Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 46
~ Selecting and "J Scheduling Compounds for Assessment INTRODUCTION Some 35,000 pesticide formulations are registered for use in this country. Many, perhaps most, of these pesticides are innocuous when used properly. Serious questions have been raised about a number of others, however, because they may present risks to public health and to the environment. Recognizing these risks, Congress, in the FIFE amendments of 1972, required EPA OPP in particular to review the registrations of all 35,000 compounds within 4 years. In the 1975 FIFE amendments, this deadline was extended a year, and in 1978 all references to a deadline were eliminated. The funds authorized to carry out the provisions of FIFE, as amended including non-RPAR activities hovered around $50 million from FY 1976 to FY 1978. In FY 1979 the authorization increased to $70 million. The oPP staff compromised approximately 450 members in early 1976, and had increased to about 700 by early 1979, but even these resources (funds and stab size) have proved inadequate to the task at hand. A single RPAR procedure namely, that for chlorobenzilate occupied an RPAR team of 13 oPP professionals for a significant part of their time for 3 years. The direct personpower costs associated with the chlorobenzilate RPAR have been estimated at $400,000 (F. Arnold, oPP, EPA, Washington, D.C., personal communication, 1979~. By early 1979 only seven RPAR procedures had been carried through to a final decision. A total of 27 compounds have entered the RPAR process 46

OCR for page 46
Selecting and Scheduling Compounds for Assessment 47 (Jellinek 1979~. About 20 compounds are in pre-RPAR stages, and another 20 are being considered or have already been selected for pre-RPAR review (Pesticide & Toxic Chemical News 1978~. Considering the current slow rate of progress in the RPAR program, the criteria by which suspicious pesticides are selected for evaluation and the order in which they are considered are matters of first importance. Realistically, it will be many years before the list of pending reviews is exhausted. Therefore, those compounds that appear most likely to have the greatest adverse impacts on public health or the environment should be selected for RPAR review before those compounds that pose less of a threat. This chapter is devoted to selection procedures. In the first section, we describe the current procedures for determining which pesticides are subjected to the RPAR process, and in the second section we present the Committee's recommendations for improving the selec- tion process. The third section recommends a procedure for determining when and how thoroughly substitute pesticides should be considered in the RPAR process. CURRENT APPROACH To facilitate the following discussion, it will be useful to distinguish between the total pool of pesticide registrations up for review, those formulations selected as candidates for the RPAR process, and those actually chosen for formal RPAR consideration. The total pool of registrations is the roughly 35,000 registered pesticide formulations that EPA iS required to review under the 1972 amendments to F'FRA. The intent of this legislated review is to subject all previous registrations to the newly established standards of safety outlined in the amendments. RPAR candidates are selected from the pool of registered pesticides; compounds in this category are subjected to a pre-RPAR review to determine whether they meet or exceed the risk criteria set out in 40 CFR 162.11 (see Chapter 2~. Finally, those compounds that will actually be subjected to the RPAR process are chosen from among the RPAR candidates on the basis of the results of the pre-RPAR review. SELECTION OF CANDIDATE COMPOUNDS FOR PRE-RPAR REVIEW The current rather informal procedure for identifying and scheduling candidate compounds for pre-RPAR review seems to be a function of several elements. Some of the selection and scheduling decisions are based on considerations internal to oPP. However, a significant number of the decisions are made in response to various external pressures.

OCR for page 46
48 REGULATING PESTICIDES Many of the initial RPAR candidates were simply inherited. When the administration of FIFED was transferred from the USDA in 1970, EPA established as a forerunner of the RPAR process the Suspect Chemical Review (SCR) program (NRC 1978~. As part of this program, oPP's Criteria and Evaluation Division (now HED) had responsibility for identifying potentially hazardous pesticides and developing a "suspect chemical" list (NRC 1978~. The identification of suspect chemicals was based largely on information extracted from various scientific publications, especially the Mrak report (U.S. DHEW 1969), which provided a comprehensive literature review and an assessment of the environmental and human health implications of a number of pesticides (NRC 1978~. The specific criteria for selecting suspect chemicals included production volume, chronic and acute toxicity, and environmental fate data, but the details of the review process were not widely and explicitly reported (NRC 1978~. Furthermore, not all of the registered pesticide formulations were reviewed. In 1975 the RPAR process replaced the SCR program, and most compounds on the suspect list were included on an initial list of RPAR candidates. Chlorobenzilate, for example, was judged to be carcinogenic in the Mrak report, was included in the SCR program's suspect chemical list, and was accepted as one of the original RPAR candidates when the SCR program was replaced by the RPAR procedure. (Chlorobenzilate was also one of the first candidate compounds actually inserted into the RPAR process.) In addition to relying upon the previous work of the SCR program, oPP also identifies RPAR candidates through reviews of applications for registration or reregistration. It is this review mechanism that constitutes the internal element in the candidate selection decisions. Finally, RPAR candidates are also identified as the result of referrals, that is, information sources external to oPP. Referrals can come from a wide variety of sources, including other branches of EPA, public interest groups (e.g., the EDF), the Congress, and even complaints by individuals (NRC 1978~. Treflan, captan, and 2,4,5-T, for instance, became = candidates as the result of such referrals (F. Arnold, oPP, EPA, Washington, D.C., personal communication, March 1979~. Referrals, which introduce an external element into the selection process, are frequently accompanied by some form of public pressure on oPP to act quickly.

