1(b)(8)(iii).

Plaintiffs contend that the FDA should not have approved Posilac because of safety concerns. For the reasons set out above in section C, supra, none of plaintiffs' proffered evidence indicating that rbST poses a danger to human health may be considered because it is outside the administrative record. The focus narrows, therefore, to plaintiffs' arguments that are grounded properly on the administrative record. Despite the limited amount of evidentiary support plaintiffs have collected, they continue to maintain that Monsanto failed to meet its burden under the Food, Drug, and Cosmetic Act. To survive summary judgment, plaintiffs must raise a genuine issue on the reasonableness of the FDA's conclusion that Monsanto met its burden of showing the safety of Posilac.

In identifying what they consider to be arbitrary and capricious FDA determinations, plaintiffs begin with the assertion that the Posilac label does not adequately counteract the health risks to cows associated with the drug's use. The FDA approved Posilac, notwithstanding its admitted risks of side effects to cows, on the theory that this safety risk is "manageable" if farmers adopt herd management techniques and ongoing herd reproductive health programs as the Posilac label recommends. Plaintiffs contend that this manageable risk criterion is arbitrary and capricious because the drug's label is vague and farmers will not know how to interpret it. Defendants respond by noting that mastitis is not a new problem to dairy farms and that the increased incidence of mastitis caused by Posilac is not significantly great (0.1 per cow per year). It was on these two findings that the agency grounded its conclusion that the risk was "manageable," and in any event not so great as to warrant denying approval of the drug. Defendants note also that under the Food, Drug, and Cosmetic Act, neither the sponsors of new animal drugs nor the FDA is held to a "zero risk" standard.

Although plaintiffs raise valid concerns about the manageable risk approach, I cannot conclude that the agency's decision was arbitrary and capricious. First, nothing in the record indicates that the agency overlooked some aspect of cow health when reviewing the Posilac application. To the contrary, the record demonstrates that the agency evaluated the target animal's safety as it was required to do under 21 U.S.C. § 321(u), considering each and every one of the known side effects associated with Posilac. Second, I cannot say it was irrational for the agency to conclude that the increased risk of mastitis caused by Posilac is not sufficiently great to warrant denying approval of the drug. This conclusion was based on the agency's determination that the risk of mastitis caused by Posilac is much less than the risk caused by other variables such as seasonal change. Finally, although mastitis is a serious health concern, it is one familiar to dairy farmers, who know how to detect and treat it. I add, however, that in the future it would be helpful to reviewing courts for the FDA to set out the factors it looks at in determining whether a particular risk is a manageable one.

Plaintiffs argue that defendants' decision to approve Posilac was arbitrary and capricious because defendants did not adequately consider the risk of higher levels of antibiotic drug residues in milk that could have an adverse impact on human health because of the greater resistance of certain bacteria, particularly human digestive bacteria. The FDA determined that the risk of exposure to increased amounts of antibiotics was slim because, among other things, the risk of increased mastitis is slight and the level of antibiotic drug residues in milk is highly regulated. Plaintiffs characterize this decision as arbitrary and capricious because many of the drugs used to treat bovine infections are not tested routinely under the current regulatory scheme. Defendants counter that the four antibiotics most commonly screened are also the ones most commonly used to treat bovine infection. The controversy boils down to whether it was proper for the FDA to rely on the current regulatory mechanism to assure that drug residues will be kept out of the milk supply.

Although plaintiffs have raised valid concerns about the adequacy of the current regulatory scheme to assure the purity of the nation's milk supply, they have not shown the agency's reliance on that scheme to be arbitrary and capricious. First and most important, the agency determined that the increased risk of mastitis was not overly great and on that basis concluded that the rise in the use of antibiotics would not be overly great either. Given these conclusions, it was rational for the agency to rest a determination of safety on the current milk purity regulatory scheme especially because milk is tested for the drugs most commonly used to treat mastitis, the beta-lactams. Although I understand plaintiffs' concerns, I conclude that the agency considered the relevant factors and that the record contains support for its determination.

Third, plaintiffs are dissatisfied with the FDA's exploration of the potential risks associated with an increased level of IGF-1 in milk from cows treated with Posilac. The FDA concluded that an increased level of IGF-1 in milk presented no significant human health concerns. This conclusion was based on Monsanto's two-week rat study indicating no adverse effects from oral ingestion of IGF-1 and on other evidence indicating that IGF-1 levels rise only slightly when the producing cow is treated with rbST, that IGF-1 exists in the human blood stream at levels much higher than those appearing in milk and that IGF-1 is denatured when milk is boiled. Plaintiffs are correct that no long term studies have been done, but they have presented no admissible evidence that would show either that longer studies were required or that the two-week rat study was not scientifically adequate to permit the FDA to conclude that IGF-1 poses no significant risk to human health. Without any admissible evidence indicating that the two-week study did not support the conclusions that the agency drew from it, I cannot conclude that the agency's decision lacked a rational basis in the record.

