[ judgement day logo ]
Justice Bell's Verdict
19th June 1997

Chief Justice Bell's 800 page judgement was handed down on Thursday 19th June 1997 after his presentation of the Summary - the whole judgement is presented here for your enjoyment.

6. McDonald's food, heart disease, cancer of the breast and cancer of the bowel.

The next section of the leaflet, after the sections which deal with starvation in the Third World, the destruction of rainforest and recycled paper, turns to McDonald's food, with the headline:

"What's so unhealthy about McDonald's food?"

    The text beneath the headline reads as follows: "McDonald's try to show in their "Nutrition Guide" (which is full of impressive-looking but really quite irrelevant facts and figures) that mass-produced hamburgers, chips, colas, milkshakes, etc., are a useful and nutritious part of any diet.

    What they don't make clear is that a diet high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals - which describes an average McDonald's meal - is linked with cancers of the breast and bowel, and heart disease. This is accepted medical fact, not a cranky theory. Every year in Britain, heart disease alone causes about 180,000 deaths.

      FAST = JUNK

      Even if they like eating them, most people recognise that processed burgers and synthetic chips, served up in paper and plastic containers, is junk-food. McDonald's prefer the name "fast food". This is not just because it is manufactured and served up as quickly as possible - it has to be eaten quickly too. It's a sign of the junk food quality of Big Macs that people actually hold competitions to see who can eat one in the shortest time.


      Chewing is essential for good health, as it promotes the flow of digestive juices which break down the food

      Page 205

      and send nutrients into the blood. McDonald's food is so lacking in bulk that it is hardly possible to chew it. Even their own figures show that a "quarter pounder" is 48% water. This sort of fake food encourages over-eating, and the high sugar and sodium content can make people develop a kind of addiction -a "craving". That means more profit for McDonald's, but constipation, clogged arteries and heart attacks for many customers".

Beneath that part of the text is the cartoon showing a man or a woman and a cow or steer, held in a burger with the legends "if the slaughterhouse does not get you" and "the junk food will!" in bubble speak."

Again, the front page subtitle "Everything they don't want you to know" and the opening section of text including the words "It's (McDonald's) got a lot to hide" are relevant to the meaning of the leaflet in this part of the case, as are the legends "McCancer", "McDisease" and "McDeadly", printed across McDonald's arches.

The meaning of this part of the leaflet, associating McDonald's food with cancer of the breast and bowel and heart disease was so hotly contested that the parties agreed that I should decide its meaning and some related matters as the trial proceeded, the better to decide just what further evidence to call.

On the 20th November,1995, I therefore gave a reasoned ruling on the meaning of the words complained of in this part of the leaflet and the question of whether they were defamatory of the Plaintiffs.

The meaning pleaded in amended form in the Statement of Claim is as follows:

    "The words in their natural and ordinary meaning meant and were understood to mean that the Plaintiffs and each of them:

    (1) Sell meals which cause cancer of the breast and bowel and heart disease in their customers.

    (2) Despite knowing that this is an accepted

    Page 206

    medical fact, deliberately and dishonestly conceal that fact from the public by publishing nutritional guides which

    (a) suppress that fact; and

    (b) falsely claim that their meals are a useful and nutritional part of any diet."

The Plaintiffs pleaded other, additional meanings, but they added nothing in my view.

Ms Steel, speaking for herself and Mr Morris argued that the leaflet meant: "McDonald's portray their food as a useful and nutritious part of any diet, when the reality is that an average McDonald's meal is high in fat, sugar, animal products and salt (sodium) and low in fibre, vitamins and minerals, and that if your diet is high in fat, sugar, animal products and salt (sodium) and low in fibre, vitamins and minerals, the medical profession considers that you are at greater risk of suffering cancer of the breast and bowel and heart disease. The type of food sold at McDonald's encourages over-eating, because it lacks in bulk and also because its high sugar and sodium content can lead to a craving for this type of food. This increases profits for McDonald's, but when people over-eat this type of food it can lead to problems such as constipation, clogged arteries and heart attacks".

The main arguments of counsel in support of the Plaintiffs' construction of the leaflet, were as follows.

One should envisage the ordinary, reasonable reader of this particular leaflet as a person of average intelligence, and average age (that is not a child or an elderly person) with an average experience of the world (that is not an expert in nutrition).

One should judge the "impression" which the leaflet would make on such a reader who would not hunt for the message as he might with a serious article in a heavyweight newspaper.

The numbers of readers who had sent off for information and received the leaflet was probably small when compared with the numbers of readers who were handed it in the street, but whether the ordinary, reasonable reader received the leaflet at an anti-McDonald's meeting, or in the street or through the post, one did not envisage him sitting down and studying the leaflet carefully

Page 207

or bothering to read it more than once. The leaflet was not a specialist publication. It was directed at the world at large, not at people with a particular interest in nutrition.

The impression made on such a reader would be that eating McDonald's food caused cancer of the breast and bowel and heart disease and that although McDonald's knew this they covered it up in their literature "with a lot of soapy stuff about how nutritious their food is." The main impression would be of the danger of eating McDonald's food, with the deception in second place.

The leaflet was nothing more than a "scare sheet" and those responsible for it could not be surprised if readers drew from it the most derogatory meaning available.

There was no conflict between the headlines and the text. "The headlines perfectly and concisely reflect, as does the cartoon, the sense of the particular passage about nutrition" in the text.

The front page and the introductory passages of the leaflet told the reader that McDonald's deliberate deceptions were to be revealed in the leaflet, and when the reader opened the leaflet out, the McDonald's arches with the words written through them would catch his eye as the leaflet's own summary of the true facts which the reader was going to learn in the leaflet.

The headline to the relevant text plainly told the reader that McDonald's food was unhealthy and that the reader would be told why.

The cartoon appeared beneath the relevant text and told the reader not to eat McDonald's food. "It may kill you".

The whole leaflet asserted certainties, with no trace of a doubt or reservation or even an argument, and it was not susceptible of careful thought at all.

The leaflet confused food with diet, and the ordinary, reasonable reader of the leaflet, the man or woman of average intelligence and knowledge, would not sit down and try to work out the pathways by which McDonald's food might cause ill-health. He would not notice any distinction made in the text between diet on the one hand and food on the other. Although the ordinary

Page 208

reader might know something about issues relating to diet and health the leaflet purported to tell him something he would not know, namely the extent of the danger involved in eating McDonald's food. The message was: "This food is dangerous. You should not eat it."

The main arguments of Ms Steel and Mr Morris, put forward jointly, in support of their construction of this part of the leaflet, were as follows.

The proper context for the remarks made in the first two paragraphs of the text was to rebut McDonald's portrayal of their food as a useful and nutritious part of any diet. The purpose of the text was to show that McDonald's food was not a useful and nutritious part of any diet.

The leaflet did not ascribe motive to McDonald's inaccurate portrayal of the qualities of their food. "It just says that they do not give the full facts".

The McDonald's arches, the headline and the cartoon did not add anything to the meaning of the words in the text nor did they change the meaning of the text in any way.

The words across the McDonald's arches were designed merely to catch the eye and to encourage people to read the rest of the leaflet. People would not read them as statements of fact; but even if they did they would have to ask what they meant because they meant absolutely nothing on their own. "Their meaning must be taken from the text. They rely on the text for an explanation of the meaning". "McCancer", for instance, could mean that McDonald's were a cancer on society, or that you could develop cancer from dioxins through the packaging process or through CFC gases destroying the ozone layer. "McDeadly" could refer to the killing of animals or starvation in the Third World. The arches could not have a stronger meaning than the text which explained them; and the ordinary, reasonable reader would read the text to find out the meaning of the arches and headline.

The cartoon depicted a symbolic crushing of both people and animals by the burger industry consuming and swallowing up anything getting in the way of it making its profits including workers in the industry. The average person in the street would not take the cartoon literally. Dead cows and dead people do not talk and McDonald's do not put human remains in their burgers.

Page 209

The cartoon did not specifically refer to McDonald's. It was consistent with closing passages in the leaflet, which referred to junk-food chains' ruthless exploitation of animals and people. The cartoon was there to break up the text, to catch the eye and to encourage the reader to read the text of the leaflet. The ordinary, reasonable reader would not think very much about it.

The text carefully explained what was meant by the headline, "What's so unhealthy about McDonald's food?"

Having seen the arches, the headline and the cartoon, the ordinary reasonable reader would go on to read the text. No such reader would assume that eating McDonald's food caused cancer, without going on to read the text of the leaflet. Most people were aware of issues of diet and health and they knew that diseases such as cancer and heart disease were not brought about by a meal which they ate or even by a week or a month of "unhealthy" eating. They knew that many, perhaps most, people were not affected at all and that where such diseases did occur they took several years to develop and were influenced by other factors. Moreover, the leaflet was clearly a reference document for people interested in the issues which it raised. It would be read carefully. If there was any confusion in the reader's mind he or she would read it more than once.

For these reasons it was necessary to concentrate on the text to find the leaflet's meaning. The text of the leaflet did not confuse food with diet. It did not say that you took a real risk of developing cancer of the breast or bowel or heart disease if you ate the Plaintiff's food, let alone that eating it caused those diseases.

The leaflet did not say that anything in McDonald's nutrition guide was untrue or a lie or deliberately deceptive. It said that the guide left out the links between diet and degenerative diseases and ill health and, therefore, that McDonald's were not telling the whole story.

The leaflet was not defamatory of the Plaintiffs since it merely criticised their products; it did not allege that they were being deliberately deceptive in their nutrition guide, and it did not allege that they purposely encouraged customers to overeat to their detriment.

Each side argued that the other relied upon the parts of the

Page 210

leaflet which encouraged its own interpretation and ignored the rest.

My conclusions were and are as follows:

The leaflet is a strong attack on McDonald's and their practices, although it also attacks giant corporations in general. It does not pull any punches. It clearly states that McDonald's food is very unhealthy. The headline to the text states this by asking: "What's so unhealthy about McDonald's food?" I note the word "so".

I do not accept that the arches or the cartoon catch the eye only, without delivering a message.

The arches with "McCancer", "McDisease" and "McDeadly" give the message that McDonald's food is dangerous. Those words, by the prefix "Mc", associate cancer, disease and death with McDonald's. Cancer is best known as a disease affecting humans. One finds the word "cancer" in the text only under the headline "What's so unhealthy about McDonald's food?" So in my view the ordinary, reasonable reader would be led to associate cancer and disease with McDonald's food. In my view the ordinary, reasonable reader of the leaflet would be led to associate "McDeadly" both with McDonald's food and with the slaughter of animals to provide that food. "McDeadly" appears above the column of text with the headline "What's so unhealthy about McDonald's food?", although that may have been unplanned.

The meaning of the cartoon with the words "if the slaughterhouse doesn't get you the junk food will", is that McDonald's food will kill you. Slaughterhouses "get" cattle in the sense that they are fatal to them. So the person in the cartoon says in effect that junk food will "get" you in the sense that it will be fatal. McDonald's food is described as junk food in the text above the cartoon. The symbolic sense urged by the Defendants is too strained to be realistic.

I bear in mind that there is a section of the leaflet which deals with food poisoning, but there is nothing which relates the arches or the cartoon to what is said in that section rather than to the column of text headed "What's so unhealthy about McDonald's food?".

On the other hand I consider that the ordinary, reasonable

Page 211

reader of this leaflet would read the text of the leaflet as well as the arches, headline and cartoon. I envisage as that notional reader a person who, handed the leaflet in the street is interested enough to read the leaflet rather than put it in the first litter bin, or who received it through the post because he or she was interested in the causes which Greenpeace (London) espoused, perhaps having asked for more information about McDonald's, or who was interested enough to go to an anti-McDonald's meeting, or who had the leaflet passed on to them by someone who had received it and thought that he or she would like to or should read it. In my judgment that notional reader would read the leaflet quite carefully, and certainly would read quite carefully the parts of the leaflet concerned with McDonald's food, since McDonald's business is food. In my view the leaflet clearly seeks to persuade the reader to abstain from eating McDonald's food, and the ordinary, reasonable reader would read the leaflet more than once if necessary, to understand why he or she should do so.

In my judgment such a reader would see from the text which I have quoted that the danger of McDonald's food was alleged to arise from its effect upon the customer's diet. In my judgment the text makes a distinction between diet and McDonald's meals (both being high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals) which the ordinary, reasonable reader of this leaflet would be alert to between 1987 and 1990 as well as now. Alleged connections between diet and ill-health have been widely broadcast for many years and no one would expect to be adversely affected by the occasional meal.

Despite this, the text does not in my judgment neutralise the message of danger conveyed by the arches, cartoon, and headline. In my judgment it leaves the message of a very real risk of suffering the serious degenerative diseases mentioned, by the effect of McDonald's food upon diet. In my judgment "linked with cancer of the breast and bowel, and heart disease" would mean "causally linked with" those diseases to the ordinary, reasonable reader who would not know of the more guarded use of words like "linked with " adopted by professional experts such as epidemiologists.

I have no doubt that the message taken by the ordinary, reasonable reader would come from arches, cartoon, headline and

Page 212

text together.

Further in my judgment the ordinary, reasonable reader would draw the inference from the text that McDonald's knew of the unhealthy nature of its food and the risk involved in eating it via its possible effect upon the consumer's diet. Why else assert that the matters spoken to are "an accepted medical fact" which, by necessary inference, any large commercial food provider would surely know ?

Moreover, there are clear and strong averments of intentional deception by McDonald's in the form of the man with the mask and the legend "Everything they don't want you to know" on the front page of the leaflet, which the ordinary, reasonable reader would, in my judgment, understand to be continued by the opening words of the relevant text: "McDonald's try to show..."

