THE ROLE OF SCIENCE AND
"OTHER FACTORS" IN CODEX DECISIONS
A DISCUSSION PAPER BY CONSUMERS INTERNATIONAL
Prepared for the
CODEX COMMITTEE ON GENERAL PRINCIPLES
Paris, France, 7-11 September, 1998
AGENDA ITEM 8
REVIEW OF THE STATEMENTS OF PRINCIPLE ON THE ROLE OF SCIENCE AND THE
EXTENT TO WHICH OTHER FACTORS ARE TAKEN INTO ACCOUNT:
APPLICATION IN THE CASE OF BST AND PST
TABLE OF CONTENTS
Other Factors Related to Health Protection of Consumers
Necessary Value Judgments Embedded in Science
Value Judgments in Risk Assessments
Value Judgments in Risk Management
Scientific Uncertainty and the Precautionary Principle
Other Factors Related to Promotion of Fair Practices in Food Trade,
Including Other Factors Related to Food Labelling
Purposes of Labelling
Food Labelling as a Risk-Management Tool
Food Labelling as Risk Communication
At What Points in the Risk Analysis Process Should "Other Factors"
Appropriately Be Considered?
ANNEX 1. A case Study: Application of General Principles Concerning
the Role of Science and Non-Scientific "Other Factors" to the Case
of Bovine Somatotrophin (bST)
Other Factors in the bST Risk Analysis Process
(1) Unusual Nature of bST Risks
(2) Structure of the Risk Assessment
Scope of Risks Considered
Depth of Analysis
(3) Composition of the Expert Body
(4) Treatment of Scientific Uncertainty
(5) Determination of "Significant" Risks
(7) Animal Welfare Issues
(8) Consumer Preference
Response to Specific Points in CX/GP 98/10
Consumers International has prepared this paper to address the role of science and the extent to which "other factors" should be taken into account in Codex decisions. Our primary emphasis is on the general principles that should apply to "other factors" within the risk analysis process; we also have appended a Case Study, applying the general approaches we discuss to the specific case of BST. We believe general policy on the use of "other factors" should be developed and then applied by Codex to the BST decision, and not the reverse.
It is evident to Consumers International that many "other factors" are now, and inevitably must be, part of risk analysis on food safety issues. Other factors arise and must be addressed in all phases of risk analysisrisk assessment, risk management and risk communicationand at all levels of the processinternational (Codex) decisions, decisions by national authorities, and decisions by consumers in the marketplace. In all cases, science is a necessary basis for decisions, but never a sufficient basis in itself. Furthermore, science itself is not value-free, and even analyses and decisions made by scientific bodies typically are based on both scientific and non-scientific considerations.
Consumers International believes that a policy on "other factors" in Codex decisions is urgently needed, to promote clarity and greater transparency of decisions. In particular, it is vital to improve the ability to distinguish between scientific and non-scientific portions of the basis for decisions. We hope this process now under way at CCGP will result in a clear policy that systematically identifies "other factors" that are appropriately addressed in Codex decisions, and provides guidance as to how they should be dealt with by Codex committees.
"Other factors" that are inescapably part of Codex decisions intended to protect the health of consumers include a variety of subjective value judgments and social choices in the application of risk analysis. Among the most important are how to treat scientific uncertainty, and perceptions as to which risks are "significant." Other factors include economic concerns, such as the feasibility of risk-management options, and the benefits of the activity or substance that poses the risks. Ethical issues, such as the rights and responsibilities of all the parties involved in the risk management process, also enter the picture.
In Codex decisions with respect to food labelling, additional "other factors" may apply. In particular, consumers need for and expressed desire for information on a food issue must be acknowledged and given weight. Consumer preference, per se, is not a basis for Codex to prohibit or limit the use of a food substance or technology. But recognition that consumers do have preferences based on many factors that cannot be considered by Codex provides a basis for labelling, so that consumers can exercise their preferences in the marketplace and manage their own risks.
With respect to the specific case of BST, discussion is constrained somewhat by the lack of availability to date of the full scientific report from JECFAs recent assessment of BST risks. However, based on extensive experience with risk assessment, with expert bodies and with the BST issue, Consumers International presents a general analysis, in Annex 1 of this paper, of some factors other than science that we believe are almost certainly part of the Codex risk analysis process on BST. In our view, such aspects as how to deal with the major uncertainties in the scientific data on BST risks, and judgments of whether the risks are "significant," are highly subjective, and likely to be based on "other factors," as well as on scientific evidence. We identify many "other factors" that appear likely to be part of decisions at some point in the Codex risk analysis on BST, and suggest ways to clarify what those factors are, and determine whether they have played a transparent and appropriate role in the process.
Consumers International recommends that Codex decisions on BST be held in abeyance until the following two conditions have been met:
(a) the full JECFA report on BST has been published and critically reviewed by all interested parties, and
(b) the Committee on General Principles has promulgated a policy on the role of "other factors" in Codex decisions, which can be applied to the BST case.
In Consumers Internationals judgment, these two conditions are essential for any final decision on BST by Codex to be perceived as valid by consumer organisations.
Consumers International also recommends that the Codex Alimentarius Commission refer the matter of BST to the Committee on Food Labelling, as well as to the Committee on Residues of Veterinary Drugs in Foods. If BST use is permitted, labelling should be provided to allow consumers to make informed choices.