OCR for page 46
Selecting and Scheduling Compounds for Assessment SELECTION OF COMPOUNDS FOR THE RPAR PROCESS 49 Once a compound becomes an RPAR candidate, the criteria for determining whether it should be subjected to the RPAR process are clear and scientific. These are the risk criteria described in Chapter 2. However, the procedures for determining the order in which the RPAR candidates receive attention or are inserted into the RPAR process (once it is determined that they exceed the risk criteria) are not clearly and explicitly defined. In some instances, high priority is assigned to those candidate compounds that involve use patterns similar to the use patterns of compounds that have previously been selected for RPAR (F. Arnold, oPP, EPA, Washington, D.C., personal communication, March 1979~. This selection criterion might be explained on the grounds that the risk- benefit analyses for such a compound are greatly facilitated if the analysts have had previous experience with the use patterns. In other instances, high priority is assigned to compounds that have relatively high exposure potential (F. Arnold, oPP' EPA, Washington, D.C., personal communication, March 1979~. The advantages of this selection criterion seem apparent: other things equal, the greater the potential exposure, the greater the potential risk to public health or the environment. Finally, external pressures on oPP are also presumably influential in some instances in determining which RPAR candidates are assigned relatively high priority, either for further evaluation as RPAR candidates or for insertion into the RPAR process. The same pressures (such as those from environmentalists, news media, or perhaps from Congress) that originally focused oPP's attention on a specific compound obviously can also result in the assignment of high priority to such compounds. EVALUATION It is difficult to describe the identification and scheduling procedure in more detail, since the basis for these determinations has not been well documented and reported. Early reregistration reviews were fraught with difficulty because of the disarray of EPA'S data files and the lack of data to satisfy current registration standards (NRC 19784. In fact, for a while in mid-1976, EPA'S registration and reregistration programs came to a virtual standstill. The type of toxicity data required by the 1972 amendments to FIFED iS not available for most older registrations, and it would take an inordinate amount of time and resources to generate such information as carcinogenicity test data. Efforts by EPA to obtain these

OCR for page 46
so REGULATING PESTICIDES data have often resulted in unavoidably long delays in regulatory activity. Compounds are put in a "holding pattern" when significant data are unavailable, and no determination is made as to whether they should be put through the RPAR process. The question is whether similar delays will again arise in the future or whether data necessary for future RPAR assessments are now being generated. Lacking a sufficiently defined and formalized internal system for identifying and ranking RPAR candidates, oPP may be more susceptible to external and even capricious influences. The lack of a logical, documented method of establishing priorities appears to be a critical weakness of this part of the RPAR process. It is sometimes suggested, for example, that oPP selects compounds for research and regulatory activity, not on the basis of a carefully reasoned decision, but on what the media uncover as the "pollutant of the week" (see Walsh 1978, for example). A serious flaw in the current procedure is that those compounds that receive the most publicity or pressure-group attention may not necessari- ly be those that present the greatest public health or environmental hazards. The current procedure does not provide for a broad comparison of the hazards posed by the large number of registered pesticides. At the same time, outside pressures to regulate a specific compound rarely arise from careful evaluations of comparative risks of alternative pesticides. To the extent that external pressures are influential in determining the order in which oPP evaluates compounds, the consequence may well be that considerable resources are devoted to regulating minor, low-risk compounds while important high-risk ones remain unreviewed for periods longer than would otherwise be the case. In an agency with limited resources, the process of deciding which activity will be the subject of a regulatory action must always be to some extent ad hoc. Nonetheless, the Committee believes that oPP has the potential to develop a more consistent scheduling policy by emphasizing the impact of the pesticide on human populations. RECOMMENDATIONS FOR ESTABLISHING A PRELIMINARY RPAR QUEUE An alternative to the current procedures for identifying and ranking compounds for in-depth evaluation is proposed in this section. It appears most important that a clear, openly documented method be established for this process. The process recommended here is an extension of the current internal oPP system of automatic RPAR "triggers" based on the risk criteria in 40 CPR 162.11, and it is designed to put somewhat more

