Chapter 3*: Prevention of Harm in the Face of Uncertainty - Toxics Regulation

*This was formerly Chapter 4.  The chapter has been renamed, reorganized, and expanded.

INFORMATION ABOUT CHEMICAL RISKS: THE REACH PROGRAM & BISPHENOL-A

THE CONSUMER PRODUCT SAFETY IMPROVEMENT ACT OF 2008

Problem Exercise: The Case of Alar - From the 4th Edition

Hot Issues: EPA Reduces TRI Reporting

 INFORMATION ABOUT CHEMICAL RISKS: THE REACH PROGRAM & BISPHENOL-A

As discussed in Note 4 on pages 199-200, the European Union has launched a far-reaching new program of pre-market chemical testing called the Registration, Evaluation and Authorization of Chemicals (REACH) program.  The program entered into force on June 1, 2007.  Canada has a similar pre-market testing program that already is having an impact on chemical use.  For example, in 2006 Canada targeted bishpenol-A (BPA), a chemical widely used in a variety of plastic materials, including baby bottles, for safety screening based on studies suggesting that it is “inherently toxic.” In April 2008 Health Canada concluded that the chemical may pose some risk to infants because of its developmental toxicity.  As a precautionary measure, the Canadian government proposed to ban the import of baby bottles containing bisphenol-A and Wal-Mart Stores, Inc. announced that it would no longer carry such products.  At the same time Nalgene, a U.S. manufacturer of popular outdoor goods including plastic water bottles, announced that it no longer would use the chemical in its products.  EPA and the Food and Drug Administration are taking the position that bisphenol-A has not been shown to be a health hazard. While the Centers for Disease Control report that 93 percent of Americans now have traces of BPA in their urine, the FDA maintains that these exposures are at levels far too low to be dangerous. FDA Scientists Say Common Chemical Is Safe, Wall St. J., Aug. 16-17, 2008, p. A2.  The American Chemistry Council is waging an aggressive campaign to keep the chemical on the market, arguing that humans who use products containing BPA are not exposed to significant amounts of it. 

THE CONSUMER PRODUCT SAFETY IMPROVEMENT ACT OF 2008

Problem Exercise: The Case of Alar - This is a popular problem from previous editions. 

The following discussion is meant to be as factually accurate as possible, although we have simplified occasionally where doing so does not materially distort the questions posed. Place yourself in the position of the EPA administrator and answer the questions raised by the text.

Alar is the trade name for daminozide, a plant growth regulator used to keep ripening apples on the tree and to keep apples firm and red during storage. Alar is considered a pesticide and was registered by EPA under FIFRA in 1963. When used on apples, some Alar penetrates the apple skin and remains after harvest and cannot be washed off. Alar decomposes over time into other chemicals including l,l-(unsymmetrical) dimethylhydrazine (UDMH), which is used as a component in rocket fuel. Alar contains about 1 percent UDMH and it hydrolyzes into additional UDMH, particularly when heated, as in the preparation of applesauce or apple juice, and when ingested in the bodies of mammals.

UDMH was first identified as a potential carcinogen in 1967 when lung tumors were found in mice exposed to UDMH. In 1973 further studies found tumors of the lung, liver, kidney, and blood vessels in mice exposed to UDMH. A study published in 1977 found that when mice were exposed to high levels of Alar, they developed a high incidence of the same type of tumors as the mice exposed to UDMH.

In 1980 EPA announced that it would conduct a special review of Alar, but later it abandoned the effort after meeting with representatives of Alar's manufacturer, the Uniroyal Company. After environmentalists sued EPA, the Agency reinstated the special review. In 1985 EPA staff proposed canceling the registration of Alar for use on food. This action would have had the effect of banning such use (following lengthy administrative hearings that could take a year or two). EPA's proposal was then reviewed by a science advisory panel that severely criticized the methodology used in the animal studies of Alar and UDMH, which had been performed prior to EPA's establishment of "good laboratory practice procedures" in 1979. Because these studies were flawed, EPA withdrew its cancellation proposal in January 1986 and instructed Uniroyal to conduct new studies using a better methodology. Because of publicity concerning the potential hazards of Alar, many apple growers announced that they would voluntarily stop using it.

In January 1989 results from the first year of Uniroyal's new two-year bioassay were reported to EPA. Although the study did not show any increased tumors in mice or rats exposed to Alar, it found tumors in those exposed to high levels of UDMH. Based on these data, the Natural Resources Defense Council prepared its own risk assessment that concluded that between 4,700 and 6,000 preschool children out of a population of 22 million (or roughly 1 in every 4,200), would get cancer from exposure to Alar in their first 6 years of life. After this information was publicized on "60 Minutes" on February 26,1989, a tremendous public outcry ensued.

