name:

Dr. Erik Millstone

section:

Nutrition

for:	The Defence

experience:

Food Additive Expert

summary:

The witness provides an expert diagnostic on the use, safety and toxicity of food additives. The witnesses assertions are objective of government and private sector funding and lead to potentially alarming conclusions.
Consequently I am satisfied that, despite the fact that all the additives referred to by the Defendants may lawfully be used in the UK, some of them may be undesirable, and that there is evidence that they may pose a hazard to some consumers.

cv:

Full cv:
(not available for this witness)

full statement:

Introduction

In my capacity as a member of university faculty, I have focussed much of my research work for the past twenty years on a study of the use, regulation, safety and toxicity of food additives. I have published extensively in this field, and am one of the very few researchers in the field in the UK who is independent of both industry and government.

In the course of my research I have concluded that when govermments, and international organisations such as the World Health Organisation and the European Commission, decide which additives to permit and which to forbid, they make their judgements by reference primarily to the results of animal studies. There is, however, very considerable scope for disagreement on the crucial question of how to interpret the results of those studies, and the relevance of data obtained from animals for humans.

supplementary statement:

Letter of 8th December 1993

In response to your request that I provide you with a review of the safety and toxicity of nine food additives, what follows is firstly a general discussion of how food additive toxicology is usually evaluated, and secondly detailed comments on the nine compounds in question.

Since all nine of those compounds may lawfully be used in the food supply in the UK my comments take the form of a critique of the evaluations provided by official national and intenational bodies. This document starts with a general introduction, then a list of abbreviations and a gloassary of terms are provided, and then the nine compounds are dealt with in turn, though I should point out that Sodium Nitrite (E250) and Potassium Nitrate (E252) and discussed together.

General Introduction

In the USA responsibility for regulating food additives lies with the Food and Drug Administration (FDA). Their procedures are significantly different from those which operate in the UK and the EC. The single most important difference is that their activities are subject to the provisions of the Freedom of Information Act, and so are not as secretive as those which occur in Europe, and furthermore their judgements can be, and sometimes are, scrutinised and challenged by Congress and in the civil courts.

While the regulation of food additives has been primarily a national activity, there has been some international co-operation for at least 30 years. With the support of most governments of the industrialised countries. The World Health Organisation (WHO) and the United Nations Food and Agriculture Organisation (FAC) established the Codex Alimentarius Commission in the early 1960s, and Codex (as it is usually called) then established a Joint Expert Committee on Food Additives (JECFA) in the mid-1960s. Of all the formal and official committees involved in the regulation of food additives, JECFA has been most thorough and most open about their evaluations of chemicals. They have published a long series of reports and it is these which have provided the basis for the production of this dictionary. They have, moreover, given a lead to national governments, so that for example the British advisory committees frequently rely upon, refer and defer to the judgements of JECFA.

The Commission of the European Community (EC) is formally advised by the Scientific Committee for Food (SCF). Since the establishment of the SCF we find that the British committees (the FACC, FAC and the CoT) have referred and deferred to the SCF; and that in turn the SCF refers and defers to JECFA. We therefore have a hierarchy of committees from the world-wide body through the European level to the national. This sequence is reflected in the structure of the text given below. A common pattern is to start from JECFA, and then deal with such comments as the SCF and the British committees have made and then we also discuss the views of other scientific commentators who are not part of the official process.

When it comes to judging the safety of a chemical, we have to turn to the science of toxicology. Experimental toxicologists seek to estimate the likely effects of chemicals on the people exposed to them. Toxicology is a very complex and troublesome science, and of course it has to deal not just with food additives but also, for example, with agricultural chemicals such as pesticides, pharmaceutical products, cosmetics and the whole range of chemicals in our environment. The science of toxicology is made up of three distinct sub-disciplines, and an overall evaluation of a chemical must take into account information from each of these.

The first two parts of toxicology do not study the effects on humans, but study instead the effects which chemicals have in model systems, in the hope that these models might shed some light on the likely effects on humans. The first part involves tests conducted on bacteria or cell cultures in glass dishes and the second part comprises tests on whole live animals.

One of the most important reasons for testing the effects of chemicals on bacteria cultured in glass dishes is because it might provide an indication that the chemical under test could cause genetic mutations to the bacteria, and so possibly to humans too. This is important for two reasons firstly because we want to avoid chemicals which would cause mutations to humans: we all want healthy and normal children and we don't want to produce deformed or damaged offspring. Secondly, a chemical wroen can cause a mutation, called a mutagen, might well also cause cancer, in other words it might be a carcinogen. If, by causing mutations in the bacteria, a chemical fails these bacterial culture tests, then we have some reason for supposing that it is not safe. If a chemical is not shown to be harmful in these bacterial tests this does not prove that it is safe, because it might well cause all sorts of harm which these tests cannot detect. It might even cause cancer by a non-genotoxic route.

The second, and in effect the most important part of toxicology, concerns the testing or chemicals on whole live animals. A range of different species are used including dogs, cats, monkeys, ferrets, rabbits and of course guinea-pigs, but most commonly it is mice and rats which are used. A wide range of different tests comprise this overall group. You can feed chemicals to animals at high or low doses for long or short periods, and even over several generations. You can inject it under their skin, or directly into their digestive tract, or you can mix it into their food. Whatever test is performed, it is important to monitor a wide range of indicators of the animals' health, and to investigate carefully their remains after they die or are killed. Animal tests provide the main source of information for industry and governments, and most official evaluations of food additives are based on the interpretation of the results of tests with animals.

