The Precautionary Principle and permitting the possible benefits of food – II Of symmetry, analogy and proportionality …

8 May 2013

The Nutrition and Health Claims Regulation is a precautionary measure. It rests on the European legislature’s acceptation of “available information” and uncertainty as a proportionate legal basis for the ex anteprohibition of health claims to ‘precaution’ the risk they might pose to human health. However, nothing less than complete certainty produced by generally accepted scientific evidence based on double blind placebo controlled randomized clinical trials is regarded as the acceptable and proportionate threshold for authorizing claims. Where uncertain risk of harm may be assumed to prohibit, benefits must be conclusively proven to authorize. In the name of a high level of protection, this duplicious approach produces a reduction of the risk on harm and a reduction of the “risk” on benefits. It is assumed, but never proved, that in this risk-free society – free of risks on harm and free of ‘risks’ on benefits – all will be well. 

A symmetrical and indiscriminate application of the Precautionary Principle would produce numerous nutrition and health claims stating, suggesting or implying that a possible or probable relationship exists between a food category, a food or one of its constituents and health.

I. Introduction

The European Nutrition and Health Claims Regulation (NHCR) is a coercive means of ex ante settling theoretical and political disputes arising out of speculative and arbitrary allegations and/or opinions in matters of nutrition and health claims. This ex ante “redress” approach to commercial communications deprives all health claims – including truthful ones – a priori of their entitlement to free movement throughout the European Union on grounds of the unproven allegation that, unless reviewed by EFSA and the Community legislature, they inherently and intrinsically constitute or produce a potential ‘wrong’, ‘hazard’ or ‘injury’.

Apart from showing a deep and ideologically driven prejudice against industry and commerce, the EU’s ex ante approach also stems from the fact that the EU Treaties have always demanded that the legislature will take as a base a ‘high level of protection’.[i] The NHCR provides for the highest level of protection, because it entails a precautionary – ex ante – approach to providing redress and protection. Placing commercial communications that concern foods under a system of prior constraint and authorisation is a precautionary measure that has its place in the context of the ‘Precautionary Principle Article’ of Regulation 178/2002/EC.[ii]

According to this Article, precautionary measures such as those laid down in the Claims Regulation, concern the possibility of harmful effects on health. Also, they should have a provisional character because they were taken on the basis of scientific uncertainty. They shall be reviewed within a reasonable period of time:

“Article 7 / Precautionary Principle
1. In specific circumstances where, following an assessment of available information, the possibility of harmful effects on health is identified but scientific uncertainty persists, provisional risk management measures necessary to ensure the high level of health protection chosen in the Community may be adopted, pending further scientific information for a more comprehensive risk assessment.
2. Measures adopted on the basis of paragraph 1 shall be proportionate and no more restrictive of trade than is required to achieve a high level of health protection chosen in the Community, regard being had to technical and economic feasibility and other factors regarded as legitimate in the matter under consideration. The measures shall be reviewed within a reasonable period of time, depending on the nature of the risk to life or health identified and the type of scientific information needed to clarify the scientific uncertainty and to conduct a more comprehensive risk assessment.” (emphases added)

The NHCR is a ‘Verbotsystem mit Erlaubnisvorbehalt‘, a regulatory system that combines an a priori prohibition with an authorization provision. Per Article 7 of Regulation 178/2002, such regulatory systems may be applied only when there is a possibility of harmful effects on human health. What is new in the NHCR, is that this ‘Verbotsystem mit Erlaubnisvorbehalt‘ is applied to precaution the potential risk posed to human health by misleading and/or incomprehensible commercial speech. In Case C-221/00, Commission v Republic of Austria,[iii] the Court determined that health claims placed on legitimate foods cannot cause a risk to health that exceeds the level of ‘residual risk’. Still and all, when health claims are placed under a ‘Verbotsystem mit Erlaubnisvorbehalt‘, the only criterion that counts is the claims’ risk to health. Applying precaution in the field of economic harm caused by possibly misleading commercial speech is not foreseen in the EU Treaty.[iv]

