2.1. FAO Manual on the submission and evaluation of pesticide residues data for the estimation of maximum residue levels in food and feed

The Meeting was pleased to note that the FAO Manual on the Submission and Evaluation of Pesticide Residues Data for the Estimation of Maximum Residue Levels in Food and Feed had been published.

The Manual gives the historical background to the operation of the JMPR and describes the object of the work, the procedures involved in the selection of compounds, and the data requirements for estimating maximum residue levels and supervised field trials median residue (STMR) levels.

The Manual incorporates all relevant information on the principles which are currently used by the JMPR to estimate maximum residue levels and STMR levels and clarifies and consolidates the procedures followed by the FAO Panel in the evaluation of the experimental results and information provided. In this way it improves the transparency of the work of the JMPR and facilitates the acceptance of Codex MRLs by national governments and their use within the WTO Agreement on the Application of Sanitary and Phytosanitary Measures.

The principles in the Manual may also provide useful assistance to member countries in evaluating residue data for the registration of pesticides and in developing their national evaluation systems.

Chapter headings in the Manual comprise:

· Selection of compounds for evaluation
· Data and information required for JMPR evaluations
· Preparation of data submissions for the consideration of the FAO Panel of JMPR
· The JMPR practice in estimating maximum residues levels and proposing maximum residue limits
· Estimation of residue levels for calculation of dietary intake of pesticide residues
· Use of JMPR Recommendations by regulatory authorities.

Appendices provide information on abbreviations, definitions, sampling and the composition of animal feeds, as well as guidance on procedures and formats for the submissions of data.

The Meeting wishes to encourage manufacturers and member countries to follow, as far as possible, the guidance given in the Manual in preparing future submissions for the JMPR.

Because guidelines by their nature are subject to revision from time to time in order to accommodate new scientific developments and standards, users of the FAO Manual are urged to keep abreast of these changes by reading future JMPR reports where such updates are recorded.

2.2 Submission of information for consideration by the FAO Panel

In view of the publication of the FAO Manual (see section 2.1), the Meeting considers that the preparation of future submissions will be much more consistent and will further facilitate the evaluation of the experimental data and information provided. In order to assist the efficient and timely estimation of maximum residue levels, some of the deficiencies and problems experienced by the current Meeting in relation to recent submissions are described below.

The Meeting repeatedly found it very difficult or impossible to interpret the summary information on GAP supplied by manufacturers (FAO Manual, 3.1.4.1). The summary should not include any information on the use which is not specifically given on the label (e.g. do not define applications in terms of kg ai/hl if only the kg ai/ha rate is specified; do not specify calculated PHIs if application is authorized only at a specific growth stage; do not give a number of applications calculated from the specified application intervals and PHI). It is stressed (FAO Manual, 3.1.4, para 2) that valid copies of current labels must be provided, together with an English translation of the relevant sections (e.g. dosage, specifying whether the concentration of the spray or the kg/ha rate is primarily defined; application methods; growth stage of plants at the time of application of the pesticide; conditions of use; any restriction of the use). Crops included in groups (e.g. leafy vegetables, fruits) should be individually named. Labels reflecting current GAP should be clearly distinguished from "proposed" labels. Indexing of labels in such a manner as to allow easy cross-reference to GAP summaries and supervised field trials would facilitate the evaluation.

In future, the specific uses of a compound will not be evaluated if the relevant labels have not been provided.

If information on GAP is provided by responsible national regulatory authorities the above detailed information is required and the submission of labels is desirable. The submission of information on GAP by national authorities is especially important in the case of generic pesticides produced by several manufacturers. In such cases information on the chemical composition of technical products and their formulations used in the reporting country is also desirable.

The description of supervised trials (FAO Manual, 3.1.5) should include: the unambiguous description of the commodity, preferably with Codex Classification Numbers; detailed information on sampling and sample preparation with special emphasis on sample size; duration and temperature of storage during successive steps from sampling to analysis; the clear description of the portion of the commodity prepared for the analysis (e.g. measured relative weights of stone and pulp; measured relative weights of peel and pulp if they are analysed separately); blank values; whether the results were adjusted for blank and recovery values (FAO Manual, page 23).

