The Unsolvable Conundrum of “Maximum Residue Limits”
[W]hat the public is asked to accept as “safe” today may turn out tomorrow to be extremely dangerous.
—Rachel Carson
“I spoke with WHO management, and you are authorized to film the beginning of the JMPR [Joint Food and Agriculture Organization (FAO)/World Health Organization (WHO) Meeting on Pesticide Residues] experts’ session, but without sound.” As I was not entirely sure I understood the instructions from Angelika Tritscher, the secretary of the Joint FAO/WHO Expert Committee on Food Additives (JECFA) and of the JMPR at the WHO, I insisted, somewhat amused, “But I’m doing a documentary for television, not a written article. I need image and sound!”
“I know,” the UN organization representative replied, “but you know that work meetings for WHO and FAO committees take place in seclusion and are closed to all outside observers. It’s already a huge privilege to be able to film a few things, even if it’s without sound. Take it or leave it, because I can’t get anything more. I will also ask you not to reveal the identities of the experts before we have published the summary of their work, because as you know, their names must not be released until they have completed their assessment.”
“But my film is going to air over a year from now.”
“In that case, no problem. You can film the cards showing their names,” concluded Angela Tritscher.
Geneva, September 2009: An Exceptional Visit to the JMPR
It’s true that I was exceptionally lucky to penetrate the WHO walls with my camera on that September 2009 day when the JMPR experts were holding their annual meeting. The authorization came after three months of intense negotiations, punctuated by e-mail exchanges and a long telephone conversation with Angelika Tritscher, a German toxicologist who worked in the United States for a long time, and without whom my filming in Geneva would have been impossible. I understood intuitively that she knew of my work on Monsanto, but she never told me her thoughts on it, or if it had helped me gain exceptional authorization to attend the JMPR work meetings or, conversely, if it had been a handicap.
As we saw in the previous chapter, the JMPR was created in 1963, using the JECFA as a model. These two bodies are charged with supplying toxicological evaluations to the Codex Alimentarius Commission, an organization created in 1963 by WHO and FAO to write recommendations and guidelines as far as food safety is concerned. The opinions produced by the Codex Alimentarius Commission have no regulatory power, but they may be adopted by national governments to set their own health standards.
As its name indicates, the mission of the JMPR is to evaluate the toxicity of pesticides by establishing an acceptable daily intake (ADI) level, but also to define the authorized “maximum residue limits” (MRLs) for every agricultural product in use, whether in a pure or processed state.1 Setting ADIs is entrusted to experts chosen by WHO, while setting MRLs falls to those appointed by the FAO. Before examining how these experts are selected to make up the panels, we need to know exactly what these MRLs are, as they are supposed to serve as the last line of defense against the noxiousness of agricultural poisons. To this end, I will take the example of chlorpyrifos-methyl, which is part of the chlorpyrifos family, an organophosphate insecticide known for its neurotoxic properties and suspected of being an endocrine disruptor. Produced by Dow AgroSciences, it has been the subject of numerous scientific studies, as the PubMed database attests: on January 29, 2011, typing the word “chlorpyrifos” into its search engine gave 2,469 results, 1,032 with “chlorpyrifos toxicity,” and 139 with “chlorpyrifos neurotoxicity.”2
At the JMPR session that took place at WHO from September 16 to 25, 2009, chlorpyrifos-methyl was included among the five pesticides submitted for reevaluation. The LD50 of the substance, which is sold under the name Reldan, is 2,814 mg per kg of body weight (for mammals, orally), and its ADI is 0.01 mg/kg. According to a fact sheet published on the European Union’s website,3 this ADI was obtained following a neurotoxicity study carried out over two years on rats, then applied with a safety factor of one hundred.4
A Complex—and Far from Reassuring—Process
As we have seen, a poison’s ADI represents the maximum quantity consumers are supposed to be able to ingest daily, for their entire life, without becoming ill. The problem is that the poison could be used to treat a multitude of fruits, vegetables, or cereals. This is especially the case with chlorpyrifos-methyl, an insecticide used in the treatment of citrus fruit (lemons, mandarins, oranges, bergamot), all sorts of nuts (pecans, pistachios, coconuts, etc.), and other fruit (apples, pears, apricots, peaches, plums, berries, grapes, etc.).5 The question risk managers are facing is thus: How can we keep a consumer from reaching his or her ADI level of chlorpyrifos-methyl quite simply because he or she has the habit of inconsiderately(!) eating several foods treated with the pesticide?
To avoid this catastrophic scenario, the JMPR’s founders followed René Truhaut’s recommendations (see Chapter 12): they decided a MRL must be calculated for each agricultural product likely to have been sprayed with pesticides. MRLs, which are expressed in milligrams of pesticide per kilogram of foodstuffs, are set at the end of a complex—and far from reassuring—process. The first step in the assessment of a given pesticide consists in measuring the quantity of its residue (and potentially of its metabolites, or what is produced when it is broken down) left on each agricultural product after harvesting. Then, experts estimate consumers’ potential exposure, based on investigations aiming to determine which fruits, vegetables, and cereals—and in what quantity—are eaten every day, taking into account the fact that food habits vary from one country or continent to another. The result: millions of figures enabling the establishment of MRLs, food by food.
“In concrete terms, how do you proceed?” I asked the Dutch toxicologist Bernadette Ossendorp, who oversaw the FAO panel at the JMPR session in September 2009.
“First, we examine the data obtained from field tests where the pesticide has been applied on crops according to what are called ‘good agricultural practices’—that is, according to the method of use recommended by the manufacturer. This allows us to establish what quantity of pesticide residue is found on the treated food and to set a maximum limit.”
“Manufacturers,” Ossendorp explained. “And in doing this, they must respect very precise specifications: tests must be conducted on different types of agricultural products and, if possible, repeated over at least two seasons, so there is no bias in connection with climate conditions, for example.”
“How can you be sure of the quality of the data?” I then asked. “Because industrial history is overflowing with insufficient, even fixed, studies . . .”
Surprisingly, the question did not seem to shock her. “We require very detailed reports that present, for example, the analytical methods used,” she replied. “We also verify that the dose of the pesticide used corresponds correctly to what is recommended to farmers or that the spray took place at the right time. If the manufacturer sprays two months before the harvest, whereas agricultural practice would have this take place two weeks earlier, there will be a lower rate of residue than what is found in the real world. It does happen—we have questioned the suitability of certain data, and asked the manufacturer for explanations. If they do not convince us, we reject the data and the product cannot be evaluated, which can impede its placement on the market.”