Consumers International is pleased to comment on this issue, which is of great interest to many of our members, in two respects. At a general level, the question of "other factors" in addition to science that are legitimately considered in risk analysis potentially affects the way all Codex decisions are made, and is thus fundamentally important. And on a specific level, with respect to the BST case, consumer organisations have long raised concerns that include, but go beyond, scientific issues, which are summarised in a paper, "Hormones in Milk Production: The Case Against" (Consumers International, October 1997; available at http://www.consumersinternational.org). The BST case is thus a suitable example for examining how Codex committees should consider "other legitimate factors" that are inescapably part of food safety policy choices.
In approaching this discussion at the Thirteenth Session of the Codex Committee on General Principles, Consumers International believes that the development of general principles on "other factors" and their role in the risk analysis process should be the central objective. In our judgment, insights on how to approach the BST decision should arise from a set of general principles, and not the reverse. We therefore have addressed these issues primarily at a general level. But we have also appended a case study in which the general concepts developed here are applied to risk analysis on BST.
Consumers International agrees with and supports the four "Statements of Principle," including both the mandate that Codex standards be based on sound science and the recognition, in the second Statement, that factors other than science are legitimately part of the protection of consumer health and the promotion of fair practices in food trade. We have identified two sets of critical questions for this discussion:
(1) Which "other factors" are in fact "relevant for the health protection of consumers and the promotion of fair practices in food trade?" What factors are legitimately weighted in Codex decisions, and what factors are outside the scope of risk analysis as conducted by Codex? What such other factors are currently being considered in Codex decisions, and how can they be better delineated as separate from the scientific basis for decisions? Are there "other factors" that are not currently part of Codex decisions, but should be?
(2) How and where in the risk analysis process do these "other factors" come into play?
Consumers International believes it is essential to develop a clear, explicit conceptual framework that indicates where in the multi-layered risk analysis process such "other factors" are appropriately weighed. Risk analysis on food safety issues encompasses risk assessment, risk management and risk communication, and includes the international decisions made by Codex, decisions by national and regional governmental food-safety authorities, and consumer decisions in the marketplace. At Codex and at the national level, some decisions establish standards for the health protection of consumers, while other decisions, especially those on labelling matters, may be more concerned with trade practices and consumer information. Not all "other factors" are appropriate to consider in all sectors of this decision-making matrix.
Consumers International believes that these questions are an important part of the larger discussion, ongoing within Codex, of the risk analysis process. The Consultation on Risk Management, held in January 1997, recommended that Codex should develop a policy on "other factors," in the context of elaborating principles of risk analysis. Consumers Inter- national endorses that recommendation, and notes that promulgation of a clear policy on "other factors" will also greatly facilitate risk communication, which was the subject of a February 1998 Consultation (the report from which should be issued later this year).
Consumers International offers this contribution to framing an approach to the needed "other factors" policy, which we believe will ultimately strengthen Codex decisions and enhance their acceptance by the public. We strongly support both objectives, and we hope to play an active role both in the discussions at this meeting, which we hope will advance the issue, and in the ongoing effort to define policy on "other factors" as this issue proceeds within the Codex system.
Consumers International welcomes the publication of CX/GP 89/10, the Secretariats background paper on "other factors." That document both addresses general principles concerning factors other than science in risk analysis, and applies some of the principles it discusses to the specifics of the BST decision. Consumers International will address the general issues here, and respond to the analysis of BST issues in the annexed case study.
CX/GP 98/10 divides the "other factors" issue conceptually into "Aspects Related to Health Protection" and those related to "Fair Practices in Food Trade." For the sake of consistency, Consumers International will adopt the same division of topics. However, we note that labelling decisions may be concerned with both health protection of consumers and promotion of fair trade practices. In these comments, we will address labelling issues under the latter heading.
Consumers International also notes, with disappointment, that CX/GP 98/10 is largely silent on the topic of "other factors" to be considered in labelling decisions. Consumers International believes labelling issues are a critical component of the discussion of "other factors," as suggested in the wording of the third "Statement of Principle." We are also disappointed that, while the section of CX/GP 98/10 on "Fair Practices in Food Trade" addresses the economic interests of member countries, it makes no attempt to identify consumer interests. Labelling plays a central role in informing consumers and enabling them to consider "other factors" that may be outside the scope of formal risk analysis when making personal food-buying decisions, which can be a significant aspect of risk management. We regard CX/GP 98/10 as incomplete in this respect, and believe Codex must address the role of other factors in labelling decisions more broadly.
Consumers International agrees with the conclusion, stated in CX/GP 98/10, that certain other factors are routinely now considered in the risk analysis process, and in at least some Codex decisions. Among the non-scientific factors identified by the Secretariat that we agree are presently part of food safety decisions are the feasibility of achieving limits set for the protection of health; the feasibility of analytical measurement and enforcement of standards; current agricultural, veterinary and manufacturing practices; and economic impacts on member nations of Codex standards. Our comments will focus primarily on certain other factors, some of which are also mentioned in CX/GP 98/10, that we believe are legitimate and important "other factors" in food safety risk analysis.
Consumers International respectfully suggests that there are many additional factors other than science that are now considered, explicitly or implicitly, at the various stages of risk analysis processes and in many Codex decisions, beyond those itemised in CX/GP 98/10. We regard the paper as a good starting point, but believe it is essential to take a more systematic approach. The many additional "other factors" involved need to be identified as clearly as possible, and policy decisions should be made to define which factors are and are not appropriately considered in food-safety risk analysis. For those determined to be legitimate, the issue of where in the overall process of risk analysis they can best be considered also must be addressed. But the first step is to reach agreement as to which factors are on the list.
Other Factors Related to Health Protection of Consumers
The World Health Organization (WHO) defines health as "a state of complete physical, mental and social well-being, and not merely the absence of disease or infirmity." When assessing how to consider "other factors relevant for the health protection of consumers" at various points in the risk analysis process, Consumers International believes it is appropriate to have in mind a similarly broad definition of health, which is more than the simple absence of hazards.
