EPA Reverses Ban on Testing Pesticides on Human Subjects

Environmental Protection Agency Administrator Christine Whitman is expected to decide soon that her agency will end its ban on accepting data from studies that intentionally dose human volunteers with toxic pesticides. Such studies have been widely condemned as unethical and unscientific. Testing pesticides on human subjects violates international agreements, including the Nuremberg Code adopted after World War II, and federal rules.

Ironically, EPA is changing its policy partly in response to pressure by a German pesticide manufacturer, Bayer, which in August submitted a human study of azinphos methyl, an insecticide derived from World War II nerve gases. Bayer scientists were among those who conducted human experiments in the Nazi concentration camps when the company was a subsidiary of I.G. Farben, the infamous manufacturer of Nazi death camp gas. Bayer submitted the azinphos methyl test in an effort to reduce otherwise applicable safety requirements for the chemical.

In late October, EPA made a tentative decision not to use the Bayer study because it was scientifically unnecessary to consider it, not because of legal and ethical questions it raised. However, there are more than 10 other human pesticide studies awaiting EPA consideration. If EPA ultimately accepts the Bayer study or any of the other human studies before it, the floodgates could open for other human pesticide studies.

Past EPA Policy, and 1988 Ban on Use of Human Pesticide Studies
Since the EPA's early years, the idea of testing pesticides on human subjects has been controversial. In 1975, when some EPA staff members suggested the agency should encourage human tests for certain pesticides, the studies were not completed due to ethical concerns. Nixon-Ford EPA Administrator Russell Train later said he was "shocked and appalled" by the proposal, and that "the thing should have been shut off at the very start without even dignifying it by a referral to an advisory board" (Washington Post, June 23, 1977). More than a decade later, 20 EPA scientists objected to a proposal to use human tests done by the Nazis in the 1940s to assess the health risks of certain chemicals. The first Bush administration's EPA head, Lee Thomas, prohibited the agency from considering the Nazi study data (Washington Post, March 24, 1988).

More recently, the Clinton-era EPA banned use of human tests of pesticides in making regulatory decisions. In 1998, EPA adopted an interim policy that human studies would not be used to establish regulatory levels (using "no observable adverse effect levels," or NOAELs) for pesticides. EPA stated in July 1998 that it was "concerned about the possibility of increased human testing as a way to potentially avoid some of the protections that the Food Quality Protection Act establishes" (New York Times, July 28, 1998; Pesticide & Toxic Chemical News, July 30, 1998; Chemical Week, August 2, 2000). EPA reiterated that policy in 2000 (Washington Post, June 28, 2000).

Bayer's Human Study Likely Will Force an EPA Decision on Human Testing
In a consent decree reached with NRDC, the United Farm Workers and other groups, EPA agreed to make a decision on the toxicity of azinphos methyl by October 31, 2001. In August 2001, Bayer, the only manufacturer of azinphos methyl, submitted to EPA a human study that its contractor conducted in Scotland in 1998. EPA wound up making its decision on the toxicity of the pesticide based on animal tests. It did not reject Bayer's human test, but said that it did not affect its decision, which was to continue to allow most uses.

Bayer's World War II History of Human Testing and Poison Gas Production
Bayer, the company pressuring EPA to reverse its policy on human testing, is the successor to I.G. Farben, the manufacturer of Zylkon B, the Nazi death camp gas. Bayer chemists also used death camp inmates to conduct human experiments on chemicals. Azinphos methyl is a Bayer-invented chemical cousin to organophosphate nerve gases developed during World War II.

The American Chemical Society summarized this troubling period of Bayer's history:

As an I.G. Farben subsidiary during World War II, Bayer entered the darkest period in its history. Recently publicized evidence suggests that I.G. Farben furnished experimental Bayer drugs for tests on concentration camp prisoners. The company stationed scientists at the camps to oversee human research, and provided at least a portion of the funds that supported the horrific experiments of Joseph Mengele, the notorious Nazi "Angel of Death." I.G. Farben produced the Zyklon B gas used in countless executions, and the company reaped handsome profits from factories set up near the Auschwitz and Maidanek prison camps to benefit from ready access to slave labor.