“Except that, quite often, the products you evaluate are already on the market?”
“That’s true, but in the long run, it isn’t good for the manufacturer to have been ‘failed’ by the JMPR.”
“But if, in the real world, the farmer does not follow the levels recommended in the instructions for use, all this work doesn’t do much, does it?”
“That does not fall under our purview,” Ossendorp admitted, “but rather that of the public authorities who have to verify that farmers are respecting pesticide use standards.”
“After examining the data on residues, you have to evaluate potential consumer exposure. How do you go about that?”
“We estimate, for each agricultural product, the average consumption by inhabitants according to thirteen dietary models that correspond to the habits of five continents, with specificities such as vegetarianism etc. Let’s take the example of apples: in order to find out how many apples a French person eats on average every day, we take annual production in France, from which we deduct the amount exported, then we add the quantity imported. Then, we divide the result by the number of inhabitants. We do this for every agricultural product. This allows us, using typical menus, to evaluate the quantity of a given pesticide that every French person is likely to ingest every day.”
“That’s a lot of work! All that to avoid our getting sick from eating?”
“Yes,” the toxicologist replied. “But, you know, we, too, are consumers.”
“Let’s take the example of chlorpyrifos-methyl, which is used as an insecticide on numerous crops. What happens if a consumer reaches the ADI, because she has eaten a few too many treated fruits and vegetables?”
“Yes, I understand your question, but you must also know that the MRL we evaluate is much higher than your actual exposure. According to our monitoring programs, we know that not all the apples you eat have been treated with chlorpyrifos-methyl. It should be understood that our evaluation of consumption really corresponds to the worst theoretical case—that is, to a day in which everything you eat has been treated with the same pesticide. This is very unlikely to happen in the real world. Because, in general, you will have on your plate a mixture of potatoes that have been treated, but also carrots or lettuce that has not been. Thus the probability that you would take in a very high level of chlorpyrifos-methyl residue in one day is extremely low.”
“Of course,” I said, “but all that isn’t really reassuring . . .”
“Of course it is!” Ossendorp replied. “Don’t forget that the potential danger of a product has nothing to do with the risk you actually face. Like with salt: if you eat five kilos of salt, you will get sick, but even so, you wouldn’t say that salt is very toxic. As Paracelsus said, the dose makes the poison. But, okay, if you really want zero risk, you’re right, pesticides shouldn’t be used. But that is a political decision. As long as politicians keep saying that they have to be authorized because farmers need them to have plentiful harvests, it’s the best we can do.”6
“The Magician-Chemists of the Postindustrial Era”
I emerged from this long interview quite annoyed. It wasn’t that I found Bernadette Ossendorp unpleasant or that I thought she was, to put things plainly, “selling me a story.” On the contrary, I got the impression that she was being very sincere, even when she proffered arguments that were hopelessly weak: for example, how can someone dare compare pesticides, which are designed to kill, to table salt? I will note, incidentally, that while suicide by pesticide ingestion is unfortunately relatively common, in contrast, suicide by voluntary table salt poisoning is unheard of! Evidently, the comparison with good old-fashioned sodium chloride is used by all risk evaluation specialists, because I also heard it from Angelika Tritscher, the secretary of the JMPR and the JECFA at the WHO: “The fact that a chemical product or pesticide residue is found in a food does not mean that your health is threatened,” she explained. “Like for salt, the question is knowing at what exposure level the danger is revealed. The problem is that food is a very emotional thing. If we add salt to it, we can control the quantity, which is not the case with pesticides. It is this unknown that frightens people, because they feel as if they don’t know what’s on their plates.”
This reasoning shocked James Huff—the carcinogenicity specialist from the National Institute of Environmental Health Sciences (NIEHS) and former director of the International Agency for Research on Cancer (IARC) Monographs Programme—who had previously denounced industry’s influence on the WHO agency (see Chapter 10). “I’m surprised that Angelika Tritscher, whose scientific qualifications I am aware of, could use the chemical manufacturers’ favorite talking point,” he declared indignantly, when I met him one month later. “Salt is a natural substance, one of whose functions is enhancing taste; it is certainly preferable to use it in moderation, but from that to comparing it to pesticides, which are specially designed to have harmful effects on living organisms and which pollute our plates, unbeknownst to us! Honestly, they can’t be serious. But this is typical of the mental confusion experts in charge of the evaluation of food pollutants are known for—we’re asking them to carry out an impossible task, because deep down, they know very well that ADIs and MRLs are only artifacts and that the only way of truly protecting people is to simply ban a large number of extremely toxic products they persist in evaluating, for better or worse.”7
This viewpoint echoes German sociologist Ulrich Beck’s enlightening analyses in his book Risk Society, in which he makes a harsh indictment of the role played by scientists in the health disaster that characterizes “advanced modernity”: “As they are constituted—with their overspecified division of labor, their concentration on methodology and theory, their externally determined abstinence from practice—the sciences are entirely incapable of reacting adequately to civilizational risks, since they are prominently involved in the origin and grown of those very risks. Instead—sometimes with the clear conscience of ‘pure scientific method,’ sometimes with increasing pangs of guilt—the sciences become the legitimating patrons of a global industrial pollution and contamination of air, water, foodstuffs, etc., as well as the related generalized sickness and death of plants, animals and people.”8
Ulrich Beck devotes a number of pages to risk specialists, scientists whom he implies are “magicians” or “acceptable level jugglers.” Since scientists are never entirely clueless,
they have many words for it, many methods, many figures. A central term for “I don’t know either” is “acceptable level.” [ . . . ] Acceptable levels for ‘permissible’ traces of pollutants and toxins in the air, water and food have a meaning similar to that of the principle of efficiency for the distribution of wealth: they permit the emission of toxins and legitimate it to just that limited degree. Whoever limits pollution has also concurred in it. [ . . . ] Acceptable values may indeed prevent the very worst from happening, but they are at the same time “blank checks” to poison nature and mankind a bit. [ . . . ] Acceptable levels in this sense are the retreat lines of a civilization supplying itself in surplus with pollutants and toxic substances. The really rather obvious demand for non-poisoning is rejected as Utopian. [ . . . ] Acceptable values make possible a permanent ration of collective standardized poisoning. [ . . . ] Acceptable levels certainly fulfill the function of a symbolic detoxification. They are a sort of symbolic tranquilizer pill against the mounting news reports on toxins. They signal that someone is making an effort and paying attention.9
Beck concludes with an acerbic commentary on the acceptable limit creators, who are, in his eyes, “late industrial chemical magicians,” gifted “seers” with “the ability of the ‘third eye’”: “It ultimately comes down to how long poisoning will not be called poisoning and when it will begin to be called poisoning. [ . . . ] No matter how benevolently one looks at it, the whole affair remains a very complicated, verbose and number-intensive way of saying: we do not know either.”10
Industry Information Is “Confidential”
“I was able to examine the list of studies you supplied to Dow AgroSciences, the producer of chlorpyrifos-methyl. It’s very interesting, because they are all ‘unpublished’ and covered by ‘data protection.’ Is this always the case?”
