In setting limits on chemicals in food and water, the Environmental Protection Agency (search) may rely on industry tests that expose people to poisons and raise ethical questions.

The new policy, which the EPA is still developing, would allow Bush administration political appointees to referee any ethical disputes. Agency officials are putting the finishing touches on a plan to take a case-by-case approach.

"It says we're going to look at each study on its individual terms and accept studies unless they are fundamentally unethical or have significant deficiencies," said Bill Jordan, a senior policy adviser in EPA's Office of Pesticide Programs (search). "We're setting the stage for making decisions about these studies. No guarantees that we will accept the data, and no guarantees that we will reject the data, either."

He added: "The system is for each program office to look at a study, and if there's any reason for concern, to bring it to the highest levels in our agency. If we need to, we'll go to outside peer reviewers, bioethicists."

Pesticide makers say human tests give more accurate results about the risks of the products to people and the environment, and that they follow safety guidelines set by Congress, EPA, courts and scientific groups.

A Nov. 3 draft of the plan, obtained by Public Employees for Environmental Responsibility, or PEER, says that anyone affected "should not assume that EPA will follow a prescribed method of reviewing a particular human study in each and every instance."

"This is a case-by-case process. As such, it binds no one to a particular result," says the draft obtained by the whistleblowers' advocacy group.

The draft has undergone several rewrites since then but there have not been any substantial changes, Jordan said. A final notice will probably be published in January and a new rule on human testing data issued by 2006, he said.

Critics say that with the draft plan, the EPA is shirking its duties to set rules now.

"By this sleazy move, EPA defers developing enforceable ethical standards," said Jeff Ruch, PEER's executive director.

The new policy would probably first be applied to pesticides such as aldicarb, carbofuran, DDVP and malathion, Jordan said.

Experiments using human subjects submitted by pesticide and other chemical makers have been a growing source of controversy at the agency. Jordan said the agency has not relied on any industry data in setting limits on pesticides or other chemicals since the late 1990s.

However, "all the studies do wind up in EPA's hands," whether they are relied on in the decision-making or not, he said. EPA also conducts its own scientific research involving people.

In February, the National Academy of Sciences recommended that EPA establish a human studies review panel to look at all such studies, both at the start and at the end.

Instead of creating a review panel, EPA plans to expand the duties of the director of EPA's National Center for Environmental Assessment and provide the center with more resources. "We think it's consistent with the spirit of the NAS report," Jordan said.

In June 2003, EPA was told that until it issues new rules, it cannot refuse to consider industry tests involving people on a case-by-base basis. The order from the U.S. Court of Appeals for the District of Columbia was the result of a lawsuit by the pesticide industry.

Just a month before the court ruling, EPA announced it would begin the process of establishing new rules — but then didn't follow through.

EPA briefly stopped accepting industry data from experiments on humans near the end of the Clinton administration. After President Bush took office, EPA documents showed that agency officials had resumed considering data from industry tests on humans.

Jordan said that policy change never took off.

"Folks said that was a bad idea and we fairly backed off that," he said. "We have not issued any risk assessments or made any regulatory decisions to approve pesticides where we have used human studies."

Citing ethical concerns, EPA earlier this month also temporarily suspended a planned government study into how children's bodies absorb pesticides and other chemicals.

EPA scientists and environmentalists said the two-year study, with $2 million in backing from a chemical makers' trade group, might encourage poor families to use more pesticides. Families that participated were to get $970 each plus a camcorder and children's clothes.