II. The a priori requirement of risk assessment

When a prohibition antecedes approval or authorisation, the burden of proof is shifted from the ‘prohibitor’ to the ‘prohibitee’, from the ‘precautioner’ to the ‘precautionee’. When this happens in the absence of legal or verifiable criteria and transparent procedures that must be respected and followed by the legislature in setting the prohibition, and/or, when the judiciary expressly accepts a prohibition as a “given” that may exist in disregard of such legal criteria and procedures, individuals and business operators find themselves defenseless against the probition. When ‘precautionees’ are thus deprived of their right to challenge the prohibition in a court of law, they are forced to do what the prohibitor is excused of doing: provide conclusiveevidence to render the prohibition nugatory.

The Food Supplements case[v] provides legal tools that help us better understand and clarify the precautionary structure of the NHCR. In Food Supplements, the Court addressed under what legally permissible circumstances all forms of vitamins and minerals might be prohibited by the Communityex ante of their authorization. In points 50 – 52, the Court held:

“50. As the Advocate General has stated at point 40 of his Opinion, it is clear from the preamble to Directive 2002/46 [the Food Supplements Directive], […..], that the Community legislature gives, as the rationale for the prohibition, considerations related to the protection of human health.

51. It remains necessary to ascertain whether the measure is necessary and proportionate in relation to the objective of protecting human health.

52. With regard to judicial review of those conditions, the Community legislature must be allowed a broad discretion in an area such as that involved in the present case, which entails political, economic and social choices on its part, and in which it is called upon to undertake complex assessments. Consequently, the legality of a measure adopted in that area can be affected only if the measure is manifestly inappropriate having regard to the objective which the competent institution is seeking to pursue […].”

Within the confines of this broad discretion, Article 7 of the EU’s General Food Law opens a wide window for precautionary measures, be it that they must be limited to precautioning the risk that nutrition and health claims pose tohuman health. In point 69 of Food Supplements, the Court held:

“It is also necessary to state in that regard that, by virtue of Article 7 of Regulation (EC) No 178/2002 […] laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety […], the Community legislature is entitled to adopt the provisional risk management measures necessary to ensure a high level of health protectionand may do so whilst awaiting further scientific information for a more comprehensive risk assessment, as is stated in the 10^th recital to Directive 2002/46 [the Food Supplements Directive].” (emphasis added)

The Court held that precautionary general prohibitions fall within the scope of the Union legislature’s broad discretion, in particular when it must make “complex assessments” in specific circumstances where, following an assessment of available information, the possibility of harmful effects on health is identified, but scientific uncertainty persists. Yet, when it comes to taking precautionary measures, the Union’s discretion is delimited in the 2^nd paragraph of Article 7 of Regulation 178/2002/EC, stating inter alia that precautionary measures “shall be proportionate and no more restrictive of trade than is required to achieve the high level of health protection chosen in the Community.”

In Food Supplements, the Court determined in points 72 and 73 that the proportionality requirement is fulfilled when a precautionary prohibition is accompanied by “a procedure designed to allow” a given substance to be added to a positive list. Transposed to the health claims situation, this would mean that a precautionary prohibition of health claims must be accompanied by “a procedure designed to allow” a given health claim to be included in a positive list. Such a procedure must not only “comply with the general principles of Community law, in particular the principle of sound administration and legal certainty.” It must also …

“… be accessible in the sense that it must be expressly mentioned in a measure of general application which is binding on the authorities concerned. It must be capable of being completed within a reasonable time. An application to have a substance included on a list of authorised substances may be refused by the competent authorities only on the basis of a full assessment of the risk posed to public health by the substance, established on the basis of the most reliable scientific data available and the most recent results of international research. If the procedure results in a refusal, the refusal must be open to challenge before the courts […].”

III. The Labyrinth, circumventing the assessment of risk

When the Court’s Judgment is superimposed over the NHCR, the Regulation emerges as a carefully designed labyrinth by which the European legislature completely absolved itself from the burden to ever perform a “more comprehensive” and “full assessment” of the alleged potential risk that health claims pose to human health. This is the – only – risk that may serve as the legal basis for the ex ante prohibition. Moreover, the NHCR makes it impossible for food business operators to get the prohibition lifted by providing proof of absence of harm.