Professor Angelo Moretto, an Italian neurotoxicologist overseeing the JMPR session in September 2009, raised his eyebrows at my question. To help him formulate his answer, I held out a sixty-six-page document, published in 2005 by the European Union, which lists the two hundred-odd studies conducted by the American producer on its insecticide.11 The document includes experiments done on animals to measure the product’s toxicity, but also field tests to evaluate residue levels on crops. For example, one measured “Residues of chlorpyrifos-methyl in tomatoes at harvest and processed fractions (canned tomatoes, juice and puree) following multiple applications of Reldan.”12 Another evaluated the “residues . . . in wine grapes at harvest following two applications of [Reldan].”13 All of these studies are described as “unpublished,” while an introductory paragraph stresses that the producer “has claimed data protection.” What’s more, some of these studies concern chlorpyrifos and not chlorpyrifos-methyl!
After taking a long look at the document, Angelo Moretto finally said, “Yes, that is quite possible. Studies supplied by industry to the JMPR or to national authorities are data that are protected by a confidentiality clause. But, if you consult the documents produced by the JMPR after evaluation meetings or by national authorities, you will find large summaries of the data.”
“Summaries, but not the raw data?”
“No, not the raw data, because they belong to the producer. So you have to trust the twenty or so JMPR experts—who come from all over the world and were chosen for their expertise, for the proper analysis and interpretation of the data.”
“And there is no reason not to trust you?”
“I should hope not!” the JMPR president answered with a forced smile.14
Here, we are touching upon one of the recurring critiques from nongovernmental organizations and civil society representatives with regard to the JMPR or the JECFA, but also to the European Food Safety Authority (EFSA) or any other public agency charged with the evaluation or management of chemical risks. This is because, without batting an eye, they all accept the dictum imposed by manufacturers, who demand their study data be covered by “trade secret.”
“The practice of keeping data secret only serves the commercial interests of chemical companies,” Erik Millstone, the British professor of science policy (see Chapter 12), told me. “It is completely contrary to the interests of consumers and public health. The WHO and regulatory agencies in no way deserve the public’s trust, as long as they don’t change their practices. Only the data concerning products’ manufacturing process can justify the confidentiality clause, because, in the context of competition, that’s what represents sensitive trade information. But all the toxicological data concerning the safety or toxicity of their products should be in the public domain.”15
I also brought up this delicate question with Angelika Tritscher, the JMPR and JECFA secretary, who, as part of her responsibilities, plays a central role in the organization of the evaluation process. One year before the committee meetings, she publicly announces which substances will be submitted for (re)evaluation, and that “governments, interested organizations, producers of these chemicals, and individuals are invited to submit data. . . . The submitted data may be published or unpublished.” In a text she put online in October 2008, in anticipation of the JMPR session in September 2009, she specified, “Unpublished confidential studies that are submitted will be safeguarded and will be used only for evaluation purposes by JMPR.”16
“Why is raw data not public?” I asked her.
“Honestly, I don’t really see what the public could do with all that data. It’s thousands of pages,” she replied.
“I’m not talking about the public in the broad sense, but, for example, for a consumer or environmental organization that wants to verify the toxicological data for a pesticide. Why is this data covered under trade secrets?”
“It is because of intellectual property protection. These are legal issues. Data are private and belong to the company transmitting them. We do not have the right to communicate them to a third party.”
“The fact that data are protected feeds the doubt surrounding their validity and undermines trust based on transparency . . .”
“Of course! I completely understand your remark, because people have the impression we have something to hide,” Angelika Tritscher acknowledged with surprising honesty.
“If we take the example of tobacco, the studies supplied by cigarette manufacturers were flawed, even manipulated or falsified, and the WHO had been misled by industry for years,” I said.
“I have no comment to make on that.”
“But is it true?”
“I have no comment to make on that, even more so because that happened before I came to the organization. I don’t know all of the details.”
“I know that it was a painful incident here,” I continued, “and led to a serious adjustment in 2000 . . .”17
“Yes, it is clearly was painful incident. But I am not sure that it can be compared to the situation with pesticides. Nevertheless, data protection is in fact the subject of intense debate here, and we will see where it takes us. You should ask the industry why it holds on to data confidentiality so tightly.”18
When Manufacturers Evade Embarrassing Questions
I hadn’t expected Angelika Tritscher’s advice to seek an interview with representatives from the pesticide industry. Naturally, as I was particularly interested in chlorpyrifos-methyl and, in a broader sense, in chlorpyrifos, one of the most controversial insecticides, I contacted the headquarters of Dow AgroSciences, one of its major worldwide producers, in Midland, Michigan, during one of the two long trips I took to the United States to carry out my investigation. On October 2, 2009, Jan Zurvalec, the head of public relations at the multinational company, forwarded my request to Sue Breach, her counterpart at the European branch, based in London. On October 13, she wrote me a very pleasant e-mail asking me to send her the questions I wanted to ask during the filmed interview: “I cannot guarantee direct participation in the program, but we will consider your request and questions with care and duly respond.”19
To be honest, I harbored no illusions, given that Monsanto had refused all dialogue during my previous investigation recounted in The World According to Monsanto; while Dow and Monsanto have always been competitors on the pesticides, plastics, and chemicals markets, they have always stuck together when it comes to defending the interests of the chemical industry. And on October 16, I received a negative response from Sue Breach: “As an organisation we are always open to media interaction regarding our products and activities, particularly in the area of health, safety and the environment. However, while we appreciate the offer to be interviewed, we regretfully have to decline on this occasion as following a review of some of your previous work, we have legitimate concerns as to whether our perspectives would be fairly represented in the proper context.” In conclusion, the representative for Dow AgroSciences offered to send me “written feedback” to my questions.
