To assess whether legal tolerances establish safe levels of exposure, the following
computation is instructive:
Compare the EPA’s current tolerances for pesticides found in the USDA’s PDP
tests to the RfD’s tests for those pesticides. To facilitate comparisons, a reference con-
centration (RfC) is calculated. The RfC is the residue concentration that gives a
twenty-kg child (forty-four pounds, about average for a five-year-old) who eats
100 grams (about 3.5 ounces) of a food item a reference dose of the pesticide. A dif-
ferent body weight or serving size of food would result in a different RfC.
The RfC can then be compared with the tolerance. If the tolerance is greater than
the RfC, it meansa twenty-kg child who ate 100 grams of this food would exceed the
RfD for that pesticide on that day from the single serving of a single food(italics mine).
Now, this does not mean the child’s health would automatically or immediately be
harmed; the RfD includes a ‘‘safety margin.’’ But it does mean that the safety margin
between actual doses and those known to be harmfulis narrower than it should be
(italics mine). The definition of ‘‘safe’’ for pesticide residues ostensibly includesensur-
ing an adequate safety margin(italics mine). By definition, then,tolerances that permit
exposures above the RfD cannot be considered ‘‘safe’’ limits. Certainly, they are not ‘‘safe
enough.’’
Unfortunately, then, most of the tolerances for these pesticides on these foods
exceed the RfD, often by a wide margin. For example, tolerances for methyl para-
thion would allow 250 times the safe dose on all six foods on which this insecticide
was detected by the PDP. Likewise, tolerances for acephate and dicofol allow up to
forty-two times the RfC, those for chlorpyrifos allow up to twenty-five times the
RfC, and most tolerances for dimethoate/omethoate permit twenty times the RfC.
The ‘‘action level’’ for dieldrin (there are no tolerances for banned pesticides such as
dieldrin) is ten times greater than the RfC.
These comparisons make it clear that there is a big difference between ‘‘legal’’ resi-
dues and ‘‘safe’’ residues. Consumers can take little comfort from the fact that most
residues are within the tolerances set by the EPA, because many tolerances currently
on the bookspermit unsafe exposures(italics mine).^19 What should the EPA do to
overcome the tendency to assume that computations from testing done by pesticide
manufacturers invariably results in both ‘‘legal’’ and ‘‘safe’’ pesticide residues on food?
There is no doubt that as scientific understanding of potential cumulative and aggre-
gate effects advances, it is certain that additional and more precise data will be
required for EPA decisions (we cannot assume that the EPA will recognize this),
along with more information on subpopulation exposure and risk. In most cases the
EPA will be able to use existing FIFRA authority to require this data. Additional data
will hopefully enhance the scientific basis and effectiveness of pesticide regulations.^20
It may seem like an incidental detail, but we cannot overlook an important limita-
tion on the accuracy of tolerance setting: the procedures used to set tolerances in the
past neglect the fact that per capita consumption of some fruits and vegetables has
risen, causing potential understatement of residue intake. For example, average con-
sumption of fresh honeydew melons, broccoli, and tomatoes increased from 1.1, 1.1,
Pesticides in Food | 81