This is a copy of a paper by David Jukes which was first published in Food Control, Volume 11(3), June 2000.

Abstract

International food trade rules, as established and agreed under the auspices of the World Trade Organisation, have enhanced the role of the Codex Alimentarius Commission. However, the increased recognition has also led to increased pressure for the Codex to define more clearly the approach to the setting of international food standards. The paper provides an account of the discussions in Codex on three complex debates: the setting of residue limits for hormones in meat and in milk and the adoption and use of four ‘statements of principle’ on the role of science in the Codex decision-making process. Arising from these debates, the paper looks at the role of science in the adoption of Codex standards and considers those issues, known as ‘other legitimate factors’, which also figure in the process.

On the 1st January 1995, the Agreement on the Application of Sanitary and Phytosanitary Measures (the "SPS Agreement") came into force with the formation of the World Trade Organization (WTO). The SPS Agreement provided detailed controls, for the first time, on issues affecting food safety measures. Although some constraints had been applied by the original General Agreement on Tariffs and Trade (GATT) in 1947, under Article XX(b), it was believed that various countries were using food safety concerns to justify maintaining or erecting food regulations against imported foods which were a barrier to trade. With no defined international agreement for verifying the justification for the measures, the justification of food safety could not be questioned or removed. Hence, when undertaking a fundamental reform of the world trade rules leading to the creation of the WTO, the SPS Agreement was conceived and ultimately incorporated into the Final Act establishing the WTO so as to precisely define the necessary discipline.

This Agreement provided new rules for the Members of the WTO such that, in the event of a trade dispute relating to food safety, the disputes procedure of the WTO would use the rules of the SPS Agreement to determine whether the complaining country was justified in its complaint and hence whether appropriate retaliatory action could be taken. The SPS Agreement therefore forms the basis upon which countries operate so as to ensure compliance with their WTO obligations.

The SPS Agreement relates to three main issues: food safety, animal health and plant health. In each case, the Agreement identifies an international body as providing the basic standards against which disputes over national regulations would be judged. This paper is concerned with the operation of the Agreement with regard to food safety issues only.

The SPS Agreement states (in Article 2(1)) that "members have the right to taken sanitary and phytosanitary measures necessary for the protection of human, animal or plant life or health, ..". It then goes on to place an obligation on members (in Article 2(2)) to ".. ensure that any sanitary or phytosanitary measure is applied only to the extent necessary to protect human, animal or plant life or health [and] is based on scientific principles and is not maintained without sufficient scientific evidence, ..". In Article 3(1) there is a requirement that countries base their SPS measures on recognised "international standards, guidelines or recommendations". For food safety, the standards of the FAO/WHO Codex Alimentarius Commission (the CAC) are specified as providing the international standards, guidelines and recommendations against which any dispute panel would judge compliance with the SPS Agreement. However, this does not imply that the Codex standards are mandatory on members of the WTO. The SPS Agreement provides a number of qualifications to the basic supposition that Codex standards are to be used by member countries in setting national food safety legislation.

The major qualification is that contained in Article 3(3) of the SPS Agreement which states that:

Linked to this permission is a footnote designed to clarify the operation of the term ‘scientific justification’. The footnote states:

A number of minor additional points need to be made clear before discussing the issue of science raised by the Agreement. First, whilst Codex food standards work has been taking place since 1962, their development did not place any obligations on countries to adopt them. It was often felt that the standards were being developed so as to assist developing countries by providing a ready made standard which they could adopt. Developing countries were prone to think that, if they adopted Codex standards into their own legislation, they might gain access to the developed world. This did not happen since most Codex standards were not implemented by the developed countries and barriers remained in place. Developed countries were unwilling to modify and disrupt their long established system of controls.

Secondly, in the area of food safety, the SPS Agreement only relates to measures which are designed to protect human health. It makes this clear by the definition of a "sanitary or phytosanitary measure" contained in Annex A to the Agreement which states, in the specific case of food, that it is "any measure applied ... to protect human .. health within the territory of the Member from risks arising from additives, contaminants, toxins or disease causing organisms in foods [or] beverages ..". Codex has often developed standards which are not based on health requirements but are valuable for other issues. An example would be a labelling standard designed to provide consumers with information.

Thirdly, although this paper is focused on the role of science as related to the food safety aspects of the SPS Agreement, another WTO Agreement controlling Technical Barriers to Trade (the "TBT Agreement") has an impact on other aspects of food standards. Although science is a factor included within the TBT Agreement, it is not the only one and the focus on science is therefore less well defined.

Although it should be made clear that the SPS Agreement places no obligations on Codex itself, only on WTO member countries, it was inevitable that Codex would examine its procedures and determine what changes were needed to take into account the new recognition given to Codex standards.

