The possibility of pesticides used in the production and storage of crops remaining on food at the point of consumption has long been acknowledged by government, and therefore there is a comprehensive programme for assessing the safety (or otherwise) of these pesticide residues. This includes both risk assessment before a pesticide is approved for use and a monitoring programme of the foods.

The risk assessment relates only to pesticides used in the UK, and therefore applies to UK-grown food, although increasingly, through harmonisation across the European Union, the same type of assessment will be carried out on all EU-grown food. The monitoring programme includes all food, UK and imported.

Many consumers are concerned about the levels of pesticide in their food and would like to reduce their consumption of pesticide residues, but are not fully aware of what steps are already being taken by the government to ensure that residues do not pose a health risk. The government’s process for considering the risk of residues in our food is a complicated one, not easily understood by someone without a significant level of expertise. The Pesticide Residues Committee (PRC) have produced a leaflet[1] aimed at the public that aims to explain something about residues in food, but is simplified to such an extent that it is not particularly informative. The PRC and the Pesticide Safety Directorate (PSD) provide a wealth of detailed information on their websites about residues in food, but piecing all of this together takes a level of determination and time that many consumers would not necessarily have.

We believe that it is important that the public is well informed so that they can make sensible decisions about what they eat. This briefing paper aims to explain to those who wish to probe a little deeper than the headlines about how the government assesses the risk to the public from eating food contaminated with pesticides, and what PAN UK’s concerns are. It includes a significant amount of detail, but is not completely comprehensive. We refer readers who want to know more to the PRC and PSD websites[2] which were the sources of information for this briefing.

1. The UK programme for monitoring pesticide residues in food
In the UK, the government checks both home-grown and imported foods for residues of pesticides. This monitoring programme is overseen by the Pesticide Residues Committee, a panel of experts and laypeople.

The purpose[3] of the monitoring programme is to check that:

The cost of the monitoring programme is around £2 million, 60% of which is funded by a levy on agrochemicals sales (i.e. paid for by the industry) and 40% by government (i.e. paid for by the taxpayer).

1.1 Maximum Residue Levels
The crucial element of the monitoring programme is the Maximum Residue Level (MRL). This is set to reflect the highest level that is likely to be achieved assuming good agricultural practice (GAP), and is generally determined from data provided by the manufacturer. (Good agricultural practice is difficult to define, but essentially means following the requirements set out on the pesticide label).

MRLs are set at National level, but there is an on-going programme to harmonise all EU countries' MRLs. The list of currently-used MRLs for the UK is given on the PSD website[4].

It is important to note that MRLs are not safety levels. "MRLs are intended primarily as a check that the pesticide is being used correctly (i.e. that the GAP is being observed) and to assist international trade in treated produce[5]". The Pesticide Safety Directorate asserts that "MRLs are not safety limits. They are always set below, often far below, safety limits".[6] This is not true, in fact.

There are many instances where the MRL is above the safety limit. For example, PAN UK has identified a number of MRLs for pesticides on apples that would exceed the allowable daily intake for toddlers, seven of which have turned up at least once over the last six years. For example, a toddler, regularly eating apples containing the pesticide carbaryl at or just below the MRL, could eat more than five times the amount that is thought to be safe. An infant eating one large portion of apple (180 g) containing carbaryl at the MRL could eat one and a half times the amount that is thought to be safe on a single day. Carbaryl is regularly detected as a residue in apples sold in the UK, at levels lower than the MRL, but still, on occasions, exceeding the safety level for daily intake. For more information about how safety levels are calculated, see section 2.

Since we estimate that there are more than 2000 pesticides/food combinations that have established MRLs, and both MRLs and safety levels are subject to change, it would be a considerable amount of work to determine exactly how many MRLs exceed safety limits at any one time. Our on-going research aims to identify those likely to cause particular problems and we will report these on our website.

1.2 Foods Monitored
Different foods are sampled each year, with some staple food items being monitored every year. Potatoes, apples, bread and milk have been sampled every year since 2000. In the last 6 years, much data will have been obtained for these foods, but others may not have been sampled at all. Generally, the number of samples that are taken are influenced by the dietary importance of the food (i.e. how much of it we eat), and the likelihood of residues occurring (with the more important and the most likely to contain residues being sampled more frequently).