Then something very amusing happened. I decided to contact the organizations that represented the chemical industry in Europe, and very quickly, I observed that their directors were all consulting each other about my “case,” exchanging several e-mails, in which a certain Thomas Lyall, from the “European Office of Governmental Affairs” at Dow in Brussels, was actively participating. I realized all this when one of the directors sent me an e-mail and forgot to delete all the exchanges that had preceded it. Eventually, the European Chemical Industrial Council declined my interview offer. As did the European Crop Protection Association (ECPA), an official lobby for large agricultural firms also based in the Belgian capital. On January 28, 2010, I received an e-mail from Phil Newton, its director of public relations, to whom I had sent my questions concerning—on a very “basic” level—the “role played by industry in the process of pesticides evaluation” and “data confidentiality.”
“Dear Marie-Monique,” he wrote. “It is important to note that all plant protection products used in European agriculture are fully assessed and tested according to relevant EU rules (Directive 91/414/EEC, to be replaced by Regulation 1107/2009/EC). Independent peer reviews of all data are conducted by the European Food Safety Authority. . . . As such, they are the most appropriate source of information on this topic.” Finally, on February 1, 2010, Ana Riley, from Croplife International—which presents itself on its website as a “global federation representing the plant science industry” and is financed by the eight main pesticides manufacturers,20 kindly blew me off, openly citing the refusal of the ECPA, which she admitted contacting.
Which left France, namely the indispensable Crop Protection Industry Association (Union des industries de la protection des plantes, UIPP), which, as we saw in Chapter 2, brings together the “nineteen business marketing and commercializing plant protection products and agricultural services.” On January 28, 2010, its press office sent me a terse response to my request to interview its managing director: “We are informing you that we do not wish to proceed with your request to interview Jean-Charles Bocquet.” I then called the UIPP headquarters directly, where I came across a very understanding person who clearly had not written the e-mail, as she gave me her director’s cell phone number. A long telephone conversation with Jean-Charles Bocquet followed, at the start of which, he admitted: “I saw your film, The World According to Monsanto, which I found to be very committed work. I have no problem with that, because you have the right to be committed, but your commitment—generally, against Monsanto—translates, for me, into a very strong commitment against all businesses that produce pesticides, and since I represent them, as it happens, it would be difficult for me to grant your interview request. Also, there are a lot of errors in your film, and I don’t think it’s right that you didn’t try to meet with Monsanto representatives.”
“What?” I interrupted. “Listening to you, I’m not sure you really saw my film, because if that was the case, you would have noticed that I went to Saint Louis, which isn’t exactly next door, and the company refused to see me, after three months of negotiations. I’ve always wondered why. Were they afraid of the questions I was going to ask? And by the way, I would like to know what errors I committed in my film.”
“Uh . . . Well I assure you I saw it . . . It’s true that Monsanto has its own communications policy, but I am generally more open.”
“Are you afraid of the questions I might ask you?”
“Not at all! I have no doubt about the questions people ask me, but more about the manner in which my answers are used.”
“I work for Arte, which is a high-quality channel, and I can’t force you to say the opposite of what you’ve already said. It is up to you to defend your point of view! For example, how would you justify the fact that the raw data sent by pesticides producers to the JMPR or the EFSA are not public?”
“Because the public is not an expert! The day it is, it will have access. We aren’t going to just give information to organizations that aren’t responsible for plant product evaluation! When we know how much the numerous studies we have to do cost . . .”
“How much does a toxicological study on the carcinogenic effects of a pesticide cost?”
“Several hundred thousand euros,” the UIPP director replied.21
In a “written response” finally sent to me on February 24, 2010, Dow AgroSciences claimed that “based on industry surveys the research needed to identify a single new pest control active ingredient and support the government registrations needed to bring it to market takes about eight years to conduct and typically costs more than $180 million.” On the subject of chlorpyrifos, the multinational company specified “chlorpyrifos has been marketed since 1965 and is currently registered in about 100 countries worldwide where it is used on more than 50 crops. The aggregate cost of the studies required to support registrations and uses for all of these crops over 45 years would be hard to determine but would certainly be well in excess of $200 million.”22
Secrecy at the JMPR
Thanks to Angelika Tritscher’s help—she made sure I understood the scope of the JMPR experts’ task—I obtained exceptional authorization to shoot some footage in the WHO basements. All of the data submitted by producers for the evaluation of their products were stored there. “Placed end-to-end, it takes up several miles of shelves,” Marie Villemin, the director of the UN organization’s archives, explained to me. “Thankfully the JMPR and JECFA are encouraging manufacturers to send their data digitally now, because otherwise it wouldn’t be manageable anymore.” Before my eyes stretched rows of shelves, carefully labeled, pesticide by pesticide: there were seven enormous cartons just for “glyphosate,” the active ingredient in Monsanto’s Roundup. I opened a number of them randomly; they contained studies “on intergenerational and reproductive effects in rats” or field tests on potatoes and carrots. Every study contained hundreds of pages, with thousands of figures, spread out over columns or tables.
“Do the experts really examine all of this data?” I asked Angelika Tritscher.
“Yes, but obviously that doesn’t happen during the session, which only lasts nine or ten days. Preparation starts a year ahead of time. The raw data are entrusted to a limited group of experts who summarize it and present that to the panel at the session.”
“And who sends the raw data to the limited group in charge of preparation?”
“Producers or sometimes the JMPR offices, it depends.”
“So, it is possible that producers might know the names of certain experts on the panel beforehand?”
“Yes . . .”
“Yet, you told me that experts’ names were kept confidential until the publication of the JMPR report . . .”
“Yes, that is a WHO rule,” the German toxicologist admitted. “This is so we avoid experts being subjected to pressure before work meetings take place, either from industry, by a state that is particularly interested in the subject at hand, or by a consumer organization.”
“But this rule has exceptions, since producers might know the identity of certain experts before the sessions?”
“Yes, it’s the most practical way to send data . . .”
“How do you choose the experts?”
“We regularly publish what we call a ‘call for experts’ to make up the JMPR or JECFA panels. Any scientist can apply by sending a detailed curriculum vitae and a list of all of his or her publications. Our choice is based on the competency and expertise of the candidates, but we also have to ensure that all the continents are represented. You should know that the experts selected are not paid—WHO and FAO only cover their travel expenses.”