During the Nuremberg trials of 1947, 13 I.G. Farben executives were convicted of war crimes and crimes against humanity and received prison sentences of up to eight years. The American Tribunal, however, concluded that I.G. Farben management had not wittingly participated in German hostilities. Nevertheless, in 1950, the Allied High Commission ordered the dissolution of I.G. Farben into 12 successor companies as part of a program to dismantle German industry. Bayer was reincorporated as Farbenfabriken Bayer Aktiengesell schaft in 1951. 1

Controversy Over Human Testing
Many physicians, ethicists and environmentalists believe intentional non-therapeutic dosing of humans with toxic pesticides is unethical, and EPA should maintain its policy of refusing to consider such testing for regulatory purposes. They believe that:

  • intentional dosing of humans with pesticides is unacceptable, particularly when it is done to advance industry interests, not to benefit the test subjects or public health;
  • such human pesticide testing has been conducted in an effort to weaken standards that would otherwise be applicable if animal tests were relied upon;
  • such testing cannot be done on children because of their higher vulnerability to pesticides, so human testing completed to date is scientifically irrelevant for infants and children, the populations of primary statutory concern under the Food Quality Protection Act (FQPA).

EPA Science Advisory Board-Scientific Advisory Panel Advice on Human Tests
EPA convened a science advisory board-science advisory panel of ethicists and scientists to advise the agency on human studies in 1998. A major split developed in the committee; the majority recommended in September 2000 that no human tests be used, except to improve health protection rather than promote industry interests, and not to set NOAELs. The minority said that human tests of pesticides are unscientific, unsafe and inherently unethical.

Substantial bias questions cast a cloud over the science advisory board's advice. For example, committee cochairman Mark Utell of the University of Rochester conducts human testing of air pollutants. According to a 1996 report by the New York State Health Commissioner, the program Utell heads was responsible for accidentally killing a volunteer in 1996 due to complications arising from human environmental research; the State Health Commissioner found the program had "insufficient safeguards to adequately minimize risks and protect individuals" and required corrective action. 2

The Nuremberg Code, International Agreements & Rules on Human Testing
Even assuming that human testing of pesticides could be theoretically acceptable in certain very limited circumstances, pesticide tests completed to-date clearly violate the testing standards established by the Nuremberg Code, Helsinki Declaration, the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA), the Common Rule, and the recommendations of EPA's science advisory board-science advisory panel.

EPA's key sources of moral and scientific guidance on the ethics of human testing include:

  1. The Nuremberg Code, which was adopted by U.S. judges in the wake of the Nuremberg "doctor trials" of Nazis after World War II. U.S. courts and others have relied on the code to establish minimum ethical standards for human testing.
  2. The Helsinki Declaration, which as adopted by the World Medical Association, governs medical testing of human subjects, but also is used to judge the ethics of other human testing.
  3. FIFRA, which prohibits any person "to use any pesticide in tests on human beings unless such human beings (i) are fully informed of the nature and purposes of the tests and of any physical and mental health consequences which are reasonably foreseeable therefrom, and (ii) freely volunteer to participate in the test." FIFRA §12(a)(2)(P).
  4. The Common Rule, which establishes ethical and scientific rules for human testing. It was issued by HHS, and adopted by EPA in 1991. 45 C.F.R. part 46; 40 C.F.R. part 26.
  5. EPA Scientific Advisory Panel-Science Advisory Board majority report and accompanying minority report. The minority found that intentional human dosing with pesticides is always unethical. The majority said bad science is always unethical, and it human testing is unethical if intended to weaken health protection, but could be ethical if intended to strengthen protection, subject to "rigorous to severe" ethical and scientific controls, and "active and aggressive scrutiny" by EPA.

Bayer's Azinphos Methyl Study Violated Basic Codes and Rules
All human tests of pesticides to date that NRDC has obtained through the Freedom of Information Act violate the principles codified by the statements, rules and regulations listed above. Bayer's azinphos methyl study presented to EPA in August 2001 is no exception:

  1. The purpose of experiment was not to improve health protection, but rather to reduce safety margins to advance interests of industry and agriculture. [Violates SAP-SAB Maj., letter at 2 ¶d;]
  2. There is no "reasonable likelihood" that the subpopulations serving as subjects would benefit from the study results. The study offered no benefit to subjects. [Violates Helsinki, ¶B-19]
  3. The experiment does not "yield fruitful results for the good of society, unprocurable by other methods or means of study," because the study would reduce protection [Violates Nuremberg Code, ¶2]
  4. The "risks to subjects [are not] reasonable in relation to anticipated benefits, if any, to subjects, and the importance of the knowledge that may reasonably be expected," since there was neither a benefit to subjects, nor important "expected knowledge." [Violates Common Rule §46.111(a)(2)]
  5. The study does not yield scientifically and statistically valid, relevant data. "Bad science is always unethical; research protocols that are fundamentally flawed, such as those with sample sizes inadequate to support reasonable inferences about the matter in question, are unjustifiable." [Violates SAP-SAB Majority, Letter at 2¶c]