E. The FDA's Decision Not to Require Labeling of Products Derived from rbST-treated Cows

Under the Food, Drug, and Cosmetic Act, food is deemed misbranded if its labeling is "false or misleading in any particular." 21 U.S.C. § 343(a)(1). 21 U.S.C. § 321(n) provides further that:

If an article is alleged to be misbranded because the labeling or advertising is misleading, then in determining whether the labeling or advertising is misleading there shall be taken into account (among other things) not only representations made or suggested by statement, word, design, device, or any combination thereof, but also the extent to which the labeling or advertising fails to reveal facts material in the light of such representations or material with respect to consequences which may result from the use of the article to which the labeling or advertising relates under the conditions of use prescribed in the labeling or advertising thereof or under such conditions of use as are customary or usual.

Plaintiffs contend that food products derived from rbST-treated cows must be labeled as such if the products are to comply with these two provisions. They argue that milk derived from rbST-treated cows differs organoleptically from ordinary milk in several respects and that these differences are "material facts" requiring labeling. In addition, they argue that there is widespread consumer desire for mandatory labeling of rbST-derived milk and that such a degree of demand is also a "material fact" requiring labeling.

An organoleptic difference is one capable of being detected by a human sense organ. See Webster's Collegiate Dictionary 953 (1991). In a different context from the one presented here, the FDA has stated:

Information disclosing differences in performance characteristics (e.g., physical properties, flavor characteristics, functional properties and shelf life) is a material fact under section 201(n) of the act because it bears on the consequence of the use of the article. Accordingly, this information must be communicated to the consumer on the product label, or labeling would be misleading, and the product would be misbranded under section 403(a) of the act.

Regarding widespread consumer demand, plaintiffs are incorrect in their assertion that by itself consumer opinion could suffice to require labeling. The FDA does consider consumer opinion relevant when determining whether a label is required to disclose a material fact, but a factual predicate to the requirement of labeling is a determination that a product differs materially from the type of product it purports to be. If there is a difference, and consumers would likely want to know about the difference, then labeling is appropriate. If, however, the product does not differ in any significant way from what it purports to be, then it would be misbranding to label the product as different, even if consumers misperceived the product as different. In the absence of evidence of a material difference between rbST-derived milk and ordinary milk, the use of consumer demand as the rationale for labeling would violate the Food, Drug, and Cosmetic Act. Because plaintiffs have not presented any evidence demonstrating organoleptic differences between regular and rbST-derived milk or of any harmful effects of rbST on consumers, they have failed to carry their burden of demonstrating the arbitrary and capricious nature of the agency's determination.

F. The National Environmental Policy Act

The National Environmental Policy Act, 42 U.S.C. § 4321-4370d, requires federal agencies to prepare an environmental impact statement if the agency plans to undertake a "major" action "significantly affecting the quality of the human environment." 42 U.S.C. § 4332(2)(c). Congress requires that these environmental impact statements include a discussion of the environmental impact of the proposed action, any adverse environmental harms that cannot be avoided if the project goes forward, alternatives to the proposed action, the relationship between "local short-term uses of man's environment and the maintenance and enhancement of long-term productivity," and any "irreversible and irretrievable commitments of resources which would be involved in the proposed action should it be implemented." Id. An "environmental assessment" is a public document that "briefly provide[s] sufficient evidence and analysis for determining whether to prepare an environmental impact statement." 40 C.F.R. § 1508.9. If, on the basis of the environmental assessment, the agency determines that the proposed action will not have a significant effect on the quality of the human environment, then the agency prepares a "finding of no significant impact" in lieu of an environmental impact statement. 40 C.F.R. § 1501.4(e). A finding of no significant impact is an agency document explaining briefly the agency's reasons for concluding that the governmental action will not have a significant effect on the environment and that preparation of an environmental impact statement is unnecessary. 40 C.F.R. § 1509.13.

Plaintiffs contend that the FDA failed to comply with the National Environmental Policy Act when it determined that an environmental impact statement was unnecessary. In particular, plaintiffs contend that the decision not to prepare and environmental impact statement was arbitrary and capricious in light of the failure to address Posilac's socioeconomic effects on the dairy farmer and Posilac's likely effects on human and bovine health in either the environmental assessment or the finding of no significant impact. In addition, plaintiffs challenge the failure to discuss feasible alternatives in the environmental assessment and the finding of no significant impact.