“I consulted your last call to experts. It specifies that they are ‘required to declare any potential conflict of interest through completion of a standard form developed by FAO and WHO.’”
As I listened to my interview later, I noted how very delicate my question had been, given that one of the biggest scandals WHO has ever experienced was looming ahead in the fall of 2009: it would involve undeclared conflicts of interest of the experts tasked with advising the organization on the “false pandemic” of the H1N1 flu. Three months before my visit to Geneva, on June 11, Margaret Chan, the director general of the WHO, had stated, in the sober tone the circumstances demanded, “The world is now at the start of the 2009 influenza pandemic,” thus triggering the chaos we are all familiar with. One year later, whereas hundreds of thousands of deaths had been anticipated, “swine flu” had created ten times fewer victims than the usual annual epidemic. But the matter was a godsend for the five main vaccine producers—Novartis, GaxoSmithKline, Sanofi-Pasteur, Baxter, and Roche—who split $6 billion in profits. It would be revealed that the “experts” who had advised the WHO were connected to the manufacturers who profited from this pathetic masquerade.23
I now better understand why Angelika Tritscher tensed up when I broached the subject of conflicts of interest: “Why are JMPR or JECFA experts’ conflicts of interest not published?” I asked her.
“It’s a WHO rule,” she replied, visibly annoyed. “You have to understand that the expert panels taking place here are not permanent, their composition changes with the cases in question. It would be a tremendous amount of work to publish on our Internet site all of the conflicts of interest of the experts we call upon.”
“But the EFSA, for example does [publish the conflict of interest of its experts].”
“That’s true, but its expert committees are permanent . . . I understand that this is an important question and, to be completely honest with you, we have discussions about this in our legal department to see how we can develop the system. This does not concern only the publication of conflicts of interest, but also the problem of scientific bias that can be found in certain studies . . .”
“To be frank, I find the functioning of the JMPR and the JECF severely lacking in transparency, because everything is secret here: study data, experts’ identities, their conflicts of interest, not to mention the sessions themselves, closed to all outside observers. Yet, it seems to me that the day before I arrived, pesticide producers were speaking before the panel?”
“Yes. We regularly call in manufacturers so they can answer questions concerning their products.”
“I can appreciate that it is important to clarify certain points, but why refuse observer status to nongovernmental organizations or academics who wish to obtain it?”
“Work sessions that take place at WHO are private by nature,” Tritscher replied. “It is not that they are closed, but to be able to participate, you have to be invited. We also think that seclusion allows the experts to express themselves more easily, free from any influence.”
“The Entire System of Food Contaminant Evaluation Should Be Revised”
“It is very difficult to make the system evolve,” Ned Groth, a twenty-year veteran biologist for Consumers International, told me with a smile. Yet, at sixty-five, this very respected, charismatic, and well-spoken scientist, who worked at the U.S. National Academy of Sciences, doesn’t seem like a troublemaker. Nevertheless his criticism of the “system” is unflinching. “You mustn’t forget,” he continued, “that the WHO and the FAO are two enormous bureaucracies that depend on the money given by UN member states, but also from private funds, the origins of which are unknown, by the way. They have no interest in clashing with their donors, who follow their activities very closely. And it is clear that the system of chemical evaluation was created by and for industry.”
“Do you really think that the experts do a detailed examination of the thousands of pages of data given by producers?” I asked.
“Of course not!” Ned Groth responded without hesitation. “That’s a well-known strategy for manufacturers: they send truckloads of data that no one can verify, unless you spend years on it! That’s why it is rare for experts who have no interest in the matter to volunteer for the thankless task, which, on top of everything, isn’t paid. If, by some very rare chance, a slightly more vigilant expert panel decides the data are questionable, it’s a good thing for industry, too, because it allows them to gain time. The JMPR will ask them to go back over their paper, which will take two years, and during that time, the standards will stay the same.”
“What is the expert profile?”
“As the work required is very complicated, the candidates are generally sent by governments that want their point of view to carry weight in the committees’ decisions. And those chosen are often people who have retired and have the time, but are not always up to date with the latest scientific advances. What scientist in the middle of his career would be willing to give up several weeks of his time for an activity where political and commercial interests take precedence over every other consideration? In general, JMPR and JECFA experts are rather mediocre scientists, because the good ones have other things to do.”
“Do you think the decisions they make are biased?”
“The problem,” Ned Groth sighed, “is that scientists who know enough about the toxicology of pesticides generally work or have worked for industry, as academics or private consultants. And, having rubbed elbows with them often enough, I know that they all come from the same school of thought. They go to the same conferences, speak the same language, and are all convinced that we could not live without pesticides.”
“Do you think experts’ conflicts of interest could truly influence JMPR or JECFA decisions?”
“Certainly! One of the most characteristic examples is that of the bovine growth hormone, which was evaluated by the JECFA in 1992 and 1998. The committee’s work was completely sealed by experts—including the panel’s recorder, Margaret Miller—who had worked for Monsanto, the hormone’s producer!”24
In The World According to Monsanto, I devoted two chapters to the exemplary case of the bovine growth hormone, or “rBGH.” I had indeed observed the efficiency of the “revolving door” practice: industry representatives take on senior-level posts in governmental or international agencies in order to defend the interests of their favorite employer, to whom they typically return once their mission is complete. The JECFA’s work on the transgenic hormone, meant to increase dairy production in cows, had been publicly lambasted by an investigative committee in the Canadian Senate, which had also revealed Monsanto’s attempted corruption of experts at Health Canada, the agency in charge of food safety. It was through working on this affair that I had, for the first time, heard of Erik Millstone, professor at the University of Sussex, who had shown in an article how Monsanto had manipulated the interpretation of its data concerning the effects of rBGH on the health of cows.25
“The entire system of food contaminant evaluation should be revised,” he told me during our meeting in Brighton. “They should get rid of the opacity that characterized both the work and selection of experts. It isn’t normal that I can’t take part in the JMPR panel because I don’t have a PhD in toxicology, even though I’ve worked on chemical food toxicology for thirty-five years! The lack of transparency is such that they refused to allow me to attend a JMPR session while I was preparing a report for a research organization at the European Union!26 Also, any person working for an industry directly concerned with committee decisions should not be able to hold a high position in their organization chart. This is unfortunately the case with Angelika Tritscher, who worked for Nestlé for several years.”