The implications of the new status was recognised well before the formal entry into force of the Agreement itself. Its development was one of the main reasons for the convening of an international food standards conference in 1991 where the likely implications of the proposed new agreement were debated and suggestions for the future work of Codex discussed. There was general recognition that the forthcoming adoption of the WTO rules meant that Codex needed to adopt new practices and procedures. Specifically, with respect to the role of science, the Conference recommended:

The obligation on countries to base their food safety measures ‘.. on scientific principles..’ required Codex to consider in more detail the methods it used to develop and adopt standards. The principles used by Codex were ill-defined and subject to different interpretation and application from meeting to meeting. This was particularly the case since much of the preparatory work of the Codex is undertaken in Codex Committees (known as ‘subsidiary bodies’) where the national delegates attending can vary significantly depending upon the subject of the committee. Although the Secretariat would attempt to ensure consistency, the decisions were (and still are) taken by consensus (or occasionally by voting) by the Codex member states. The new status therefore required Codex to recognise that, since countries had an obligation to use scientific principles, it was incumbent upon the delegations from member countries of Codex to make full use of science when adopting Codex standards relating to food safety measures.

In many areas of Codex work, there is little dispute as to the importance of science in establishing Codex standards. Science has formed the basis of many discussions and decisions. However, there have been two issues which have caused major difficulty and both relate to the use of hormones as veterinary substances. The specific substances are:

• growth promoting hormones used in beef cattle: estradiol 17-beta, progesterone, testosterone, zeranol and trenbolone acetate
• ‘production aids’ used to increase milk production in cows: recombinant Bovine Somatotropin (rBST or, more simply, BST)

In both cases the USA, along with certain other countries, has permitted their use following examination of the scientific evidence by their relevant national committees. However, in both cases, the European Union and certain other countries have resisted their use. In the case of the meat hormones, the EU imposed a ban on both its use within the EU and on the import of meat produced using the hormones. In the case of BST, the EU has placed a moratorium on its use pending more detailed information. However, the import of milk and dairy products from other countries has not been subject to any specific prohibition.

Within the Codex system, approval for the use of veterinary drugs (including hormones) is by the adoption of a maximum residue limit (MRL). The recommendation for an MRL comes from the FAO/WHO Joint Expert Committee on Food Additives (JECFA) and is based many factors including:

The toxicological data is used to establish the appropriate Acceptable Daily Intake (ADI). JECFA then uses a defined decision process to allow the incorporation of the other data so as to establish a recommended MRL which meets the required criteria (JECFA 1990). The MRL can be set at either a quantified level or can be ‘not specified’ indicating that ‘the presence of residues in the named animal product does not represent a health concern and there is no need to specify a numerical MRL.’

Linked to these specific discussions, there have been moves to define more precisely the role of science in Codex decisions. These have led to the development of certain ‘Statements of Principle’. Discussion of the issues relating to the two types of hormones and the ‘Statements of Principle’ have overlapped and been directly related. However, it will be useful to set out separately the various stages in their adoption in the following sections - looking first at the beef hormones, then the ‘Statements of Principle’ and finally the milk hormone. A summary of the chronology of the discussions is given in Table 1.

The use of the hormones was examined at the international level by the Joint Expert Committee on Food Additives (JECFA) in 1987 and an ADI ‘not specified’ was adopted (JECFA, 1988). Subsequently it was progressed through the stepwise procedure for the adoption of Codex maximum residue limits (MRLs). It was adopted as a proposed Codex MRL (at Step 5) by the Commission meeting in 1989 (FAO/WHO, 1989). It was then reconsidered by the Codex Committee on Residues of Veterinary Drugs in Food (CCRVDF) in October 1989 (CCRVDF, 1989). Although there were some concerns expressed at the speed of adoption, the CCRVDF agreed to advance the MRLs to Step 8 for adoption by the 19th Session of the Commission in 1991.

At the 19th Session, positions had hardened and the view of the EC was expressed by the delegation of the Netherlands. Whilst "not objecting to the evaluation of JECFA", the report of the Session records the following areas of concerns (FAO/WHO, 1991b):

• concerns about the different aspects of the use of hormones;
• concerns about the practical possibilities to enforce and control the prescribed administration of growth-promoting hormones independent of these concerns;
• the need to take into account the clear concerns of European consumers on the safety of meat produced with growth hormones;
• the opposition to the use of hormones for growth-promoting purposes as expressed on repeated occasions by the European Parliament as well as consumer organizations in the Community.

The opposing view was that the evaluation was based on sound science. As recorded in the report:

" Other delegations noted that the work of Codex was consistently based on sound scientific principles and that the JECFA evaluation confirmed that health issues in relation to residues of these substances in food were not a cause for concern. It was stated that consumers were often unaware of the scientific validity and logic supporting the use of these compounds, and that regulatory control procedures for their use were similar to those for the application of any chemical substance." Reference was also made to the recognition being given to Codex with the development of the WTO and it was suggested that Codex had been recognised as a body which works "strictly within scientific principles and procedures". It concludes with the statement: "Deviating from these basic and founding principles of the Commission was felt to have serious implications for Codex." The proposal to adopt the MRLs as Codex standards was then put to a vote and 12 countries voted in favour with 27 against. The MRL was therefore not adopted and this decision would be reported back to the CCRVDF.

