EPA Pesticide Program Dialogue Committee Testimony

The following written testimony was presented to the Pesticide Dialogue Committee in Alexandria, Virginia on April 16,1998.

Dr. Carl K. Winter is the Director of Foodsafe Program and Associate Cooperative Extension Food Toxicologist in the Department of Food Science and Technology at the University of California Davis, CA 95616

To The Pesticide Program Dialogue Committee

My name is Carl Winter. I am an Associate Extension Food Toxicologist on the faculty of the Department of Food Science and Technology at the University of California at Davis and I direct the University's FoodSafe Program. I hold a Ph.D. in Agricultural and Environmental Chemistry and a B.S. in Environmental Toxicology, both from Davis. In ten years on the faculty of the University of California, my program has focused on pesticide residues, natural food toxins, and risk assessment. During that period, I have not received any funding support from the agricultural, food, or chemical industries.

I also represent the Institute of Food Technologists, a scientific society composed of 28,000 members working in food science, technology, and related professions in industry, academia, and government. I serve the Institute as a designated Food Science Communicator as well as the Chair-Elect of the Toxicology and Safety Evaluation Division. I recently co-authored an Institute Scientific Status Summary on assessing, managing, and communicating chemical food risks which has been distributed to PPDC members as a primer discussing many of the risk assessment issues being considered by this committee such as uncertainty factors, threshold levels, and high-to-low dose extrapolation.

I am here today to address two issues:

1) the extremely low risks posed by pesticide residues in food, and 2) my concerns with the overly conservative risk assessment methods that may be used by the EPA in its interpretation of the FQPA that may lead to scientifically unjustified decisions that could significant and adversely impact agriculture of California and the nation.

I'll start with the second issue first. When the bill was passed in the summer of 1996, I was initially pleased. The spirit of the bill, in my opinion, was to provide greater scientific flexibility in assessing the risks posed by pesticides in food to aid in improved regulatory practices. The bill replaced the rigid and anachronistic zero-risk Delaney Clause with a framework that allowed input of the best possible science into the risk assessment process. After following FQPA developments over the past twenty months, I have become very concerned with the direction of EPA's efforts to implement it. My concerns are that the conservative nature of the assumptions used to calculate risks may result in the development of phantom risks that exist only on paper as a result of stacking or multiplying layers of conservatism; such exaggerations of risk may ultimately lead to unsound regulatory decisions.

Obviously, risk assessment is a complicated process that requires a multitude of assumptions to be made. Most of us in this room today are aware of the 100-fold uncertainty factor that is typically used in the assessment of threshold risks from chemicals such as pesticides that assumes that 1) humans are ten times more sensitive to chemicals than the most sensitive laboratory animals tested and that 2) some humans are ten times more sensitive than the average human. What fewer people are aware of, I believe, is that a variety of lesser-known assumptions are also inherent in the risk assessment process that also may dramatically exaggerate risks. The determination of pesticide use, residue levels, and food consumption estimates typically use conservative assumptions and it is likely that aggregate exposure estimates considering water and residential exposure will also be exaggerated. It is critical to realize that even the No Observed Effect Level (NOEL) from which the reference doses are derived also tend to be conservative. As an example, if a chronic animal toxicology study yielded a NOEL of 10 mg/kg/day and a Lowest Observed Effect Level of 200 mg/kg/day, the "true" NOEL could be anywhere between 10 and 200 mg/kg/day and certainly much higher than 10.

Such exaggerations of risk become even greater when considering cumulative risks from toxicologically-related pesticides such as the organophosphates; in practice, this results in stacking all of the conservatism of the individual pesticide risk assessments, leading to a "super-exaggerated" risk. And, on top of all of this, there's consideration given to applying an additional 10-fold safety factor to the risk assessment of sensitive population subgroups such as infants and children.

In my opinion, the compounding of the various conservative assumptions leads to the generation of phantom risks that exist only on paper but not in reality. What this means is that the proposed "risk cup" is more likely taking on the shape of a medicine dropper.

With all of the focus on the treatment of the uncertainties inherent in the risk assessment process, it is convenient to overlook what is known about pesticide residues in foods. Results from hundreds of thousands of food residue analyses conducted by state and federal agencies and the food industry consistently indicate that the levels of residues, when indeed detected, remain extremely low.

The risks posed by residues are perhaps reflected best by the results of the Food and Drug Administration's Total Diet Study, a comprehensive market basket survey performed annually that analyzes food for residues at the time of consumption. As an example, consider this scenario:

1) Take the typical human daily exposure to a pesticide obtained from the Total Diet Study.

2) Now feed laboratory animals 10,000 times the typical human daily exposure (on the basis of body weight) every day throughout their lifetimes.

3) What happens to the animals?

In general, nothing happens. For any noticeable effects to be observed, animals generally need to be exposed to doses greater than 10,000 times our typical daily dose. Does this prove the safety of pesticide residues? Certainly not. But it does explain why there is strong skepticism among many members of the health community, myself included, over whether pesticide residue controls need to be tightened.

So to summarize, I am concerned that the paths EPA may take in assessing risks according to its interpretation of FQPA may create theoretical phantom risks that exist only on paper. Basing regulatory actions on such exaggerated estimates of risk may have immense consequences. By unnecessarily restricting the uses of many pest control products, there may be risk tradeoffs through substitution of less effective chemicals which could increase worker safety, environmental, and resistance management concerns. Food production and quality may also be affected, leading to lower availability and higher consumer cost for fruits and vegetables. Effects could be dramatic in states such as California, where I live, which is by far the leading agricultural producer in the nation. Almost all of California fruit and vegetable production centers on its 250 "minor" crops which may be most affected by unnecessary pesticide restrictions.

It is hard to argue against the safety of infants and children. As the father of four and six year-old boys, I strive to do everything I can to ensure their health and safety. But if we're really concerned about what they eat, we should encourage their consumption of healthy fruits and vegetables rather than possibly restrict their access due to lower availability and increased cost that may result from regulatory decisions that emanate from exaggerated estimates of risk. It is critical that regulations are based on the best estimates of risk and not the worst.

I appreciate the opportunity to address this committee. If I may be of assistance to the Committee or to the Agency, I welcome such an opportunity. Thank you for your attention.

Return to Table of Contents for the June 1998 issue

Return to Agrichemical and Environmental News index