This wasn’t new information. Before meeting the German toxicologist, I noted that she had worked at a scientific research center for the Swiss firm, which is a heavy user of food additives, including aspartame (see Chapters 14 and 15). I also noted that she participated in a congress in January 2009 in Tucson, Arizona, organized by the International Life Sciences Institute (ILSI), the “scientific” organization financed by multinational chemical, food industry, and pharmaceutical companies.
“Why should my previous position at Nestlé keep me from working at WHO?” Tritscher was outraged when I questioned her on the subject. “Moreover, the in-house legal department carefully reviewed my CV and decided I had the appropriate background. I have nothing to hide! Know that before being hired by Nestlé, I applied for a position at Greenpeace, in Hamburg, but I wasn’t hired. So please don’t judge my motives—in life, not everything is black and white.”
“Of course,” I said, “but could you imagine a former member of Greenpeace in your current job?”
“If he has the necessary scientific qualifications, why not?” Angelika Tritscher replied. “As to my participation in the ILSI congress, it was decided upon with WHO management, who thought I should represent the organization at a round table devoted to risk evaluation. Where’s the problem with that?”
I must say that the German toxicologist convinced me of her good faith and her wish to move the system toward more transparency. I will also note that it was thanks to her that I was able to penetrate the very secretive walls of the JMPR, which is a telling sign. My impression was confirmed by Ned Groth’s comments: “I know Angelika very well, and I admire her. I don’t believe that she could be considered an industry ‘mole’ inside the JMPR or the JECFA, because she truly does care about public health, and she does everything she can so the committees fulfill their mission as best as possible. There are a lot of excellent scientists who have worked for industry, and a lot of bad scientists who work outside of industry. Beyond the people, it’s the system that doesn’t work, because it doesn’t protect consumers.”
“What the JMPR Does Is Not an Exact Science”
To prepare for my trip to Geneva, I reviewed a document published by the FAO and WHO entitled “Principles and Methods for the Risk Assessment of Chemicals in Food.” A sentence caught my attention: “JECFA and JMPR determine ADIs based on all the known facts at the time of the evaluation.”27 It echoed what René Truhaut had written in one of his retrospective articles, that ADIs are not fixed and unchangeable, and that any new information could lead to their revision.28 Then I asked myself: if ADIs and MRLs are not definitive values, since they depend on the experts’ level of knowledge at the time they set them, how can we claim that they protect us? My doubt as to the efficiency of these infamous standards was reinforced by a document from the EFSA concerning procymidone, a fungicide made by the Japanese company Sumitomo. It was on the list of pesticides submitted for “reevaluation” at the JMPR session in September 2009, for “doubts expressed by the European Union.” Actually, the European Union had banned its use in 2008, because it is an extremely strong endocrine disruptor (see Chapter 19), which led the EFSA to reduce its ADI and MRL. While the poison is no longer used in Europe, it is still present in a number of countries that export agricultural products to the EU, hence the necessity of maintaining standards to be used in (potential) inspections.
In the 2009 document, the EFSA explains that it decided to lower procymidone’s ADI from 0.025 to 0.0028 mg/kg, and consequently “proposes to change the . . . MRLs in order to reduce the acute and/or consumer exposure to a level where no negative consumer health effects are expected.”29 Before analyzing this very troubling passage, it’s worth noting that procymidone is used in the cultivation of some forty fruits and vegetables, including pears, apricots, peaches, plums, table grapes, grapevines, strawberries, kiwis, tomatoes, peppers, eggplants, cucumbers, zucchini, melons, lettuce, garlic, onions, etc. If the substance was banned in Europe after more than twenty years of use, it was due to a “multigenerational” study showing that male descendants of rats who had been exposed to 12.5 mg/kg during gestation presented “reduced anogenital distance, hypospadias,30 testicular atrophy and undescended testes.” At a dose five times lower, “increased weight of the testes and decreased weight of the prostate, epididymis31 and seminal vesicles” were observed. All of this information is revealed in the EFSA document, which specifies that, to establish the new ADI, the experts applied a safety factor of one thousand in relation to the no-observed-adverse-effect level (NOAEL).
I showed the EFSA opinion to Angelo Moretto, the JMPR president, when I met him in Geneva in September 2009: “I have here an EFSA opinion, published January 21, 2009, which concerns procymidone, a fungicide you are set to reevaluate.”
“Yes, the European Union expressed doubts about the limits we set,” he confirmed, before attentively examining the document I held out to him.
“Did you read the sentence I underlined? It says: ‘The EFSA proposes to change the MRLs in order to reduce the acute and/or consumer exposure to a level where no negative consumer health effects are expected.’ Does this mean that the ADIs or MRLs set by the EFSA or the JMPR are never definitive?”
“Yes, in life nothing is ever definitive, even science,” Moretto eventually replied, after an long, embarrassed silence. “So, if there are new data obliging us to change our previous decisions, we do.”32
I posed the same question to Angelika Tritscher, who also took a long pause before answering: “I really don’t like the way that phrase is formulated,” she commented, “because it gives the impression that the previous values were not protecting anyone at all and that consumers were in danger. Which isn’t true! People must understand the difference between a chemical’s potential danger and the actual risk a consumer faces, because everything depends on the level of exposure. Don’t forget that there are significant safety factors . . .”
“Of course, I understand the difference between danger and risk. But let’s take the example of lindane, an organochlorine insecticide that’s been on the market since 1938. It is a powerful neurotoxin that was classified in 1987 by IARC as possibly carcinogenic to humans and is considered a persistent organic pollutant. It was definitively banned in Europe in 2006. In 1977, the JMPR set an ADI of 0.001 mg per kilo. This standard was completely deceptive, then, even more so because lindane has the ability to accumulate in organisms.”
“The problem is that at the time the JMPR did its evaluation, no one was talking about persistent organic pollutants yet,” Angelika Tritscher replied. “If you mean to say that what the JMPR does is not an exact science, then I agree with you. Its decisions are based on the scientific knowledge available at the time it carries out its evaluation.”
“Excuse me for being a bit trivial, but I find that this whole process seems a bit shoddy.”
“The word you just used is very offensive to the experts, who really do everything they can to put out the best scientific judgment,” the German toxicologist shot back with a disapproving look.