In the last 6 years, around 146 food types have been monitored. The total number of samples per head of population taken in the UK was around 0.04 – the lowest in the EU apart from Portugal. Denmark samples more than 5 times as many per head – 0.29 – and finds a lower proportion of samples that do not contain residues.

1.3 Pesticides tested
According to the Pesticide Residues Committee website, there are around 1,000 chemicals that could be tested for. However, the cost of testing for all of these would be excessive, and most of the tests would be unnecessary. A more focused approach is taken by selecting a smaller number of pesticides to test for.

While this is sensible in order to keep the costs and the quantity of data to reasonable quantities, there is a possibility that under-testing occurs, with the presence of some chemicals going unchecked.

Across the EU, there is a wide range in the numbers of pesticides tested for – in the UK the maximum number any sample was checked for during 2003 was 149, whereas in Germany it was 519. The number of pesticides found in the UK was 68 (46% of the maximum number sought) whereas in Germany it was 246 (47%). This suggests that the more you look for, the more you find. In the UK there may be more than 100 pesticides that could be in our food but are never checked for.

1.4 Published Residue Data
The Pesticide Residues Committee publishes quarterly data on the residues measured under their monitoring programme on their website[7]. They also ask the food industry to submit residue data to them and publish that too. This is the most comprehensive source of information about residues available in the UK.

Some retailers[8] may also publish some information about the residues they have detected from their own suppliers, but this is generally less detailed.

The residues will be given as quantity of pesticide per quantity of food, usually expressed as mg/kg. Thus if a residue of 0.5 mg/kg is given for a batch of apples, this means that for every kilogramme of apples eaten, 0.5 milligrammes of pesticide would also be consumed. (a milligramme is a millionth of a kilogramme).

Each sample consists of a specified amount of the food – this might be around 10 apples for example, or several bunches of grapes. The entire fruit is included in the sample – even when the skin or the core is not normally consumed. Because all the food in the sample is blended together for the analysis, the residue level that is measured is an average and the amount of pesticide on individual fruit within the sample will vary.

The quarterly reports include details of the quantities of pesticides found in each sample, and an analysis of the percentage of samples containing residues, the percentage exceeding the MRLs, the brand names of foods exceeding MRLs and an analysis of the health implications for the residues measured.

1.5 Improvements PAN UK would like to see in the monitoring programme:

2. Assessing the intake of pesticide residues in our food
Section one showed that we have some knowledge, from the government’s monitoring programme, of the residues that are found on the food we buy. In addition to this, before a pesticide is approved for use on food crops, tests are undertaken to determine what levels of residues are likely to occur in the crops on which it is to be applied.

According to the Pesticide Safety Directorate, they “set conditions of use to ensure that these residues are not at levels which may cause harm”[9]. It is necessary, therefore to estimate how much of the pesticide present in our food people actually consume.

This process is a complex one, and more details can be found on the Pesticide Residues Committee (PRC) website[10] , although they are not all easy to find. This section aims to outline the basic principles involved, rather than the specific details, starting with the safety assessment that is carried out for a pesticide before it is approved for use.

2.1 Measurement of residues in supervised trials
Before a pesticide is approved for use in the UK, tests have to be made to establish what levels of residues are likely to occur. These tests are carried out by the company applying for product approval under highly controlled conditions, called “supervised trials”. The pesticide will be applied according to “good agricultural practice” (i.e. following the instructions on the pesticide label) and under “worst case conditions”, which means the highest number of applications at the highest dose, and with the shortest harvest interval (the time between the last application and harvest) that the label permits.

Samples of the crop are taken, and the residues present are measured. These trials have to be undertaken on each of the crops for which the pesticide is to be approved, at a range of different sites (different countries, or regions with different climates) The measured residues are then analysed and are used to calculate two particular values – the supervised trials median residue (STMR) and the highest residue (HR). These two values are then used in safety assessments.

2.2 Dietary intake
Next, the amount of the pesticide that might be consumed needs to be estimated. This depends on how much of the particular food on which the residue is found is consumed. In order to estimate this for the UK, dietary surveys of ten different consumer groups are used – adults; infants; toddlers; children 4-6 years; children 7-10 years; children 11-14 years; children 15-18 years; vegetarians; elderly in their own home; elderly in residential care.