Consumers International has identified several factors other than strictly scientific analysis that we believe are typically part of the risk analysis process by which limits designed to protect public health are established, in the food safety context. We believe, further, that in many cases it may not be evident where, how or to what extent these other factors have been considered. Both to support discussion of the legitimacy of considering individual other factors, and to make decisions based in part upon them more transparent, procedures need to be developed to identify these factors, to agree that they play a role in risk analysis, and to determine how they affect decisions. The factors include:
- Necessary value judgments embedded in primarily scientific processes;
- Scientific uncertainty and the application of the precautionary principle;
- Benefits of substances, practices and food technologies that also pose risks;
- Good practices in agriculture, food processing and veterinary medicine.
Necessary Value Judgments Embedded in Science
Sound science is an essential basis for standards, guidelines, and recommendations that set limits to protect public health. But science is never a sufficient basis for these decisions, in and of itself. Subjective choices and value judgments always need to be made in order to apply the best scientific knowledge to the setting of safety limits. This decision process is in fact a social policy choice, not merely a scientific judgment. In the Consultation report on Risk Management and in other contexts, Codex has recognized that risk management decisions involve the application of many non-scientific factors. Consumers International is pleased to note that CX/GP 98/10 also recognizes that, to a degree, elements of risk management, and therefore of decision-making based on non-scientific as well as scientific factors, are frequently present in risk assessments.
Consumers International believes it is critical to extract and examine value components, at each stage of risk analysis. This is essential so that the scientific basis for decisions and the other factors involved can be viewed as distinct, separate entities.
Value Judgments in Risk Assessments. Risk assessment, which in the Codex context may include those performed by JECFA and JMPR, and possibly in the future by a similar Joint Expert Committee on Microbiological Hazards, is primarily a scientific process. At the same time, scientific experts carrying out risk assessments routinely make a range of judgments on matters such as which risks are within the scope of their assessment and which are not; how much depth of scientific analysis is justified on a particular question or risk; where adequate scientific data do or do not exist to support an assessment; which evidence is most central or most credible on a particular point; and many other questions. In deciding these matters, experts are guided in part by scientific criteria and in part by their own subjective beliefs as to what is important, relevant, reasonable, and so forth. These subjective factors are at least partly non-scientific in nature.
In a classic report from the U.S. (Risk Assessment In The Federal Government: Managing The Process; National Academy of Sciences/ National Research Council, National Academy Press, Washington, DC, 1983), a committee of experts on what is now called risk analysis addressed points in a risk assessment where judgments must be made regarding the analytic approach to be taken. For example, what relative weights should be given to studies with differing results? Should dose-response relationships be extrapolated according to best estimates or according to upper confidence limits? Which population groups should be the primary targets for protection, and which provide the most meaningful expression of the health risk?
The NRC publication notes,
" that a scientist makes the choices does not render the judgments devoid of policy implications. Scientists differ in their opinions of the validity of various options, even if they are not consciously choosing to be more or less conservative. In considering whether to use data from the most sensitive experimental animals for risk assessment, a scientist may be influenced by the species, strains, and gender of the animals tested, the characteristics of the tumor, and the conditions of the experiment. A scientists weighting of these variables may not easily be expressed explicitly, and the result is a mixture of fact, experience (often called intuition), and personal values that cannot be disentangled easily. As a result, the choice made may be perceived by the scientist as based primarily on informed scientific judgment. From a regulatory officials point of view, the same choice may appear to be a value decision as to how conservative regulatory policy should be, given the lack of a decisive basis for choice."
A recent similar report by the Interdepartmental Liaison Group on Risk Assessment of the United Kingdom (Use of Risk Assessment Within Government Departments, Health and Safety Executive, HMSO, 1996), concludes:
"Risk assessment provides a necessary structure for rationalising decisions on the risks that society is prepared to accept. Such acceptance has to take account of values established by political debate and public willingness to tolerate risks in return for benefits. However, for the tool to work, those who rely on it for informing their decisions have to understand its nature and its limitations; that it relates to hypothetical rather than real persons; and that the process is assumption and value laden and subject to uncertainties which need to be exposed."
In summary, even risk assessment, the most "scientific" aspect of risk analysis, cannot be kept entirely free of values and other non-scientific factors. Scientists, regardless of how excellent their scientific credentials and how sincere their good faith efforts, can and do interpret scientific evidence differently, in light of uncertainties and the necessary value judgments described above. This is a reflection of the nature of science and scientific inquiry; science is not absolute.
Consumers International believes that, in the Codex context, it is essential to make such normal and necessary value judgments more explicit. National governments considering adopting Codex standards may conduct their own risk assessments, frequently using one done for Codex by JECFA or JMPR as their starting point. To enable such authorities to develop a sound scientific basis for their own decisions, it is critical to distinguish what is science and what is value judgment within the risk assessment relied upon by Codex.
Value Judgments in Risk Management. In the risk management sector of risk analysis, it is generally more evident that trade-offs among values and social policy choices must be made, not simply scientific decisions. One central question risk managers must decide is what is an appropriate level of health protection. This complex decision requires a basis in science, of course, but also involves other factors such as feasibility, economic costs of risk reduction, and social consensus on how much protection is desired. Considerations that affect the choice of protection level include judgments as to how much risk would be acceptable in conjunction with associated benefits, whether every individual member of society should be protected and whether certain subpopulations (such as young children or the fetus) require special protection, and the like. These factors clearly require ethical and social judgments, as well as a scientific risk assessment. Another common question risk managers must decide is what risk-management strategy to pursue. This choice also involves non-scientific factors, such as beliefs about who should bear the responsibility for managing a particular risk, a decision with both pragmatic and ethical dimensions.
