Pesticide Exposure to Children Committee Report

Click here to access an ascii version of this document.

Click here to view press advisory

To Table of Contents

To Background

Executive Summary

Assembly Bill 2161 (Chapter 1200, Statutes of 1989, sometimes called the Food Safety Act of 1989, [Food Safety Act]) added and expanded several sections in the Food and Agricultural Code and the Health and Safety Code. The bill allowed the creation of additional regulatory efforts that build upon and enhance California's food safety programs to ensure that food continues to meet existing rigorous standards.

The Food Safety Act also required the Department of Pesticide Regulation (DPR) and the Office of Environmental Health Hazard Assessment (OEHHA), both of the California Environmental Protection Agency (Cal/EPA), to jointly review how well existing federal and state programs protect infants and children from exposure to pesticide residues in food, and recommend potential areas where the existing pesticide regulatory system could be improved. DPR was to work with other designated state agencies, and was to submit a report to the Legislature on the study and recommendations. To be included in this report was an analysis of a study then being prepared by the National Academy of Sciences (NAS).

While the NAS study was being conducted, many issues raised in the Food Safety Act were under discussion among California regulators. With the Academy's release in June 1993 of its report, Pesticides in the Diets of Infants and Children (NAS Report), work could begin on the completion of the report required by the Food Safety Act, incorporating the NAS Report findings in the NAS Report in the review. Cal/EPA organized a working group, called the Pesticide Exposure to Children Committee (PECC). Chaired by DPR, PECC included representatives from the OEHHA, state Department of Health Services (DHS), California Department of Food and Agriculture (CDFA), University of California at Davis (UCD), and USEPA Region IX. During the preparation of the report, PECC members met in formal meetings as well as in informal discussions. Drafts of the report were circulated among the members for comments. Staff from the executive offices of DPR, OEHHA, DHS, and CDFA also met to make final decisions on some issues. Cal/EPA also solicited public comments on the NAS Report. Many comments were received and were considered in preparing this report.

PECC recognizes that efforts in pesticide use reduction, reduced-risk pesticide use, and alternative pest management strategies will undoubtedly affect the overall profile of pesticide residues in foods. These subjects are, however, beyond the scope of this review. They are currently being addressed in other forums at both the federal and state levels. PECC also agrees that a comprehensive effectiveness evaluation of the food safety regulatory program should be considered in the future. The scope of the evaluation should be defined by the restructured Pesticide Advisory Committee, chaired by DPR.

This report by PECC contains six chapters: (1) Introduction, (2) Toxicity Assessment, (3) Dietary Exposure Assessment, (4) Risk Assessment Issues, (5) Summary of Evaluation, and (6) Summary of Recommendations and Implementation.

I. INTRODUCTION

A. Pesticide Registration Programs

Pesticides are used to control pests that cause damage or economic loss or that transmit or cause diseases. A pesticide must be registered by both USEPA and DPR before it can be offered for sale, used, or possessed in California. The goal of the USEPA and California registration programs is to ensure the proper, safe, and efficient use of pesticides, and to eliminate pesticides that produce harmful effects.

USEPA is responsible for the registration of pesticides under various sections of the U.S. Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and for the implementation of pesticide regulations. USEPA is also responsible under the Federal Food, Drug, and Cosmetic Act (FFDCA) for setting tolerances for pesticide residues in raw agricultural commodities (Section 408 of FFDCA) and processed commodities (Section 409 of FFDCA). Tolerance is the legal maximum residue concentration of pesticide allowed in raw agricultural commodities and processed foods. For a pesticide to be used on food or feed crops, a tolerance or an exemption from a tolerance must be established. To ensure that existing pesticide data meet current standards, USEPA reevaluates pesticides through the process of reregistration. In addition, a pesticide may also be reevaluated through the Special Review process when there are new data that suggest a potential for a health or environmental hazard.

DPR is the state's lead agency for the registration, evaluation and mitigation of environmental and human health effects of pesticide use in California. In addition to USEPA-required studies, DPR may require studies to address specific areas of concern or unique growing conditions in the state. These data submitted for registration are reviewed and used by DPR scientists to ensure the safe use of pesticides in California. Pesticides are ranked for risk assessment according to their toxicity and the potential for exposure. The health risks of pesticides are evaluated under the authority of Food and Agricultural Code and statutory mandates (e.g., the Food Safety Act, Birth Defect Prevention Act of 1984, the Pesticide Contamination Prevention Act, and the Toxic Air Contaminant Act of 1983.) DPR has an extensive follow-up program which includes residue and worker monitoring, enforcement, illness reporting, and reevaluation. DPR may reevaluate a pesticide when new data show additional potential for concern, such as health hazards. When the risk for exposure is unacceptable, DPR has the authority to mitigate the risk through regulation to reduce exposure or to stop the use.

DPR and DHS have the regulatory authority to set California-specific tolerances for pesticide residues in raw agricultural commodities and processed foods, respectively. While California has the statutory authority to set state tolerances for raw agricultural commodities and processed foods, the state by regulation has adopted the tolerances established by the federal government as contained in the Code of Federal Regulations (CFR) Titles 21 and 40.

B. Evaluation of Food Safety

Federal and California agencies evaluate the safety of pesticides in the food supply following the mandates of their respective laws.

USEPA evaluates the safety of pesticides in foods and feeds as part of the tolerance-setting process. The U.S. Food and Drug Administration (FDA) and the U.S. Department of Agriculture (USDA) are responsible for the monitoring and enforcement of the tolerances.

In California, the food safety program involves the data review and dietary risk assessment programs already discussed, as well as use enforcement and residue monitoring. DPR and OEHHA, under their respective mandates, are responsible for the evaluation of pesticides in the food supply. Under the Food Safety Act, DPR evaluates the safety of pesticide residues in fresh produce and processed foods in the total diet, and examines the established tolerance levels for raw agricultural commodities. OEHHA is mandated by Assembly Bill 2848 to determine whether tolerances for processed foods are protective of public health.

To assure compliance with the nation's toughest pesticide laws, California has the largest and best-trained state enforcement organization in the nation. DPR oversees licensing and certification of dealers, pest control advisors, pest control businesses and applicators; has overall responsibility for pesticide incident investigations; and coordinates pesticide use reporting.

Pesticide use enforcement activities in the field are largely carried out by the county agricultural commissioners and their staffs, with training, coordination, oversight, and technical and legal support provided by headquarters personnel, as well as DPR field staff. Enforcement options include administrative actions; criminal and civil actions; and crop quarantine, crop seizure, and crop destruction.

DPR administers the nation's largest state pesticide residue monitoring program for fruits and vegetables. The results show that residues very rarely approach tolerance levels. Among the approximately 20,000 samples taken in 1990 and 1991, less than 1 percent exceeded tolerances and less than 1 percent contained residues between 50 percent and 100 percent of tolerances. Data from DPR's residue monitoring program are used in dietary risk assessments for the evaluation of food safety.

Under the Food Safety Act, DHS has the responsibility for monitoring processed foods for pesticide residues and other contaminants.

C. The National Academy of Sciences Report

The key issues addressed by the NAS Committee on Pesticide Residues in the Diets of Infants and Children (NAS Committee) were: (1) assessment of current information on pesticide exposure in the diets of infants and children; (2) adequacy of the current risk assessment methods and policies; and (3) identification of the toxicological issues of greatest concern and research priority. The NAS Committee made recommendations in the following areas: (1) toxicity testing, (2) consumption data, (3) residue data, (4) exposure approaches, (5) risk assessment methodology, and (6) tolerance setting.

The evaluation of the overall safety of the food supply was not one of the charges of the NAS Committee. Nonetheless, in a statement announcing release of the report, Dr. Philip J. Landrigan, who chaired the NAS Committee, said, "Parents should continue to emphasize fruits and vegetables in their children's diets." Elsewhere in his statement, Dr. Landrigan said, "The goal of our report is to make the very good food supply of the United States even better."

The Pesticide Exposure to Children Committee reviewed the NAS Report and agreed that the NAS Committee has done an excellent job in addressing the key issues. PECC agreed with the NAS Committee that there is a need to further improve the nation's already very good system of pesticide monitoring, especially for the protection of infants and children. PECC is also in general agreement with many of the methods the NAS Committee suggested to improve the system.

II. TOXICITY ASSESSMENT

Evaluations of human health risks of pesticide exposures, including dietary exposures, rely on a variety of factors. These include toxicity studies performed on laboratory animals (which are required as part of the application for pesticide registration), reports on pesticide-related illness in humans, and epidemiological studies.

The NAS Report has identified additional toxicity data that are necessary for a more thorough evaluation of risk of dietary exposures in infants and children, and recommended approaches to obtain these data. These data would provide information on: (1) the susceptibility of infants and children to pesticide toxicities; (2) the pattern of pesticide disposition (e.g., absorption, distribution, metabolism, elimination) in infants and children versus adults, and between laboratory animals and humans; (3) a broader understanding of the developmental toxicity and the toxicity to the nervous, immune, visual, and endocrine systems, and (4) acute and subchronic developmental and functional neurotoxicity for all food-use pesticides.

Both USEPA and DPR evaluate pesticide toxicity under their respective mandates. Both agencies consider the toxicological data from required submissions, from data published in the open literature and by governmental agencies, and from pesticide-related illness reports in humans. To evaluate the potential risk associated with pesticide exposures, the most sensitive toxicity endpoint identified in the most appropriate species tested, for respective exposure scenarios (e.g., acute and chronic exposures) is used to characterize the risk for each pesticide of concern. If data do not permit the determination of the most appropriate endpoint or species, data on the most sensitive of each of these factors are used. The most sensitive toxicity endpoint is the effect which is of significance to humans and observed at the lowest level of exposure from toxicological studies.

PECC agrees with the NAS conclusion that toxicity studies as currently required, while sufficient to address many health concerns, are inadequate for others. Furthermore, PECC agrees that these potential health concerns cannot be fully addressed until new testing guidelines can be developed, which in turn may lead to generation of additional data. PECC agrees that, since federal and state mandates for data are similar, DPR should continue to work with USEPA to study ways to implement the NAS Report recommendations. In so doing, DPR will coordinate with OEHHA and/or DHS on issues that fall within their respective statutory responsibilities. Guidelines for testing the susceptibility of immature laboratory animals to pesticides as well as pesticide toxicity in the nervous, immune, visual, and endocrine systems are currently undergoing development or revision at USEPA. USEPA is preparing a plan that will identify research needs on age-related differences. Such research would yield a set of general principles concerning age-related toxicity. The recommendations by PECC are:

  1. DPR should work closely with USEPA to identify the necessary toxicity data for pesticides.

  2. DPR should coordinate with USEPA to obtain information on the physiological and pharmacological differences both among infants, children, and adults, and between humans and animals. The goal is to generate data in laboratory animals that can be better used to assess the risk of pesticide exposures in infants and children.

  3. DPR should provide inputs to USEPA regarding the testing guidelines for toxicity in nervous, immune, visual, and endocrine systems. As the need for new data requirements is assessed, consideration must be given to the manner in which these data will be used and the additional value of the information.

  4. Acute and subchronic neurotoxicity should be tested for all food-use pesticides because it is not always possible to predict the potential of a chemical to cause neurotoxic effects. DPR should coordinate with USEPA in implementing this recommendation, especially concerning the specific type of study to be required.

  5. DPR should consider applicability of the testing guidelines being developed by USEPA to the risk assessment guidelines coordinated by OEHHA for Cal/EPA, when available.

III. DIETARY EXPOSURE ASSESSMENT

The potential exposure of humans to pesticide residues in foods can be estimated by determining the types and amounts of food eaten and the levels of residues in the food.

The NAS Report identified deficiencies in the food consumption and pesticide residue databases and emphasized that: (1) data should be sufficient for an estimation of data distribution, and (2) sample and data analysis as well as data reporting should be standardized for establishing a nationwide database. Specifically, the NAS Committee made recommendations in the following areas:

  1. Regarding consumption data: (a) Food consumption and water intake data for infants and children should be surveyed periodically. (b) A uniform method for the conversion of prepared foods to their components is needed. (c) Consumption for infants and children should be considered in yearly age intervals up to age 5, and two separate subgroups for age 5 to 10 and 11 to 18 years old.

  2. Regarding residue data: (a) More representative samples of raw agricultural commodities and processed foods consumed by infants and children should be monitored, using individual units instead of composite samples for some commodities. (b) Samples should be analyzed for both the pesticide and its degradation products at the levels of health concern. (c) The use of percentage of crop treated for the adjustment of residue profiles should be justified. (d) A national residue database should be established.

  3. Regarding exposure assessment: (a) Distributional analysis instead of point estimates should be used to account for variation in both consumption rates and residue levels. (b) Methods should be developed for assessing the combination of exposures from dietary and non-dietary sources.

In conducting dietary exposure assessment, USEPA and DPR use similar approaches, but different databases. This is because each available consumption and residue database has its own limitations and no single database is distinctly more appropriate than the others for exposure assessment. Of the two Nationwide Food Consumption Surveys (NFCS) databases, USEPA and DPR use those collected in 1977-1978 and 1987-1988, respectively. The latter survey has low response rate, but may better reflect current consumption patterns. Because of the concerns on the current NFSCs, PECC agreed that new consumption surveys should be considered to characterize better the consumption patterns for infants and children, and California-specific concerns. The new surveys for California should be conducted in coordination with the nationwide surveys, to the extent possible. The design and conduct of the surveys should be discussed in an interagency forum.

Data on residue levels in food and water are available from federal and state monitoring programs, and from studies conducted by pesticide registrants, trade organizations, and academia. Both USEPA and DPR evaluate available residue databases and select the most appropriate ones for use in risk assessment. Commodities are generally analyzed for residues as composites (more than one unit of a commodity, e.g., five bananas). The samples may be analyzed by either multiple- or single-residue analytical methods depending on the type of monitoring program. Since most monitoring program data are designed for enforcement purposes, the detection limits have historically been set relative to the tolerance levels. These detection limits, however, may not be sufficiently low for dietary risk assessment. USEPA may make adjustment of the detected residue data to account for the percent of crop treated and/or decrease in residue level due to food processing.

DPR uses best available data to conduct dietary exposure assessments. There are constraints and limitations to the current databases. DPR uses conservative approaches and default assumptions that tend to overestimate the exposure, and they are:

  1. All samples of each label-approved crop (crops on which the pesticide of concern is permitted to be used by the pesticide product label) are assumed to contain a certain amount of residues. Residue levels are not adjusted for percent of crop treated. In chronic exposure assessment, samples in which residues are not detected are assumed to contain levels at half the detection limit.

  2. Residue reduction through loss and degradation from storage, processing, and home food preparation is not routinely considered unless well supported by pertinent data.

  3. In estimating acute exposure, residues are assumed to be at the highest detected level at or below the tolerance in all samples of each crop.

  4. When monitoring data for the raw agricultural commodity are not available, the residue level is assumed to be at the tolerance. With processed foods, the residue level is assumed to be at the tolerance level or at a calculated level (the residue level of the raw agricultural commodity multiplied by a concentration factor).

For the determination of exposure, USEPA and DPR use computerized programs with similar analysis strategies. Both agencies recognize that there are variations in eating patterns within a population. Dietary exposures are routinely evaluated for population subgroups of different region, age, race-ethnicity, and sex. The exposures of infants and children are addressed in four subgroups: nursing infants (less than one year old), non-nursing infants (less than one year old), children 1-6 and 7-12 years old. However, the consumption of certain commodities by certain subgroups may not be appropriately characterized by the nationwide surveys because of the limited number of individuals surveyed and responded. In general, the risk of acute exposure for each population subgroup is evaluated based on the high-end of daily exposures. For example, DPR uses the 95th percentile of exposure. The non-oncogenic risk for chronic non-lifetime exposures for each population subgroup is evaluated based on the average daily exposures. The acceptability of risks for the whole population is then based on the population subgroup(s) with the highest exposure levels. These are often the exposures of infants and children subgroups. The lifetime oncogenic risk is evaluated based on the exposure averaged over a lifetime.

In addition to the evaluation of exposure from food, DPR also addresses exposure of pesticides from other sources when data are available. USEPA currently does not routinely consider exposure via multiple routes of exposure.

In response to the NAS Report, federal agencies have already proposed plans to improve the consumption and residue data. DPR is currently working with USEPA on the proposed plans. In an ongoing effort to harmonize pesticide regulations between the state and federal agencies, DPR has also been working with USEPA to compare and coordinate the respective dietary exposure assessment processes.

PECC recommends that further refinement of the current dietary exposure assessment process considers the following:

  1. PECC recommends that DPR should consider the supplemental use of existing smaller-scale surveys (such as the Continuing Survey of Food Intake by Individuals [CSFII] and National Health and Nutrition Examination Survey [NHANES]), as well as further subdivision of the 1-6 years old children subgroup into 1-3 and 4-6 years old subgroups when data are adequate.

  2. The federal agencies and DPR should reorganize the components of residue monitoring programs which generate data for dietary exposure assessment to provide representative residue profile of foods generally consumed by infants and children. These monitoring programs should also consider using lower detection limits at levels pertinent to risk assessment. They should also consider, besides composite analyses, determining residues in individual units of commodities, when appropriate.

  3. DPR should coordinate with federal agencies in the development of a national residue database.

  4. DPR and DHS should collect more representative samples of drinking water.

  5. DHS should evaluate how best to address the NAS recommendation for increased pesticide residue testing of processed foods, particularly those consumed in large quantities by infants and children.

  6. DPR should establish criteria for the use of information on the percentage of crop treated in adjusting regional residue profiles for chronic exposure assessment.

  7. DPR should use a distributional approach for characterizing dietary exposure whenever data are adequate. The approach should also consider (a) methods to estimate residue levels of samples below the detection limits, and (b) the validity of using residue values exceeding the tolerances.

  8. DPR should continue to evaluate multiple routes of exposure for all population subgroups including infants and children. When data become available, DPR should expand the scope of the assessment regarding both the possible routes of exposure and other population subgroups of concern (e.g., farm children).

  9. USEPA should routinely conduct assessments of acute exposures for all pesticides. USEPA should also address multiple routes of exposure and consider requiring the submission of appropriate exposure data.

IV. RISK ASSESSMENT ISSUES

Risk assessment is a complex process which involves the extrapolation of toxicity data from animal studies to humans and the use of default assumptions in areas where information is incomplete. The approaches and methods used in dietary risk assessment continue to be revised as more information becomes available. Many issues highlighted in the NAS Report are applicable to all population subgroups and are not limited to addressing the risks of infants and children to pesticide exposures.

A. Pharmacokinetics and Physiologically-Based Pharmacokinetic Models

Pharmacokinetics and physiologically-based pharmacokinetic models are used to estimate the concentration of the chemical and/or its active metabolite(s) at the target tissue of toxicity in laboratory animals and humans and, thereby, provide better estimates of the exposure. The NAS Report concluded that there is a need for validated models for predicting the health risks to infants and children from dietary exposures to pesticides.

PECC recommends that OEHHA should develop general Cal/EPA guidelines for the use of physiologically-based pharmacokinetic models. DPR should work with USEPA to establish guidelines for these models to be used in pesticide risk assessments. The necessary biochemical, physiological, and pharmacokinetics parameters should be obtained.

B. Uncertainty Factors

Uncertainty factors are used to account for the inherent uncertainties in the toxicological database as well as the response within and between species. Threshold exposure standards, such as the Reference Dose, are calculated by applying uncertainty factors to scale downward the exposure level at which no effects are observed in a toxicity study. Usually, a no-observed-effect level determined in animals is lowered by a 100-fold uncertainty factor to determine a threshold exposure standard for humans. This factor of 100 includes the assumption that humans can be 10 times more susceptible than the most susceptible animal species tested at the most sensitive endpoint of toxicity, and that there could be up to 10-fold variation in susceptibility between different people. The NAS Report recommended that an additional uncertainty factor of up to 10 should be considered for postnatal developmental toxicity and when data from toxicity testing relative to children are incomplete.

PECC concludes that DPR should coordinate with USEPA to assess the routine use of an additional uncertainty factor as defined in the NAS Report. General guidelines must be established to ensure consistency in its application.

C. Dose-Response Models

Depending on whether the toxicological effects are oncogenic (causing tumor or cancer) or non-oncogenic, separate sets of default assumptions have been used to describe the relationship between the level of exposure (or dose) and the magnitude of response. For non-oncogenic effects, it is assumed that the exposure has to reach a certain level (a threshold dose) before an effect can be expected. For oncogenic effects, the default assumption is that a threshold of exposure does not exist. In another words, any increase in the exposure level will result in increased probability of tumor occurrence. Four areas highlighted by the NAS Report are discussed below.

1. Model for Non-Oncogenic Effects

For the assessment of non-oncogenic effects, the NAS Report recommended that the benchmark dose approach should be considered. In this approach, the relationship between the level of exposure and the magnitude of response is first described by a mathematical model. Then, a dose corresponding to a "benchmark response" (a low level of response, e.g., five percent, considered to be either lowest-observable or non-observable response level) is estimated by the model. This benchmark dose approach has been used occasionally by DPR when a toxicity study does not include a dose level at which effects are not observed.

PECC recommends that DPR should continue the current use of the benchmark dose approach. General Cal/EPA guidelines should be established by OEHHA to ensure the consistency of application. DPR should also work with USEPA to establish guidelines for this approach in the evaluation of pesticides.

2. Model for Oncogenic Effects

For assessing oncogenic effects, the NAS Report recommended that the two-stage clonal expansion model should be considered. The model provides a possibility for incorporating knowledge concerning the biological processes of tumor formation such as cell transformation and cell turnover. The model also can be used to accommodate differential susceptibilities during various stages in life (e.g., early age). However, much of the necessary information about tumor formation processes and parameters (e.g., target cell numbers, rate of cell transformation) for the model has not been developed.

PECC recommends that, as the necessary information becomes available, the use of the two-stage clonal expansion model should be considered.

3. Age-Specific Susceptibility

Existing data show that infants and children may be more or less susceptible to effects of oncogenic chemicals. The NAS Report discussed the application of models to accommodate the age-specific susceptibility. The NAS Report used hypothetical scenarios in which higher risk may be possible if the exposure and susceptibility during early stages in life is higher than during later stages in life. However, a practical application of this type of analysis would require development of chemical-specific information on exposure and susceptibility for each stage in life for both laboratory animals and humans.

As necessary data become available, PECC recommends that the susceptibility issue should be appropriately addressed.

4. Interspecies Dose Extrapolation for Oncogenic Effects

For oncogenic effects, different approaches to extrapolate dose levels from laboratory animals to humans have been used by various federal and state regulatory agencies. Each approach has its underlying scientific support.

PECC recommends that OEHHA should pursue a unified approach for the interspecies dose extrapolation within California. DPR should coordinate with USEPA in deriving a unified approach for the evaluation of pesticides.

D. Multiple-Chemical Exposures

Individuals may be exposed to more than one pesticide simultaneously. The exposure scenarios are dependent on the amount of residues remaining on the commodity at the time of consumption and the potential for co-occurrence of pesticides in the diet. The NAS Report illustrated an approach of using the toxicity equivalence factor to evaluate the total risk of exposures to multiple chemicals with the same mechanism of action.

PECC recommends that DPR should continue to explore all possibilities in determining the most appropriate approach to address dietary risk of multiple pesticide exposures. DPR should coordinate with USEPA on this issue.

E. Inert Ingredients

Inert ingredients are used in pesticidal products and, if they remain in the commodities at harvest, may present a health concern. USEPA has placed existing inert ingredients in groups based on their known toxicity and the need for additional toxicity testing. USEPA is concentrating its attention on the higher priority inert ingredients. Of some 50 substances identified by USEPA as presenting potential toxicological concern, all but a few have now been eliminated by registrants from their products. In the interim, manufacturers must relabel products to identify the presence of these toxic inert ingredients. In addition, DPR has developed an extensive database on the use of these inert ingredients in formulated products. DPR also plans to initiate studies to investigate the environmental fate of inert ingredients.

PECC recommends that the information on use, toxicity, and the potential for residues remaining on commodities should be considered in prioritizing inert ingredients into the existing risk assessment activities.

F. Tolerance Assessment

Tolerances for pesticides on food and feed are established by USEPA for the active ingredient, metabolites, and degradation products of concern. They are based on agricultural practices and evaluated for potential hazards to human health. USEPA assesses dietary exposure and tolerances using a tiered approach. The first step is an evaluation assuming the exposure was to residue at the tolerance. If the resulting estimated risk appeared not acceptable, then a second tier of analysis is conducted using other kinds of residue data (e.g., field trial, monitoring, processing, etc.). The information on percentage of crop treated may be used to adjust residue levels for chronic exposure assessment. USEPA has made legislative proposals to change the tolerance-setting process to place more emphasis on health considerations.

To evaluate whether exposures at the tolerance level present an acute health risk, DPR evaluates risk to population subgroups, including infants and children, at the tolerance level for each individual label-approved commodity. In the assessment, the consumption of all food forms associated with the specific raw agricultural commodity are also included in the exposure analysis. Acute and chronic exposure assessments of multiple commodities, all at tolerances, are not addressed because the occurrence of such scenarios would be highly improbable. For the same reason, chronic exposure to a single commodity at tolerance level is also not addressed.

DPR also conducts expedited acute dietary risk assessments to evaluate risk from the consumption of commodities containing residues over the tolerance level. Whenever an incident of overtolerance is determined to present a potential unacceptable risk of exposure, DPR works with DHS, OEHHA, and USEPA to determine appropriate action.