OCR for page 46
Selecting and Scheduling Compounds for Assessment 51 emphasis on exposure and to establish comparative rankings of com- pounds in a priority queue. By following this system for identifying and ranking RPAR compounds, EPA would shift initial emphasis from toxic ejects (see Chapter 2) to allow greater weight to be placed on potential exposure of humans and other nontarget species (livestock, crops, and the natural biota). These modifications are designed to reduce the effects of haphazard and extraneous considerations that currently influence oPP's scheduling procedures, and to provide a workable scheme for a preliminary screening of all 35,000 registered formulations. The proce- dure is intended to ensure that compounds that present the greatest overall risk to human health and the environment are identified, given appropriate priority for a thorough benefit-risk assessment, and assigned the necessary resources to complete a review expeditiously. The following discussion of the recommended process is phrased in terms of risks to human health. It should be understood, however, that the possibility of significant adverse environmental effects is implicitly taken into account in our recommended procedure, although, as noted in Chapter 1, the report focuses on public health. The Committee recommends the establishment of a two-stage pre- RPAR selection and ranking system. The first stage concentrates on classifying compounds according to their acute or chronic toxicity. The second stage establishes a preliminary ranking that indicates the approximate order in which specific compounds are to be evaluated. Exposure is emphasized in the second stage. The initial assignment of priorities may be altered somewhat as the RPAR evaluations proceed; this possibility is discussed in the next section. Both stages are designed to employ the fragmentary and limited data that are likely to be available before a pesticide has been studied seriously. Thus, the resultant rankings are tentative, and many are likely to be altered as additional information accrues. The remainder of this section considers both the substantive aspects of the proposed ranking system and the procedural issue of making the recommendation operational. PRE-RPAR CLASSIFICATION OF COMPOUNDS About 514 technical-grade ingredients are used in active registrations (Jellinek 1979~. The first substantive step in ranking pesticides for review is to undertake a preliminary toxicity analysis of these 514 compounds (less those that have already been subjected to the generic standard process). The purpose of the preliminary toxicity analysis is to determine whether evidence suggestive of acute or chronic toxic activity that may present unique hazard exists. The preliminary analysis would be

OCR for page 46
52 REGULATING PESTICIDES conducted according to the Agency's Health Risk and Economic Impact Assessments of Suspected Carcinogens: Interim Procedures and Guidelines (U.S. EPA 1976), and would be based only on data easily accessible in oPP or company files or in major scientific journals. It would not replace the worldwide literature search and the more thorough toxicity analysis conducted for the RPAR candidates (see Chapter 2~; these would still be done, but at a later stage in the process. The rough, preliminary analysis is intended only to provide some indication of the compounds' inherent toxicity, not to serve as a substitute for later, more careful evaluation (if it is undertaken). One problem recognized but not addressed by the PEAP Committee is that of impurities introduced into the active ingredients or the pesticide formulations by the complexities of industrial synthesis processes. Generally, the data on which preliminary toxicity analyses would be based do not include an assessment of impurities contained in commer- cial preparations. In fact, animal bioassays typically employ pure preparations, although in some instances it is the impurity in a pesticide formulation that poses the human health hazard. Thus, it is apparent that consideration must be given in the preliminary toxicity assessment to the possibility of potentially hazardous impurities entering commer- cial preparations. On the basis of the preliminary toxicity assessment, each formulation should be placed in one of three classes: Class A, apparently a potential toxic hazard (an RPAR candidate); Class B. insufficient data to permit a reasonable judgment; and Class C, no evidence to suspect potential hazard when used as directed. This classification scheme, illustrated in Figure 3.1, is not fixed and irreversible, but will be subject to further review as additional pertinent data become available. It should be noted at this point that, in addition to the acute and chronic toxicity criteria, a third type of risk can currently trigger an RPAR proceeding. This is the absence of an antidote or other emergency treatment for toxic erects in humans from a single exposure to a pesticide (see Chapter 2~. Technically, this criterion alone, whether or not the acute toxicity criterion is met or exceeded, can trigger an RPAR proceeding. The Committee has not factored this third criterion into its preliminary selection and ranking scheme, but has chosen rather to concentrate on the acute and chronic toxicity criteria. It is expected that if information came to light during the selection and ranking procedures indicating that this third criterion may have been reached, the pesticide in question would be placed in Class A (an RPAR candidate) to be subsequently ranked within that grouping according to the scheme recommended in the next subsection.