In my view, the text under "Paying For The Habit" does no more than provide colour and confirmation to the case presented against McDonald's and its food elsewhere in the leaflet, and the only relevance of the text under "Fast = Junk" is to associate McDonald's with the junk food referred to in the cartoon.

For these reasons, and having regard to the impression made on me by the leaflet as a whole, I found in my ruling given on the 20th November,1995, that the leaflet bears the meaning:

    "that McDonald's food is very unhealthy because it is high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, and because eating it may well make your diet high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, with the very real risk that you will suffer cancer of the breast or bowel or heart disease as a result; that McDonald's know this but they do not make it clear; that they still sell the food, and they deceive customers by claiming that their food is a useful and nutritious part of any diet."

When deciding the meaning of the leaflet, the Court is not concerned with the merit or demerit of any possible defence of justification; and I was careful to keep this in mind when making my interlocutory decision on the 20th November,1995, by which time I had heard only part of the evidence on this part of the

Page 213

case, and none of the final arguments on the merits of the Plaintiffs' claims that the defamatory meaning was untrue and the Defendants defence that it was justified.

When all the evidence had been heard and the parties arguments were deployed, it became clear that the meaning which I had found needed elaboration in two respects, so that the merits of the matter could properly be decided.

Firstly, the words "eating it (McDonald's food) may very well make your diet high in fat..." might be thought to mean, "eating it at all, even just once", whereas it is clear from what I have already said that the reader of the leaflet would not expect to be adversely affected by the occasional McDonald's meal, at least so far as degenerative diseases were concerned; nor would he or she expect his or her diet to be adversely affected by the occasional meal. So fairness to the Defendants and loyalty to my reaction to what the relevant parts of the leaflet, taken together, mean, require that the words "eating it" be read as "eating it more than just occasionally" in the meaning as I expressed it on the 20th November,1995. On the other hand there is nothing in the leaflet to suggest to the reader that he or she has to eat McDonald's meals often, once, twice or even several times a week for instance, before diet is affected and there is a very real risk of degenerative disease.

Secondly, the words "very real risk" might be argued to mean very real in the sense of "existing" whereas I meant it in the sense of a serious or substantial risk although falling short of probability; a risk which the ordinary, sensible reader of averagely robust temperament would worry about; not a minimal or bare risk which is there, but which one can get through life without undue concern for.

In my judgment, the meaning which I have found falls within the scope of the meaning pleaded by the Plaintiffs and it is defamatory of both Plaintiffs and damaging to their trading reputations - the First Plaintiff as the corporation with overall responsibility for McDonald's everywhere, and the Second Plaintiff with responsibility for McDonald's in England where publication is complained of. The message goes beyond mere disparagement of the Plaintiffs' food products. Both an allegation that the Plaintiffs sell food which they know to be very unhealthy and an allegation that

Page 214

they nevertheless deceive customers by claiming that their food is a useful and nutritious part of any diet reflect badly on the Plaintiffs' conduct of their trade and businesses and would tend to deter potential customers from buying their food and make other people reluctant to deal with them.

Although the Defendants pleaded alternative defences of justification and fair comment in respect of the words complained of in this part of the case, I see the defamatory meaning as a series of statements of fact and not opinion. The overall sting, that the Plaintiffs' food is very unhealthy and known to be so, could be opinion if expressed as opinion, but in my judgment it is expressed as simple fact. It follows that if the defamatory meaning of this part of the leaflet is to be successfully defended it must be shown to be true, and it seems to me that the essential issues in this part of the case are as follows.

Firstly, is McDonald's food "high" in fat, sugar, animal products and salt (sodium), and "low" in fibre, vitamins and minerals?

Secondly, if so, is it right that eating McDonald's food more than just occasionally might well make your diet high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals?

Thirdly, if so, will it bring the very real risk that you will suffer cancer of the breast or bowel or heart disease as a result, i.e. does a diet high in fat, sugar, animal products and salt (sodium) and low fibre, vitamins and minerals lead to a very real risk of those degenerative diseases .

Fourthly, in the light of the answers to those questions, is McDonald's food very unhealthy as the First and Second Plaintiffs must know?

Finally, do McDonald's (including the First and Second Plaintiffs) knowingly deceive customers by claiming that their food is a nutritious part of any diet despite it being so unhealthy?

Those questions accord with the analysis of counsel for the Plaintiffs and Ms Steel, speaking for herself and Mr Morris.

The evidence on these issues was given by a number of expert witnesses whose opinions one side or the other invited me to

Page 215

adopt. I was also referred to a large number of papers written by experts who did not give evidence.

The witnesses put forward as experts were mostly scientists or professionals by training and experience in medicine or nutrition. One had no relevant scientific training, but was widely read in the field, in which he had moved for a number of years.

Opinion evidence of the proffered experts was admissible on matters relevant to the first three issues because resolving them called for expertise on matters which were not within my own experience or knowledge or which were, at best, only partly so.

Evaluation of each particular expert's opinion, however, remained a matter for me, especially where opinions were expressed on an ultimate issue which I have to decide. It seemed best that they should all express their opinions leaving me to assess the extent to which their expertise had been achieved from reading source material which I could just as well read and understand for myself. The degree of expertise actually revealed by a witness in his evidence was obviously an important part of the process of attributing weight to his opinion.

All the expert witnesses referred to writings on the subject of diet and disease. The evidential status of books, reports or articles written by experts or groups of experts who do not actually give evidence has always been a grey area, in my experience. Often what is said in a scientific article - medical articles are the ones with which I am most familiar forensically - is adopted by a witness to support his own evidence. There is no problem about his own evidence, but does the article stand as evidence in its own right, for instance on matters upon which the witness has not done research, or where he has not treated or questioned the patients in the sample upon which the article was based?

I adopt what Bingham J. said in H. v. Schering Chemicals Ltd [1983] 1 WLR 143. One of the major issues in that case was whether a drug manufactured and marketed by the Defendants had caused the injuries of which the Plaintiffs complained, and the question arose as to whether large numbers of documents summarising the results of research, and articles and letters published in medical journals concerning the drug, could be adduced in evidence. At page 148 A to G, Bingham J. said:

Page 216

"It is, as I have said, common ground that these articles can be referred to by experts as part of the general corpus of medical knowledge falling within the expertise of an expert in this field. That of course means that an expert who says (and I am looking at it from the plaintiffs' point of view for the purposes of my example) 'I consider that there is a causal connection between the taking of the drug and the resulting deformity', can fortify his opinion by referring to learned articles, publications, and letters reinforcing the view to which he has come. In doing so, he can make reference to papers in which a contrary opinion may be expressed but in which figures are set out which he regards as supporting his contention. In such a situation one asks: Are the figures and statistics set out in such an article strictly proved? and I think the answer is no. I think that they are nonetheless of probative value when referred to and relied on by an expert in the manner in which I have indicated. If an expert refers to the results of research published by a reputable authority in a reputable journal the Court would, I think, ordinarily regard those results as supporting inferences fairly to be drawn from them, unless or until a different approach was shown to be proper. Let me apply that to this case. Mr Beldam submits that there are great dangers in relying on those results contained in this material. For example, he says that certain of them refer to pills having been prescribed but leave it uncertain whether the pills were taken. If the pills were taken the results as published often leave it unclear at what stage of a pregnancy they were taken. Were they, for example, taken at a stage when it was too late for the foetus to be affected by the pills, even if they were capable of having injurious effect in other circumstances? How were the control cases matched? How were the histories taken? How were the cases identified, and so on? All of these are valid points which will fall to be considered and assessed when they are made and when they are put to and discussed with any expert who relies on the articles. It may be that some of the

Page 217

answers will be found in the articles themselves. It may be that other matters will be left in doubt. It may well be that grounds will emerge for viewing the results of the research with caution or scepticism. But in my judgment the proper approach of this court is to admit the articles, in the sense of reading them, and to give the factual assertions in those articles such weight as appears to the court, having heard any cross-examination or other evidence, to be proper."

Bingham J. accordingly ruled that the Plaintiffs were entitled by means of expert evidence to incorporate the contents of the articles in their evidence. It would be given such weight as in the light of any other evidence and any cross-examination appeared proper.

Since 1983 it has become increasing popular for international bodies and Government Departments, including Government Departments in the U.K., to commission reports by panels of experts on topics, including diet, which relate to public health. The panels bring their own expertise to their conclusions, having considered and analysed large amounts of research and large numbers of articles on the topics with which they are concerned. The panels, for example, the Panel on Dietary Reference Values of the Committee of Medical Aspects of Food Policy (COMA) and the Cardiovascular Review Group of COMA then report to the Department of Health which publishes the reports in book form. COMA 41 in 1991 and COMA 46 in 1994 are reports made by the Panel and the Review Group to which I have referred. An equivalent report of a Cancer Review Group is expected soon.

In my judgment I should treat such reports in the same way as the learned articles which Bingham J. was considering, and for the same reasons.

Having said all this however, I am bound to be most influenced by the direct evidence of the witnesses whom I have seen and heard, especially in an area where the literature is so copious and where there is something, indeed an awful lot, for everyone

The evidence relating to this part of the case was heard in two parts: firstly in July, September and October, 1994, as the trial got under way, and secondly in May and June, 1996, to cope with my November,1995, ruling on the meaning of the leaflet and with advances in the science of diet and degenerative disease.

Page 218

So I come to the first issue: is McDonald's food "high" in fat, sugar, animal products and salt (sodium) and "low" in fibre, vitamins and minerals?

The Defendants contended that the measure of what was "high" or "low", should be the various dietary recommendations made by national and international bodies. If a food or meal was above the upper recommended limit for something like fat, it was to be treated as high in fat. If a food or meal was below the lower recommended limit for a particular nutrient, it was to be regarded as low in that nutrient. Even given that the recommendation related to diet, and not to a particular food or meal, it was natural and sensible to relate a food or a meal to the levels to be aimed for in the diet as a whole when judging whether the food or meal was high in one component or low in another.

Recommendations differed but there was broad agreement, particular so far as fat was concerned.

In September,1983, the Health Education Council produced a discussion paper on proposals for nutritional guidelines for health education in Britain, prepared by a working committee for the National Advisory Committee on Nutrition Education (NACNE).

NACNE did not like the phase "balanced diet" which had been associated with problems of vitamin, protein or mineral deficiency which were no longer the major nutritional diseases in Britain.

The paper recommended a "healthy varied diet" and made various long term proposals. Of particular relevance are the proposals that fat intakes should be on average 30% of total energy intake and saturated fatty acid intake should be on average 10% of total energy intake. This could not be achieved rapidly and the aim in the 1980s should be to reduce the average total intake fat intake by 10% from the current 38% of total energy Kcal to 34% and to reduce the average saturated fatty acid intake from the current 18% of total energy to 15% of total. The aims for the 1980s gave the figures for fat and saturated fat in grams before giving them in percentages of energy.

The long term proposals also included reducing average sucrose intake to 29kg per head per year, and causing salt intakes to fall by an average of 3g per head per day.

Page 219

Fibre intake should increase on average to 30g from 20g per day, the increase to come mainly from the increased consumption of whole grain cereals. An increase in vegetables and fruit consumption should also be advocated.

Protein intake should not alter, but a greater proportion of vegetable protein developing from the other recommendations was appropriate.

NACNE made no recommendations to reduce dietary cholesterol, due to a variation in current scientific opinion.

The World Health Organisation's 1990 report on Diet, Nutrition and the Prevention of Chronic Disease set "population nutrient goals."

For total fat the upper limit was 30% of energy and the lower limit 15%.

For saturated fatty acids the upper limit was 10% of energy. The lower limit was zero because saturated fatty acids were not essential nutrients and they did not need to be included in the diet.

For salt (sodium chloride) the upper limit was 6 grams per day. No lower limit was defined.

For free sugar the upper limit was 10% of energy. The lower limit was zero.

For dietary fibre expressed as non-starch polysaccharides (NSP) the upper limit was 24 grams per day and the lower limit was 16g.

For fruit and vegetables the lower limit was 400 grams per day, of which at least 30g should be in the form of pulses, nuts and seeds.

For dietary cholesterol the upper limit was 300mg per day. The lower limit was zero.

In 1991, COMA 41 considered the range of conditions to which dietary fat might be related and recommended that saturated fatty acids should provide an average for the population of 10% of total dietary energy, and that total fatty acid intake should

Page 220

average 30% of total dietary energy including alcohol or 35% of energy derived from food.

The Panel decided that it was not possible to establish an Estimated Average Requirement (EAR) of sodium but it suggested that the Lower Reference Nutrient Intake (LRNI: the amount of the nutrient that was enough for only the few people in a group who had low needs) be set at 575mg per day with a Reference Nutrient Intake (RNI: the amount that was enough, or more than enough, for about 97% of people in a group) of 1600mg per day. The Panel was unable to offer guidance on high consumption but noted that usual high intakes were in excess of 3.22g (3220mg) a day. The Panel saw no physiological advantage in exceeding that intake and considered that it would not be appropriate to increase it further. Later the report said that according to a 1990 survey mean intakes of sodium were 3,376 and 2351mg/d in British men and women respectively.

The Panel proposed that adult diet should contain an average for the population of 18 grams per day (individual range 12 to 24g per day) of non-starch polysaccharide from a variety of foods whose constituents contained it as a naturally integrated component. The Panel saw no advantage in increasing intakes of NSP in excess of 32g per day.

The Panel gave guidance on intakes of some vitamins and some minerals in addition to sodium. They were RNIs.

The Panel's investigations covered research up to and including 1990.