In the NHCR, proof of absence of harm was silently replaced by proof of presence of benefit. Food business operators who enter the labyrinth can only find the way out by providing uncontested and generally accepted proof of presence of benefit in the form of a convincing demonstration of a cause and effect relationship between a food and a beneficial effect. Operators who would seek a way out by providing proof of absence of harm will never get out of the labyrinth. The initial ex ante prohibition was effaced by placing full emphasis on the approval or rejection of an application for authorization.

However, a rejection of an application for authorization of a health claim lacks the force of law, because it is the authorization that lifts the prohibition that preceded the rejection. Without authorization, the prohibition remains in force and it does not require the rejection to remain in force permanently. The rejection of an application for authorization does not establish the permanency of the prohibition, because it is not “a full assessment of the risk posed to public health by the health claim, established on the basis of the most reliable scientific data available and the most recent results of international research.”

The result of this, is that NHCR will remain a “permanently provisional prohibition.” The permanency of the provisional prohibition is created by circumventing the requirement laid down in paragraph 7.2 of the Regulation 178/2002/EC, which says that such prohibitions “shall be reviewed within a reasonable period of time.” Thus, the NHCR is a legal construct that contains a precautionary ex ante prohibition – of health claims – that will never be open to challenge before or review by the courts. Moreover, the NHCR “template” makes all review of precautionary measures impossible. Still, Article 7 of Regulation 178/2002/EC and the applicable EU case law would require that the Commission shall base a permanent refusal to authorise a claim on a full assessment of the risk the claim poses to public health.

IV. Reversing the burden of proof and what must be proven

It will be argued that a demonstration of benefit replaces or should replace the assessment of the potential harm. That’s fine, except that the NHCR expressly prohibits a precautionary approach when uncertainty exists concerning potential health benefits of foods or foodstuffs. When it comes to benefits, the NHCR demands absolute certainty. Still, the ex ante prohibition of health claims was not based on an assessment that produced absolute certainty. In addition, the prohibition does not rest on an assessment of the scientific uncertainty concerning a food product’s healthbenefits. It rests on an assessment of uncertain risks that health claims pose to health.

The logical application of the precautionary principle as the pillar supporting the prohibition, demands that the potentialdanger posed by a claim must first be demonstrated and then, if possible, disproven. In case of a precautionary prohibition, the proving as well as the disproving must take place within the same context and they must address the same subject, formulated in Article 7 of EU’s General Food law not as the beneficial, but as the “harmful effects on health.” Certainly, under the umbrella of precaution, the burden of proof is reversed from proof to disproof, but that does not change the fact that an applicant/precautionee must first of all demonstrate the harmlessness of the health claim. Someone accused of stealing a car is not acquitted when he provides evidence that concerns helping old ladies cross busy streets.

Curiously, if the operator could convincingly prove the absence of harm of a health claim – which would not be very difficult in light of the fact that in Case C-221/00 the Court flatly determined that health claims can only pose residual risks to health – this in itself would lead to the lifting of theex ante prohibition. In turn, the lifting of the ex anteprohibition would wipe out the authorization process laid down in the NHCR. What would remain is an ex post benefit-assessment and benefit-management procedure in which the competent authorities may request no more than that a food business operator making a claim shall prove it. In view of what the Court held in point 73 in Food Supplements, this would also imply that the competent European authorities would no longer be in a position to authorize a claim. Their powers would be strictly limited to refusing a claim.

To paraphrase the Court’s point 73 in Food Supplements, in an ex post assessment of a health claim, the claim may be definitively refused by the competent authorities only on the basis of a full assessment of the alleged benefits to public health, “established on the basis of the most reliable scientific data available and the most recent results of international research.” However, this is not what the NHCR entails. In it, precaution is applied asymmetrically and arbitrarily.