PECC believes that tolerances should be based on health with considerations for agricultural practices. DPR should review the current and future updates of the USEPA tolerance-setting process. OEHHA, under the mandate of Assembly Bill 2848, should continue to evaluate tolerances for pesticides in processed foods.

V. SUMMARY OF EVALUATION

In previous sections, current federal and state pesticide regulatory programs on food safety are described. For the identification of potential health hazard, toxicological data are evaluated by USEPA and DPR for pesticide registration. In addition to the health evaluation, both federal agencies and DPR have monitoring programs to ensure compliance. The DPR regulatory program also include enforcement, full pesticide use reporting system, and illness reporting system to ensure the safe use of pesticides.

DPR evaluates risks of dietary exposures to pesticides using two separate processes: the assessment of detected residue levels, and the assessment of residues at the established tolerances. The approaches for the two processes are different based on the nature of the two sets of residue data. Both the acute and chronic exposures to residues at the detected levels are routinely evaluated.

Under the Food Safety Act, DHS has the responsibility for monitoring processed foods for pesticide residues and other contaminants. Because of budget constraints, the monitoring program for processed foods has been limited, with low numbers of samples being monitored. The NAS Report recommends increased pesticide residue testing of processed foods, particularly those consumed in large quantities by infants and children. DHS is considering how best to address this concern.

OEHHA had previously initiated program activities to evaluate the adequacy of USEPA tolerances for processed foods. Due to staff redirection (i.e., metam sodium spill at Dunsmir and Medfly program) and budget cuts, no further evaluations have been conducted. Summary reports were drafted. To date, no recommendations for new tolerances have been made. Therefore, full evaluation of the OEHHA program for processed food tolerances can not be made at this time.

Both USEPA and DPR conduct dietary exposure assessments using similar approaches. In a limited comparison, DPR has identified two specific areas of differences in the two programs: (1) USEPA does not routinely conducts acute exposure assessment for all pesticides, and (2) USEPA does not evaluate multiple routes of exposure, dietary as well as non-dietary routes.

Based on the considerations discussed in this report, PECC concluded that given the available databases, the current California and federal programs of food safety adequately protect infants and children from unacceptable risks posed by pesticide residues in the diet. However, questions remain in areas where data are incomplete or absent. It is not known how the susceptibility of infants and children compares to that for adults, for which pesticide toxicity profiles are estimated from studies in animals. It is also not known whether health-based standards established for protecting against the endpoints established under current data requirements will also protect against additional toxicity endpoints for which tests have not previously been required. In addition, the effects of exposures to more than one pesticide in the diet will not be known until a scientific approach can be established for the evaluation. On the other hand, the estimated dietary exposures may have been unrealistically high because the limitations of current data necessitate the repeated use of high values and the use of assumptions that tend to yield high-exposure estimates. Therefore, the conclusion that the current California and federal programs adequately protect infants and children can only be drawn based on the currently available data. By answering the remaining questions and, with additional data, further refining the dietary risk assessment process and methodology, there would be even greater assurance that the state and federal programs adequately protects infants and children.

VI. SUMMARY OF RECOMMENDATIONS AND IMPLEMENTATION

As a result of the review of the current programs, PECC made recommendations regarding: (1) additional toxicity data and further investigation of the susceptibility of infants and children to toxicological insults; (2) additional data on food consumption and residue monitoring for better characterization of dietary exposures of the population, including infants and children; (3) development of risk assessment methodology for addressing various areas of concern regarding pesticide exposures, and (4) exploration of new risk assessment approaches that may be applicable as additional scientific information becomes available.

The following PECC recommendations are specific to federal agencies, Cal/EPA-DPR, Cal/EPA-OEHHA, and DHS. These recommendations build upon the foundation of existing programs and are directed toward respective agencies and departments with specific mandates on food safety evaluation. These recommendations, when implemented, should provide further assurance that human health will be protected from risks of dietary pesticide exposures. General guidelines for risk assessment will need to be developed. PECC recognizes that under current Cal/EPA policy, OEHHA has the responsibility to develop guidelines for risk assessment. The discussions of issues pertaining to general risk assessment guidelines should continue in the restructured standards and criteria work group chaired by OEHHA in accordance with Senate Bill 1082. DPR will work with USEPA on pesticide testing and risk assessment guidelines, utilizing appropriate Cal/EPA guidelines, as they become available. DPR will coordinate with OEHHA and/or DHS on issues which fall within their statutory responsibilities.

A. Federal Agencies

USEPA has formed six working groups within the federal regulatory agencies for studying the implementation of the NAS Report recommendations. PECC recommends that federal program goals should be to:

  1. Identify research needs for obtaining comparative data regarding the manifestation of toxicity between immature and mature animals for representative pesticides.

  2. Finalize toxicity testing guidelines for neuro-developmental, immuno-, visual, and endocrine toxicities. When determining additional data requirements for pesticide registration, the anticipated use of data in risk assessment must be considered.

  3. Require neurotoxicity testing for all food-use pesticides.

  4. Consider the supplemental use of the existing smaller-scale consumption surveys (such as the CSFII and NHANES).

  5. Develop a nationwide residue database for dietary exposure assessment.

  6. Investigate the use of distributional approach for characterizing exposures.

  7. Explore the use of the benchmark dose approach for developmental effects.

  8. Routinely conduct the assessments of acute exposures for all pesticides. USEPA should also address multiple routes of exposures and consider requiring the submission of appropriate exposure data.

C. Cal/EPA - DPR

  1. DPR should work closely with USEPA to identify the necessary toxicity data and to provide inputs regarding testing guidelines.

  2. DPR should coordinate with USEPA to obtain information on physiological and pharmacological differences both among infants, children, and adults, and between humans and animals.

  3. DPR should consider the applicability of the testing guidelines being developed by USEPA to the risk assessment guidelines being coordinated by OEHHA for Cal/EPA, as they become available.

  4. As an interim measure before more comprehensive consumption data can be collected, DPR should consider the supplemental use of existing smaller-scale surveys (such as the CSFII and NHANES), as well as further subdividing the children subgroup into 1-3 and 4-6 years when data are adequate.

  5. DPR should reorganize the component of its residue monitoring program which generates data for dietary exposure assessment to provide representative residue profiles of foods generally consumed by infants and children.

  6. DPR should ensure that residue detection limits are set at levels pertinent for risk assessment and develop methods to estimate residue values below the detection limit.

  7. DPR should determine the residue levels in composite as well as individual units of commodities when appropriate.

  8. DPR should coordinate with federal agencies in the development of a national residue database.

  9. DPR should collect more representative samples of drinking water.

  10. To refine the estimation of residue levels for chronic exposure assessment, DPR should establish criteria for the use of information on percentage of crop treated in adjusting the regional residue profiles. The data on the percentage of crop treated are available through California's unique full pesticide use reporting system.

  11. DPR should use a distributional approach for characterizing dietary exposure whenever data are adequate. The approach should also consider (a) methods to estimate residue levels of samples below the detection limits, and (b) the validity of using residue values exceeding the tolerances.

  12. DPR should continue to evaluate multiple routes of exposure for all population subgroups including infants and children. When data become available, DPR should expand the scope of the assessment regarding both the possible routes of exposure and other population subgroups of concern (e.g., farm children).

  13. DPR should work with USEPA in establishing guidelines for physiologically-based pharmacokinetic models and the benchmark dose approach to be used in the evaluation of pesticides.

  14. DPR should coordinate with USEPA to assess the routine use of an additional uncertainty factor to address postnatal developmental toxicity and when data from toxicity testing relative to children are incomplete.

  15. DPR should consider the applicability of biologically-based models (i.e., two-stage clonal expansion model) in addressing age-specific susceptibility for oncogenic effects, when necessary information becomes available.

  16. DPR should coordinate with USEPA in deriving a unified approach for the evaluation of pesticides.

  17. DPR should develop guidelines and methods for addressing risks from multiple routes of pesticide exposures. DPR should also explore all possibilities in determining the most appropriate approach to address exposure to multiple pesticides.

  18. DPR should consider the information on use, toxicity, and the potential for residues remaining on commodities in prioritizing inert ingredients into the existing risk assessment activities.

  19. DPR should review the current and future updates of the USEPA tolerance-setting process.

  20. In addition to what can be done within California, DPR should continue to coordinate with USEPA on federal plans for implementing NAS recommendations through active participation in the six implementation working groups.

D. Cal/EPA - OEHHA

  1. OEHHA should continue to evaluate tolerances for pesticides in processed foods.

  2. OEHHA should give higher priority to the development of general Cal/EPA risk assessment guidelines, in coordination with appropriate departments within the agency. These guidelines may include physiologically-based pharmacokinetic models, benchmark dose approach, and interspecies dose extrapolation approach.

E. DHS

  1. DHS should evaluate how best to address the NAS recommendation for increased pesticide residue testing of processed food, particularly those consumed in large quantities by infants and children.

  2. DHS should collect more representative samples of drinking water.

CONCLUSION

In conclusion, this report describes approaches used by federal and state agencies to evaluate the safety of pesticides in the diet, especially in the diets of infants and children. Based on the considerations presented in this review, the current California and federal pesticide regulatory systems adequately protect infants and children from risks posed by pesticide residues in the diet. However, there are potential areas for improvement of the state's pesticide regulatory program. Uncertainty remains in areas where data are incomplete or absent. In addition, the scientific basis of some of these issues requires further research and discussion. Many of the PECC recommendations are consistent with those in the NAS report. These recommendations, if adopted, will provide a better scientific basis for approaches and assumptions currently used. This, in turn, will improve the accuracy of the risk estimates and allow greater consensus to be achieved concerning the magnitude of pesticide risks in the diet.

The implementation of these recommendations in California can proceed in parallel with the plans at the federal level. The monetary cost of implementing some recommendations is expected to be significant. In addition, considerable scientific effort will be required. In this era of limited resources, choices must be made on the benefits to be derived from additional testing requirements and data acquisition.

This report should also serve to provide the framework of scientific discussions between participating agencies in California that are involved in food safety programs. DPR, in its role as the lead agency for pesticide regulation, should continue its on-going project with USEPA to achieve greater harmony in pesticide regulatory programs. In this capacity, DPR can serve as liaison between the state and federal agencies.


Contents

Background

Chapter 1. Introduction

A. Pesticide Registration

B. Evaluation of Food Safety

C. The National Academy of Sciences Report

Chapter 2. Toxicity Assessment

A. Introduction

B. Recommendations by the NAS Committee

C. Databases Available

D. Current Approaches

E. Discussions and Future Direction

Chapter 3. Dietary Exposure Assessment

A. Introduction

B. Recommendations by the NAS Committee

C. Databases and Computer Programs Currently Used

D. Current Approaches

E. Discussions and Future Direction

Chapter 4. Risk Assessment Issues

A. Introduction

B. Pharmacokinetics and Physiologically-Based Pharmacokinetics

C. Uncertainty Factors

D. Dose-Response Models

E. Multiple-Chemical Exposures

F. Inert Ingredients

G. Tolerance Assessment

H. Conclusions

Chapter 5. Summary of Evaluation

A. Highlights of Current Programs

B. Remaining Questions

C. Analysis of Current Programs

Chapter 6. Summary of Recommendations and Implementation

A. Summary of Recommendations by the PECC

B. Implementation of Recommendations

Conclusion

Appendices

A. List of Abbreviations

B. Glossary of Terms

C. List of Contributors of Public Comments


Background

Assembly Bill 2161 (Chapter 1200, Statutes of 1989, sometimes called the Food Safety Act of 1989 [Food Safety Act]) added and expanded several sections in the Food and Agricultural Code and the Health and Safety Code. The bill allowed the creation of additional regulatory efforts that build upon and enhance California's food safety programs to ensure that food continues to meet existing rigorous standards. The statute required increased priority pesticide residue monitoring; established a scientific advisory committee to review pesticide residue analytical methods; established a committee to fund research into alternative pest management practices; made possible expanded pesticide use reporting; required risk assessments on the dietary exposure to pesticides in both raw and processed foods; provided authority to call-in acute toxicity studies where needed to support the risk assessments; required the state Department of Health Services (DHS) to commence a processed foods monitoring program; and required private laboratory accreditation and reporting by private laboratories of findings of illegal pesticide residues in the channels of trade. The bill also established a number of funding mechanisms that were later augmented by a General Fund budget enhancement.

The Food Safety Act also required the Department of Pesticide Regulation (DPR) and the Office of Environmental Health Hazard Assessment (OEHHA), both of the California Environmental Protection Agency (Cal/EPA), to jointly review the existing federal and state pesticide registration and food safety system and determine if the existing programs adequately protect infants and children from dietary exposure to pesticide residues. DPR was to consult with the University of California and other qualified public and private entities in conducting the joint review. The review was also to include an evaluation of the National Academy of Science (NAS) study on the scientific and policy issues concerning pesticides in the diets of infants and children. The study was commissioned by the U.S. Congress and U.S. Environmental Protection Agency (USEPA) in 1988. DPR was to report to the Legislature on the results of the review including recommendations for modification of the existing regulatory system in order to adequately protect infants and children.

California has a special interest in the NAS Report. California already has the most comprehensive state food safety program in the nation. The program begins with evaluation of pesticide safety, and follows through with enforcement of pesticide use in the field and residue monitoring. While the NAS study was being conducted, many of the issues raised in Food Safety Act were under discussion among California regulators. These issues include the use of dose-response models, interspecies extrapolation factors, multiple-chemical exposures, and the use of percentage of crop treated information to adjust the residue profiles. With the Academy's release in June 1993 of its report, Pesticides in the Diets of Infants and Children (NAS Report), work began on the completion of this report, as required by the Food Safety Act and incorporating finds in the NAS Report. The long-awaited report received widespread coverage in the news media, was the subject of Congressional hearings, and elicited many comments from persons representing public interest groups, academia, the chemical industry, and grower groups.

Cal/EPA organized a working group, called the Pesticide Exposure to Children Committee (PECC) and chaired by DPR, which included representatives from OEHHA, DHS, California Department of Food and Agriculture (CDFA), University of California at Davis (UCD), and USEPA Region IX. During the preparation of the report, DPR convened four formal meetings with PECC members to discuss the outline of the report and the content. In addition, there were numerous informal discussions between DPR members and other members. Three drafts of the report were circulated among the members for comments. The staffs from the Executive Office of DPR, OEHHA, DHS, and CDFA met to make final decisions on issues.

Cal/EPA also solicited public comment on the NAS Report, and held a hearing on September 3, 1993 at which interested parties were invited to testify. By the time the comment period ended on September 8, 1993, the interagency working group received a wide variety of comments. These comments were taken into consideration in preparing this report.

Scope of This Report:

This document focuses on programs and issues pertaining to the evaluation of dietary exposure of infants and children to pesticides. PECC recognizes that pesticide residues in foods can be reduced through efforts in pesticide use reduction, reduced-risk pesticide use, and alternative pest management strategies. Although discussions of these efforts are beyond the scope of this review, three major programs should be noted because they could well have a significant impact in reducing the dietary risk of pesticides:

  1. USEPA recently issued a Voluntary Reduced-Risk Pesticide Initiative (USEPA, 1993) and is planning an use/risk reduction initiative. The purpose of these initiatives is to encourage the registration of lower risk pesticide products containing new active ingredients.

  2. DPR has begun to carry out many of the recommendations in Challenge and Change. A Progressive Approach to Pesticide Regulation in California (Benbrook and Marquart, 1993). Cal/EPA commissioned the study as a critical evaluation of the pesticide registration process in California, and to recommend ways to move toward registration and use of safer and more environmentally benign pest control systems.

  3. DPR and CDFA have formed the Pest Management Advisory Committee to evaluate alternative pest management strategies and needs, and to find ways to reduce the environmental burden associated with pesticide use.

These and other efforts to increase the use of reduced-risk chemicals and to find pest management strategies that reduce overall pesticide use will help further reduce the risk associated with dietary exposure to pesticides, reinforcing the recommendations outlined in the NAS Report and in this document.

PECC agrees that a comprehensive effectiveness evaluation of the food safety regulatory program should be considered in the future. The scope of the evaluation should be defined by the restructured Pesticide Advisory Committee, chaired by DPR.

Organization of this Report:

This document is organized into six chapters and follows the subject sequence in the NAS Report.

Chapter I, INTRODUCTION, is a general discussion on the state and federal food safety programs and the recommendations in the NAS Report.

Chapter II, TOXICITY ASSESSMENT, examines the adequacy of the toxicity study requirements for registration.

Chapter III, DIETARY EXPOSURE ASSESSMENT, reviews the approaches used by regulatory agencies in assessing dietary exposures to pesticides.

Chapter IV, RISK ASSESSMENT ISSUES, discusses issues which may impact the risk assessment of dietary exposures.

Chapter V, SUMMARY OF EVALUATION, evaluates the adequacy of the current programs for the protecting infants and children from exposure to pesticide residues in food.

Chapter VI, SUMMARY OF RECOMMENDATIONS AND IMPLEMENTATION, summarizes the recommendations in previous chapters of this document and organizes them based on approaches used in existing programs.

The appendices are: Appendix A - List of Abbreviations; Appendix B - Glossary of Terms; and Appendix C - List of Contributors of Public Comments.

Back to Table of Contents


Chapter 1. Introduction

Introduction

This introductory chapter describes in general terms the state and federal pesticide registration and food safety evaluation processes and the recommendations in the NAS Report. Specifics on the evaluation of dietary exposures to pesticides and the proposed implementation of the NAS Report recommendations are addressed in subsequent chapters.

A. PESTICIDE REGISTRATION PROGRAMS

"Pesticide" is a general term used to describe substances that control pests. The target "pest" is any living organism that causes damage or economic loss, or transmits or produces disease. Therefore, pesticides include insecticides, fungicides, herbicides, rodenticides, disinfectants, as well as insect and plant growth regulators.

Because all pesticides are designed to be toxic to their target organisms and can present a risk to nontarget organisms if used improperly, pesticides are strictly regulated by the state and federal government. A pesticide must be registered by both USEPA and DPR before it can be offered for sale, possessed, or used in California. The goal of the registration programs at USEPA and California is to ensure proper, safe, and efficient use of pesticides, and to eliminate pesticides which produce harmful effects.

1. U.S. Environmental Protection Agency

USEPA is responsible under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) to register (or license) pesticide products on the basis that when used according to label directions they will not result in any unreasonable adverse effects to human health or the environment. FIFRA requires USEPA to balance the risks of pesticide exposure to human health and the environment against the benefits of the pesticide use to society and the economy. USEPA, under FIFRA, regulates pesticide use in many ways.

Experimental Use Permits (Section 5): Experimental use permits (EUP) are granted to permit gathering information necessary to register a pesticide product. Section 5 of FIFRA provides that any person wishing to test an unregistered pesticide or a registered pesticide for an unregistered use may apply for an EUP. The purpose of an EUP is to facilitate the generation of data necessary for registration and yet provide sufficient regulatory control to prevent adverse environmental effects. An EUP authorizes limited use of a pesticide on a limited number of acres, under specific controlled conditions, to develop the necessary data. Applicants for an experimental pesticide use that may result in residues occurring in or on food or feed must also apply for a tolerance or exemption from tolerance, or certify that the affected crop will be destroyed and not available to consumers.

Pesticide Registration (Section 3): Under FIFRA, USEPA bases registration decisions for pesticides on its evaluation of a battery of test data provided primarily by applicants for registration. Detailed data requirements for pesticide registration are described in the Code of Federal Regulations Title 40, Part 158 (Code of Federal Regulations, 1992). For most pesticides, the requirements include studies in product chemistry, acute and chronic toxicology, residue or dietary chemistry, farm worker exposure, wildlife and non-target organism toxicology, and environmental effects.

Of the studies required for registration, toxicology data in mammalian species and residue chemistry data are generally the most pertinent in the evaluation of the safety of the pesticide residues on foods. Detailed discussions on these subjects are presented in Chapters II and III. A product may be classified for restricted use only by certified applicators, if it warrants special handling due to its toxicity.

Tolerance Requirement: If a pesticide is used on a food or feed crop, a tolerance (defined as the legal maximum residue concentration of pesticide allowed) or exemption from the requirement of a tolerance is required. For any pesticide use involving food or animal feed crops, USEPA is responsible under the Federal Food, Drug, and Cosmetic Act (FFDCA) for setting tolerances for pesticide residues in raw agricultural commodities (Section 408 of FFDCA) and processed commodities (Section 409 of FFDCA). USEPA uses the established tolerances of foods to estimate total dietary exposures to pesticides. A more detailed discussion on the issue of tolerances is presented in Chapter IV.G. (Tolerance Assessment).

Pesticide Reregistration: USEPA is required by law to reregister existing pesticides that were originally registered prior to 1984, when standards for approval and scientific testing requirements were less stringent than they are today and before current scientific and regulatory standards were formally established. This comprehensive re-evaluation of pesticide safety was accelerated by the 1988 amendments to FIFRA and is required to be completed by 1997. The reregistration process ensures that up-to-date databases are developed for these older chemicals and that all conditions of registration, including labels and tolerance reassessments, are reflective of contemporary science.

Emergency Exemption and Special Local Needs (Section 18 and Section 24[c]): Section 18 of FIFRA authorizes USEPA to exempt state and federal agencies from any provision of FIFRA if it is determined that an emergency condition exists. Emergency exemptions allow the use of an unregistered pesticide to treat a sudden or significant pest infestation. The four types of exemptions are: specific exemption, quarantine exemption, public health exemption, and crisis exemption. Section 18 provides the authority for state and federal agencies to issue crisis exemptions in situations involving unpredictable emergency situations when there is insufficient time to allow USEPA to review and authorize an emergency exemption. FIFRA Section 24(c) authorizes a State to register additional unregistered uses of federally registered pesticides for distribution and use within its borders to meet a special local need. No special local need registration for a food or feed use can be issued unless a tolerance or tolerance exemption under FFDCA has been granted.

Special Review: USEPA has instituted a formal procedure for determining whether the use of a registered pesticide poses an unreasonable risk to humans or the environment. This Special Review process is set into motion when USEPA has reason to believe that the use of a pesticide poses a greater risk than previously determined. Special Review entails the call in of additional or customized studies considered necessary to evaluate the safety and benefit associated with the use of the pesticide of concern and the availability of alternatives. A Special Review can result in a decision to continue the use or a decision to cancel, restrict, or modify the use in order to reduce the overall risk to an acceptable level. This process involves opportunities for public comments and participation in the decision-making process. While Special Review is being conducted, the pesticide remains on the market and may be used until a regulatory decision is made

2. California Agencies

In addition to the registration process at the federal level, California independently evaluates pesticides under its own registration programs. The following section describes the programs.

Pesticide Registration: DPR is responsible for registering and regulating the use of pesticides in California. Beyond the data required by USEPA, DPR may require additional studies to address specific areas of concerns or the unique growing conditions in the state. DPR may require supplemental information for the analysis of residues in commodities. A food-use registration cannot be approved unless there is a residue analytical method that can be completed within 24 hours for the proposed use. DPR also requires extensive environmental data that are designed to test a chemical's potential to leach into ground water. The submitted data are reviewed and used by state scientists to evaluate various routes of exposure such as occupational, residential, dietary, and ambient air, in order to ensure the safe use of pesticides in California.

Risk Evaluation: Risk assessments to evaluate pesticides for potential adverse health effects are conducted under various authorities and mandates. General authority for risk assessment as part of pesticide regulation is in the Food and Agriculture Code. Statutory mandates include the Food Safety Act, the Birth Defect Prevention Act of 1984 (Senate Bill 950, [Birth Defect Prevention Act]), and Assembly Bill 1807 (sometimes called the Toxic Air Contaminant Act of 1983, [Toxic Air Contaminant Act]). The pesticides are prioritized for evaluation based on toxicity and potential for exposure, with priority given to those which are considered to be highly toxic. Regardless of the route of exposure, when the risk is unacceptable, DPR has the authority to mitigate the exposure by changing use practices through regulation or permit conditions. If the risk cannot be mitigated, DPR may suspend or cancel the product registration.

In addition to the food safety evaluation program described in this report, California has the nation's strictest and most comprehensive worker protection program. Before a chemical can be registered, DPR requires studies to be conducted indicating the potential amount of exposure to persons mixing and applying the pesticide, as well as workers and the public who are likely to enter treated areas.

These and other data are the basis for determining potential risk and adequate margins of safety for establishing reentry intervals for workers, protecting clothing requirements, and use restrictions which are frequently more stringent than those on the USEPA-registered pesticide label.

To ensure safe use in the field, California has the nation's largest and best-trained field enforcement organization. California is the only state in the nation with agricultural commissioners in each county who have the authority to enforce pesticide laws. In addition, DPR conducts field studies each year to monitor pesticide exposure to workers performing routine tasks to determine if additional measures are necessary to mitigate any unacceptable exposures. Pesticide mixer/loader/applicators of Toxicity Category I (the highest acute toxicity category as designated by USEPA) pesticides are required to be tested for cholinesterase activity depending on extend of exposure.

DPR's program to protect workers also includes mandatory safety training for all pesticide handlers, strict licensing and certification requirements for applicators. Under state law, all illnesses suspected of being related to pesticide use must be reported to state authorities (a more detailed description is in Chapter II.C. Databases Available). DPR, in consultation with OEHHA, has the primary responsibility on the development of regulations relating to pesticide and worker safety.