In 1992 the U.K. Government White Paper "The Health of the Nation" set dietary targets to help reduce coronary heart disease (CHD). The targets were to reduce the average percentage of food energy derived by the population from saturated fatty acids by at least 35% by 2005 (from 17% in 1990 to no more than 11%), and to reduce the average percentage of food energy derived by the population from total fat by at least 12% by 2005 (from about 40% in 1990 to no more than 35%).

In 1994, COMA 46, the Report of the Cardiovascular Review Group on Nutritional Aspects of Cardiovascular Disease recommended a reduction in the average intake of common salt (sodium chloride) by the adult population from the current level of about 9g/day (150 mmol/d) to about 6g/day (100 mmol/d) representing about 7g/day (115 mmol/d)

Page 221

for men and 5g/day (85 mmol/d) for women. 1 mmol = 23mg sodium (Na): 1g salt (NaCl) contains 17.1 mmol Na. So, counsel for the Plaintiffs calculated, the recommended level for a man was 2358mg/d of sodium itself.

COMA 46 recommended that the average contribution of fatty acids to dietary energy be reduced to no more than about 10%. It recommended a reduction in the average contribution of total fat to dietary energy in the population to about 35%. It recommended that the average dietary intake of cholesterol should not rise.

A number of the Defendant's witnesses measured McDonald's meal combinations or food items against those recommendations.

Ms Jane Brophy took a degree in chemistry before qualifying as a State Registered Dietitian. Her work involves working with other professionals within the NHS including general practitioners, practice and district nurses and dietitians, giving them information and advice on how to prevent diet-related diseases.

Measuring various McDonald's meal combinations in 1993 against the various recommendations, Ms Brophy said that in her opinion an average McDonald's meal was high in fat (particularly saturated fat found mainly in animal products), sugar, and salt (sodium). It was low in fibre and some vitamins. This could lead to diseases such as heart disease and certain cancers.

Using the rule of thumb that one gram of fat produces 9 Kcalories of energy, Ms Brophy gave analyses of two meal combinations published by the Second Plaintiff in Good Food, Nutrition & McDonald's in October,1995. The analyses were:

	Meal Combination 1
Big Mac, Large French Fries, Apple Pie and Regular Cola: Energy 1283 Kcalories Fat 64.4 grams (579.6 kcal or 45% Fat) Sodium 1488 m/grams (mg)

Meal Combination 2
Hamburger, Regular French Fries and Milk: Energy 707 Kcalories

Page 222

	Fat		35.1 grams (315.9 kcal or 45% Fat)

Those calculations came from figures in the Second Plaintiff's "Good Food, Nutrition and McDonald's" leaflet of October,1985.

Ms Brophy said that the sodium content of Meal Combination 1 was 1,488mg which was close to the maximum of 1,600mg per day recommended by the Government. In my view that involved a misreading of COMA 41 which referred to an RNI (the amount which was enough for 97% of a group) of 1600mg.

Ms Brophy said that similar calculations could be done which showed that a typical McDonald's meal was lacking in fibre, fresh fruit and vegetables and certain vitamins and minerals. She appeared to relate that to the fact that the WHO recommendation of 400g/d of fresh fruit and vegetables was far above the amount found in a McDonald's meal. 400g/d would mean something like five whole pieces like a whole banana, apple, or pepper. McDonald's meals were displacing more positive types of food.

Ms Brophy thought that the COMA figures of 35% of energy from total fat was really an arbitrary one, picked as achievable. The actual target which would benefit people's health would be nearer the 15 to 30% recommended by the WHO.

She thought that if someone ate once a week at McDonald's and high fat meals the rest of the time, one could attribute the unhealthy effect on his diet to the McDonald's meal, because the McDonald's meal was a style of eating.

Dr Neal D. Barnard M.D. came from the U.S.A. to give evidence, primarily on the first and third issues. He confirmed his written witness statements dated 27th July,1993, and 12th April,1994. He spent four days in the witness box in October, 1994. He made a third witness statement dated 1st July,1996, which was admitted under the Civil Evidence Act.

Dr Barnard, is a licensed physician, a graduate of the George Washington University School of Medicine, and a diplomat of the National Board of Medical Examiners. Since 1985, he has been President of the Physicians Committee for Responsible Medicine, a non-profit making organisation of more than 3,000 physicians in the United States. It was formed in 1985 to address the issue of preventing chronic disease with particular reference to diet. Dr Barnard has written books and articles on nutrition and health. He lectures frequently on the role of nutrition in preventive medicine. He is Editor-in-Chief of Good Medicine, a quarterly magazine on medical issues including nutrition. He said that in those

Page 223

capacities he had extensively reviewed the medical literature on nutrition issues. He said that his medical training gave him an advantage in analysing the literature. Although his clinical practice was in psychiatry, that was now limited to one morning a week, and he thought that he had a particular expertise in nutritional factors which affected chronic disease.

The essence of Dr Barnard's evidence on the first issue was that many products sold at McDonald's restaurants were high in fat and cholesterol, and low in fibre and certain vitamins. For example, according to Bowes and Church's Food Values of Portions Commonly Used (1989) by J.A.T. Pennington, 55% of the calories in a Big Mac came from fat, along with 83mg of cholesterol. For a cheeseburger, fat content was 45% with 41mg of cholesterol. French fries were 47%, while a hamburger was 39% fat and held 29mg of cholesterol. Many other McDonald's products were similarly high in fat and cholesterol. McDonald's products were in general particularly high in saturated fat although they were also high in total fat. They tended to be low in fibre and because they were meat based they tended to be low in certain vitamins as well. They presented more fat than an average and unmodified diet.

Dr Barnard said that the average fat intake in the U.S. was approximately 37% of calories. He noted that in the report of Professor D.J. Naismith, a witness to be called by the Plaintiffs, the fat content of a typical McDonald's meal was put at 43%, and that a typical McDonald's meal contained both cholesterol and fat.

However, in Dr Barnard's view the appropriate yardstick by which to judge foods of the type sold at McDonald's and their contribution to health risk was not the average American diet, but the dietary guidelines recommended by health authorities. The U.S. government and most major private health organisations had for many years recommended limiting dietary fat intake to less than 30% of calories, 10% from saturated fat. Dr Barnard thought that those

Page 224

figures were too high. The figure of 30% was an arbitrary round figure which did not require much change on the part of the American public. It had a scientific and political element as a result of lobbying by the food industry and others. International comparisons, such as those from various sections of China, clearly showed advantages in reducing fat intake to levels which were substantially lower than those currently recommended by U.S. health authorities.

A number of McDonald's products were "really quite high" in sodium, by which he meant over a gram of sodium in a single food.

Some of McDonald's products were high in sugar. Their sodas had sugar as virtually their sole nutrient.

They were also low in fibre, some vitamins and minerals.

Professor Michael Crawford is the Director of the Institute of Brain Chemistry and Human Nutrition at Queen Elizabeth Hospital for Children at Hackney. His first degree was in chemistry. He then studied biochemistry before obtaining a doctorate in chemical pathology in 1960. Since then his research programme has focussed on the relationship between nutrition and disease. His particular field is dietary fat and fatty acids and their relationship to human disease. He has published a very large number of papers. He has been a member of a number of consultative bodies and committees investigating questions which are relevant to issues in this part of the case. He is clearly a distinguished man and he was an engaging witness.

Professor Crawford gave evidence on the issues of whether McDonald's products are high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, and whether a diet high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, created the very real risk of heart disease or cancers of the breast and bowel.

So far as the first issue was concerned, Prof. Crawford produced analyses of McDonald's milk shakes and a beefburger in August,1990. They differed from figures produced by independent analysts for McDonald's, but this was probably because Mcdonald's had reduced their fat content since 1990.

Prof. Crawford's 1990 analysis of a chocolate milk shake showed total fat providing 25% of energy. Saturated fat provided about 16% of energy; 17.4% if one included trans isomers (due to biohydrogenation in ruminant animals) which acted in the same way


as saturated fatty acids. McDonald's analysis was 17.3% for fat, and 12.3% for saturated fat. Prof. Crawford's figures were slightly higher for a strawberry milk shake. McDonald's were slightly lower.

McDonald's analysis of a Big Mac showed total fat providing 48.5% of energy, and saturated fat 23.3%. Prof. Crawford calculated that DHA and EPA, the principal n-3 fatty acids which Prof. Crawford described as "anti-thrombotic and anti-cancer proliferation", were present in only trace amounts in the Big Mac.

Prof. Crawford's analysis of a beefburger showed fat providing 56% of energy: saturated fat 26%; 29% including trans isomers. McDonald's analysis showed fat providing 31.7% of energy. The figures compared unfavourably with the various recommendations, and in summary, Prof. Crawford said, there was a high proportion of the types of fats and fatty acids which people were recommending be reduced and low proportions of those which people were recommending be increased.

There was nothing wrong in someone going into McDonald's for a burger, in Prof. Crawford's view, but what concerned him was its overall influence on food choices around the world, exerted by effective advertising.

By the time of his return to the witness box in 1996, Prof. Crawford had done some calculations of the nutritional content of two home-made chicken and beef meals with two Mcdonald's meals.

The McDonald's chicken meal was a McChicken sandwich, medium french fries and medium coke. The home-made meal was pasta with diced chicken, peppers and coriander.

The McDonald's beef meal was a hamburger, medium french fries and medium coke. The home-made meal was a ground beef burger in pitta bread made with lean beef.

Generally speaking the McDonald's meals were higher in energy (kcal), fat and saturated fat, and lower in protein, fibre, vitamins and minerals than the home-made meals.

However, the McDonald's meals did have protein, vitamins and minerals which were important nutrients. Potatoes from which french fries were made

Page 226

were an important source of vitamin C. Fat generally was important as a source of energy, but Prof. Crawford would not describe saturated fat as an important nutrient. He would describe it as an "anti-nutrient". Prof. Crawford said that there was no doubt that McDonald's meals did provide nutrients. His point was that you could obtain them without having the high saturated fat load. In fact, if a 20 year old man's energy requirement for each day was about 2,500 kcal the 10% DRV for saturated fat in COMA 41 would allow 28 grams of saturated fat a day at 9 kcal per gram. The McDonald's chicken meal provided 5.58g and the McDonald's beef meal provided 6.32g.

McDonald's chicken meal provided 3.6 grams of fibre and the beef meal 3.15g.

Prof. Crawford had not included sodium and free sugars because he was not making a point about them.

Dr Timothy Lobstein was called by the Defendants to give evidence on the first and second issues.

Dr Lobstein's degree was in psychology. His doctoral thesis concerned the overlapping area between physiology and psychology. He is editor of The Food Magazine and co-director of The Food Commission, a small limited company which is a consumer orientated watchdog on food issues. He is a freelance journalist and consultant on food policy. When he gave evidence in October, 1994, he had recently joined the fast food panel of the Government's Nutrition Task Force. He described his area of expertise as food policy issues, in particular social science and food issues and their bearing on nutrition.

So far as the first issue was concerned, Dr Lobstein said that he had examined the nutritional make-up of a variety of fast-food meals including all eight meal combinations published by the Second Plaintiff in "Good Food Nutrition & McDonald's" of October,1985, and he found that they were generally unbalanced with regard to their nutrient content. The eight meal combinations published by McDonald's had higher fat levels than recommended for an overall balanced diet while also having low levels of certain essential vitamins and minerals.

If they were relied on for a substantial part of the diet they would not be providing sufficient vitamins and

Page 227

minerals to the amount of fat which they provided. They were excessively fatty and salty.

As an example Dr Lobstein took Meal Combination 1: Big Mac, Large French Fries, Apple Pie and a Regular McDonald's Cola. For a child taking in 2000 Kilo calories a day, a 35% fat allowance (per COMA) would mean 700 kcal or 78 grams of fat, yet Meal Combination 1 provided 64.4 g, or over 80% of the daily quota.

That meal combination sounded to me to be more appropriate to a young man that to a child. With a daily intake of 2500 kcal a young man would have a fat allowance of 875 kcal or 97g, of which Meal Combination 1 would provide about two thirds.

Of course there is more to consider, like not taking one day on its own on the one hand, or 35% being liberal on the other, but the figures demonstrate Dr Lobstein's approach. He did not at first seem to allow for the fact that a whole day's eating was still only a part of one's diet, because he referred to 78 grams of fat being the maximum amount of fat which the child should consume for the day.

He developed the exercise to say that only 8.2% of the day's vitamin A, 34% of its calcium and 32% of its iron came from Meal Combination 1. In order to eat healthily a child or an adult would have to find an extremely low fat meal, very rich in nutrients to end up with an average daily intake that met government guidelines.

However, Dr Lobstein did go on to say that he was not claiming that Meal Combination 1 was eaten every day. If it were, he thought children would be extremely at risk. He said that the point was to show the imbalance. Fat levels were high compared with the low levels of many of the essential nutrients. It was the same for all eight meal combinations.

The position was very similar for salt. As a rule of thumb: 40% of salt was sodium. So out of 6g of salt a day, 2.4g was sodium. The WHO recommended 2,400mg of sodium as "being sufficient for a daily intake." Dr Lobstein appeared to treat this as a maximum, because he said that Meal Combination 1 contained 1488mg of sodium which was approximately two-thirds of the maximum recommended by the WHO. (It is 62%, in fact.)

Page 228

The Defendants' main witness on the first two issues was Prof. Vernon Wheelock. He has a first degree in chemistry and agricultural chemistry. He has a PhD in physiological chemistry. He is Head of the Food Policy Research Unit at the University of Bradford and Special Professor in Food Science at the University of Nottingham.