V. In search of analogy and symmetry

One of the NHCR’s general conditions, laid down in Article 5.1(a), is that the use of nutrition and health claims shall only be permitted if “the presence, absence or reduced content in a food or category of food of a nutrient or other substance in respect of which the claim is made has been shown to have a beneficial nutritional or physiological effect, as established by generally accepted scientific evidence.” In and by itself, thisconditio sine qua non is vague and not very meaningful, especially because the legislature omitted providing definitions of ‘beneficial’ and ‘generally accepted’.

But, the condition is qualified and better explained in point 3.a of the General Principles for Scientific Substantiation, Annex to Regulation 353/2008/EC.[vi] There, we are advised that “[t]he substantiation of health claims shall take into account the totality of the available scientific data and, by weighing the evidence, shall demonstrate the extent to which the claimed effect of the food is beneficial for human health.” This clearly implies a probabilistic approach to beneficiality, a grading of the beneficial effects, supposedly ranking these effects from highly beneficial to not beneficial at all. It also implies that, in the eyes of the Union, generally accepted scientific evidence can be the result of the probabilistic evaluation of evidence.

When one introduces into legislation that concerns public health the notion of ‘the extent to which something is beneficial’, we should approach beneficial effects in a manner analogous to the one we use in approaching risk and adverse effects. To achieve similarity between probabilistic methods of assessment, forecasting the (un)certainty of a future event must take place irrespective of its qualification in terms of good (beneficial) or bad (adverse). Apparently, this is what the Community legislature proposed in Regulation 353/2006/EC, where it seeks to be informed about the extent to which a physiological effect is beneficial for human health. ‘Come rain or come shine’ must be forecast, so it seems, by applying analogous methods.

Philosophically speaking, when opposing subjects – beneficialversus adverse effects – are involved, the value judgments about each of the opposites would have to be identical. If the value judgments were to be different, there would be no way of determining whether the subjects are opposing or similar. Determining what is good and what is bad requires the application of one single value judgment method. So does determining the degree to which something is bad or good. One cannot correctly value – the degree of – an adverse effect of something by applying value judgment type A and value – the degree of – a beneficial effect of an opposite something by applying value judgment of type B.

In the NHCR, different effects (risk and benefits) are valued by applying different value judgment methods. While uncertainty is the judgment value applied in precautioning risks, certainty is the judgment value applied on permitting benefits. The result of this asymmetrical approach is that of the 4.637 ‘Article 13.1’ health claims that were submitted for review to the European Food Safety Agency (EFSA), 222 survived the test of beneficiality. The result of considering the application of symmetry between the NHCR and the Traditional Herbal Medicinal Products Directive – accepting traditional use as a threshold for market entry of a traditional health claim – is that of the 4.637 claims some 1.500+ health claims were placed on hold. This, however, creates a discriminatory asymmetry within the NHCR between non-traditional and traditional health claims. That asymmetry could have been prevented, had the EU legislature followed a symmetrical approach from the beginning. A symmetrical NHCR would also have prevented that traditional health claims had to be placed on hold.

VI. Effects of asymmetrical assessment and management

The problem that a foodstuff – e.g. sugar – is undesirable in the eyes of consumer advocates and most policy makers and desirable in the eyes of most consumers, can only be resolved by assuming identical value judgments. The consumer advocates resolve this problem by exerting political pressure and the force of law to impose their value-of-sugar judgment on industry so that consumers who don’t share their value judgment will still have to behave and live by the advocates’ values. At least insofar as sugar is concerned.

Legally speaking, lack of analogy would cause a prejudice for both opposite effects. The adverse effects could claim that they are being unduly overexposed and the beneficial effects could claim that they are being unduly underexposed. Both effects could claim a violation of the principle of equal treatment and non-discrimination. Commercially speaking, if analogy would be missing, food business operators would be exposed to a regulation that would take much more adverse than beneficial effects into consideration. In terms of public health, if analogy would be missing, consumers would be asymmetrically exposed to much more negatives than positives, and that would, in the case of foods, seriously disturb the making of dietary choices.

