2. WORKING GROUP I - ESSENTIAL PRINCIPLES, DEFINITION AND SCOPE GOVERNING THE PROVISION OF SCIENTIFIC ADVICE

2.1 Scope and definition of scientific advice

The Group discussed at some length the meaning of advice. Dictionary definitions of judgement, opinion, analysis and advice were obtained to assist in furthering the discussion. There was agreement that advice was more than information or just opinion. Advice could include one or more options reflecting the potential outcome of different risk management options. It was agreed that the purpose of providing scientific advice was to help in decision making, often by risk managers or policy makers, but also possibly by others. It was recognized that advice is usually sought by a client, e.g. Codex, or a Member country of FAO or WHO. The advice provided should reflect the needs of the client. However, effective dialogue with the client is essential to ensure that the question asked provides advice of a nature and form that would help address the issue of concern most effectively.

The Group discussed whether the output should be named differently, depending on the level of confidence in the conclusion. It was agreed that all output should be considered to be scientific advice, but that the nature of the output would vary, reflecting the degree of uncertainty. The Group discussed various definitions of scientific advice that incorporated these concepts, and finally agreed on a formulation, together with the purpose of scientific advice.

The scope of scientific advice was then discussed. The Group recognized that advice could be sought on a wide range of issues with different levels of urgency, and that the adequacy of the information available could vary with the request and with the urgency of the need for advice. The assessment could be limited to data available in a very short time, when the need for advice was urgent. Clearly, such advice would have a greater degree of uncertainty than when a full assessment was carried out. The possibility of identifying advice provided under such circumstances as "preliminary" or "provisional" should be considered. Scientific advice was often requested on issues that had a direct or indirect impact on human health. However, the Group agreed that the scope should include the possibility of addressing issues relating to environmental health or fair practices in the food trade.

The scope of scientific advice should be broad enough to range from simple responses to specific questions to a complete risk assessment. While advice was often sought prior to risk management, the Group considered that it was appropriate to include the option for provision of advice anywhere throughout the risk analysis procedure. The Group recognized that risk analysis is an ongoing process and that this advice may also include effective impact assessment of risk management strategies.

The issue of scientific opinion was raised. It was noted that this was not a term used within the FAO/WHO process, but that it was relatively widely used elsewhere. However, it was also noted that such use varied considerably, and that in some instances it was used synonymously with scientific advice.

Scientific advice is the conclusion of a skilled evaluation taking account of the scientific evidence, including uncertainties. It may comprise an appraisal of the consequences of one or more options based on an analysis of the available scientific knowledge and on scientific judgement. Such advice should include explicit recognition of any uncertainty either in the current state of knowledge or in the adequacy of the available data. If necessary, it should include any alternative interpretations of the data.

The purpose of scientific advice is to help risk managers, policy makers and others in decision making.

Scope of scientific advice. Advice may take many different forms, from a response to a specific question, or provision of scientific information related to specific needs, to a full quantitative risk assessment. Depending on the degree of uncertainty, advice may range from a clear conclusion on risk to a recommendation to obtain additional data. Advice may be sought at any time throughout the risk analysis process or even subsequently.

Optimal advice requires effective dialogue between risk assessors and risk managers. In some instances advice may include conclusions regarding the need for information and activities necessary to enable capacity building or technical assistance.

2.2 Essential core principles for the provision of scientific advice

The Group discussed all of the core principles raised in the e-forum, and identified one or two additional potential principles. Each of the principles was evaluated to ascertain whether it was essential to the effectiveness of the process, and whether it was a primary or an enabling principle (i.e. a means to achieve a primary principle). The question of duplication was addressed, and on this basis a number of the principles were combined. Principles were also assessed to establish whether they applied to the participants, the process or the advice itself. The Group discussed the possible distinction between those principles that would advance the quality of the advice scientifically, and those that would contribute to improving the credibility of the process. Transparency was the principal subject of discussion. One view was that transparency was a means to an end, to ensure soundness and responsibility, and hence should not be a core principle. An alternative view was that transparency should require complete access and comment at all stages of the process. However, the views of the Group converged, and it was agreed that transparency should be a core principle, but that its description should not be too prescriptive. It was agreed that it was more the process than the information that needed to be transparent.

