Tobacco Documents Online

REGULATORY DECISION-MAKING UNDER UNCERTAINTY: THE CASE OF ALAR* * This is a teaching case commissioned by the Harvard Center for Risk Analysis for use by the Office of Continuing Education, Harvard School of Public Health, for a course on Risk Analysis in Envirozunental & occupational Health, September 5-7, 1990. The case was written by Ms. Susan Moses under the supervision of Dr. John D. Graham.

As Jack Moore, Acting Administrator of the U.S. Environmental Protection Agency (EPA), hung up the phone, he wondered whether he should agree to an interview with Ed Bradley for a 60 Minutes segment on;pesticides. It was January 1989, George Bush had just been inaugurated as President, and the Administration had not put forth its policies on environmental issues. Jack Moore was Acting Administrator of EPA. He had most recently held the position of Assistant Administrator for Pesticides and Toxic Substances at EPA. As an advocate for the use of sound science in the regulatory decision- making process, he had good working relationships with both industry and environmentalists alike. Twice before Dr. Moore had been asked to appear on 60 Minutes (prior to being in the position of Acting Administrator), and both times he declined. He was under no pressure from The White House; it was his decision whether to grant 'L-he interview or not. Not too long ago, Dr. Moore had received an informal copy of a not-yet-released report on pesticides in children's food, written by the Natural Resources Defense Council (NRDC), an advocacy group. He now began to wonder whether or not there might be some connection between this report and the:scheduled 60 Minutes segment, particularly since Ed Bradley had referred to the pesticide Alar, a registered trademark for the chemical daminozide that is sprayed on apples. Alar, and its metabol ite unsymmetrical dimethylhydrazine (UDMH), were highlighted in the NRDC report as potential hazards. 2

Knowing that 60 Minutes is watched by millions of viewers, Jack Moore began to evaluate the implications of his appearance on the show and wondered how he should prepare himself if he agreed to the interview with Ed Bradley. THE F7PMERAL INSECTICIDE, FUNGICIDE, AND RODENTICIDE ACT (FIFRA) The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRik), enacted in 1947, is the federal law regulating pesticide producats and their use in the United States. Until 1972, the law focused on the proper labeling of pesticide products. In 1972, amendrncants Were passed changing FIFRA from a labeling law to a more comprehensive statute that charged EPA with the responsibility of premarket data review and registration. These changes reflected public: concern about potential adverse health effects and the need to evaluate the "reasonableness°t of any of these risks. Since 1972 there have been a series of FIFRA amendments, and the debate over the adequacy of the current law in protecting human health and the environment;continues today. F]:FRA is a "risk-balancing" statute. EPA weighs any potential adverse effects of the product against its benefits as part of the decision-making process. The operating words of the statute are that the pesticide, when used as directed, "will not cause any unreasonable risk to man or the environment, taking into account the economic, social, and environmental costs and benefits of the use of' the pesticide." This risk/benefit mandate is in contrast 3

to other laws, such as the Delaney Clause of the Federal Food, Drug, and Cosmetic Act, which bans the use of any intentional food additive that is shown to be a carcinogen in humans or animals, regardless of any benefits. All pesticide products must be registered by EPA prior to being marketed and distributed in commerce. Manufacturers must submit toxicological and environmental data to the Agency as part of their application for product registration. EPA reviews these data for short-term and long-term toxicity, mutagenicity, oncogenicity, fetotoxicity, teratogenicity, for effects on wildlife and other organisms, and for environmental fate and residues in food or feed. 'Under the Federal Food, Drug, and Cosmetic Act, EPA sets tolerances for pesticide residues in food. A tolerance is the maximu,m level of residue permitted in the raw harvested commodity and on processed forms of the commodity. If EPA approves a pesticide registration application, the product may be manufactured for.commerce. However, each product is approved to control specific pests, for use on specific crops, and for use in specific concentrations and frequency of appl ication . A:Ll pesticides that were on the market prior to the enactment of FIFIZA were /0grandfathered" under the statute. Therefore, it is possib:Le that many of these products would not be approved today under the current pesticide registration procedure. However, EPA is in the process of reviewing these "existing" pesticides to 4