Suppose that you are the EPA administrator. Your staff analyzed the NRDC report and the Uniroyal data and performed their own risk assessment. They find that of the 22 million children in the cohort examined by NRDC, 5.5 million are currently predicted to get cancer eventually from causes other than Alar. Children, of course, are not the only exposed population. Practically everyone eats apples. NRDC, however, has studied food consumption patterns as part of its analysis, relying largely on Department of Agriculture surveys. It found that fruits of all kinds comprise 34 percent of a preschooler's diet but only 20 percent of a mother's. Furthermore, the average preschooler consumes 6 times more fruit, overall, than an adult woman between 22 and 30, 7 times more apples and applesauce, 7 times more grape products, and 18 times more apple juice. Your staff calculates that the risk of cancer attributable to Alar is 9 in 1 million, 25 times lower than NRDC's estimate. They say the different calculations stem from several sources:

1.       Exposure Period. EPA has calculated exposure only for a period of 18 months, which it estimates is the likely length of the administrative hearings that would have to be held before Alar uses could be halted unless you determine that an "imminent hazard" exists; NRDC's figures are for exposure during the 6-year period when children are between birth and the age of six.

2.       Dose-Response. Quantitative risk assessment requires an estimate of how much harm will come from given doses. NRDC thinks Alar a more potent carcinogen than does your staff. Part of the difference comes from different models of cancer causation employed. NRDC used a "time-dependent multistage model" that takes into account the longer latency periods available when one is exposed to a carcinogen at a young age. EPA's model is time-independent and assumes that the risk from a given dose is the same whether one is 5 years old or 70 years old. NRDC also argues that children are more vulnerable to cancer risk for physiological reasons (and hence any exposure amount is more potent for them). For one thing, children weigh less than adults. Therefore, the weight of fruit and its accompanying chemicals relative to body weight is greater than for an adult, and such differences increase the risk of cancer under most current models of cancer causation. NRDC also claims that children are even more susceptible to cancer than adults because their cells are dividing rapidly as a natural part of the growth process and because their enzymatic detoxification systems are not as fully developed as an adult's. There is some evidence these latter claims are true, although there is no scientific consensus. Finally, estimates of potency are extrapolations from the discrete findings of animal experiments. EPA used the interim data from the Uniroyal study, which NRDC adjusted upward because Uniroyal is just one year into a two-year study. NRDC claims that doubling the duration of exposure can be expected to increase the cumulative incidence of tumors by a factor of 32 rather than the factor of 8 used by EPA. Thus, because a far greater proportion of the adverse effects are likely to appear in the second year than EPA had adjusted for, NRDC's estimate of the slope of the dose-response curve is higher, consistent with the findings of the pre-1985 studies that the science advisory board found methodologically flawed.

3.       Exposure. NRDC assumes that children consume more apples than EPA assumes. EPA's apple consumption estimates rely on a survey of 30,000 people conducted by the U.S. Department of Agriculture (USDA) in 1977-1978; NRDC is using the results of a USDA survey of 2,000 people conducted from 1985-1986 that shows a 30 percent jump in fruit consumption since the 1977-1978 data were compiled. EPA discounts the significance of the more recent survey because the sample size is small. Data from USDA's larger 1986-1987 study were not yet available. FIFRA gives EPA the authority to ban the use of Alar on apples if such use poses "unreasonable adverse effects" to human health or the environment. However, the statute permits the manufacturer of Alar to challenge such a decision during lengthy administrative hearings that sometimes last for years (see 7 U.S.C. §136d(b)). During these hearings the manufacturer can keep the product on the market unless you determine that an "imminent hazard" exists (see 7 U.S.C. §§136d(c)(1), 1361). Even then, the product may remain on the market during expedited hearings unless you declare that an "emergency" requires its immediate removal (see 7 U.S.C. §136d(c)(3)). EPA has been studying Alar for more than 17 years.

QUESTIONS

Question One. Several members of Congress have called your office demanding to know what you are going to do about Alar. Can you declare an emergency and order that Alar be removed from the market immediately? What would you tell the farmers who may lose an entire apple crop if you take such an action?

Question Two. As EPA administrator, how do you evaluate the differences between NRDC's estimates of risk and the estimates by your own staff? Recall the previous discussion about the impact of data uncertainties on risk assessments. What uncertainties does EPA face for each of the four elements of the Alar risk assessments? Which type of uncertainty affects each of them? Which differences between EPA's estimates and NRDC's estimates stem from the use of different models? Which stem from different assumptions? Whose models, assumptions, and estimates are more reasonable, NRDC's or EPA's?

Question Three. School districts are calling you, wondering whether they should pull apples off school cafeteria lines. Nationwide, purchases by schools account for nearly 25 percent of apple sales. What do you tell them? (As it happened, several school districts, including New York City's and Los Angeles', did pull apples for a time, then reinstated them.)