Prof. Wheelock is also the Managing Director of Vernon Wheelock Associates Ltd, consultants in food policy. Since February,1991, he has been engaged to advise the Second Plaintiffs in the U.K. on nutritional issues and they have made changes to some of their products, in line with his advice. About 20% of the income of Prof. Wheelock's company comes from the Second Plaintiff.

Prof. Wheelock said that he was happy with the recommendations in COMA 41 because the type of exercise done by the Panel which produced the report had been undertaken many times by efficient bodies in different countries and by committees set up by learned societies, and by and large there was a remarkable degree of consensus on the dietary recommendations that were put forward.

A significant part of Prof. Wheelock's evidence was devoted to describing how the second Plaintiff had reduced the total fat, saturated fat and salt content of various food items, yet without losing their distinctive taste, particularly since Prof. Wheelock had begun to advise the company in 1991. McDonald's food still did not come within the recommendations, but there was no reason why every food which everybody ate should come within the recommendations.

Prof. Wheelock said that so far as salt was concerned things had moved on since COMA 41 in 1991. There was a growing awareness of the links between dietary sodium and hypertension, and in his view it would be prudent to reduce the daily average for a man of 3376mg, but not below 1600mg. Needs for salt varied from person to person and climate to climate. I believe that COMA 46 may not have been published when he gave his evidence.

Prof. Wheelock accepted that there was no question that some meal combinations which a person would have in McDonald's might well come above the proportions of fat, saturated fat and salt recommended

Page 229

for the total diet; but there was a range of different compositions available. It did depend on what you chose.

Ms Steel specifically asked Prof. Wheelock: "A typical meal of cheeseburger, fries and milk shake, would you agree that that was going to be high in fat, saturated fat and sodium?" He answered: "I am pretty sure it would be if I was to do the calculations. You know, we would end up with a fat content which is higher than 35, a saturated fat content which is higher than 10; yes, I do not think there is any question of any argument about that." He was clearly relating McDonald's food to the COMA 41 recommendations for diet in judging whether the meal was "high in" fat and saturated fat, although he did add that there was quite a difference between the average diet and the recommendations which we were trying to achieve.

So far as salt was concerned, Prof. Wheelock said that it was virtually impossible to reach the COMA 41 recommendations for sodium with commonly available processed foods, by which he meant foods which people bought to take home. McDonald's food was in line with other processed foods. The concentration of sodium in McDonald's foods was about the same as in a typical British diet. I think that in saying that, when referring to the COMA 41 recommendations, Prof. Wheelock must have had the RNI of 1600mg per day in mind, because the typical British diet, and by Prof. Wheelock's account, therefore, McDonald's food, must have been about the average daily consumption given in the report.

So far as fibre content was concerned, Prof. Wheelock said that the concentrations in McDonald's foods were about the same as a typical British diet.

The ten McDonald's meals which Prof. Wheelock chose [five children's meals and five adult meals] and then analysed in order to support his view that one McDonald's meal per week or even per day could be assimilated into the diet of a child and adult without adverse effect on the diet so far as intake of fat, saturated fat, extrinsic sugar, fibre and sodium were concerned, showed that the percentage of calories provided by fat varied between 31.3% and 51.4%, most of the meals being in the middle 40s. The percentage of calories provided by saturated fat varied between 15.8% and 25.4%, seven of the ten meals being 20% or over.

An adult meal of cheeseburger, regular fries, diet coke and apple pie provided 987mg of sodium which was just over 60% of

Page 230

COMA 41's RNI of 1600mg and just over a quarter of the national daily average. Prof. Wheelock expressed the opinion that it was not high in sodium. However, the meal chosen was the lowest for sodium of the meals in the Table and Prof. Wheelock expressed this view before he conceded that it was virtually impossible to reach COMA 41 recommendations for sodium with commonly available processed foods.

The same meal provided 12.37g of fibre so Prof. Wheelock said that it just crept into the bottom of COMA 41's daily bracket of 12 to 24 grams. In fact this bracket was for NSP, but of course it was only one of the meals in the day. Prof. Wheelock said that the meal was not low in fibre.

The above calculations appeared to be done at about the beginning of 1994, and Prof. Wheelock accepted that since 1989/1990 the fat content in McDonald's cheese had been reduced. Salads featured in some of the meals, but they were not available in all McDonald's restaurants. Orange juice was introduced in about 1989. Frying oil was changed from beef tallow to vegetable oil in 1989. There had been a switch from lard to vegetable oil in the buns.

Prof. Wheelock said that McDonald's meals were not low in vitamins and minerals.

Dr Sidney Annott, whose prime task was to give evidence on the position of diet in the causation of cancer of the breast and bowel, said that he thought the COMA 41 recommendations for daily reference values for total fat and saturated fat were very fair.

On the basis of this evidence it seems clear that a number of McDonald's products, including a sufficient number of typical meal combinations, presumably put forward in the Second Plaintiff's own nutritional leaflets because they are typical, have higher concentrations of fat, saturated fat and salt than the levels generally recommended for a healthy diet.

Taking fat and saturated fat as an example, the question then arises as to whether those recommendations, made as they were for diets, and not for individual meals, are a valid test of whether a food item or a meal is "high in" one component or "low in" another.

In my view they are not only a valid test, but the

Page 231

appropriate one, in this part of the case, where the concern is for health, the whole scenario is that recommended levels will improve public health, and levels in individual food products or meals can only sensibly be measured against levels to be aimed for in the diet as a whole.

An alternative approach would be to take the present average levels in diet in the country where publication is complained of, but in my view that would be absurd in a situation where the national diet is not as healthy as it might be and the whole point of looking at dietary fat levels is to get them down by, for instance, eating less fatty food. The national diet is high in fat and saturated fat so far as health issues are concerned.

The approach urged by counsel for the Plaintiffs was to take the COMA daily recommendations and then see how a McDonald's meal fitted into them. If fat intake did not exceed the recommended allowance by weight for the day, the meal could not be said to be "high" in fat. But this did not help me so far as fat was concerned. Food items which were high in fat, salt or sugar, by any other test one could imagine, for example a boiled sweet high in sugar, might come out as "low" by this test. All would depend on the fat content of what one ate for the rest of the day which must logically be irrelevant to the fat level in the McDonald's meal.

Using the recommendations for diet as a test of whether a food item or meal is "high in" fat or saturated fat, does not mean that one has lost sight of the fact that the recommendations relate to diet. That is allowed for by the need to go on to ask the further question (part of the second issue): of whether eating McDonald's food which is high in fat and saturated fat might well make your diet high in fat and saturated fat?

If I had any doubt about the appropriateness of this test it disappeared when Prof. Wheelock quite clearly accepted the COMA 41 recommended levels for fat and saturated fat in the diet, as the test of whether a meal was high in fat and saturated fat. That was not a "give-away" on his part, in my view. It was no more than good sense.

I also find it difficult to see why the Plaintiffs should have taken the trouble to try to reduce the fat content and saturated fat content of some menu items, unless they thought them to be "high" in fat.

Page 232

In my view the arguments which apply to fat also apply in judging whether McDonald's food or a McDonald's meal is high in animal products or salt (sodium) or sugar, or low in fibre, vitamins and minerals. Where there are clear recommendations for dietary levels in a nutrient or a food component it is appropriate to use those recommendations as the test of what is "high" or "low" in the context of this case.

In my judgment if follows from what I have so far said that McDonald's food is "high" in fat which includes a "high" component of saturated fat.

In my view McDonald's food is also "high" in salt (sodium) in that a typical meal appears to provide rather more than its share of the recommended daily quantity of salt (sodium) although the position is less clear than with fat and saturated fat where concentrations in the meal can be judged by percentage of calories provided, and the recommendations are more precise. Prof. Wheelock's evidence placing McDonald's food on a par with so much other food in a typical diet, means that, on balance, it is high in salt, judging by the recommendations which clearly involve a reduction in present levels nationally.

Sugar is more difficult, and I have not been persuaded to the necessary standard that "McDonald's food is high in sugar." Some food items such a sweet drinks and, no doubt, pies must be high in sugar; and there was evidence that fries were sprayed with a sugar solution for consistent colour. But the predominant part of the fry is still potato. Prof. Wheelock told me that sugar free Coca Cola was popular at McDonald's. The main meal items are savoury rather than sweet. Nothing was really made of the sugar content of McDonald's Meal Combinations. The Defendants spent some time discussing the high sugar content of McDonald's birthday cake without taking any account of how often a customer is likely to eat it.

Fibre is most difficult to judge. COMA 41 recommendations related to non-starch polysacchorides, and not fibre as such. The total weight of fibre in a typical U.K. McDonald's 1989 meal of cheeseburger, regular french fries and apple pie was 4.9g which is about 27% of the recommended weight of NSP. By 1993 it was 16.5g or 92% of the recommended weight of NSP. But no one really educated me on the relationship of NSP to dietary fibre. COMA 41 says that "NSP are the major factor of "dietary fibre" whatever definition [of dietary fibre] is used", so it seems reasonable to infer that 4.9g or 16.5g of fibre

Page 233

contains less than 4.9g or 16.5g of NSP. On this basis I would judge the 1989 meal to be low in fibre, but not necessarily the 1993 meal. There was less fibre in the McDonald's chicken and beef meals analysed by Prof. Crawford's team, but that did not answer the question of whether they were low in fibre in themselves. I do not consider that it was right for Ms Brophy to judge McDonald's food low in fibre because it was low in fruit and vegetables. Potatoes appeared to be excluded from "vegetables", but they contain fibre: so do buns.

My conclusion is that on mere balance of probabilities McDonald's food was low in fibre at the relevant time of publication (September,1987, to September,1990,) because a typical McDonald's meal provided only 27% of the recommended daily levels of NSP in fibre even; but it has not been proved that McDonald's food has remained low in fibre since.

It has not been proved that "McDonald's food is low in vitamins and minerals." Some foods are clearly lacking in certain vitamins and minerals; but the requirement of certain vitamins and minerals for a healthy diet is very small. They are clearly picked up here and there in a variety of items of food which go together to form one's diet. McDonald's products provide various amounts of some of the necessary vitamins and minerals. Unlike the case of salt (sodium) which I assume to be present in all meals to a greater or lesser extent, I do not consider it accurate to say that a meal or a particular food is "low in vitamins and minerals" just because it does not have its fair share of the full diet's requirement of all vitamins and minerals which are essential to a healthy diet. The evidence on this part of the case was very sparse. The final submissions of counsel for the Plaintiffs included some detailed calculation based on the limited information on the vitamin and mineral content of some McDonald's products. They were designed to show that McDonald's products provided sufficient amounts of some significant vitamins and minerals, but they were not put to witnesses and counsel fairly said that some of them should be treated with caution. The burden of proving that McDonald's food is low in vitamins and minerals lay on the Defendants and, in my view, they did not discharge it.

I have no doubt that McDonald's food is fairly to be described as "high in animal products" since so many of the most

Page 234

often sold products include beef, or chicken, or sausage, or cheese, as well as milk in shakes. I do not need government recommendations to tell me this.

In summary I find that at the material time of publication of the leaflet between September,1987, and September 1990, McDonald's food was high in fat (including saturated fat) and salt (sodium) and animal products and that it has continued to be so. I find that it was low in fibre at the material time of publication of the leaflet, but that it has not been proved to be so now . It has not been shown that McDonald's food generally is high in sugar, although some individual items are. It has not been shown that McDonald's food is low in vitamins or minerals.

Although I will largely leave sugar behind at this point, I should say that the evidence of any harmful effects of dietary sugar was weak and really limited to the risk of dental caries and, indirectly, to a possible increase in body weight if too much sugar contributed to excessive energy intake.

Turning to the second essential issue, the findings which I have just made must mean that it has not been proved that eating McDonald's food more than occasionally might well make your diet high in sugar or low in vitamins or minerals. The issue remains as to whether it is right that eating McDonald's food more than occasionally might well make your diet high in fat, animal products and salt and, when the leaflet was published by the Defendants, low in fibre.

The essence of the Defendants' case was that a significant number of people ate McDonald's food which was high in fat, animal products and salt, and low in fibre, often enough to make their diets high in fat, animal products and salt, and low in fibre.

The Plaintiffs took issue with this.

They said that a typical McDonald's customer would balance McDonald's food with a variety of other foods. Indeed the Plaintiffs asked him to do so. If he did not do so but ate too much other food which was high in fat, animal products and salt and low in fibre, then his diet could not be laid at the door of McDonald's which made a limited contribution to the diet.

They contended that even if some people did eat McDonald's food often

Page 235

enough to make their diets high in fat, animal products and salt (sodium) and low in fibre, however balanced the rest of their meals were, that did not justify the thrust of the leaflet to its ordinary reader who would not be a very heavy McDonald's eater, to the effect that eating McDonald's food more than occasionally might well make his diet high in fat, animal products and salt (sodium) and low in fibre. If the ordinary reader was an ordinary McDonald's customer he would eat there only now and again and what he ate at McDonald's would not affect his diet significantly.

There was a lot of evidence on the frequency with which people ate McDonald's food.

In my view the relevance of this evidence of "eating frequency" was that if a large proportion of people ate McDonald's food frequently enough to make their diets high in fat, animal products and salt (sodium) and low in fibre, so that it was the norm to do so, then the allegation that eating McDonald's food might well make your diet high in fat, animal products and salt (sodium) though not in my view sugar, and low in fibre, though not in my view vitamins and minerals, would be substantially justified; it would be shown to be true in substance and in fact.

On the other hand if only a small proportion of people could be shown to eat McDonald's food frequently enough to make their diets high in fat, animal products and salt (sodium) and low in fibre, so that result would be exceptional, then the allegation would not in my view be justified, because the leaflet does not say if you eat McDonald's food very often it may very well make your diet high in fat, animal products, and salt, or low in fibre.

