Blinded by Pseudoscience


Last May, Environmental Protection Agency chief William Reilly asked a panel of four outside scientists to review the quality of research conducted at his agency. In March, the scientists reported their findings. The EPA flunked.

Among the panel's conclusions: Agency insiders and the general public believe the EPA adjusts its science to fit its policies; the agency doesn't subject all its research to peer review or other quality checks; and, when assessing human health risks, EPA researchers rely too heavily on animal tests.

To correct these deficiencies, Reilly has asked Congress to boost his agency's research budget by 50 percent. A poker player would call this "throwing good money after bad."

Unverified, internally generated studies do not merely waste money. They can become dangerous weapons for activists or regulators with an agenda. Consider, for example, the battle over the EPA's reevaluation of dioxin. Once deemed "the most toxic synthetic chemical known to man," dioxin has recently been on the road to scientific redemption.

Infinitesimal amounts of dioxin can cause cancer in some animals. But humans and animals respond differently to some toxic substances; dioxin appears to be one.

At an October 1990 conference on dioxin cosponsored by the EPA, researchers reported that, unlike most carcinogens, dioxin does not damage DNA. Instead it links with proteins inside cells. If dioxin accumulates in these cells, and they are damaged by other carginogens, cancer may result. The conference chairmen stated that "there exists a [dioxin] dose level below which no biologically significant effects can occur."

Environmental activists were not amused by this revisionist line. Green groups have long sought expensive regulations on industrial processes using chlorine and producing dioxin as a byproduct; the new scientific view of dioxin may stop these regulations. And proving one environmentalist scare story wrong casts doubt on others—from Alar to the greenhouse effect.

The EPA's dioxin research continues. And it seems that selected sections of the review, loaded with scary rumors and exaggerated assertions, have been pitched to the popular press. A story in the February 20 Wall Street Journal points to "research suggesting dioxin could be more, not less, dangerous than previously thought." The April 6 U.S. News & World Report says "dioxin looks more dangerous to human health than ever."

But most dioxin researchers disagree. While dioxin is dangerous, they say, there is no cause for alarm. An epidemiological study published last January in The New England Journal of Medicine shows no direct correlation between cancer rates in animals exposed to trace levels of dioxin and rates in dioxin-exposed humans.

The recent press stories, however, rely on animal studies linking dioxin with immune-system damage and abnormal production of such hormones as testosterone. They list no corresponding research on humans to back up these assertions.

Interestingly, dioxin panelist George Carlo wrote a scathing letter to The Wall Street Journal, noting that its reporter relied heavily on two toxicologists "whose views on dioxin have proved to be quite inconsistent" with mainstream research.

Medical researchers link exposure to toxics with specific diseases. Skeptics can then challenge these connections with research of their own. Medical science depends upon regular patterns and recurring behaviors.

Environmental activists who shun the rigors of science focus instead on ill-defined infirmities, such as "immune-system suppression." They allege that people whose immune systems are even temporarily impaired become hypersensitive to chemicals and can develop all manner of diseases.

Thus, they can connect any unexplained ailment—the U.S. News story links "pollutants like dioxin" with Alzheimer's or Parkinson's disease—with chemical exposure. No medical research supports this.

The unfounded frenzy over dioxin suggests a fatal flaw in all EPA research: It isn't independent. Funded by an agency dedicated to regulating industry rather than understanding the natural world, it is politicized from its very inception. Publishing in peer-reviewed journals could ameliorate this flaw. Leaking scare stories to a credulous press does not.

Rather than giving Reilly his 50-percent increase, Congress should stop funding in-house research conducted by any government regulatory agency. The alternative is more political science.