While there appears to be general recognition that the risk management sector of the risk analysis process involves factors other than science, Consumers International believes that much more needs to be done to identify specific non-scientific factors in food-safety risk-management decisions, and to make the role they play more transparent. Within Codex, risk-management decisions are made by committees such as CCRVDF, CCFAC, CCPR and others that may recommend maximum limits for contaminants, residues or additives. As CX/GP 98/10 points out, correctly, expert bodies such as JECFA and JMPR also enter the realm of risk management when they recommend MRLs. Setting an MRL is determining an "appropriate level of health protection," and depends on many non-scientific factors, as just noted. Those other factors must be made explicit in Codex decisions, so that the scientific basis and the non-scientific basis for a given decision can be distinguished readily from each other.
Scientific Uncertainty and The Precautionary Principle
Consumers International is pleased that the Secretariats background paper CX/GP 98/10 identifies the Precautionary Principle as another factor that might be considered in Codex decisions.
At the 22nd session of the Codex Alimentarius Commission, Dr. Hiroshi Nakajima, then Director-General of the World Health Organization, proposed to the Commission that Codex standards should be set on the basis of both "sound science and a precautionary approach." A precautionary approach counsels decision-makers to err on the side of public health protection when formulating standards in cases where there is significant scientific uncertainty, or where the potential public health consequences of making an error are significant. This is a prudent approach, and rests on the ethical principle that it is "better to be safe than sorry" when public health is at stake. In terms of both health hazards and economic impacts, the cost of not taking action because of uncertainty may be far greater, in the long run, than the costs of prudent precautionary actions. The case of BSE (or "Mad Cow Disease") is one recent example in the food safety context of an instance in which early preventive measures based on the Precautionary Principle could well have averted extraordinary costs to society.
As a practical matter, the most critical question with respect to the Precautionary Principle is probably not whether it is appropriate to consider it, but rather when it is appropriate to invoke it. Consumers International agrees in general with the suggestion in CX/GP 98/10 that a precautionary approach is justified when there is uncertainty as to the exact nature and magnitude of the risk. The Secretariat offers some criteria that might determine when to apply the Precautionary Principle, including having an incomplete risk assessment and having difficulty in qualifying the risk or in identifying appropriate risk-management options.
All risk assessments and most risk-management strategies contain scientific uncertainty, some more than others. In some cases not enough may be known even to properly tell what the critical uncertainties are. How much uncertainty is acceptable in any given case is a social value judgment, and non-scientific factors must be considered in making that decision in the risk management process. A decision by risk managers that the degree of uncertainty is too large, given the potential consequences of an error, would be a sound basis for invoking the Precautionary Principle.
If Codex accepts the Precautionary Principle as a legitimate factor in decisions (which Consumers International believes it should do), it would then be necessary to make greater efforts to ensure that the scientific uncertainties in risk assessments are transparent to risk managers. In the Codex context, for example, expert advisory bodies might be asked to provide more explicit assessments of the uncertainties inherent in the scientific data they have reviewed and the analytical approaches they have chosen, and to indicate the impact of the uncertainties. For example, expert bodies can be asked to identify and quantify key uncertainties regarding risks; assess the degree of confidence they have in their key scientific conclusions; and explain the possible public-health consequences if any of those conclusions later turn out to be incorrect.
The benefits, as well as the risks, of a particular food substance, technology or activity inevitably are considered, explicitly or implicitly, in the risk management process, and sometimes in the risk assessment component of risk analysis as well. It is appropriate to consider benefits at some point in the process, since the amount of benefit associated with a risk affects the acceptability of the risk both to individual consumers and to societies.
Closely related to the relative magnitude of risks and benefits is the distribution of risks and benefits. Distribution of risk is an important factor in determining how to manage the risk. For example, if risks fall disproportionately on infants or pregnant women, or on those with a low-income, or on children with poor diets, that information is critical to developing and implementing risk management options. The question of who benefits and who is at risk also is crucial to public perception of the equity of the risk, which is a factor in the social acceptability of a risk.
Consumers International believes a more focused effort is needed to identify benefit issues embedded in each step of risk analysis, and in Codex decisions. Assumptions about the value of benefits may color expert judgments and are implicit in the selection of an MRL. For example, in determining MRLs for veterinary drug residues or pesticide residues, CCRVDF and CCPR typically consider the level remaining in foods when the drug or pesticide is used as intended to obtain the economic benefit of using it as part of the basis for the MRL. Decisions are not based solely on a scientific assessment of risks to health. Another assumption, generally unstated, is that use of the drug or pesticide is necessary; i.e., that there are no alternative ways for the producer to obtain the same benefits. A better assumption would be that alternatives do exist, including perhaps alternatives that involve smaller risks to public health. Such assumptions need to be more transparent, so that their validity and appropriateness can be examined and so that there is a clearer demarcation between the scientific and non-scientific components of the basis for risk-management decisions.
Another reason to examine benefit issues as an explicit "other factor" is that doing so could improve the rigor of benefits assessment. In Consumers Internationals experience, the value of benefits is often merely assumed or asserted; the degree of scientific data and analysis applied to estimating benefits is nothing like that applied to assessing risks. If the ultimate risk management decision is to reflect the balanced bests interests of all affected, a more scientific approach to assessing benefits seems essential.