Place this analogy/symmetry template over the market for ‘mobility’ and one can see right away that when analogy is lacking in measuring adverse and beneficial effects, e.g. in comparing systems of private and public transportation, or combustion- and electricity-driven cars, consumers won’t be able to make ‘symmetrically informed’ choices. Place the analogy/symmetry template over the Claims Regulation and one can immediately see its prejudicial character. By overexposing the risk of uncertain adverse effects, all claims – even the non-misleading ones – are a priori prohibited. By underexposing the probability of beneficial effects, few claims pass assessment and adoption. Consumers are asymmetrically informed.

The grading of beneficiality must not be mistaken for the grading – forecasting – of cause and effect, which concerns the pure and simple demonstration of the extent to which a relationship exists between a foodstuff and a claimed effect. It is the weighing of the totality of the evidence of effect as convincing, probable, possible or insufficient. Weighing the beneficiality of the forecast effect is a different thing, it’s something that now belongs primarily in the domain of the Union’s public health managers. After the cause and effect relationship has been weighted and graded, they must weigh the beneficiality of the weighted effect with public health as the benchmark.

VII. Transforming research traditions

When a foodstuff’s effect has been graded as possible or probable, this does not at all preclude that, in grading itsbeneficiality, public health managers would nevertheless welcome such claims as very or sufficiently beneficial and therefore acceptable for use in commercial communication. After all, when it is likely that a foodstuff enhances one or more essential biochemical pathways in growth, development or function, or reduces the influence of a serious disease risk factor, its relative beneficiality for human health seems undisputable. Nothing should then prohibit the use in commercial communication of a claim that correctly informs consumers about the likelihood of the health benefit.

This might be the reason why the Community legislature proposed – at least in theory – the same probabilistic approach to the assessment and management of benefit as the one it standardly applies in the case of risk. This approach modifies the absoluteness of benefit and moderates the demand for convincingness of beneficiality. Still, in the real world, including in particular the real world imposed on us by the European Union and EFSA, while risk is routinely approached in terms of probability, health, especially in the context of the Claims Regulation, is predominantly assessed and framed in terms of certainty, not of probability. In practice, asymmetry and prejudice rule.

The issue of analogy is quite elegantly addressed in Nutrition and health – transforming research traditions.[vii] In their article, authors Jaap Hanekamp,[viii] Aalt Bast[ix] and Edward Calabrese[x] state:

“… Food is not medicine in terms of its direct (pharmacological) impact, and fortunately so. A different approach is required that better captures the polyvalent nature of most nutrients, the numerous small, potential beneficial, physiological changes within the defined range of biological plasticity occurring upon exposure to certain food-components such as polyphenols. Ironically, toxicology does give us tools to better explore and quantify these small, potentially beneficial, physiological changes. The full range of these tools gives us the means to mimic the risk assessment procedure the EFSA only partially framed. But we should be mindful that ‘benefit’ is not the opposite of ‘risk’, as incorrectly suggested by EFSA. Etymologically, risk is a general term that captures both opportunity and danger (Althaus, 2005).

Now, it is a relative small step to construct an outline of a benefits-analysis that is analogous to the well-known risk assessment procedure. Defining ‘risk’ as capturing both opportunity and danger is in line with biphasic dose-responses that seem to underlie the connection between food and health: too little or too much of a ‘good thing’ is detrimental to health, whereas the optimum intake of that same ‘good thing’ confers benefits. As a result, the traditional risk assessment procedure can simply be partitioned into two analogous procedures: risk-‘as-danger’ and risk-‘as benefit’.