    Bayer's azinphos methyl study was scientifically invalid because:

    • It tested only eight adult males, whereas a test of more than 2,500 people is needed to yield statistically valid results for certain effects [SAP-SAB Majority App. B; Minority Rep];
    • It did not yield results relevant to children or women or non-healthy adult males;
    • There were at least 67 "adverse events," including symptoms often associated with organophosphate poisoning; all eight dosed subjects suffered from such "events" (including chronic headaches, nausea, abdominal pain, etc.). All "events" were attributed, without detailed medical explanation, to a "virus" or the "ward environment," even though most of these events occurred in the dosed group, and two of for placebo subjects suffered no such events.
  6. Bayer conducted the study to try to establish a less protective no adverse effects level. The SAB-SAP "in general, would not support human experimentation primarily to determine a no adverse effects level." [Violates SAP-SAB majority report at 11, ¶3.1.1]
  7. The study of adults did not yield results of scientific relevance to children, the population of legal relevance under the Food Quality Protection Act. Since children are more vulnerable, and cannot ethically be tested, the results are scientifically and legally irrelevant. [Violates Nuremberg Code ¶2 & Helsinki ¶B-19 requirement for scientific benefit to test; SAP-SAB minority says all human tests of adults are irrelevant to children, p. C2] The Nuremberg Code and other applicable principles prohibit non-therapeutic testing of toxic chemicals on children. [Grimes v. Kennedy Krieger Inst., No. 128, Md. Ct. of App., slip op. at 47-52, 77-88 (August 16, 2001), citing court decisions].
  8. Bayer apparently tested poor subjects who likely suffered from economic hardship and likely did not understand the risks, and therefore did not give truly informed consent.
  9. The company Bayer conducted the azinphos methyl test on eight "volunteers.," but did not clearly document they gave their fully informed consent or that they freely volunteered. The company hospitalized them for a month, dosed them with a known poison, took repeated blood and urine tests, and paid them £1500, which suggests the participants were economically disadvantaged. The company did not provide evidence that the subjects understood the consent form or "volunteer information" sheet. According to the Nuremberg Code, a subject "should be so situated as to be able to exercise free power of choice, without the intervention of any element of force, fraud, deceit, duress, over-reaching, or other ulterior form of constraint or coercion; and should have sufficient knowledge and comprehension of the elements of the subject matter involved as to enable him to make an understanding and enlightened decision." [Nuremberg Code ¶1]. Consent must be "fully informed" and subjects must "freely volunteer." [FIFRA §12(a)(2)(P)][Common Rule §46.116] [Helsinki B 20-22] "Some research populations are vulnerable and need special protection. The needs of the economically… disadvantaged must be recognized." [Helsinki ¶A-8] The lack of clear documentation of fully informed and voluntary consent thus violates several principles of applicable codes and rules.
  10. Moreover, the evidence available indicates that Bayer did not obtain fully informed consent because the study's subjects lacked knowledge and comprehension of the goals, risks, and known or possible hazards. The azinphos methyl study fails this test in several ways:

    • The consent form mentions no specific adverse effects that are possible. None of the long-term effects associated with organophosphates is mentioned on any form Bayer gave to the subjects.
    • The "volunteer information" (VI) form says that previous studies show azinphos has "no side effects other than the intended effect" of cholinesterase inhibition (it is unclear whether subjects were fully briefed on the VI form). Although the VI form mentions short-term acute effects that could occur, the form says they are "not anticipated" in the study.
    • The actual consent form signed by subjects misleadingly states that the study may be disclosed "for medicines in the UK and elsewhere" when the company had no intention to use the study to develop medicines. While the VI form refers to azinphos as a pesticide and says the goal of the study is to set safe levels, this form was not signed by volunteers and it is unclear whether the company fully briefed them about it.
    • Bayer never told the subjects the study's true goal, which was to establish rules for azinphos methyl that likely would have been weaker than those imposed if the company had used animal studies.