As CX/GP 98/10 notes, Codex standards, guidelines, and recommendations frequently now consider "good agricultural practice" (GAP), "good veterinary practice" (GVP), and "good manufacturing practices" (GMP). GAP, GVP, and GMP, as defined by Codex, reflect practices currently permitted or authorized at the national level. Such practices differ between member countries, for many legitimate reasons, often reflecting traditions, technological and economic conditions in particular countries; they are not scientifically determined, for the most part.
Consumers International believes that consideration of "good practices" frequently has resulted in risk-management decisions that accept greater risk to public health than would result from strictly science-based decisions. For example, Codex standards for pesticide residues in foods (MRLs) are based on GAPs that result in the highest level of residues, providing that exposure to the resulting residues does not exceed the acceptable daily intake (ADI). There is ample scientific basis for concluding that pesticide residues may pose non-negligible risks to health at levels below the ADI. Recognized risks of this sort include the additive effects of multiple residues with the same mechanism of toxicity (e.g., the organophosphate insecticides); exposure to pesticides by non-dietary routes (drinking water, household pest control) as well as in foods; and the extra sensitivity of very young individuals to certain toxic effects (e.g., neurodevelopmental toxicity). Risk assessors also acknowledge that there is an insufficient basis at present to quantify some suspected risks of pesticides in the diet, such as potential effects on the endocrine and immune systems.
Based strictly on scientific analysis of the need for health protection, then, a strong case can be made that MRLs for certain pesticides should be set with as wide a margin of safety below the ADI as possible, to accommodate both the known risks of multiple exposures and the uncertainties in estimating certain risks. When CCPR proposes MRLs based on "lowest-common-denominator" GAPs, it appears that non-scientific factors in the decision that run contrary to the scientific basis for health protection have over-ridden or at least offset the scientific basis for the decision. Consumers International believes this example, as well as the BST issue, would be a suitable case study for examining the appropriate role of factors other than science in Codex decisions.
Consumers International also proposes that considerations of GAP, GMP and GVP might better be reconceptualized as examinations of "best practices" (BP). In some countries, national GAP may result in far lower residues, and thus national standards for pesticides may justifiably be set much lower than the Codex MRL. It is important for the GAP(s) upon which the Codex MRL is based to be stated explicitly, for transparency. And to the extent possible, we believe Codex committees should also specify BP, to clarify the extent to which better practices exist that permit selection of a higher level of health protection by countries capable of achieving such a public-health goal.
This reconceptualization process is under way already, to some extent, at CCPR. For example, CCPR at its 29th session agreed that "attention should be paid to measures to reduce residues, including Integrated Pest Management (IPM), availability of quality pesticides, and training in safe and efficient use as part of IPM. IPM should be taken into account when developing GAP information for JMPR." IPM can be thought of as "best agricultural practice" since it almost always utilizes lower application rates of pesticides while effectively controlling pests, which is more economical to growers and lowers the probability of risk to consumers, agricultural workers, and the environment.
Other Factors Related to Promotion of Fair Practices in Food Trade,
Including Other Factors Related to Food Labelling
Consumers International accepts the Secretariats assessment in CX/GP 89/10 that economic interests of member countries are considered in Codex decisions, and we believe that is proper, as well as inescapable.
Consumers Internationals chief concern under this heading is with "other factors" that are not obviously relevant to health protection, but which nonetheless may be very important to consumers, and which we believe therefore need to be addressed, at some level of the risk analysis process. Many of these factors might be better addressed not in Codex decisions, but in decisions at the national level, or in food-buying decisions of individual consumers. A discussion is needed, however, to identify the factors and to reach agreement as to which are and are not appropriately within the scope of Codexs standards, guidelines and recommendations.
In this regard, Consumers International believes that the third "Statement of Principle," which explicitly refers to labelling decisions as an appropriate arena for considering many "other factors," offers the best guidance for this discussion.
We will first enumerate some other factors that consumers consider important, and then address some general principles concerning how they might be dealt with in labelling decisions. Consumers International regards the work of CCFL as one of the most critical sectors of the risk analysis process. We believe enabling consumers to make informed choices in the marketplace is one primary function of labelling and is an essential, pre-eminent component of "promoting fair practices in food trade."
Some "other factors" involved in labelling decisions are also "relevant to the health protection of consumers." Among the most important of these is the need to empower individuals to exercise their personal right to accept or avoid particular risks. Individual consumers differ in their willingness to accept risks of different kinds; such differences may also exist between different societies. From an ethical perspective, consumers have a right to know what risks they are exposed to; this is closely related to the fundamental right to bodily integrity. Individuals also have a right and a responsibility to play a role in managing their own risks, by deciding which risks they are and are not willing to expose themselves to.
While establishing a uniform national (or international) level of health protection is both necessary for practical reasons and beneficial in the interests of trade, it is important and feasible to preserve the rights of individuals who disagree with a nationally determined level of risk to choose their own "level of health protection." The best way to promote harmonisation while protecting the ethical values also involved is to inform consumers, so that they can make their own risk-management decisions. Labelling is often one of the best ways to convey the information consumers want, need and have a right to have.
Consumers and societies often also consider many other factors relevant to decisions as to whether to consume or permit consumption of certain foodsfactors such as religious and philosophical beliefs, concern for the welfare of animals, concern about the environ- mental impacts of food production practices and desire to support more sustainable food production, desire for more "natural" and less "technological" production methods, and nutritional, dietary and personal-preference factors. Most of these factors are not clearly "relevant for the health protection of consumers," but in Consumers Internationals view, they are relevant for the "promotion of fair practices in food trade."
To determine to what extent these other factors should be considered in making labelling decisions in food-safety risk analysis, agreement needs to be reached on the purposes of labelling and how best to use it to achieve the agreed-upon objectives.