Pesticide Reevaluation: DPR may reevaluate the use of a pesticide when new data show additional potential for concern, such as health hazards. These new data may be the result of routine monitoring programs (for example on produce, worker, and ambient air) conducted both at the federal and state levels, illness incidents, and additional data submitted under various state mandates (such as Birth Defect Prevention Act, Food Safety Act, Air Toxic Contaminant Act; Proposition 65, the Safe Drinking Water and Toxic Enforcement Act of 1986; and the Pesticide Contamination Prevention Act of 1985, Assembly Bill 2021 [Pesticide Contamination Prevention Act]).

Tolerances: California has adopted USEPA established tolerances for food and feeds. DPR and DHS also have the authority, by regulation, to set tolerances for raw agricultural commodities and processed foods, respectively, if the established tolerances are found to be unacceptable as a result of the evaluation by DPR on raw agricultural commodities and by DHS on processed foods.

B. EVALUATION OF FOOD SAFETY

Federal and California agencies evaluate the safety of pesticides in the food supply following the mandates of their respective laws. The following is a general discussion of the regulatory mandates. Specific details on the residue monitoring programs and the evaluation process are in Chapter III (Dietary Exposure Assessment) and Chapter IV.G. (Tolerance Assessment).

1. Federal Agencies

USEPA regulates the safety of pesticides in the food supply by setting tolerances for pesticide residues in or on foods and animal foods sold in commerce. The tolerance program is intended to ensure that consumers are not exposed to unsafe levels of pesticide residues from their foods.

To assess the overall short and long-term dietary risk of a pesticide, USEPA uses data on toxicological effects, dose levels, residue studies, as well as data on the types and amounts of foods people eat. In its risk assessments, USEPA takes into account the eating habits of various subgroups of the U.S. population, e.g., infants and children, or people of Hispanic origin.

The U.S. Food and Drug Administration (FDA) and the U.S. Department of Agriculture (USDA) monitor the food supply and enforce the tolerances established by USEPA for domestically produced and imported foods.

2. California Agencies

In California, the food safety program involves the data review and dietary risk assessment programs, as well as use enforcement and residue monitoring.

Under its general regulatory authority and the Food Safety Act, DPR evaluates the safety of pesticide residues in fresh produce and processed foods in the total diet, and examines the established tolerance levels for raw agricultural commodities. Because of the diversity in food consumption patterns in a population, DPR routinely addresses the exposure of population subgroups based on region, age, sex, race-ethnicity, and seasons (when applicable). The exposures of infants (less than one year old) and children (1-6 years old and 7-12 years old) are evaluated separately from the exposures of adults. In conducting the health risk assessment, all pertinent information submitted to DPR under state mandates, as well as published reports and journal articles, are reviewed.

OEHHA is mandated by Assembly Bill 2848 to evaluate tolerances for processed foods and determine whether they are protective of public health. If they are not protective, OEHHA can develop and recommend new tolerances to be adopted by DHS.

To ensure that foods available in California do not contain residues that may pose an unacceptable risk, DPR has an extensive monitoring program for fruits and vegetables (DPR, 1991 and previous annual reports). More than 10,000 samples are currently collected each year. Residues that are found are usually at levels that are measured at a fraction of a part per million (ppm). Only a very small percentage of samples - about 1 percent - has residues over the tolerance levels. DHS, under the Food Safety Act, also has the mandate to collect residue data for processed foods.

DPR also has enforcement authority to ensure that pesticides are used according to the labels. DPR, working with the county agricultural commissioners, has wide-ranging authority to deal with violators of pesticide laws and regulations. Enforcement options include administrative actions; criminal and civil actions; and crop quarantine, crop seizure, and crop destruction. Administrative actions can be taken by DPR or the county agricultural commissioner to refuse, revoke or suspend the right of a pest control operator to do business, or a farmer to use certain pesticides.

Both the DPR and the commissioners also have the authority to levy agricultural civil penalties to enforce certain pesticide regulations, including those prohibiting the packing, shipping or selling of produce containing illegal pesticide residues.

Criminal and civil proceedings are considered for repetitive or intentional violations, or violations that have created a hazard to human health or the environment. Crop quarantine, crop seizure, and crop destruction are considered for produce with illegal residues, those over the tolerance levels or with residues with no established tolerances. DPR investigates every case of illegal residue detected in California by the DPR and FDA residue monitoring programs.

C. THE NATIONAL ACADEMY OF SCIENCES REPORT

The Committee on Pesticide Residues in the Diets of Infants and Children (NAS Committee) within the National Research Council of the NAS conducted the study and submitted the report Pesticides in the Diets of Infants and Children. The key issues addressed by the NAS Committee in the NAS Report were:

  1. assessment of the current information on pesticide exposure in the diets of infants and children,

  2. adequacy of the current risk assessment methods and policies, and

  3. identification of the toxicological issues of greatest concern and research priority.

The recommendations made by the NAS Committee are outlined and discussed in detail in each respective chapter in this document. The recommendations with key subjects (in boldface) presented in the Executive Summary of the report are summarized below:

  1. Toxicity testing should address the vulnerability of infants and children.

  2. Additional food consumption data for children and infants should be collected within narrow range of age groups.

  3. Regarding residue data,

    1. data should be collected using comparable analytical methods and standardized reporting procedures;

    2. a computerized database containing pertinent information should be established;

    3. sampling of targeted commodities consumed by infants and children should be conducted; and

    4. the effects of processing on residue levels should be addressed.

  4. Regarding exposure,

    1. when data are available, the exposure should be characterized by exposure distributions rather than point estimates.

    2. all routes of pesticide exposures (diet, water, non-diet) should be addressed;

  5. Regarding risk assessment,

    1. physiologically based pharmacokinetic models should be developed for describing the unique features of infants and children;

    2. an additional uncertainty factor up to 10 should be considered when there is evidence of postnatal developmental toxicity and when data from toxicity testing relative to children are incomplete;

    3. a different approach to cancer risk assessment should be considered to account for the potential differences in susceptibility to toxic effects and dietary exposures with age; and

    4. exposures to multiple pesticides should be addressed.

    5. Tolerances should be based more on health considerations than on agricultural practices.

The evaluation of the overall safety of the food supply was not one of the charges of the NAS Committee. Nonetheless, in a statement announcing release of the report, Dr. Philip J. Landrigan, who chaired the NAS Committee, said, "Parents should continue to emphasize fruits and vegetables in their children's diets." Elsewhere in his statement, Dr. Landrigan said, "The goal of our report is to make the very good food supply of the United States even better."

PECC agreed that the NAS Committee has done an excellent job of evaluating all the data relevant to the issue of pesticides and children. While not in agreement with every technical detail of the NAS Report, PECC is in full agreement with the need to take steps to further improve the nation's already very good system of pesticide monitoring, especially for the protection of infants and children. PECC is also in general agreement with many of the methods the NAS Report suggested to improve the system, although PECC believes that further analysis is required to fully assess the economic impact of generating the additional data the NAS Report envisioned.

REFERENCES

Benbrook, C.M. and D.J. Marquart, 1993. Challenge and Change. A Progressive Approach to Pesticide Regulation in California. Prepared for the California Environmental Protection Agency, Department of Pesticide Regulation. Sacramento, CA.

Code of Federal Regulations, 1992. Protection of Environment. Title 40 parts 150 to 186, July 1, 1992. The Office of the Federal Register, National Archives and Records Administration, Washington, D.C.

DPR, 1991. Residues in Fresh Produce - 1991. Pesticide Enforcement Branch, Department of Pesticide Regulation, California Environmental Protection Agency, Sacramento, CA.

NAS Report, 1993. Pesticides in the Diets of Infants and Children, National Research Council, National Academy Press, Washington, D.C.

USEPA, 1993. Voluntary reduced-risk pesticide initiative. Pesticide Regulation (PR) Notice 93-9. Office of Prevention, Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Washington, D.C.

Back to Table of Contents


Chapter 2

Toxicity Assessment

A. INTRODUCTION

Data from toxicological studies form the basis for identifying the potential health hazard inherent in a pesticide. Such data are also necessary for characterizing the relationship between the dose (or exposure) level and the magnitude of biological response. Toxicological data pertinent to human health effects are obtained from observations and studies made in humans as well as in laboratory animals and microorganisms. Because of the uncertainty involved in extrapolating toxicity data from laboratory animals to humans, dose-response relationships delineated in humans are generally preferred for use in the risk assessment process. However, for obvious ethical and practical reasons, human studies are rarely available. Epidemiological studies, when available, may provide useful descriptions of toxicological effects in humans, but they often lack definitive data on exposures. Therefore, data obtained from laboratory animals are most often used.

In this chapter, the types of toxicological data for the evaluation of toxicity and the assessment of dietary risk are described. The chapter begins with a summary of the NAS Report recommendations regarding the toxicity data necessary for addressing the potential dietary risk of pesticide exposures in infants and children. Then, databases currently available for conducting risk assessment are presented. The evaluation of toxicological studies, both for identifying the potential hazard and for characterizing the dose-response relationship is subsequently discussed. Finally, a discussion of the NAS Report recommendations is presented.

B. RECOMMENDATIONS BY THE NAS COMMITTEE

In the course of evaluating the types of toxicological studies routinely required for assessing the potential dietary risk of infants and children, the NAS Committee has identified a number of additional and modified research needs aimed at further ensuring the protection of infants and children from pesticide risks. The recommendations can generally be summarized into the following three areas: (1) age-related variation in susceptibility, (2) age-related pharmacokinetics and physiology, and (3) evaluation of additional toxicological endpoints. (Detailed discussion of these areas are in Chapters 2 through 4 of the NAS Report).

1. Age-Specific Variation in Susceptibility

Concerns about pesticide exposures during perinatal and pediatric periods have been raised because damage sustained during these developmental stages may have long-lasting effects throughout the lifetime. Data for determining the susceptibility of infants and children as well as the stages of differential susceptibility are generally lacking. According to the NAS Report, while data extrapolation may be possible from adults to developing humans, extrapolating data from animals to humans (especially those for developing organisms) may be inaccurate; (NAS Report p.109). This is because "non-primate species are generally less mature at birth than are humans" (NAS Report p.51).

For evaluating the vulnerability of infants and children, the NAS Report recommended that standard toxicity testing protocols in appropriate animal models be established for performing tests during the developmental period and for identifying any adverse effects that may become evident over the lifetime (NAS Report pp.9 and 109). In addition to the toxicity endpoints currently required for evaluation, overall growth (NAS Report p.41) and organ-specific functional measures should also be included (NAS Report p.43). Specific testing recommendations made by the NAS Committee include:

  1. Compare the critical endpoints (e.g., neurotoxicity, immunotoxicity, effects on endocrine system) of several representative classes of pesticides between adult and immature animals (NAS Report p.153).

  2. Include in utero exposures in the regimen for the chronic and oncogenicity studies (NAS Report pp.153 and 154).

2. Age-Specific Pharmacokinetic and Physiological Considerations

Differences in metabolic and kinetic characteristics between infants, children, and adults can affect the ultimate toxicity of a pesticide. Data are lacking not only on age-related pharmacokinetics of pesticides but also on the difference in pharmacokinetic and metabolic characteristics between laboratory animals and humans. The NAS Committee recommended that age-related biochemical and physiological changes and their potential interactions be examined (NAS Report p.43). Specific tests recommendations by the NAS Report include:

  1. Assess the influence of immaturity on pesticide kinetics and toxicity (NAS Report p.110).

  2. Elucidate the metabolism and disposition of pesticides in infants, adolescents, and young adults (NAS Report p.153).

3. Additional Toxicological Endpoints

Except for developmental and reproductive toxicity studies, current pesticide registration guidelines for toxicity studies do not require the examinations of effects as a result of exposures during the perinatal and neonatal periods. The NAS Committee recommended that studies be conducted to develop broader understanding of the principles guiding developmental toxicity, especially in humans, following birth and during critical periods of postnatal development, including infancy and puberty (NAS Report p.109).

In addition, toxicity to the nervous system (NAS Report p.110), immune system, visual system, and endocrine system have either not been rigorously evaluated or are not included in the routine toxicity evaluation protocol for all pesticides. The NAS Committee recommended that testing guidelines be established in appropriate animal models for neuro-developmental (NAS Report pp.110 and 155), immune system (NAS Report p.110), and visual system (NAS Report p.156) toxicity. In addition to establishing guidelines for studies, the NAS Committee made recommendations in the following specific areas:

  1. Neurotoxicity: Testing for acute and subchronic developmental and functional neurotoxicity should be required for all food-use pesticides and testing guidelines should be revised as more information becomes available (NAS Report p.155). The NAS Report provided a modification of the current teratogenicity study protocol for evaluating neuro-developmental effects up to 60 days post partum (NAS Report p.155).

  2. Immunotoxicity: Specific tests should be conducted when histopathological abnormalities are found in spleen, lymph nodes, thymus, and bone marrow (NAS Report p.155).

  3. Endocrine system: Serum thyroid hormones T3, T4 and serum thyroid stimulating hormone (TSH) should routinely be measured in subchronic and chronic/oncogenicity studies (NAS Report p.154).

C. DATABASES AVAILABLE

Three types of toxicological database that are available for evaluating the toxicity of pesticides are presented in this section. Toxicity testing is required for pesticide registration under both federal and California regulatory authorities. Toxicological data available in the open literature and government publications are also considered by DPR in the assessment of the potential risk of dietary exposures. In addition, reports on pesticide-related illness are valuable in identifying toxicological responses when human exposures have occurred.

1. Regulatory Mandates

The studies required by USEPA under the authority of FIFRA and submitted by registrants constitute the principal source of toxicological data for the assessment of human health risk associated with the use of a pesticide, including the risk from dietary exposures for infants and children.

a. USEPA

The toxicological studies required by USEPA for pesticide registration under the authority of FIFRA are published in the Code of Federal Regulations (CFR) Title 40, Sections 158.340, 158.690, and 158.740. For pesticides used on food, data are generally required for: acute (oral, dermal, inhalation) toxicity; primary eye and dermal irritation; dermal sensitization; acute delayed neurotoxicity (in hens); subchronic toxicity; chronic toxicity through dietary exposures (in rodents and non-rodents); oncogenicity through dietary exposures (in two rodent species); teratogenicity (or developmental toxicity; in two species); two-generation reproductive toxicity; genotoxicity (gene mutation, structural chromosomal aberration and other genotoxic effects); and general metabolism. Data on environmental fate and ecological effects are also required. Additional data can be required if needed, or data waived if there is sufficient justification.

b. Cal/EPA - DPR

The California Code of Regulations (CCR, Section 6159) states that DPR director finds that data required by USEPA under Title 40 substantially meet data requirements for the evaluation of pesticide use, including other supplemental data requirements. However, data waived by USEPA are required to be submitted unless justification can be shown and the director makes a written finding to waive the requirement. Registrants are required to submit a full set of valid mandatory chronic health effects studies on: reproductive toxicity, chronic toxicity, genotoxicity, neurotoxicity, oncogenicity, and teratogenicity. These studies have been required for the full registration of pesticides in California since July 1, 1983 (FAC Section 131323). The submission of toxicological data is near completion for the 200 pesticides determined to have the most significant potential for causing adverse human health effects. As with the federal program, data on efficacy, residue chemistry, worker safety, environmental and ecological effects are also required.

Under regulation and mandates of California legislation, DPR may also require submission of additional data. For example, under the Food Safety Act, DPR shall require the submission of additional data on the acute effects of pesticide active ingredients if the available data are insufficient for a reasonable estimation of dietary risk.

2. Electronic Database

Toxicological databases, both factual and bibliographic, are available electronically for search and retrieval. In conducting health risk assessment of pesticide exposures, including dietary exposures, DPR routinely conducts database searches through Toxline Plus (includes Toxline, Toxlit, and Risk Line). Additional databases routinely consulted are: Medline, National Technical Information Service (NTIS), Cancerline, Chemical Carcinogenesis Research Information System (CCRIS), Hazardous Substance Data Bank (HSDB), Integrated Risk Information System (IRIS), and Registry of Toxic Effects of Chemical Substances (RTECS). Studies that provide pertinent toxicological information are also retrieved for review. These data are considered in hazard identification and can be used to characterize the dose-response relationship.

3. Illness Reports

Illness incidents related to pesticide exposures provide post-registration information and serve as support for various regulatory actions. This information is used to evaluate the effectiveness of the DPR regulatory program and to determine the need for enforcement activity or changes in use practices. Information is also used to identify toxicological responses in humans that may differ from those observed in laboratory animals.

a. Federal Agencies

USEPA established the National Pesticide Hazard Assessment Program (NPHA) in the mid-1970's to collect and analyze illness incident data and to conduct follow-up investigations. The computer program, Pesticide Incident Monitoring System (PIMS) was established. PIMS was in operation for approximately six years before termination in 1981. NPHA was dismantled a few years later. In 1992, the Incident Data System (IDS) was implemented to record pesticide incident information (USEPA/USDA/FDA, 1993).

After the registration of a pesticide, registrants are required to report to USEPA any new information indicating that their products may cause unreasonable adverse effects (FIFRA Section 6 [a] [2]). Pesticide incident data can also be obtained from the various toll-free telephone hot-lines (e.g., National Pesticide Telecommunications Network, NPTN), that provide medical emergency services for humans and domestic animals, National Institute of Occupational Safety and Health (NIOSH) surveillance in three states (NY, TX, OR), and the California review of organophosphate poisoning. Additional sources of information available in 1993 are: USEPA data call-in for 28 pesticides (1985-1992 data from poison control centers) and the National Agricultural Workers Survey conducted by the Department of Labor.

b. California Agencies

Reporting by physicians of pesticide-related illnesses has been mandatory in California since 1971 (California Health and Safety Code Section 2950). In this program, illness reports received by county health officers are distributed to the county agricultural commissioners, the Department of Industrial Relations (DIR), DPR, and OEHHA. The agricultural commissioners conduct investigations and submit their reports to DPR. DPR evaluates the reports and analyzes the investigations to determine if regulations or use practices must be altered to prevent further incidents. DPR publishes an annual summary of the collected information and the report is available upon request to the DPR Worker Health and Safety Branch.

All formal complaints by workers are investigated within 48 hours under a Memorandum of Understanding with DIR, DPR, and the agricultural commissioners and cooperative agreement with USEPA. If violations are found based on supportable evidence, civil penalty or criminal action may be taken.

D. CURRENT APPROACHES

This section presents the process by which the toxicological data are evaluated and used in assessing the potential risk of pesticide exposures, including dietary exposures.

1. Review of Toxicological Studies

Toxicological studies are reviewed to determine adequacy for filling the requirements and to identify possible adverse health effects. The FIFRA pesticide assessment guidelines in Subdivision F (Hazard Evaluation for Humans and Domestic Animals, USEPA, 1984 and subsequent revisions) and Draft Subdivision M (Hazard Evaluation for Microbial and Biochemical Pest Control Agents, USEPA, 1989) are generally used as the criteria for decisions regarding study adequacy. The scientific criteria for the evaluation of toxicological studies are generally similar for the federal and the California regulatory agencies. This section provides a brief description of the review processes within USEPA and Cal/EPA - DPR.

a. USEPA

To ensure uniformity in data evaluation, USEPA has standard evaluation procedures that provide guidelines for reviewers on what to look for in the data and how to reach consistent conclusions. Processes are also available for internal peer review (e.g., cancer peer review, reference dose peer review). USEPA also has Good Laboratory Practices specifications and audit programs to ensure the quality and integrity of data. When reviewers have different opinions on whether studies can be used to fill data requirements and/or in risk assessment, the evaluations may then be reviewed by an independent Scientific Advisory Panel. The scientific bases for USEPA's regulatory decisions are subject to public review (USEPA, 1991).

b. Cal/EPA - DPR

The review of toxicological data for health effects is mandated under DPR general authority to register pesticides. These evaluations are peer-reviewed within the Department. Procedures are also available for interested parties to rebut the conclusions of evaluations by interested parties. Study reviews and summaries of toxicological data for each pesticide are available by request at the DPR Pesticide Registration Branch. Due to differences in scientific judgement, interpretation of guidelines under FIFRA, and the possibly different additional data submitted, DPR may reach a different conclusion from USEPA about the same study regarding the acceptability of the study, toxicity endpoints of significance, and other considerations.

2. The Use of Databases in Risk Assessment

Data from the toxicological studies form the basis for identifying the inherent potential of health hazards for pesticides. Without evidence to the contrary, it is assumed that the toxicological effects observed in laboratory animals can also occur in humans, with similar patterns of dose-response relationships. The most sensitive endpoints identified in all the studies from all species are used in evaluating the potential risk of pesticide exposures through dietary intake and other routes of exposure.

Two sets of default assumptions regarding dose-response relationships are currently used by federal and California regulatory agencies. These default assumptions are not used when there is evidence to the contrary. For evaluating a non-oncogenic effect, the assumption is that a threshold dose exists below which the effect is not expected to occur. Therefore, in the process of data evaluation, the lowest-observed-effect level (LOEL) and, when available, the no-observed-effect level (NOEL) are identified. For evaluating an oncogenic effect, the assumption is that a threshold dose does not exist. Any incremental increase in dose or exposure will result in an increase in the probability of the response. The NAS Report made recommendations regarding the mathematical models available for assessing dose-response relationships. Not all of the models considered by the NAS Report are designed to address exclusively the possible age-specific susceptibility. Discussions on these issues are presented in Chapter IV.D. (Dose-Response Models).

The different default assumptions result in two separate outcomes in risk characterization. The risk for a non-oncogenic effect is often characterized in terms of a percentage of the reference dose (RfD) or in terms of the margin of safety (MOS). The RfD is an estimated daily exposure to the human population that is likely to be without an appreciable risk of deleterious effects during a lifetime (USEPA, 1993). A threshold level for short-term exposures, sometimes referred to as the relevant reference dose (RRD), can also be established. In deriving the RfD or RRD, numerical uncertainty factors (UFs) are used to account for uncertainties in the database and the inter- and intra-species extrapolation of data. The MOS is the ratio of the NOEL to the estimated exposure. The NAS Report made recommendations regarding the use of uncertainty factors when toxicity data of a pesticide are insufficient for infants and children. A discussion of the inter-relationship between the percentage of the RfD and the MOS as well as other issues regarding UFs is presented in Chapter IV.C. (Uncertainty Factors). The risk for oncogenic effects is commonly expressed in terms of the excess lifetime probability of tumor occurrence. The probability of risk is a product of potency and the dose (or exposure).

Pharmacokinetic data, to the extent that they are available, are most frequently used in estimating the route-specific absorption factors. These factors are used in route-to-route dose extrapolation and in evaluating the total exposure through multiple routes. The importance of addressing the total exposure from all routes has been highlighted in the NAS Report. Multiple routes of exposure to a pesticide has been included in the evaluation of risk of pesticide exposures by DPR. Comparable pharmacokinetic data on absorption factors in humans can also be useful for interspecies dose extrapolation. This subject will be discussed in greater detail in Chapter IV.B. (Pharmacokinetics and Physiologically-based Pharmacokinetics).

E. DISCUSSIONS AND FUTURE DIRECTION

PECC agrees that areas of health concern, as identified by the NAS Report and summarized in section B of this chapter, cannot be fully addressed until additional data are obtained. The approach to obtaining these data is discussed in this section.

The NAS Report recommended that protocols be established for testing the susceptibility of immature animals for various endpoints (i.e., neuro-developmental, immune, vision, and endocrine effects) that are not currently part of the general battery of toxicity testing requirements. The NAS Report also made specific recommendations concerning the implementation of these recommendations (e.g., modification of the current study protocols).

PECC considers it essential to work closely with USEPA in obtaining the necessary data since the federal and the state mandates for toxicological data are similar and generally follow the FIFRA guidelines. USEPA has established a working group for the implementation of the NAS Report recommendations regarding toxicology issues. PECC support the effort and DPR has been providing input for the implementation plans. Specific PECC recommendations can be categorized into the following three areas:

1. Age-specific Susceptibility

The NAS Report recommended that critical toxicity endpoints (e.g., neurotoxicity, immunotoxicity, effects on endocrine system) and pharmacokinetic disposition in mature and immature animals be compared for representative classes of pesticides. PECC agrees with this recommendation. USEPA, together with other federal agencies, has begun to prepare a plan which will identify research needs on age-related differences. Such research would yield a set of general principles concerning age-related toxicity. Funding needs for this research plan are likely to be significant. The results of these studies will be useful for understanding the potential differences between mature and immature animals. In addition, USEPA is investigating the feasibility and regulatory need for hormone measurements as well as in utero exposures in oncogenicity studies.

PECC also recommend that physiological and pharmacological data indicating differences among infants, children, and adults and between animals and humans be collected, so that the findings from these investigative studies can be used in risk assessments for infants and children.

The results of these investigations will be useful to both USEPA and DPR. Because such investigations are expensive, it is important that DPR works with USEPA to plan the investigative studies to avoid duplication of efforts and to ensure that California needs are addressed.

2. Testing Guidelines

The NAS Report recommended that study guidelines be established for investigating additional toxicity endpoints. Guidelines for evaluating neuro-developmental effects were established in 1991. In response to the recommendation made in the NAS Report, USEPA proposed that additional data gathering may be integrated into the existing testing requirements (e.g., reproductive toxicity study). In addition, USEPA guidelines for several toxicological endpoints recommended by the NAS Report have been drafted and are in various stages of review and revision. Among these are:

  1. guidelines for immunotoxicity testing,

  2. FIFRA and TSCA (Toxic Substance Control Act) harmonized guidelines for reproductive toxicity testing that include additional endpoints, and

  3. neurophysiology guidelines, including those for sensory testing to reliably predict visual system toxicity.

