Глобальный форум по продовольственной безопасности и питанию (Форум FSN)

CONTRIBUTIONS

The questions posed to guide this consultation are:

1. When you think about advancing an SPI for agrifood systems in your country, what is the greatest challenge that the FAO guidance, such as presented here, can help address? What suggestions do you have to make the guidance more practical and useable at the country level?
2. Are the sections/elements identified in the draft guidance the key ones to strengthen SPIs at the national level? If not, which other elements should be considered? Are there any other issues that have not been sufficiently covered in the draft guidance? Are any sections/topics under- or over-represented in relation to their importance?
3. In order to make the guidance as concrete as possible, we are including numerous boxes/cases studies on real-life use cases. In this context, please contribute 300-450 words on examples, success stories or lessons learnt from countries that have/are strengthening SPIs for agrifood systems, including addressing asymmetries in power, collaboration across knowledge systems, connecting across scales, capacity development activities and fostering learning among SPIs?
4. Is there additional information that should be included? Are there any key references, publications, or traditional or different kind of knowledges, that are missing in the draft and which should be considered?

There is a great deal in this draft concerning agrifood systems’ science-policy interfaces that deserves to be welcomed, including the recognition that the scientific basis of advice to policy makers is often characterised by “high levels of uncertainty”, and that “A narrow view of what counts as evidence favors specific expertise over others, and a wide range of evidence remains undocumented and unpublished, leading to bias.”

Moreover, the text recognises that many of the most important science-policy interactions occur within official institutional bodies, and that those bodies have often been designated as having responsibility for ‘risk assessments’, which are portrayed as ‘scientific’, and which are delivered to official policy decision-makers, or ‘risk managers’. Those risk managers are expected to integrate the scientific risk assessment advice with ‘other legitimate factors’ when making agrifood policy decisions. Those other legitimate considerations include for example an understanding of what is affordable, practicable and acceptable.

Questions 2 and 4: the importance of risk assessment policies

This response primarily addresses questions 2 and 4 above; it identifies an important gap in the draft and provides the missing element to fill that gap. The missing element that should be taken into account can be found in the rules of a UN Food & Agriculture Organisation body that the draft did not reference; the body in question is the international food standards body, the Codex Alimentarius Commission (or CAC).1

The CAC has made an important contribution to our understanding of the interactions between scientific and policy considerations in agrifood standards-setting, with important implications for what is needed to strengthen both national and international science-policy interface institutions with responsibility for agrifood systems. The key provisions introduced by Codex refer to ‘Risk Assessment Policy’, which is the term used to refer to the policy- judgements that frame the deliberations of scientific advisory risk assessment bodies, and which consequently frame the advice provided to policy-makers.

In 2003, the Codex Alimentarius Commission introduced an innovative provision into its Procedural Manual which was directly concerned with the interface between science and policy-making. The innovative text referred to what it characterised as ‘Risk Assessment Policy’ in the following terms:

• Determination of risk assessment policy should be included as a specific component of risk management.
• Risk assessment policy should be established by risk managers in advance of risk assessment, in consultation with risk assessors and all other interested parties. This procedure aims at ensuring that the risk assessment is systematic, complete, unbiased and transparent.
• The mandate given by risk managers to risk assessors should be as clear as possible.
• Where necessary, risk managers should ask risk assessors to evaluate the potential changes in risk resulting from different risk management options.2
  • what is to count as a relevant risk?

The implications of this innovation were profound although their significance was not, and is not yet, widely appreciated. Initially the provisions applied just to the way in which Codex scientific advisory bodies, such as the Joint Expert Committee on Food Additives3 and the Joint Meeting on Pesticide Residues4, interacted with the Codex risk management bodies that they advise, namely the Codex Committee on Food Additives and the Codex Committee on Pesticide Residues.5

The importance of the provisions regarding risk assessment policy were substantially enhanced at the July 2007 plenary meeting of the Codex Alimentarius Commission when a text on the Working Principles for Risk Analysis for Food Safety for Application by Governments was formally adopted.6 Under the provisions of that agreement, and in a section headed: Proposed Draft Working Principles for Risk Analysis for Food Safety for Application by Governments, all Codex Member States, and regional jurisdictions like the European Commission, in its interactions with the European Food Safety Authority, accepted the obligation for their domestic food safety risk managers to provide their scientific risk assessors with explicit risk assessment policies prior to the start of the deliberations of those risk assessors.7 The text the Codex Member States adopted on ‘risk assessment policy’ is identical to that in the 2003 13th edition of the Codex Procedural Manual, cited above at reference 2.

Whether or not the CAC and UN Member States fully appreciated the implications of those RAP provisions, they have all, at least implicitly, acknowledged that scientific advice on agrifood risks cannot be fully separated from risk policy-making considerations.

The change accomplished by the adoption of those provisions constituted to a repudiation of an orthodoxy that had prevailed since the mid-1980s. That orthodoxy emerged in the USA in 1983 in a report from the US National Research Council called Risk Assessment in the Federal Government: Managing the Process. That report was published with a bright red cover, and consequently it has often been referred to as the Red Book model.

The Red Book model: science first, policy-making second (the scheme please find in the attachment)

A defining feature of this model is that, while it acknowledged that both scientific and non- scientific considerations are indispensable ingredients to policy-making, it portrays the science as influencing policy-making, but with the direction of influence going only one way; it portrays science as if it were entirely independent of all policy considerations.

The adoption of the RAP provisions by the CAC and by all of its Member States constituted a repudiation of the Red Book model and, at least implicitly, by the acceptance of an alternative that can be represented graphically in Figure 2, which is called a co-dynamic model.