Regarding socioeconomic effects, the regulations promulgated under the National Environmental Policy Act provide that "economic or social effects are not intended by themselves to require preparation of the environmental impact statement." 40 C.F.R. § 1508.14. Plaintiffs are wrong when they assert that Posilac's socioeconomic effects on the dairy farmer require the preparation of an environmental impact statement. See Missouri Coalition for the Environment v. Corps of Engineers of United States Army, 866 F.2d 1025, 1031-32 (8th Cir.), cert. denied, 493 U.S. 820, 110 S.Ct. 76, 107 L.Ed.2d 42 (1989). It is true that an environmental impact statement must discuss economic or social effects of the proposed action to the extent those effects interrelate to its natural or physical environmental effects, but the regulations do not contemplate an independent consideration of socioeconomic effects when there is no determination that the proposed agency activity will significantly effect the environment. Id.

Plaintiffs do not argue that the socioeconomic effects of Posilac would cause secondary environmental effects to such a significant degree as to require the preparation of an environmental impact statement. Rather, they contend merely that Posilac will cause both environmental and socioeconomic harms and that for this reason an environmental impact statement was necessary. This assertion is insufficient to satisfy the "interrelatedness" requirement of § 1508:14. I read 40 C.F.R. § 1508.14 to mean that it is only after an agency determines that the socioeconomic impact of the proposed agency action is likely to cause environmental harms itself that the agency needs to discuss the socioeconomic effects in the environmental impact statement. See Breckinridge v. Rumsfeld, 537 F.2d 864, 866 (6th Cir. 1976) (accord), cert. denied, 429 U.S. 1061, 97 S.Ct. 785, 50 L.Ed.2d 777 (1977). This reading fully comports with the plain language of the regulation and the FDA's stand on the issue. (The FDA's policy is not to include purely socioeconomic considerations in its environmental assessments. FDA, National Environmental Policy Act; Policies and Procedures; Final Rule, 50 Fed.Reg. 16636, 16648 (Apr. 26, 1985).) Because defendants were not required to include a discussion of socioeconomic effects in their environmental assessment, it was not arbitrary and capricious to omit consideration of those nonenvironmental effects from the environmental assessment.

The next question is whether, to comply with the National Environmental Policy Act, the FDA was required to take an independent and second look at the effects Posilac might have on the health of cows and consumers or whether the agency could rely on the safety evaluations it made pursuant to its statutory duties under the Food, Drug, and Cosmetic Act. Plaintiffs do not seem to argue that the analysis of the human and bovine health issues required under the two statutes differ in some way or that the analysis required under the Food, Drug, and Cosmetic Act is somehow less rigorous than that required under National Environmental Policy Act. Rather, plaintiffs' argument appears to be simply that the agency must review the issues twice to satisfy the requirements of both statutes.

40 C.F.R. § 1506.4 provides that "[a]ny environmental document in compliance with [the National Environmental Policy Act] may be combined with any other agency document to reduce duplication and paperwork." In Citizens for Safe Power v. Nuclear Regulatory Comm'n, 524 F.2d 1291 (D.C. Cir. 1975), the issue was whether the NRC had to consider separately and formally the risks and benefits associated with the grant of a nuclear power plant license under both the National Environmental Policy Act and the Atomic Energy Act. The court answered the question in the negative, explaining its position as follows:

[I]t would be pointless, and a waste of agency resources, to require the [Atomic Energy Commission] to reapply efforts that have already gone into its basic health and safety regulations . . . in the absence of some evidence that a particular facility presents risks outside the parameters of the original rule making. And in evaluating the sufficiency of agency determinations in particular cases it would be stultifying formalism to disregard the whole record and test the [Atomic Energy Commission] compliance by only the evidence received at so-called `health and safety' hearings; or [National Environmental Policy Act] compliance only on the basis of socalled `environmental' hearings.

Plaintiffs contend that at a minimum, the agency should have been required to incorporate by reference its health and safety findings into the environmental assessment and the finding of no significant impact. This argument is not without merit. Such incorporation of the health and safety data by reference in the environmental assessment and finding of no significant impact would provide an interested party (or reviewing court) with a complete picture of all analysis bearing on the agency's obligations under the National Environmental Policy Act.