The decision was discussed at the 6th Session of the CCRVDF in October 1991. The United States had prepared a paper setting out the questions which needed to be considered further in view of the decision at the Commission meeting. The Secretariat clearly stated the position as recorded in the report of the meeting:

The matter was again discussed at the 7th Session of the CCRVDF in October 1992. However, since it was agreed that the Commission decision had only held the Draft MRLs at Step 8, no action was needed by the CCRVDF as the possible adoption would be on the Agenda of the next Commission meeting in July 1993 (CCRVDF, 1992). Also, detailed discussion on the implications of the decision had been referred to the Codex Committee on General Principles (CCGP) and the decisions of that Committee would be relevant to any further debate.

The US discussion paper was considered and discussed (under ‘Other Business’) at the 10th Session of the Codex Committee on General Principles (CCGP) in September 1992. In indicating their concerns at some of the Codex decisions (in the case of the beef hormones for example) the United States stated that: "Codex should demonstrate that it was a genuinely neutral intergovernmental body and that it must show that its standards, guidelines and other recommendations rested on a sound scientific basis." (CCGP, 1992). They then asked that Codex consider the possible modification of procedures to help in this process. The report of the meeting indicates that Canada and New Zealand expressed support for this.

However, other delegations, whilst expressing support for "the principle that Codex recommendations should be based on the best scientific opinion available" also indicated that there would be "other non-scientific criteria, such as economic factors or consumers’ concerns, which needed to be taken into account in arriving at Codex recommendations." It was agreed that the matter needed to be discussed further and that the Secretariat would prepare a discussion paper for the next CCGP meeting.

The matter was however next discussed at the 20th Session of the full Commission meeting in July 1993. The specific topic for the debate related to an MRL for Trenbolone Acetate, another growth-promoting hormone. Since the matter was being discussed within the CCGP, the meeting felt it best to hold the MRL for Trenbolone Acetate at Step 8 with the other substances. The Commission requested the CCGP to consider developing guidance on how science and other factors should be integrated into its elaboration procedures and decision-making processes. This development is discussed further in the next section. (FAO/WHO, 1993).

The matter was again on the Agenda of the 21st Commission meeting in July 1995. The report summarises the position as follows: ‘The Commission was split into two opinions: (1) in favour of adopting these MRLs at Step 8 without delay; and (2) in favour of postponing the consideration to the next session of the Commission awaiting an international conference organized by the EC scheduled later [in 1995] to study this issue. After a lengthy debate on the issue in relation to whether to base a decision on currently available scientific evidence or to take into account factors other than health concerns, the Delegation of Spain, on behalf of the member countries of EU, proposed a roll-call vote on the adjournment of debate on the adoption of certain growth-promoting hormones at Step 8.’ (FAO/WHO, 1995) The motion for adjournment was however lost (28 votes in favour; 31 votes against; 5 abstentions). It was then agreed to hold a secret ballot on the adoption of the MRLs. In this case, as a result of the secret ballot, the Commission adopted the MRLs (33 votes in favour, 29 votes against with 7 abstentions). The debate was therefore closed with the MRLs being adopted and becoming available for use in any dispute procedure under the rules of the WTO involving the SPS Agreement. This has subsequently happened but is not part of the discussion in this paper.

As already mentioned, the importance of science had been stressed at the 1991 FAO/WHO conference on food standards. The matter was carried forward at the 10th Session of the CCGP in September 1992 which asked the Secretariat to prepare a paper on the issue (CCGP, 1992). At the 20th Commission meeting in July 1993, arising from the discussions of hormones for beef, the Commission asked the Codex Committee on General Principles to consider developing guidance on how science and other factors should be integrated into its procedures and decision making processes. Other factors in addition to science would need to be considered. Such other factors included legitimate consumer concerns, animal welfare, fraudulent or unfair trading practices, labelling and other ethical and cultural considerations although it was recognised that science was to be preeminent in Codex procedures. (FAO/WHO, 1993).

A paper was therefore submitted to the 11th session of the CCGP in April 1994 on ‘Consideration of the Role of Science in the Codex Decision-Making Process’ (CCGP, 1994a). The debate at this meeting appears from the report to have been lengthy and difficult. Although the Commission had specifically asked the CCGP to consider the issues relating to science and the elaboration of standards, the Secretariat’s paper suggested the inclusion of a statement relating to science in the General Principles of the Codex Alimentarius. However, this section does not constitute formal rules for the work of the Commission and several delegations wanted a more drastic redrafting of the rules and procedures to ensure full integration of science. After some discussion, the Committee established a Working-group to consider the different options presented during the debate. Later in the meeting, the Working Group presented its proposals which took the form of a listing of the sections of the Procedural Manual where revision might be considered. After even more debate the Secretariat indicated that it would draft some proposed revisions, circulate them for comment and prepare a redraft for the next meeting of the CCGP. (CCGP, 1994b)

The matter was however subsequently discussed at the 41st Session of the Executive Committee in June 1994. Whilst grateful for the work of the CCGP, the Committee felt that the matter needed to be resolved more rapidly and accordingly drafted four ‘Statements of Principle’ which they decided should be put to the Commission. Amendments to the Procedural Manual in relation to these matters would not therefore be considered further. (EXEC, 1994)

At the 21st Commission meeting in July 1995, the Commission therefore adopted the four ‘Statements of Principle concerning the Role of Science in the Codex Decision making Process and the Extent to Which Other Factors are Taken into Account’ (FAO/WHO, 1995). The Statements are given in Table 2. The inclusion, in Statement 2 of the phrase ‘other legitimate factors’ should be noted.

