4. WORKING GROUP III - PROCEDURES AND MECHANISMS FOR THE PROVISION OF SCIENTIFIC ADVICE

The general approach taken was to group the various numbered items under the following four general headings:

1. Selection and effectiveness of experts
2. Data
3. Processes and procedures
4. Communication of scientific advice

Under each of the four general areas the Group prepared an extensive list of topics for possible further discussion. In developing its report, the Group made a concerted effort to identify the positive and negative implications of the recommendations.

4.1 Selection and effectiveness of experts

Considerations

Selection process and criteria. The Group considered that it was important to select experts not only on the basis of their scientific achievements, but also to consider experience in risk assessment (national, international) and knowledge of food safety issues. In addition, proven effectiveness in work with expert bodies and ability to contribute effectively to group work are required, and in order to achieve a balanced representation in the expert body, a broad range of expertise and perspectives. The omission of these elements might compromise quality and acceptance of the expert body's analysis and advice even if the expert body is composed of distinguished scientists.

A set of criteria for the selection of experts should be identified and harmonized across different expert bodies as appropriate. The selection process should be transparent, including dealing adequately with conflicts of interest. The determination of whether a conflict of interest exists should rest with FAO/WHO. Information on organizational affiliation, government service, research support, public statements and positions, financial interest and other interests (e.g. professional affiliations) should be provided. This information, as appropriate, should be available publicly before the meeting (e.g. via Internet), consistent with rules of privacy. It should be made a condition of participation that all interests relevant to the subject matter of the meeting would be publicly disclosed. This information should include a statement that experts are to be contacted only through the secretariat. Experts should inform the secretariat of any attempts to contact them direct. The current disclosure forms should be amended accordingly.

Efforts should be made to increase participation of experts from developing countries by targeted efforts to identify experts, provision of financial or other support (infrastructure) and specific training programmes. This could be supported by creating a network of competent scientific institutions. Other means would be fellowship or visiting scientist programmes to FAO/WHO which establish and maintain a network of such institutions.

Training. FAO/WHO should make available training opportunities for experts in the following areas: writing, communication, English as a second language, the ability to recognize conflicts of interest, framework of work by expert bodies (roles of the advice, key institutions), working procedures, leadership, working/facilitation techniques.

In addition, training is needed to enable experts to better recognize value judgements made during expert meetings, and to better distinguish between risk assessment and risk management.

Recognition and reward. FAO/WHO currently have limited ability to recruit experts worldwide. At present, expert panel activities are not regarded in the same way as scientific peer-reviewed publications, and this reduces the attraction of participation, particularly by younger scientists. FAO/WHO should work with academic institutions to encourage them to recognize expert meeting participation and outputs as equivalent to other scientific achievements. Letters and official documents should be issued by FAO/WHO to the expert and expert's institution documenting the value of the work and its equivalence to peer-reviewed work and publications. Institutions that provide experts should be publicized (e.g. on the Internet). Increased constraints on institutions which release experts may require that FAO/WHO consider compensating experts or their institutions, particularly if their involvement covers an extended period, or if they would otherwise be deprived of salary. FAO/WHO should recognize expert resource contributions from Member countries as equivalent to financial contributions. This will have implications in terms of the fiscal needs of FAO/WHO.

Language as related to selection of experts. It was recognized that meetings to develop scientific advice for a global audience are typically held in English only, which puts experts who do not speak English at a disadvantage, or even excludes them. The ideal would be for all languages to be accommodated. Practical considerations, however, are likely to limit severely the ability to conduct scientific meetings in multiple languages. The availability of expert interpreters who are sufficiently trained in the sciences to achieve accurate simultaneous interpretation is severely limited, and the need for interpretation was viewed as significantly limiting the speed at which expert panels can proceed. This has a negative impact on the working structure/informal communication needed to achieve effective deliberations. However, inclusion of experts with specific knowledge may on occasion warrant the need to support them with interpretation services. FAO/WHO should alternatively consider providing experts that do not have English as their mother tongue with access to training that would enable them participate more efficiently, not only as experts but also in other roles (e.g. rapporteur, chairperson). This would have consequences for the structure of expert panels, favouring one where at least a portion of the experts serve as standing members for a specified period of time.