Two different types of pesticide intake need to be calculated – firstly, that which results from regularly eating contaminated food (this is known as chronic exposure), and that which results from eating a single portion with a particularly high level of residues (acute exposure).

Dietary surveys are used to estimate the average quantities of each food that we consume per day as well as the “high levels” – which is the quantity that people who eat a particularly large amount of that food might consume. There are in fact two “high levels” – one is the highest amount that some people regularly eat, and the other is the highest amount that someone might eat on a single day.

These “high levels” are calculated on what is called the 97.5th percentile – i.e 97.5% of people are estimated to eat less than that value. However, this means that 2.5% of people eat more than the high level and are potentially not protected by the risk assessment.

The models that are used to estimate pesticide intakes are contained in two spreadsheets that are available on PSD’s website, together with guidance notes, which explain how the calculations are done[11]. They also contain the data on mean and high consumptions from the dietary surveys.

When assessing daily intake (chronic exposure), residues on all the foods that might contain the pesticide will be taken into account. It does not, however, consider intake from imported food, where the same pesticide may be appear on different foods, or from other sources of exposure such as water, and general environmental contamination.

When the intake for a single day (acute exposure) is assessed, only one food type is considered and no other sources of exposure.

2.3 Calculation of pesticide intake
The actual calculations are more complicated than we can show here but the spreadsheets used by PSD are available to be downloaded and can be used to calculate how much pesticide might be consumed.

As a rough guide, the calculation estimates the amount of a pesticide consumed per unit bodyweight as:

Pesticide intake (mg/day) =

quantity of food consumed (kg/day) x quantity of pesticide on that food (mg/kg)
Body weight (kg)

2.4 How good is the risk assessment?
The risk assessment for pesticide residues in food before the pesticide is approved for use is relatively comprehensive and makes a good attempt at estimating intakes. However, there are a number of areas where it makes simplifications without ensuring that “worst cases” are included and PAN UK believes that steps could be taken to make the risks lower. A paper published in 2002 showed, using a less simplistic model, how a significant number of children could be eating more than the safety limit of some pesticides that are present on apples[12].

A much wider range of exposure routes need to be included for chronic exposures. All sources of dietary intake – including imported foods and water – should be included and some assessment of potential environmental exposure should be made. Acute exposures also ignore the underlying daily exposure from regular low-level consumption.

Dietary surveys are based on data that goes back as far as 1986 for infants, and as recent as 2001 for adults. Since infants and children are generally the most vulnerable because of their high consumption to body weight ratio, it is important that their diet is accurately represented. Nutritional advice to parents has changed dramatically in recent years, in particular in encouraging everyone to eat five portions of fruit and vegetables each day and changes in school dinners may also influence diet. We would suggest regularly updating surveys on all consumer groups and also developing an assessment of an “ideal” diet of at least five portions of fruit and vegetable to ensure that those eating more healthily are not exposed to unacceptably high levels of pesticides.

Currently, an estimate of mean body weight is used[13], suggesting that those under average weight may be less well protected. The data should be analysed differently so that mean and high levels of “consumption per unit body weight” are used, rather than mean and high levels of consumption, divided by mean body weight as at present.

The 97.5th percentile levels of consumption fails to protect 2.5% of consumers. We would suggest that 2.5% is too high, that a 99% level of consumption per unit body weight would be preferable and that those who consume larger quantities for their body weight than this should be warned that they may not be protected.

Use of the “median” residue level to calculate long term intakes gives an artificially low value, particularly when less than 50% of samples have detectable residues (see Glossary for an explanation and example). WHO guidelines[14] suggest that “the use of the median or the mean generally gives similar results” which is patently untrue, and also says “the median is preferred because it simplifies calculations …” It may be that agrochemical industry pressure has contributed to this choice of statistical measure – the only reason the median is simpler is because it allows, in many cases, long term intakes to be considered negligible, even when as many as 49% percent of samples contain relatively high levels. A European Commission report on monitoring of pesticide residues in the EU[15] suggests that the arithmetic mean or the 90th percentile values should be used instead.