Plaintiffs' argument fails, however, for two interrelated reasons. First, at bottom, their argument is a purely informational one, which fails because they allege no harm stemming from the asserted procedural defect. Cf. Foundation for Economic Trends v. Lyng, 943 F.2d 79, 84-85 (D.C. Cir. 1991) (informational injury alone is not enough to satisfy standing requirements under NEPA). Second, environmental assessments and findings of no significant impact are informal agency actions subject to judicial review under the arbitrary and capricious standard. As the Court of Appeals from the Seventh Circuit has stated, "a reviewing court may — without violating the rule of SEC v. Chenery Corp., 332 U.S. 194, 67 S.Ct. 1575, 91 L.Ed. 1995 (1947), against the court's supplying a rationale for the agency's decision — `uphold a decision of less than ideal clarity if the agency's path may be reasonably discerned.'" Cronin v. United States Dept. of Agriculture, 919 F.2d at 442 (quoting Bowman Transportation, Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 286, 95 S.Ct. 438, 442, 42 L.Ed.2d 447 (1974)). Judicial review of a decision of an agency not to prepare an environmental impact statement requires a court to decide whether the agency's determination is "sustainable on the [entire] administrative record made," and is not limited to a review of the agency's statements regarding the lack of a need for an environmental impact statement. Missouri Coalition for the Environment v. Corps of Engineers, 866 F.2d 1025 at 1031. Because the record reveals that defendants have done health and safety analyses, their failure to reference those analyses in the environmental assessment or the finding of no significant impact does not undermine the adequacy of the conclusions reached in those documents.

The final issue plaintiffs raise is the omission of any discussion of feasible alternatives from the environmental assessment and the finding of no significant impact. Plaintiffs argue that the agency should have considered the possibilities of a delay on approving the Posilac application until further studies on health and safety could be conducted, a lower recommended dosage to the cows, or approval of Posilac only as a prescription drug. (Plaintiffs proposed as fact that the environmental assessment did not address the alternative of approving the drug for marketing only when accompanied by detailed instructions or educational seminars regarding methods to minimize health risks, but they have not pursued this argument in their briefs.) The first alternative, a delay in approving the drug, was suggested to the agency. During the review process, the Foundation on Economic Trends, a non-profit consumer organization and a former plaintiff in this action, suggested that approval of the drug be postponed until further safety research could occur. However, plaintiffs have failed to offer evidence that the other two alternatives (a lower recommended dosage or prescription drug status) were ever presented to the agency for consideration. If an alternative is not presented to the agency for consideration, the agency cannot be attacked in federal court for not considering it. Upper West Fork River Watershed Assoc. v. Corps of Engineers, 414 F. Supp. 908, 920 (N.D.W.Va. 1976) (when challenging agency's consideration of alternatives, plaintiffs may not rest upon naked conclusion that alternatives should have been considered). Plaintiffs have not shown that the agency was made aware of the two possible alternatives relating to lower dosages or making Posilac available only by prescription. Therefore, they may not challenge the agency's failure to consider them.

Even if I assume that plaintiffs made a presentation of their alternative of delayed approval pending further research into health and safety that was adequate to require the agency to consider the proposal, see Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978) (proponents of alternatives must make a preliminary showing that proposed alternative merits review before agency required to consider alternative during agency proceedings), plaintiffs would still not meet their burden of showing that an environmental impact statement should have been prepared. When an agency determines that a proposed action poses a significant risk of harm to the environment, the agency must consider alternatives to the proposed action. See 42 U.S.C. § 4332(2)(c)(iii); 40 C.F.R. § 1502.14. However, if the agency determines that a proposed action will not significantly effect the environment, it is required to consider alternatives only when the proposal involves "unresolved conflicts concerning alternative uses of available resources." 42 U.S.C. § 4332(2)(E); River Road Alliance, Inc. v. Corps of Engineers of United States Army, 764 F.2d 445, 452 (7th Cir. 1985), cert. denied, 475 U.S. 1055, 106 S.Ct. 1283, 89 L.Ed.2d 590 (1986). In this situation, I cannot perceive any unresolved conflict concerning uses of available resources. Plaintiffs' proposed alternative involves only a delay in approving the drug until further tests have been conducted. This situation does not involve an "available resource" as the term is used in § 4332(2)(E).

ORDER

IT IS ORDERED that the motions for summary judgment of defendants Donna Shalala, Secretary of Health and Human Services, and David Kessler, Commissioner of the Food and Drug Administration, and intervenor defendant Monsanto Company's motions for summary judgment are GRANTED. Plaintiffs' motion for summary judgment is DENIED. The clerk of court is directed to enter judgment for defendants and the intervenor defendant and close this case.