While the limitations described above are likely to restrict the conduct of expert panels to English for the foreseeable future, FAO/WHO and the panels themselves should make every effort to not allow language to limit the range of data and publications considered in their deliberations.

Inclusion of observers and stakeholders in expert bodies. The inclusion in expert bodies of observers/stakeholders, or others not traditionally invited as scientific experts, raised a number of issues that require careful examination. These included (1) costs of participating might preclude some interested parties from attending, resulting in biased representation of stakeholders; (2) the need to maintain balanced representation on expert bodies; (3) potential enhancements through the addition of communication and "plain language" experts; and (4) development of a code of ethics regarding improper influencing of the decision-making process.

A variety of mechanisms are available to improve the openness of the process including meetings, and there is a need to conduct an additional analysis of the advantages and disadvantages of various approaches. During this analysis, two principles that should be considered in assessing the appropriateness of various approaches are: (1) ensuring balanced representation of different interests and regions; (2) communicating transparently the reasons for inviting or not inviting non-experts. Associated with those principles, the analysis should consider whether NGOs could be represented by experts or non-experts, whether observers should be nominated by their organizations and/or by the secretariat, the desirability of including a communications or "plain language" expert on expert panels, and the inclusion of standing representatives of specific interest groups such as consumers and industry.

Secretariats of expert bodies. There was a need to consider whether current resources were adequate to support the activities of the expert panels. It should be recognized that strengthening procedures may modify the requirements and tasks of the officers of the secretariat which may have training, budgetary and staffing implications. The role and mandate of the secretariat in general and for a specific expert body needed to be reviewed, defined and communicated.

Key findings

The Group noted that there was a need for clear, harmonized, publicly available selection criteria for the appointment of expert bodies, and noted that the training of experts in the roles that they would play on an expert panel was of paramount importance, followed by skill base and communication abilities. The Group identified a lack of contributions by experts from developing countries as a key issue needing resolution. It was also noted that there was a lack of recognition for the contributions provided by experts, and a potential need to compensate experts or their institutions for their involvement in expert bodies was identified. This was particularly important if the involvement covered an extended period, or if experts would be deprived of salary. Potential mechanisms for the inclusion of non-experts and/or observers at expert meetings should be considered.

4.2 Data

Considerations

Relevance of data. Data should be adequate for the purpose of scientific evaluation and the provision of scientific advice. Where possible, harmonization of data requirements by international and national bodies should be promoted (e.g. the minimum number of pesticide trials).

Data quality. Data used as the basis for the provision of scientific advice should be of acceptable quality and well documented (e.g. following GLP and QC/QA). Data should also be timely and representative of populations and subgroups as well as of geographical distribution, to reflect the actual situation. In those instances where data of insufficient quality were the only data available, they should be so identified and used by taking account of the uncertainty and limitations of the data.

Sources of data. FAO/WHO should encourage and support submissions from developing countries, and develop means to overcome:

• the lack of understanding in developing countries of both the need to provide data and the types of data required;
• deficiencies in the infrastructure and mechanisms in developing countries to generate the data necessary for the provision of scientific advice by FAO/WHO;
• the limited facilities, personnel, and operating funds.

In order to overcome these barriers:

1. Codex Contact Points should disseminate information to researchers and other stakeholders in their countries related to scientific advice from Codex bodies and from FAO/WHO, e.g. calls for data.
2. Training of Codex Contact Points and scientists to be involved in risk assessments is needed through regional offices of FAO and WHO, by utilizing existing networks such as GEMS/Food, Salm-Surv, etc.
3. Sustainable funding should be available to establish and support required facilities, to develop human resources, and for operational expenses.
4. Twinning between a developing country and an experienced country could be useful in data generation.
5. Interaction between food, clinical and epidemiology experts should be encouraged to promote generation of data.