Discussion was then continued at the twelfth session of the CCGP where consideration was given to incorporating the four statements into the Procedural Manual. Although certain delegations initially expressed a view that this would be inappropriate, it was agreed to recommend to the next Commission that they should be added to an Appendix to the Procedural Manual entitled ‘General Decisions of the Commission’ (CCGP, 1996). This was subsequently put to the Commission at its 22nd Session in June 1997 and it was agreed to introduce the new Annex (FAO/WHO, 1997a). The Annex can be found in the 10th Edition of the Procedural Manual published later in 1997 (FAO/WHO, 1997b).

The second major issue which has developed relates to the use of hormones to increase the milk yield from dairy cows.

At the 4th Session of the CCRVDF in October 1989, discussions were held on the priority listing of substances for JECFA evaluation. The United States pressed for BST to be evaluated rapidly. The Report records that: "It was noted further that issues unrelated to science were threatening to affect registration of the somatotropins in several countries, which could create trade problems." (CCRVDF, 1989). However, other countries (principally from Europe) believed that additional information was needed before JECFA could evaluate BST. It was decided to place BST on the provisional list for JECFA evaluation in 1992.

JECFA duly evaluated the substances in June 1992 (JECFA, 1993) and their conclusions were considered at the 7th Session of the CCRVDF in October 1992 (CCRVDF, 1992). JECFA had concluded that "the margin of safety was so large taking into account proposed use, potential intake of residues and available toxicity data that [residues of BST] represented no hazard to human health". As a result, the CCRVDF agreed to move the proposed draft maximum residue limits ("not specified") forward for consideration by the full Commission meeting. At that meeting, in July 1993, BST was adopted at Step 5 with a draft Codex MRL (FAO/WHO, 1993).

Consideration then progressed to the 8th Session of the CCRVDF in June 1994. The report of that meeting indicates that the MRLs were advanced to Step 8 for final adoption by the Commission at its next session. However, the report contains the following statement suggesting possible future problems:

It was therefore presented to the Commission at the 21st Session in July 1995. No indication of any discussion is recorded in the Report of the meeting. It simply records that Member States of the European Union proposed a roll-call vote on the adjournment of debate on the adoption of the MRLs until the next sitting. The result was that 33 countries voted to adjourn the debate, 31 countries voted against adjournment and there were 6 abstentions (FAO/WHO, 1995).

No decision was therefore taken with the matter returning to the 22nd Session in June 1997 and, not surprisingly, little had changed. The report makes it quite clear when it states: "The Commission was split into two options: (1) in favour of adopting the draft MRLs at Step 8 without delay; and, (2) in favour of postponing their adoption." (FAO/WHO, 1997a). Those wanting to adopt the MRLs were of the view that the scientific assessment had been completed by JECFA and the CCRVDF and as additional scientific information had not been provided, a re-evaluation was not necessary. If new data were to become available, then a re-evaluation would be appropriate. Of the opposite point of view, various delegations believed that new information was available which warranted a review. However, they also indicated that there were other issues which, in their view needed to be considered - as stated in the report: ".. the use of BST was opposed by consumers and had no benefit for them, and would not improve milk quality or safety. The importance of taking other legitimate factors than sound scientific analysis and evidence as stated in the Statement of Principle on the Role of Science into consideration was also emphasised."

To draw the debate to a conclusion, a proposal was put to the Commission to do two things:

This proposal was subject to a vote by the Commission and was agreed by 38 votes to 21 with 13 abstentions. It is interesting to note that the 38 votes were, with the exception of Singapore, all from Europe (including all 15 countries of the EU), Africa or the Middle East. The 21 votes against included the USA, Canada, South Africa, Nigeria, Australia and New Zealand as well as countries from South America and Asia. The vote therefore reflected a strong a regional split.

The two things mentioned above then followed. Firstly JECFA was asked to re-evaluate BST when it next met in February 1998. This it did and re-affirmed the original view that both the ADI and the MRL were "not specified". The conclusions of the JECFA meeting were published in time for the subsequent CCRVDF meeting in September 1998. It should be noted however that the detailed discussion and associated documentation were not available then and were only published in early 1999 (JECFA, 1999).