Question Four. In other contexts, EPA has suggested that it will aim to reduce excess cancer risk to the level of one in one million or lower. NRDC is pushing for an immediate ban on Alar; your staff has recommended proceeding with administrative hearings to cancel the use of Alar on apples while awaiting the completion of Uniroyal's study and the release of the larger USDA survey data on food consumption before making a final decision. What should you do? Does the law provide a clear answer concerning the proper course of action?

Question Five. On April 5, 1989, a full-page advertisement sponsored by the American Council on Science and Health, a nonprofit organization funded largely by industry contributions, appeared in several major newspapers. Declaring that "OUR FOOD SUPPLY IS SAFE!," the ad condemned "efforts of certain environmental groups to exaggerate risk and cause needless anxiety." In order to "set the record straight," the ads made the following four assertions: (1) "There is no scientific evidence that residues in food from regulated and approved use of pesticides have ever been the cause of illness or death in either adults or children." (2) Charges that pesticide residues cause cancer have "no scientific merit" because they "are based exclusively on studies that expose rats and mice to enormous doses of chemicals. Humans would have to eat tons of produce every day for many years to ingest similar amounts." (3) Pesticides have helped make our food supply "the safest and most plentiful in the world," and if Americans "go 'back to nature,' they will have to coexist with vermin and insects and the diseases they bring." (4) Citing AIDS, smoking, and drug abuse, the ad argued that "[i] is dangerous to focus on unsubstantiated claims when real risks to our health abound." N.Y. Times, April 5, 1989, at A11. Based on what you know abut risk assessment procedures, how would you evaluate the accuracy of each of these claims? Are they misleading in any respects? How would you respond to them?

A Postscript on the Alar Controversy

The denouement of the Alar controversy came quickly. Alar was removed from the apple market by its manufacturer, not because of any regulatory requirements imposed by EPA, but because of consumer pressure. The rapid decline in apple consumption that followed the "60 Minutes" report on February 26, 1989, led apple growers to pressure the manufacturer of Alar to remove it from the market.

Uniroyal, the sole manufacturer of Alar, agreed in June 1989 to halt voluntarily all domestic sales of Alar for food uses. EPA had issued a preliminary determination to cancel all food uses of Alar under FIFRA on May 24, 1989. 54 Fed. Reg. 22,558 (1989). But rather than force EPA to go through cancellation proceedings, Uniroyal formally requested a voluntary cancellation of all food use registrations of Alar on October 11, 1989. EPA issued a cancellation order that became effective on November 17, 1989. 54 Fed. Reg. 47,493 (1989).

Cass Sunstein and Timur Kuran have analyzed the Alar controversy as an example of an availability cascade, "through which expressed perceptions trigger chains of individual responses that make these perceptions appear increasingly plausible through their rising availability in public discourse. . . .Once Alar became identified as a threat to social welfare and, in particular, the lives of children, it became risky to urge restraint in interpreting the figures. No official, and certainly no politician, wants to develop a reputation for softness on matters vital to children's health.... If some doubters and dissenters shade their views or mute their voices in response to such reservations, the relative content of the publicly available information will obviously get distorted. Specifically, information pointing to danger will become ever more available, raising its plausibility in the eyes of the many people who use availability as a criterion for evaluating the validity of such information. In the process, moreover, the burden of justifying one's beliefs will shift increasingly toward dissenters. Thus, as the Alar cascade took off, the proponents of banning Alar could increasingly take for granted that their safety concerns would be treated with respect. For their part, the opponents had to carry an increasingly heavy burden of being suspected of insensitivity to human suffering." Kuran and Sunstein, Availability Cascades and Risk Regulation, 51 Stan. L. Rev. 683, 685, 701 (1999).

The Alar controversy left many apple growers bitter. See Egan, Apple Growers Bruised and Bitter After Alar Scare, N.Y. Times, July 9, 1991, at Al. In November 1990, a group of apple growers in Yakima County, Washington, filed a product defamation suit against NRDC, "60 Minutes," and NRDC's public relations firm. The lawsuit sought $200 million in damages for dissemination of "false, misleading, and scientifically unreliable statements about red apples." Ironically, because the apple industry was then enjoying record harvests and high prices, "[o]ne tricky point the growers will have to explain is why the apple business has done so well since Alar was taken off the market." Marshall, A Is for Apple, Alar, and . . . Alarmist?, 25 Science 20, 21 (1991).

CBS subsequently won summary judgment dismissing the growers' claims because the growers failed to produce evidence sufficient to create a triable issue of fact as to the falsity of the broadcast. Auvil v. CBS "60 Minutes," 836 F. Supp. 740 (E.D. Wash. 1993). This decision was affirmed on appeal in October 1995. Auvil v. CBS "60 Minutes," 67 F.3d 816 (9th Cir. 1995). The court explained that:

The growers' only challenge to the scientific studies is their claim that animal studies cannot be relied on to indicate cancer risks for humans. Because animal ? studies can be relied upon, their evidence that no studies have been conducted on the effects of daminozide on humans does not create a genuine issue for trial on the falsity of the broadcast's assertions regarding daminozide's carcinogenicity. [67 F.3d at 821.]

