3. Dietary risk assessment for pesticide residues in food

Assessment of risk from long-term dietary intake

Risks associated with long-term dietary intake were assessed for compounds for which MRLs were recommended and STMRs estimated at the present Meeting. Dietary intakes were calculated by multiplying the concentrations of residues (STMRs, STMR-Ps or recommended MRLs) by the average daily per caput consumption estimated for each commodity on the basis of the GEMS/Food diet^{[10] [11] [12]}. Theoretical maximum daily intakes (TMDIs) were calculated when only recommended or existing MRLs were available, international estimated daily intakes (IEDIs) were derived when STMR or STMR-P values could be used. Dietary intakes were estimated from combinations of recommended MRLs and STMR or STMR-P values. Codex MRLs that have been recommended by the JMPR for withdrawal were not included in the estimations.

Long-term dietary intakes are expressed as a percentage of the ADI for a 60-kg person, with the exception of the intake calculated for the Far East, in which a body weight of 55 kg is used^[13]. The estimates are summarized in Table 1. Percentages up to and including 100% are rounded to one significant figure and values above 100% to two significant figures. Percentages above 100 should not necessarily be interpreted as giving rise to a health concern because of the conservative assumptions used in the assessments. The detailed calculations of long-term dietary intakes are given in Annex 3.

The dietary intake of thiophanate-methyl was considered together with carbendazim, as the compound is determined as carbendazim, which also has the lower ADI.

Malathion and phosmet were considered by the Meeting only for the assessment of the acute RfD. Risk assessments of long-term dietary intake for these compounds were considered by previous meetings (1999 for malathion and 2002 for phosmet).

The evaluation of parathion-methyl considered only nectarines, which were included with peaches in the assessment of the 2000 JMPR.

No new recommendations were made for diphenylamine, which was last evaluated by the 2001 JMPR.

Estimates for residues of carbofuran arising from the use of carbosulfan made at this Meeting do not affect the assessment conducted by the 2002 JMPR for this compound.

Calculations of dietary intake can be further refined at the national level by taking into account more detailed information, as described in the Guidelines for predicting intake of pesticide residues^[14].

Table 1. Summary of long-term dietary of risk assessments conducted by the 2003 JMPR.

Assessment of risk from short-term dietary intake

Risks associated with short-term dietary intake were assessed for compounds for which STMR and HR values were estimated at the present Meeting and for which acute reference doses (acute RfDs) had been established, in commodities for which data on consumption were available. The procedures for calculating the short-term intake were defined primarily in 1997 at an FAO/WHO Geneva Consultation², refined at the International Conference on Pesticide Residues Variability and Acute Dietary Risk Assessment sponsored by the Pesticide Safety Directorate^[15] and at subsequent JMPR Meetings. Data on the consumption of large portions were provided by the governments of Australia, France, The Netherlands, Japan, South Africa, the UK and the USA. Data on unit weights and per cent edible portions were provided by the governments of France, Sweden, the UK and the USA. The body weights of adults and children aged £6 years were provided by the governments of Australia, France, The Netherlands, South Africa, the UK and the USA. The consumption, unit weight and body weight data used for the short-term intake calculation were compiled by GEMS/FOOD and are available at www.who.int/foodsafety/chem/acute data. The documents are dated 01/01/2003 (large portions and body weights) and 05/02/2003 (unit weights).

International estimated short-term intake (IESTI)

Depending on the data on consumption, the IESTI for each commodity is calculated from the equation defined for each case, as shown below. The following definitions apply to all equations.

Case 1

The concentration of residue in a composite sample (raw or processed) reflects that in a meal-sized portion of the commodity (unit weight is < 25 g). This case also applies to meat, liver, kidney, edible offal and eggs, and for grains, oil seed and pulses when the estimate of the HR or HR-P was based on post-harvest use of the pesticide.

Case 2

The meal-sized portion, such as a single piece of fruit or vegetable, might have a higher residue than the composite (unit weight of the whole portion is > 25 g). A default variability factor of 3 is applied in the equations (see Section 2.10). When sufficient data are available on residues in single units to calculate a more realistic variability factor for a commodity, the calculated value replaces the default value.

When data are available on residues in a single unit allowing estimation of the highest residue in a single unit, this value should be used in the first term in the numerator of the equation for case 2a, with no variability factor. The HR value derived from data on composite samples should be used in the second term. For case 2b, the estimated highest residue in a single unit should be used in the equation with no variability factor.

Case 2a

The unit weight of the whole portion is lower than that of the large portion, LP.

Case 2b

The unit weight of the whole portion is higher than that of the large portion, LP.

Case 3

When a processed commodity is bulked or blended, the STMR-P value represents the probable highest residue. This case also applies to milk, and to grains, oil seed and pulses when the estimate of the STMR or STMR-P was based on pre-harvest use of the pesticide.

A risk assessment for short-term dietary intake was conducted for each commodity-compound combination by assessing the IESTI as a percentage of the acute RfD. When the maximum residue level was estimated for a Codex commodity group (e.g. citrus fruits), intakes were calculated for individual commodities within the group. The selected commodities should include the one(s) that would lead to the highest intake.

The Meeting concluded that acute RfDs might be necessary for pirimiphos-methyl, carbendazim and thiophanate-methyl, but these have not yet been established. The Meeting recommended that these compounds should be evaluated for the establishment of acute RfDs in the near future.

Acute RfDs were established for pyraclostrobin, terbufos and paraquat, but short-term intakes were not calculated as information on STMRs and HRs was not available for these compounds.

On the basis of data received by the present Meeting, the establishment of an acute RfD for cyprodinil was considered to be unnecessary. The 1998 JMPR concluded that an acute RfD was unnecessary for dicloran. The intake of these compounds was therefore not estimated.

The short-term intakes as percentages of the acute RfDs for the general population and for children are summarized in Table 2. They are rounded to one significant figure for values up to and including 100%, to two significant figures for values above 100%. Percentages above 100 should not necessarily be interpreted as giving rise to a health concern because of the conservative assumptions used in the assessments. The detailed calculations of short-term dietary intakes are given in Annex 4.

Table 2. Summary of short-term dietary risk assessments conducted by the 2003 JMPR.

¹ from the use of carbosulfan

Refinement of the short-term dietary risk assessment

In view of the default variability factor of 3 which the Meeting agreed to use for the calculation of Case 2 IESTIs (Section 2.10), the Meeting revised all the assessments conducted since 1999 for commodity-compound combinations for which the acute RfD had been exceeded. The results of this process are shown in Table 3. They are rounded to one significant figure for values up to and including 100%, to two significant figures for values above 100% and to 3 significant figures for values above 1000%. Percentages above 100 should not necessarily be interpreted as giving rise to a health concern because of the conservative assumptions used in the assessments. The detailed calculations of short-term dietary intake are given in Annex 4.

Table 3. Summary of short-term dietary of risk assessments conducted by the 2003 JMPR for compounds considered previously, using a variability factor of 3.