Dioxin: death for objectivity

Shortly before hurricane Katrina struck Mississippi, a devastating jury decision there struck the company DuPont, awarding a man $15million because of his scientifically groundless claim to have been injured by the chemical dioxin.

This was only the latest in a long line of bogus assaults on dioxin. Greenpeace and other environmental organisations damn the US Environmental Protection Agency (EPA) for moving too slowly to eliminate dioxin from meat and dairy products, air and water, and, most poignantly, from mothers’ milk. Greenpeace says that dioxin causes cancer, miscarriages and birth defects, developmental disorders and lasting mental impairment. Despite moving too slowly for Greenpeace, however, EPA agrees that dioxin must be carefully regulated. And the Department of Veterans Affairs (VA) pays millions of dollars to thousands of veterans each year as compensation for the diseases ostensibly caused by the dioxin that contaminated Agent Orange.

Greenpeace, EPA and VA cite the results of hundreds of animal tests to bolster their claims about dioxin’s deadliness. But they ignore or downplay the absence of human evidence for harmful effects.

That absence is not from lack of trying to find human effects. EPA, VA, the US Department of Health and Human Services, the International Agency for Research on Cancer, the Occupational Safety and Health Administration, the National Institute of Occupational Safety and Health, the United States Air Force, state agencies, and industry, have all spent millions studying populations of people who have been or might have been exposed to dioxin.

Sifting through the results of studies of workers exposed to high levels of dioxin reveals scattered reports of a particular cancer or disease being found at above-expected levels in one study or another. It also reveals that some cancers or diseases are found at below-expected levels in some studies. There’s little consistency – cancers elevated in some studies are decreased in others.

The results support the conclusion that dioxin exposures to human populations are without effect, and that the sporadically found higher and lower disease rates are the result of random fluctuations in disease occurrence. Even the International Agency for Research on Cancer’s (IARC) widely proclaimed assessment that dioxin causes cancer in humans relies on animal test results; IARC concedes the evidence from studies of humans is less than convincing.

What the dioxin studies really tell us

There are fewer studies of ‘environmentally exposed’ people, primarily because there are few populations that have been exposed to higher than ‘background’ levels. The two most studied populations are those that lived around a chemical plant in Seveso, Italy, which blew up in 1976; and the residents of Times Beach, Missouri, which was exposed to dioxin-contaminated oil. These studies find that there are no increases in overall disease rates or cancers. It is possible to associate dioxin exposures with elevated occurrence of a specific cancer or disease in some studies, but the occurrence of other cancers and diseases is lower than expected, and there is little consistency among studies’ results. This suggests that the varying disease rates result from fluctuations of occurrence observed whenever small populations are studied.

The most publicised dioxin-exposed group is Vietnam veterans. Agent Orange, a mixture of two herbicides, was used to defoliate trees in the jungles of Vietnam. It was always contaminated with traces of dioxin, present at levels from less than one part per million to perhaps 50 parts per million, so anyone exposed to Agent Orange was exposed to dioxin. Dioxin persists for years in the fatty tissues of animals, and measuring current levels of dioxin in those tissues provides information about past exposures to dioxin and dioxin-containing substances, such as Agent Orange.

The two active ingredients of Agent Orange are far less toxic than dioxin in animal tests, but one was removed from the market because it was always contaminated with dioxin. The other is still available in any hardware store.

The ranch hands

Two groups of Vietnam veterans have been intensely studied. The ‘Ranch Hands’ are the 1,200 Air Force personnel who serviced and flew the airplanes that sprayed 90 percent of the Agent Orange used in Vietnam. Along with 1,200 other Air Force personnel – a control group referred to as ‘Comparisons’, who were not exposed to Agent Orange – they were subjected to a week-long physical and psychological examination every five years between 1982 and 2002. This finds no difference in overall disease or mortality rates between the Ranch Hands and Comparisons.

Much has been made of a supposed elevated rate of diabetes and a cancer in the Ranch Hands. But the widely proclaimed increase in diabetes among Ranch Hands results from wilfully ignoring the results in the Comparisons: in fact, diabetes occurs at the same rates in Ranch Hands and Comparisons. The increase of melanoma seen in the Ranch Hands is balanced by the decrease in digestive cancers – consistent, again, with no effect of dioxin on cancer rates. Random fluctuations in the numbers of particular cancers – elevations in melanoma and decreases in digestive cancers – are to be expected when small populations are examined.

Exposed ground troops?

The 1,200 Ranch Hands represent a tiny fraction of the 2.8million Vietnam veterans. Many of those millions understand that they were exposed to Agent Orange; thousands are receiving compensation for diseases that the VA associates with Agent Orange exposure. Thousands more can expect compensation when they develop the diseases – lung and prostate cancers and diabetes – common to men in their 50s, 60s, and 70s. But this compensation is based on ignoring science.

