EPA report thrusts dioxins back into the spotlight

A long-awaited draft reassessment of the health risks posed by dioxins by the US Environmental Protection Agency (EPA) has left little room for complacency. The EPA has strengthened its characterisation of dioxins as probable human carcinogens, and warns that they may have adverse effects on development, reproduction and immune suppression at levels close to those to which the population is currently exposed. The reassessment contrasts strongly with the reassuring picture painted five years ago in the UK by the Departments of Environment and Health - and may cast a shadow over the Government's hopes to increase reliance on incineration as a waste disposal route.

Dioxins have been in the public eye for over 20 years. Initial concern was prompted by the exposure of the Vietnamese population and US armed forces personnel to Agent Orange - the herbicide 2,4,5-T which was heavily contaminated with dioxins - following defoliation operations during the Vietnam war. This was reinforced in 1976 when a runaway reaction in a trichlorophenol plant in Seveso, Italy, released 12 kilogrammes of 2,3,7,8-TCDD, the most toxic dioxin, over the surrounding countryside. Several less severe industrial accidents over the years have also involved significant dioxin releases, including one at Coalite Chemicals' plant in Derbyshire in 1968 (see box  ).

Despite this long history of concern over dioxins, an understanding of their true significance to the environment and human health has remained elusive. It is ironic that one of the driving forces behind the EPA's decision to reassess their health effects was a growing feeling in the scientific community that the risks had been strongly overstated.

The term "dioxins" is applied as a catch-all to 210 polychlorinated dibenzodioxins (PCDDs) and dibenzofurans (PCDFs). Their toxicity is commonly rated as toxic equivalents (TEQs) of 2,3,7,8-TCDD. Any attempt to assess the risk posed by dioxins must grapple with the diversity of sources, differing patterns of contamination, the compounds' ubiquity in the environment, and the problems of detecting effects at levels which are extremely low compared to most pollutants.

The cancer question
In 1985, the EPA characterised dioxins as proven animal carcinogens and, tentatively, as probable human carcinogens. However, a 1989 report by the UK Department of the Environment (DoE) adopted a considerably more reassuring tone (ENDS Report 173, pp 18-19). It found no "convincing evidence of a link between exposure to dioxins and cancer" and suggested that as 2,3,7,8-TCDD was not itself a mutagen, a threshold must exist for its carcinogenic effects.

Overall, the DoE's report concluded, "human studies are reassuring in that no major adverse health effects have so far been seen in people exposed to much higher levels of PCDDs than the general population is ever likely to encounter". Nevertheless, the Government took the unprecedented step of accepting a guideline level for daily human dioxin intake of 1pg TEQ per kilogramme of body weight - lower than the level to which much of the population was already exposed.

The UK guideline was derived from the lowest "no-effect" concentration observed - for reproductive toxicity in monkeys. However, the Government stressed that the level was precautionary. If exceeded, it should be used "to trigger further local investigation" so that "appropriate measures should be taken to reduce environmental levels generally". The World Health Organization (WHO) has since suggested a tolerable daily intake for dioxins of 10pg TEQ/kg.

The EPA's latest risk assessment makes for less comfortable reading - although it must be stressed that the document is still a draft, and that the conclusions of the final report, due in a year's time, may be altered by the peer review process. Over 100 scientists have been involved in the three-year preparation of the draft, which reviews the current knowledge of dioxin sources and health effects. A further paper on the ecological risks posed by dioxins in aquatic ecosystems is due shortly.

Background exposures
According to EPA Assistant Administrator Lynn Goldman, the earlier characterisation of dioxins as probable human carcinogens remains "essentially unchanged", though it is stated "with greater confidence". Based on a "weight-of-evidence" approach, the EPA has proposed a risk-specific daily dose estimate of 0.01pg TEQ per kilogramme of body weight for a cancer risk of one in a million. The estimate is little changed from the figure of 0.006pg TEQ offered by the 1985 report.

Typical daily "background" exposures to PCDDs and PCDFs in the USA and Europe are 1-3pg TEQ/kg. However, the EPA says that inclusion of PCBs which exhibit "dioxin-like" health effects increases the background exposure to 3-6pg TEQ/kg. On this basis, it concludes that "plausible upper-bound risk estimates" for cancer in the general population may be as high as one in a thousand. The assessment also assumes that there is no no-effect level, and suggests that cancer risk is linearly related to TCDD exposure at low doses.