OCR for page 46
Se~- ~d ~^~ ^~ ~r ~' Stage I: Underlake a quick, preliminerv risk assessmen1 of the ~or-so aclive pesllcide ingredienls Involving ~1- -1~1~ wl~ ~e in~ndon ~ ~n~-ly ~l~i~ -~ ~id~ ~mul~i~ ~ ~e ~ ~e hree celenorles ~low: ! + CLASS A CLASS B Apparenllv Dangerous ~Insufficienl Dala R~R ~ ~ a Delermlnallon T . 1 I _ ~ _ _ _ __. . ___ St~ I I : 1 F Schedule for addilional sludv, _ ~- i~ he grealesl exp~ure receiving . 1he earliesl alle~lion. Does furlher sludv reveal 1ha1 a ~ ~ d~ _ 1 r ~ , YES 1 NO ~ 1 r . + CLASS C No ~1 lo Suspecl Polenlial Hazard When Used as Direcled ~hedule ~slicides 1ha1 use i_~ ~ R~ review in accordance ~e u~n~ ~ he sllualion (ur~ncv is a funclion of 1oxlc polencV and lhe exlenl of exposure). ~ ~ ' ~ ~ ~e l1 Regisler or re- regisler all incorporale 1he aclive ingredienl. FIOU~ 3.1 R=o~nded pro~ss ~r sel~ting and ra^g ~shci~ ~at~= ~r assessment (see tex1 ~r ~sc~sion). 1be co~ounds in Class C- ~bicb include tbose co~ounds of apparendy low to~ciV and ~r ~bicb e~c~ve an~dotes are av~able require no ~rtber consideration by oPP (u~es~ of course, new evidence reveals ~ classiRcation e~o0. Accordingly, all cu~ently active registra- hons involving tbese compounds sbould be renewed ~db~itb. In contrast tbe Class A ~nd B co~ounds do re-he ~r~er attention. Howeve~ ~nce tbere m~ be bundreds of co~ounds ~ tbese

OCR for page 46
54 REGULATING PESTICIDES two categories, the order in which they are considered is especially important. We now turn to this issue. PRE-RPAR RANKING OF COMPOUNDS Class B Compounds Since oPP can devote resources to only a few compounds at a time, it is necessary to rank Class A and B compounds in some order of priority. Class B compounds require further study to determine whether they pose problems of acute or chronic toxicity. The order in which these compounds receive attention should be directly related to their exposure potential. Other things being equal, those compounds to which many people (or many members of other significant nontarget species) are likely to be exposed should be assigned highest priority for further study. Compounds that are little used should be assigned relatively low priorities, unless there is some evidence suggestive of high toxic activity. Efforts should then be initiated to acquire the preliminary data needed to reclassify these compounds, beginning with those with the greatest exposure potential. When a compound has been reclassified, it either takes its place in the Class A list or is eliminated from further consideration, as a Class C compound. Much of the work necessary to rank the Class B compounds (and probably the Class A compounds as well) has already been completed; oPP recently classified all 514 actively registered technical-grade ingredi- ents as compounds of "major," "average," or "minor" significance (Jellinek 1979~. The criteria for judging significance are production volume, numbers of registered products, residue tolerances issued, and volume and type of use. According to these criteria, only about 65 compounds are of major significance. Another 120 compounds are considered of average importance, and some 300 technical-grade ingredients are of minor significance. In accordance with these designa- tions, the compounds in Class B that are of major significance would be the first to receive closer scrutiny in order to more clearly define the risks. If further testing is needed, oPP would probably have to seek the assistance of a research agency such as the National Cancer Institute, contract with a private testing laboratory, or require the testing of a registrant. (Prior to 1978, FEPCA did not give EPA the right to require additional testing by a registrant. However, as part of the 1978 revision of the standards for data sharing among applicants, EPA now has the power to require the submission of additional data that EPA feels are