To ensure quality and uniformity in studies conducted by different laboratories, the importance of establishing study guidelines cannot be overstated. PECC agrees that USEPA should finalize these guidelines and DPR should provide input to ensure that the recommended studies also meet requirements in California.

PECC also recommends that there should be discussion on how the results of these studies should be used in risk assessment. This issue seems to be generally missing in the discussions of developing testing guidelines. As testing methodology becomes more refined, dose-related effects or changes that can be detected tend to become more subtle. These subtle effects may or may not be of toxicological significance. Consequently, the judgement of the adversity of an effect may no longer be obvious. Therefore, in the future, if studies in these areas are added to the battery of toxicity testing requirements, the manner in which these additional endpoints may be used in risk assessment should be considered in conjunction with the development of the guidelines. DPR should consider the applicability of testing guidelines to the risk assessment guidelines coordinated by OEHHA for Cal/EPA, when available.

3. Testing Requirements

The NAS Report recommended that acute and subchronic functional and developmental neurotoxicity be tested for all food-use pesticides. Presently, acute neurotoxicity tests are required for cholinesterase inhibitors and other pesticides of neurotoxicity concerns. In addition, a 90-day neurotoxicity testing is conditionally required for any chemical which produces functional and behavioral neurotoxicity or neuropathology in rats and hens in any of the acute toxicity studies.

PECC agree with the NAS Report recommendation that these tests should be required for all food-use pesticides because it is not always possible to predict the potential of a chemical to cause neurotoxic effects. DPR should coordinate with USEPA to implement this recommendation.

REFERENCES

USEPA, 1993. Integrated Risk Information System.

USEPA/USDA/FDA, 1993. Testimony of Carol M. Browner, Administrator of USEPA, Richard Rominger, Deputy Secretary of USDA, and David A. Kessler, Commissioner of FDA, before Committee on Labor and Human Resources, United States Senate and Subcommittee on Health and the Environment, Committee on Energy and Commerce, U.S. House of Representatives. September 21, 1993.

USEPA, 1991. EPA's Pesticide Programs. Pesticides and Toxic Substances (H-7506-C), United States Environmental Protection Agency, 21T-1005.

USEPA, 1989. Pesticide Assessment Guidelines. Subdivision M, Microbial pest control agents and biochemical pest control agents. USEPA, Office of Pesticide and Toxic Substances, Washington, D.C. Publication D-10425.

USEPA, 1984. Pesticide Assessment Guidelines. Subdivision F, Hazard evaluation: Human and domestic animals. USEPA, Office of Pesticide and Toxic Substances, Washington, D.C. PB86-108958.

Back to Table of Contents


Chapter 3

Dietary Exposure Assessment

A. INTRODUCTION

The purpose of a dietary exposure assessment is to determine the potential exposure of humans to pesticides in the diet. There are three components in the assessment:

  1. consumption data (the types and amounts of food consumed);

  2. residue data (residue levels present in foods consumed); and

  3. exposure calculated as follows:

This chapter describes the databases and approaches used by USEPA and DPR in conducting dietary exposure assessments. In the first section of the chapter, the recommendations of the NAS Committee pertaining to dietary exposure assessment are summarized. Next, the databases on food consumption and pesticide residues in food as well as the computer programs currently used for calculating exposure are described. The final section is a discussion of some of the technical and practical issues of exposure assessment in light of the recommendations in the NAS Report.

B. RECOMMENDATIONS BY THE NAS COMMITTEE

The NAS Committee made recommendations under five areas: (1) consumption data, (2) residue data, (3) exposure assessment, (4) data standardization and accessibility, and (5) multiple routes of exposure. These recommendations are summarized below. Potential exposure to multiple pesticides is discussed in Chapter IV.E. (Multiple-Chemical Exposures).

  1. Food consumption and water intake data should be collected for infants and children. These data should be collected periodically to ascertain changes in consumption patterns. There should be adequate numbers of individuals surveyed to allow determination of the distribution of responses. This is important especially if the exposure is to be calculated in yearly age intervals. In the analysis of the consumption survey data, there is a need to have a uniform method for the conversion of prepared foods to its components. (NAS Report pp.10, 196-197, and 271-272)

  2. Residue sampling plans should include more representative samples of commodities consumed by infants and children, and more information on the effects of processing procedures on residue levels. The analytical methods used in the analysis of the samples should be sensitive to detect the pesticide as well as degradation products at exposure levels which are of health concern. In addition, individual units should be analyzed for some commodities. The adjustment of residue profiles for percentage of crop treated should be considered only when information on use pattern justifies such use. There should be sufficient number of samples collected to determine the distribution of residue values. (NAS Report pp.10, 261-262, and 317-318)

  3. Exposure should be characterized as probability distributions to reflect variations in the data. (NAS Report pp.11 and 317)

  4. Food consumption surveys and residue data analysis should be standardized among organizations. The standardization of information will allow the development of nationwide databases. A repository of information about the effects of processing should also be established. (NAS Report pp.10, 197, 261-262, and 318)

  5. Methods should be developed for assessing the combination of exposure from dietary and non-dietary sources. (NAS Report pp.11 and 319)

C. DATABASES AND COMPUTER PROGRAMS CURRENTLY USED

1. Consumption Data

Food consumption databases currently used in exposure assessment are the Nationwide Food Consumption Surveys (NFCSs) for 1977-1978 and 1987-1988. The NFCSs are conducted at approximately 10-year intervals by USDA. The 1977-1978 NFCS was based on geographically stratified probability sampling of households in the U.S. A unique sample of households was selected from each stratum (pre-determined characteristic) in each of the four seasons. Every individual in a selected household was surveyed for three consecutive days. The first-day survey was a personal interview for the 24-hour recall of foods consumed the previous day, and the second- and third-day surveys were based on individual records. In the 1977-1978 NFCS, there was a total of 30,770 individuals surveyed (GAO, 1991) with a response rate of 72% (Petersen, 1991).

There are three major differences in the designs of the two NFCSs. For the 1987-1988 NFCS: (1) Individuals instead of households had equal probability of being included; therefore, the results should be more representative of the population. (2) More food items are identified. (3) Total sample size of the survey was reduced to 10,172 individuals and the response rate for individuals was below 34%. Statistical weights developed by Loughin and Fuller (1990) were used by the USDA on the survey data to compensate for differences between the survey sample and the U.S. population due to non-response and missing data. Because of the low response rate and other survey design concerns, the U.S. General Accounting Office (GAO) recommended that the 1987-1988 NFCS should be used with caution (GAO, 1991).

Comparison of the two NFCSs has been made by Technical Assessment Systems, Inc. (TAS, Petersen et al., 1991) and USEPA. The average consumption rates for the U.S. general population, nursing infants, non-nursing infants, children 1-6 years old, and children 7-12 years old, were similar for some commodities and different (defined as differences by more than 25%) for other commodities (Petersen et al., 1991). These differences may have been due to actual changes in eating habits or inadequacies in either survey. Therefore, the impact of the difference between the two NFCSs on the total exposure depends on the number and the types of commodities being assessed.

2. Residue Data

Data on residue concentrations in raw agricultural commodities and processed foods are available from federal and state monitoring programs, and from other sources such as pesticide registrants, trade organizations, and academia.

a. Residues of Pesticides in Food

(1) FDA: FDA has three monitoring programs: Regulatory Monitoring, the Total Diet Study (TDS), and Incidence/Level Monitoring. Depending on the program, raw agricultural commodities and/or processed foods are collected for analysis. Composite samples are analyzed by either multiple-or single-residue methods. Pesticide residue levels found in each composite sample subjected to multi-residue methods are reported so that information on multiple residues is available. The limit of detection for each pesticide is provided either as a range or a specific level depending on analysis procedures. Data for individual samples from each of the programs are available upon request to the FDA Division of Contaminants Chemistry.

The following is a brief description of the FDA programs. (Detailed descriptions and the results are available in FDA annual reports, including FDA, 1991 and 1992; Yess et al., 1993.) For Regulatory Monitoring, surveillance samples are collected from individual lots of domestic and imported foods at the source of production or at the wholesale level. Imported food samples are collected at the point of entry into the U.S. In 1992, there were 313 pesticides which could be detected by the multiple- and single- residue methods used in this program (FDA, 1992). Detection limits are below the tolerance levels and range from 0.1 to 50 ppm.

Pesticide residue data are also collected in the TDS (Pennington, 1983). In contrast to the regulatory monitoring program, the TDS monitors residue levels in the form that the commodity is commonly eaten or in prepared meals. These are typical meals for various age groups based on the 1977-1978 NFCS. For 1992, the food list was revised based on the 1987-1988 NFCS (Pennington, 1992). Samples for each food are collected four times per year, once from each of the nation's four geographical regions. The detection limits of the analytical methods used for this program are 5 to 10 times lower than those used for Regulatory Monitoring. The residue data generated for raw agricultural commodities and some processed foods, such as prepared infant foods, are useful for the determination of exposure. However, the data for mixed dishes such as stew or soups are not useable to evaluate raw agricultural commodities since the contribution of the individual components to the total residue of the prepared food item is not known. Another limitation of the data is the low sample size; only one sample of each food is taken each quarter with a total of four samples per year.

FDA also conducts the Incidence/Level Monitoring program which is designed to address specific concerns. For example, in 1990-1991, USDA Animal and Plant Health Inspection Service National Monitoring and Residue Analysis Laboratory performed assignments which included the analysis of pesticides in commercially prepared infant foods; benomyl/thiabendazole in foods with postharvest uses and/or food additive tolerances; daminozide in processed foods; ethylene thiourea in selected processed foods; aldicarb in bananas and citrus fruits; and organohalogen pesticide residue in whole pasteurized milk.

In 1992, FDA initiated a statistically-based monitoring project to improve its pesticide monitoring efforts (FDA, 1992). For the first year, pears and tomatoes were chosen because each has a significant domestic and import component, is consumed raw (and often unpeeled), is available year-round, and has tolerances for about 90 different pesticide chemicals. For 1993, two more commodities, apples and rice, were added to the program.

(2) USDA: The Food Safety and Inspection Service of USDA is responsible for ensuring that USDA-inspected meat and poultry products are free of illegal residues. The samples are collected randomly through a statistically based selection from healthy-appearing animals under inspection. The results from this database are limited to specific pesticides of concern to USDA. The numbers of samples collected and pesticides analyzed are generally low and thus the data are not particularly useful for dietary exposure assessment.

USDA is also responsible for the Pesticide Data Program (PDP), implemented in 1991 to collect objective, comprehensive pesticide residue data for fresh produce for risk assessments conducted by USEPA (USDA, 1991). The pesticides and produce to be sampled are selected based on the toxicity of the pesticide as well as the need for residue data to determine exposure and to support minor use registration. Minor use registrations are for those pesticides with limited uses. The samples are collected at produce markets and chain store distribution centers close to the consumer level in several states including California. The samples are composited and only the edible portions are prepared for analysis by laboratories which meet quality control standards set by USDA. Under the contract to USDA, DPR personnel collects samples in California for the PDP. Samples are analyzed by the CDFA Chemistry Laboratory Services laboratories, and the results are available from DPR.

(3) Cal/EPA - DPR: There are two elements in the DPR residue testing program which are currently used for dietary exposure assessment: Marketplace Surveillance and Priority Pesticide. Marketplace Surveillance is designed to ensure that the pesticides are used according to the California laws and regulations. Sampling in this program is weighted toward such factors as patterns of pesticide use; relative number and volume of pesticides typically used to produce a commodity; relative dietary importance of the commodity; past monitoring results; and extent of local pesticide use. To complement the surveillance monitoring, the Priority Pesticide program was instituted to target commodities known to have been treated with pesticides of toxicological concern.

The samples for both elements of residue testing are analyzed by CDFA laboratories. Composite samples containing pre-determined amounts of food material are subjected to three multi-residue screening methods in Marketplace Surveillance and single-residue methods in Priority Pesticide. Single-residue methods may also be used on Marketplace Surveillance samples for confirmation of results obtained from multi-residue methods. There are a total of 204 compounds detected by the multi-residue screens, including 44 metabolites of 22 parent pesticides. Minimum detection limits for all pesticides in the screens range from 0.02 to 0.2 ppm (CDFA, 1993). The concentration of each residue detected in a sample is recorded in the laboratory report. All detected residues, whether above or below tolerance, are recorded. All residue data are available by request from the DPR Pesticide Enforcement Branch.

(4) DHS: The DHS monitoring program targets processed foods collected either at processing facilities or retail markets. Samples from different lots are generally not composited before testing. The DHS laboratory analyzes processed foods for pesticide residues using a multi-residue method modified from that used by CDFA. This method is capable of detecting 84 residues. However, data generated under this program are not useable for dietary exposure assessment because of the low number of samples analyzed. Both DPR and OEHHA are consulted in the implementation of this program

(5) Other sources: Additional sources of residue data are those submitted to USEPA and DPR by registrants or the Interregional Research Project Number 4 (a USDA-funded program that generates data for minor use crops). The data are generated from field trials on agricultural commodities for the registration of pesticides and the establishment of tolerances. Field trials are conducted to determine the highest residue levels possible under the proposed use. More than one field trial may be required for commodities which are grown in different geographical/climatic regions or under different agricultural practices.

Because of the need for residue data which reflect actual consumer exposure, registrants and trade organizations have conducted large-scale market basket surveys. Two recent surveys which have been used by USEPA and DPR to estimate exposure are the National Food Survey (1989-1990) for the ethylenebisdithiocarbamates (EBDC) and ethylene thiourea (ETU) by the EBDC/ETU Task Force (Slesinski, 1990), and the Aldicarb National Food Survey by Rhone-Poulenc Ag Co. (Petersen and Gregorio, 1988).

b. Residues of Pesticides in Water

(1) USEPA: Pesticide residues in well water were evaluated by USEPA in the five-year National Survey of Pesticides in Drinking Water Wells. In this survey, samples were collected from 1,349 drinking water wells (both community water systems and rural domestic wells) and analyzed for 126 pesticides and breakdown products as well as for nitrates (USEPA, 1990). Only the data for California may be applicable for use to assess risks in California.

USEPA also establishes permissible levels for some inorganic and synthetic organic chemicals in the water (USEPA, 1991). These levels, known as the maximum contaminant level (MCL) or the proposed maximum contaminant level (PMCL), are mandated by the U.S. Safe Water Drinking Water Act and amended in 1988 (USEPA, 1991) and are determined based on both toxicology and available technologies for the removal of the contaminants.

(2) California Agencies: In California, well water and surface water are monitored by DPR and other state and local agencies for a variety of purposes. As required by the Pesticide Contamination Prevention Act, DPR maintains a statewide database for all wells tested for pesticides. The database includes results of many different studies conducted by various governmental agencies and private industry using different sampling and testing methods. These differences limit the usefulness of the data to determine the extent of contamination and to apply the data for risk assessment for the general population. Also required by law, DPR maintains a list of chemicals that tend to leach into ground water and monitors wells for chemicals on that list. The priority for monitoring of a particular pesticide is based on the leaching potential in soil, toxicity, the amount sold, and previous history of detection (DPR, 1988). DPR uses the database to identify areas potentially sensitive to pesticide leaching, and to design studies for future sampling (DPR, 1992).

In 1989, The DHS Office of Drinking Water, under Assembly Bill 1803, completed the sampling of public drinking water systems to determine contamination by organic pollutants. Since the completion of the project, the Office of Drinking Water has continued to routinely monitor public drinking water systems derived from either wells or surface water. Selection of the system for monitoring is based on past history of contamination and local usage. The water samples are analyzed for more than 200 pesticides and the results are included in the statewide database. Each region of the California Regional Water Quality Control Board also monitors the surface water for pesticide contamination.

In addition, OEHHA evaluates toxicology studies and proposes recommended public health levels (RPHLs) for chemicals including pesticides. The Office of Drinking Water sets California-specific MCLs based on RPHLs and other factors such as cost and technology, under the mandate of the California Safe Drinking Water Act.

3. Exposure Programs

Computer-based programs are used to calculate dietary exposures because of the complexity involved in calculating exposure for multiple commodities and distribution of consumption rates. USEPA uses the Dietary Risk Evaluation System (DRES) (Saunders, 1987; Saunders and Petersen, 1987), while DPR uses the TAS EXPOSURE 1™ and EXPOSURE 4™ programs (TAS 1992a and 1992b).

The analysis strategies for the food consumption data are similar for the DRES and TAS programs. Each exposure analysis has the capability to estimate the total exposure of one pesticide in multiple commodities. However, the use of a single residue value for a specified food form assumes that all samples of that food form in the diet contain the assigned residue value.

Since the NFCS reports the consumption of some food items in forms different from those for which residues are generally measured, the food items are broken down into components in terms of raw agricultural commodities (Petersen et al., 1991). For example, apple pie is converted into cooked apples, flour, sugar, and other ingredients. Adjustment factors to account for possible increase in residue levels due to processing such as concentration (loss of water) of some commodities are included in the programs.

The programs estimate dietary exposure for the general U.S. population, the general population for each of the four seasons and four regions in the U.S., and subgroups based on ethnicity, sex, and age. The ethnic subgroups are: Hispanics, non-Hispanic whites, non-Hispanic blacks, and non-Hispanic others. The subgroups based on age and sex are: nursing infants (less than one year old), non-nursing infants (less than one year old), children (1-6 years, and 7-12 years), females who are pregnant but not nursing (13 years and older), females who are nursing (13 years and older), females who are neither pregnant or nursing (13-19 years old), females who are neither pregnant or nursing (20 years or older), males (13-19 years old), and males (20 years or older). Options are available in the TAS programs (but not DRES) to customize the subgroups by specifying the criteria, for example, children at yearly age intervals.

For short-term exposures to pesticides, daily consumption is considered. Estimates of consumption answer the question "When one eats a commodity on a given day, how much of that commodity is likely consumed on that day?" (TAS, 1992b). A consumer is defined as an individual who consumed at least one of the food forms included in the assessment. The programs generate a consumption distribution which is the distribution of user-days, defined as individual consumption per day; that is, an individual who responded to all three days of the survey is considered as three user-days. The consumption rate on each user-day in the consumption distribution is multiplied by the pesticide residue value for each commodity. The exposures from the consumption of at least one commodity that contains residue of concern on each user-day are then summed to give the total exposure distribution for a pesticide.

For the determination of long-term exposure, the combined consumption data for consumers and non-consumers are determined. Since individuals are not surveyed every day of the year in the NFCS, consumption estimates are calculated as annualized average consumption and answer the question, "What is the daily consumption of a given commodity as the total annual consumption averaged over 365 days?" (TAS, 1992b). Exposure is calculated by multiplying the annualized average consumption for the population by the pesticide residue level for each commodity. The programs can also be used to calculate commodity contributions; that is, the contribution of individual foods, food-forms, and crop groups to the total exposure estimate.

D. CURRENT APPROACHES

The primary goal of the dietary exposure assessment conducted by USEPA and DPR is to ensure that the consumption of foods containing pesticide residues does not pose significant health hazards. USEPA uses a tier approach by first determining the exposure by calculating the theoretical maximum residue contribution (TMRC) using the tolerance level (discussed in Chapter IV.G. Tolerance Assessment). As a second step, the TMRC may be adjusted using detected residue data, field trial data, information on crop treated, and processing effects on residues to calculate the anticipated residue level (the best estimate of the pesticide residue likely to be consumed). The risk from potential exposure to the pesticide of concern in the diet is then determined. Complete description of the anticipated residue approach is in the USEPA Guidelines for the Use of Anticipated Residues in Dietary Exposure Assessment (USEPA, 1993a).

DPR evaluates risks to residue and tolerance levels using separate processes: (1) risk is determined for total exposure in the diet based on detected residue levels for all label-approved crops and (2) risk is determined for exposure to individual commodity at the tolerance level. In this section, only the dietary exposure determined by residue data is discussed. The evaluation of proposed and existing tolerance levels by USEPA and Cal/EPA (DPR and OEHHA) is discussed in Chapter IV.G. (Tolerance Assessment).

1. USEPA

a. Consumption Data

For consumption estimates, USEPA uses the 1977-1978 NFCS in the DRES program primarily because of the concerns about the low response rate in the 1987-1988 NFCS.

b. Residue Data

USEPA uses tolerance levels based on field trial data for acute dietary exposure assessment. Monitoring data are not used for acute exposure because they may not adequately reflect the highest possible residue levels to which consumers may be exposed.

USEPA conducts assessments of risk at the anticipated residue levels if the theoretical exposure assuming 100 percent of the crop treated and residues at tolerance exceeds the acceptable levels for chronic exposures. To calculate anticipated residues for chronic exposures, field trial or monitoring data are used. When field trial data are used, the average value of all the residue data is used. Residue levels below the detection limit are assumed to be 1/2 of the detection limit expressed as the limit of quantitation (LOQ). This average value is then adjusted by the fraction of the crop treated with the pesticide based on percent of crop treated information. For example, if it is known that only 25% of a particular crop is treated with a particular pesticide, then 75% of the commodity can be assumed to contain no residues, rather than having residues at a certain fraction of the LOQ.

When monitoring data are used, the calculation is different depending on the endpoint and expected exposure period. For example, a lifetime exposure is calculated for pesticides with oncogenic effects. Only long-term exposure, not a lifetime, is required to evaluate those with non-oncogenic effects. For pesticides with oncogenic effects, the average of all residue values is used. Samples below the LOQ that are expected to contain residues (based on percentage of crop treated data) are assumed to be at 1/2 of the LOQ. Samples which are not expected to contain residues are considered to have no residues. For non-oncogenic effects, the 95th percentile residue level is used because USEPA considers it possible that higher-than average residues could be consumed over the less-than-lifetime period.

c. Exposure Assessment

Using the DRES program, USEPA assesses acute exposure for pesticides when an acute toxicological effects of concern such as developmental toxicity or neurotoxicity has been identified. The high end of exposure is used to evaluate risk. Chronic exposure is evaluated for all pesticides.

Exposure via multiple routes is not currently considered by the USEPA Dietary Risk Evaluation Section (Health Effects Division in the Office of Pesticide Program). The Occupational and Residential Exposure Branch of USEPA addresses occupational and non-occupational exposures to pesticides. These assessments generally do not include food or drinking water as exposure media.

2. Cal/EPA - DPR

The Medical Toxicology Branch of DPR conducts dietary exposure assessments as part of the health risk assessment for pesticides registered for use on food crops, as well as new active ingredient registrations in California. Water is included as a commodity when there are monitoring data which indicate possible water contamination. The degradation products and/or metabolites of the active ingredient which have established tolerances or whose toxicity is of concern are also considered in the assessment. The completed dietary exposure assessments are reviewed by other branches of DPR and peer-reviewed by OEHHA.

a. Consumption data

The primary food consumption database used by DPR for dietary risk assessment is the 1987-1988 NFCS. This NFCS has been found to reflect the current eating habits, in spite of the low sample size and response rate (Chaisson, 1990). The 1977-1978 NFCS is used to supplement the 1987- 1988 NFCS when the user-days for a particular population subgroup are inadequate.

When there is no consumption information recorded for the commodity or when the number of user-days from both NFCSs is too few, an individualized consumption estimate is determined. This estimate is a plausible upper limit for daily consumption of individual commodities by representative population subgroups using serving size information (Pennington, 1989) or surrogate commodities based on the USEPA crop group (Code of Federal Regulations, 1992).

b. Residue data

(1) Food: Generally, DPR selects residue data from DPR residue testing programs, federal monitoring programs, as well as registrant field trials and market basket surveys. When there is residue information from more than one source, the usefulness of each source is evaluated based on several criteria. The criteria include the following:

  1. the quality (Were the results generated according to standards of conduct such as Good Laboratory Practices established by USEPA?),

  2. analytical method (Were single residue or multi-residue screening methods used? What was the detection limit?),

  3. number of samples (Were representative samples collected?),

  4. sample form (Were composite or individual units analyzed?).

In general, databases with large sample sizes, low detection limits, and well-documented sampling and analysis methods are used. The selection of the appropriate database also considers the exposure scenario. For example, residue data from field trials may be more appropriate than from market basket surveys for the acute exposure of commodities which may be eaten unpeeled and purchased at the site of harvest, e.g., at the fields.

All label-approved crops or commodities (crops on which the pesticide of concern is permitted to be used by the pesticide product label) are assumed to contain a certain residue levels and these levels are not adjusted for percent of crop treated. Detected residue values are also not routinely adjusted for possible loss and degradation from storage, processing, and home food preparation unless well-supported by pertinent data.

For acute exposure, the highest measured residue value at or below the tolerance for each commodity is used. For samples with residue levels below the minimum detection limit (MDL), the MDL is used as the default. For chronic exposure, the average of the measured and 1/2 of the MDL for "below detection limit" residue values for each commodity is used. USEPA uses the 95th percentile value instead of the average value. Depending on the database, the MDL determined in the analysis is not necessarily the LOQ or limit of detection (LOD). LOQ and LOD are more strict standards than MDLs and are defined as the lowest concentration level that can be quantified reliably, or that can be statistically different from a blank, respectively.

When no residue data are available, either tolerance values or surrogate data on similar pesticides and similar commodities in accordance with the USEPA specification of representative commodity grouping (Code of Federal Regulations, 1992) are used. The tolerance level is used for both acute and chronic exposures. For processed foods, the default is either at the tolerance or the residue level on the raw agricultural commodity multiplied by a concentration factor (greater than one) and not exceeding the tolerance. For example, when the actual residue level for apple cider is not available, a concentration factor (more than one) is used to multiply the residue level for fresh apples to obtain the estimated level for cider. Since pesticides are likely to be degraded and not concentrated during processing, the use of the default concentration factor is likely to result in an overestimation of the actual exposure.