The UK's view of dioxins' role in cancer continues to be more conservative than the EPA's, but has evolved since the 1989 report in the light of recent epidemiological studies. This summer, the Department of Health quietly published a report in which one of its advisory committees concluded that despite insufficient evidence for a "clear causal link", "it would be prudent at present to regard TCDD as a possible human carcinogen". A recent study at the University of Milan among 37,000 people living near Seveso added weight to the concerns, finding an unusually high incidence of certain rare cancers.

However, the most important - and controversial - findings of the EPA's reassessment concern non-cancer health effects of dioxin exposure. Ms Goldman said that these "may include developmental and reproductive effects, immune suppression and disruption of regulatory hormones".

Broad spectrum of effects
In its final draft, the EPA is careful to use more cautious language in describing these risks than in an earlier, leaked version (ENDS Report 232, pp 9-10 ). Effects such as enzyme induction, changes in hormone levels and altered cellular function have, it says, "been observed in laboratory animals and humans at or near levels to which people in the general population are exposed." This suggests the "potential for adverse impacts on human metabolism and developmental and/or reproductive biology and, perhaps, other effects in the range of current human exposures."

According to Professor Claude Hughes of Duke University, North Carolina, who was involved in the preparation of the draft reassessment, "the general levels in the human population are right at the point where effects are seen in animal studies. This means there is no margin of safety". The reassessment warns that "the lack of a clear indication of disease in the general population should not be considered strong evidence for no effect of exposure to dioxin-like compounds. Rather, lack of a clear indication of disease may be a result of the inability of our current data and scientific tools to directly detect effects at these levels of exposure."

The EPA's assessment is based on the hypothesis that most of the typical US daily intake of 120pg TEQ is received through food, leading to typical body burdens of 40 parts per trillion when dioxin-like PCBs are included. However, it estimates that intakes among the most highly exposed 10% of the population may be three times higher.

A focus of concern is the elevated exposure of infants. Because dioxins accumulate in body fat, breast milk contains relatively high levels. The EPA estimates that the average daily exposure to a breast-fed infant in its first year is some 60pg TEQ/kg - over 20 times higher than the background exposure to adults. Breast feeding could account for 4-12% of the typical lifetime dose of dioxins - and the EPA says that "it is reasonable to assume that developing organisms may be particularly sensitive to adverse impacts from temporary increases above background exposure levels."

Once again, the EPA has gone considerably further than the DoE's 1989 report. This offered "reassurance that no major adverse effects on human reproduction have been detected as a result of exposure to dioxins", and said that there was no evidence that observed changes to the immune system "had resulted in clinical disease". Furthermore, the report found it "extremely unlikely that any harm will result from the presence of PCDDs and PCDFs in human milk", and repeated advice from the WHO that any possible risk was far outweighed by the benefits of breast feeding.

Sorting out sources
Assuming that the EPA's risk assessment survives the peer review process without radical surgery, it will present policy makers with a difficult dilemma. Dioxin releases from major sources should decline in most industrialised nations as new regulations start to bite. Recent studies also suggest that dioxin levels in the environment, and possibly in the general population, are starting to decline. However, if the health effects are as worrying as the EPA suggests, regulators will come under strong pressure to introduce tougher controls.

The main problem is the continuing uncertainty over the relative significance of the numerous dioxin sources. In a bid to plug the gaps, the EPA has called for all interested parties to supply new information on dioxin releases, including voluntary emissions testing programmes with potential industrial sources. A smaller-scale research programme was launched in the UK following the 1989 report, but few of its findings have been published.

A major headache is the debate over the relative contribution of new emissions and redispersion of existing, or "reservoir", contamination towards human exposure. Studies in the UK and other countries have suggested that known emissions account for only around 10% of estimated dioxin deposition, though large uncertainties are at work (ENDS Report 228, pp 11-12 ).