Descriptions of analytical methods should include (FAO Manual, section 3.1.3), but not be limited to, the clear indication of the compound(s) determined, whether they were free or conjugated, the levels of fortification used in the validation of the methods, the limits of determination, and the recoveries obtained.

Processing studies are among the critical supporting studies which are required for the evaluation of a new compound or a periodic review (FAO Manual, 3.1.6.2). The results of processing studies are used for dietary intake calculations and estimations of maximum residue levels where appropriate. Because the studies submitted are often inadequate, guidance is provided in this section of the Manual, where it is stressed that studies which simulate commercial practices as closely as possible are required. Additional studies such as the effects of household jam preparation and small-scale laboratory studies may be submitted as useful supplementary information, but small-scale laboratory studies will not be accepted as a substitute for properly designed and implemented studies simulating large-scale industrial processes. The processing studies should always be on the raw agricultural commodity as it is marketed (e.g. a processing study on residues in fruits should always include the determination of the residues in or on the original unwashed fruits). The studies should be reported according to well-established guidelines such as the US EPA Hazard Evaluation Division's Standard Evaluation Procedure Magnitude of the Residue: Processed Food/Feed Studies EPA 540/9-86-145, or the Commission of the European Communities’ "Guidelines for the generation of data concening residues as provided as provided in Annex II, part A, section 6 and Annex III, section 8 of directive 911/414/EEC concerning the placing of plant protection products on the market." 1607/VI/97. 7th January 1997.

2.3 MRLS for pesticides for which JMPR estimates of dietary intake exceed the ADI

PROPOSAL

That MRLs for which the available information is insufficient for the JMPR to conclude that the ADI would not be likely to be exceeded be designated as MRLMs (maximum residue limits for monitoring)¹.

¹ Maximum residue levels for use by government laboratories in their monitoring and/or enforcement activities. The JMPR estimates of dietary exposure based on these residue levels exceed the ADI.

BACKGROUND

The Codex Alimentarius Commission has agreed in principle to recommendations of the 1995 FAO/WHO Consultation² on strengthening the consideration of risk assessment in the elaboration and use of Codex MRLs, in particular making a clear distinction between risk assessment and risk management responsibilities. The subject was considered further at the 1997 FAO/WHO Consultation on Risk Management and Food Safety³. For pesticide residues in food, Codex has the primary responsibility at the international level for risk management and the JMPR for risk assessment. Both have attempted to implement steps to effect the recommendations of the two Consultations.

² FAO/WHO. 1995. Application of risk analysis to food standards issues. Report of a Joint FAO/WHO Expert Consultation, 13-17 March 1995. Geneva. WHO/FNU/FOS/95.3 (E.S.T). World Health Organization, Geneva.

³ FAO/WHO. 1997. Risk management and Food Safety. Report of a Joint FA/WHO Expert Consultation, 27-31, January 1997. Rome. FAO, Rome, Italy.

Until recently, estimates by the JMPR of the chronic dietary intake of pesticides have generally been gross over-estimates. Recently introduced procedures using supervised trials median residues (STMRs) and supervised trials median residues for processed products (STMR-Ps) have substantially improved the ability of the JMPR to make better estimates of dietary intake at the international level and thus improve the value of those estimates for risk assessment and risk management decisions.

When the JMPR cannot conclude that the dietary intake will not exceed the ADI (e.g. owing to the lack of sufficient data for the more refined estimation of dietary intake) uncertainty remains as to the toxicological significance of residues in food resulting from the application of pesticides according to GAP. This has resulted in the reluctance of countries to accept some MRLs recommended by the JMPR at the national level, has created risk management problems in Codex, and has the potential to create problems in the use of Codex MRLs in the WTO for the resolution of trade disputes.