Figure 2: Co-dynamic model: scientific risk assessments framed by risk assessment policies (the scheme please find in the attachment)

The co-dynamic perspective recognises that conducting risk assessments for official policy- making bodies routinely entails presupposing a set of evaluative judgements. They relate, for example to:

• what kinds of evidence should be deemed relevant?
• how much of which kinds of evidence should be variously deemed necessary or sufficient to justify advising risk managers that some product or process should be permitted, restricted or banned?

For example, when the Joint (WHO and FAO) Expert Committee on Food Additives sets an ‘acceptable daily intake’ for a food additive, it is making an assumption about ‘acceptability’, which is an irredeemably evaluative consideration about how much evidence to require and/or accept.

The co-dynamic model also acknowledges that interactions between scientific and policy considerations are reciprocal, rather than unidirectional. The Codex RAP provisions also acknowledge that, while scientific and policy-considerations reciprocally interact, their separate contributions can and should nonetheless be explicitly identified and acknowledged by accountable institutions following transparent decision-making processes.

If the Codex RAP provisions were implemented, then the resultant changes could substantially contribute to meeting many of the aims and objectives of the Draft Guidance. It could, moreover, substantially enhance both the scientific and political legitimacy of the processes by which agrifood policies are decided.8

A study of the risk assessment policies that prevailed in six different jurisdiction, which was supported by the European Commission’s Joint Research Centre’s Institute for Prospective Technological Studies, comparatively reviewed four agrifood debates across those six jurisdiction. That study reported that, notwithstanding the wide diversity of RAP assumptions, in all cases those RAPs included at least 3 distinct types of up-stream framing assumptions, namely those concerning substantive, procedural and interpretative issues, although they were also often inter-dependent.9

Substantive RAPs are firstly concerned with delineating which potential changes and effects are to be included within the scope of risk assessments and which are outside their scope, but also judgements about which kinds of evidence are admissible and which are not. For example, when the risks posed by food additives are considered should they focus solely on toxicological issues, or should they be extended also to consider possible impacts on public health nutrition?

Studies have recently been increasingly indicating that consuming chemical emulsifier additives can have adverse effects on intakes of oils and fats, gut microbiomes and nutritional public health by increasing the risk of Type II diabetes.10 But official risk assessments of the putative risks of emulsifiers, singly and collectively, have been and remain confined to toxicological considerations, to the exclusion of all other possible types of adverse consequences. That is a substantive RAP assumption, and an increasingly problematic one.

Procedural RAPs are concerned with characterising the processes by which risk assessments are conducted and reported. For example, should risk assessment deliberations be conducted in open or closed meetings, and how should risk assessors respond to conflicting evidence and uncertainties?

Interpretative RAPs are concerned with the ways in which data are interpreted. Data and documents do not interpret themselves. In policy contexts, interpretation routinely involves value judgements and assumptions. For example, are laboratory rodents treated as good or as poor models for the effects of chemicals on humans – and for all types of adverse effects at all sites, or only for some? Should equal weight be assigned to identifying and evaluating putative false positives and for false negatives, or should one type receive preferential consideration?11

Furthermore, the processes by which risk managers should establish their RAPs, should be completed before particular risk assessments commence. Deliberations on selecting the RAPs should also be conducted in consultation with risk assessors and all other interested parties. Those procedures should help ensure that the risk assessments are systematic, complete, unbiased and transparent.

Despite the importance and ubiquity of such risk assessment policy assumptions and judgements, the commitments made by the CAC in 2003 and all UN Member States in 2007 are only rarely fulfilled.12 That is important because if they were, that would substantially contribute to enhancing both the scientific and political legitimacy of official agrifood policy decisions.13

In practice, what frequently happened, and frequently continues to happen, is that risk assessment bodies, which are composed of scientists and their official secretariats, take selected RAP assumptions for granted, without explicit engagement with the risk managers who will be the recipients of their advice, or the numerous ‘other interested parties’ that should have had an opportunity to comment on those RAPs. In other words, the RAPs that are adopted are routinely chosen in opaque and unaccountable processes, often behind closed doors. Moreover, they are so closely integrated in the deliberations, reports and recommendations that are provided, that those evaluative RAP assumptions are routinely misrepresented as if they were purely scientific.14

Given that the draft Guidance on strengthening national science-policy interfaces for agrifood systems fails to acknowledge the risk assessment policy requirements that have been established by the Codex Alimentarius Commission, I strongly recommend that the Guidance be revised to include references to those requirements as their implementation could significantly strengthening national science-policy interfaces and enhance the scientific and political legitimacy of agrifood policy regimes.

Conflicts of Interest

A distinct set of considerations relates to the challenges posed by the risk that scientific risk assessors might have commercial conflicts of interest. It is not always possible to discover what conflicts of interest prevail, because in many jurisdictions that information has never reached the public domain. A recent study in the UK, however, where declarations of conflicts of interest have to be publicly disclosed (although contractual details and levels of payments remain undisclosed) reported that there was extensive evidence of conflicts of interest across the UK’s entire range of official food and agriculture advisory and policy- making bodies.15 The rules governing scientific advisors to the UK government stipulate that conflicts of interest should be declared, and those declarations published. Evidence shows however that declarations are not sufficient to ensure that they don’t influence judgements, recommendations and decisions.16

By contrast, a 2021 report to the UN General Assembly from the UN Human Rights Council entitled Right to science in the context of toxic substances, said: “In order for the science that forms the basis of policy to be trusted, conflicts should be avoided rather than simply managed through disclosure processes.”17 (emphasis added) That stipulation should also be incorporated into a revised Guidance on strengthening national science-policy interfaces for agrifood systems.