Indeed, the latter would be quite logical to characterise as benefit = intrinsic health capacity × exposure. Such a definition does justice to probabilism that so dominates the risk-‘as-danger’ assessment paradigm and should be central to the benefits assessment. Ironically, despite EFSA’s commonplace description of the risk assessment procedure with its well-known probability statements, for benefit, as a conservative means to protect public health, ‘the manager frequently requires the evidence to be convincing.’ (Scientific Opinion EFSA, 2010)

This is remarkable and illogical on multiple levels: (i) it essentially forgoes, in EFSA’s terms, the ‘mirroring’ of the risk assessment paradigm, whereby; (ii) it simply and without evidential basis proffers the notion that, in order to protect public health, potential benefits from certain foods and food-components should not be rated in terms of probability but in terms of scientific absolutes, whereby; (iii) it highly overestimates the potential of science to be straightforwardly ‘convincing’ in its factfinding, and inconvertibly; (iv) makes all food-advice, including coming from the EFSA itself, contentious, and finally; (v) ad absurdum, makes virtually all research results within nutritional science, or any other scientific field for that matter, moot. Things are aggravated even further by the limitation of scientific methodology designated as some kind of gold standard.”

VIII. Revising the text of Article 7 of Regulation 178/2002/EC

The beauty of ‘transformed research traditions’, is that they also open the door for a symmetric and analogous approach in disease risk reduction research (health risks) and research concerning growth, development and function (health benefits). It not only provides the common scientific ground for assessing and managing health risks and health benefits, but also for researching, assessing and managing Article 13 and Article 14 claims on an equal footing. The effect of a foodstuff in disease risk reduction can now be measured and evaluated in the same way as measuring the effects in the enhancement of growth, development and function.

The new approach gives us credible scientific ground for making “inverse” precaution possible in communicating benefits well ahead of the moment when evidence has become convincing and conclusive. Assuming that by “inverting” precaution the inversion of the burden of proof, so firmly embedded in the Precautionary Priciple, would not be lost, the maker of the “inverted” claim would have the protection of the benefit of the doubt. The critics would have to convincingly prove the claim’s falsity. If they do, the benefit of the doubt can be ended and the health claim can definitely be refused.

Conversely, if this precautionary approach to benefits would be denied to makers of claims that, in conformity with Regulation 353/2006/EC, truthfully expound the extent to which a product produces a beneficial effect, that denial should also hit the precautionary approach to risk. Since the Precautionary Principle has become part and parcel of the Union’s policies and policy making, and since it found unconditional expression in the “Verbotsystem mit Erlaubnisvorbehalt” that is the legal nucleus of the Claims Regulation, there is no reason why the Union would refuse to apply “inverse” precaution in the management of health benefits.

The effect of the Precautionary Principle Article 7 of Regulation 178/2002/EC, is that all foods placed on the market are intrinsically very safe under normal conditions of consumption. An important side-effect of Article 7 is that nutrition and health claims placed on foods can only cause a negligable residual risk on harmful effects on health.[xi]Whether they do must be established on a case-to-case basis.

If the European Union would symmetrically and indiscriminately apply the Precautionary Principle, not just to precaution the possibility of suffering harm or loss, but also to permit, following an assessment of available information, the possibility of enjoying ‘uncertain’ benefits or gains, we would have numerous nutrition and health claims stating, suggesting or implying that a possible or probable relationship exists between a food category, a food or one of its constituents and health. To make this happen, all that is required is the changing of a few words in the text of Article 7 of Regulation 178/2002/EC and the addition of that revised text as paragraphs 3 and 4 to this Regulation to make this happen:

‘Article 7 – Regulation 178/2002/EC
Precautionary principle applied to possible benefits
3. In specific circumstances where, following an assessment of available information, the probability or possibility of beneficial effects on health is identified but scientific uncertainty persists, provisional benefit management measures necessary to ensure the high level of health chosen in the Union may be adopted, pending further scientific information for a more comprehensive benefit assessment.
4. Measures adopted on the basis of paragraph 1 shall be proportionate and no less permissive of trade than is required to achieve the high level of health chosen in the Union, regard being had to technical and economic feasibility and other factors regarded as legitimate in the matter under consideration. The measures shall be reviewed within a reasonable period of time, depending on the nature of the benefit to life or health identified and the type of scientific information needed to clarify the scientific uncertainty and to conduct a more comprehensive benefit assessment.’

If the Precautionary Principle would be indiscriminately applied to risks on harm and “risks” on benefits, we would as many “risks” on benefits as we now have risks on harm.

Bert Schwitters