Residue values exceeding the tolerance levels are not used because the incidents are investigated by the DPR Pesticide Enforcement Branch and are relatively infrequent (consistently less than 1 percent incidence according to the DPR residue monitoring programs). DPR evaluates the potential risk of over-tolerance samples under a separate process, expedited acute risk assessment (discussed in Chapter IV.G. Tolerance Assessment).

(2) Water: An assessment of exposure to pesticide residues in drinking water is not routinely conducted for all pesticides because it is difficult to estimate the residue level of drinking water in households. The drinking water in households often represents a mixture from more than one source. Therefore, the data generated for an individual source or localized site may not be appropriate for the general population. DPR included the exposure to atrazine in drinking water in the dietary risk assessment because it has a history of ground water contamination and is a potential health concern. In this case, the average residue level of sampled wells was used.

c. Exposure Assessment

Depending on the use pattern of the pesticide and its potential toxic effects, the dietary exposure assessment may address acute, subchronic, chronic, and lifetime exposures. For the assessment of acute effects using TAS EXPOSURE 1™ program, exposure is calculated based on the distribution of consumption rates and the highest measured residue value at or below the tolerance level for each commodity. DPR generally uses the 95th percentile of exposure. Higher percentiles may be used for highly toxic pesticides or in cases of inadequate toxicity, consumption or residue data. The assessment of chronic effects is based on annualized average exposure for each population subgroup using the annualized average consumption rate and the average residue level. Lifetime exposure is determined for pesticides which have been found to be oncogenic in experimental studies. For this assessment, the exposure level for the U.S. general population is used because the exposure approximates the time-weighted lifetime exposures. In each exposure scenario, the highest exposure level among the population subgroups, often the infants and children subgroups, is used to determine the acceptability of risks for the entire population.

In the exposure assessment, DPR calculates combined exposure from all the commodities for which there are established tolerances for the pesticide of concern. DPR also conducts assessments on potential exposures from non-food sources in the work place and at home in Risk Characterization Documents prepared under the Birth Defect Prevention Act. In the Risk Characterization Document, each of the potential routes of exposure (dietary, occupational, residential, ambient air) is considered alone and in combination with the other routes. Exposure to pesticides in the ambient air is also separately addressed under the mandate of the Toxic Air Contaminant Act. When the risk for exposure is unacceptable, DPR has the authority to mitigate the risk through regulation to reduce exposure or to stop use.

In summary, DPR uses best available data to conduct dietary exposure assessments. There are constraints and limitations to the current databases. DPR uses conservative approaches and default assumptions which tend to overestimate the exposure. They are:

  1. All samples of each label-approved crop contain certain amount of residues. Residue levels are not adjusted for percent of crop treated. In chronic exposure assessments, samples in which residues are not detected are assumed to contain residues at half of the detection limit.

  2. Residue reduction through loss and degradation from storage, processing, and home food preparation is not routinely considered unless well-supported by pertinent data.

  3. In estimating acute exposure, residues are assumed to be at the highest detected level at or below the tolerance in all samples of each crop.

  4. When monitoring data for the raw agricultural commodity are not available, the residue level is assumed to be at the tolerance. With processed foods, the residue level is assumed to be at the tolerance level or at a calculated level (the residue level of the raw agricultural commodity multiplied by a concentration factor).

E. DISCUSSIONS AND FUTURE DIRECTION

Both USEPA and DPR conduct dietary exposure assessments using similar approaches. When residue data are used in the assessment, both agencies are likely to reach the same conclusion as to whether the exposure to a particular pesticide is acceptable. DPR is currently working with USEPA to develop means to directly compare dietary exposure assessments. The following discussion addresses some of the recommendations made by the NAS Committee.

1. Consumption data

Both USEPA and DPR recognize the inadequacy of current food consumption surveys, especially for infants and children, and agree with the NAS Committee that better data are needed. In addition to those currently used in dietary exposure assessment, there are data from other consumption surveys which potentially could be used to provide additional information. The Continuing Surveys of Food Intake of Individuals (CSFII), conducted by USDA, collects information on individual dietary intake from the four geographical regions of the U.S on a yearly basis. Some of the CSFII data may be useful for comparison of consumption rates with the NFCS but combining the databases may not be appropriate. The Gerber Infant Nutrition Survey and the National Health and Nutrition Examination Surveys (NHANES, conducted since 1971 by the National Center for Health Statistics of the U.S. Department of Human Health Services) are also potential sources of data.

USEPA and FDA are working with USDA in the design of the next set of CSFII. These agencies are also working with the U.S. Department of Human Health Services and the U.S. Bureau of Census to design special supplemental surveys for infants and children. The federal agencies, working in concert, can conduct continuing surveys to monitor changes in eating habits as recommended by the NAS Committee. As part of the survey design development, USEPA is working with the U.S. Human Nutrition Information Service on the Food Grouping System to develop a simple and uniform method for converting prepared foods to raw agricultural components.

In addition to the need for more consumption data at the national level, PECC agreed that California should obtain consumption data specific to California populations to address the diets of infants and children of different cultural and ethnic backgrounds, and other concerns. The designed and conduct of the surveys should be discussed in an interagency forum. Limited dietary surveys for Californians have been carried out by participants of the Interagency Nutrition Coordination Council of DHS. For example, the Nutrition and Cancer Prevention Program conducted mail and telephone surveys to characterize fruit and vegetable consumption in adults and, more recently, in children 9 to 11 years old. These surveys formed the basis for the development and implementation of California 5-A-Day-Campaign; however, they were not designed for exposure assessment purposes. Data for consumption distributions for various age groups, as recommended by the NAS Committee, will require a very large survey based on a carefully designed sampling plan. To minimize cost and maximize efficiency, California should work with USEPA, USDA, and U.S. Department of Human Health Services. These federal agencies have the experience in designing consumption surveys as well as established mechanisms for conducting surveys. The state survey, to the extent possible, should be conducted as a part of nationwide surveys and follow the federal agency protocol so that data can be shared between the state and federal regulatory agencies.

The NAS Committee also recommended that food consumption of children should be profiled separately for each year of age up to five years. The current approach used by DPR can be modified by designing a custom population profile. However, subdividing the age subgroups into individual years will further reduce the number of user-days available in each subgroup to a point that it may not be valid for use. This is especially true for commodities consumed by a small number of individuals. A year-by-year analysis of selected commodities and the consumption rates by children using the 1977-1978 and the 1987-1988 NFCS suggests substantial changes in the consumption rates after the age of 3 years old. Therefore, an alternative to the yearly age analysis is to further subdivide the 1-6 age subgroup into 1-3 years and 4-6 years, whenever the available data are adequate. These consumption rates may be compared with those obtained from the CSFII data for validation.

Therefore, given the limitations of both 1977-1978 and 1987-1988 NFCS, PECC recommends that USEPA and DPR should consider the supplemental use of existing smaller-scale surveys (such as the CSFII and NHANES), as well as further subdividing the children subgroup into 1-3 and 4-6 years old when data are adequate.

2. Residue data

The NAS Committee recommended that the residue sampling plans should include more representative samples of commodities consumed by infants and children. Federal and California agencies are committed to improve the current residue programs for dietary exposure assessment. The current programs should be evaluated to focus monitoring efforts to generate data more appropriate for exposure assessment without compromising existing programs for enforcement purposes. FDA has amended the TDS to include more age groups and more food items. FDA also plans to expand the statistically designed incidence level monitoring program to sample additional commodities. The USDA PDP will also continue to expand by including more pesticides and commodities, including some processed foods. The DPR Marketplace Surveillance and Priority Pesticide programs should also be reorganized to include more pesticides and commodities which are important in the diets of infants and children.

The NAS Committee recommended developing more sensitive analytical methods for detecting pesticide residues with lower detection limits. PECC recommends that the determination of the limit should consider matrix interference and the acceptable exposure level. As sensitivity increases, natural chemicals may interfere with the detection of the pesticide of concern. The detection limit should be at a level relevant for risk assessment.

The NAS Committee discussed the need to analyze individual units of a sample instead of composite samples. The current practice of composite sampling and analysis determines the average residue level in the sample and assumes that all individual units are subjected to uniform treatment. This assumption may not be valid and can underestimate potentially high levels in individual units, for example, for an individual banana in a bunch. USEPA has proposed guidelines for the analysis of individual units of commodities treated with acutely toxic pesticides (USEPA, 1993b).

In addition, the NAS Committee recommended that data collection and reporting be standardized, and that a national computerized residue database be created. Certain aspects of the FDA monitoring data reporting system are already being used by many states and by the USDA PDP. The FDA FoodContam program is a nationwide residue database for state-generated residue data. The USEPA Pesticide Residue Information System (PRIS) contains data from the FoodContam program from 1988-1991, data from eight Canadian governmental agencies, member companies of the National Food Processors' Association and Scientific Certification Systems under contract to private food retailers (Nies, 1993). Federal and state agencies have begun the process toward the development of the National Pesticide Residue Database (NPRD). California is a participant in the federal efforts and will continue to provide input toward the development of the national residue database.

While the NAS Committee recommended that more information on water intake should be collected, additional data on pesticides in the drinking water are also important. The assessment of exposure from pesticides in the drinking water requires representative residue data at the consumption level. The USEPA National Survey of Pesticides in Drinking Water Wells and the DPR statewide database of pesticides in wells provide useful residue data for acute exposure or for population subgroups living in regions with water contamination. PECC recommends that current California monitoring programs conducted by DPR and DHS should be modified to provide representative sampling of drinking 0water, both from private wells and public water systems.

The NAS Committee recommended that more information is needed on the effect of processing on residue levels. USEPA requires processing studies to be conducted for some pesticides. FDA has conducted monitoring studies on processed foods. DPR has received processing data from the registrants as part of the registration process and in response to the DPR request under statutory mandates. The processed food industry, especially producers of baby foods, has routinely monitored its products, and those data may potentially be used, especially if they are incorporated in the national residue database. Residue monitoring for processed foods available in California has been limited because of budget constraints in the DHS program for monitoring. DPR plans to implement a limited sampling program of designated processed commodities (i.e., sweet corn and sweet peas) as part of its participation in the USDA PDP. PECC recommends that more residue data on processed foods should be obtained.

The NAS Committee did not recommend the routine use of percentage of crop treated to adjust residue levels as used by USEPA. DPR currently assumes all crops are treated. Understandably, the availability and usefulness of the percentage of crop treated information for the entire nation is limited because the percentage is likely to vary greatly nationwide. In California, however, reporting is required for all agricultural pesticide uses. With this unique pesticide use reporting database, a realistic estimation of percentage of crop treated can be made for California-grown commodities. For many fruits, vegetables, and nuts, a majority of them is grown in California. The information can then be used to adjust the residue profile for chronic exposure assessment, especially when a pesticide is not detected in a majority of residue samples. The potential exposure by rural residents and consumers of locally-grown and treated crops should be considered in the criteria. DPR is in the process of developing criteria for appropriate use of the information.

In summary, PECC recommends that both federal agencies and DPR should enhance the component of residue monitoring programs which generate data for dietary exposure assessment to provide a representative residue profile of foods available to infants and children. This monitoring program should also consider (1) lower detection limits at levels pertinent to risk assessment, and (2) ability to determine residue level in individual units of a commodity as well as in composite samples. California should coordinate with federal agencies in the development of a national residue database and collect more representative samples of drinking water. In addition, DHS should evaluate how to best address the NAS Report recommendation for increased residue testing of processed food particularly those consumed by infants and children. Making use of the data uniquely available through California's full pesticide use reporting database, DPR should establish criteria for the use of the percentage of crop treated information in adjusting regional residue profiles for chronic exposure assessment.

3. Exposure Assessment

The use of an exposure distribution in risk assessment based on the variability in both consumption rates and residue levels is conceptually an improvement over using single values to characterize exposure. The NAS Committee examined the existing databases and concluded that more detailed data are needed for such an approach. The distributional analysis approach requires extensive databases and appropriate linkage of the consumption and residue data. PECC recommends that DPR should utilize a distributional approach for characterizing dietary exposure whenever data are adequate. General criteria for data requirement should be established for the approach. The validity of using residue values exceeding the tolerance levels should also be considered.

The NAS Committee also recommended the use of statistical techniques to determine numerical values for levels below the limit of detection. The use of statistical methods is appropriate when there are sufficient data to estimate a distribution. For example, when measurable values are found to be log-normally distributed, it is reasonable to assume that the values below the detection limit are also log-normally distributed. However, when more than half of the data are below the detection limit, statistical techniques are not applicable. PECC recommends that DPR should continue the use of a default of 1/2 of the detection limit for estimating the average residue levels, and develop guidelines on when a distributional approach should be used for residues below the detection levels.

DPR currently routinely conduct dietary assessments for acute exposures. USEPA conducts acute dietary assessments only when an acute toxicological effect of concern has been identified. However, PECC recommends that unless criteria for determining the circumstances under which an acute toxicity are not of concern, it is prudent to conduct acute risk assessment routinely for all pesticides.

4. Multiple-route exposures

USEPA does not routinely address multiple-route exposures. DPR routinely addresses exposure of pesticides from the diet and other sources such as ambient and indoor air at the work place and in homes. In specific instances, other sources of potential exposure are also included. For example, avermectin in breast milk, atrazine in drinking water, and permethrin on treated clothing were addressed during respective risk assessments conducted at DPR. USEPA, other California state agencies (OEHHA and DHS), and academia are also working on approaches to evaluate total exposure to pesticides. Preliminary projects completed include studies of pesticide residues in house dust and residential soil, and development of a methodology to determine dermal exposures in toddlers.

A concern regarding the multiple routes of exposure has been raised about the exposures of farm children because of the specific circumstances under which exposures can occur. Some of the potential exposure scenarios are: (1) exposures from being in the agricultural fields (e.g., dermal contact, soil ingestion); (2) exposures at home from pesticides carried on tools, clothing and shoes; (3) exposures to pesticides specifically used in the fields (e.g., diverse pesticides to which migrant farm workers may be exposed); and (4) exposures through maternal exposures of farm workers which may be different from the maternal exposures of the general population. Further investigation of pesticide exposure of farm children is needed. A preliminary investigation indicates that specific data for a quantitative assessment of exposure are not currently available and will be difficult to generate.

PECC recommends that USEPA should assess multiple routes of exposure. USEPA should consider additional requirements to obtain the needed data for the various potential routes of exposures. DPR should continue to evaluate multiple-routes of exposure for all population subgroups including infants and children. When data become available, DPR should expand the scope of the assessment both regarding the possible routes of exposure and other population subgroups of concerns (e.g., farm children).

REFERENCES

CDFA. 1993. CDFA Multi-Residue Screen Update. California Department of Food and Agriculture Chemistry Laboratory Services. May 23, 1991.

Chaisson, C.F., 1990. Survey reveals significant changes in American diet. TAS Review 2:1-2.

Code of Federal Regulations, 1992. Subpart A-Definitions and interpretive regulations in Tolerances and Exemptions from Tolerances for Pesticide Chemicals in or on Raw Agricultural Commodities. Part 180.1 (h). Office of the Federal Register National Archives and Records Administration, U.S. Government Printing Office, Washington, D.C.

DPR, 1988. Protocol for Ranking the Ground Water Protection List for Contamination Potential and for Subsequent Monitoring under Commercial Agricultural Conditions, November 30, 1988. Environmental Monitoring and Pest Management, Department of Pesticide Regulation, Sacramento, CA.

DPR, 1992. Sampling For Pesticide Residues in California Well Water. 1992, Well Inventory Data Base, Cumulative Report 1986-1992. Seventh Annual Report to the Legislature, State Department of Health Services, and the State Water Resources Control Board. EH93-02. Environmental Monitoring and Pest Management Branch, Department of Pesticide Regulation, California Environmental Protection Agency, Sacramento, CA.

FDA, 1991. Residue Monitoring 1991. Food and Drug Administration Pesticide Program. Food and Drug Administration, Washington, D.C.

FDA, 1992. Residue Monitoring 1992. Food and Drug Administration Pesticide Program. Food and Drug Administration, Washington, D.C.

GAO, 1991. Pesticides. Food Consumption Data of Little Value to Estimate Some Exposures, GAO/RCED-91-125, United States General Accounting Office, Washington, D.C.

Loughin, M.M. and W.A. Fuller, 1990. Construction of regression weights for the National Food Consumption Survey Data. (Progress report for Human Nutrition Information Service, USDA -July 1990). Iowa State University, Ames, Iowa.

Nies, J.D. 1993. Enhancement of the Pesticide Residue Information System (PRIS), Volume 1. Final Report on EPA Contract No. 68-D2-0053 submitted to U.S. EPA Office of Policy, Planning and Evaluation, Pesticide Policy Branch, July 14, 1993.

Pennington, J.A.T., 1983. Revision of the Total Diet Study food list and diets. J. Am. Diet. Assoc. 82:166-173.

Pennington, J.A.T., 1989. Bowes and Church's Food Values of Portions Commonly Used, 15th edition. Harper & Row, Publishers, New York.

Pennington, J.A.T., 1992. The 1990 revision of the FDA Total Diet Study. J. Nutritional Education 24(4): 173-178.

Petersen, B., 1991. Personal communications from Barbara Petersen to T.R. Hathaway. May 8, 1991.

Petersen, B., and C.A. Gregorio, 1988. Final Report. Aldicarb Acute Dietary Exposure Analysis. Rhone-Poulenc Ag Company. Department of Pesticide Regulatory Vol. 269-133.

Petersen, B.J., J.R. Tomerlin, and S. Chew, 1991. Comparison of changes in dietary consumption of raw agricultural commodities between 1977-78 and 1987-88. Technical Assessment Systems, Inc., Washington, D.C.

Saunders, D.S., 1987. Briefing paper on The Tolerance Assessment System (TAS) for presentation to the FIFRA Science Advisory Panel. Hazard Evaluation Division, Office of Pesticide Programs. U.S. Environmental Protection Agency, Washington, D.C.

Saunders, D.S., and B.J. Petersen, 1987. Introduction to the Tolerance Assessment System. U.S. Environmental Protection Agency, Washington, D.C.

Slesinski, R.S., 1990. EBDC/ETU National Food Survey. Project No. ETU 89-01. Report No. ETU 90-09. EBDC/ETU Task Force (Atochem North America, Inc., BASF Corp., E.I. du Pont de Nemours and Co., and Rohm and Haas Co.).

TAS, 1992a. Exposure 4™. Detailed Distributional Dietary Exposure Analysis, Version 3.1. Technical Assessment Systems, Inc., Washington, D.C.

TAS, 1992b. Exposure 1™. Chronic Dietary Exposure Analysis Version 3.1. Technical Assessment Systems, Inc., Washington D.C.

Trichilo, C.L. and R.D. Schmitt, 1989. Tolerance setting process in the U.S. Environmental Protection Agency. J. Assoc. Off. Anal. Chem. 72 (3): 536-538.

USDA, 1991. Pesticide Data Program, Calendar Year 1991 Report. Agricultural Marketing Service, U.S. Department of Agriculture, Washington, D.C.

USEPA, 1990. National Survey of Pesticides in Drinking Water Wells. Phase I. NTIS Doc. No. PB-91-125765. National Technical Information Service. Springfield, VA.

USEPA, 1991. National Primary Drinking Water Regulations-Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for Unregulated Contaminants; National Primary Drinking Water Regulations Implementation; National Secondary Drinking Water Regulations. Federal Register 56(20):3526-3597.

USEPA, 1993a. Guidelines for the Use of Anticipated Residues in Dietary Exposure Assessment. U.S. Environmental Protection Agency, Washington, D.C.

USEPA, 1993b. Addendum to Subdivision O (Residue Chemistry) of the Pesticide Assessment Guidelines: Residue Data Collection for Acutely Toxic Pesticides. Federal Register 58(12):5390-5391.

Yess, N.J., E.L. Gunderson, and R.R. Roy, 1993. U.S. Food and Drug Administration monitoring of pesticide residues in infant foods and adult foods eaten by infants/children. J. Assoc. Off. Anal. Chem. 76 (3): 492-507.

Back to Table of Contents


Chapter 4

Risk Assessment Issues

A. INTRODUCTION

The risk assessment process is used to evaluate the potential health hazards from exposure to pesticides. There are four components to the risk assessment process: hazard identification, dose-response assessment, exposure assessment, and risk characterization. Hazard identification is the review and evaluation of experimental studies to identify the toxicological properties of the pesticide. The dose-response assessment then estimates the relationship between the amount of the pesticide exposure and the severity or probability of a toxic effect. The exposure assessment includes an estimation of the potential exposure of humans to pesticides. For estimating dietary exposures, the degree of exposure is the product of the amount of pesticide residue on specific commodities and processed foods, and the amount of the commodity consumed. The risk characterization then integrates the toxic effects observed in laboratory animals with potential human exposures to estimate the risk of pesticide exposures from residues in the diet.

Assumptions are used in each of the components of the risk assessment process because of the basic uncertainty in the extrapolation of toxicity data from animal studies to humans and the limitations of the databases for determining actual dietary exposures from single or multiple chemicals. Some of the contemporary, and occasionally controversial, issues that are considered when conducting a risk assessment are discussed in this chapter, including pharmacokinetics and physiologically-based pharmacokinetics, uncertainty factors, dose-response models, multiple chemical exposures, inert ingredients in pesticide products, and tolerance assessment.

B. PHARMACOKINETICS AND PHYSIOLOGICALLY-BASED PHARMACOKINETICS

1. Introduction

Pharmacokinetics (PK) is the study of the movement (i.e., rates and concentration) and biotransformation of chemicals within a biological system. Pharmacokinetics provides a quantitative evaluation of the absorption, distribution, metabolism and excretion of a chemical and/or metabolite(s). The pharmacokinetic data are generally developed from studies using rodent species. Pharmacokinetic data for humans and non-human primates are generally limited unless research were conducted to address specific questions.

Physiologically-based pharmacokinetic (PB-PK) models are designed to predict the movement and biotransformation over a wide range of dose (or exposure) levels based on physiological and biochemical parameters. This modeling requires extensive information on the anatomy and physiology of the biological systems in question, since the overall goal is to determine the dynamics of the chemical and/or metabolite in a specific biological "compartment", i.e., tissue or organ. For example, some chemicals require metabolic activation to exert a carcinogenic effect, whereas other carcinogens are inactivated through competing metabolic enzyme systems. A proper application of the model would provide an estimate of effective dose of the parent chemical and/or its active metabolite(s) that reach the target tissue. Consequently, the extrapolation of toxicity data from animals to humans can be refined so as not to be based on the dose administered to animals or the exposure experienced by humans but on the internal dose effective at the site of toxicity. This approach may lead to more accurate predictions of risks in humans. Another important application for PB-PK modeling is to account for physiological and biochemical changes that occur during pregnancy. Changes in body weight, total body water, plasma proteins, body fat, and cardiac output can alter the distribution of many chemicals. Currently, most of the PB-PK modeling efforts have focused on volatile organic chemicals, such as methylene chloride and trichloroethylene, with little information available for pesticides. Comparative data on physiological and biochemical parameters for mature and immature animals and humans, which may be chemical-specific, are also very limited.

2. Recommendations by the NAS Committee

  1. Physiological and biochemical characteristics of infants and children that influence metabolism and disposition of pesticides need to be considered in risk assessment. For example, physiological parameters, such as tissue growth rates, and biochemical parameters, such as enzyme induction, may be important in evaluating the response of infants and children to pesticide residues in food (NAS Report p.361).

  2. Pharmacokinetic models that provide for the unique physiologic characteristics of infants and children should be developed. PB-PK models used to estimate the target tissue dose can be extended to evaluate potential exposure through breast milk (NAS Report p.361).

3. Current Approaches

a. USEPA

Metabolism and pharmacokinetic studies are required for pesticide registrations by USEPA under FIFRA (USEPA, 1991a). USEPA occasionally adjusts NOELs from an animal toxicity study based on route-specific relative absorption factors derived from these studies. The USEPA is considering developing computer programs for PB-PK modeling in infant animals with the eventual capability of extrapolating to human infants (i.e., comparative pharmacokinetics). As part of this methodology, a parameter database for estimating chemical partitioning into the fetus and breast milk would be developed.

b. Cal/EPA

Pharmacokinetic and metabolism data are reviewed by the DPR Medical Toxicology Branch and Worker Health and Safety Branch for registration of the pesticide in California. In conducting dietary risk assessments, DPR routinely applies a route-specific absorption factor in estimating the dose that enters the body when extrapolating dose-response relationships of systemic effects from one route of exposure to another. In general, systemic effects are the effects observed in sites located distant from where the chemical enters the body. On the other hand, the adjustment for absorption is not applied when extrapolating the dose from animals to humans of the same route of exposure. In this case, it is assumed that the route-specific absorption is the same for animals and humans, unless data indicated otherwise. Pharmacokinetic information in humans can also be obtained from biological monitoring. Biological monitoring of urinary metabolites of workers who are exposed under occupational settings can be used to determine the total absorbed dose of a pesticide from all potential routes of exposure, including occupational and non-occupational (e.g., dietary, ambient air, water). DPR has occasionally used PB-PK models to address specific pesticide exposures. Other Cal/EPA groups (e.g., OEHHA, Department of Toxic Substances Control) have also started developing PB-PK models for their programs.