Despite the re-affirmation of the "not specified" status of BST, at the CCRVDF meeting there was again a division. Several delegations considered that JECFA had failed to undertake a satisfactory analysis, the failure to publish the full report was regarded as a problem since delegations had been unable to consider the scientific basis of the JECFA decision, and, there were once again arguments about the "other legitimate factors" which should be considered and which could be used as an argument for deciding not to adopt the MRL. At the meeting, 12 delegations spoke in favour of recommending the MRL to the Commission for adoption whilst 12 delegations (11 EU countries and Norway) spoke against. The Chairman of the CCRVDF decided on his own initiative to advance the MRL to the Commission since in the view of the Chairman there was no scientific argument for rejecting the JECFA recommendation (CCRVDF, 1998). The proposal therefore goes forward to the Commission meeting in June 1999.

The general issue of how to operate the principles, including the interpretation of ‘other legitimate factors’, was referred to the thirteenth CCGP meeting in September 1998 (CCGP, 1998). The meeting was involved in a long debate based around a Secretariat paper which attempted to set out the issues and to provide a basis for developing conclusions. Although the discussions were focused on BST (since that had been the specific topic refereed to it by the Commission) broader issues were raised. Details of the arguments will be discussed below. It was again apparent that there was no consensus and that the meeting was split. To try to resolve things, it was agreed to return to the discussion at the next meeting but to base it on two separate papers - one which would set out the general issues relating to "other legitimate factors", the other would consider them in relation to BST. In that way it was hoped to try and provide a more structured discussion.

The Committee held its fourteenth Session in April 1999 and had the two papers prepared by the Secretariat (although the one on general issues only appeared a few days before the start of the meeting). The first paper (CX/GP 99/9), relating to the general issues of "other legitimate factors", will be discussed below (CCGP, 1999a). The second paper (CX/GP 99/10) relating to BST had, as one of its proposed conclusions, the statement that: "No clearly identified factor has been put forward to prevent the establishment of a MRL for BST at the international level, for the purpose or international trade." (CCGP, 1999b)

On this second paper the meeting was again divided with no consensus being achieved. Although the meeting had been specifically asked to consider the issue of "other legitimate factors" it was apparent that the decision of the CCRVDF decision to recommend that the Commission adopt the MRL for BST at the next meeting (in June 1999) was a complicating factor. On several occasions delegations included reference to arguments over the scientific basis of the decision in their presentations. Documents made available to the conference included two reports from EU scientific committees prepared in March 1999. The Committee recongised that it was unable to reach a consensus and agreed to report this fact to the Commission at its meeting. During the discussion it can be noted that the US delegation stated that "Failure to support CX/GP 99/10 could have profound, far-reaching and possibly fatal consequences" for the work of Codex.

Although the discussions at the Committees had suggested that there could be a major dispute at the Commission meeting in June, in practice the discussion was limited. The Codex Commission has increasingly been stressing the need for consensus in the decision making process. Early in the meeting, it had been agreed to incorporate a reference to this in the Rules of Procedure of the Codex Commission. Since it was apparent that there was no consensus on the issue of hormones, no decision could be reached if this change was to be applied to this discussion. The final report of the Codex meeting gives the following account of the resulting debate (FAO/WHO 1999):

The phrase ‘Other Legitimate Factors’ comes about from its use in the four Statements of Principle agreed by Codex setting out how science is to be used as the basis of Codex standards (as described above). The phrase has therefore to be seen in the context of science providing the key factor used in establishing standards but, ‘other legitimate factors’ (OLFs) can be used.

Before presenting and discussing the issues associated with OLFs, it must be appreciated that the development of the wording in Codex documents is not subject to the scrutiny of trained legal draftsmen. Words, phrases and other terminology are often agreed on the basis of consensus. Delegations in Codex meetings are often heard expressing a view that a certain text is not ideal but that they would be willing to accept it to achieve compromise. The precise meaning of a phrase is always open to interpretation in different ways by different delegations. By contrast, this is not the case with the texts of the World Trade Organisation including the SPS Agreement. When considering disputes within the WTO, a panel of legal experts is established who are used to precisely interpret the rights and obligations of member countries. This lack of precision in the terminology is at the basis of the current debate over OLFs and the lengthy period that it has so far taken to even establish what the phrase means.

The concept of an OLF must be seen in the context of both the overall mandate of the Codex and the first two principles agreed by the Codex. Article 1 of the Codex Alimentarius Statutes gives the Commission responsibility for all aspects relating to the FAO/WHO Food Standards Programme, the purpose of which includes "the protection of the health of consumers and ensuring fair practices in the food trade." (FAO/WHO, 1997b). These issues are carried forward into the second Statement of Principle under which Codex states that it will ".. have regard, where appropriate, to other legitimate factors relevant for the health protection of consumers and for the promotion of fair practices in food trade." It can be noted that, even in these two important phrases, the terminology used is not identical - an example of the texts being developed without legal advice.

If the second principle is to be taken at face value, then the concept of an OLF must be "relevant" either "for the health protection of consumers" or "for the promotion of fair practices in food trade." This concurs with the views of the Secretariat in their discussion paper CX/GP 99/9 which also put forward this interpretation as the basis of their suggested OLFs.

There are various possible OLFs which have been identified in various papers and discussion documents. In each case, there are arguments both in their favour and against.

