Published on Wednesday, June 7, 2000 in the Washington Post
Going Forward:
US Rejects Testing Pesticides On Humans
by Joby Warrick
The Clinton administration has decided to sidestep a major political and ethical quagmire by rejecting the use of human experiments in setting regulatory limits for pesticides.

Worried about a resurgence in human experiments by pesticide companies--some of which have been testing products on students and other volunteers for decades--the Environmental Protection Agency will adopt a policy of officially ignoring such studies in establishing legal limits for pesticides in food and water, agency officials said yesterday.

The decision essentially preempts a long-awaited report by an EPA scientific panel that had been deadlocked for months over the morality of administering pesticides to people to test their safety. A draft of the report released to panel members this week concludes that certain kinds of human experiments may be acceptable, and even desirable.

The prospect of even a limited EPA endorsement of human experiments had deeply troubled several scientists on the panel and outraged environmentalists and some medical ethicists.

"Studies that dose people intentionally with pesticides are scientifically and morally bankrupt," said David Wallinga, a physician and senior scientist with the Natural Resources Defense Council, which has been lobbying against the use of humans in pesticide tests. "This is a rerun of government tests in the 1950s, when we lined up soldiers in front of nuclear blasts to see what would happen."

The EPA panel's draft report, obtained by The Washington Post, supports limited, carefully controlled experiments to determine how pesticides are processed by the human body. But both the panel and top agency officials rejected the use of human subjects merely to establish a pesticide's toxic threshold--the legally crucial level where harmful effects are first observed.

"There is nothing in the report that will change our policy," said Steven Galson, director of science and policy in EPA's pesticides division.

While the EPA does not directly regulate scientific research by private companies, it traditionally relies on industry studies in establishing safe limits for pesticides. In most cases the regulations are derived from experiments on laboratory animals or on people inadvertently exposed to chemicals, such as farm workers. Several industry groups have advocated more human studies, arguing that regulations based on animal research are often excessively strict.

"The lab animal data could overstate human risk or it might understate it," said Roger McClellan, president emeritus of the Chemical Industry Institute of Toxicology. "It all speaks to using data from humans if at all possible--but only under ethical circumstances."

Since the 1960s, chemical companies have quietly submitted to the EPA scores of studies in which humans were knowingly exposed to pesticides. In one typical experiment in 1973, volunteers at a New York state prison were fed small amounts of the potent insecticide chlorpyrifos and monitored for weeks. In a more extreme example, pregnant women and newborns were exposed to the compound DDVP in fly strips at a hospital in Italy.

Questions over the ethics of such tests returned to the spotlight two years ago when the Environmental Working Group, a Washington-based advocacy organization, reported a surge in human experiments in Britain in the 1990s. The researchers, working mainly for U.S. companies, paid youthful volunteers about $600 each to ingest small amounts of pesticides over several weeks, according to documents obtained by the environmental group. Last April, a similar study was conducted on student volunteers in Lincoln, Neb., by pesticide manufacturer Dow Agrosciences Ltd.

Public outcry over such experiments led to the creation of the special EPA advisory committee on human experiments in July 1998. But instead of offering ethical guidance on the issue, the panel quickly bogged down into bitter and sometimes public disagreement over whether human tests could ever be ethically acceptable.

The draft released this week by the panel's leaders was described as a compromise, though several members continued to disagree sharply yesterday over its central conclusions. The report states that the use of humans in pesticide experiments "is acceptable, subject to limitations ranging from 'rigorous' to 'severe.' "

While volunteers in such experiments are not likely to see direct benefits, the knowledge gleaned from human studies benefits society as a whole, the draft concludes. Ideally, administering very small amounts of pesticides to humans could yield important medical insights--for example, by helping scientists understand how the chemicals are processed and stored within the human body.

But the draft also urges "active and aggressive scrutiny" of all human tests by the EPA, and it cautions against giving pesticides to young children under any circumstances. Human tests should never substitute for animal experiments in answering basic questions about a pesticide's toxicity, it said.

Arthur Kaplan, a panel member and one of the nation's most prominent medical ethicists, described the debate as "one of the toughest ethical dilemmas I've ever faced." Unlike most drug trials, human experiments with pesticides offer no direct benefit to the participant, and typically attract subjects only through coercion or cash.

"It seems from the public's point of view that testing is absolutely essential; from the point of view of protecting the individual it looks absolutely abhorrent," said Kaplan, director of the Center for Bioethics at the University of Pennsylvania.

But another panel member said the report failed to reflect the views of several members who are disturbed by both structural flaws and moral issues raised by previous human tests.

"The questions were: Should these tests be done, are they scientifically defendable and are they ethical," said the scientist, who asked not to be identified. "The answer to all three is no."

© 2000 The Washington Post Company