So it was necessary to see what numbers of people, if any, ate frequently enough at McDonald's to make their diets high in fat, animal products and salt, and low in fibre. This involved looking at frequency of eating at McDonald's and at the effect of any given frequency on diet.

The main strands of evidence on the second issue were as follows.

Dr Lobstein said that there was a lack of balance in the diet of the general population of this country. He said that the imbalance in the diet of the general population, which bodies like Department of Health and

Page 236

the World Health Organisation were currently concerned with, was a diet with excess fats and sugars, insufficient dietary fibre and insufficient fruit and vegetables and nutrients which were delivered from fruit and vegetables. The McDonald's meal combinations which he examined had much of this imbalance. If such meal combinations were being eaten frequently they would need to be balanced with low-fat, high-nutrient meals to compensate. In his experience this was unlikely to be the case. He thought that for an increasing number of people McDonald's meals represented an increasing part of the diet, and that for some parts of the population it already represented a large part of the diet. The continued promotion of McDonald's meals contributed to the maintenance of unbalanced diets of the kind which the Department of Health and the World Health Organisation warned against.

Dr Lobstein referred to evidence that meals such as those promoted by McDonald's were being eaten frequently.

Although frequency of eating various types of meal varied between groups of the population, government figures showed what Dr Lobstein described as a "tragic" tendency for younger eaters to be eating large quantities of fast food such as burgers and french fries. The reference for this was Diets of British Schoolchildren, HMSO, 1989, a report of the sub committee on National Surveillance of the Committee on Medical Aspects of Food Policy (COMA 36).

In fact the parts of the report to which Dr Lobstein referred said that all the children in the sample consumed large quantities of bread, cakes, biscuits, puddings, milk, meat products, crisps, potatoes and particularly large quantities of chips. This came as no surprise to me. But the reference to fast food turned out to be a reference to 68 out of 513 older boys and 54 out of 461 older girls who took term time lunches out of school at cafes, take-aways and fast food outlets as opposed to eating school lunches or bringing a packed lunch.

Approximately a quarter to a third of children had fat intakes contributing more than 40% of their energy intakes.

Dr Lobstein thought that the proportion of children eating school lunches had fallen since 1989.

Page 237

On reflection he thought that his comment about the tragic tendency for younger eaters to be eating large quantities of fast food such as burgers was based on The Food Commission's own survey in Peckham.

A market research survey showed fast food to be eaten most often by those aged 15-24, and those young people ate fast food twice as often as people aged 45-64.

The authority for this was a British Market Research Bureau/Mintel survey in 1985, but Dr Lobstein no longer had a copy, so I could not see what the actual frequency was. Dr Lobstein did produce a British Nutrition Foundation report "Eating in the Early 80s", published in 1985, which showed 34% of adults eating take-away foods at least once a week and 45% "less often". The highest group eating take-aways at least once a week was 63% of single 16-24s. Over 35s were around 25%.

Dr Lobstein thought it important that the BNF report showed that a percentage of children bought hamburgers with their pocket money, but only from 1 to 5% of the sample said that they did so.

A survey by the trade association HOTAG (The Hot Takeaway Group) (Consumer Catering Report, 1986) found that 43% of those aged 15-24, 54% of unemployed people and 36% of students ate fast food more than twice a week. That report could no longer be traced.

In 1987 a survey by The Food Commission called Grazing in Peckham, a survey of 354 fast food eaters in the London Borough of Southwark, found that 87% of respondents had eaten fast food at least once during the previous week, and 30.5% said that they ate fast food on average once every day. People were questioned as they came out of four fast food stores (a McDonald's, a Wimpy, a Kentucky Fried Chicken and a fish and chip shop) at lunch time and between 3.30 and 5 pm. It was a low income area. Dr Lobstein agreed that the sample might include a number of people who had fallen into the habit of eating fast food at lunchtime, but the point of the survey had been to see if there were "people who were using fast foods at quite high levels of intake." 32% of those coming out of McDonald's said that they ate fast food somewhere once a day, and a further 49% said that they did so more than once but less than five times a week. The average of those coming out of McDonald's

Page 238

ate fast food 4.21 times a week of which about 2.99 were meals and about 1.23 were snacks. The same point about regular lunches might apply and a school run dropped off there in the afternoon.

30% of the sample thought that their general diet was poor, and nearly 42% thought that the food from fast food outlets was poor, yet they were still eating in that way.

Burgers were the most common items eaten by those who ate fast food every day.

Dr Lobstein also expressed views on the relationship of diets high in saturated fat and sodium to heart disease, but in my view he was less qualified than the scientific witnesses on this issue.

Ms Fiona Winter, then Carruthers, the dietician who carried out the survey "Grazing in Peckham" in 1987, carried out a further survey in 1988 called, "Who is under the Golden Arches?" That survey was carried out in the centre of Leeds. Ms Carruthers was a Dietetic Student in her final year at Leeds Polytechnic at the time.

In the discussion at the end of her research report Ms Carruthers very fairly pointed out that the samples of the population used in the study were not wholly representative of fast food eaters. Interviews only took place over the lunchtime period, and the sample profile might have been very different if interviewing was carried out in the evenings. The location of the outlets was also a source of bias. The centre of town was chosen because a range of outlets was found in a small area, offering the greatest choice to the customer. However, the customers using those outlets in town at lunchtime might be very different to those using outlets in a residential area in the evening.

Ms Carruthers found that a large proportion of the consumers interviewed in the Leeds study were eating fast food at least weekly, with a quarter of respondents using the outlets daily. This showed that fast food was contributing significantly to the diets of some individuals, in her view.

She said that recommendations for nutritional intakes were made for the whole day, so each individual meal need not necessarily be balanced.

Page 239

Most of the Plaintiffs' evidence on eating frequently was given by Mr Alastair Fairgrieve, the Second Plaintiff's Marketing Services Manager since 1991. For seven years before that he had worked for the Second Plaintiff in store management and marketing.

Mr Fairgrieve prepared a number of tables from U.K. marketing research surveys.

The first, called "Fast Track" was by Harris Research, based on interviews of about 5,000 people a year. It looked primarily at who McDonald's users were and what they thought of it and its competitors and how opinions shifted.

It was cross-checked against the second by Taylor Nelson Market Research, which was based on telephone interviews of about 60,000 people a year for a syndicate of which McDonald's was a major member.

Both sets of figures tended to accord with the information which the Second Plaintiff got through its tills.

None of the figures included customers up to the age of fifteen.

Mr Fairgrieve produced various tables or abstractions from the surveys. His table produced from the results of both surveys showed that zero percent of McDonald's customers visited McDonald's nearly every day, although he knew that some people ate at McDonald's every day. 4% visited several times a week. 11% visited once a week. This meant that 15% were what were described as "heavy eaters", eating McDonald's food once or more a week. Another 37% ate one a month or more. A real margin of error had to be allowed for.

Later, Mr Fairgrieve produced another table which showed zero percent eating at McDonald's daily, about 0.18% or about 40,000 customers eating nearly every day, about 1.77% or 385,000 eating several times a week, an additional 8.76% or about 1.9 million eating once a week, and an additional 7 million or about 32.32% eating once a month or more.

As part of his final submissions, counsel for the Plaintiffs relied upon figures from the large survey which showed that about 22.5 million people (about 40% of the U.K. population) ate at McDonald's sometime during the year. No discernible percentage of

Page 240

all customers ate there every day. About 55,000 or 0.24% of all customers ate at McDonald's nearly every day. About 477,000 or 2.1% of all customers ate at McDonald's several times a week. About 2.14 million customers or 9.5% of all customers ate at McDonald's once a week. So something like 12% of McDonald's customers ate there once a week or more.

Those figures were in the same broad brackets as Mr Fairgrieve's.

The survey figures which I have given did not include McDonald's employees. A survey in 1992 of a sample of the Second Plaintiff's restaurant staff showed that 6% ate a McDonald's meal more than six times a week; 61% four to six times a week, and 31% zero to three times a week.

Mr Fairgrieve said that a typical McDonald's customer visited an informal eating place about sixty times a year and about one third of his visits were to McDonald's. In my view this must have been a stab at the canvas with a very broad brush in the light of the figures which I have given.

The surest ground was that the heaviest usage group was 16 to 24 followed by 25 to 34. Whichever way one looked at the various use by age group figures, frequency of use fell off quickly once one was out of the 16 to 24 age bracket, and more quickly still once one was out of the 25 to 34 age bracket.

By far the most common reason given for visiting McDonald's less often among those who did visit less often was "change of circumstances" which included changes in age, family, work and financial situation.

About half the visitors took children to McDonald's with them.

Among the attributes of McDonald's mentioned by McDonald's users, "Place Kids Enjoy" came consistently top and "Nutritious Food" was consistently one of the lowest mentions.

Economic and social classes were evenly balanced.

The number of McDonald's restaurants in the U.S. is considerably greater and

Page 241

considerably greater in relation to the population. Calculations done by the Plaintiffs' counsel on the basis of a U.S. survey showed children up to thirteen visiting McDonald's about once every three weeks on average for the population. Customers aged fourteen to forty-four visited about once every eleven days to a fortnight, and older people visited more rarely. The U.S. heavy users tended to eat at other quick service restaurants also, which has some consistency with Ms Winter's surveys.

I was shown a report prepared in December,1991, for the First Plaintiff. It dealt with use of McDonald's restaurants in the U.S.

Its drafting is confusing in parts. For instance it says: "Overall, McDonald's customers are likely to be:...Heavy users (72% visit McDonald's once a week or more)". Later it says: "Nearly three-fourths of McDonald's visits are made by heavy users". This clearly refers to the proportion of visits. This is immediately followed by the statement: "Of these heavy users, about one third are super heavy users". This refers to the proportion of users. It also states: "Nearly 3 out of 4 visits to McDonald's are made by heavy users", and I think that the frequency visit figures which are given must be percentages of visits by categories of users (super heavy, heavy, medium and light) and not percentages of users who fall into those categories.

On this basis the report says that 23% of visits in 1991 were made by super heavy uses (visiting 4 or more times a week). 49% of visits were made by heavy users (visiting once to three times a week), thus giving the "nearly-three fourths" (72%) of visits. 23% of visits were made by medium users (once to 3 times a month) and 5% by light users (less than once a month). The figures were broadly the same for the years since 1986.

My interpretation of the report is confirmed by a page which states that 22% of all customers visit at least once a week. The figures were 22% for 1986, 25% for 1987 and 27% for 1988. The 22% for 1991 was divided equally (11% and 11% super heavy and heavy users; but I wonder if this is accurate because in 1986, 1987 and 1988 only 4% were super heavy users and the much greater balance (18%, 21% and 23%) were heavy users. However this may be, it appears from the report that at the time of relevant publication of the leaflet complained of about three-quarters of

Page 242

McDonald's U.S. customers ate its food less than once a week, and that the majority of that three-quarters ate its food less than once a month.

The McDonald's U.K. and U.S. surveys were done to help guide their marketing rather than to provide reliable figures on eating frequency, and I have misgivings about the accuracy of the survey figures when turned into hard numbers or percentages of customers eating McDonald's food at certain frequencies. But the Plaintiffs introduced the surveys or abstracts from them to help me get a picture of how often people used McDonald's, so they can at very least be used by the Defendants if they help them. Ms Winter's surveys had biases of their own.

The overall picture which I took from the plethora of figures put before me was that in the U.K. where publication of the leaflet is complained of, there was and is a very small proportion of all McDonald's customers eating their food nearly every day; a still very small proportion eating their food more than once a week; about 10% of all McDonald's customers eating their food about once a week and the remainder, probably about 85%, eating their food less often than once a week.

There are about 30,000 hourly paid McDonald's restaurant staff in this country so their regular eating at McDonald's does not change the picture if there are over 20 million people using McDonald's.

The important gloss on the picture is that there was really no evidence that those who eat McDonald's food once a week or more often do so or have done so for a significant proportion of their lives, or will continue to do so.

In the U.S. the proportion of frequent users appears to be greater and heavy use continues into the 35 to 44 age bracket, no doubt because McDonald's has been established there for longer than it has been here, but also no doubt because of slightly different lifestyles. Even then the great majority of McDonald's U.S. customers eat its food less than once a week. Again, however, there was really no evidence as to how long the habitual McDonald's customers have been so.

On the general issue of the effects of McDonald's meals on diet, Prof. Wheelock stressed time and again that although recommendations for total fat, saturated fat, salt, vitamin and mineral intakes, where made, were made on a daily basis, it was their effect on diet which mattered so far as health was concerned. If your intake was

Page 243

in excess of the maximum recommended, or short of the minimum recommended, on one day, you could balance that by eating more or less over the next day or two. In fact Prof. Wheelock said that when looking at diet one was looking at intakes over a week or a month.

Prof. Wheelock said that there was nothing wrong with the occasional high fat meal. He provided Tables of meals and analyses with a view to demonstrating how McDonald's food could be eaten quite frequently without adversely affecting the diet.

Table 1 took five different but typical McDonald's meals and placed them one by one, once a week, in the diet of a child requiring 2,000 kcal of energy a day or 14,000 per week. The weekly totals came out at just under 14,000 kcal. Tables for each of the days when a hypothetical McDonald's meal was taken came out at around 2,000 kcal.

Table 2 took five different but typical McDonald's meals and placed them one by one, once a week, in the diet of an adult requiring 2,500 kcal of energy a day or 17,500 per week. The weekly totals came out at between 17,200 and 17,300 kcal. Tables for each of the days when a hypothetical McDonald's meal was taken came out at around 2,500 kcal.

