7.1 Safety assessment of pesticides
7.2 Residue studies and recommended MRLs
7.3 Interpretation of residue analytical results in comparison with MRLs
The summaries and evaluations contained in the reports and evaluations are, in most cases, based on unpublished proprietary data submitted for the purpose of the JMPR assessment. In this context the JMPR documents are a unique source of information. Regulatory authorities and other interested specialists are encouraged to make use of the critical evaluations of the JMPR.
The JMPR monographs and reports should be of help to FAO and WHO Member States in the safety assessment of pesticides and their residues. However, two major problems are encountered when some Member States attempt to use these assessments: (1) the JMPR assesses the toxicology of active ingredients and not formulations, which are controlled at the national level, and (2) relationships between the purity and specifications of the active ingredients that were tested and evaluated by the JMPR and technical materials of commerce are often unknown.
The purity of technical active ingredient depends on, among others, the route and conditions of synthesis, the purity of raw materials used for the manufacture, and the packing and storage conditions. The toxicity of certain impurities can be several magnitudes higher than that of the active ingredient, and therefore their presence even in very small concentrations may substantially affect the toxicity of the pesticide product.
The Joint Meeting evaluates toxicological studies on test materials that in most cases correspond to active ingredients that are sold by the company(ies) which provided the data. The purity and specifications of active ingredients that regulatory authorities are asked to approve may or may not correspond to those that were tested and summarised in the JMPR monographs. For this reason, registration authorities should carefully consider the extent of similarity between any active ingredient being considered for registration and the technical material assessed by the Joint Meeting. To be able to make this determination, registration authorities should seek information on manufacturing impurities in pesticide products, as emphasised in Sections 6.2.2 and 6.2.3 of the FAO International Code of Conduct on the Distribution and Use of Pesticides. The safety of other components of formulations should, of course, also be considered when registering pesticides. For these reasons the JMPR does not recommend use of JMPR Evaluations as the sole basis for safety assessment for national registrations.
If the evaluations are used for registration purposes, authorities should use documentation provided by manufacturers in accordance with national laws relating to the submission and use of unpublished proprietary data to ensure that the JMPR evaluations are of pesticides manufactured by the same routes, of comparable purity and with similar impurities to the pesticides that are being registered.
The information relating to pesticide residues (e.g. results of supervised trials, metabolism, animal transfer, processing studies) can be used more generally than the safety assessment of pesticides.
The comparability of the trial conditions discussed in detail in chapters 5 and 6 should be assessed for deciding on the applicability of JMPR conclusions and recommendations for the particular national use conditions.
Codex MRLs are meant to be used primarily to enforce and control compliance with nationally authorised uses of pesticides on commodities moving in international trade. The applicability of Codex MRLs for national use depends on the relation of GAP on which the maximum residue level estimates were based to the national GAP. In making decisions on comparability of national use conditions to the trial conditions described in the monographs, the results of a few supervised trials carried out under typical growing conditions of the country can be of great value.
In accordance with the principles of FAO International Code of Conduct on the Distribution and Use of Pesticides, governments "should promote the use of safe, efficient and cost effective application methods" in order to reduce the exposure of consumers and the environment resulting from the use of pesticides. When the national use conditions lead to substantially lower residues than the Codex MRL, the establishment of lower national MRLs may be considered for enforcing domestic uses since higher MRLs would encourage unauthorised use of the pesticide, which is against the principle of GAP. However, for imported commodities the national authorities have an obligation to accept higher Codex MRLs which afford an acceptable level of consumer protection, in accordance with the provisions laid down in the Sanitary and Phytosanitary (SPS) agreement of the Uruguay Round of GATT.
A question frequently asked is whether the Codex MRLs, which are based on the limits recommended by the JMPR, should be considered either as strict limits or with the allowance of a further margin when considering the analysis of samples for enforcement purposes.
By definition an MRL is a limit not to be exceeded. The burden of proof is on the monitoring authority to establish, with a high degree of assurance, whether the residue in the lot being examined exceeds the MRL in order to make any regulatory actions.
The uncertainty of the analytical results (SR) deriving from the random variation of the consecutive procedures comprises the uncertainties of sampling (SS), sample preparation (SSp) and analysis (SA).
(SR) = Ö [(SS)2 + (SSp)2 + (SA)2]
Since the average residue is the same the equation can be written as:
(CVR) = Ö [(CVS)2 + (CVSp)2 + (CVA)2]
The uncertainty of the final analytical result (CVR) cannot be smaller than that of any step of its measurement.
The experiments show that, on average, the expectable minimum coefficient of variation of residue results of supervised trials is around 0.3-0.4. In this estimate the variation of replicate analyses accounted for is only 10% (Ambrus, 1996b).
However, much larger variations of analytical results are encountered in practice. A summary of answers to the question "What are considered acceptable values for reproducibility (deviations between laboratories)?", sent out by the Working Group on Methods of Analysis of CCPR in 1987, produced the information summarised in Table 7.3.1.
Table 7.3.1. Reproducibility of representative residue methods at different residue levels.
|
Concentration, mg/kg |
Coefficient of variation, % |
|
|
Mean |
Range |
|
|
0.01 |
77 |
20-200 |
|
0.1 |
45 |
10-100 |
|
1 |
22 |
5-50 |
The figures in Table 7.3.1 are typical of the range of errors routinely encountered (using accepted analytical methods) in analyses for pesticide residues at the concentrations indicated. In view of the variability inherent in an analytical method, a decision is needed on which analytical results are required to be sure that the residue concentration in the sampled product exceeds the MRL.
Furthermore, international collaborative studies revealed that, in the comparison of an analytical result with the MRL, accuracy (influenced by mainly systematic errors) is more important than precision.
In order to obtain reliable results, the laboratories performing regulatory enforcement analysis are encouraged to:
· establish internal quality control measures which enable them to assess the within laboratory variation of results;· participate in international sample check programmes to assess the accuracy of their analysis;
· pay attention to information on storage stability of residues and the definition of residues;
· strictly adhere to Codex guidelines for preparing the portion of commodity for analysis;
· validate the sampling procedures used for obtaining samples, and ensure proper training of sampling officers.
The same precautions should be applied in performing supervised trials or selective surveys to provide data for estimating maximum residue levels.