The FAO and WHO had jointly organised, in 1997, a joint expert consultation on risk management and food safety. Although hey did not discuss in detail the issue of other legitimate factors, one of their recommendations was that: "The CAC should clarify for the guidance of Codex Committees how to apply the [second principle]. In particular, this clarification should include explicit description of the factors which may be considered, the extent to which these factors should be taken into account, and the procedures to be used in this regard." (FAO, 1997).

The Codex Secretariat tried to identify (in CX/GP 99/9) those OLFs which had previously been used in Codex work and where they felt that it could see justification in its use under the second Statement of Principle. The Secretariat based their initial discussion on the use of OLFs in the risk analysis process but their discussion extended beyond this specific issue. Whilst the issue had formed one recommendation of the expert consultation, the use of the phrase in the second principle does not restrict an OLF to this issue alone.

The following represent those OLFs which were identified in the Secretariat paper (CCGP, 1999a):

Having identified a risk in food safety, it is then up to risk managers (regulators for example) to implement an appropriate management strategy. However, the decision may be based on factors which go beyond science. They may need to taken into account what is "practically achievable in the production and processing processes in view of current technology". A control measure needs to be feasible and economically sustainable. In setting a level for mycotoxins in food, whilst a very low level is considered appropriate, it has been necessary to recognise that if the level were to be set too low, it might be impossible (particularly in developing countries) to meet the level.

Prior to agreeing a limit for a contaminant, pesticide or veterinary drug it is necessary to recognise approved methodology by which the level of the substance concerned is monitored and controlled.

This recognises that it may not be feasible to fully implement a particular control or processing strategy. It is linked to the question of economic sustainability given in a) since perhaps most things are technical feasible these days but may not be economic in practice. The Codex paper suggests that the use of the concept of ‘As Low As Reasonably Achievable’ (ALARA) applied to certain maximum contaminant levels is an example of this approach.

Another approach to the consideration of technically achievable is used in the process of adopting MRLs for pesticides. Initial assessment is on the level of the pesticide residue which remains if ‘Good Agricultural Practice’ (GAP) is used during the application of the pesticide. The resulting level is then compared with the level regarded as being safe based on scientific evidence and including consumption data. If the level determined under GAP is below the safe level determined by science, then the GAP level is adopted. This encourages good practice in the use of the pesticide and provides added safety.

In evaluating and deriving numerical values for the ADI of an additive, contaminant or residue, the JECFA has developed the application of a ‘safety factor’. The safety factor tries to provide enhanced safety to take into account an awareness that the scientific evidence may not have provided a precise demonstration of a safe level in all cases (e.g. age of the consumer, the overall diet, variations in metabolism within individuals). The value of it is not based on science although science is sometimes used to modify it where information suggests the need for an enhanced safety level or where a lower level is considered appropriate and possible without risk to health.

There are various other factors which have been included in the discussion of ‘other legitimate factors’. Some of these have been mentioned earlier in the discussions on hormones. The following are the issues most usually raised.

The inclusion of ‘technological need’ derives from a well established part of the process to consider whether a food additive should be adopted into a Codex standard. It is often the case that food additives can only be permitted for food use if, firstly, there is a demonstrated technological need for the additive and, secondly, it is regarded as safe under the intended conditions of use. This idea is incorporated into the Preamble to the Codex General Standard for Food Additives which states that: ‘The use of food additives is justified only when such use has an advantage, does not present a hazard to health of and does not mislead the consumer and serves one or more of the technological functions set out by Codex and [defined] needs, and only where these objectives cannot be achieved by other means which are economically and technologically practicable.’ (CCFAC, 1997). The application of the criteria to establish the need for the additive certainly goes beyond the strict evaluation of the safety of that additive (which is the second component of the approval process - if the need has not be demonstrated, then safety is not considered). It is interesting to note that this issue was not included in the discussion paper CX/GP 99/9 probably because it does not appear to fall within the strict terminology of the statements of principle.

The criteria of need has been well established as a basis for trying to ensure that food additives are not used excessively and unnecessarily. In most cases its use has been well accepted and fairly uniformly applied. Perhaps the greatest area of dispute has arisen over the question of ‘need’ when applied to food colours. Here it has been argued that colours are used to disguise the lack of certain ingredients in a processed food - the ‘need’ for the colour would not exist if a greater quantity of the original coloured ingredient had been used (e.g. strawberries in a strawberry flavoured ice cream).

The idea of technological need is therefore pertinent to the discussion of OLFs and has been raised by various bodies. In particular, the issue is of direct relevance to the BST debate since BST is used to enhance milk production. Its use is therefore to provide more milk from an individual cow. Is it therefore ‘needed’? Within the EC, where there has been an over-supply of milk, the answer is likely to be no. However, in other countries, operating under different economic and geographical constraints, the use of BST may well be helpful. The evaluation of need is therefore difficult in the context of an international standards setting body. The paper from the Codex Secretariat indicates that, in their view, a Codex MRL for a veterinary drug should be solely based on the scientific evidence. If a country subsequently decides not to permit the substance due to national criteria, then that is for the country to defend.