After the Alar incident, representatives of food producers petitioned their state governments to enact laws that would discourage public assaults on food. So-called veggie libel laws have been passed or are under consideration in over 20 states. Don't Diss Your Vegetables, or Your Beef Either, Salt Lake Tribune, June 8, 1997, at E4. Oprah Winfrey tangled with the Texas version of such food disparagement laws after a show of hers devoted to mad cow disease in which Oprah exclaimed, "It has just stopped me from eating another burger!" Babineck, Suit Gets to Meat of 'Veggie Libel": Feeders, Oprah Clash Over "Mad Cow" Show, Arizona Republic, June 17, 1997, at A11. In February 1998, she was eventually acquitted before a Texas judge and jury. On the food disparagement laws generally, see Bederman et al., Of Banana Bills and Veggie Hate Crimes: The Constitutionality of Agricultural Disparagement Statutes, 34 Harv. J. on Legis. 135 (1997).

Now that the Alar bioassay has been completed, EPA reportedly has estimated that Alar is only half as potent a carcinogen as the Agency had estimated in its 1989 risk assessment. 25 Science, at 20. If this report is accurate, EPA's risk assessment would indicate that Alar would pose an individual risk on the order of 4 in 1 million for an exposed person. Should this have changed the outcome of the Alar controversy? EPA's deputy administrator for pesticides and toxics says no. The new data do "not change EPA's earlier conclusion that, although Alar has some benefits, its dietary risks from historical exposures are unacceptably high." While "Alar appears less risky than [EPA] thought in 1989, that lower cancer potency estimate would not lead EPA to reverse [its] regulatory position." Kimm, Alar's Risks, 254 Science 1276 (1991).

The battle over how the Alar incident is to be interpreted rages on, with the critics of NRDC's "60 Minutes" approach apparently gaining momentum. In the course of an editorial decrying the Winfrey lawsuit, the Washington Post referred to the Alar episode as a "scare" that was "later determined to be unfounded." Hamburgers and Free Speech, Wash. Post, Jan. 5, 1998, at A18. This produced a letter from two prominent scientists, noting that "[i]n fact, the Alar episode was based on solid science," showing "unequivocally that Alar is carcinogenic." Rall and Landrigan, Of Apples and Alar, Wash. Post, Jan. 13, 1998, at A14, followed by a rebuttal from the director of research and development at Uniroyal. Lacadie, Of Apple and Alar (cont'd), Wash. Post, Feb. 4, 1998 at A18. When the Alar controversy's tenth anniversary rolled around in early 1999, the Environmental Working Group issued a report on pesticides in fruits and vegetables to take advantage of the timing. Bruce Ames countered the report's criticism of pesticide residues in fruits and vegetables by arguing that eating lots of fruits and vegetables is more important in preventing cancer than the increased cancer risk from "minuscule amounts" of residues. Blake, Pesticides: Are Kids Getting Too Much?-Ten Years After the Alar Scare, the Controversy Continues, Seattle Times, Feb. 24, 1999, at El.

Along one dimension of NRDC's disagreement with EPA, the NRDC position has clearly gained the upper hand. Responding to an intervening NAS study recommending that the physiological differences between children and adults counseled for greater protection of children, the Food Quality Protection Act of 1996 requires EPA to take into account children's "special susceptibility" in issuing pesticide tolerances, including building in a safety factor of 10. 21 U.S.C. §408(b)(2)(C). President Clinton also issued Executive Order 13,045, instructing all agencies to place a high priority on addressing risks that disproportionately affect children. 62 Fed. Reg. 19,885 (1997).

Had EPA decided instead that Alar posed an "imminent hazard" warranting immediate suspension of its registration, would its decision have been upheld in court? See National Coalition Against the Misuse of Pesticides v. EPA, 867 F.2d 636 (D.C. Cir. 1989).

 EPA Reduces TRI Reporting

One of the most successful parts of EPCRA has been section 313 which requires businesses meeting certain criteria to report releases of toxic chemicals.  In December 2006 EPA finalized a new rule which has the effect of reducing the number of companies providing detailed reports and thus the amounts of released chemicals reported as well as the circumstances surrounding their release.

EPA says the new rule will encourage some large producers of releases to lower their releases below the raised threshold.  Critics argue that the net result will be fewer companies reporting fewer releases with no reduction in the amount of chemicals actually released.  EPA argues that the rule change will reduce burdens on small business, but companies with 10 employees are fewer are already exempt from TRI reporting.

Click here for a Christian Science Monitor story on the rule change.