The Centers for Disease Control (CDC) determined dioxin levels in 600 Army veterans of the ground war in Vietnam who had served near places Agent Orange had been sprayed, and compared those to levels in 100 veterans who had not served in Vietnam. CDC found no elevated dioxin levels in the Vietnam veterans and concluded there was no evidence for exposure to Agent Orange. This was no surprise. Agent Orange spraying was restricted to areas away from US troops, and its components, including dioxin, degrade in sunlight, of which there was plenty on top of the trees in the Vietnam jungles. CDC gave up on studying the effects of Agent Orange because it could not identify ground troops who had been exposed (if indeed any had).

CDC’s comparison of the health of Vietnam veterans with the health of veterans who had not served in Vietnam revealed some differences related to the stress and noise of combat. CDC concluded, however, that there was no convincing relationship between service in Vietnam and health effects. Congress had directed the Office of Technology Assessment (OTA), where I worked at the time (in the 1970s and early 80s), to review and comment on CDC’s findings. OTA agreed with CDC.

CDC’s results outraged many veterans and members of Congress, who insisted that they must be in error. Congress turned to the Institute of Medicine (IOM), a part of the National Academy of Science (NAS), for another review. However, when IOM reviewed the information, it did not use standard scientific criteria. It based its conclusions on statistical association, not on causality, as is common in scientific reviews. Dr Christopher Portier of the National Institute of Environmental Health, a frequent contributor to EPA risk assessments, characterised the IOM reviews as ‘political’ rather than scientific.

IOM found evidence of associations between Agent Orange exposure and more than a dozen diseases. Based on those decisions, VA now pays compensation to any Vietnam veteran who develops an ‘Agent Orange disease’, despite there being no evidence that those diseases are more common among Vietnam veterans than among the general population.

Uncertainty on both sides of the Atlantic

What about the general population? Has it been exposed? There is no doubt that everyone has been exposed. Veterans who did not serve in Vietnam, Air Force veterans who did not work with Agent Orange, prospectors who have spent their lives in the desert – all have detectable levels of dioxin in their tissues. The question is: how risky are those levels?

European countries, which like the USA have been wrestling with what to do about exposures to dioxin, have decided that current background exposures are acceptable and contribute little if anything to the risk of disease. What level of dioxin exposure poses a risk has not yet been settled in the USA.

For more than 15 years, EPA has failed in its attempts to convince its Science Advisory Board that its estimates of the risks of dioxin, including the ‘background levels’ present in all of us, are correct. Some members of the SAB have concluded that dioxin is not likely to have caused human cancer or other disease. Others agree with EPA that background levels increase human risks. Because of the impasse, EPA’s dioxin risk assessment has been passed to the National Academy of Science (NAS) for review and critique, and the NAS report is expected at the end of the year. If the NAS supports EPA’s risk assessment, we can expect that regulatory and litigation costs associated with dioxin will balloon far above the more than $100billion estimated a few years ago.

One thing is certain: dioxin exposures will continue to fall. From historically low levels, dioxin exposures began to increase in the 1940s and peaked in the 1970s. Since then, exposures have fallen and are expected in the next decade to reach levels last seen in the 1930s (and earlier).

The reasons for the decline are not completely clear, but they include better combustion processes, which minimise or prevent dioxin formation in the burning of rubbish and other materials, the elimination from commerce of dioxin-contaminated chemicals, and, probably, the phasing out of leaded gasoline, because the chemicals used in it to scavenge lead are no longer available for combustion. Whatever risk (if any) existed from environmental exposures in the past few decades will fall with exposures.

Occupational exposures to dioxin have fallen more precipitously. The dioxin-contaminated chemicals that were the source of occupational exposures are no longer being manufactured.

Despite the lack of support for risks from environmental exposures to dioxin, it’s likely that more and more litigants will make their way to court to claim that their diseases (or fear of diseases) are rooted in exposures to dioxin. After all, a Vietnam veteran can be compensated for an ‘Agent Orange disease’ without having to present any evidence of exposure – simply having been in Vietnam is sufficient.

Plaintiffs’ lawyers are assembling classes of ‘exposed’ people who lived or worked near a source or alleged source of dioxin. They parade the IOM decisions and the VA compensation program as evidence of a connection between dioxin exposures and disease. But neither bureaucratic decisions nor jury decisions determine scientific fact.

Michael Gough, Ph.D., a biologist, has advised the congressional Office of Technology Assessment and think-tanks including the American Council on Science and Health (also see HealthFactsAndFears.com).