    Taken together, it is clear Bayer did not give the study's subjects a full or fair explanation of the study's main goal -- reduce regulatory protections that would otherwise have been imposed -- nor did the company tell them about "all inconveniences and hazards reasonably to be expected, and the effects upon his health or person which may possibly come from his participation in the experiment." [Nuremberg Code ¶1][ FIFRA §12(a)(2)(P)] [Helsinki ¶B-22][Common Rule §46.116(a)(2)]

  11. Bayer did not adequately inform the subjects of "any possible conflicts of interest and institutional affiliations of the researcher. Bayer apparently did not inform the azinphos study subjects that the researchers conducting the study were Bayer contractors, or that Bayer stood to gain financially from weaker health protections that could result from use of the study. [Violates Helsinki ¶B-22]
  12. Subjects did not have the right to revoke their informed consent at any time without any penalty or loss of benefit. The azinphos methyl consent form included wording that amounted to not-so-subtle pressure not to withdraw, indicating that while the subject could withdraw from the study, if he did so for non-"medical" reasons, there was no assurance that he would receive any compensation, no matter how long he had participated. Withdrawal without loss of benefit is central to true consent. [¶Common Rule §46.116(a)(8)][Helsinki ¶B-22][Nuremberg Code ¶9][FIFRA §12(a)(2)(P)]
  13. Informed consent was absent because the consent form included exculpatory language or a waiver, or the appearance of a waiver, of the subject's legal rights. The azinphos methyl consent form required subjects to agree that if they were injured, rather than suing, they would submit any dispute to an arbitrator (who had to be agreed to by the company), that damages for injury would be limited to "the amount of damages commonly awarded for similar injuries in an English court," and that the law that would govern any dispute was U.K. law exclusively in English courts, which tend to be defendant-friendly. (For example, under the English Rule unsuccessful plaintiffs are liable to defendants for all costs.) This apparently means subjects waived other tort remedies, or redress in U.S., German, or other courts. [Violates Common Rule §46.116]
  14. The Ethical Review Committee's (or "Institutional Review Board's" (IRB)) qualifications, possible biases or conflicts of interest, and independence from Bayer or the investigators are neither documented nor mentioned. This makes any judgment about their compliance with applicable conflict of interest and independence rules impossible may have conflicts of interest and may not be independent of the investigator and sponsor. Bayer did not provide resumes or any other information on the qualifications, resumes, institutional affiliations, sponsors, education, financial interests, or qualifications of for members of the IRB for the azinphos methyl test. Moreover, the IRB chair has the same last name as a lead investigator on the study. Therefore it is impossible to determine whether the IRB chairman is related to a lead investigator, or the IRB meets the requirements of independence, balance, and lack of conflict of interest. [Violates Common Rule §46.107(e); Helsinki, ¶B-13]
  15. EPA did not conduct "active and aggressive scrutiny" of the IRB, and did not ensure that it had "adequate staff and financial resources" to conduct such scrutiny. EPA apparently has not visited or reviewed the IRB at issue. [SAP-SAB Majority, Letter at 3][Common Rule §46.103]
  16. The report expressly prohibits dissemination of results, contrary to requirements that results of human studies are to be published or disseminated. The azinphos methyl report states, "This report may not be reproduced or communicated to a third party without written permission of Bayer Corporation." This is contrary to the requirement that for human studies, "negative as well as positive results should be published or otherwise publicly available." [Helsinki ¶B-27]

EPA Has Funded Certain Human Tests of Air and Other Pollutants
In the past EPA has quietly funded studies to support stricter standards for air pollution in which people were exposed to levels of pollutants similar to those encountered in polluted cities. While NRDC has not carefully reviewed all of these studies for their scientific validity and compliance with ethical standards, it is clear that those studies were fundamentally different in many ways from the pesticide studies. First and most important, they were intended to improve public health protection and thus could directly benefit the health of the volunteers and the public. The EPA Science Advisory Board has taken the position that human tests may be ethically acceptable if they are conducted to improve health protection, not to advance industry interests. Similarly, the Nuremberg Code and Helsinki Declaration state that human tests are justifiable only if they benefit the subject or would lead to health benefits for others.