The issue of environmental concern has been raised as an OLF. For Codex however this presents problems. Codex Committees are not competent to evaluate the risk to the environment of any food related activity. This does not prevent environmental issues being addressed by Codex standards - the need for any decision though will have to come from another competent body. The phasing out of methyl bromide and the development of methods of analysis which do not use ozone-depleting substances have been given as examples of environmental concerns addressed by Codex. The issue is likely to become increasingly prominent with the current concerns relating to genetically modified crops.

When trying to interpret the second statement it could be argued that the phrase ‘for the health protection of consumers’ goes beyond the question of health relating to food safety issues. This view was expressed at the CCGP meeting. The discussion at the CCGP meeting is recorded in the Report as follows:

Most veterinary drugs are designed to prevent disease in animals and are therefore largely of major benefit to an animal and its welfare. In the case of BST however, it is argued that the use of BST actually increases the prevalence of disease in the treated animals. Codex is however not a suitable forum for a discussion of the issues affecting animal health welfare. With regard to animal health, the assessment of the suitability of any drug in treating a disease is assessed by the OIE. This function is recognised under the SPS Agreement in a similar way to the recognition given to Codex in food safety matters. However, it is not directly competent to assess the animal health effects. JECFA has also clearly stated, when considering BST, that it is not competent to assess animal health issues. As stated by the French Government: ‘It is to be emphasized that contrary to most other veterinary compounds assessed by the CCRVDF, BST does not improve the health status for animals by helping to prevent or cure pathologies. This factor must be taken into account despite the absence of a formal group of scientific experts to review the consequences of BST on animal health.’ (CCGP, 1999d)

Consumers International (CI) believe that there is growing evidence suggesting a scientific basis for animal welfare concerns with BST. They cite an EU report and the decision by Canada not to approve BST use. This does leave a problem for Codex - where should these issues be incorporated into the approval process? According to CI, ‘It may now be feasible for CCRVDF to conduct a review of animal health and welfare impacts of BST use, as it appears that the scientific basis for doing so is emerging.’ (CCGP, 1999e). The Secretariat also seem to support this: ‘If no specific expert advice could be provided in this framework [JECFA/OIE] but the results of risk assessment carried out at the national level in a number of countries gave similar results, on the basis of scientific studies, this question might be discussed and solved in the framework of the CCRVDF, provided consensus existed.’ (CCGP, 1999b).

Discussion at the CCGP meeting was inconclusive. In the general discussion of OLFs it is reported that: ‘The Committee recognized that there was no consensus on the integration of a number of other factors including animal health [and] animal welfare’. When considering BST separately, the Report states: ‘Other delegations referred explicitly to animal health and welfare as legitimate factors that had to be taken into account in relation to BST. It was noted that the welfare of animals was included in the Codex International Recommended Code of Practice for the Control and Use of Veterinary Drugs (CAC/RCP 38/1993). Furthermore, it was suggested that different considerations should be given to the evaluation of substances used for therapeutic purposes that those for production efficiency and growth promotion.’ (CCGP, 1999c).

The issue of consumer concern was clearly stated in the report of the CCRVDF meeting in October 1991 where, in relation to the beef hormone issue, the Secretariat pointed out that: ‘The principle problem was not with the safety of the substance and their residues, but in the public perception of their safety’ (CCRVDF, 1991). Within Europe, consumer concerns have been heightened by various real issues, not least with BSE and the continuing uncertainty over the possible long-term human consequences. Any proposal to modify the food supply is questioned and subjected to often hostile media attention. The British press is currently reporting concerns over genetically modified foods and consumers have doubts about their safety. Given this background, it is not surprising that politicians, who are the people’s representatives, are cautious is expressing approval of the use of new technologies including hormones.

Consumers wish to feel that the legislation provides protection against risk. Whilst acknowledging that zero risk is impossible, increasingly reference is being made to the ‘precautionary principle’ as an approach promoted by consumer groups. At the recent CCGP, the use of the precautionary principle was discussed and a possible working definition was proposed:

This interpretation makes it clear that the objective is to base measures on science but that there will be occasions where there is a lack of scientific evidence. Does this then make it into an OLF? Within the context of the consideration of OLFs, the Codex Secretariat makes it clear that, in their view, the precautionary principle applies when the science base is still being developed - it is therefore pertinent to the first Statement of Principle relating to the use of science rather than the second relating to the consideration of OLFs. In the particular case of BST, since JECFA have conducted two assessments, the science base has been clearly evaluated and the precautionary principle is not, in their view, applicable. The validity of the assessment has been questioned but that does not in itself make the subject an OLF.