Purposes of Labeling
Codex labelling actions have at least two distinct, and in Consumers Internationals judgment, equally important purposes.
One purpose of labeling is to protect the health of the consumer. For example, advice to "refrigerate after opening" may inhibit growth of pathogens (as well as preserve sensory quality). Nutrient labelling enables consumers to obtain nutrients essential for health and to avoid excessive intake of those (such as fat or sodium) that may have adverse effects if not consumed in moderation. Labelling to identify known allergens in foods may help people sensitive to those allergens avoid a potentially life-threatening adverse reaction. Labelling in some countries identifies food additives and permits consumers to avoid eating foods that contain particular additives that concern them, for health or for other reasons. Labelling is thus a potentially valuable risk-management tool.
Labelling has a second fundamental purpose as well: communication. The basic function of a food label is to give consumers information they need to know about the food. As just noted, that may include safety information. But food labels also inform the consumer about the identity and ingredients of the food (that is, they address the consumer’s right to know what one buys and eats); provide advice on proper handling and preparation of the food; tell by whom it was made and marketed. Labels also may tell consumers how food was produced (for example, organically grown); how it has been processed (for example, frozen, pasteurized, or irradiated); and even whether the foods packaging was made with recycled materials.
Food Labelling as a Risk-Management Tool
Consumers International agrees with the general proposition that food safety regulation should keep unsafe foods off the market. Allowing a food that poses substantial risks to public health to be marketed while relying on labelling to advise consumers of the risks is generally not an acceptable alternative strategy. However, there is a large "gray area" that applies to many current food safety debates. Some foods pose small risks, or risks that are decidedly uncertain in nature or magnitude. There may be an insufficient scientific basis to conclude that a food poses hazards and should be kept off the market; at the same time, there may be sufficiently plausible scientific questions about risks to preclude being sure that the risks are negligible. Faced with such (normal) scientific uncertainties, many consumers tend to be risk-averse, and would prefer not to accept personally even risks that fall below the threshold officially considered "significant."
Risk-averse behavior is more likely in cases such as BST use, where consumers see no personal benefit, but only possible risk. In this context, labelling can allow individuals either to avoid or to knowingly accept risks that the expert community has defined as "acceptable" from a public-health standpoint, but about which individual consumers may have doubts. Labelling lets individuals make their own value judgments about risks and benefits and how much uncertainty they are willing to tolerate. Labeling is thus one way to exercise the precautionary principle.
Dr. Karen Dodds of Health Canada wrote a background paper on the role of government risk managers in risk communication, for the WHO/FAO Joint Consultation on "Risk Communication in Food Safety Decision Making," held in Rome this past February. In her paper, Dodds stressed the need for risk managers to identify and address legitimate concerns of the public, and to take those concerns into account in determining how best to manage the risk.
An important non-scientific attribute of risks, addressed in Dodds’ paper and in others prepared for the same Consultation, is whether a risk is voluntarily chosen, or whether it is perceived by consumers as being imposed on them without the individual’s knowledge or consent. Regardless of the size of a risk, involuntary risks are more worrisome and more likely to be objectionable to consumers than risks over which the consumer has a choice, and feels in control. One of the key strategies that Dodds urged governments to pursue in managing food-related risks is, "Try to make risks more voluntary."
Labelling can make risk more voluntary, by enabling individuals to choose whether or not to take whatever risk they perceive to be associated with a particular food. Knowing that they have a choice and the power to avoid the risk makes the risk less worrisome and less objectionable in the eyes of most consumers. Such a desire to make risks more voluntary possibly underlies the overwhelming preference for labeling of food produced through biotechnology, expressed by consumers repeatedly in public-opinion polls on that topic in many countries.
Consumers International also notes, however, that use of labelling as a risk-management tool has definite limitations. There may be situations where the risk is unquantifiable and consumers cannot make truly informed choices, even with label information. There also may be consumers who because of education level, language or other cultural factors, may not read or would not fully understand risk-related information on food labels. Selectivity is essential; labelling is not the right risk management option for every risk.
In Consumers Internationals judgment, what risks concern consumersas opposed to merely what risks the expert community judges importantis a legitimate "other factor" to consider in labelling decisions. Where objective data exist, in the form of scientifically sound opinion polls (as they do, for example, regarding genetically engineered foods), Consumers International believes the information wants and needs of consumers should be acknowledged and given great weight in labelling decisions.
Food Labelling as Risk Communication
Whether a labelling decision is based on the need to inform consumers about health concerns related to a food or to inform them about aspects of a food other than its risks to healthi.e., when considering labelling in its broader function of communicating facts about foodsthe most important "other factor" is knowledge of what facts consumers need, want and have a right to know about the foods they eat.
Consumers International believes that the most appropriate way to determine what information consumers want and need on labelsi.e., which facts are "material" to consumersis to ask consumers, through opinion polls, focus groups and other forms of dialogue between producers, processors, and government on one hand, and consumers on the other. To promote effective consideration of consumer information needs in the risk analysis process, greater efforts need to be made to gather this kind of data. Information-gathering interactions with consumers are also an important part of risk communication, an acknowledged critical component of risk analysis.
Decisions by expert authorities or by food processors that consumers "do not need" or do not have a right to get certain information about foods are likely to be flawed in several ways. Such decisions may be uninformed, based on no data describing what information consumers actually want and need. They are also likely to be seen as ethically unsound (paternalistic), and they are frequently biased by other concernssuch as that consumers may use the information as a reason not to buy the product.