Once recommendations by the JMPR for MRLs are taken up in the Codex procedure, the draft MRLs for chemicals whose ADIs might be exceeded are not readily distinguishable from those whose ADIs would not be exceeded according to the dietary intakes estimated by the JMPR. An "implicit" JMPR endorsement is therefore attached to MRLs of both types, although one is clearly a better estimate for risk management decisions than the other. There is clearly a need to ensure that risk managers in Codex and at the national level are able to recognize which dietary estimates have more, and which less, uncertainty. In view of the importance of Codex standards in WTO mechanisms for the resolution of disputes concerning pesticides, the JMPR considers it especially important that its estimates of dietary intake and their relation to the ADI be clearly conveyed to Codex in such a way that potential health concerns are not overlooked.

One approach would be for the JMPR to continue to estimate maximum residue levels when there is insufficient information to determine that the ADI would not be exceeded, but not to recommend them to Codex for use as MRLs. That would provide the strongest incentive for the submission of better data for the estimation of dietary intake. However in some cases this might not be equitable for those countries who are capable of making more refined national estimates of dietary intake than are possible at the international level, or those for which there would be no dietary intake problem simply because their food consumption does not include significant quantities of those commodities which contribute most to the total dietary intake of the pesticide.

Another option (and the one preferred by the JMPR) is to coin another term in order to distinguish those compounds whose intakes can be concluded to be below their ADIs from those for which there is insufficient information to come to that conclusion. The JMPR believes that this approach would better convey the significance of JMPR estimates of dietary intake and would contribute to better risk management decisions within Codex and at the national level. The JMPR therefore proposes a new term: MRLM (maximum residue limit for monitoring), to be defined as an MRL for a pesticide for which an ADI has been allocated, but for which insufficient information has been provided for the JMPR to estimate whether its dietary intake would be below the ADI. The JMPR recommends that MRLMs be applied to new or periodic review chemicals reviewed by future FAO Panels of the JMPR, and that they be clearly indicated as such, and further recommends that the information needed for the JMPR to refine its estimates of dietary intakes continue to be clearly stated in JMPR reports and evaluations.

2.4 The estimation of maximum residue and STMR levels for products of animal origin when residues are transferred from feed items

1. The 1996 Meeting agreed that guidance was required on the estimation of maximum residue levels and STMRs for products of animal origin to aid in the preparation of future JMPR evaluations and to improve the transparency of the procedures. The Meeting agreed that this guidance should supersede the recommendation made in the report of the FAO Panel Workshop¹, which met in the Hague in April 1996.

¹ Anon 1996. Report of an informal workshop on data evaluation in the estimation of dietary intake of pesticide residues for the JMPR. The Hague, Netherlands, April 1996.

2. The present Meeting recognized that it is appropriate to use a different residue value for the estimation of acute intake from that used for the estimation of chronic intake. For chronic dietary exposure the York Consultation² had developed the concept of using the median residues from supervised field trials (STMRs) to represent the most likely residue level if the pesticide is used according to maximum GAP conditions. However, the scope for the estimation of a median residue for products of animal origin directly from the feeding trials is limited, since usually only one or very few feeding studies are available at an appropriate level.

² Anon 1996. Report of an Joint FAO/WHO Consultation on the Revision of the guidelines for the estimation of dietary intake of pesticide residues. York, UK, May 1995.

3. The Meeting agreed that the problems of estimating STMRs and maximum residue levels for products of animal origin are different from those of estimating these values for crop commodities from supervised residue trials. For example, the main residue in feed items may be a metabolite and this may need to be taken into account. The Meeting concluded therefore that this recommendation of the Workshop, although providing useful guidance, would not be appropriate in all situations.

4. The Meeting noted that the continuous consumption of feed items containing residues at the MRL is unlikely and therefore recognized that the magnitude of the residues in animal products following the consumption of treated feed items would depend on the rate at which the compound reached a steady state in tissues, milk and eggs. The residues of some compounds transfer into milk or eggs fairly rapidly, whereas those of other compounds do not reach a plateau unless dosing is sustained for 1-2 weeks or more.