OCR for page 46
Selecting and Scheduling Compounds for Assessment 55 necessary to support a continued registration.) Attention would be directed next toward the Class B compounds in the average-significance category, and finally toward those whose significance is considered minor. Of course, it may be desirable to further order the Class B compounds within each of these three groups of significance. The main steps in this procedure recommended for dealing with Class B com- pounds are illustrated in Figure 3.1. Class A Compoumls The Class A compounds are the RPAR candidates. oPP must determine which of these compounds (or the products in which they appear) reach or exceed the risk criteria set forth in 40 CFR 162.11. Those compounds identified as reaching or exceeding the risk criteria will be subjected to the RPAR process. According to the Committee's recommended proce- dure, the order in which these compounds (or the pesticides that contain them) should be examined more closely would be based on a preliminary risk assessment that accounts, in a rough fashion, for both potential exposure of human populations (or other nontarget species) and a compound's toxicity. That is, the preliminary risk assessment should consist of both a preliminary toxicity analysis and a preliminary exposure assessment. The preliminary toxicity analysis has already been discussed in the section on Pre-RPAR Classification of Compounds. The purpose of the preliminary exposure analysis would be to identify those active ingredients to which significant numbers of people are exposed. Such exposures may occur either through persistence of residual quantities of pesticides in the food supply, or by contamination of water supplies or other environmental media. Briefer but more intense exposures may also occur during the manufacture, transport, and application of the pesticide formulation. To the extent permitted by existing, easily accessible data, both the amount of active ingredient that is likely to reach a typically exposed person in a given population and the number of people exposed to each dosage should be taken into account in this preliminary survey. The key factors in this accounting should be the chemical's environmental chemistry and the use patterns of the major products in which the active ingredient appears. As noted in the discussion of Class B compounds, oPP has already completed a substantial part of the work required for this preliminary exposure analysis and, thus, for the preliminary risk assessment. The preliminary risk analyses should be used in ranking the Class A compounds. The preliminary ordering should take into account both the

OCR for page 46
56 REGULATING PESTICIDES compound's exposure potential and its toxicity. Accordingly, highly toxic compounds with relatively high exposure potential should be assigned high priority and considered first, whereas weakly toxic, low- exposure compounds should be classified as low priority and considered last. Assignment of these priorities would necessitate a number of difficult and arbitrary decisions. It is not obvious, for example, how to order pesticides posing mainly environmental hazards relative to those likely to impair human health. The current practice of placing the greatest weight on human health hazards should probably be continued until such comparisons can be made more soundly than at present. Nevertheless, whatever ranking system emerges from these choices will involve less arbitrariness than the current selection and scheduling procedure. More important, the suggested ranking system would enhance greatly the likelihood that oPP's limited resources will be devoted systematically to regulating those compounds posing the greatest hazards to human health and the environment. Although the ranking system would become a subject of controversy, it would serve an important function by broadening the preliminary screening to include all active ingredients and by reducing the influence of unwarranted pressure on oPP's decisions. The procedure described above determines the approximate order in which Class A compounds should receive further consideration from oPP. The ordering should not be regarded as fixed and irreversible: it would no doubt be revised as new data become available. Moreover, in order to reach correct RPAR decisions on specific compounds, it would sometimes be necessary to promote certain other pesticides to a higher position on the priority list. The latter possibility involves a number of complexities that are considered below. First, however, we discuss briefly one possible procedure for undertaking the preliminary risk screening discussed in this section. CONDUCTING THE PRELIMINARY RISK SCREENING In operation, the preliminary review, classification, and ranking shown in Figure 3.1 could be performed by a group of outside experts. Relying upon outside experts for an independent review and ranking would avoid placing significant new demands on oPP personnel and would enable oPP to enlist the skills of specialists not on its full-time staff. A review system similar to that used by the National Library of Medicine (NEM) in its Toxicology Data Bad program might be used. In that program, an outside review committee is appointed, and each member is given data summaries for 1~15 chemicals. The summaries

OCR for page 46
Selecting and Scheduling Compounds for Assessment 57 are supplied in computer format by the NEM through a contractor. Each committee member reviews the data on his or her assigned compounds for completeness and accuracy and is responsible for evaluating the data when the committee meets. In fact, the NEM has already reviewed many pesticides. In this same spirit, oPP could establish a pesticide screening committee consisting of about a dozen consultants including toxicologists, pharma- cologists, oncologists, and agricultural economists. The committee would have two functions and would be disbanded when those functions had been discharged. The first would be to classify the 500-odd active ingredients used in pesticide formulations into the three classes described earlier: Class A apparently a potential toxic hazard (an RPAR candi- date); Class B insufficent data to permit a reasonable judgment; and Class C-no evidence to suspect potential hazard when used as directed. The second function would be to rank the Class A ingredients in order of importance. It is visualized that such a committee would meet approximately one day a month and could be expected to classify active ingredients at the rate of about 10 at each session. The procedure envisioned is roughly as follows: At the first meeting or two the committee would have available the list of actively registered technical-grade ingredients and limited information about them, including the classifications already established into ingredients of"major," "average," or "minor" significance. On the basis of this information the committee would decide on the order in which ingredients are to be taken up for classification; ingredients of major significance that appear likely to be toxic on the basis of readily available information would be dealt with first and those of minor significance that are judged likely to be innocuous would be placed at the bottom of the list. Then the work would begin. The classifications are to be based on information already available in EPA files, in published literature, in the NEM'S Toxicology Data Bank and similar sources, and especially in the accumulated knowledge of the members of the committee. Obviously, the accuracy of the preliminary risk assessments and subsequent ranking of compounds will depend significantly on the quality of the data available and would be subject to modification as later data came to light. Each of the ingredients would be assigned to one or two committee members, who would be responsible for reviewing the sources of information and for reporting a digest of the information to the committee, with a recommended classification. The entire committee would then consider the evidence reported and arrive at a classification. Subsequent consideration would then diner depending on the class in question.