4. Discussions and Future Direction

PECC recommends that OEHHA should develop general Cal/EPA guidelines for physiologically based models, such as the PB-PK models, that can provide a better quantitative estimate of human exposure and the internal dose. DPR should work with USEPA to establish guidelines for these models to be used in pesticide risk assessments. Furthermore, DPR should coordinate with USEPA to obtain and compile normal baseline values for biochemical, physiological and pharmacokinetic parameters in infants and children. This information is important and necessary to adequately evaluate the response to pesticide exposure. In addition, these parameters should be developed for immature laboratory animals (e.g., rat, mouse, rabbit, dog, monkey) to facilitate extrapolation of test results to humans. These biochemical and physiological parameters can be used in the PB-PK models in order to better define the absorption, distribution, metabolism, and excretion in young animals and in humans, including infants and small children.

C. UNCERTAINTY FACTORS

1. Introduction

Uncertainty factors (UFs) are numerical values that are used to account for the uncertainty in the toxicological database for evaluating the toxicity of chemicals. In assessing the risk of effects other than tumors, an UF of 10 is commonly used for interspecies variation. This means it is assumed that humans can be up to 10 times more susceptible than the most susceptible animals at their most sensitive toxicological endpoints. Another UF of 10 is commonly added for intraspecies variation, that is, there could be up to a 10-fold variation in susceptibility between different people, due to factors such as age, sex, race-ethnicity, and genetic predisposition.

UFs can also be used to account for other areas of uncertainty. UFs can be used when the quality of a toxicological study is poor in design and conduct. An UF can also be used when a study is inadequate for establishing a NOEL (e.g., an UF of 10 is used to estimate a NOEL from a lowest level at which effects are observed, or LOEL). Another application of UF is when the duration of exposure in a study is not long enough for addressing the specific exposure scenario experienced by humans (e.g., an UF of 10 is used to estimate a NOEL for a long-term exposure from a NOEL determined in a shorter-term study).

Additional UFs can be applied when the toxicological database is inadequate. While the use of UFs to allow for inter- and intraspecies variation is more or less conventional (a factor of 10 for each consideration), the rules for additional use of UFs are not as rigid. The overall UF is the multiplication product of all of the UFs.

UFs are used by USEPA in setting threshold standards of exposure for humans, such as the RfD. RfD is an estimate of daily exposure for humans that is not likely to result in appreciable risk of deleterious non-oncogenic effects for an exposure duration of a lifetime (USEPA, 1991b). The RfD is calculated by dividing the NOEL by the overall UF:

RfD = NOEL/Overall UF (Equation 4-1)

For determining if the risk associated with an estimated or expected exposure level is acceptable, a margin of exposure (MOE, term used by USEPA) or margin of safety (MOS, term used by DPR) is calculated. It is the ratio of the NOEL to human exposures:

MOS (or MOE) = NOEL/Human Exposure (Equation 4-2)

Alternatively, risk can also be expressed as a percentage of the RfD. However, as RfD is a reference level for a long-term exposure, its use generally would not be directly relevant to a short-term exposure scenario.

A comparison between the equations 4-1 and 4-2 shows that when the "Human Exposure" is at the "RfD," the MOS would equal the overall UF. When the MOS exceeds the intended overall UF for determining RfD, the human exposure would be deemed acceptable. For example, if an overall UF of 100 (UF of 10 for intraspecies and 10 for interspecies variations) is to be used to establish the RfD based on a NOEL of 10 mg/kg/day determined in laboratory animals, the RfD would be 0.1 mg/kg/day (one-hundredth of 10 mg/kg/day). On the other hand, if an exposure level experienced by humans is at 0.05 mg/kg/day, the MOS would be 200 (10/0.05) which exceeds the intended overall UF of 100 by two-fold and is thereby considered acceptable. In this case, the human exposure is said to be 50% of the RfD; or the MOS for the human exposure is said to be 200 which is two-fold above the MOS that is considered acceptable.

2. Recommendations by the NAS Committee

For effects other than tumors, the NAS Committee recommended the continued use of an UF of 10 each for interspecies and intraspecies variations, and an additional UF of 10 for fetal developmental effects (NAS Report pp.9 and 361). The Committee also recommended that the latter UF of up to 10 should be considered for postnatal developmental toxicity and when data from toxicity testing relative to children are incomplete (NAS Report pp.9 and 361).

3. Current Approaches

a. USEPA

In addition to the conventional use of UFs, an additional UF has been used to account for scientific uncertainty surrounding the adequacy of the database for evaluating the toxicity of the chemical (e.g., an incomplete set of required studies) (Dourson and Stara, 1983; Barnes and Dourson, 1988). These factors are used in setting RfDs using Equation 4-1. On the other hand, when the MOE is calculated (using Equation 4-2) for evaluating the acceptability of a human exposure scenario, an MOE less than the intended overall UF is considered not acceptable.

b. Cal/EPA - DPR

DPR routinely applies the UFs of 10 for each of the intraspecies and interspecies variations in drawing conclusions regarding the MOS. Additional UFs of 10 have also been used in estimating a NOEL from an LOEL and in estimating a NOEL for a long-term exposure from a shorter-term study.

For developmental effects, the current understanding is that they can potentially be elicited by a single exposure. Therefore, a developmental NOEL established from a study after repeated dosing throughout the organogenesis period is used to assess acute exposures, not subchronic or chronic exposures in humans. An additional uncertainty factor may be used on a case-by-case basis to address specific concerns such as the severity of effects.

Under California Proposition 65, a MOS of 1000 for the purpose of providing public warning of exposure is mandated for chemicals listed as known to the state to cause reproductive or developmental toxicity. The exposure is to be calculated based on "reasonably anticipated rate of intake or exposures for average users" of the product (CCR Section 12821).

4. Discussions and Future Direction

Threshold standards of exposure, such as the RfD, are calculated based on certain assumptions and are constrained by the availability of data; therefore, care must be taken in interpreting the results of a risk assessment when the risk is expressed in terms of a percentage of an acceptable standard. When an exposure level is determined to exceed the RfD, the exposure does not necessarily present an imminent health hazard. This is because uncertainty factors are designed to provide ample margins of safety for human exposures. The assumption that humans are 10 times more susceptible than the most susceptible animals tested likely would not be the case for all chemicals. The assumption that there are 10-fold variation in susceptibility within the human population is generally considered adequate for the protection of infants and children (NAS Report p.9). The NAS Committee concluded that "quantitative differences in toxicity between children and adults are usually less than a factor of approximately 10-fold" (NAS Report p.3). It is also important to bear in mind that the RfD is usually calculated from the NOEL, a level at which no effect has been observed in an experiment. The difference between the NOEL and the next higher dose at which an effect is observed would depend on the dose levels selected for the experiment. This is usually between 2- to 10-fold. Therefore, the RfD should be viewed as a goal for the protection of health that, when exceeded, would increase the potential of health hazard.

The current practice by USEPA and DPR to account for intraspecies, interspecies, and data extrapolation is in general agreement with the NAS recommendations. However, an additional UF of up to 10 has not been routinely applied by USEPA or DPR specifically for postnatal developmental toxicity or when data from toxicity testing relative to children are incomplete, as recommended by the NAS Committee. Instead of being specifically for developmental toxicity, an extra uncertainty factor of up to 10 has been used by USEPA to account for incomplete knowledge of the toxicity of a pesticide. USEPA is in the process of evaluating this practice to determine if it should be modified. DPR is also investigating the application of this additional UF.

PECC recommends that DPR should coordinate with USEPA to assess the routine use of such factor. If a factor is to be used, general guidelines must be established to ensure consistency in its application, especially if a variable rather than a fixed uncertainty factor is to be used. Also, the basis for judging toxicity data as "incomplete" regarding the toxicity in children needs to be defined.

D. DOSE-RESPONSE MODELS

1. Introduction

Separate sets of default assumptions regarding dose-response relationships and extrapolation of dose between species (intraspecies extrapolation) are currently used for non-oncogenic (effects other than tumors) and oncogenic effects (effects of tumors or cancer). Default assumptions are used in the absence of evidence to the contrary. For a non-oncogenic effect, the default is that a threshold dose exists, below which no effects are expected. Accordingly, the LOEL and NOEL for each endpoint are established to address the various exposure scenarios (e.g., acute, subchronic, chronic). For an oncogenic effect, the default is that a threshold dose does not exist. Any increase in the level of exposure, however minute it may be, presents an increased probability of tumor occurrence. At the relatively low levels of environmental exposures commonly experienced by humans, the dose-response relationship is assumed to be linear.

In this section, these default assumptions as well as the alternative models recommended by the NAS Committee for consideration are presented. The potential of these methods to accommodate any differences in susceptibility of infants and children is also discussed.

2. Recommendations by the NAS Committee

The NAS Committee recommended the consideration of the following approaches in risk assessments. These methods are pertinent to risk assessment in general rather than specific to the risk evaluation of dietary exposures to pesticides.

  1. Explore the use of benchmark dose approach and consider it a means to bring consistency to non-oncogenic and oncogenic risk assessment (NAS Report p.338).

  2. Consider biologically-based models, specifically the two-stage clonal expansion model (see glossary in Appendix B for description), for characterizing the oncogenic dose-response relationship (NAS Report pp.12 and 351-357).

  3. Consider different methods for interspecies dose extrapolation, particularly pertaining to oncogenic effects (NAS Report p.110).

3. Current Approaches

In general, the approaches used by the federal agencies and in California for dose-response assessment are similar. Issues relevant to the NAS Report recommendations are presented in this section.

a. Model for Non-Oncogenic Effects

The method most frequently used to describe the dose-response relationship involves a determination of a NOEL. The NOEL is the highest experimental dose at which an effect is not identified. With this conventional approach, the assessment of dose-response relationship is summarized by one data point, the NOEL. It neither makes use of the entire data set nor reflects the shape of the dose-response relationship (whether the response increases sharply or gradually with the exposure). The dose level for the NOEL is dictated by the dose selection in a toxicity study. As a result, the NOEL cannot be used directly for comparing the toxicity of two chemicals, the sensitivity of two toxicological effects, or even between two studies of the same type. This approach may produce a NOEL that is far below the dose at and above which effects would appear (LOEL). For example, a lower-than-necessary NOEL may be established when the response at the LOEL is marginally different from the controls but the NOEL is considerably lower than the LOEL. A higher NOEL would likely have been established had an extra dose level between the NOEL and LOEL been included in the test. On the other hand, this approach has the potential of "rewarding" a poorly conducted study. For example, a NOEL determined from a study using a small number of animals and having great sample variations due to poor analytical protocols may indeed have been a LOEL had the conduct of the study been improved to detect the effects. DPR toxicologists consider these aspects in determining NOELs for assessing the potential risk of pesticide exposures, including dietary exposures. When study design or conduct indicates considerable uncertainty, the default is to err on the side of the protection of health.

Another approach for characterizing the dose-response relationship is the benchmark dose (BMD) method. In this approach, the relationship between the level of exposure and the magnitude of response is first described by a mathematical model. The benchmark dose at a predetermined level of response (benchmark response) within the experimental range is then estimated by the mathematical model. Theoretically, the benchmark dose can be an estimation of either LOEL or NOEL, depending on whether the benchmark response is defined as a response that is observable or not. Since the use of BMD approach is not for extrapolating the dose-response relationship far below the experimental dose range, the choice of the mathematical models is not critical as long as a model adequately describes the experimental data. Nevertheless, separate models should be used in characterizing quantal (discrete yes or no response) and continuous (varying degree of response) data for maximizing the use of experimental data. The concept of this approach has been introduced in the Guidelines for Developmental Toxicity Risk Assessment (USEPA, 1991c). However, numerous critical issues for ensuring a consistent use of this approach have yet to be addressed. The BMD approach has been used occasionally by DPR in determining an estimated-no-effect level (ENEL) for pesticides when an apparent NOEL for the most sensitive endpoint cannot be established by the conventional method.

b. Model for Oncogenic Effects

For a chemical that has sufficient weight of oncogenic evidence, the current default model for characterizing the oncogenic dose-response relationship is the linearized multistage model (USEPA, 1986). In addition to utilizing the data on overall tumor incidence, time-to-tumor data are also considered. In an animal study, the early appearance of tumors (shortened time-to-tumor) is detected most frequently when animals die before the end of the lifespan or study. The dose-response relationship is described by an exponential polynomial equation and the coefficients are estimated using the statistical technique of maximum likelihood. The model is constrained to linearity in the low-dose regions. The "potency" is defined as the maximum likelihood estimate (MLE; Q1) of the linear term in the model equation and/or its upper 95% confidence limit (UCL; Q1*). The risk is calculated as the potency multiplied by the exposure. It is an estimate of the excess cumulative probability of tumor occurrence in a lifetime (70 years for humans). The federal and California regulatory agencies have not routinely used the two-stage clonal expansion model (sometimes referred to as the Moolgavkar, Venzen, and Knudson model; MVK model) because much of the necessary data for its use have not been available.

c. Age-specific Susceptibility

A greater susceptibility to toxic insult during early stages in life has been demonstrated for some chemicals, both for non-oncogenic and oncogenic effects. However, there are also cases in which the opposite is true. A general pattern for all chemicals has not been obvious. The lack of pertinent data for pesticides is discussed in Chapter II of this report. Consequently, neither the federal nor California agencies have established default methods to address this issue.

A separate issue frequently mentioned in the context of susceptibility of infants and children is the greater exposures per unit body weight (NAS Report p.362). DPR and USEPA routinely include exposures of different age groups in assessing potential non-oncogenic risks of dietary exposures. However, for oncogenic effects, human exposures are averaged over a lifetime for calculating the lifetime oncogenic risk. Apart from not being able to address the differences in susceptibility to oncogenic insults during perinatal and neonatal periods, the use of the lifetime average dose as a default is reasonable because the same exposure scenario is used in deriving a potency value from an animal study. In a mandatory oncogenicity study, pesticides are typically included in the diet of rodents. Under this condition, the exposures are generally greater during the early period of life for the same reason as in humans (higher consumption per body weight). However, the potency is derived from these studies using the average exposure over a lifetime as inputs for exposures to the mathematical models.

d. Interspecies Dose Extrapolation for Oncogenic Effects

Currently, three different approaches for extrapolating dose from animals to humans for oncogenic effects are used in federal and California programs. The three approaches are: to assume the equivalence of dose between species based on body weight, body weight to the 2/3 power, and body weight to the 3/4 power. Each assumption has its underlying support. For pesticide programs, USEPA is currently using the assumption of body weight to the 2/3 power based on the 1986 guidelines on carcinogenicity risk assessment (USEPA, 1986). DPR is currently using the assumption of body weight to the 3/4 power. This is based on the best estimates of interspecies differences in metabolic rates and many physiological parameters. The difference between the outcomes of risk estimates from these two assumptions is usually within two-fold. The difference is not substantial compared to the uncertainties in the extrapolation of the dose-response relationship from the high- to low-dose range.

4. Discussions and Future Direction

PECC recognizes the potential usefulness of the approaches recommended by the NAS Committee. However, much of the criteria or necessary data for their use are currently lacking. Some practical aspects are discussed in this section.

a. Benchmark Dose Approach

The BMD approach offers many advantages. It defines the dose-response relationship based on the complete set of experimental results, rather than selecting a dose level based on one data point, as in the conventional NOEL determination. It also provides a uniform approach in selecting a level of response for a LOEL or NOEL and allows for a better comparison of results between different studies and endpoints.

Criteria for the application of the BMD approach to ensure consistency are presently lacking. An intra-agency working group within Cal/EPA (BMD working group) has outlined some of the key issues such as: criteria for data selection, determination of benchmark response level, use of best estimate or lower bound of the benchmark dose, models for quantal and continuous data, extent of extrapolation beyond the experimental dose range, and UFs to be applied to the benchmark dose in setting threshold standards of exposure. USEPA is also currently working with other scientific bodies in defining the use of the BMD approach within the context of developmental toxicity assessment.

PECC recommends that the current use of the benchmark dose approach by DPR should be continued. General Cal/EPA guidelines should be established to ensure the consistency of application. DPR should also work with USEPA to establish guidelines for this approachin the evaluation of pesticides.

The BMD approach may bring limited consistency to non-oncogenic and oncogenic risk assessment. With the BMD approach, the probability of a non-oncogenic response can be quantitatively specified at the benchmark dose. The same mathematical models for characterizing an oncogenic dose-response relationship can also be used in fitting quantal data for non-oncogenic effects. However, the fundamental difference between the default assumptions for non-oncogenic and oncogenic effects remains. For non-oncogenic effects, mathematical models are used to estimate a threshold dose below which no effects are expected. Whereas, for oncogenic effects, mathematical models are used to estimate the slope of a dose-response relationship at low dose region, at which a certain probability of effects still exists.

b. Two-Stage Clonal Expansion Model for Oncogenic Effects

The two-stage clonal expansion model provides a possibility for input on certain biological processes (e.g., cell transformation, division, and turnover) and has the potential to address the differences in susceptibility during different stages in life. However, much of the biological information on the rates of mutations that lead to cancerous lesions as well as the rates of clonal expansion are still largely unknown. PECC recommends that as the understanding of a specific oncogenic process increases and the necessary parameters become available, the use of this model should be considered.

c. Age-Specific Susceptibility

The NAS Report illustrated the application of two models for oncogenic effects by comparing the estimated oncogenic risks of dietary exposures from four individual foods under various combinations of assumptions (e.g., number of stages of neoplastic cell transformation, stages that are dose-dependent, clonal expansion rates). The consumption rates for all four foods varies with age throughout the lifetime. The two models used were linearized multistage and two-stage clonal expansion models. It was shown in these model simulations that for an oncogen having high oncogenic potential and exposures at early stages in life, the use of a lifetime average dose may underestimate the risk by five-fold. Theoretically, the magnitude of the possible underestimation of risk would be lower if the total exposure from all foods instead of a single commodity was used in the examples. This is because the age-related difference in total exposures from all foods is not likely to be as great as from individual foods that have high consumption in infants and children.

In order to utilize the model's capability for considering biological processes of tumor formation to yield a more realistic analysis of risk, many of the necessary data and parameters as used in the model simulations given in the NAS Report need to be obtained for both animals and humans. These types of data are not currently available. PECC recommends that the susceptibility issue be appropriately addressed as chemical-specific information on age-specific susceptibility and mechanism of oncogenicity becomes available.

d. Interspecies Dose Extrapolation for Oncogenic Effects

There is no clear scientific evidence that lends better support for any one of the three approaches currently in use by federal and state regulatory programs. As presented earlier, the greatest differences between the outcome from assuming dose equivalence based on 2/3 and 3/4 power of body weight is only approximately two-fold. Therefore, this controversy appears to be a policy issue, rather than a scientific one. For the sake of having a unified approach within the federal agencies, USEPA (USEPA, 1992a) proposed the use of the interspecies dose equivalency based on 3/4 power of body weight. PECC recommends that OEHHA should pursue a unified approach for the interspecies extrapolation within California. DPR should coordinate with USEPA in deriving a unified approach for the evaluation of pesticides.

E. MULTIPLE CHEMICAL EXPOSURES

1. Introduction

While pesticide food safety evaluations routinely address exposures from the consumption of multiple commodities, the focus generally remains on an individual active ingredient. The reason for this approach is partly because the registration process is structured to register an individual active ingredient and its formulation products. The evaluation during the registration process considers only the properties and effects of an individual active ingredient and the formulations.

Frequently more than one pesticide is used on each individual food commodity. The amount of pesticide residues remaining on the commodity at the time of consumption depends on several factors, e.g., application rate, physical/chemical properties of the active ingredient and inert ingredients in the formulation, the time between application and harvest, and the handling of the commodity after harvest. Humans can also be exposed to more than one pesticide when consuming more than one commodity that contains at least one pesticide per commodity.

The NAS Report illustrated an approach using a toxicity equivalence factor (TEF) for estimating human exposures to multiple pesticides with a common mechanism of action, cholinesterase (ChE) inhibition. Although ChE inhibition is regarded as a valid indicator of exposure to organophosphate pesticides (OPs), the association between plasma and blood ChE inhibition and the functional and structural changes of the nervous system is unclear (NAS Report p.301). Therefore, the analysis and discussion of TEF method was "intended as an assessment of methodology, rather than a specific attempt to characterize the proportion of children at risk" (NAS Report pp.306-307).

For the illustration of TEF approach given in the NAS Report, five OPs used on eight food commodities that are representative of the foods eaten by a two-year-old child were selected. The assumptions for this approach were that the toxic potencies of the OPs can be standardized by developing estimates of relative potencies using a reference standard, and that the effects of exposure to the combination of these pesticides were additive. A "strategic simulation" by a computer model was used to generate the distribution of exposures. An exposure analysis was performed initially with the eight commodities. A second analysis included not only the eight raw agricultural commodities but also three juices (orange, grape and apple) in order to determine the potential impact on residue levels from processing the raw agricultural commodity into juice. However, the NAS Report did not present specific details or assumptions used in the illustrations for further analysis of this approach.

2. Recommendations by the NAS Committee

The NAS Committee recommended that exposures to multiple commodities and multiple chemicals with common mechanism of action should be properly addressed (NAS Report p.318). The TEF approach illustrated in the NAS Report is one possible way to address the multiple chemical exposures. Multiple pesticide residues detected in a food sample should be recorded together so that the exposure distributions for multiple pesticide exposures could be more accurately estimated (NAS Report p.318).

3. Current Approaches

a. USEPA

Several programs within USEPA evaluate exposures to mixtures, and this topic has been previously addressed in USEPA's Risk Assessment Guidelines for Mixtures (USEPA, 1985). The Office of Pesticide Programs has not adopted this methodology primarily because it has been difficult to establish the appropriate mixture of pesticides that could be used to characterize potential exposure on representative food commodities.

USEPA had previously developed TEFs for congeners of dibenzo-p-dioxins and dibenzofurans regarding their oncogenic potential. These TEFs were developed based on an extensive evaluation of human data, carcinogenicity and reproductive studies, and in vitro tests. A similar approach is currently being expanded to other chemical classes and endpoints of toxicity. Pesticides are also being considered for inclusion in a similar application of this methodology.

b. Cal/EPA - DPR

In dietary risk assessments, DPR has considered only an individual active ingredient and its degradation/metabolic products of concern on all commodities for which there are established or temporary food tolerances. For the same reason presented by USEPA, DPR has not developed or used a method for the dietary assessment of multiple pesticide residues on individual or multiple commodities.

In addressing the risk of pesticide degradation/metabolic products that are of toxicological concern, a TEF approach similar to that recommended by the NAS Report has been used. For example, a TEF of 10 relative to ethyl parathion was assigned for ethyl paraoxon detected in the air. This is based on the comparison of acute toxicity of ethyl parathion and paraoxon.

4. Discussions and Future Direction

Two areas of considerations are important to the application of the TEF method in dietary risk assessment: the toxicological considerations and the estimation of exposures. Toxicologically, the assumption inherent in the TEF method is that the toxicity of the chemicals under TEF consideration is additive. The ideal candidates to which this additivity assumption can justifiably be applied would be chemicals that have the same or similar mechanism of toxicity. OPs as a class of pesticides would be an appropriate candidate. Although the mechanism of chronic toxicity of OPs is largely unknown, many OPs are known to exert their acute toxicity through the inhibition of acetyl cholinesterase (AChE) in the nervous system. OPs are also the group of pesticides used for the illustration in the NAS Report. The extent to which the approach can be applied to effects of multiple chemicals not known to have similar mechanisms of action is unclear without adequate scientific support.

In developing the TEFs for addressing the acute toxicity of OPs, DPR has been exploring different options for comparing the toxicity of a group of chemicals. The NAS Committee proposed the TEF as a ratio of a criterion of toxicity (e.g., NOEL or LOEL) between a reference standard and the chemical of concern. Instead of using the NOEL or LOEL as the criterion of toxicity, DPR is currently investigating the option of using the benchmark dose for the criteria of toxicity so that the comparison of toxicity between two chemicals can be standardized.

PECC recognizes that the approach to exposure assessment also needs to be explored. It is not appropriate to assume that all the pesticides under the TEF consideration which are permitted to be used on commodities of concern have been applied in a manner that will result in residues in these commodities. California has the unique advantage of developing a more accurate estimate of multiple chemical exposures by utilizing the data collected by DPR under the full pesticide use reporting requirement effective since 1990. For addressing the acute exposures, the data from this use reporting system can be used to identify pesticides that are expected to be present in foods during the same period of time. This information, coupled with the residue data, would enable a more realistic estimation of exposures. A thorough methodology is currently being developed at DPR.

Two opinions have been expressed within the PECC regarding approaches to address the multiple pesticide exposures in California. One opinion is to expedite the process and adopt the policy of summing hazards of chemicals on the same target organ, regardless of whether the mechanism of toxicity has been elucidated for each chemical. The other opinion is to continue with the DPR investigation and to use chemicals with the same mechanism of action as a prototype to explore the application of TEF approach for evaluating multiple chemical exposures before a risk assessment policy is made.

A consensus within the PECC regarding this issue cannot be reached at this time. PECC agreed, however, that the issue will be taken up after the completion of this report through an appropriate interagency forum in which the agencies and departments that have expressed interest in this issue are represented. This would allow thorough discussion of all perspectives without the constraints of the tight time frame mandated for this report. DPR should explore all possibilities for determining the most appropriate approach and coordinate with USEPA on this issue.