determine whether any of them should be withdrawn from the market because of potential harmful effects. The Agency is under Congressional pressure to move more quickly in its evaluation of these: grandfathered substances. The burden of proof as to the "safety" of a new pesticide rests with the manufacturer submitting the registration application. If for whatever reason the Agency determines that the terms of the registration are not being met, EPA may begin the "Special Review" process for canceling the registration. At this point, the burden of finding "unreasonable risk" is shifted to the Agency. This review is a process, formerly called the Rebuttable Presumption Against Registration (RPAR), whereby EPA collects and evaluates information on the pesticide and can request additional information from the manufacturer to determine whether any uses of the pesticide 00cause unreasonable adverse effects to human health or the environment." Depending on the nature of the new data, the Agency may propose changes to the terms of the registration under the rationale that such changes reduce risks to the level where the benefits outweigh the risks; or, EPA may proceed with cancellation by issuing a "Notice of Intent to Cancel" if the Agency finds that the risks outweigh the benefits. Throughout this entire process, the palalic has the opportunity to submit comments in an effort to affect any regulatory action. 5

THE 8K`IEISTIFIC ADVISORY PANEL (SAP) The S.cientific Advisory Panel (SAP), a standing advisory committee, was mandated in 1975 by FIFRA to review, for potential regulatory action, EPA's evaluations of environmental and health risks posed by specific pesticides. Regulatory History of Alar Alar, the Uniroyal Chemical Company trade name for daminozide, was first registered with the Environmental Protection Agency (EPA) as a plant growth regulator for potted chrysanthemums in 1963. The first registered food use was for apples in 1968. To apple growers, Alar was a major breakthrough; ripe apples stayed on the tree longer, and remained firmer and redder (better market quality) during harvest and storage. The tolerance (maximum permissible residue level) for Alar in apples was set at 30 ppm. From 1968- 1985 Alar was also registered for use on cherries, nectarines, peaches, pears, grapes, peanuts, tomatoes, and ornamental plants. However, in 1985 it was estimated that approximately 75% of the damino.ide in commerce was used on apples; since that time usage has ds:clined significantly (EPA, 1989). ]:ra the summer of 1984, EPA issued a Notice of Initiation of a Special Review of pesticide products containing daminozide, which indicated that the Agency was going to investigate potential harmful, effects of the pesticide. Of particular concern was a 6

degradation product of daminozide, unsymmetrical dimethylhydrazine (UDMH). Data from animal studies indicated that both daminozide and L;rLaNiH elicited "statistically and biologically significant oncogenic responses at multiple organ sites in multiple species and straina of animals. UDMH was believed to be a very potent animal carcinogen and mutagen." (EPA, 1989) Although the database was limited, the Agency decided to proceed with a cancellation action„ A year later, in the fall of 1985, EPA developed a combined Prelininary and Final Determination and Draft Cancellation Notice. The process was accelerated in light of the potentially high dietary exposure of daminozide and UDMi to children. The EPA Scientific Advisory Panel (SAP), required under FIFRA to review this documentation, believed that while the data raised concerns, they were not sufficient to support a quantitative risk assessment for either daminozide or UDMH. The Department of Agricul.ture (USDA) also reviewed the report and argued that EPA had underestimated the benefits of daminozide use, and therefore, should reassess its call for cancellation. Although not legally bound by the SAP decision, the Agency decided to reassess its position based on the SAP recommendation, and chose not to proceed with the cancellation action. The Agency did require Uniroyal to conduct additional testing and collect additional data on the oncogenic risk of daminozide and UDMH. In the interim, to reduce exposure, EPA lowered the tolerance for daminozide residues on apples from 30 ppm to 20 ppm. However, this tolerance was set to 7