Submission of data. Data providers often may not be aware of when and to whom to submit data. In addition, details of how to submit data (format and media) are not known:

1. The submission of data should be in electronic format and hard copy as required.
2. Analytical data on chemical contaminants in food should be submitted in the GEMS/Food format.
3. A format for the submission of data related to microbiological risk assessment should be developed.
4. Issues related to trade-sensitive data need to be overcome. For example, data to only be identified by region and not country.

Confidentiality. FAO/WHO should ensure that data are well secured before, during and after the evaluation. Guidelines for handling such data should be developed.

Ownership of data. The use of data by FAO/WHO and their expert panels must be consistent with the precepts of intellectual property rights. This issue should be addressed by appropriate international mechanisms, such as those of the World Intellectual Property Organization (WIPO).

Ethical considerations. Data involving human subjects should be handled so as to respect the right of the subject to informed consent and privacy. Studies involving human subjects should be reviewed by relevant authorities, such as WHO and national bodies.

Key findings

The Group noted that the key issues concerning the adequacy of data were source and relevance. Many data are derived from industry sources and the Group noted that there were substantial differences among the various expert committees in accepting such data. There needed to be better harmonization among the scientific advisory groups, which was likely to require development of transparent inclusion/exclusion criteria for judging the suitability of data. Limited communication infrastructures, training, and potentially fiscal support were recognized as significant limiting factors in the increased acquisition of scientific data from developing countries. The Group also recognized the need to establish clear yet flexible data quality assurance criteria. This included the need to harmonize these approaches with other international, national and professional guidelines for assuring data quality. Such guidelines were likely to vary among different disciplines. For example, toxicological studies used to provide scientific data are generally conducted under GLP, while microbiological studies more often rely on use of validated standard analytical methods. The Group noted that Codex, FAO and WHO could play a much more prominent role in the collection and dissemination of scientific data, especially through their Contact Points in developing countries.

4.3 Processes and procedures

Considerations

Procedures for enhancing the participation of developing countries. There needed to be enhanced processes in place to ensure effective communication with experts, data providers and users in developing countries. This could be achieved by using languages other than English in such communications. Mechanisms of communication other than e-mail and Internet need to be used, particularly in developing countries where such modes of communication may be limited.

FAO/WHO should make better use of their regional and country offices, and the Codex Contact Points should improve the dissemination of information in developing countries.

Interaction with Codex. The Group considered that there was a need for clear articulation of the questions from Codex and Member countries. In the context of Codex, the expert bodies could not rely solely on text in the official Codex reports. Improved communication with representatives from Codex committees was needed between the annual meetings of these committees.

It was important to negotiate/clarify the questions between the representatives of the expert bodies and the Codex committees. Questions should be relevant and within the scope of the respective body. This needed the active involvement of the expert body, the requesting Codex committee and the FAO/WHO secretariat.

FAO/WHO should consider means of enhancing such interactions, such as scheduling back-to-back meetings between expert body meetings and Codex committee meetings. They should also consider making members of expert bodies available at the Codex committee meetings to provide technical advice and participation of representatives of Codex committees in expert meetings. FAO/WHO should also establish means for increased interaction among expert bodies in matters/questions that impact multiple committees (e.g. considerations involving risk-risk trade-offs).

A simple and rapid mechanism was needed to address small queries from Codex to experts and to provide clarification on reports outside the meetings of expert bodies.

Meeting arrangements and logistics. Advance preparation for scientific meetings was considered to be essential, since most experts have time constraints and their availability is limited. FAO/WHO should urge governments to allow their experts to take time to work before and during international expert meetings. Mechanisms for industry and other stakeholders to provide resources without raising issues of conflict of interest needed to be developed. FAO/WHO should encourage allocation of national resources for permanent assignment of experts to international activities. A strong secretariat is needed to manage the preparation process, monitor progress and quality of preparatory work, and ensure the adequacy of all background material. FAO/WHO should provide a clear description of the role of all meeting participants, including the extent to which secretariat members should actively participate in the deliberations of expert panels. FAO/WHO should formulate realistic expectations in relation to the amount of work that can be achieved in the designated timeframe and resource constraints. The format or type of meeting should be determined according to the task and the scientific advice required. Operational procedures should be developed for expert meetings where these are not available. Convening more meetings per year, while addressing the workload, would require increased resources, secretariat capacity and expert time.