So all those meals could be fitted once a week into the week's diet or once a day into a day's diet without any or any significant excess of calories. Moreover if the adult was a young, active man with a requirement of about 3,000 calories a day, the typical McDonald's meal provided roughly one third of the COMA 41 recommended fat limit and half the recommended saturated fat limit for one day. So, Prof. Wheelock said, it would not be a problem for the young man to accommodate a McDonald's meal into every day, provided that he ate sensible things at other times of the day.

The same exercise, however, showed that the child would get almost half its day's allowance of total fat and just over two-thirds of its day's allowance of saturated fat from its McDonald's meal.

I have already referred to the calculation which showed the adult taking in 987 of sodium from a typical cheeseburger meal. Prof. Wheelock's conclusion was that he could take in a lot more

Page 244

sodium before reaching the COMA 41 RNI of 1600mg for the day. His intake from the McDonald's meal was just over a quarter of the national daily average.

I have also referred to the similar calculation which showed the adult taking fibre at the bottom of the COMA 41 RNI for NSP for a day, from a typical cheeseburger meal.

Prof. Wheelock was cross-examined at some length on his Tables. I have already mentioned changes made to the constituents of some of the meals between the time of relevant publication of the leaflet and his analyses. Prof. Wheelock did not think they made any significant difference to the contribution of one McDonald's meal to the week's diet.

If Prof. Wheelock had chosen a woman's daily requirement of about 2000 kcal per day for the adult Table the picture of total weekly intake would be different.

Prof. Wheelock produced his schedules of the other meals in each week and in each day of each week, which were taken with the chosen McDonald's meals, to show that total energy intake could be less than 14,000 or 17,500 kcal and around the recommended daily intake, and that the weight of fat and saturated fat and salt and fibre could be within COMA 41 recommended limits. The other meals were challenged as being carefully chosen, with some meals of poached or grilled fish, lots of vegetables, semi-skimmed milk and few fried potatoes, for instance. In my view there was some merit in this criticism, but the other meals were mostly the sort of meals which one might well find in a U.K. home where some real attention was paid to producing a varied diet without much fried food.

Prof. Wheelock agreed with Miss Steel's suggestion that as a nutritionist he would advise anyone who ate a McDonald's meal every day that unless they were very careful balancing the rest of their diet they were going to risk their health in the long run. The question was posed and the answer was given in the context of whether a diet high in fat (including saturated fat) and sodium, and low in fibre, was linked with cancers of the breast and bowel and heart disease, but I took Prof. Wheelock's answer to be acceptance that it would be difficult to avoid making one's diet high in fat and sodium, if not low in fibre, if one ate a McDonald's meal every day.

Page 245

Having considered all the evidence on the point and all the calculations put forward by the Defendants, I find that it is possible to eat a typical McDonald's meal several times a week, without making the diet high in fat (including saturated fat) and salt (sodium) or low in fibre, by the standard of the recommendations which I have applied, provided that one takes a lot of care with the rest of one's meals. If one eats several McDonald's meals a week without taking a lot of care with the rest, one's diet is likely to be high in fat (including saturated fat) and salt (sodium) and it may be low in fibre, and since several meals a week must in my judgment form a significant part of one's diet, the frequent McDonald's meals would bear a real part of the responsibility for that result, for so long, but for so long only, as they continued.

On the other hand, if one ate a typical McDonald's meal, high in fat (including saturated fat) and salt (sodium) to the extent which I have found, and low in fibre to some extent in 1989 but not now, just once or twice a week, one would not in my view have to take much trouble with the remainder of one's meals in order to keep to the COMA 41 recommendations. If one ate quite a lot of the same kind of food during the rest of the week, one would be in trouble; but in that case I do not consider that it would be right to hold the one or two McDonald's meals responsible. Whether one's diet was high in fat (including saturated fat) and salt (sodium) and low in fibre would be far too dependant on what one ate for the rest of the week to blame the one or two McDonald's meals for the result.

The same obviously applies even more strongly to the effect or lack of effect, of eating McDonald's meals less than once a week.

In reaching this view I have taken account of the evidence to which I will come that eating a certain kind of food can be habituating, and thereby affect overall diet. Whatever merit there may be in this if a person falls into the habit of eating fast food virtually every day, I cannot accept that eating a McDonald's meal once or twice a week, or less, can habituate the ordinary person to eating similar food frequently enough to affect his diet adversely.

Only a small proportion of those who visit McDonald's in the U.K., and therefore an even smaller proportion of the general public, eats McDonald's food more than once a week. The proportion of those who visit McDonald's in the U.S. more than

Page 246

once or twice a week appears to be larger, but still small, if one accepts that a fair proportion of the quarter who visit once or more a week must visit once or twice only.

In my view it must follow that it is not true to say that eating McDonald's food, albeit more than just occasionally, might well make your diet high in fat, animal products and salt (sodium), let alone sugar, or low in fibre, let alone vitamins and minerals. Such a statement is not justified. It is not true in substance and in fact because it is only true (so far as fat, animal products, salt and fibre are concerned) in relation to a small proportion of people, who eat McDonald's food several times a week. The leaflet does not say that if you eat McDonald's food several times a week it might well make your diet high in fat, animal products and salt (sodium), and low in fibre. It leads the reader to believe that this is to be expected from anything more than the occasional McDonald's meal.

It follows that it cannot be right to say that eating McDonald's food more than just occasionally will bring the very real risk that you will suffer cancer of the breast or bowel or heart disease as a result of making your diet high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, even if such a diet carries such a risk.

However, in case there is any appeal against the findings which I have just made I will go on to consider whether it has been proved that a diet high in fat (including saturated fat), animal products and salt (sodium), and low in fibre, leads to a very real risk of heart disease or cancer of the breast or cancer of the bowel.

Before referring to the main strands of evidence on this question, it is necessary to describe the nature of the scientific studies which bear on the question of diet and its part, if any, in the causation of heart disease or cancer of the breast or bowel, and some of the difficulties which stand in the way of drawing confident conclusions from them.

There have been numerous epidemiological studies which include population studies, group studies, case control studies and prospective or cohort studies. There have been experimental studies on animals and in vitro, and to a lesser extent on humans. There have been clinical studies, largely

Page 247

observations of morbidity in humans.

Each of the various kinds of study has its own advantages and disadvantages in the context of the issue here.

Population studies, where the incidence of disease and the way of life, including diet, in one country are compared with those of another, are said to have the advantage of clear distinctions for instance in the proportion of energy in the diet provided by fat, but also in other matters such as the amount of total energy intake or the intake of fresh fruit and vegetables, or exercise, or body mass, which may or may not be related to the proportion of energy provided by fat. On the other hand population studies can be criticised as crude, based on food disappearance or food balance figures which may or may not accurately reflect consumption of a particular food; or based on diagnoses or records of disease, which may not be reliable.

Case control studies where a group of people who suffer from a disease are compared with a group of people in similar conditions, who do not, may involve groups where differences in diet are narrow, and may involve poor recollection of diet, especially diet some years before.

Prospective or cohort studies where as large a number of people as possible are followed for a number of years for a number of factors, including diet, to see who suffers a particular disease, lead to more accurate records of the various factors; but again differences in diet may be small and time spans may be short or they may not cover a potentially relevant time of life, for instance pubity.

Animal physiology is different to that of man. The animals may be specially bred for susceptibility to a suspected carcinogen, and doses of the carcinogen may be high, the sooner to achieve results.

There are ethical and practical difficulties in the way of what might be the most productive human experiments. One can hardly invite one part of a group to take a diet which is suspected of increasing the risk of disease, and it is difficult to make people change their diets, even for what might be the better. Time scales might have to be long.

Quite apart from the arguments about the specific advantages or

Page 248

disadvantages of one or other type of study, there are problems of assessment which apply to them all.

Firstly, the greatest difficulty seems to me to arise from the fact that many so-called degenerative diseases are multifactorial and, as their description as "degenerative" indicates, they develop over a period of time which is often considerable. So a number of factors may have to be in play for a long time before any risk of disease is noticeably increased. This is bound to make identification of the role of a certain kind of diet difficult. Identification of the role of a particular component of a certain kind of diet is even more difficult. If a particular component is causative of disease, is it directly or indirectly so? At what stage or stages of life does the diet have to be of a particular kind or contain a particular component, and for how long?

In most studies an important part of the exercise is to allow for non-dietary factors which may "confound" the conclusion. But this may be difficult to do and there may be room for argument as to whether the "confounding factor" is related to the suspect and, if so, to what extent. For instance, let us assume that a high fat diet is suspected of being a risk factor for cancer of the breast: so are early age menarche, late age at birth of first child, low parity (number of children) and late menopause, all features of western society where the incidence of a "western disease" like cancer of the breast is high. But early age of menarche may be related to nutrition in the sense that the better fed a population is the earlier the average age of menarche becomes. Is this due to high fat diets or just "over-nutrition" (high intake of energy/calories generally)? Late age at birth of first child and low parity are social rather than dietary factors. How does one balance their influence? Which category does age at menopause fit into?

Secondly, it is not always easy for a lay outsider to have faith in the experts who give evidence or in the experts upon whose work they rely, when the studies are so copious and varied that there is something for everyone, and when faults or perceived faults in the methodology as well as the general nature of studies may provide ample opportunity to "explain away" their apparent conclusions if they are unattractive to the reader. It is clear

Page 249

to me as a comparative newcomer to the area, that the topic of diet and disease stirs strong feelings. There can be a perfectly human tendency enthusiastically to adopt a point of view and to hold it. Once a stand has been taken, support can always be found and objections explained away.

On the other hand the comparative imprecision of some of the science enables a sceptic to say all too easily that an hypothesis is interesting, but not proved. An attractive route to the balance between the points of view of enthusiast and sceptic would be to follow any consensus between the national and international bodies which have given evidence on diet. But much of the guidance is given on the basis that it is possible that a high intake of a certain kind of food may be a causal factor or risk factor in a particular disease, whereas a lesser intake will not cause harm. It is then obvious good sense to recommend its diminution, so the fact of a recommendation may mean no more than "it will do no harm and it may do some good."

Thirdly, over all the population or group studies, so far as the issue which I have to decide is concerned (very real risk or not, of actual disease), lies the shadow that even if a particular diet or component of diet can be shown to increase the incidence of a particular disease in a population or group which broadly shares that diet, it may still be difficult to show that people who follow that diet take a very real risk of suffering that disease as a result, as opposed to having a statistical share in a countrywide risk which may be important from the point of view of public health, but minimal or unquantifiable from an individual point of view.

Finally, there are clearly important limits to the science of the causes of cancer of the breast and cancer of the bowel in particular. Family history is the greatest epidemiological risk factor for both. But we appear to be a long way from understanding the genetics and the part they play.

It may be easy to propound a plausible biochemical mechanism between suspect and disease, but difficult to prove or disprove it to any standard remotely approaching the preponderance of probability in the present state of knowledge.

The evidence of Dr Barnard was the high water mark of the Defendant's case on the relationship of diet to heart disease and cancers of the breast and bowel.

Dr Barnard said that as a result of their high content of fat and cholesterol

Page 250

McDonald's products contributed to heart disease, certain forms of cancer, and other diseases. A diet which was high in fat and cholesterol, sodium and sugar, and low in fibre, certain vitamins and minerals, was clearly linked to a high risk of heart disease, cancer and other chronic diseases as well as to a worse prognosis when those diseases arose. The links were causal.

Dr Barnard would agree with the conclusions of the 1990 World Health Organisation Report on Diet, Nutrition and the Prevention of Chronic Disease that: "Dietary factors are now known to influence the development of a wide range of chronic diseases, e.g. coronary heart disease, various cancers, hypertension, cerebrovascular disease, and diabetes." He believed that most other physicians would agree with the WHO position. There had been calls for changes in diet including the reduction of fat in diet since a report of the National Cancer Research Council in 1982.

So far as heart disease was concerned, Dr Barnard said that common heart disease or atherosclerosis was a disease in which plaques composed of cholesterol, fat, debris, and cells gradually obstructed the arterial passages. Foods that were high in fat and high in cholesterol, such as those served at McDonald's, tended to increase blood (serum) cholesterol levels and to lead to a higher risk of developing heart disease and a higher likelihood of death related to such disease. Research had established that higher cholesterol levels led to a higher risk of heart disease. On average every one percent increase in the amount of blood cholesterol raised the risk of a heart attack by two percent or more. (Lipid Research Clinics, 1984.) Meat-eaters had higher average cholesterol levels than vegetarians, clearly leading to a higher risk of heart disease. (Kestin 1989, Fisher 1986.)

Beef, cheese and other animal products contained cholesterol itself and saturated fat, both of which increased serum cholesterol levels.

There were 100 milligrams of cholesterol in four ounces of beef. Every 100 milligrams of cholesterol in one's diet added roughly 5mg/dL to the total serum cholesterol level of the average person.

The saturated fat in beef and cheese stimulated the liver to make more cholesterol. Meats had fat not only on the outer edge, but also marbled through the lean. Approximately fifty

Page 251

percent of beef fat was saturated, which was far higher than for vegetable oils. Those who avoided meat products such as those sold at McDonald's had reduced cholesterol levels.

The prevalence of ischemic heart disease was markedly reduced in populations which avoided meat. Dr Barnard said that he had looked into hundreds of studies into the links between diet and heart disease and there was a great deal of consistency to the effect that the less fat, the less saturated fat, the less cholesterol and, the less meat and dairy products that one consumed, the lower one's risk of heart disease was likely to be.