Indeed, information on a food label may lead some consumers not to buy a product, based on any of the numerous factors listed above. Other consumers who want to buy that food product might not be able do so without the label information. Consumers International believes the most fundamental issue is informed choice in the marketplace. At the heart of the "promotion of fair trade practices," in our view, is the principle that consumers have a right to know what they are buying, and the market cannot operate as it should when they are denied the information they need.
At What Points in the Risk Analysis Process Should Identified
"Other Factors" Appropriately Be Considered?
Consumers International believes it is important to view the risk analysis process as a whole system, in addressing this set of issues. Risk analysis on a particular food safety matter may include risk assessment, risk management and risk communication. It also is likely to include international (Codex) health-standard setting and labelling activities, the same two activities at the national level, and consumer decisions in the marketplace. It is essential to identify and address factors other than science that occur in all of these sub- processes of the larger risk analysis, and to determine where in the overall process each is nest condidered. The answer may be different for different factors.
Determining how to deal with "other factors" at each level requires a clear conceptual approach to how the system as a whole is supposed to function. Decisions at all levels need to be aware of how these non-scientific factors have been or will be dealt with in other parts of the system. Decision-makers at each level have their own roles, rights and responsibilities to fulfill. Codex, in carrying out its own role, must be sensitive to how the consideration it may give to "other factors" affects the decisions of other players in the full system of risk analysis.
Under the terms of the SPS agreement, national food safety authorities have the right to establish the appropriate "level of health protection" for their citizens. Governments are not bound to accept Codex standards as their own; specifically, they are free to choose a "higher level of sanitary or phytosanitary protection," as long as they have a "scientific justification" for doing so. It is evident, of course, that many "other factors" addressed here that bear directly on the selection of the appropriate level of health protection can be considered at the national level, not merely at the Codex level.
As long as there is a sound scientific basis for the chosen level of health protection that a national government selects, other factors, such as consideration of the acceptable level of risk in view of the benefits associated with the substance or activity that poses the risk, or differing degrees of willingness to accept scientific uncertainty, are also appropriate parts of the basis for national food-safety standards. Both the magnitude and the distribution of benefits and risks of a food substance or activity may differ from one country to another, because of characteristics of the population, the culture or the food-production system in different societies. Perceptions of the appropriate ratio of risk to benefit and of the need to take a precautionary approach also are likely to differ from nation to nation. Efforts to harmonise food standards notwithstanding, different countries, each with a sound basis in science and each giving different weights to factors other than science, may choose levels of health protection for their citizens that differ from Codex standards. As we understand the text of the SPS agreement, this is not only clearly permitted, but also expected.
In this context, transparency of Codex decisions seems especially important. There seems to be no disagreement that factors other than science are already part of the basis for many Codex standards (though there is still a need to determine what those factors are, and how they affect decisions). It is essential that national authorities considering whether to adopt a Codex standard or to adopt another standard should be able to see clearly both the basis in science for the Codex decision, and the non-scientific basis for it. Countries that do not accept Codex standards may have a different scientific basis for estimating the risk, or may rely on the same scientific basis but have a different societal basis for determining the appropriate level of health protection. In order to ensure that differing national standards are not unjustified trade barriers, it is important to be as clear as possible on what the basis is for each decision.
Consumers International does not believe Codex committees should attempt to balance the risks and benefits of food substances or activities, or to give weight to many of the other non-scientific factors identified here, when setting standards. Making such trade-offs requires a political, not a technical process, and Codex is primarily a technical, not a political, entity. It would be exceptionally difficult for Codex to balance all the varied national perceptions of non-scientific factors, and arrive at "consensus" international judgments. This clearly is not the role assigned to Codex under the SPS agreements.
At the same time, Codex decisions are not based purely on science, because many of the "other factors" are relatively inextricable from the scientific components of risk analysis. In our judgment, the most useful step Codex can take at this point is to develop a policy that identifies the non-scientific factors that are either explicitly or implicitly considered in risk assessments and risk management decisions, and requires that they be documented as they occur and are addressed throughout the Codex risk analysis process.
Consumers International also believes that the Codex Committee on Food Labelling is the appropriate forum in which to consider a wide range of "other factors." Although Codex may not decide to ban or to restrict the use of a substance or food technology on the basis of consumer preference or myriad "other factors" that determine such preferences, CCFL may certainly recognize that those other factors are important to consumers. In our view, when there is objective evidence from several countries that consumers actively require and demand label information about some food attribute, CCFL should either adopt international standards for labelling that inform consumers of that attribute or adopt guidelines that strongly encourage national governments to require such labelling.
Consumers International believes that developing a better conceptual approach to what "other factors" are legitimate parts of the basis for decisions in food-safety risk analysis will improve the quality of decisions, improve the process for reaching decisions, and greatly facilitate risk communication about the subjects of decisions. With those goals in mind, we offer these recommendations for the Codex Alimentarius Commission and its committees and subsidiary bodies:
- Consideration of how to address "other factors" should proceed within the larger context of the ongoing discussion of risk analysis in Codex. How to deal with the "other factors" needs to be addressed in terms of all three major components of risk analysisrisk assessment, risk management, and risk communication.
- Codex should approach the role of "other factors" in risk analysis systematically, taking into account the appropriate roles of Codex itself, national authorities, and consumer buying behavior in the marketplace in weighing these factors.
- Codex should elaborate its general principles on how to consider "other factors" before it makes final decisions on BST, certain MRLs for pesticides, labelling of foods produced through biotechnology, and other issues with clearly-identified, important non-scientific components of the debate.
- The Codex Committee on General Principles should designate the precautionary principle as one of the principal "other factors" to be explicitly considered in setting Codex standards, and develop policy guidelines on when it may be appropriate to invoke this principle.