2.5 Improvements PAN UK would like to see in the assessment of dietary intake
Include all sources of exposure, including all foods, water and environmental routes for both chronic and acute intakes

3. Assessing the safety of pesticide residues in food
In order to assess the safety, or otherwise, of pesticide residues in food either before pesticides are approved for use, or when residues are monitored in the food we eat, some comparison of the quantity of pesticide we consume with the safety levels for that pesticide needs to be made.

This section explains how those safety levels are determined, and how the comparison is made.

3.1 Safety levels for pesticides
There are two safely levels for human consumption of pesticides, the Acceptable Daily Intake (ADI) and the Acute Reference Dose (ARfD).

The ADI is defined as “the amount of a substance which can be ingested every day of an individual's entire lifetime, in the practical certainty, on the basis of all known facts, that no harm will result”.[16] It is determined by feeding different doses of the pesticide to laboratory animals over either two years (for rodents) or one year (for non-rodents)[17] and assessing the highest dose at which no effect is seen. This, the “No Observable Adverse Effect Level, (NOAEL)” is then assumed to be the safe level for those animals.

In order to extrapolate to humans, it is reduced by a factor of 10. Since not all humans are the same, and some (e.g. the very young, the elderly or the already ill) may be more vulnerable than others, another factor of 10 is introduced. No account is taken of the fact that the lifespan of humans is so much more than the two years over which the pesticide is tested, nor of the fact that there may be a reaction in humans that is not observable in animals. Therefore for humans, the ADI will be set at 1/100th of the NOEL for the most sensitive laboratory species tested.

The acute reference dose (ARfD) is defined[16] as “(on the basis of all known facts at the time of the evaluation) an estimate of a chemical substance in food (or drinking water), expressed on a bodyweight basis, that can be ingested over a short period of time, usually during one meal or one day, without appreciable health risk to the consumer.” A set of tests are undertaken to estimate the ARfD similar to those for the ADI, with the NOAEL this time determined over a short period of time – either a single dose or up to 90 days[17]. Again, there is a factor of 100 to extrapolate to humans and to take account of the most vulnerable, but no account is taken of the possibility of effects in humans not being observable in laboratory animals.

All of these tests are carried out with one pesticide at a time. There is no assessment of the safety of one pesticide in the presence of other chemicals.

The ADI is the safety level for how much we can eat every day of our lives without any harmful effect. The ARfD is the safety level for a single high level of consumption. Long term intake is generally more dangerous than a single dose[17] and so the ARfD is usually higher than the ADI, although not all pesticides have an ARfD.

The ADI and ARfD are given as quantity of pesticide divided by body weight and will be expressed as mg/kg.

It is important to note that although residues and safety limits are both given in mg/kg, you cannot compare them directly since the residues are per kg food, and the safety limits are per kg bodyweight.

In order to determine whether a residue is safe or not, we need to know how heavy the person eating it is, and how much of the food he or she consumes, as explained in section 2.

3.2 Safety assessments – during the approval process for pesticides

Acute Exposure
The highest amount that might be consumed on one day of a particular pesticide per unit bodyweight[18] is calculated for each consumer group (infants, toddlers, children, adults, elderly, vegetarians) and compared with the Acute Reference Dose.

Chronic Exposure
The highest amount that might be consumed regularly over a long period of a particular pesticide per unit bodyweight[19] is calculated for each consumer group and compared with the Allowable Daily Intake.

The pesticide will only be approved for the proposed use if the acute exposure is less than the Acute Reference Dose for all consumer groups, and the chronic exposure is less than the Allowable Daily Intake for all consumer groups[20].

3.3 Safety assessments of the food we buy
Similar assessments are made of the food that is tested under the government’s monitoring programme to establish whether the levels of pesticide residues that occur in practice may give unacceptable risks to consumers.

However, the residue levels that are the highest are compared with the acute reference dose; there is no assessment of the long term intake in order to compare with the allowable daily intake.