Another aspect of consumer concerns is the option of providing food labelling as a means of giving consumers the ability to choose a particular product. The third statement recognises this and states that: ‘In this regard it is noted that food labelling plays an important role in furthering both of these objectives.’ It can be noted that the Statement opens with the phrase ‘In this regard ..’ which appears to relate it to the idea of other legitimate factors. It has been argued that labelling to satisfy consumer concerns is wrong. One submission to the CCGP meeting stated the following: ‘There is a difference between a consumer’s need to know, and a consumer’s right or desire to know. a consumer’s need to know specific information about a product in order to ensure health and food safety would be an appropriate Codex consideration. In contrast, using labelling simply to satisfy a consumer’s purported desire to know or right to choose, on the other hand, would be outside of Codex’s purview, which rests on protecting consumer health.’ (CCGP, 1999f). This represents a very strict interpretation of the work of Codex.

The alternative view expressed by Consumer’s International in their paper relating to the consideration of OLFs in the case of BST states the following: ‘Consumers International rejects the view, voiced by industry participants, that in the absence of a clear risk to health, labelling with respect to BST use in milk production cannot be justified. We also reject the often-stated view that mandatory labelling for BST status would constitute a barrier to trade. One purpose of labelling, recognised by Codex and by national authorities, is to identify food products, and to avoid deceptive trade practices. .. Consumers International believes that labelling for BST use does, in fact, permit consumers to avoid buying milk and milk products so labelled. Considering that a ‘trade barrier’, however, perverts the term. .. Consumers have a fundamental right to decide which risks and which social impacts of products they should accept, and which attributes of products are important to them.’ (CCGP, 1999e). Precisely the same arguments are already being heard in the discussion about genetically modified foods.

Of course, a requirement to label does place a responsibility upon the food industry to monitor so as to ensure that products are correctly labelled. The French government in their submission to the CCGP on BST issues uses this as one of their reasons for opposing the adoption of a Codex standard on BST in that it would ‘go against fair trade since it would then be necessary to set up a very costly guarantee and control system in the dairy product processing business to ensure fair practices.’ (CCGP, 1999d).

It is apparent that the fundamental approach to the development of international food standards is being argued about based on two major issues where there are many scientific, technical, social and cultural aspects involved. The adoption of the residue levels for the meat hormones was difficult but was finally agreed by Codex in 1995. It has subsequently been carried forward into the WTO dispute settlement procedure and, at the time of writing, the differences between Europe and North America are still unresolved. The discussion on the milk hormone has taken even longer and continues to occupy much Codex time. The same parties are in dispute although Canada has in practice currently imposed a national ban on the use of BST for animal welfare reasons.

In a bid to try to define more clearly the issues involved, Codex in 1995 adopted the Statements of Principle which are now supposed to define the approach of Codex to its work. However, the use of the term ‘other legitimate factors’ continues to cause uncertainty.

From one perspective, the application of science to the evaluation of the meat and milk hormones is no different from any other consideration. The argument, simply stated, is that having conducted a safety assessment, on the basis of the result, acceptable residue levels should be set. Opposition to the setting of the residue limit must be motivated by other factors. Since the opposition results is a loss of potential market to the manufacturers of the hormones, it is presumed that the motivation is economic protection. Any other arguments are considered to be an attempt to disguise this prime factor.

From the other perspective, rapid evaluation of the scientific evidence and a reliance on data which may in the future prove insufficient creates a risk. When considering the food supply, consumers do not wish to take risks and expect governments to protect them from these. The level of risk which is acceptable is difficult to define but where the use of the substance confers little benefit, even a low risk may be considered unacceptable by many consumers. People who promote the use of hormones in the face of consumer opposition are thought to be motivated by greed and a excessive determination to make money.

In both cases, there is a perception that the appropriate procedures are being manipulated. The industry is seen to be attempting to manipulate the scientific evaluation process and consumer activists are accused of manipulating the media and promoting scare stories to influence consumer opinion. The correct position is likely to be that both views have some element of truth. For the adoption of international standards, where the resulting standard can be used to interpret provisions of international treaties, the resulting dispute tends to divide the parties rather than move people together towards a consensus.

The use of labelling might be regarded as placing an excessive burden upon industry but it does provide an opportunity for consumers to express a choice. Two alternative approaches are actually available and may not have been fully explored in the debate. Consumer organisations have been pressing for products to be labelled with an indication that they come from treated animals. This is the approach now being used within Europe for GM ingredients - products have to be labelled as containing ‘genetically modified maize/soya’. The alternative is to indicate on the label that a product is believed to be free from ingredients derived from treated animals. This approach is the equivalent of the marketing of organic foods - products which have been produced under specified organic conditions may be labelled ‘organic’. In the same way, ‘hormone-free meat/milk’ could provide a marketing advantage.

The question for Codex is whether, when faced with a difficult and complex issue, it is better to avoid the adoption of the standard and recognise that the parties do not agree. From the American perspective, as quoted above, this could have ‘profound, far-reaching and possibly fatal consequences’. It is however difficult to see what alternative approach can be taken by the European governments. There is undeniable consumer resistance to scientific advances in the food supply - the European Parliament and national Parliaments find it hard to ignore this and seek means to reassure their constituents.

What is apparent from all of the above considerations is that science is not the only determining factor is the establishment of international food standards. Societies are willing to accept the role of science in the process. The level of acceptance does however vary and no amount of discussion within Codex will be able to alter this fact.