On 21st July,1990, the medical journal, The Lancet, published research findings of Dr Dean Ornish of the University of California at San Francisco with whom Dr Barnard had worked. The findings showed that heart disease could actually be reversed in 82% of patients using a combined regimen of vegetarian diet along with other lifestyle measures. (Ornish, 1990). But a diet containing lean meat had been demonstrated to make plaques worse over time. (Ornish 1990, Blankenhorn 1987, Brown 1990).

In Dr Barnard's view a diet including food products such as those sold at McDonald's could not accomplish this result. It tended to encourage the progression of heart disease. Heart disease was irrefutably connected with high-fat, high cholesterol diets. Foods served at McDonald's were precisely the type of foods which caused heart disease and encouraged its progression.

High fat and low fibre diets tended to go together and fibre helped to eliminate cholesterol from the body.

Dr Barnard said that hypertension (raised blood pressure) was related both to salt intake and to meat consumption. "There is, quite clearly, a causal link between diets that are high in fat and sodium, and hypertension."

One quite obvious and well accepted causal link was that a diet high in fat made obesity more likely. It increased body weight, partly because a diet that was high in fat was high in calories; but that was not the only reason. Dietary fats required almost no chemical conversion in order to be added to the body fat, unlike carbohydrates which required very extensive biochemical change within the body if their calories were going to be stored as body fat.

Page 252

If two individuals consumed precisely the same number of calories but one was getting those calories predominately from carbohydrates and the other from fat, the one who got the calories from fat would have a greater tendency towards obesity and a higher body weight than the one who was getting precisely the same number of calories but from carbohydrate. According to Dr Barnard, 23% of the calories in carbohydrate are expended simply trying to make fat.

In addition, when fat was consumed it had no influence on body metabolism. In other words, fat did not increase the speed with which the cells of the body consumed calories (energy) but carbohydrates did.

Sodium had a tendency to hold water and when individuals were on high sodium diets there was a tendency to draw water into the blood vessels raising the pressure of fluid within the blood. When individuals reduced their sodium intake their blood pressure fell. If you increased the sodium intake of someone with hypertension his blood pressure would be driven up further.

Hypertension was a serious condition. If you had excessive pressure in your arteries, that pressure accelerated atherosclerosis. When the arteries in the neck, into the head and to the brain were under higher than normal pressure, the likelihood of a haemorrhagic stroke was increased. Hypertension was a well known risk factor for stroke.

Turning to cancer, Dr Barnard said that cancer risk was elevated by foods which were high in animal fat, such as those sold at McDonald's. The U.S. National Cancer Institute estimated that more than 30 percent of cancers were linked to foods. He quoted with approval the U.S. Surgeon General's Report on Nutrition and Health published by the U.S. Department of Health and Human Services in 1988, to the effect that: ".....a comparison of populations indicated that death rates for cancers of the breast, colon, and prostate were directly proportional to estimated dietary fat intakes. Estimates varied, but dietary factors accounted for anywhere from 30% to perhaps 60% or, some authorities would say, 70% of cancers."

Dr Barnard gave an outline of the evidence for those conclusions.

For nearly twenty years epidemiological studies had shown that if you compared many different countries, those with high fat intakes had higher cancer rates. The 1988 report by the U.S.

Page 253

Surgeon General called this a direct correlation. It was quite striking that there was a very, very high correlation between fat intake and cancer rates. Dr Barnard said that that alone was an association. It did not prove cause and effect, but epidemiologic studies, as well as other kinds of studies , could be refined to give more evidence of what the association might be, and to shed some light on whether the factors might be causal.

When given his head to refer to reports or papers to which he attached particular importance Dr Barnard first returned to the Surgeon General's report of 1988.

On pages 194 and 195 under the heading "Role of Dietary Fats in Cancer" the report said:

    "Despite some inconsistencies in the data relating dietary fat to cancer causation animal studies show an effect on carcinogenesis and support a cancer-promoting role, and international epidemiologic studies have suggested that differences in dietary fat intake may provide a meaningful key to prevention of cancer. For example, substantial epidemiologic and animal evidence supports a relationship between dietary fat and the incidence of both breast cancer (Kakar and Henderson, 1985) and colon cancer (Kolonel and Le Marchand, 1986). Indeed, a comparison of populations indicates that death rates for cancer of the breast, colon, and prostate are directly proportional to estimated dietary fat intakes (Wynder et al. 1981; Rose, 1986)........Considerable uncertainties remain to be resolved about these relationships. For example, the effects of different types of dietary fat (i.e. saturated vs. unsaturated; animal vs. plant origin) have not been separated in most human studies. But the weights of the studies to date are strongly suggestive of the role of dietary fat in the aetiology of some types of cancer"

Dr Barnard said that the words "carcinogenesis" (which Dr Barnard described as the beginnings of cancer including the causation of cancer), "cancer promoting role" and "the role of dietary fat in the aetiology" were all references to causation and not simply to association.

Page 254

    Moving to "Human Epidemidogical Studies" the report said:

    "The risk of breast cancer is correlated with total fat consumption in comparisons of countries (Armstrong and Doll,1975; Grey, Pike, and Henderson,1979; Rose, Boyar and Wynder,1986), districts in Japan (Hirayama,1977 and query 1979), and ethnic groups in Hawaii (Kolonel, Hankin et al.,1981). The risk of cancers of the colon and prostate is also correlated with total fat consumption in international comparisons (Armstrong and Doll,1975; Knox,1977; Liu et al.,1979). A worldwide correlation between breast cancer mortality and total fat consumption has been demonstrated (Carroll and Khor,1975)......

    Although further epidemiologic study is needed to verify the association between diet and breast cancer and to elucidate its biologic basis, the consistency of the evidence derived from the epidemiologic and animal studies suggests that the association may be causal (Miller,1986). Table .... summarises certain key (although limited) dimensions of the human epidemiologic studies of diet and breast cancer."

At first Dr Barnard said that this meant that back in 1988 the Surgeon General was saying that the evidence strongly suggested that the relationship was causal. He agreed, when it was put to him, that the Surgeon General had used the words "may be causal."

Dr Barnard did not have any comment on what followed which in fact read:

    "Correlation studies show the strongest association. Migrant studies show that people who move to a country with higher incidence of breast cancer than their native country tend to acquire the dietary habits of their new country of residence and may experience a cancer incidence that changes with the change in dietary fat (Kolonel, Nomura et al.,1981; Gori,1979). Case-control and cohort studies relate the risk for breast cancer to total fat consumption in some (Miller et al.,1978) studies. While methodological problems may have obscured a true risk association in these negative studies (Willett et al.,

    Page 255

    1987; Herbert and Wynder,1987; Self et al.,1988), they reinforce the need for cautious interpretation and additional study of diet and breast cancer risk."

Dr Barnard went through tables in the report, which summarised epidemiological studies where positive or inverse correlations between Total Fat, Animal Fat, Meat, Eggs and Vegetable Fat and Breast Cancer had been found, or where no correlation had been found at all, before saying: "When the studies show a statistical relationship, that is quite important. When they fail to do so, it quite often is an error in method rather than meaning that there is simply no relationship between the two."

Dr Barnard would not agree that this put a sceptic in a "no win" situation. He said that a large majority of studies found a positive correlation; there were plausible mechanisms for dietary fat causing cancer, and as the Surgeon General went on to say:

    "Animals fed a high-fat diet often have higher rates of carcinogen-induced cancer of the breast, colon, and pancreas than those fed on low-fat diets (Carroll, 1986.)"

The report's table of epidemiological studies relating to dietary fat and colon cancer showed a large number where an inverse relationship was found between various types of dietary fibre and the incidence of colon cancer. In other words they indicated a protective effect of fibre intake for colon cancer.

Dr Barnard attached importance to what had happened in Japan as fat intake grew. He said that if one compared Japan and the United States, Japan historically had quite a low intake of fat, with fat providing as low as seven to ten percent of calories, compared to the United States' average fat intake of about 37 percent of calories. Japan had had a correspondingly low incidence of cancer, breast cancer in particular. However, Japan and the United States differed in many ways, not just in diets but in other aspects of lifestyle. Within Japan itself, where everyone was presumably breathing the same air and drinking the same water and had similar access to medical care and other factors, researchers had looked at dietary factors and their relationship to cancer. They found that higher fat intake was linked and that it went hand in hand with a higher risk of breast cancer.

Page 256

Researchers had initiated and completed other kinds of studies such as cohort studies. In a cohort study the researcher identified a group of subjects, identified their diet and then followed them prospectively to see their rate of cancer.

Dr Barnard went through the abstract of Hirayama's 1978 paper "Epidemiology of Breast Cancer with Special Reference to the Role of Diet" as follows:

    "Breast cancer is still relatively infrequent in Japan. However, both mortality and morbidity rates have sharply increased in recent years, especially in ages 45 - 59. The risk was noted to be 8.5 times higher in women of high socioeconomic strata eating meat daily compared with women of low socioeconomic strata who do not eat meat daily, when 142,857 women age 40 were followed for 10 years. A high positive correlation was found between per capita fat intake and adjusted death rates of breast cancer in different districts of Japan. It was estimated that the breast cancer death rate will rise to the U.S. level when Japanese dietary fat intake approaches present day U.S. levels. The close correlation with fat intake was noted to come mainly from the consumption of pork and animal fat. The ratio of recent increase in breast cancer death rates was also found to be under the combined influence of animal fat AF2, a highly mutagenic food additive widely used in Japan from 1965 to 1975 and shown to produce mammary carcinoma in rats. A series of case-control studies reveals the higher risk of breast cancer with the increase in body size especially in postmenopausal women. The recent breast cancer increase could therefore be a reflection of the fact that women in Japan are becoming heavier, especially after age 30."

Dr Barnard interpreted this as finding associations between fat and cancer, particularly meat intake and cancer, but with a doubt as to whether animal fat operated directly or was mediated by obesity. Increase in body weight did not necessarily involve obesity. The subjects had a tendency to be larger in size and a greater proportion of them were obese.

In Dr Barnard's view Hirayama's work eliminated genetic factors from consideration for those changes. There

Page 257

was a role for genetics in breast cancer. However, when you had a population that had remained the same and their cancer rate had gone up rather dramatically, yet their DNA had not undergone some massive change, one had to presume that there was an environmental factor, and the key one seemed to be fat.

Studies were also done by Kolonel in Hawaii showing that with the same geographic area, where again one would have similar exposure to carcinogens and pollutants, again a high fat intake was linked with a high rate of breast cancer.

Migrant studies had also shown the same sort of thing; that people who changed from one geographical location to another quite quickly adopted the diets of their newly adopted land and tended to assume the risk of breast cancer of that area.

Dr Barnard particularly relied upon a 1989 paper of Toniolo and others on "Calorie-Providing Nutrients and Risk of Breast Cancer." The abstract read:

    "A case-control study was conducted in Italy to investigate the role of diet in breast cancer. Cases were 250 women with breast cancer, and controls were a stratified random sample of 499 women from the general population. A dietary history questionnaire was used to measure the intake of total fat, saturated fat, animal proteins, and other macronutrients. In multivariate analyses, the relative risks of breast cancer for women in highest quintile of consumption of saturated fat and animal proteins were 3.0 (95% confidence interval, 1.9-4.7) and 2.9(1.8-4.6), respectively. A reduced risk was found for women who derived <28% of calories from fat versus >36%. A similar reduced risk was found for women who derived <9.6% of calories from saturated fat or <5.9% from animal proteins. These data suggest that during adult life, a reduction in dietary intake of fat and proteins of animal origin may contribute to a substantial reduction in the incidence of breast cancer in population subgroups with high intake of animal products."

Dr Barnard said that a "95% confidence interval 1.9-4.7" meant that the chances were that 95 times out of 100 the true

Page 258

value was somewhere between 1.9 and 4.7 higher risk. The paper suggested that the women who were in the highest fifth of the group of consumption of saturated fat and animal proteins had three times the risk of breast cancer. Women on a lower fat diet had substantially less risk. If those on high fat diets reduced their fat intake "there is an expectation there may be a reduction in subsequent breast cancer risk."

Dr Barnard said that an additional line of evidence linking fat and breast cancer was the fact that there was very clear cut evidence of mechanisms by which fat influenced cancer risk. A high fat diet and a diet that was low in fibre, because the two tended to vary inversely, increased oestrogen levels in a woman's body. Oestrogen levels that were higher were clearly and unquestionably linked to a higher rate of breast cancer. The breast cells were responsive to oestrogens. So, when a high fat diet elevated oestrogen levels, this was believed to be at least one important mechanism for increasing the risk of breast cancer.

In addition, a high fat diet encouraged obesity by the fat laying mechanism which Dr Barnard described in relation to obesity and hypertension. Adipose (fat) tissue increased oestrogen levels in the body because the fat allowed the production of oestrogen. Thus a high fat diet increased the risk of breast cancer. The relationship between the two was causal. National Cancer Institute books on cancer stated that if fat intake was dropped from the current American average of about 37% of calories, to 20%, that could be expected to lead to a 17% reduction in oestrogen levels, and that was believed to be at least one mechanism by which a low fat diet could be protective against breast cancer.

When Dr Barnard spoke of "obesity" in relation to hypertension and breast cancer, it was not always clear whether he used the word in the medical sense of being grossly overweight or in the lay sense of overweight. I learned from other evidence that there is a standard calculation of "body mass index" which involves dividing weight in kilograms by the square of height in metres. 25 and over is "overweight." 30 and over is "obese." So a man who is 1.83m (6 feet) tall becomes "overweight" at about 84 kilos (13st 3lbs or 185 lbs), but he does not become "obese" until he reaches

Page 259

about 100 kilos (15st 10lbs or 220lbs), if my arithmetic is correct. For a woman who is 1.6m (5ft 3ins) tall the equivalents would be 64 kilos (10st or 140lbs) and 77 kilos (12st 2lbs or 170lbs).