In addition, in the EPA air studies the volunteers are exposed to the types and quantities of pollutants that they would experience in a smoggy city on a bad day. The air studies we have seen would yield stricter standards to improve public health -- not weaker standards to benefit industry. Therefore, to the extent that these experiments show that such exposures cause problems, they could directly help people. In addition, the EPA air studies appear to be scientifically distinguishable from the pesticide studies, because the air studies do not use a small number (e.g. eight) people to reach conclusions about the supposed safety of a whole population for whom they cannot possibly be fully representative. Rather, the air studies use volunteers to document biological mechanisms of reversible health problems triggered by the pollutants at levels the public could encounter in every day activity.

By contrast, unlike the air studies, pesticide studies generally expose people to chemicals for the sole purpose of weakening standards that would be established if animal tests were used, and often these tests expose people to far more of the pesticide than consumers would encounter in their food or the ambient environment.

A landmark court decision from Maryland's highest court in August 2001 found that an EPA-funded study by Johns Hopkins researchers, in which families with young children became lead poisoned after they were encouraged to move into partially cleaned up apartments containing lead paint, was unethical, violated the Nuremberg Code, and violated EPA's Common Rule. In this case, the court found the study highly unethical because parents were not fully and timely informed about the risks to their children or about toxic lead hot spots in their apartments. The court found that parents cannot ethically consent to intentionally exposing their children to such toxic risks. In addition, the court found that the Hopkins "institutional review board" that was supposed to review the ethics of the study was a failure. Grimes v. Kennedy Krieger Inst., No. 128, Md. Ct. of App., slip op. at 47-52, 77-88 (August 16, 2001).

The Natural Resources Defense Council is a national, non-profit organization of scientists, lawyers and environmental specialists dedicated to protecting public health and the environment. Founded in 1970, NRDC has more than 500,000 members nationwide, served from offices in New York, Washington, Los Angeles and San Francisco.


1. American Chemical Society, http://pubs.acs.org/journals/pharmcent/company5.html. For further information on I.G. Farben and Bayer, see, http://abcnews.go.com/onair/2020/2020_990611bayer.html (ABC News summarizing a recent lawsuit filed by Holocaust survivors against Bayer for alleged abuses in concentration camps); http://www.oregonlive.com/news/99/12/st120501.html (Bayer took assets from IG Farben, developed azinphos methyl, and is pushing for human testing of pesticides); http://www.earthisland.org/eijournal/new_articles.cfm?articleID=204&journalID=47; http://www.corpwatch.org/trac/feature/humanrts/history/mimkes.html (both summarizing Bayer-IG Farben links); http://home.earthlink.net/~x288files/I.G.intro.htm; http://www.mega.nu:8080/ampp/bayer.html; http://www.healthwatcher.net/Bayer/bayercrimes.html; http://abcnews.go.com/sections/us/DailyNews/terrorism_chemical.html (Zyklon -B developed as pesticide); http://www.nizkor.org/faqs/auschwitz/auschwitz-faq-06.html (Zyklon B used in Auschwitz); http://www.rrz.uni-hamburg.de/rz3a035/ZyklonB.html (Zyklon B used in Auschwitz). See also, Borkin, Joseph (1978), The Crime and Punishment of I.G. Farben, Free Press, New York.

2. New York State Health Commissioner Barbara DeBuono, M.D., Press Release accompanying "University of Rochester Case Report," September 26, 1996; see also, "Investigative Report on New York State Department of Health," 1996. The EPA committee's co-chairman, Dr. Mark Utell of University of Rochester, tests the effects of air pollutants on human subjects. He was a lead investigator in a 1996 test evaluating the effects of environmental exposure to pollution and smoking on lung tissue. A healthy student volunteer died in April 1996 as a complication arising from one of these tests. The Medical Examiner found she was overdosed with an anesthetic administered during the test. This occurred when a tube was "jammed down her throat," in the words of one press account. The death prompted state and U.S. Health and Human Services Administration (HHS) investigations and restrictions on the investigators' work for a time and extensive recommendations for changes in the research. See, ibid; "Student Dies After Taking Part in Research Study," New York Times, April 4, 1996.; "New York Seeks to Tighten Rules for Medical Research, New York Times, September 27, 1996; "Teen Dies in Medical Study," New York Daily News, April 4, 1996; Correspondence between HHS, NIH Office for Protection from Research Risks, and University of Rochester, 1996-1997.