My remark was certainly a bit harsh, even if it conveyed my sentiments exactly, which were, once again, confirmed by reading Risk Society. Save one nuance: contrary to Ulrich Beck, who vilifies the “acceptable level magicians,” I think that the main responsibility of the health disaster we are experiencing falls upon politicians, because they are the “risk managers” who we expect will see the big picture and protect our long-term health accordingly. As for the rest, I agree with the German sociologist: “It is not that one could not know how the toxic rations affect people individually or in total. One does not want to know it. [ . . . ] Even the already published statistics on such things as diseases or dying forests apparently do not appear eloquent enough to the acceptable level magicians. We are concerned, then, with a permanent large-scale experiment, requiring the involuntary human subjects to report on the accumulating symptoms of toxicity among themselves, with a reversed and elevated burden of proof. Their arguments need not be heeded, because, after all, there are acceptable levels that were met!”33
January 2010: An Edifying Visit to the EFSA
“I have here an EFSA opinion announcing a decrease in the ADI and MRL for procymidone, because of worries regarding consumer health. Does this mean that the previous ADI, which we thought was protecting us, was in fact not protecting us?” Predictably enough, I posed my perennial question to Herman Fontier, the head of the pesticides unit at the EFSA. There was a palpable silence in the Belgian toxicologist’s office. He shot several desperate looks at the three members of the PR team seated behind me. As it happens, they were very nice, and scrupulously recorded the four interviews I completed at the EFSA, on January 19, 2010, in Parma. If they were to listen to their recordings, they would easily see that I am transcribing my interview subject’s very muddled answer verbatim.
“It was not protecting for that . . . It did not have the same . . . It did not offer the same protection. Once again, there are safety values that are applied, a value of one hundred in relation to the no-effect level, so there are safeties that are inserted a bit all over in the system. So it is very improbable that the ADI that was set before had health effects . . .”
At first, I was amused by the European civil servant’s uneasiness, but then I became quite sad, faced with the extreme fragility of “acceptable level magicians” who are forced to walk a rope so tight that it could snap at the slightest snag: “If I gave you an apple with procymidone and chlorpyrifos residues, would you eat it?” I asked.
“That depends on the residue levels. If they are in compliance with legislation, with a pesticide content below the MRL, yes, I would eat it,” he replied, obviously relieved by this new question.
“Even if you know that, in three years, the MRL will be reduced because there will be new data?”
“Yes. We never know what the future has in store for us, but I am confident in the work we do. Absolutely!”34
It was precisely in order to “restore and maintain confidence in the EU food supply”35 that the EFSA was created in January 2002, “following a series of food crises in the late 1990s,” as its website explains. And it should be noted that in this domain, the EFSA’s task is immense. According to a Eurobarometer survey published in February 2006, “40% of people . . . think that their health could be damaged by the food they eat or by other consumer goods. . . . [and] the association of food with health is only made by one person in five.”36 At the top of the list of “external factors” considered particularly “dangerous” by Europeans are “pesticide residues” (71 percent), followed by “residues in meats, like antibiotics and hormones” (68 percent). Finally, the last revelation from the survey: “54% think that their health concerns are taken seriously by the EU. . . . 47% of citizens think that when deciding on priorities, authorities would favor the economic interests of producers over the health of consumers.”
Located in Parma, Italy, the EFSA is responsible for evaluating risks connected with the use of chemical products in the food chain. Lacking any regulatory power, it settles for producing “scientific opinions and advice” to “support the European Commission, European Parliament and EU Member States in taking effective and timely risk management decisions.” To understand the EFSA’s function, it should be viewed as part of the European system of regulation of “plant protection products,” governed by directive 91/414 of July 17, 1991. This directive requires that every pesticide be found on a “positive list” of authorized products, the famous “Annex I,” before it can be used legally. To secure this listing, the producer must submit a marketing authorization request to one of the Member States of the European Union, considered the “rapporteur Member State,” which is tasked with gathering and evaluating the toxicological and ecotoxicological studies on the active substance supplied by the producer. To this end, it calls upon the expertise of the EFSA, which intervenes on two levels.
Firstly, the EFSA gives an opinion on the classification of the molecule as to its potentially carcinogenic, mutagenic, and reprotoxic effects. Adding to the confusion of an already extremely complex subject, the European Union’s classification of carcinogenic substances is not the same as that of IARC (see Chapter 10). Category 1 corresponds to the UN agency’s group 1, “known to be carcinogenic to humans.” Category 2 corresponds to group 2A (“probably carcinogenic to humans”); and category 3 corresponds to group 2B (“possibly carcinogenic”).37 The same principle is used for mutagenic and reprotoxic substances. Secondly, since September 1, 2008, the EFSA has been tasked with proposing ADIs and MRLs for every pesticide submitted for evaluation, which are then enacted by the European Union and are from then on applied/shared by its twenty-seven Member States.
Ultimately, it is the rapporteur Member State that grants the first marketing authorization for a pesticide. Valid for ten years and renewable, it is generally adopted by the other European Member States, following the principle of “mutual recognition,” even if each country maintains the ability to “provisionally limit or ban the movement of a product on its territory.” By virtue of regulation 1107/2009, which replaced directive 91/414 as of June 14, 2011, the European Commission is now able to “adopt emergency measures in order to restrict or prohibit the use and/or sale of a plant protection product if it is likely to constitute a serious risk to human or animal health or the environment, and that such a risk cannot be contained by the Member State(s) concerned.”
“How many active substances in pesticides38 are currently authorized in Europe?” I asked the head of the EFSA pesticides unit in January 2010.
“You should know that in the 1990s, there were nearly 1,000,” he explained. “But today, there are only 300. The European Union led a vast revision program, and a lot of the chemicals did not survive, namely because manufacturers did not defend them, by deciding not to send the data requested of them. In certain cases, the dossier submitted was not complete and inclusion of the products in the new approved list was denied.”
“That means that 700 chemicals have been recently banned?”
“Yes, the revision program ended in 2008.”
“Does the EFSA take into account the JMPR’s work to set ADIs and MRLs?”
“Of course, we follow the JMPR’s recommendations very closely. We don’t always arrive at the same conclusions, because we sometimes have new studies at our disposal that the JMPR did not have when the evaluation was made. Generally, when there is a difference, our ADI is lower.”
“In that case, for the consumer, it would be better to have the EFSA’s ADI than the JMPR’s?”
“Of course, we think that the EFSA’s ADI is the one you should follow!”