OCR for page 46
58 REGULATING PESTICIDES Class C ingredients do not require further consideration unless information indicating adverse affects on public health or the environ- ment should come to light at a later date. Class B ingredients do require further work. When the committee assigns a chemical to this class, it should specify the minimal information it would need in order to reclassify it into one of the other classes. Then oPP should attempt to acquire the requisite information, and when it is obtained, the chemical should be returned to the committee's agenda for reclassification. Class A compounds are the candidates for possible RPAR proceedings and the committee should assign a priority rank to this list according to the procedures recommended earlier. By the time the committee has been operating in this manner for half a year, an agenda of some 5~60 Class A ingredients will have been accumulated, all of them considered to be seriously toxic and to have the potential for widespread exposure. This agenda could then serve to schedule the more thorough reviews needed to determine whether pesticides incorporating that ingredient should be registered or rere~s- tered, or whether the RPAR risk criteria set forth in 40 CFR 162.11 have been tripped. MODIFICATIONS TO THE PRELIMINARY RANKING: THE ROLE OF ALTERNATIVE PESTICIDES Once the tentative rankings have been completed, oPP should consider the compounds in order of priority. First, the pre-RPAR review would be conducted on the top-priority compounds to determine whether to initiate an RPAR proceeding. After an RPAR proceeding is begun, it will sometimes become necessary to promote some of the lower priority compounds (or at least some of their uses) to the front of the RPAR queue. The reasons and procedures for making such modifications to the initial priority queue are described in this section and presented in Table 3.1. THE PROBLEM The purpose of the benefit-risk assessments is to determine, for each use of an RPAR pesticide, how various regulatory options are likely to influence the level of risks and benefits arising from pest control activities. The public health, environmental, and economic ejects of any regulation depend not only on the extent to which the regulation changes the use of the pesticide in question, but also on the changes it induces in alternative methods of pest control and in the public health, environmen- tal, and economic ejects of the alternative control measures. It is quite

OCR for page 46
Selecting and Scheduling Compounds for Assessment 59 TABLE 3. ~ Decision Sequence for Uses of the RPAR Pesticide for Which Class A or B Pesticides Are Substitutes Result of Analysis on Initiating Action on Initiating Assumption Compound Compounda No Class A or B Risks > benefits Disallow use or substitutes will impose restrictions rema n available Benefits > risks Go to Step 2 All Class A or B Benefits > risks Reregister for this use substitutes will remain available Risks > benefits Go to Step 3 Obtain RPAR-type data on risks and benefits of major Class A substitute pesticides. Go to Step 4. Evaluate Class A substitutes sufficiently to form confident expectations as to which will be available. Assume Class B substitutes in use for 3 years or more will continue to be available. Go to Step 5. Derived from Step 4 Benefits > risks Risks > benefits Reregister for this use Disallow for this use a The only final decisions listed explicitly are "disallow" and "reregister." In many cases, restrictions on use (such as relabelling) may be feasible. A use subject to such restrictions should be regarded as a new use, and the same sequence of decisions should be followed. b The 3-year time period is chosen for illustrative purposes and is intended to separate those compounds that have been in use long enough for some health or environmental problems to have surfaced. Further consideration may indicate that another measure is preferable. possible for a regulation to have an adverse effect on public health if, for example, it induces more widespread use of a substitute chemical that is more toxic than the one being regulated. Even aside from this extreme possibility, a regulation may be unwarranted if it stimulates the use of alternative pesticides that are nearly as harmful to public health or the environment and that have substantially greater economic costs. To guard against these contingencies, two sorts of information are required: (1) the extents to which pesticide users are likely to resort to specific pest control alternatives, and (2) the effects of those alternatives on public health, the environment, and economic costs of protection. It is already routine to include in the analysis of any regulatory option an estimate of its ejects on the use of alternative methods (chemical and biological) of pest control to which users are likely to resort. Such an estimate,