The third part of toxicology is information on the direct effects which chemicals have on humans. In principle, but not yet in practice, this could be the most crucial segment of the science. It consists or two parts, namely epidemiology and clinical reports. The term epidemiology has the same root as the word `epidemic', and it is the study of patterns of health and illness within and between human populations. Clinical information is provided by doctors and hospitals when they report the effects of chemicals on particular individuals who come to them as patients, or very occasionally as experimental subjects.

If human evidence is available it can be particularly informative because it tells us directly what we want to know, whereas the other parts of toxicology give us information about non-human organisms from which we have to draw inferences as to the likely effects of the chemicals on people. One major problem, however, is that the methods and techniques of epidemiology are currently too insensitive to enable us to detect the effects which additives may be having over the tong term. We consume many different chemicals, in numerous complex combinations, and only rarely can we distinguish two population groups which differ only in that one group avoids a chemical which the other consumes. Consequently the science of epidemiology has very little to contribute to food additive toxicology. Clinical studies have their limitations because the people who are investigated are at best a very small and unrepresentative sample of the population as a whole. As a result almost all our evaluations of the possible effects of food addittives have to depend on inferences drawn from toxicological models. The most fundamental problem with human studies is precisely that we want to know what effects chemicals would have on humans before they come into use, and we don't want to have to wait until after people have been poisoned.

When they evaluate and review the safety of a food additive, international agencies and national governments often review the same information but they usually make different kinds of judgements. JECFA and the SCF consider specific chemicals in abstraction from their uses, and seek to assess how much if any, of a chemical can be safely consumed in food and drinks on a regular basis. National agencies consider, in the light of the evaluations provided by JECFA and the SCF, whether to approve the use of specific food additives in specific groups of products.

The main regulatory concept employed by both JECFA and the SCF is the notion of an 'acceptable daily intake' or ADI. Underlying their approach is the assumption that the results of animal tests can provide a reliable guide to the likely effects which chemicals will have on humans. JECFA and the SCF define an ADI as a dose level at which the chemical is safe for humans. It is derived from a dose level at which it can be administered to a group of animals without causing any observable adverse effects to those animals. In the animal tests they seek to identify what is called a `no effect level' or NEL, and they define the ADI directly in relation to the NEL.

To be precise, the ADI is equal to the NEL divided by a safety factor (SF). According to the textbook accounts, this safety factor is normally set at 100. This figure is obtained by assuming that humans as a whole may be 10 times more sensitive to chemicals than the laboratory animals, and that there may be a 10-fold variation in sensitivity within a population of humans. The safety factor of 100 is then simply the product of the two factors of 10. In practice, we find that when different safety factors are employed by JECFA and the SCF, but these other numbers are no less arbitrary than the figure of 100. The units in which both the NEL's and ADI's are specified is milligrams of chemical per kilogram of body weight of the animal or human, and in this dictionary, as in the official publications, it is abbreviated as `mg/kg bw'. In exceptional cases, instead of setting an ADI, JECFA will approve the use of a chemical up to a particular level of use, in parts per million (ppm) by weight of the food.

In the reports of JECFA and the SCF we find brief summaries of the results of toxicological tests, and wherever possible an NEL is identified, and from it an ADI is set, but it is important to remember that these numbers change through time, and the two committees do not always agree with each other.

The British advisory committees have only recently started to set ADIs, and so far they have set only a very few. The CoT also lists chemicals into five groups, from A to E. These are defined as follows:

• Group A: Substances that the available evidence suggests are acceptable for use in food.

• Group B: Substances that on the available evidence may be regarded meanwhile as provisionally acceptable for use in food, but about which further information is necessary and which must be reviewed within a specified period of time.

  Group C: Substances for which the available evidence suggests possible toxicity and which ought not to be allowed in food without further evidence establishing their acceptability.

  Group D: Substances for which the available evidence suggests probable toxicity and which ought not to be allowed in food.

  Group E: Substances for which the avaliable evidence is inadequate to enable an opinion to be expressed as to their suitability for use in food.

For those chemicals which the CoT list in Groups A or B, the FACC (or as it has now become, the FAC) will recommend to ministers that they approve the inclusion of those substances in specific groups of products.

Toxicology is a very uncertain and unreliable science. It is not possible to take at face value all of the information which is generated by each of the three parts of toxicology. There are innumerable contradictions both within and between the results of different studies in all parts of toxicology. There is no consensus amongst experts, and the availabie evidence is not by itself sufficient to enable experts to settle the question of what should be permitted and what banned.

From the point of view of the consuming public, the crucial question is: who is getting the benefit of the doubt? If experts, officials and government are giving the benefit of the doubt to consumers then they will assume that a chemical is toxic until it is proved to be safe, but if they are giving the benefit of the doubt to industry then they will assume it to be safe until proved harmful. The text which follows is based on the assumption that consumers deserve the benefit of the doubt.

List Of Abbreviations