The Group concluded that it was possible to identify six core principles which incorporated all relevant considerations, and that it was not necessary to subdivide these in any way: soundness, responsibility, objectivity, fairness, transparency, inclusiveness. The Group then discussed which concepts should be covered by each principle.

It was recognized that in some instances it might not be possible to apply all of the principles proposed, and agreed that the principles should be applied to the extent possible with respect to the nature of the advice sought, and that the reasons for any deviation from these principles should be documented as part of the process.

The Group did not reach agreement on whether excellence in communication should be a core principle. However, there was general agreement that excellence in communication was necessary in all steps of the risk analysis process and between all parties.

Soundness is the need for scientific excellence, and applies to both the participants and the process. It includes consideration of adequacy of competence, recognized standing in the discipline represented, the ability of opinions and advice to withstand scrutiny by peers, the application of current scientific knowledge in reaching a conclusion, and ensuring that those producing advice represent a suitable balance of expertise.

Responsibility encompasses the various aspects of accountability, and applies both to the need to safeguard the integrity of the process and to consider scientists answerable for their views. This includes the responsibility of participants to justify their views by adequate citation of reputable sources, the application of a suitable level of caution in data interpretation, timeliness, compliance with agreed task description, efficiency in conducting the assessment, cost-effectiveness of the process and maintenance of confidentiality as mandated. An additional aspect of responsibility is that sponsors should be asked to submit all appropriate data, and not just those necessary to comply with the data requirements. Ideally, responsibility should include updating scientific advice on the basis of new knowledge (i.e. review of conclusions).

Objectivity is considered to include neutrality and applies both to the participants and to the advice provided. While participants may be drawn from different sectors, including those where there may be potential conflicts of interest, wherever an opinion is provided it should be independent and unbiased. Both the opinions of individuals and the advice provided should be based only on scientific evidence. In reaching a conclusion it is necessary to balance the opinions from participants with different perspectives and to seek a scientifically-based consensus. The views expressed should be weighted according to the degree of certainty underpinning them. Where scientific advice is the outcome of a risk assessment there should be adequate separation from risk management.

Fairness applies to the conduct of the assessment process, and requires respect of all participants for each other and for their scientific views. Participants should be given adequate and equal opportunities to express their views. Minority views should be properly considered. Participants themselves should contribute appropriately to the process. The selection of participants should be objective and inclusive to the extent possible. The process should be conducted in an ethical manner.

Transparency involves the design and implementation of mechanisms that ensure that the process whereby advice is formulated and that the advice itself is clearly understandable to others. Transparency could involve the provision of access to pivotal scientific information that is comprehensive, understandable and timely, while respecting legitimate concerns to preserve confidentiality. It could also involve the provision of explicit documentation of all procedures, policies and practices. Transparency may also involve review of both the advice and the procedures involved in providing advice.

Inclusiveness is considered to include group balance. Two aspects of inclusiveness were identified: minority scientific opinion and the balance of skills and expertise necessary for the assessment. Inclusiveness requires that due respect and consideration be given to minority scientific opinion. In the selection of participants, in addition to their expertise, due consideration should be given to geographical and socioeconomic balance, but not to the extent that it compromises scientific integrity. Particular emphasis should be placed on improving the participation of developing countries. Where participation is limited by a skill or knowledge gap, appropriate capacity building activities should be undertaken.

2.3 Advantages of developing a set of core principles

Development of a set of core principles and their adoption will better guide the work of FAO/WHO in producing scientific advice by ensuring the integrity, consistency, quality, transparency and accountability of the process. This will help improve the credibility and utility of the advice.