OCR for page 46
60 REGULATING PESTICIDES however, will often be deficient because some of the regulatory options considered entail the use of pesticides that are awaiting RPAR proceed- ings, with results that cannot be foreseen. The systematic screening procedure recommended above will reduce the prevalence of this difficulty but cannot be expected to eliminate it entirely. The second type of information mentioned above relating to possible adverse erects of alternative methods of pest control is not generated adequately under current procedures. At present, the RPAR team must make the best conjectures that it can about erects of substitute pesticides. The substitutes usually have not been subjected previously to systematic review. The difficulties just described can present evaluators of a contemplat- ed regulation with baffling perplexities. While the best information that can be obtained about the pesticide being evaluated may be at hand, it is usually the case that comparable information about alternative pesti- cides is lacking. Therefore, it is extremely difficult to estimate changes in benefits and risks that would result from the adoption of various regulatory options. RECOMMENDATIONS It follows from the forgoing discussion that assessment of the conse- quences of any regulatory option requires information about the public health, environmental, and economic effects of the regulated pesticides and comparable information about the effects of its principal alterna- tives. To generate this information for chemical alternatives, the Committee recommends that as soon as one of the high-priority compounds is assigned to the RPAR process, the RPAR team should identify all of the compound's uses and the alternative pesticides for each use. In some instances, restrictions associated with data, time, or budget may limit attention to a compound's major uses and the related alternative pesticides. For any one use of the RPAR compound, there may or may not be chemical alternatives. For those uses for which there are no economically viable alternative pesticides, the RPAR evaluation should proceed as it currently does (amended by some of the Committee's recommendations in this and subsequent chapters). If the alternatives for a specific use are all Class C compounds (that is, presumed to be safe when used as directed), the same rule should apply: the RPAR evaluation for that use can proceed as at present, since it is realistic to assume that Class C compounds will continue to be available (unless the registrants voluntar- ily choose to withdraw them because they are not profitable).

OCR for page 46
Selecting and Scheduling Compounds for Assessment 61 If some of the alternative pesticides are in either Class A (RPAR candidates) or Class B (compounds for which there are insufficient data), the decision-making procedure becomes more complicated. A helpful tabular summary is presented in Table 3.1. It will be seen that the decision sequence is designed to minimize the amount of attention that has to be paid to substitute formulations. Consider first those uses for which one or more of the alternatives are either Class A or Class B compounds (some may also be Class C compounds). In these cases, the RPAR evaluations for the initiating compound should first explore the possibility that the Class A or Class B alternatives will not continue to be available (i.e., that their registrations for the specific use in question will be cancelled). On Table 3.1, this is Step 1. If the risks of using the RPAR compound appear to outweigh the benefits when no Class A or B alternatives are available (and the benefits of using the RPAR compound are therefore at a maximum), the same state of affairs will prevail when one or more such alternatives are available. There is no need to explore further; the registration or reregistration for that use should be denied or severely restricted. The scenario differs considerably if oPP determines that the benefits of a specific use of the RPAR compound outweigh the risks under the assumption that none of the Class A or B substitutes will be available (Table 3.1, Step 1, second possible result of RPAR analysis). The initiating compound should not necessarily be reregistered in this instance, since one or more of the Class A or Class B alternatives may be available and the advantages of the RPAR compound over the alternative may be insufficient to justify the additional risks entailed. To determine whether this possibility is, in fact, realistic, the benefit-risk estimates for the RPAR compound should next be recalculated under the assumption that all of the Class A or Class B alternatives for the use in question will be available (i.e., reregistered). In Table 3.1, this is Step 2. If, under this assumption, the RPAR compound continues to over benefits in excess of the risks, then it should be reregistered for this specific use. In this instance, it is not important to have a well-founded judgment about the outcome of eventual RPAR proceedings against the Class A alternatives (or against those Class B alternatives that become Class A compounds). If, however, in Step 2 the risks of continued use of the RPAR compound outweigh the benefits, it now becomes essential to have information about risks and benefits of the alternatives. If such information is not already available, the Class A alternatives will have to be pulled out of their places in the queue and given instant attention. They might be assigned to the same RPAR team, which will already be familiar with the economic and technical aspects of their use, or other teams may be