F. INERT INGREDIENTS

1. Introduction

An inert ingredient is defined as any substance (or group of structurally similar substances), other than an active ingredient, which is intentionally included in a pesticide product (Code of Federal Regulations, 1992). They may include solvents, carriers, diluents, and conditioning agents. Depending on label claims, a chemical may be the active ingredient in one product and be an inert ingredient in another.

2. Recommendations by the NAS Committee

The NAS Committee discussed the lack of regulation for inert ingredients. While many are innocuous, some may be of toxicological concern (NAS Report pp.342-344).

3. Current Approaches

a. USEPA

In 1987, USEPA categorized the 1,200 inert ingredients currently used in pesticide formulations into four categories based on known or suspected toxicity (USEPA, 1987). The categories are referred to as lists. List 1 includes 57 compounds that are of toxicological concern. List 2 includes 67 compounds that are potentially toxic and have a higher priority for testing than Lists 3 and 4. List 3 includes 800 compounds which have no basis for being on Lists 1, 2, or 4. List 4 contains compounds considered as Generally Recognized As Safe (GRAS) as defined by USEPA. There are approximately 300 inert ingredients in this final category.

Data Call-In Notices (DCI's) were issued for products containing List 1 inert ingredients in April 1988 and March 1989. Of the 57 inert ingredients on List 1, DCI's were issued for 30 chemicals. The other chemicals were determined to either not to be used as inert ingredients or were deferred from data requirements. Of the 30 chemicals subject to DCI, 23 were eliminated from pesticide products as a result of cancellation, suspension or reformulation, one chemical is being considered in conjunction with its companion active ingredient, and six chemicals have pending data waivers.

b. Cal/EPA - DPR

Prior to the registration of a pesticide product for use in California, the inert ingredients in the formulated product are identified as confidential business information in the application by name, amount, and chemical purpose. Under Section 6190 of the California Code of Regulations, DPR, after consultation with the Pesticide Registration and Evaluation Committee, is required to establish a list of inert ingredients which are to be exempted from chronic toxicity data requirements. In 1988, DPR adopted the 1987 USEPA policy regarding the use of inert ingredients in pesticide products with known toxicity (DPR, 1988).

In 1990, under a grant from USEPA, DPR began the development of an inert ingredient database for all pesticide products registered for sale in California. The information from this database is to be linked with the pesticide use reporting data to estimate the total amount of individual chemicals used, and the location of that use. The first phase of the project was the development of a dictionary of the inert ingredients. The types of information in the dictionary include: California chemical code numbers, Chemical Abstracts Service registry numbers, names of synonyms and trade names, name of the reference from which the preferred name was determined, chemical formulation, type of inert ingredient, chemical function, USEPA inert ingredient list category, California Proposition 65 listing, and volatility. This dictionary is now complete with the list of 1,525 inert ingredients, and will be continually updated as new inert ingredients are published by USEPA. The second phase of the ongoing project is to describe and encode the inert ingredients from the actual product files of each pesticide currently registered in California into a computer database. The third phase of the project is the linking of the inert ingredients database with the pesticide label file database, and the California pesticide use reporting database. Initial links between the inert ingredients database and the pesticide label file database have already been established to identify products containing volatile organic compounds. A complete link between the three databases is expected by 1995.

The registration process for formulated pesticide products requires the submission or citation of acute toxicity studies on the pesticide formulation, which generally includes both active and inert ingredients. These studies may be performed on the subject product or on a test article sufficiently similar in formulation as to represent a minimal difference in the considered acute toxicological properties when compared to the product to be registered. These studies evaluate both the acute systemic and the primary irritation potential of the subject product. The acutely toxic potential of the inert ingredients are evaluated, therefore, in the context of the actual mixture to which the user will be exposed. Since DPR dietary risk assessment is conducted only on the active ingredients, the exposure to inert ingredients in the diet is not evaluated.

4. Discussions and Future Direction

PECC recommends that health effects of inert ingredients, and the environmental, toxicological, and exposure databases on these compounds be evaluated or new data be developed. Such databases for some chemicals do exist under other regulatory mandates such as the Toxic Substances Control Act. DPR plans to initiate studies to investigate the environmental fate of inert ingredients. On the other hand, the collection of residue data for inert ingredients in foods may be difficult in that they are often mixtures of many compounds. In addition, the residue levels of some are likely to be negligible. Most solvents would evaporate before harvest. To best utilize limited resources, the assessment of dietary exposures to inert ingredients should be prioritized considering the information on use, toxicity, and the potential for dietary exposures.

G. TOLERANCE ASSESSMENT

1. Introduction

Much of the debate on the health-protective aspect of the tolerance revolves around the approach used to conduct assessments on the tolerance levels. Tolerances established by USEPA are based primarily on agricultural practices. The food supply has been considered safe because evaluation of potential health concerns are included in the process of setting tolerances and the majority of the monitored samples contain residues well below the tolerance levels. On the other hand, some consider the tolerances as not health protective because the exposure at the TMRC for some pesticides has been found to exceed the RfD for chronic exposure or negligible risk standards for lifetime exposures (Fisher, 1992). In addition, tolerance assessments for acute exposures are conducted only for pesticides determined to be of concern regarding developmental and neurotoxicity. In order to determine how the current tolerance-setting process should be modified, some critical issues are presented in this section.

2. Recommendations by the NAS Committee

The NAS Committee recommended modification of the establishment of the tolerance based more on health considerations than on agricultural practices (NAS Report pp.1, 2, and 8). The modified approach should incorporate improved estimates of exposure and more relevant toxicology, along with continued consideration of the requirements of agricultural production (NAS Report p.8).

3. Current Approaches

a. USEPA

As discussed in the Introduction (Chapter I under Pesticide Registration Programs), the use of a pesticide on a food or feed crop requires a tolerance or an exemption from a tolerance. USEPA sets tolerances for raw agricultural products under the authority of Section 408 of FFDCA and for processed food or feed additive tolerances under Section 409 of the Act.

Tolerance levels are proposed by registrants or USDA based on intended uses of the pesticide. In its evaluation of proposed tolerance, USEPA calculates the potential exposures using the TMRC from all existing and proposed uses, and assumes that 100% of each crop is treated with the pesticide, and all contain residues at the tolerance levels. USEPA evaluates the proposed tolerance using the established threshold standards of exposure (e.g., RfD) and risk levels for non-oncogenic and oncogenic endpoints, respectively, and determines the exposures for population subgroups using the DRES (as discussed in Chapter III. Dietary Exposure Assessment). Tolerances which cause the TMRC to exceed the threshold standards are granted if there are sufficient data (e.g., actual residue data and percentage of crop treated) to show that the use of the TMRC results in overestimates of the expected exposure. Tolerance levels are set for all residues of toxicological concern, including metabolites and degradation products. Tolerances serve as enforcement tools to determine deviations from label use which may result in illegal residues or inadvertent contamination of food or feed.

As part of the reregistration process, USEPA is also reassessing the established tolerances to determine if changes are necessary based on new information on toxicity and/or residue levels. The tolerances may be lowered to reduce exposure, or changed to better reflect the current use practice or harmonize with the international food standards established by the CODEX Alimentarius Commission (USEPA, 1992b).

b. Cal/EPA - DPR

DPR has adopted USEPA-established tolerances by regulation. The tolerances for pesticides on raw agricultural commodities are currently being evaluated as part of the DPR mandate under the Food Safety Act. The tolerance for a pesticide on each label-approved commodity is evaluated individually for potential adverse health effects using the TAS acute exposure software programs as described in Chapter III. Dietary Exposure Assessment. For pesticides used on many commodities, the tolerance assessments are conducted for selected fruits and vegetables. Generally, 10 fruits and 10 vegetables are selected from all the uses based on high potential for exposure due to high frequency of consumption for the U.S. population (FDA, 1992), high tolerance values, and other considerations. The acute tolerance assessment does not routinely address multiple commodities since the probability of consuming multiple commodities all at the tolerance levels significantly decreases as the number of commodities included in the assessment increases.

If the established tolerance for a pesticide on a particular raw agricultural commodity yields an insufficient margin of safety for acute exposures, mitigation measures are taken. DPR has authority for setting tolerances within the state, but because USEPA has the authority to establish tolerances for the entire country, the first option for mitigation is to notify USEPA of DPR's findings. In the event that no actions are taken by USEPA to lower the tolerance, DPR can exercise its authority to establish separate tolerances for the protection of human health.

A chronic exposure assessment using residue values equal to the established tolerances is not conducted because it is highly improbable that an individual would habitually consume single or multiple commodities with pesticide residues at the tolerance levels. These conclusions are supported by data from both federal and California pesticide monitoring programs which indicate that less than 1 percent of all sampled commodities have residue levels at or above the established tolerance levels.

DPR also conducts expedited acute dietary risk assessments for pesticides which have been found to occur at residue values exceeding the tolerance levels. Overtolerance incidents may occur due to uses not approved on the labels as well those according to labels. In this assessment, the acute exposure from the consumption of the affected commodity at the detected residue by population subgroups is evaluated. If the assessment shows that the potential risks may be unacceptable, DPR, working with OEHHA and DHS, determines the appropriate regulatory action that should be taken. DPR also informs USEPA of those incidents when the uses were according to the labels and works with USEPA to reevaluate the established tolerances.

c. Cal/EPA - OEHHA

OEHHA, due to budget constraints, has not formalized a program to evaluate the adequacy of USEPA tolerances for processed foods. Summary reports were drafted at the initial implementation of the legislative mandate of Assembly Bill 2848. No recommendations for new tolerances have been made.

OEHHA plans to reinstate program activities to evaluate the adequacy of USEPA tolerances for processed foods. The selection and prioritization of chemicals for evaluation will be based on available information such as that compiled under the Birth Defect Prevention Act or other indications of concern due to chemical toxicity or high exposure potential. OEHHA plans to evaluate both acute and chronic toxicity for various age groups based on varying food consumption rates.

4. Discussions and Future Direction

In response to the NAS Report recommendations and the overall evaluation of the food safety program and pesticide laws, USEPA recently presented the following proposals (USEPA, 1993):

1. Tolerance-setting would be based on a strong, health-based standard, defined as "a reasonable certainty of no harm" to consumers of the food.

2. Tolerance setting would include specific findings that a tolerance is safe for infants and children.

3. All existing tolerances would be reviewed, and they would have to meet the health-based standard within seven years.

4. Time-limited transitional tolerances could be granted for no more than five years for pesticides which do not meet the standards, if the loss of the pesticides would result in significant disruption in the food supply.

5. Tolerance revocation procedures would be amended with a notice-and-comment cancellation process instead of the time-consuming, formal, trial-type proceedings.

It is not possible at this time to thoroughly evaluate the adequacy of the above USEPA proposal in providing a more health-based tolerance-setting process because specific details on the approach to the assessment are not available. PECC recommends that DPR should review the current and future updates of the USEPA tolerance-setting process.

PECC agrees with the NAS Report recommendation that the tolerance-setting process should be health-based. However, the process of setting tolerances should take into consideration agricultural practices as well as human exposure. When a tolerance set to meet the needs of pest control results in an unacceptable risk to humans, the tolerance should be lowered or not granted. On the other hand, a tolerance that poses a risk substantially below the significant risk level should not be raised to a level corresponding to field application greater than that needed for pest control purposes. DPR should continue to evaluate tolerances for pesticides in raw agricultural commodities. OEHHA should continue to evaluate tolerances for pesticides in processed foods.

H. CONCLUSIONS

Risk assessments conducted at USEPA and Cal/EPA are currently based on established approaches. There are issues, as discussed in this chapter, which are subject to different scientific interpretation. PECC supports the development of consistent federal and state approaches and default assumptions for performing risk assessments for dietary as well as residential and occupational exposures. Pharmacokinetic and physiologically-based pharmacokinetic modeling should be encouraged in order to develop better means of estimating absorbed dosages. Procedures such as the benchmark dose approach for non-oncogenic endpoints should be established with clearly defined guidelines. Biological parameters, extrapolation assumptions, and modeling procedures should be uniform within California regulatory agencies and between federal and California regulatory agencies.

REFERENCES

Barnes, D.G., and M. Dourson, 1988. Reference Dose (RfD): Description and use in health risk assessments. Regul. Toxicol. Pharmacol. 8:471-486.

Dourson, M.L. and J.F. Stara, 1983. Regulatory history and experimental support of uncertainty (safety) factors. Regul. Toxicol. Pharmacol. 3:224-238.

Code of Federal Regulations, 1992. Inert ingredients. Title 40, part 158.153 Definitions. Office of the Federal Register. U.S. Printing Office, Washington, D.C.

DPR, 1988. Policy on Inert Ingredients. Pest Management Division Policy 88-1. Department of Pesticide Regulation, formerly California Department of Food and Agriculture, California Environmental Protection Agency.

FDA, 1992. Identification of the 20 Most Frequently Consumed Raw Fruit, Vegetables, and Fish. Federal Register 56(127): 30468-30482.

Fisher, L., 1992. List of Food Use Carcinogens and their Risk Numbers based on the Theoretical Maximum Residue Contribution in a Memorandum from Linda Fisher to Keith Pitts. U.S. Environmental Protection Agency, Washington, D.C.

USEPA, 1986. Guidelines for carcinogen risk assessment. Federal Register 51(185):33992-34012.

USEPA, 1987. Inert ingredients in pesticide products; Policy statement. Federal Register 52(77): 13305-13309.

USEPA, 1991a. Pharmacokinetics and metabolism studies; Proposed Rule. 40CFR798; Fed. Reg. 56 (137): 32541-32544

USEPA, 1991b. Reference Dose: Description and Use in Health Risk Assessments. Integrated Risk Information System (IRIS): Background Document. Online. Intra-Agency Reference Dose (RfD) Work Group, Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH.

USEPA, 1991c. Guidelines for developmental toxicity risk assessment; Notice. Federal Register 56(234):63798-63826.

USEPA, 1992a. Draft Report: A cross-species scaling factor for carcinogen risk assessment based on equivalence of mg/kg3/4/day; Notice. Federal Register 57(109):24152-24173.

USEPA, 1992b. Pesticide Reregistration. Public. no. 700-K92-004, May 1992. Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Washington, D.C.

USEPA, 1992c. Press Advisory. EPA proposes to revoke tolerance of pesticide aldicarb in bananas. July 17, 1992. Communications, Education, and Public Affairs, U.S. Environmental Protection Agency, Washington, D.C.

USEPA, 1993. Executive Summary of Testimony: Administration Pesticide/Food Safety Legislative Reforms. Sept. 21, 1993. EPA Press Office.

Back to Table of Contents


Chapter 5

Summary of Evaluation

The purpose of this study is two-fold: (1) to review and determine if the existing federal and state programs adequately protect infants and children from exposure to pesticide residues in food, and (2) to recommend potential areas for improving the existing pesticide regulatory system. A summary of the program evaluation is presented in this chapter, and proposed improvement to the existing program will be presented in the next chapter, SUMMARY OF RECOMMENDATIONS AND IMPLEMENTATION.

A.HIGHLIGHTS OF CURRENT PROGRAMS

In previous chapters, current federal and state pesticide regulatory programs on food safety are described. For the identification of potential health hazard, extensive toxicological data are required by USEPA and DPR for pesticide registration. DPR may also require additional studies for specific areas of concern. The toxicity data are independently evaluated by USEPA and DPR to determine the potential risks from exposures to pesticides. In addition to the health evaluation, both federal agencies and DPR have extensive residue monitoring programs to ensure compliance with approved pesticide use and provide data for health evaluations. DPR also has the nation's only full pesticide use reporting system of actual pesticide use data which can be utilized in assessing and managing risks of pesticide exposures. The enforcement program in DPR further ensures that pesticide users follow established laws and regulations for the safe use of pesticides. DPR also has the illness reporting system to track pesticide-related illness and to provide information for post-registration effectiveness of safety evaluations.

Two risk assessment procedures are used by DPR in evaluating the risk of dietary exposures to pesticides. One procedure is to conduct risk assessments based on the residue levels detected in monitoring programs and field studies. The other is conduct risk assessments based on the established tolerance levels. The purpose of the dietary risk assessment program is to ensure that the food supply will not pose unacceptable risks to humans, including infants and children. Detailed discussions on issues of the dietary risk assessment process have been presented in the preceding chapters (Chapters II to IV). Only those aspects pertaining to the programs' adequacy for protecting infants and children are highlighted in this section. For each type of assessment, the California program is presented first, followed by a brief discussion of the USEPA program.

1. Assessment of Detected Residue Levels

The potential exposure of humans to pesticide residues in foods can be estimated by the types and amounts of food eaten and the levels of residues in the food. The evaluation of potential hazard to humans relies mainly upon the results of animal toxicity studies that are required for the registration of pesticides.

a. Cal/EPA - DPR

In conducting dietary risk assessments, DPR strives to obtain best available data and utilize them in a manner that would afford protection of human health. Some of the highlights are:

Toxicity Assessment:

  1. The submission of data required under the Birth Defect Prevention Act is near completion for the 200 pesticides determined to have the most significant potential for causing adverse human health effects.

  2. The potential short-term, long-term, and lifetime of exposures that often have different toxicological effects of concern are routinely addressed.

  3. Risk is characterized based on the relationships between exposures and health effects on the most sensitive toxicity endpoints identified in the most susceptible species tested.

  4. It is assumed that humans may be up to 10 times more susceptible than the most susceptible laboratory animals and that there is up to 10-fold variation in susceptibility among the human population. As a result, the upper limit allowed for human exposures to protect against non-tumor effects is usually 100 times below the highest level at which no adverse effects were observed in laboratory animals.

  5. For pesticides that cause tumors, it is assumed that any minute increase in exposure will result in an increase in the probability of tumor occurrence.

Exposure assessment:

  1. The main database for consumption is the most recent National Food Consumption Survey (NFCS 1987-88). When it is inadequate, the previous survey data (NFCS 1977-78) with a larger sample size is utilized.

  2. DPR uses the most appropriate detected residue data from its own program, and those of federal agencies and registrants.

  3. Multiple routes of exposures, though not a part of the dietary risk assessment program, are routinely addressed in the DPR full pesticide risk characterization when data are available. In these assessments, occupational exposures in addition to the exposures through various routes experienced by the general population (e.g., ambient air) are also addressed.

Moreover, when data are not adequate or available, a number of conservative approaches and default assumptions which tend to overestimate the exposure are used. They are:

  1. All samples of each label-approved crop contain certain amount of residues. Residue levels are not adjusted for percent of crop treated. In chronic exposure assessments, samples in which residues are not detected are assumed to contain residues at half of the detection limit.

  2. Residue reduction through loss and degradation from storage, processing, and home food preparation is not routinely considered unless well-supported by pertinent data.

  3. In estimating acute exposure, residues are assumed to be at the highest detected level at or below the tolerance in all samples of each crop.

  4. When monitoring data for the raw agricultural commodity are not available, the residue level is assumed to be at the tolerance. With processed foods, the residue level is the tolerance level or calculated residue level based on the raw agricultural commodity multiplied by a concentration factor.

DPR recognizes the differences in dietary exposures to pesticides due to the differences in food consumption pattern and the kinds of foods consumed among different subgroups in a population. Population subgroups based on age, race-ethnicity, and sex are individually evaluated. Four subgroups for infants and children are addressed: nursing and non-nursing infants less than 1 year old, and children 1-6 and 7-12 years old. In general, the risk of acute exposure for each population subgroup is evaluated based on the 95th percentile of daily exposures. The non-oncogenic risk of chronic exposures for each population subgroup is evaluated based on the average daily exposures. The acceptability of acute and chronic non-oncogenic risks for the whole population is then based on the population subgroup(s) with the highest exposure levels. These are often the exposures of infants and children subgroups. The lifetime oncogenic risk is evaluated based on the exposure averaged over a lifetime.

b. Cal/EPA - OEHHA

Under the Food Safety Act, OEHHA, in addition to DPR, is consulted regarding the DHS monitoring program on processed foods. OEHHA also assesses risks from ingestion of pesticides and other chemicals in drinking water.

c. USEPA

USEPA conducts dietary exposure assessments on a routine basis. This is done as a matter of course, for example, each time a new or revised food use is requested under Section 3 of FIFRA, when a FIFRA Section 18 Emergency Exemption is requested, during the tolerance reassessment process associated with Reregistration, and when a pesticide is being considered for or is in the Special Review process because of potential dietary exposure concerns. While most of these assessments are directed toward the evaluation of long-term or chronic exposures/risk, acute assessments are conducted when acute toxicologic effects of concern are identified.

One aspect of the USEPA chronic assessment methodology that is comparable to the DPR dietary risk assessment process is the "anticipated residue" analysis. If in the first step of the chronic analysis it appears that there may exist an unacceptable level of exposure/risk when comparing the tolerance-level value against the health standard, USEPA will proceed to the step of "anticipated residue" analysis. The analysis further clarifies and refines the exposure estimate by incorporating additional information such as residue data and percentage of crop treated. However, these data are used somewhat differently in the USEPA assessment process than in the DPR assessments. Also, USEPA uses food consumption data from the 1977-1978 NFCS because USEPA determined that there was greater uncertainty to using the data from 1987-1988 survey.

DPR has begun to compare the results of the federal and California dietary risk assessment programs in order to identify possible differences due to the use of different food consumption survey data. Based upon preliminary comparisons, the results from the two approaches are similar.

2. Assessment of Tolerances

a. California Agencies

DPR is responsible for evaluating the tolerances established by USEPA for raw agricultural commodities. The food consumption and toxicity databases used in the dietary risk assessment of residues are also used for the tolerance assessment. The DPR program addresses acute exposures to individual raw agricultural commodities at tolerances. In the assessment, the consumption of all food forms associated with the specific raw agricultural commodity are also included in the exposure analysis. Acute and chronic exposures of multiple commodities, all at tolerances, are not addressed the occurrence of such scenarios would be highly improbable. For the same reason, chronic exposures to a single commodity at tolerance is also not addressed. The ongoing residue monitoring programs for raw agricultural commodities by DPR indicate that pesticide residue levels near the tolerance levels are uncommon. In 1990 and 1991, a total of approximately 20,000 samples were analyzed in the California Priority Pesticide and Marketplace Surveillance programs. Residues were not detected in 75% to 92% of the samples in these two programs. Less than 1 percent of samples contained residues above tolerance and less than 1 percent of the samples contained residues between 50% to 100% of tolerances.

If the tolerance established for a pesticide on a particular raw agricultural commodity yields an insufficient margin of safety for acute exposures, mitigation measures are taken. As a matter of policy, DPR notifies and consults with USEPA when considering the use of its authority to establish separate tolerances specific for California.

Under Assembly Bill 2848, OEHHA is responsible for evaluating the tolerances established by USEPA for processed foods. Summary reports were prepared at the initial implementation of the mandate. Due to staff redirection for higher priority projects such as the metam sodium spill at Dunsmir and Medfly program, further evaluation was not undertaken.

b. USEPA

USEPA assesses dietary exposure and tolerances using a tiered approach as discussed in Chapters III and IV. Whether it be for an acute or chronic exposure assessment, the first step is an evaluation assuming the exposure was to residues at the level of the tolerance. This is done for an individual crop/commodity combination or for the sum of all uses of the chemical under evaluation. If the resulting estimated risk appears not be acceptable, then a second tier of analysis which uses other kinds of residue data (e.g., field trial, monitoring, processing, etc.) is conducted. In chronic assessments, the percent of crop treated may be included.

B. REMAINING QUESTIONS

The NAS Report raised several issues pertaining to the protection of infants and children for which scientifically defensible approaches cannot be made until key studies are conducted. The following are some of these issues.

1. Susceptibility of Infants and Children

A major issue raised by the NAS committee was the susceptibility of infants and children to toxicological insults. NAS Report also concluded that the current database does not show a consistent pattern with regard to what age group in a population may generally be more susceptible. Infants and children may be more susceptible to some pesticides, while persons of other ages (including adults) or ethnic groups may be more susceptible to other pesticides. It appears that the susceptibility issue most likely would have to be addressed on a case-by-case basis when specific data become available.

2. Additional Toxicity Testings

A number of toxicity endpoints, such as effects to nervous, immune, visual, and endocrine systems, have not been rigorously tested or routinely included in the battery of testing requirements for the registration of pesticides. Therefore, it is not known whether these aspects of toxicity might be more or less sensitive than the most sensitive endpoints currently identified for each pesticide. In other words, it remains to be seen whether the measures used to protect against the toxic effects currently considered as the most likely to occur at the lowest exposures would also afford protection against those effects that have not been tested. These questions will remain unanswered until data are available.

3. Multiple-Chemical Exposures

Another area of concern highlighted by the NAS Committee is the consequence of exposures to multiple pesticides. The exposure scenarios can potentially exist either due to multiple pesticides present in a single commodity or single or multiple pesticides present in multiple commodities. The issue is a valid one and historically has not been addressed by DPR or USEPA. Further research is necessary to derive a scientifically defensible approach. One option is the Toxicity Equivalence Factor approach recommended by the NAS Committee. In considering the exposures to a mixture of chemicals, one can reasonably argue for also considering the concomitant exposures to naturally present toxic chemicals. Depending on the amount and the toxicity potential, these natural toxic chemicals may be of greater toxicological significance than the synthetic pesticides. The regulatory perspectives for the exposures to natural and synthetic chemicals, however, would understandably be different.

C. ANALYSIS OF CURRENT PROGRAMS

Section A of this chapter highlighted the current food safety system at both federal and state levels. There are a number of assumptions used in estimating dietary exposures. The use of these assumptions generally results in estimates of exposure at or above the high-end range of potential exposures.