However, the EPA's draft assessment is quick to discount the theory that natural sources of dioxins may explain the gap. From studies of dioxin-like compounds in sediments and old human tissue samples, it concludes that their presence in the environment "has occurred primarily as a result of industrial practices". Some industry bodies have been keen to point the finger at "natural" sources such as forest fires and leaf composting - but the EPA says that these are probably secondary sources which were initially created by dioxin deposition from the atmosphere.

Despite the large uncertainties in current understanding, the EPA has attempted for the first time to assess the contribution of several major sources to the USA's dioxin pool. Any nation's assessment of sources will depend strongly on its pattern of industry, regulatory controls and perceived priorities. However, the draft assessment contains an interesting breakdown of known air emissions in the USA, and compares them with estimates for the UK in 1989 supplied by the industry toxicology organisation ECETOC (see table ).

Considerable uncertainty surrounds these estimates, as they are based on emissions testing at relatively few facilities. The total known US emission to air may be anywhere between 3,300 and 26,000g TEQ. The 6,000 existing hospital waste incinerators currently do not employ sophisticated control technologies, and are believed to be the USA's most significant source. An estimated 1.8 tonnes of dioxins were sent to landfill in residues from municipal incinerators.

Chemical industry gaps
However, the table is equally interesting for what it does not say. The chlorine industry, for example, has come under growing scrutiny from Greenpeace, which claims that it is a significant dioxin source (ENDS Report 223, pp 7-9 ). However, the European chemical industry federation CEFIC claims that the sector accounts for less than 1% of the dioxin releases in Europe.

The UK Chemical Industries Association responded to the EPA report with the statement that "the only well documented effect of dioxins on human health is chloracne [a skin disorder] which is reversible", while the European Council of Vinyl Manufacturers welcomed the EPA report as "confirmation that PVC's contribution to global dioxin formation is minimal." In fact, the EPA declined to put a figure on dioxin emissions arising from chemical manufacture on the grounds that "insufficient emission data are currently available to make an independent evaluation". A key aim of the EPA's data call-in will presumably be to plug this gap.

Threat to incineration?
Other dioxin sources have only come to light fairly recently in the UK. A series of studies have found significant dioxin contamination around non-ferrous metal smelters (ENDS Report 228, pp 11-12 ). A recent German study has implicated effluents from dioxin-contaminated textiles as another source (ENDS Report 235, pp 3-4 ). Varying estimates have been made for the contribution of domestic and industrial coal combustion, but no clear picture has yet emerged.

Inevitably, however, the EPA document has focused attention on incinerators, generally accepted as the largest known source in the UK. In a report last year, the Royal Commission on Environmental Pollution (RCEP) advocated a greater reliance on incineration with energy recovery, particularly for the disposal of municipal waste (ENDS Report 220, pp 13-15 ). The RCEP argued that because existing municipal incinerators must be closed down or upgraded to meet new EC emission standards by 1996, total dioxin emissions would be reduced to about 2% of the 1991 level even if the amounts of waste burned remain broadly constant. And even with a substantial increase in incineration capacity, the RCEP concluded that dioxin emissions would not exceed 6% of the 1991 level.

This summer, the Government broadly accepted the RCEP's call for greater emphasis on incineration as part of an agreed national waste management strategy (ENDS Report 234, pp 21-25 ). However, its response contained a potentially important caveat. If, said the Government, the Chief Medical Officers judge that the EPA reassessment and the latest evidence from Seveso "establish some significant uncertainty about the implications of environmental dioxin levels for human health, that would represent a constraint on decisions about the future use of incineration as a waste management option". The extent of the constraint would be for consideration in the light of the Chief Medical Officers' advice."

HM Inspectorate of Pollution has already commissioned a review of the contribution made by processes under its control to the UK's total dioxin releases. Its Director, Dr David Slater, said that "our current process of authorisations is aimed at reducing dioxin emissions - in particular from municipal incinerators - by up to 50%, and we will continue to look at ways in which emissions can be reduced yet further." In the light of the EPA report, other industries which are potential sources can expect to feel the squeeze from further controls on dioxins - the health issue that refuses to go away.

Please sign in or register to continue.

Sign in to continue reading

Having trouble signing in?

Contact Customer Support at
or call 020 8267 8120

Subscribe for full access

or Register for limited access

Already subscribe but don't have a password?
Activate your web account here