Openness. Mechanisms for bringing in a greater range of stakeholder comments should be explored and consideration should be given to incorporating a public comment round into the draft assessment advice of the expert groups. This could increase trust in the scientific advice provided, make the process more transparent and ensure that input is received from a wider group which would add value to the output. However, more time, resources and secretariat support would be needed to organize the meetings. The meetings would be more formal in nature and a higher level of control would be required to ensure that the meeting is productive. While some participants considered that open meetings might inhibit discussion and free expression by experts, others felt that outside comments could stimulate discussion. FAO/WHO should undertake a review of existing procedures for expert committees that are available at the national level, particularly in relation to openness of meetings, and consider how these might apply to international expert meetings.

The current process for delivering documented scientific advice is slow, often as a result of the need for technical editing after the expert meetings have reached their conclusions. FAO/WHO should consider making adopted draft reports available on the Internet immediately after the meetings, and making resources available to recruit qualified technical editors to begin work during the meetings.

Differences among experts/need to reach consensus/dealing with minority opinions. Procedures for the resolution of scientific conflicts should be documented. Members should be made aware of these procedures prior to the meeting. Voting could be used where consensus cannot be reached. Meetings should strive for consensus wherever possible, but where consensus cannot be achieved, this should be documented. These procedures would give the opportunity for the publication of minority opinions in such situations.

Work sharing and use of reports from national/regional bodies. The Group concluded that more use of national and regional reports should be encouraged. This would reduce duplication of effort in areas where there was already agreement, but the challenge to maintain the independence and impartiality of FAO/WHO scientific advice remained. However, it was important to recognize that only certain components of national and regional reports could be used for international deliberations. FAO/WHO, in consultation with Codex, should develop guidance on the use of national and regional reports. To enhance work-sharing, FAO/WHO should examine the possibility for coordination between national, regional and international bodies to achieve a coordinated evaluation or elaboration of scientific advice on specific issues.

Structured interaction with data providers and access to stakeholder expertise. Procedures for the interaction of data providers and other stakeholders with expert panels should be developed. This could be accomplished through interaction with data providers including before the meeting, the development of permanent mechanisms of exchange for technical questions controlled by the secretariat, and timely feedback to data providers. Efforts should be made to establish trust with data providers to facilitate provision of data.

Key findings

The Group noted that many Member countries had developed procedures and expertise in the acquisition and dissemination of scientific advice, and that FAO/WHO/Codex may benefit from studying these approaches in order to enhance their related activities at the international level. In particular, FAO/WHO could benefit from reviewing the way national governments had successfully enhanced the openness of their scientific advisory panels while safeguarding both the scientific integrity and efficiency of advisory panel deliberations. The Group noted that the transparency of FAO/WHO mediated acquisition and dissemination of scientific advice could be enhanced through articulation and implementation of written protocols (i.e. standard operating procedures) that formally communicate processes and criteria that would be used by expert meetings. The Group noted a critical need for timely, structured, iterative communication with the body(ies) requesting the scientific advice for the purposes of clarifying scientific questions to be addressed; negotiating specific tasks, approaches, timelines, and decision points; and providing initial conclusions and interpretation of the results. Particularly in relation to Codex, these may require a reconsideration of the current reporting structure to avoid the need to wait for extended periods to acquire needed input from the requesting Codex committee. Such improvements in the standardization and support of the expert bodies were likely to increase the level of logistical support required. Similarly, such iterative processes were likely to increase the need for a percentage of scientific experts that served as members of a standing body for a designated time period. This is particularly pertinent if FAO/WHO were to recognize the need to increase the degree of training that members of expert panels receive.

4.4 Communication of scientific advice

Considerations

Timeliness was defined as the time between the conclusion of a meeting and the publication of a final report. At present, this time period may be 6 months, one year, or longer in some cases.