1. California Agencies

DPR, in using the toxicity database to assess risk, follows the conventional approach of basing its assessment on the most sensitive measures of toxicity in the most susceptible species. DPR also assumes humans may be up to 10 times more susceptible than the most susceptible laboratory animals to the most sensitive endpoint, and that there is up to 10-fold variation in susceptibility among the human population (e.g., age, race-ethnicity, sex). The judgement on whether dietary exposures pose an unacceptable risk is based on the highest exposures from the population subgroups evaluated. Because the exposures per body weight of infants and children tend to be higher, conclusions are often driven by the evaluations of infants and children. In addition to the dietary risk assessment of residues and tolerances, DPR's extensive monitoring programs and strong enforcement are mechanisms for ensuring that pesticides are used properly according to the approved labels.

Under the Food Safety Act, DHS has the responsibility for monitoring processed foods for pesticide residues and other contaminants. Because of budget constraints, the monitoring program for processed foods has been limited, with low numbers of samples being monitored. The NAS Report recommends increased pesticide residue testing of processed foods particularly those consumed in large quantities by infants and children. DHS is considering how best to address this recommendation.

OEHHA had previously initiated program activities to evaluate the adequacy of USEPA tolerances for processed foods. Due to staff redirection and budget cuts, no further evaluations have been conducted. Summary reports were drafted. To date, no recommendations for new tolerances have been made. Therefore, full evaluation of the OEHHA program for processed food tolerances can not be made at this time.

2. USEPA

Both USEPA and DPR conduct dietary exposure assessments using similar approaches. When residue data are used in the assessment, both agencies are likely to reach the same conclusion as to whether the exposure to a particular pesticide is acceptable. Because of the complexity of the dietary exposure assessment (e.g., database used, mathematical aspects of exposure estimation), a comprehensive comparison of the USEPA and DPR dietary assessments cannot be completed at this time. However, DPR has identified two specific areas of differences in the two programs: (1) USEPA does not routinely conduct acute exposure assessment for all pesticides, and (2) USEPA does not evaluate multiple routes of exposures.

3. Result of the Analysis of Current Programs

Based on the considerations discussed in this report and summarized in this chapter, PECC concluded that given the available databases, the current California and federal programs of food safety adequately protect infants and children from unacceptable risks posed by pesticide residues in the diet. However, questions remain in areas where data are incomplete or absent. It is not known how the susceptibility of infants and children compares to that for adults, for which pesticide toxicity profiles are estimated from studies in animals. It is also not known whether health-based standards established for protecting against the endpoints established under current data requirements will also protect against additional toxicity endpoints for which tests have not previously been required. In addition, the effects of exposures to more than one pesticide in the diet will not be known until a scientific approach can be established for the evaluation. On the other hand, the estimated dietary exposures may have been unrealistically high because the limitations of current data necessitate the repeated use of high values and the use of assumptions that tend to yield high-exposure estimates. Therefore, the conclusion that the current California and federal programs adequately protect infants and children can only be drawn based on the currently available data. By answering the remaining questions and, with additional data, further refining the dietary risk assessment process and methodology, there would be even greater assurance that the state and federal programs adequately protect infants and children.

Back to Table of Contents


Chapter 6

Summary of Recommendations and Implementation

California has been recognized as having the most comprehensive state-level pesticide regulatory program in the nation (Benbrook and Marquart, 1993). We have evaluated the current programs and, in the previous chapter (Chapter V), concluded that based on the available databases, the current California food safety system adequately protects infants and children from unacceptable risk posed by pesticide residues in the diet. This does not mean, however, that the system cannot be improved.

As a result of the review of the current programs, PECC made recommendations for improving federal and state pesticide regulatory programs to provide additional assurance that infants and children are protected from dietary exposures to pesticides. Some of these recommendations are obviously costly to implement, while others necessitate coordination with the on-going implementation plans at the federal level.

This chapter provides a summary of the recommendations by the PECC as presented in Chapters II through IV of this document. The feasibility of implementation is also discussed.

A. SUMMARY OF RECOMMENDATIONS BY THE PECC

Toxicity Database

  1. DPR should work with USEPA to gather data that quantify physiological and pharmacological differences both among infants, children, and adults, and between humans and animals so that data generated with laboratory animals may be better used in risk assessments of pesticide exposure in infants and children.

  2. Because the federal and state mandates for toxicity testing are similar, DPR should continue to work with USEPA in:

    1. obtaining necessary toxicity data for pesticides;

    2. providing inputs to USEPA for the finalization of testing guidelines for neuro-developmental, immuno-, visual, and endocrine effects, and including considerations of how data obtained can be used in risk assessments; and

    3. instituting the requirement of neurotoxicity testing for all food-use pesticides.

Consumption Data

  1. USEPA and DPR should consider the supplemental use of existing smaller-scale surveys as well as the further subdivision of the age group of children 1-6 years old into two groups of 1-3 and 4-6 years old, whenever the available data are adequate.

Residue Data

  1. FDA, USDA and DPR should reorganize the components of residue monitoring programs which generate data for dietary exposure assessment to provide representative residue profile of foods available to infants and children. These monitoring programs should also consider: (a) lower detection limits at levels pertinent to risk assessment, and (b) ability to determine residue level in individual units of a commodity as well as in composite samples.

  2. DPR should coordinate with federal agencies in the development of a national residue database.

  3. DPR and DHS should collect more representative samples of drinking water.

  4. DHS should evaluate how best to address the NAS Report recommendation for increased pesticide residue testing of processed foods, particularly those consumed by infants and children.

  5. DPR should establish criteria for the use of information on the percentage of crop treated in adjusting regional residue profiles for chronic exposure assessment.

Exposure Assessment

  1. DPR should use a distributional approach for characterizing dietary exposure whenever data are adequate. The approach should also consider: (a) methods to estimate residue levels of samples below the detection limits, and (b) the validity of using residue values exceeding the tolerances.

  2. DPR should continue to evaluate multiple routes of exposure for all population subgroups including infants and children. When data become available, DPR should expand the scope of the assessment both regarding the possible routes of exposure and other population subgroups of concerns (e.g., farm children).

  3. USEPA should routinely conduct the assessments of acute exposures for all pesticides. USEPA should also address multiple routes of exposures and consider requiring the submission of appropriate exposure data.

Risk Assessment

  1. DPR should work with USEPA to establish guidelines for physiologically-based pharmacokinetic models that can provide a better quantitative estimate of human exposure and effective internal dose for pesticide risk assessments. The necessary biochemical, physiological and pharmacokinetic parameters should be obtained.

  2. DPR should coordinate with USEPA to assess the routine use of an additional uncertainty factor as defined in the NAS Report. General guidelines must be established to assure consistency in its application.

  3. DPR should continue the current use of the benchmark dose approach. DPR should also work with USEPA to establish guidelines for the benchmark dose approach in the evaluation of pesticides.

  4. DPR should consider the use of biologically-based models to appropriately address age-specific susceptibility to oncogenic risks when necessary data are available.

  5. DPR should coordinate with USEPA in deriving a unified approach for the interspecies dose extrapolation for the evaluation of pesticides.

  6. DPR should continue to explore all possibilities in determining the most appropriate approach to address dietary risks of multiple pesticide exposures and coordinate with USEPA on this issue.

  7. DPR should consider the information on the use, toxicity, and potential for residues remaining on commodities to prioritize the inert ingredients into the existing dietary risk assessment.

  8. DPR should review the current and future updates of the USEPA tolerance-setting process.

  9. OEHHA should continue to carry out an adequate program to evaluate tolerances for pesticides in processed foods.

  10. OEHHA should develop general Cal/EPA risk assessment guidelines, e.g., physiologically-based pharmacokinetic models, benchmark dose approach, and interspecies dose extrapolation approach.

B.IMPLEMENTATION OF RECOMMENDATIONS

The implementation of the above recommendations can be categorized into California-specific and federal program goals. The California-specific goals include those which can proceed in parallel with the implementation of the federal plans.

Federal Agencies

USEPA has formed six working groups within the federal regulatory agencies for implementing the recommendations by the NAS Report. PECC recommends that federal program goals should be to:

  1. Identify research needs for obtaining comparative data regarding the manifestation of toxicity between immature and mature animals for representative pesticides.

  2. Finalize the various toxicity testing guidelines. When determining additional data requirements for pesticide registration, the anticipated use of data in risk assessment should be considered.

  3. Require neurotoxicity testing for all food-use pesticides.

  4. Consider the supplemental use of the existing smaller-scale consumption surveys (such as the CSFII and NHANES).

  5. Develop a nationwide residue database for dietary exposure assessment.

  6. Investigate the use of distributional approach for characterizing exposures.

  7. Establish the use of the benchmark dose approach for developmental effects.

  8. Routinely conduct the assessments of acute exposures for all pesticides as well as address multiple routes of exposures.

California Agencies

These recommendations for California agencies (Cal/EPA-DPR, Cal/EPA-OEHHA, and DHS) are built upon the foundation of the existing programs and are directed toward respective agency/department with specific statutory mandates on food safety evaluation. They should provide further assurance of protection of human health from risks of exposures to pesticides in the diet.

Several general guidelines for risk assessment need to be developed in California. Different approaches for the various aspects of risk assessment have been used by various regulatory programs in California prior to the formation of Cal/EPA. The need to unify these approaches became more apparent after the formation of one agency. Several risk assessment issues, not exclusively pertaining to dietary risk assessment, have been in discussion within Cal/EPA for some time. PECC recognizes that under current Cal/EPA policy, OEHHA has the responsibility to develop guidelines for risk assessment. The discussions of issues pertaining to general risk assessment guidelines should continue in the restructured standards and criteria working group chaired by OEHHA in accordance with the Senate Bill 1082. DPR will work with USEPA on pesticide risk assessment guidelines, utilizing appropriate Cal/EPA guidelines, as they become available. DPR will coordinate with OEHHA and/or DHS on issues which fall within their statutory responsibilities.

DPR

  1. As an interim measure before more comprehensive consumption data can be collected, DPR should consider: (a) the supplemental use of other consumption surveys and, (b) the further subdivision of the age groups of children to ages 1-3 and 4-6 years old. Other consumption survey databases are available. Data analysis and compilation into a usable format will be resource-intensive. The use of data to augment the current NFCS data needs to be defined. Interim measure (b) is being considered by DPR.

  2. DPR should establish guidelines and criteria for: (a) considering the use of the percentage of crop treated information, (b) considering the use of an additional uncertainty factor and, (c) addressing multiple pesticides exposures. DPR has begun to investigate the critical issues in these areas. For item (b) and (c), considerable data are needed to develop the scientific approaches. DPR should work with USEPA on these issues.

  3. DPR should continue to evaluate multiple routes of exposure for all population subgroups including infants and children. DPR will expand the current practice when data become available.

  4. DPR should reorganize the component of residue monitoring program which generates residue data for food and water to be used in dietary exposure assessment. Because of the inherent differences between residue monitoring for enforcement purposes and monitoring done for risk assessment, the implementation of this recommendation is essential for improving the current program toward a more realistic assessment of dietary exposure. However, the implementation of this recommendation may require substantial additional funding resources.

  5. DPR should coordinate with the federal effort in the development of a national residue database. California is already a participant of federal residue monitoring programs such as the FoodContam and the USDA PDP. California should continue to work with USEPA and other federal agencies to develop procedures to standardize the residue analysis and reporting systems.

  6. DPR should utilize a distributional approach for characterizing dietary exposures whenever data are adequate. General criteria for data requirements should be established for the approach.

  7. DPR should prioritize inert ingredients into the existing dietary risk assessment. To ensure the best use of limited resources, DPR should consider the information on the use, toxicity, and potential for residues remaining on commodities for the prioritization.

  8. DPR should work with USEPA to compile data that indicate physiological and pharmacological differences both among infants, children, and adults, and between animals and humans. A preliminary search for the information available in the open literature will establish the likelihood for achieving this goal. The implementation will be resource-intensive.

  9. DPR should consider the use of biologically-based models to appropriately address the age-specific susceptibility to oncogenic risks. Much of the necessary data and parameters for these approaches are currently lacking. For the intended use of addressing the differential risks of infants and children, data on differential susceptibility are also lacking. The use of this approach needs to be defined, both in terms of needed resources and effectiveness in better protecting infants and children. Possible alternatives should also be explored.

  10. DPR should coordinate with USEPA in establishing guidelines for physiologically-based pharmacokinetic models and the benchmark dose approach. USEPA and Cal/EPA have been working on these guidelines for some time. Because of many unresolved issues (some due to lack of data), no clear plans to establish guidelines have been made. It is important that DPR works with USEPA to establish guidelines for use in pesticide risk assessments.

  11. DPR should derive a unified approach with USEPA on interspecies dose extrapolation for the evaluation of pesticides. This subject has been under discussion within federal agencies and Cal/EPA for some time. A decision needs to be made for a unified approach with federal agencies.

  12. In addition to what can be done within California, DPR should also coordinate with USEPA on its implementation plans. USEPA has formed six working groups within the federal regulatory agencies for implementing the recommendations by the NAS Committee. It is essential that DPR continues to participate in those efforts. DPR should review the current and future updates of the USEPA tolerance-setting process.

OEHHA

  1. OEHHA should continue to evaluate tolerances for pesticides in processed foods. Preliminary efforts have been underway toward the implementation of this mandate. A functional program should be established.

  2. OEHHA should give a higher priority to the development of general risk assessment guidelines. The guidelines should be developed in coordination with appropriate departments within the agency. These guidelines may include physiologically-based pharmacokinetic models, benchmark dose approach and interspecies dose extrapolation approach.

DHS

  1. DHS should evaluate how to best address the NAS Report recommendations for the increased residue testing of processed foods, particularly those consumed by infants and children. Residue data for processed foods that are useful for dietary exposure assessment should be obtained.

  2. DHS should collect more representative samples of drinking water. Representative data are needed for better characterizing the total exposures to pesticides.


    CONCLUSION

    In conclusion, this report describes approaches used by federal and state agencies to evaluate the safety of pesticides in the diet, especially in the diets of infants and children. Based on the considerations presented in this review, the current California and federal pesticide regulatory systems adequately protect infants and children from risks posed by pesticide residues in the diet. However, there are potential areas for improvement of the state's pesticide regulatory program. Uncertainty remains in areas where data are incomplete or absent. In addition, the scientific basis of some of these issues requires further research and discussion. Many of the PECC recommendations are consistent with those in the NAS report. These recommendations, if adopted, will provide a better scientific basis for approaches and assumptions currently used. This, in turn, will improve the accuracy of the risk estimates and allow greater consensus to be achieved concerning the magnitude of pesticide risks in the diet.

    The implementation of these recommendations in California can proceed in parallel with the plans at the federal level. The monetary cost of implementing some recommendations is expected to be significant. In addition, considerable scientific effort will be required. In this era of limited resources, choices must be made on the benefits to be derived from additional testing requirements and data acquisition.

    This report should also serve to provide the framework of scientific discussions between participating agencies in California that are involved in food safety programs. DPR, in its role as the lead agency for pesticide regulation, should continue its on-going project with USEPA to achieve greater harmony in pesticide regulatory programs. In this capacity, DPR can serve as liaison between the state and federal agencies.

    Back to Table of Contents


    Appendices


    Appendix A

    List of Abbreviations

    ADI Acceptable Daily Intake

    BMD Benchmark Dose

    Cal/EPA California Environmental Protection Agency

    CCRIS Chemical Carcinogenesis Research Information System

    CDFA California Department of Food and Agriculture

    ChE Cholinesterase

    CSFII Continuing Survey of Food Intake by Individuals

    DCI Data Call-In Notices

    DHS Department of Health Services, California

    DIR Department of Industrial Relations

    DPR Department of Pesticide Regulation, Cal/EPA

    DRES Dietary Risk Evaluation System

    EBDC Ethylenebisdithiocarbamates

    ENEL Estimated-No-Effect Level

    ETU Ethylene thiourea

    FAC Food and Agricultural Code

    FDA U.S. Food and Drug Administration

    FFDCA Federal Food, Drug and Cosmetic Act

    FIFRA Federal Insecticide, Fungicide, and Rodenticide Act

    GLP Good Laboratory Practices

    HSDB Hazardous Substances Data Bank

    IDS Incident Data System

    IRIS Integrated Risk Information System

    LOEL Lowest-Observed-Effect Level

    LOQ Limit of Quantitation

    MDL Minimum Detection Limit

    MLE, Q1 Maximum Likelihood Estimate

    MOE Margin of Exposure

    MOS Margin of Safety

    MVK Moolgavkar, Venzen, and Knudson Model

    NAS National Academy of Sciences

    NFCS Nationwide Food Consumption Survey

    NHANES National Health and Nutrition Examination Surveys

    NFPA National Food Processors' Association

    NOEL No-Observed-Effect Level

    NPHA National Pesticide Hazard Assessment Program

    NPRD National Pesticide Residue Database

    NPTN National Pesticide Telecommunications Network

    NTIS National Technical Information Services

    OEHHA Office of Environmental Health Hazard Assessment, Cal/EPA

    OP Organophosphate

    PDP Pesticide Data Program

    PB-PK Physiologically-Based Pharmacokinetics

    PIMS Pesticide Incident Monitoring System

    PK Pharmacokinetics

    PRIS Pesticide Residue Information System

    RPHL Recommended Public Health Level

    RED Reregistration Eligibility Document

    RfC Reference Concentration

    RfD Reference Dose

    RRD Relevant Reference Dose

    RTECS Registry of Toxic Effects of Chemical Substances

    TAS Technical Assessment Systems, Inc.

    TDS Total Diet Study

    TEF Toxicity Equivalence Factor

    TSCA Toxic Substance Control Act

    TMRC Theoretical Maximum Residue Contribution

    TSH Thyroid Stimulating Hormone

    UCL, Q1* Upper 95th Confidence Limit

    UF Uncertainty Factor

    USDA U.S. Department of Agriculture

    USEPA U.S. Environmental Protection Agency


    Appendix B

    Glossary of Terms

    Acceptable daily intake (ADI) is the acceptable consumption level for life-time exposure. USEPA has replaced its use by the terms reference dose (RfD) or reference concentration (RfC).

    Active ingredient is an agent responsible for the pesticidal activity. Pesticides are regulated primarily based on the active ingredients.

    Anticipated residue level is the residue level derived from the tolerance level. It accounts for an increase or loss of residue due to processing such as peeling or cooking, and for the probability that the sample has not been treated.

    Benchmark dose (BMD) is the dose for a response of biological significance and is selected from the dose-response curve based on the full set of experimental data.

    Cancer is a general term for a cancerous tumor or for forms of new tissue cells that lack a controlled growth pattern. Cancer cells usually invade and destroy normal tissues. A cancer tends to spread to other parts of the body by releasing cells into the lymph or bloodstream.

    Carcinogenesis is the process of causing cancer.

    Cholinesterase (ChE) is an enzyme in the body which is responsible for the breakdown of the chemical, acetylcholine, and is involved in the control of the nervous system. The mode of action of many pesticides is to inhibit the ChE in the target pest.

    Composite means that individual units of a sample are mixed and a portion of the mixture is then analyzed for pesticide residues. For example, instead of a single head of broccoli being analyzed, several heads of broccoli are chopped together and samples are taken from the mixture for analysis.

    Distributional analysis is a method to examine the dispersion or range of the data points in a dataset. For example, for consumption data, the analysis would answer the questions who eats how much of what foods.

    Dose-response relationship is the relationship between the level of exposure (or dose) and the magnitude of response.

    Emergency exemption is an permit (allowed under Section 18 of FIFRA) for use of a pesticide, not registered for the intended use, to address urgent, non-routine situations when no effective alternative pesticide is available.

    Endpoint is an effect observed in a toxicity study. The effect is selected because it is of biological significance and indicates a health concern.

    Estimated-no-effect level (ENEL) is an estimate of the highest dose level which causes no effect. It is determined by scientists when the experimental data do not provide a no-effect level and the ENEL is estimated from the levels with effects.

    Experimental Use Permit is a permit issued by USEPA for the testing of unregistered pesticides or new uses of registered pesticides.

    Label-approved crop is a crop on which a pesticide is permitted to be used by the pesticide product label.

    Lowest-observed-effect level (LOEL) is the lowest dose level in which an effect was observed in the experimental studies.

    Limit of quantitation (LOQ) is the lowest level detected which can be determined to be statistically different from the background from samples without treatment. It is different from the MDL in that it requires more precise and accurate determination of the detection limit.

    Minimum detection limit (MDL) is a generic term to represent the lowest level detected which is determined to be different from the background. This term is used when statistical methods are not used to determine the actual detection limit with certain levels of confidence.

    Maximum likelihood estimate (MLE, Q1) is the slope of the dose-response curve for oncogenic effects observed in animal studies. The steepness of the slope is used to compare the potency of chemicals. For example, given the same dose, a chemical with a steeper slope has a greater probability of causing tumors than a chemical with a smaller slope.

    Margin of exposure (MOE) is a term used by USEPA to express the acceptability of an exposure level with respect to the no effect level observed from experimental studies.

    Margin of safety (MOS) is used by the Cal/EPA, and has the same definition as margin of exposure.

    No-observed-effect level (NOEL) is the level at which no effect was observed in experimental studies.

    Oncogen is a chemical which causes the formation of tumors which may be malignant or benign.

    Oncogenesis is the process of the growth of a tumor through the action of biological, chemical, or physical agents.

    Pesticide Advisory Committee is a DPR committee to bring together all public agencies with jurisdiction by law over the use of pesticides or over activities or resources which may be affected by the use of pesticides to consult on policy issues regarding pesticides.

    Pesticide Registration and Evaluation Committee is a committee chaired by the Department of Pesticide Regulation to bring together all public agencies with jurisdiction by law over the use of pesticides or over activities or resources which may be affected by the use of pesticides to consult on pesticide product registration, renewal, and reevaluation issues.

    Physiologically-based pharmacokinetics (PB-PK) is a biological model to predict the movement of chemicals in the body and is based on physiological and biochemical parameters.

    Pharmacokinetics (PK) is the study of the movement of chemicals in the body.

    Potency is the slope of the dose-response relationship for an oncogenic response. The statistical 95th confidence limit of the slope is usually taken as the potency.

    Probability distribution is used to express the variability in a dataset and is used to predict the occurrence of a certain outcome.

    Raw agricultural commodity includes fresh fruits, vegetables, grains, nuts, eggs, raw milk, meats, and similar agricultural produce as defined by USEPA in the Code of Federal Regulations.

    Reference concentration (RfC) is an estimated daily exposure by inhalation for the human population that is likely to be without an appreciable risk of deleterious effects during a lifetime.

    Reference dose (RfD) is an estimated daily exposure by ingestion for the human population that is likely to be without an appreciable risk of deleterious effects during a lifetime.

    Relevant reference dose (RRD) is an estimated daily exposure for the human population that is likely be without an appreciable effect in a short-term period.

    Special local need registration is a state registration (under the authority of Section 24 (c) of FIFRA) of a pesticide product registered for other commodities by USEPA, to address a need not met by the federally approved label.

    Special Review is the process used by USEPA to evaluate a greater risk to humans or the environment than previously determined.

    Statistical weights are correction factors used to adjust for inadequacies of a dataset based on additional information.

    Stratified area probability sampling is an approach where the population to be sampled is divided based on pre-determined characteristics. Examples of the characteristics are geographical regions, age groups, and sex. Each subgroup within the group is then called a stratum. Samples are then drawn randomly from each stratum.

    Toxicity equivalence factor (TEF) is a factor used to average out the effects of chemicals to a pre-determined common endpoint for the comparison of toxicity.

    Theoretical maximum residue contribution (TMRC) is the total exposure level calculated based on the assumptions that 100% of a crop is treated, and that consumers are chronically exposed to all the commodities with the residue levels equal to the tolerances.

    Tolerance is the legal maximum residue concentration of pesticide allowed in raw agricultural commodities and processed foods. It is not the anticipated residue level. Tolerances are established to ensure the protection of the public, including infants and children, from health risks posed by eating foods that have been treated with pesticides according to the label directions.

    Two-stage model (clonal expansion) is a biologically-based mathematical model that can be used to describe the dose-response relationship of toxicological effects of causing tumors. It allows input for biological processes pertaining to tumor formation, such as cell transformation and turnover rate.

    Upper 95th confidence limit (Q1*) is the upper confidence limit for the slope of the dose-response curve for oncogenic effects observed in animal studies.

    Uncertainty factor (UF) is a numerical value to express the uncertainty in the toxicological database, as well as uncertainty in the variability of response within (intraspecies) and between (interspecies) species. For example, to extrapolate the NOEL obtained from data conducted with animals to humans, the NOEL is divided by an UF, usually a factor of 10, to estimate the NOEL for humans.


    APPENDIX C

    CONTRIBUTORS OF PUBLIC COMMENTS

    Action Now
    American Council on Science and Health
    Ames, Bruce, University of California, Berkeley
    California Certified Organic Farmers
    California Citrus Mutual
    California Farm Bureau
    California Indian Basketweaver's Association
    Children Advocacy Institute
    Council for Agricultural Science and Technology
    Edwards, Gordon, San Jose State University
    Gold, Lois, University of California, Berkeley
    Iceberg Lettuce Advisory Board
    Jukes, Thomas, University of California, Berkeley
    National Resources Defense Council
    Pesticide Watch
    Sirkin, Gerald and Natalie, Connecticut
    The Keystone Center
    Valent U.S.A. Corporation
    Western Agricultural Chemicals Association
    Western Grower Association

    Back to Table of Contents