OCR for page 46
62 REGULATING PESTICIDES organized to develop the requisite data. The hard fact has to be faced that information about the alternatives is needed for a sound decision, and that decision can only be imperiled by waiting until the alternative compounds reach the top of the queue. Thus, to obtain necessary information, it is necessary to issue "qualified" RPAR'S for at least some of the Class A alternatives (Table 3.1, Step 3~. These ancillary RPAR'S are qualified in that they apply only to the use in question. For example, in connection with the chlorobenzilate RPAR, this recommendation could well lead to an auxiliary RPAR being issued for the use of dicofol as a citrus miticide, but not for dicofol's other uses. The ancillary RPAR'S need be issued only for the crucial Class A alternatives, that is, those alternatives responsible for the ambiguity in the benefit-risk test for the initiating compound. In some instances all of the Class A alternatives will be crucial. However, in other instances some may be of such minor importance for the use in question that they would not reach or exceed the risk criteria in 40 CFR 162.11 and thus can be ignored. Clearly, the issuance of these additional RPAR'S implies a selective reordering of the preliminary priority queue discussed above. The auxiliary RPAR'S will delay a final decision on the initiating compound. However, these delays are unavoidable if correct decisions are to be reached. Moreover, the length of the delays should be considerably shortened over time as the data base for all Class A compounds improves. It becomes important to have a well-founded judgment about the availability of Class A and Class B alternatives if a use of the RPAR compound appears to entail risks in excess of the benefits when the alternatives are assumed to be available (i.e., at Step 2 in Table 3.1, second result of the RPAR analysis). The benefit-risk test implies different courses of action depending upon whether some or all of the alternatives continue to be available. Consequently, the question of continued availability of the crucial Class A alternatives determined by conduct- ing ancillary RPAR'S must be addressed before oPP can reach a correct decision on the initiating compound (Table 3.1, Step 4~. Finally, there remains only the situation in which (for a specific use) one or more of the alternatives to the initiating compound are Class B pesticides (that is, those pesticides for which the existing data do not permit their assignment into either Class A or Class C). The Class B alternatives should be incorporated into the benefit-risk evaluations of the initiating compound as follows. First, they should be divided into two subgroups: (1) those that have been in use for, say, 3 or more years, and (2) those that have been in use for less than 3 years. The dividing line is intended to separate those compounds that have been in use long

OCR for page 46
Selecting and Scheduling Compounds for Assessment 63 enough for at least some health or environmental problems to have surfaced. The 3-year time period is an arbitrary suggestion; further consideration may indicate that some other period of time or a measure including the extent of use is preferable. Next, the benefit-risk evaluations for the initiating compound should tentatively assume that those Class B alternatives that have been In use for the critical number of years will continue to be available (Table 3.1, Step 4~. In contrast, it should be assumed that the recently registered Class B alternatives will not continue to be available unless there is at least some weak evidence to the contrary. The number of recently registered compounds that fall into Class B is likely to be rather small because of the testing requirements of the FIFRA legislation (as amended). As soon as the qualified RPAR evaluations for the crucial Class A alternatives are completed and the appropriate assumptions for the Class B alternatives are adopted, oPP can reach a sound final decision as to which toxic compounds to reregister for the use in question (Table 3.1, Step 5~. For present purposes, the relevant aspect of the preceding discussion, table, and earlier figure is the guidance they give concerning the order in which pesticides should be subjected to the RPAR procedure. To summarize, an initial ordering is established on the basis of preliminary information about the toxicity of active ingredients and the extent of human (and other) exposure to pesticides that incorporate them, with widely used compounds that contain the most toxic ingredients receiving the highest priority (Figure 3.11. When any pesticide is reached on the priority list, substitutes tor it In its major uses are identified. If any of these substitutes are in Class A or B and have not already been issued an RPAR, then the decision sequence summarized in Table 3.1 has to be followed for each such use. At Step 2, if it is reached, the pertinent substitute pesticides have to be pulled out of their low places in the queue and scheduled for early attention. An important component of several steps is the determination of "Risks > Benefits" or vice versa. We have seen that such determinations require that the risks and benefits of alternative methods of pest control be taken into account, and therefore that adequate data on those alternatives be obtained by issuing qualified RPAR'S or by other means. REFERENCES Jellinek, S.D. (1979) Regulations and the national interest. Chemical and Engineering News 57(8):37-39.

OCR for page 46
64 REGULATING PESTICIDES National Research Council (1978) Pesticide Decision Making. Vol. VII, Analytical Studies for the U.S. Environmental Protection Agency. Committee on Pesticide Decision Making, Commission on Natural Resources. Washington, D.C: National Academy of Sciences. Pesticide and Toxic Chemical News (1978) RPAR Nov. 2 status report from special pesticide reviews division noted. November 8, 1978, pages ~4. Washington, D.C.: Food Chemical News, Inc. U.S. Department of Health, Education, and Welfare (1969) Report of the Secretaty's Commission on Pesticides and Their Relationship to Environmental Health. Parts I and II. Washington, D.C.: U.S. Government Printing Office. U.S. Environmental Protection Agency (1976) Health Risk and Economic Impact Assessment of Suspected Carcinogens: Interim Procedures and Guidelines. 41 Federal Register (102)21402-21405. Walsh, J. (1978) EPA and toxic substances law: Doling with uncertainty. Science 202:598- 602.