- Codex committees that establish standards to protect public health should endeavor to make all aspects of their risk assessment and risk-management processes in which non-scientific factors are considered as transparent as possible in terms of the other factors that were considered, and how they affected decisions.
- Codex committees that depend on risk assessments by JECFA, JMPR and other expert bodies should ask those advisory panels to include explicit statements of where and how their assessments are uncertain, and how defensible their conclusions on key issues may be, in light of the recognized uncertainties.
- Expert bodies should also be asked to provide more explicit explanations of value judgments embedded in their risk assessments, such as the relative importance of the risks and benefits of the substance or activity being assessed, and the nature of the safety margin chosen and reasons for choosing it.
- Recognizing that individual experts make value judgments that may color their recommendations, Codex should seek to include in risk assessments on which it bases its decisions the views of a range of scientists, including academic experts and those who work for or with consumer and other public-interest NGOs, in addition to the government and industry scientists typically involved at present.
- CCGP should elaborate more fully on the meaning of the third "Statement of Principles," clarifying in detail the extent to which and the manner in which factors other than science should be weighed in Codex labelling decisions.
- Expressed consumer interest in "other factors" related to a food safety issue should be one legitimate basis for Codex labelling decisions.
- CCFL should consider labelling foods as a means that permits consumers to act in their own behalf and make their own legitimate choices, both to manage risks to their health that concern them and to serve whatever other values they wish to bring to bear on their purchase decisions.
A CASE STUDY:
APPLICATION OF GENERAL PRINCIPLES CONCERNING
THE ROLE OF SCIENCE AND NON-SCIENTIFIC "OTHER FACTORS"
TO THE CASE OF BOVINE SOMATOTROPIN (BST)
Consumers International is pleased to present this case study, applying the general principles articulated in our accompanying paper to the specific case of BST.
Although the Codex Alimentarius Commission in its request to the Committee on General Principles asked that the review be applied to the case of BST and PST, in Consumers Internationals judgment, the PST issue has not yet advanced to the point within the Codex risk analysis process where such an assessment would be fruitful. Our case study therefore focuses only on BST. We note that the Secretariat, in preparing CX/GP 98/10, took the same approach.
Consumers International conceives the purpose of this case study as being to examine the risk analysis process on BST and to identify factors other than science that are, have been, or perhaps should be part of the basis for decisions, by Codex and other authorities. We hope the case study will illustrate general principles on the role of "other factors" in risk analysis and will expand understanding of what is and is not the basis for decisions that are to be made by Codex on BST. It is not a purpose of this case study to prescribe how Codex should resolve the BST decision.
We note that CX/GP 98/10 does reach conclusions and make recommendations regarding the "correct" Codex decision on BST. In the view of Consumers International, such conclusions and recommendations are premature and unjustified, for a number of sound reasons we will elaborate below.
"Other Factors" in the BST Risk Analysis Process
CX/GP 98/10 points out that the complete scientific report on the evaluation of BST risk issues by JECFA at its 50th meeting is not yet available. CX/GP 98/10 also notes that although JECFA issued a summary statement of its conclusions at the end of its meeting in February, JECFA also asked that that summary not be quoted or cited, pending release of the complete report. Consumers International will honor that prohibition. For these reasons, we will address the JECFA risk assessment process only in general terms, based on our extensive past experience with and our general knowledge of the risk assessment process, the functions of expert advisory bodies, and the BST issue.
One salient fact about JECFAs review of BST is known: JECFA has determined that it will recommend that no ADI and no MRL need to be established for BST. That is the same recommendation JECFA made in 1993; i.e., JECFA recommends no change in its previous position.
CX/GP 98/10 also states its intention to refer to the JECFA process and its outcome only in general terms. Nevertheless, the authors of the background paper state some important conclusions about JECFAs risk assessment and its recommendations. For example, at Paragraph 19: " the precautionary principle cannot apply in the case of BST as the scientific basis clearly exists." And at Paragraph 24, the paper states that while "other factors" would have to be considered in conjunction with scientific evidence as a general principle, "This does not appear to be applicable to the particular case of BST where recommendations covering both risk assessment (ADI) and risk management (MRL) have been made on a scientific basis." Paragraph 19 also states "JECFA has identified no significant risk" from BST and that "no safety concern has been put forward" with respect to BST.
With all due respect, Consumers International does not accept these interpretations of JECFAs action. Other interpretations are also possible. In our judgment, until the full JECFA report on BST has been published, it is not possible to know exactly what JECFA concluded, or what the actual basis for JECFAs conclusions may have been. Nor can it be determined whether JECFAs risk assessment is scientifically sound. The report must be published, reviewed by the expert community and interested parties, subjected to normal scientific criticism and debate, and discussed in the context of its implications for risk management. More than one view of those implications seems likely to be tenable. The conclusion in CX/GP 98/10 that JECFA has provided "a basis in science" which then allows Codex to follow only one course of action on BST is simply not supportable.
Consumers International believes, as outlined in our comments on general issues related to the role of "other factors" in risk analysis, that one cannot conclude at present that the basis for JECFAs recommendations was "science." In all likelihood some non-scientific factors also entered into the basis for the "no ADI/no MRL" recommendations, and in fact those other factors may have received great weight.
In the interests of a sound risk analysis (including effective risk communication), Codex, and in particular CCRVDF, should take no action on the JECFA recommendation until the full report has been published and reviewed. If, as Consumers International expects, a careful review of the basis for JECFAs recommendation indicates that it was based in part on science and in part on non-scientific other factors, those other factors need to be identified and examined. Codex