Greenpeace’s Criticisms of the New European Toxicological Standards
Since September 1, 2008, MRLs have been set by the European Commission, which has led a vast program of harmonization of the existing standards in the twenty-seven Member States of the European Union. In fact, until that date, each country set its own acceptable levels for each agricultural product (vegetables, meats, fruit, milk, eggs, grains, spices, tea, coffee, etc.) and there were close to 170,000 different MRLs for the whole of the European territory! Which was a real headache for the European Commission, who hoped to streamline the process by aligning all the European Union countries with the same standards.
“It was a really good idea, because it allowed all European consumers to be assured the same level of protection,” said Manfred Krautter, a chemist who worked for eighteen years for the German branch of Greenpeace, on October 5, 2009, in Hamburg. “Unfortunately, instead of choosing the smallest common denominator, the Commission generally kept the highest MRLs. For Germany and Austria, for example, who had the most ambitious standards, harmonization brought an increase in the authorized residue levels up to 1,000 times higher for 65 percent of pesticides used.”39
In a report published in March 2008, Greenpeace and Friends of the Earth stressed that “for apples, pears and table grapes, 10% of the set acceptable levels are potentially dangerous for children,” who are major consumers of these fruits. As we saw in Chapter 12, toxicological standards are expressed in the quantity of the substance in question in relation to body weight. If an adult consumes quantity X of pesticides residues, he or she will experience fewer effects than a child will. In other words: a child weighing 12 kilograms who eats two apples and a bunch of grapes runs a proportionally higher risk than an adult who weighs 60 kilograms. In their report, the ecological organizations note that “a child who weighs 16.5 kilograms reaches the acceptable level of procymidone by eating just 20 grams of grapes, and that of methomyl (an insecticide) with 40 grams of apples or 50 grams of plums.”40
“How do you explain that the harmonization led to an increase of numerous MRLs rather than a decrease?” I asked Herman Fontier.
I found his answer rather unconvincing: “First, I’d like to note that the EFSA did away with a certain number of national MRLs that it considered problematic,” began the head of the EFSA pesticides unit. “Sometimes, there are indeed differences from one country to another. For example, in State A, the MRL for an agricultural product was, let’s say, 1 mg/kg, and in State B, 2 mg/kg. We check whether 2 mg/kg poses any health problems, and if this isn’t the case, we decide to take this MRL as a reference, so as to allow country B to continue cultivating the product with the necessary pesticide amount, because obviously agronomic and plant health conditions are not as favorable as in country A. But it should be noted that, in State A, they continue to use the minimum efficient amount, allowing them to stay below 1 mg/kg. This can seem paradoxical, but the increase resulting from harmonization does not lead to a rise in consumer exposure; on the contrary, the fact that we have eliminated certain MRLs has actually increased their safety.”41
This is what is called “having it both ways.” Because, through the simple game of commerce, products from country B arrive in country A, and they contain twice as much residue as those grown at home. So, to claim that an increase in the MRL does not lead to a rise in consumer risk is, at the least, a half-truth and, incidentally, completely contrary to the very principle of acceptable levels. “The increase in a certain number of MRLs means they can embellish the big picture in Europe,” Manfred Krautter, the Greenpeace chemist, explained to me, “because the higher the standards are raised, the less chance there is to exceed them! This is what we saw with the publication of the first annual EFSA report on pesticides residues, in which they prided themselves on an observed decrease in the number of times standards were exceeded.”
Published on June 10, 2009, the report consisted of a summary of observations made in the twenty-seven Member States of the European Union. In total, 74,305 samples were taken from 350 kinds of food: 354 different pesticides were detected in fruits and vegetables, and 72 in grains. MRLs were exceeded by one or more pesticides in 3.99 percent of the samples, and 26.2 percent of samples contained residues of at least two pesticides (and 1 percent contained more than eight different pesticides). As the report’s authors underline, “the percentage of samples of fruit, vegetables and cereals with multiple residues has increased over the years from 15.4% in 1997 to 27.7% in 2006. In 2007, the percentage of samples with multiple residues slightly decreased.”42
On paper, these results seem more or less reassuring, but it’s important to note that the use of a European average hides huge disparities from one country to another.43 In fact, the number of pesticides sought out varies from 709 in Germany, which is by far the leader, to 14 in Bulgaria (265 in France and 322 in Italy). The number of pesticides detected also varies considerably: 287 in Germany and 5 in Hungary (122 in France and in Spain). Finally, the number of samples analyzed totaled more than 16,000 in Germany, but only a few hundred for Malta or Luxembourg (4,000 for France). “The problem,” Manfred Krautter explained, “is that detection of pesticides residues is very expensive, and a number of European countries are not equipped to properly carry out the task. If it was being honest, the EFSA should have specified that the figures they were putting forward fall very short of reality.”
In October 2009, I visited the best German laboratory for analyzing pesticides residues and veterinary products, in Stuttgart. Thanks to ultramodern equipment using chromatography and mass spectrometry, the public center can detect more than a thousand molecules (pesticides and their metabolites). “We are one of the rare European labs to have access to this equipment,” Eberhard Schüle, its director, explained to me. “And, on average, 5 percent of the food we regularly analyze at the German authorities’ request exceeds the standards in effect.”
“Do you eat organic food?” I asked, surprising the lab director.
“I could give a personal answer to that question, but as a representative of a public establishment, I prefer to abstain,” he replied.44
Meanwhile, if several positive clues (I will come back to this in the last chapter of this book) indicate that Europe is setting off on the right track, at the time of writing (in 2010) we still have a long way to go. Going through the EFSA report, I discovered that, among the twelve pesticides that were the most often detected in the samples, two were classified as or suspected of being reprotoxins, one neurotoxin (chlorpyrifos), five carcinogens, and two endocrine disruptors (including procymidone).
“Can carcinogenic pesticides still be found on the market?” I asked Herman Fontier.
“Yes, there are still a few,” the EFSA pesticides unit head admitted. “But that will change with the new European regulation 1107/2009, which soon will replace directive 91/414. Because from now on, all category 1 substances classified as mutagenic, carcinogenic, or toxic for reproduction, or suspected of disrupting the endocrine system, will have to be taken off the market.”45
This is, indeed, good news. However, the EFSA still needs to base its chemical evaluations on high-quality studies, or else the pressure exerted by manufacturers must not completely skew the process. This unfortunately is too often the case, as shown in a sadly exemplary way, and in a field completely different from pesticides, by the unbelievable controversy surrounding an infamous artificial sweetener—aspartame.