2.4 Mechanisms to enhance participation of developing countries

The Group recognized that effective participation of developing countries in the provision of scientific advice is a prerequisite for the full application of the core principles, and emphasized the principle of balanced participation of developing countries (e.g. data provision, expert group participation). Appropriate specialists in specific areas exist in developing countries, and they should be identified and given the opportunity to participate in relevant assessment activities. In the least developed countries, however, there is often a smaller research community and limited scientific infrastructure, leading to scientific isolation. Effective capacity building is essential to ensure full participation in the future. Specific principles relating to this issue have been considered further under the relevant headings below.

2.5 Harmonization of terminology, methodologies and outputs

It was agreed that harmonization of terminology, methods, inputs and outputs for the provision of scientific advice should, to the extent possible, be a priority for FAO/WHO. Harmonization should not be perceived as standardization but rather as an understanding of the methods and practices used for a variety of scientific advice activities by various countries and organizations, so as to develop confidence in an acceptance of assessments that use different approaches. Achieving harmonization of approaches will afford a number of opportunities, including an improved framework for elaborating credible science, comparing information on risk assessments, understanding the basis for exposure standards, and promoting transparency, effective communication and capacity building. Harmonization is particularly important for developing countries in order to enable them to understand those issues that affect food safety and help promote fair practices in the food trade. Harmonization would be especially important insofar as the issues on which scientific advice is sought become more complex, e.g. in risk comparisons, risk-benefit analyses or integrated risk assessments.

The Group agreed that harmonization was highly desirable and should be given a high priority. However, it should be recognized that complete harmonization would not be possible across all areas. While the Group recognized that there were several ongoing initiatives tending towards harmonization through vertical integration (e.g. FAO/INFOODS on food composition databases; UNU/FAO harmonization of methodologies on nutrient requirements), it considered the need for horizontal integration as an important priority (e.g. the International Programme on Chemical Safety (IPCS) programme on the harmonization of approaches to the assessment of risk from exposure to chemicals from all sources).

2.6 Potential establishment of new programmes

The Group considered this under three different headings. Firstly, the areas needing additional activity within FAO/WHO were discussed, including topics already being addressed in an ad hoc manner and those where other organizations had perceived a concern. Secondly, the issue of criteria for establishing a new area in which scientific advice might be needed was reviewed, including mechanisms both for identifying such areas and for pursuing them within FAO/WHO. A third topic discussed was foresight and self-tasking. The Group agreed that there was a need to be much more proactive and that sufficient resources should be available to enable established expert groups within FAO/WHO to identify and place critical items on their agendas, including general items.

Vertical and horizontal issues identified as requiring further action, refinement or improvement within FAO/WHO, perhaps requiring a new programme, were for example.:

Vertical issues

• Nutrition (general principles)
• Biotechnology products, including GMOs (or as a separate activity)
• Novel foods

Horizontal issues

• Risk-benefit and risk-risk assessment. This activity might include consideration of nutrition or other benefits versus adverse effects, the criteria for health benefit claims, and the extent to which such assessments should take into account wider implications (e.g. socioeconomic).
• More holistic assessments. These might include consideration of multiple routes of exposure, exposure to mixtures, multifactorial (including polygenic) susceptibility to potential adverse effects, proposals for and the evaluation of the review/updating of scientific advice (which could include surveillance, periodic review and analysis of the impact of risk management decisions).

Mechanisms for identifying the scope of new programmes should be established. These might include electronic fora. Criteria for establishing new areas in which to provide scientific advice might include timeliness of the issue, the absence of adequate information on the consequences of exposure (i.e. perceived risks), the weight of evidence for concern (based on health surveillance, epidemiology, experimental studies), and the potential impact that advice provided in this area might have. These criteria should be subject to consideration at a consultation, at which time the need for a formal evaluation process and the nature of this process should be agreed. Foresight and self-tasking should involve regular horizon scanning by established expert groups, and possibly by other groups convened for this purpose. Some formal mechanism should be established to evaluate the output of this exercise, and to consider prioritization and resourcing of new areas. This process should involve established expert groups, enabling a degree of self-tasking.