expire on July 31, 1987 at which time EPA believed it would have additional data to evaluate the tolerance further. The Agency also instructed Uniroyal to include a use advisory with its product warning not to use the chemical on apples intended for use in apple sauce and apple juice. (When apples are processed into apple sauce and juice, the heating process causes daminozide to break down into UDMH. Therefore, these products have higher concentrations of UDMH.) At the time, Jack Moore felt that there was enough evidence for EPA to be concerned about the carcinogenic potential of daminozide and its metabolite UDMH, but not enough from a legal point of view to regulate Alar under FIFRA. Unlike the requirements for new pesticides where the registrant must bear the burden of proof that "the intended use will not present an unreasonable risk," for currently registered chemicals such as daminos,ide, the burden of finding "unreasonable risk" lay with EPA. In 1985-86, when the carcinogenic potential of Alar was made public:,, consumers acted predictably--they stopped buying apples until they were assured by their grocers and food processors that the apples in their stores and products were Alar-free. The protest was relatively calm, and short-lived. Civ,er the next several years (1986-88), NRDC, Public Citizen, and the: States of New York and Maine petitioned and then sued EPA for noi> amending the tolerance for daminozide residue to zero. The Ags:ncy claimed it did not have sufficient data to determine 8

whethe:r these residues (20 ppm) posed a health hazard to the U.S. population. The case was dismissed. In 1987, the available residue and toxicological data were not sufficient to determine if a new tolerance would be adequate to protect public health. Therefore, the 20 ppm tolerance was extended to January 31, 1989 when new data would be available. Beginning in 1988, tests by independent laboratories revealed that claims by one grocery chain of selling Alar-free apples and products were false. Once again, fear of eating Alar-treated products caused some consumers to stop buying apples and apple products. In early January 1989, EPA staff recommended to Jack Moore to seek cancellation of all food uses of daminozide. This decision was based on the new data collected that indicated "that dietary exposure to UDMH represents a significant carcinogenic risk which outweighs the benefits of use of daminozide on food crops and therefore warrants the cancellation of the food uses of daminozide" (EPA, 1989, p.ii). Agency staff estimated the lifetime risk of cancer for the general population from dietary exposure to UDMH to be 4 in 100,000. Infants up to one year are considered the highest exposure group. EPA also estimated the benefits derived from the use of daminozide. The calculations were based on the economic impacts IU ~ that would result if daminozide were banned. The greatest impact ~ ~ would be on the apple industry, as there are "no alternatives to ~ damino2;ide -that alone will accomplish all of the growth regulator A ~ ~ ~ 9 ~

benefits attributed to daminozide" (many of which relate to the appearance of the fruit). The overall effect on all growers is estimated to be "an annual income increase of $1.5 million" resulting from higher market apple prices, with daminozide users losing $14.5 million and non-users gaining $16.1 million. Growers of certain apple varieties (particularly Eastern McIntosh and Stayman), however, may experience annual income losses of $5.7 and $1.8 million, respectively. In addition, a cancellation of Alar was estimated to reduce the supply of fresh apples. "The net social cost (total society cost) of cancellation of daminozide use on apples based on 10 percent of the crop treated is estimated to range from $18 to $81 million as compared to $44 to $198 million for 1985 usage levels. Economic impacts of a cancellation for other uses of daminozide, such as on cherries, grapes, peanuts, and ornamentals, are predicted to be much less significant. In addition to the apple industry, peach growers' losses were estimated to range from $1.5 to $5.!5 million (EPA, 1989). '.rhe Agency staff did not recommend issuing an emergency suspension of daminozide use on food crops because while the data did indicate cause for concern, "the level of risk during the time neces;u.,dry to complete a cancellation action is not unreasonably high." (EPA, 1989). According to FIFRA, an immediate suspension is warranted only if EPA determines that the risks present an immediate hazard. In the interim it was also expected that usage would decline, thereby lowering the risk of exposure. 10