EPA's Proposed Cancer Risk Assessment Guidelines
A Junk Scientist's Dream Come True
At long last, the wait is over. After almost nine
years, the U.S. Environmental Protection Agency has finally proposed to update its cancer risk
assessment guidelines. After reviewing them, I think I would have waited a lot longer,
perhaps forever. If you're a fan of junk science, though, you couldn't
ask for more.
EPA's cancer risk guidelines were first issued in 1976 and updated
in 1986. The 1986 guidelines were viewed so dimly, however, that
efforts to revise them began as early as 1988. Since then, the public has been continually teased
by rumors about the "imminent" release of new and improved guidelines. Now, new proposed
guidelines are out. But, for one major reason, they are a giant step in the wrong
direction.
As rightfully acknowledged by EPA, potentially the most valuable
tool to the risk assessor is epidemiology, the study of
distributions and causes of disease in actual human populations. In
short, epidemiology is the study of real people in the real world.
However, there is good epidemiology and there is bad epidemiology.
To help distinguish good from bad, criteria have been developed and employed. For example,
epidemiologic risk estimates are more credible when they are large and precise. Exposures of
concern should precede the onset of diseases by biologically
reasonable amounts of time. Observed risks should make sense in
terms of biological knowledge. Risks should be observed for similar
exposures in different populations in different studies. There
should be a dose-response relationship where the greater the doses
the greater the observed risks.
Of critical importance among these criteria, is the requirement
that epidemiologic results be statistically significant. That is,
and for the sake of simplicity, there should be some level of
confidence, usually 95 percent, that observed study results did not
occur by luck or chance. Study results that are not statistically
significant, are deemed to be a fluke and are not suitable for
concluding that a risk exists.
The traditional requirement of statistical significance has long
been the Achilles heel of epidemiologists and, hence, risk
assessors. Because of statistical significance, epidemiologists
have not been able to convincingly associate electromagnetic
fields, dioxin, environmental tobacco smoke, dietary pesticide
residues, and hazardous waste sites, to name a few, with cancer
because the results of their epidemiologic studies often have not
been statistically significant.
But if EPA has its way, this problem will be no more. Answering the
prayers of many epidemiologists and risk assessors, EPA, through
its proposed cancer risk assessment guidelines, would deliver these
epidemiologic studies from the plague of statistical significance.
That is, in listing criteria for evaluating the quality of an
epidemiologic study, guess what criterion EPA has carefully
omitted? Statistical significance!
In its proposed cancer guidelines EPA specifically listed a number
of criteria, but not statistical significance. Why? Was this an
oversight? Hardly! With statistical significance out of the way,
EPA's latitude in using epidemiologic studies to associate various
substances and conditions with cancer has been significantly
increased. Although included in the 1986 cancer risk assessment
guidelines and, thus part of EPA's current epidemiologic criteria,
the requirement of statistical significance has now been quietly
deleted. Although EPA mentions that significance analyses should be
conducted, achieving statistical significance is no longer an
explicit requirement.
This actually comes as no surprise. For years, epidemiologists have
been trying to do away with the hurdle of statistical significance.
Thanks to EPA, it appears that they are well on their way to
victory.
As proposed, the guidelines further enhance EPA's already virtually
unfettered ability to label as cancer-causing whatever substance or
condition it chooses. More important, however, given the public's
acquired immunity to cancer fearmongering, the proposed cancer risk
assessment guidelines pave the way for EPA to take a new
stranglehold on society through the future assessment of potential
health risks from environmental estrogens, the subject of Our Stolen Future.
Material presented on this home page constitutes opinion of the
author.
Copyright © 1996 Steven
J. Milloy. All rights reserved. Site developed and hosted by WestLake
Solutions, Inc.