Samples are regularly detected that have residues that would give a one-off intake greater than the ARfD. Examples of these in 2004 and 2005 include the pesticides cyprodinil, inorganic bromide, methamidophos, azoxystrobin, chlorothalonil and dithiocarbamates, all of which appeared on samples of lettuce at levels that exceed the ARfD. The usual response of the Pesticide Residues Committee is that “there is no cause for concern” and a wide variety of reasons for this is given. No action is taken with the retailer unless the legal limit (MRL) is exceeded (see section 1 for an explanation of the MRL).

3.4 Improvements PAN UK would like to see in the safety assessment

4. Summary of conclusions and recommendations
The way pesticide residues in food are assessed for safety is very complicated, and a simplified explanation, which we have attempted to provide in this briefing, will inevitably result in some omissions. PAN UK welcomes the very open approach that the UK government takes with the risk assessment it carries out for residues in food and is grateful for the published sources of information – particularly the Pesticide Safety Directorate and the Pesticide Residues Committee websites – that have been used in writing this briefing.

During our journey through monitoring and the intake and safety assessment procedures, we have identified a number of areas where we believe there is scope for improvement, to ensure that the risks to consumers are reduced to more acceptable levels and to increase customer confidence in their food. The most important are reiterated here: we would like to see government action to address these.

Glossary
For explanation of other terms, see the glossary on the PSD website[16] where many of these definitions were obtained

Acute exposure Exposure (to a pesticide) in a short time -one day or less

ADI - Acceptable Daily Intake The acceptable daily intake is the amount of a substance which can be ingested every day of an individual's entire lifetime, in the practical certainty, on the basis of all known facts, that no harm will result. The ADI is expressed as milligrams (mg) of chemical per kg body weight of the consumer. The ADI is derived from the most appropriate NOAEL by applying an assessment factor normally 100.

ARfD - Acute Reference Dose This is intended to define (on the basis of all known facts at the time of the evaluation) an estimate of a chemical substance in food (or drinking water), expressed on a bodyweight basis, that can be ingested over a short period of time, usually during one meal or one day, without appreciable health risk to the consumer.

Chronic exposure Exposure (to a pesticide) over a long period of time

GAP - Good Agricultural Practice The way products should be used according to the statutory conditions of approval, which are stated on the label.

Median The middle number of a group of numbers; that is, half the numbers have values that are greater than the median, and half the numbers have values that are less than the median. The average, or arithmetic mean, and is calculated by adding a group of numbers and then dividing by the count of those numbers. For example, for the group of numbers 0,0,0,0,1,2,3,3,6,7 the median is 1.5, the mean is 2.2 and the 90th percentile (i.e. the value that 90% of samples would be expected to be less than and 10% greater than) is 6.1.

MRL - Maximum Residue Level The maximum concentration of a pesticide residue (expressed as mg/kg), permitted in or on food commodities and animal feeds. MRLs are primarily a check that Good Agricultural Practice is being followed and to assist international trade in produce treated with pesticides. MRLs are not safety limits and exposure to residues in excess of an MRL does not automatically imply a hazard to health.

NOAEL - No Observed Adverse Effect Level The highest exposure level in a toxicity study at which there are no statistically significant and/or biologically significant increases in the frequency of adverse effects between the group of animals exposed to the test substance and its respective control group.

PRC - Pesticide Residues Committee An independent Non-Departmental Public Body. The Committee is made up of some members with a technical background (e.g. food safety/toxicological risk assessment), some with experience in the horticulture industry, and some with consumer/lay interests. The PRC is responsible for overseeing the Government's monitoring programme for pesticide residues.

PSD - Pesticide Safety Directorate An Executive Agency of the Department for Environment, Food and Rural Affairs (Defra). responsible for:

Residue The amount of pesticide that may be present in fruit, vegetable and animal products following treatment with that pesticide. In addition to the pesticide that was applied, degradation or reaction products and metabolites that may be of toxicological risk, might also be found. These levels are expressed as milligrams of the chemical in a kilogram of crop (mg/kg), or parts per million (ppm).

Risk Assessment A risk assessment determines whether pesticides at specific levels present a concern for consumer health. Consumer risk assessments are routinely assessed as part of the approval process for pesticides and are based on residue trials. Approval of a pesticide is only recommended whent the assessment shows that consumer risk is acceptable.

References