The use of a peer review process could increase the delay between completion of a draft report and its publication. Factors which impacted the speed with which peer review occurred included the delay for the review itself plus the delay on the part of the authors in responding to the reviews. One issue which had an impact on both components of the delay was FAO/WHO staff workload. Overburdened staff members have less time to follow up with tardy reviewers and authors. Author response time is affected by issues of recognition and compensation. An author who has devoted time to a meeting and writing a draft, but who receives little recognition and/or compensation for the work, may not be motivated to respond to reviews in a timely manner.

Another factor which could affect an expert's willingness to volunteer time also relates to timeliness. If a report is not released in a timely manner (and assuming this was not under the expert's control), stakeholders (from industry or consumer groups) may contact the expert to ask about the findings. If the expert was bound by confidentiality, this could lead to a great deal of frustration which might dissuade continued participation. Models which avoided this problem included that used by the United States Food and Drug Administration (FDA), where petitioners could be in direct contact and get results quickly, and the European Union model (EFSA) where minutes, opinions and the draft report appeared within one week of the meeting. One simple solution would be to publish a draft report immediately after the meeting (assuming the meeting participants were all in agreement with the draft).

Transparency. A transparent report was one which included: (a) a clear explanation of the question asked; (b) the data available to answer the question; (c) the data used to arrive at the conclusions; (d) the reasons why some data were excluded; (e) an explanation of the reliability of the data used (including any uncertainty); (f) the strength of the hypothesis on which the advice is based, including perhaps a qualitative description of that hypothesis (e.g. plausible, possible or unlikely).

The inclusion of minority opinions was an important component of transparency. Minority opinion in providing scientific advice must be based on science. Minority opinions were more likely to arise with fewer or shorter meetings, as reaching consensus often requires more time for discussion.

If mathematical models or computer simulations were used to provide scientific advice, the communication of that advice should include access to the mathematical models or computer simulations as a component of transparency.

Meeting the needs of the requestor. The scientific advice delivered at the conclusion of the process should meet the needs of the requestor. This can be enhanced by procedures that foster the interaction between the risk assessors and risk managers at the start of the process so that the question is properly framed. This interaction should be iterative, and include a review of the draft of the expert panel's report by those requesting the scientific advice.

Plain language. Although it might be important for a final report providing scientific advice to be written in plain language, the level of complexity in the final report depended upon the needs of the requestor. If the communication needs of multiple audiences were to be met, the final document could be released in several forms (e.g. technical document, interpretive summary and executive summary), each with the appropriate level of detail. Release of the findings in several forms with different levels of detail might also improve the ability of the public and stakeholders to comment on the findings of the report.

The challenge of translating complex technical language may increase when the uncertainty surrounding a particular issue is greater. The inclusion of risk communication experts on the expert panel from the start of the process often facilitates the translation of the findings into plain language at the end of the process.

The technical complexity of the final report (or reports) is closely linked to the "roll out" of the findings, and this is further addressed in the process section. Briefly, roll-out concerns include: determining who to contact, flagging any sensitive areas, coaching report communicators on those areas, issuing appropriate press releases and formulating a policy for press contact with expert panel chairs.

It may not be possible to translate a complex technical document into multiple languages. If plain language interpretive and/or executive summaries are prepared, however, the translation of these documents may be possible and would help to address the needs of developing countries.

Key findings

The Group noted the need to increase the transparency and effectiveness of the communication process. In particular it was noted that significant gains in transparency could be achieved through considering the audience to which the scientific advice was being provided and the need to provide information in plain language or multiple versions. This approach could also simplify issues around the translation of reports and findings. Achievement of transparency should also include a review of work products to ensure that all assumptions were explicitly stated, and that the impacts of these assumptions were articulated. There was a need for processes that allowed for the recording of alternate interpretations of the scientific data if consensus could not be reached by an expert body. FAO/WHO should consider advantages that may be gained through the initial release of reports in draft and the solicitation of public comments.

The Group also noted the need for FAO/WHO to pay increased attention to the manner in which scientific advice was released. Opportunities to document its significant impact in advancing both science and public health were often missed.