5. The Meeting agreed that in deciding which dosing level to use from feeding studies on farm animals, account must be taken of whether the dosing levels in the feeding studies are expressed on a wet or a dry weight basis. The Meeting recommended that reports of future feeding studies should record the feeding levels primarily on a dry weight basis. It agreed that interpolation between dosing levels from the feeding studies was appropriate in cases where reasonable linearity of the relation between residue level and dosing level could be assumed (where three or more dosing levels show reasonable linearity), whereas extrapolation substantially beyond the range of dose levels was not appropriate.

6. The Meeting agreed that maximum residue levels and STMRs for products of animal origin should normally be derived as shown below and summarized in Table 1.

a) For compounds which reach a plateau rapidly in milk or eggs.

Maximum residue levels should be calculated from the MRL for each feed item, the maximum feed incorporation rates and the highest residues determined in the milk, eggs and tissues in the feeding studies.

STMRs should be calculated from the STMR for each feed item, the maximum feed incorporation rates and the mean residues determined in the milk, eggs and tissues in the feeding studies.

b) For compounds which reach a plateau slowly in milk or eggs.

Maximum residue levels should be calculated from the STMR for each feed item, the maximum feed incorporation rates and the highest residues determined in the milk, eggs and tissues in the feeding studies.

STMRs should be calculated from the STMR for each feed item, the maximum feed incorporation rates and the mean residues determined in the milk, eggs and tissues in the feeding studies.

In deciding between options a) and b), the persistence of the residue in the feed items should be taken into account.

7. The Meeting recommended that worked examples should be developed in time for the 1998 JMPR.

Table 1. Summary of procedures for the estimation of maximum residue levels and STMRs for products of animal origin when residues are transferred from feed items.

	Residue reaches plateau rapidly		Residue reaches plateau slowly
	Max. Residue level	STMR	Max. residue level	STMR
Feed item residue level	MRL	STMR	STMR	STMR
Feed incorporation rates	maximum	maximum	maximum	maximum
Feeding study residue level	highest	mean	highest	mean

2.5 Extrapolation of residue data to minor crops

The 29th Session of the CCPR (1997, ALINORM 97/24A paragraph 99.3) requested the JMPR to give particular consideration to the concerns of developing countries when elaborating criteria for extrapolating residue data to minor crops in the proposed revision of the FAO Manual on the Submission and Evaluation of Residue Data with particular attention to the commodities identified in CX/PR 97/16 Appendix 1 and CX/PR 97/17 Appendix 1, question 4.

The issue was previously considered by the 1989 JMPR (report, Section 2.11) and briefly by the 1996 JMPR (report, Section 2.5) as part of the estimation of group maximum residue levels. The report of the 1996 JMPR formed the basis for Section 5.4.2 of the FAO Manual. The 1996 JMPR also drew attention to the development of minimum data requirements which are under consideration by governments, industry and the Organisation for Economic Cooperation and Development (OECD). Minimum data requirements and situations where extrapolation is valid are closely related.

The 1989 JMPR confirmed that decisions to extrapolate are on a case-by-case basis when adequate relevant information is available. Adequate information would include information on GAP for the relevant crops, a reference to the residue data used to support the original MRL, and an explanation of the logic for the extrapolation.

The FAO Manual provides advice on extrapolations and the nature of the information needed to support an extrapolation.

Section 5.1.4.2 of the Manual states that selective surveys may provide supplementary information which will assist in extrapolations. The approved use pattern of the pesticide on the minor crop should be the same as or similar to that on a major crop for which an MRL is already adequately supported by data. The selective survey should be fully documented and should focus on samples of the minor crop produced under typical commercial conditions where the pesticide is known to have been used.

Section 5.4.2 discusses the estimation of group maximum residue levels and gives examples and limitations. Adequate data for the major crops of a group may be sufficient to estimate maximum residue levels for the whole group.

