Thousands of chemicals approved
By Tom Avril
PHILADELPHIA - Without requiring lab tests to determine
their safety, the U.S. government has approved thousands
of chemicals for use in such products as sofa cushions,
soaps, paints and baby bottles.
On average, two more chemicals are approved every day.
The result: consumers are unwittingly part of a kind
of vast, uncontrolled lab experiment.
"We're treating (people) worse than lab rats,"
said Karen Florini, an attorney with the non-profit group
Environmental Defense. "At least with lab rats, somebody
bothers to collect the data."
The U.S. system of regulating chemicals is under renewed
scrutiny as European officials voted last week to adopt
much tougher rules, which would require substantial testing
of many substances before they can be sold there.
With growing amounts of synthetic substances detected
in human blood and breast milk, U.S. critics have stepped
up calls for similar rules here.
Chemical makers counter that the European proposal would
cost billions, stifling research in an industry whose
products are overwhelmingly safe and perform valuable
functions in society. With strong support from the Bush
administration, the companies favor a more voluntary approach
Aside from saving time and money, conducting fewer tests
on lab animals means less criticism from animal-rights
As the debate continues, new concerns emerge regularly:
Polybrominated diphenyl ethers, used as flame retardants,
have been linked to brain damage and lower fertility in
lab animals. The amounts in human breast milk in the United
States are doubling every two to five years, in some cases
nearing levels linked to health problems in animals.
In a rare move Monday, the chief U.S. maker of two such
chemicals agreed to stop making them by the end of 2004.
Bisphenol-A, used in baby bottles, dental sealants and
linings of food cans, has been tied to lower fertility
in rats and defective chromosomes in mice eggs.
Perfluorinated compounds, used to make Teflon pans and,
formerly, Scotchgard, have been turning up in human blood
and breast milk. Some members of this chemical family
have been linked to cancer in lab animals; human risks
Methyl tertiary-butyl ether (MTBE), a gasoline additive
that reduces air pollution, readily infiltrates groundwater
because of its high solubility. It is linked to cancer
in mice and is labeled a potential human carcinogen.
With most chemicals, there is no proven risk to humans;
manufacturers often sponsor research that contradicts
the findings of possible risk.
And in most cases, that's the end of the story.
Even if health concerns are raised after a chemical is
on the U.S. market, the government often does not require
that it be tested in any systematic way.
And manufacturers can continue making it for years.
When companies do come across possible risks, they must
submit data to the U.S. Environmental Protection Agency.
But companies can - and for most new chemicals, do - claim
that the information is confidential.
As a result, the information is never available to the
This system for screening chemicals, enacted 27 years
ago, allowed continued production of the 62,000 chemicals
then in use and required little, if any, lab testing for
the 18,000 approved since then. (Other laws govern chemicals
in pesticides, food and drugs, which do require tests.)
Though criticized now, the Toxic Substances Control Act
was seen in 1976 as a big improvement, said a lab technician
at DuPont's Chambers Works plant in Deepwater, Salem County,
"They used to put `XYZ' on the drums," said
Corliss Sheppard, former head of Local 2-943 of the Paper,
Allied-Industrial, Chemical & Energy Workers union.
"You couldn't even say what it was. We were tickled
to death to at least make them have to tell us what it
Difficulties in implementing the law soon became apparent.
EPA has the power to require testing of any chemical,
old or new, which it believes may pose a "reasonable
risk" to human health or the environment.
Yet the burden of demonstrating risk lies with the agency,
a lengthy process that involves estimating exposures for
each way a substance might be used. The agency rarely
requires tests through this process; more often it negotiates
with companies to conduct tests.
A 1997 EPA study found that of the 3,000 chemicals imported
or produced in the United States in amounts above 1 million
pounds, 43 percent had no publicly available data on toxicity.
Just 7 percent had a full set of basic data, which includes
information on how long a chemical persists in the environment
and its short- and long-term impacts on the health of
Soon after, manufacturers volunteered to test these high-volume
chemicals, an effort that is ongoing. Industry officials
say they already had some data for most of the high-volume
chemicals, though they concede it was not publicly available.
"This allegation that these chemicals haven't been
evaluated isn't necessarily accurate," said Steve
Russell, an attorney for the American Chemistry Council,
an industry group.
Moreover, Russell said, most of the 80,000 approved chemicals
are not made in significant amounts. Council officials
estimate 15,000 chemicals are actively made, but could
not provide an exact figure, nor could they say how many
have been tested.
If requiring tests is hard, restricting a chemical's
use is even harder.
To take such action, the EPA must determine that any
risks are not outweighed by a substance's economic and
societal benefits, again for each way in which it might
be used. In practice, this provision is so difficult and
expensive to implement that it is rarely used.
Even with asbestos, a substance known to cause a rare
lung cancer, the EPA spent a decade documenting the risks
but ultimately could not make a proposed ban stand up
Restricting new chemicals is easier, through a 90-day
screening process. Using computer models, the EPA predicts
which new chemicals may be harmful by comparing their
structures to those of existing substances - yet again,
safety data on the old chemicals can be limited.
In a handful of cases, manufacturers have volunteered
to stop making chemicals. In 2000, 3M agreed to phase
out a chemical used in Scotchgard. Yesterday, Indianapolis-based
Great Lakes Chemical Corp. agreed to phase out two flame
retardants by the end of next year.
California had previously banned the two flame retardants
as of 2008; a European ban takes effect next year. Great
Lakes also announced a replacement product for use in
furniture, dubbed "Firemaster 550," which it
had voluntarily submitted for an unusual EPA review.
Officials declined to reveal the exact chemical name
or structure but said it was not in the same family as
the original products. EPA officials said the new substance
did not appear to be toxic but said they would monitor
A representative of flame-retardant makers maintained
there is no risk from the old varieties.
"Just detecting something in the environment doesn't
mean it's going to have any effect on human health, or
on animal health," said Peter O'Toole, U.S. program
director for the Bromine Science and Environmental Forum.
Critics counter that evidence of harm will be elusive
without required testing.
"How do you know?" said Joel Tickner, an occupational
health professor at the University of Massachusetts in
Lowell. "Unfortunately, the lack of proof of harm
is very often misinterpreted as proof of safety."
That may change in Europe. Wednesday, the European Commission
voted to require various tests on more than 10,000 chemicals,
depending on the amounts produced. Other data are required
on 20,000 more.
Final approval of the proposal could take two years.
The commission estimates the direct cost to chemical
makers at $2.7 billion; other companies that use chemicals
in their products would spend billions more. Industry
officials say the costs would be higher.
Upon evidence of potential risk, Europe could restrict
or ban chemicals without waiting years for proof, a concept
known as "the precautionary principle."
Proof of risk is a high hurdle, one cleared by only a
handful of substances to which humans are exposed at significant
levels. Some scientists think the evidence against polybrominated
flame retardants, for example, is headed in that direction.
How the chemicals enter the body is unclear. What is
not in dispute is that the chemicals are "persistent,"
meaning they don't break down in the environment for years,
All the more reason to play it safe, Tickner said.
"If we wait until we have perfect data on the chemicals,"
he said, "and we're wrong (about their safety) -
then you deal with them for a long time."