The Meeting examined the list of pesticides frequently encountered in commodities from developing countries in Appendix 1 of document CX/PR 97/16. Many of the commodities listed, e.g. apples, grapes, oranges, potatoes, are major commodities and extrapolation to major crops is generally not acceptable. Where residues are a problem or potential problem for trade, national governments are invited to submit data for review by the JMPR. A number of the pesticides listed in document CX/PR 97/16 are scheduled for periodic review in the near future and the ideal time to submit data on a pesticide is at the time of its periodic review.

The data submitted to support extrapolation to a minor crop must include the following information.

1. Background information on the reasons for describing the crop as minor, the importance of the use of the pesticide in terms of pests controlled, the extent of its use on the minor crop, and the nature of the problems or potential problems for international trade

2. A description of the cultural practices for the production of the major crop and the approved or registered uses of the pesticide on the major crop from which extrapolation is proposed.

3. A description of the cultural practices for the production of the minor crop, the approved or registered uses of the pesticide on the minor crop, and the reasons for expecting similar residue levels on the minor crop to those on the major crop.

4. Supervised residue trials on the major crop supporting the MRL or reference to the JMPR Evaluations if trials data have already been reviewed by the JMPR.

The data submission should also include the following supporting information where available.

1. Data on supervised trials with approved or registered uses on the minor crop.

2. A copy of the label describing the registered or approved uses and an English translation of the instructions for use.

3. Monitoring data from selective surveys on the minor crop produced under typical commercial conditions where the pesticide is known to have been used.

The Meeting recommended that the CCPR request national governments to provide information on situations where extrapolation of residue data to minor crops is considered feasible at the national level.

The Meeting welcomed the initiative by the CCPR ad hoc Working Group on Problems Relative to Pesticide Residues in Food in Developing Countries and recommended that national governments prepare data submissions for commodities of concern when the specific pesticides are scheduled for review by the JMPR.

2.6 Calculation of dietary intake of pesticide residues

Theoretical Maximum Daily Intakes (TMDIs) were calculated for the JMPR by WHO (GEMS/Food) using the methods described in Guidelines for predicting dietary intake of pesticide residues (WHO, 1997). When information was available, International Estimated Dietary Intakes (IEDIs) were also calculated. Dietary intake assessments were not performed for amitrole or fipronil because no MRLs exist and none have been proposed. The ADI for guazatine has been withdrawn. Excepting fenamiphos and lindane, the intake assessments for all the pesticides evaluated at the present Meeting were below their ADIs. The results are summarized in Annex III. Details of the calculations will be made available at the 30th Session of the CCPR in April 1998.

The TMDI exceeded the ADI of fenamiphos in one of the five GEMS/Food regional diets. However, residue aspects of fenamiphos are scheduled for re-evaluation by the 1999 JMPR, at which time the dietary intake concerns are likely to be resolved. The Meeting recommended that a full re-evaluation of the toxicological and residue aspects of lindane be undertaken at a future Meeting, after which an assessment of intake can take into account the latest information.

Acute hazard intake assessments were not performed on fenthion or methidathion, inter alia, even though acute RfDs were established for them at the present Meeting, because a large portion size database has not yet been established by WHO. An FAO/WHO Consultation on Food Consumption and Exposure Assessment to Chemicals held in February 1997 provided further guidance in calculating acute hazard dietary intake, particularly to address the existence of single commodity units with residues well above the MRL. The Meeting looked forward to the publication of the report of this Consultation.

In response to concern expressed at the 29th Session of the CCPR about the MRL of 3 mg/kg for fenthion in virgin olive oil, the Meeting noted that a 60 kg person would have to consume 200 ml of virgin oil before the acute RfD would be exceeded. In the absence of data on the consumption of virgin olive oil relevant to acute hazard intake assessment at the international level, the Meeting invited governments to make available appropriate information on the consumption of food commodities by high-percentile consumers.