Dr Barnard said that the age of menarche was a risk factor in breast cancer; the earlier the age of menarche, the higher the risk of breast cancer. As the fat content of the diet had increased and the fibre content had decreased in many western countries, the age of menarche had dropped. The same was true of Japan. Earlier menarche was the result of overall diet over a prolonged period.

Dr Barnard referred to the 1985 paper of Gregorio and others on "Dietary Fat Consumption and Survival Among Women with Breast Cancer" which was a prospective study of 953 women with breast cancer and which showed that when controlling for disease stage and patient age at diagnosis the estimated risk of death increased 1.4 fold, i.e. by 40%, for each 1000g increase in monthly fat intake. Dr Barnard said that 1000g of fat could easily be the difference between a low fat diet and a typical English and American diet over a month. The paper was also authority for oestrogen which influenced normal breast tissue development being a suspected risk factor in breast carcinogenesis. Dr Gregorio said that persistent oestrogenic stimulation in women increased their risk of cancer. Dr Barnard said that this was important when considering the effects of high fat diets and of obesity on oestrogen levels. Dr Gregorio said that high-fat diets had been hypothesised to contribute in several ways to sustained elevation of serum oestrogen levels.

Dr Gregorio also said that other factors which had been shown to increase the risk of developing breast cancer were early age at menarche, low parity, late age at first pregnancy and late age of menopause.

Dr Barnard said that what he got out of Gregorio was that there was cogent evidence for the level of oestrogen and the length of time over which it was a factor, which could itself be related to early menarche and late menopause, having a causal effect on breast cancer; and that a high fat diet contributed to that.

Dr Barnard referred to Dr L.J. Kinlen's 1987 paper

Page 260

"Fat and breast cancer". Dr Kinlen works at the Cancer Research Campaign's Cancer Epidemiology Unit at the University of Edinburgh. The summary of the paper read:

    "Breast cancer mortality and incidence in different countries show a strongly positive correlation with the per caput consumption of fat. In addition, the disease has increased among the Japanese, both in Japan and in the United States, and in both groups fat consumption has been increasing. In contrast, both case-control and prospective studies have on the whole failed to confirm the relationship. Despite these negative findings, the hypothesis that fat causes breast cancer has continued to be popular. The evidence for fat as a cause of breast cancer seems to have been exaggerated, and insufficient attention given to alternative explanations for the geographical correlations and for the changes among the Japanese in the frequency of the disease. These include the effects on breast cancer risk of body weight, body size, age at menarche (all influenced by excess calories) and age at the birth of the first child, as well as effects of obesity on the fatality rate in breast cancer. Evidence is lacking that the source of calories is important."

However, Dr Barnard wished to draw my attention to two paragraphs of the paper which read:

    "There is a way of resolving the differences between the international correlational and the individual-based studies - other than always attributing the negative findings of the latter to methodological defects - and which involves postulating no new risk factor for breast cancer. The associations with fat may be indirect and may reflect the effects of 'excess calories' and other fertility linked risk factors.

    There is much evidence for the role of 'overnutrition' in breast cancer aetiology, and the more calorigenic nature of fat then either carbohydrate or protein suggests a connection. In the large prospective study by the American Cancer Society (Lew and Gartinkel, 1979), a significant trend in relative risk was

    Page 261

    found with increases in weight, as in a smaller study in the Netherlands (De Waard and Baanders, 1974). Case-control studies have also produced evidence of this effect at postmenopausal ages. Since nutritional status also influences age at menarche, the effects of calorie intake on the incidence of breast cancer are mediated not only by obesity but also by age at menarche, a much longer established risk factor for the disease."

    In Dr Barnard's view Dr Kinlen agreed with Dr Gregorio and the effect of these papers was that there were causal links between dietary fat and cancer. Dr Barnard referred to the rule of thumb that a gram of fat provides nine calories. A gram of carbohydrate or of protein has only four calories. This was the reason why it was well known and long established that high fat diets encouraged obesity. It was what Dr Kinlen was referring to when he stated that high fat foods were, by their very nature, much more calorie dense, higher in calories on a per weight basis, than high carbohydrate foods or high protein foods.

    Dr Barnard then took me to a later 1991 paper by Dr Kinlen on "Diet and breast cancer".

      The introduction to the paper read:

      "The issue of diet as a cause of breast cancer has been dominated by fat. Some have judged this convincing enough to warrant dietary recommendations aimed at prevention. Others find the evidence so far rather weak.

      This hypothesis arose because fat appeared to cause mammary tumours in rodents and because in women, mortality from the disease in different countries strongly correlates with the corresponding per caput fat consumption as crudely measured in food balance tables. These correlations concern animal, and not vegetable, fat and are stronger for postmenopausal than pre-menopausal breast cancer. Other early observations that seemed relevant are the increase in breast cancer among the Japanese in the USA, a group whose intake of fat has also increased. However, all such evidence is crude and indirect and for causation to be invoked persuasively, individual women must be shown to have a higher than average fat intake. Moreover since this may simply reflect high energy requirements as determined by body size, physical activity or individual metabolism, affected women should ideally be shown to obtain a higher proportion of calories from total fat then do controls."

    However, Dr Barnard was interested in two short passages on menarche and obesity. The paragraph on menarche read:

      "Age at menarche is a well established risk factor for breast cancer and later menarche (and consequently lower risk), like height, can reflect caloric restriction during the years of growth."

    Page 262

    Dr Barnard said that this was entirely resonant with Dr Kinlen's earlier paper.

      I note that the next paragraph in the paper read:

      "Other fertility factors. Late age at first birth, low parity and late menopause (all risk factors for breast cancer) are commoner in high fat-consuming countries. They may therefore indirectly contribute to the international correlations."

      Then a paragraph on obesity included the statement: "In contrast to the findings at pre-menopausal ages, post-menopausal breast cancer has shown a positive relationship with body weight in nearly all the (more than 20) studies of the subject. This is probably an oestrogen effect since androstenedione is converted in adipose tissue to oestrone."

    Dr Barnard said that Dr Kinlen was simply pointing out that studies have consistently shown that for postmenopausal breast cancer obesity was a risk factor and he offered a mechanism which linked them in a causal way.

    I assume that most obese, post menopausal women have been following the same sort of diet for a considerable of length of time.

      I note that Dr Kinlen's conclusion to the whole paper read:

      "There are well founded reasons concerning obesity and coronary heart disease for reducing consumption of animal fat. As a cause of breast cancer, however, the contradictory nature of the evidence prevents any definitive conclusion. Much of the international data that suggests a relationship may simply reflect the effects of calorie restriction during growth in certain countries, or over-nutrition in others, respectively reducing or increasing breast cancer risk. The largely negative findings of prospective studies, the great scope for bias in case-control studies, and the possible confusion between the

      Page 263

      effects of fat per se and those of excess calories all make a conservative position prudent in evaluating the findings to date about fat. There is a dearth of data about individuals with a very low fat intake and a need also for a combined analysis of all the prospective studies. A stringent approach is required in designing future studies of the role of fat in this disease. A protective effect of green vegetables suggested by certain studies requires confirmation."

    Dr Barnard also referred to a 1991 paper by de Ridder and others: "Dietary habits, sexual maturation, and plasma hormones in pubertal girls: a longitudinal study". Omitting the statistics, the Abstract read:

      "The relationships between dietary components and physical or hormonal sexual maturation in 63 pubertal girls were examined. The effects of vegetable protein and dietary fibre on breast development and menarche became more pronounced in a multivariate analysis, after elimination of the linear effects of body height and energy intake. From the multivariate analysis with combinations of vegetable protein, polysaccharides, and fibre in the equation, fibre appeared to be the most important factor. The gonadotropin and estradial plasma concentrations were higher in girls who consumed less grain fibre. We conclude that a diet rich in vegetable products, especially fibre, may affect the rate of physical and hormonal sexual maturation, possibly mediated by the hypothalamus-pituitary-gonad system."

    Dr Barnard said that this meant that girls on higher fibre diets had lower levels of estradiol and reached menarche later, meaning that their risk of breast cancer would be lower. Those girls who had less fibre in their diet and higher levels of estradiol reached menarche significantly earlier, indicating that their risk of breast cancer would be higher.

      Dr Barnard pointed out that the paper went on to say that:

    "An adequate intake of nutrients together with good health is probably responsible for the shift in the onset of puberty and menarche to an earlier age. The high-calorie Western diet increases body fat mass, most notably intra-abdominal fat. In turn, the amount of body fat modulates the timing of the growth spurt and menarche as postulated by Fisch and others about 20 years ago."

So Dr Barnard said that what Dr de Ridder was bringing to our attention was that the links between adiposity and earlier menarche are already established. The causal links between a high fat diet and obesity, even not necessarily frank obesity, but increased body weight, was also well established. When girls were on higher fibre diets menarche occurred later, reducing the risk of cancer. If one looked at the differences in fibre intake that were used in the study and that were shown to have a significant difference, they varied between about 17.5 grammes per day and about 19.5 grammes per day, suggesting that small differences in fibre of the range that were commonly consumed in western countries had a pronounced and significant effect on the age of menarche and hence had an effect on breast cancer risk.

Dr Barnard said that dietary fat shortened the survival of patients diagnosed with cancer. In a Canadian research study, women with cancer were more likely to have lymph node involvement if they had a higher intake of saturated fat. (Verreault 1988)

Dr Barnard said that vitamins and minerals played a role in protecting against breast cancer. Gladys Block who was a very well known and well respected cancer researcher in the United States, had reviewed the literature on vitamin C and had reported that breast cancer occurred less frequently in women who consumed diets that were high in vitamin C. Vitamin C was found in vegetables and fruits. There was no vitamin C in meat products or in animal products in general. Dr Barnard did not refer to potatoes, despite the popularity of McDonald's "fries."

Dr Walter Willett had shown that women whose diets were low in vitamin A which was produced in the body primarily from Beta-carotene but could also come from vitamin A in the diet, had a higher risk of breast cancer. Meat based diets tended to be low in Beta-carotene.

So far as cancer of the colon was concerned Dr Barnard said that both fat and fibre appeared to play a role.

Strong links had been found between the consumption of meats and other fatty foods and colon cancer. (National Research Council,1982; Willett,1990; Gerharsson,1991). When the past diets of cancer patients were studied, it was clear that meat-based diets were linked to colon cancer.

Dr Barnard referred to the very extensive, prospective study of Dr W.C. Willett and others into the relation of meat, fat and

Page 265

fibre intake to the risk of colon cancer. He said that according to Dr Willett's study of 88,751 women nurses, meat and animal fat were clearly correlated with colon cancer risk. Women who ate beef, pork, or lamb daily had a relative risk of colon cancer of 2.49, i.e. two and a half times the risk of colon cancer compared to those who ate those products less than once a month. (Willett,1990). Fruit fibre had a non-significant negative association with colon cancer risk, and there was no demonstrable association between vegetable oils and cancer.

Another report found that the risk of developing colon cancer for Adventists in the upper third of animal fat consumption (among Adventists) was 1.95 times the risk for those in the lower third.

Dr Barnard said that lower colon cancer incidence rates in vegetarians had been attributed to lower levels of cholesterol and bile acids, particularly secondary bile acids, in their intestinal tracts. (Aries,1971; van Faassen,1987; Nair,1984; Turjman,1984; Reddy,1973). The ratio of secondary to primary bile acids was reduced as meat consumption was reduced (Turjman,1984), at least in part due to favourable changes in the gut flora when vegetarian diets were adopted. (Aries,1971; van Fassen,1987). Evidence of dietary links with colon cancer was not only epidemiologic. There were also biological mechanisms that explained the observed associations. For example, meat-eaters tended to have gut flora which converted the primary bile acids in digestive secretions to secondary bile acids which were carcinogenic.

Dr Barnard also said that heating meat in the process of cooking caused carcinogens, i.e. cancer causing hydrocarbons, to form on its surfaces.

The reference to carcinogens brought Dr Barnard to the protective role of fibre in relation to colon cancer also. This had been mooted since a review of epidemiological evidence in 1971. The theory was that carcinogens that might be in the digestive tract could be deluded by a high fibre intake. A high fibre intake made stools considerably more bulky and the fibre absorbed water rather like a sponge soaked up water. The carcinogens were absorbed and deluded in the process. That reduced the concentration of carcinogens on the wall of the large bowel. This was the presumed mechanism, or one of the mechanisms.

Page 266

In addition, a diet which was high in fibre and low in fat modified the digestive flora.

Dr Barnard said that the international comparisons of colon cancer parallelled those with breast cancer, in that countries that had a higher fat intake had a higher rate of colon cancer in general terms. Likewise, countries that had a higher fibre intake had a lower rate of colon cancer. Case-control studies in which individuals who had cancer were compared to individuals who were like them in other ways but who did not have cancer, showed an association between fat, saturated fat and total calories and cancer of the colon.

Dr Barnard said that, as with breast cancer, there had been some studies that had found no relationship, but the preponderance of evidence had been such that cancer authorities in the United States had felt that it was long past the time to make recommendations that Americans increase fibre consumption and reduce fat consumption.

In Dr Barnard's view the link between a high fat, low fibre diet and cancer of the colon was "almost certainly causal".

Dr Barnard referred to a 1991 paper by De Verdher and others: "Meat Cooking methods and colorectal cancer: A case reference study in Stockholm", which contained the statement:

    "Fat tends to increase frying temperature and results in increased production of mutagens during cooking. Therefore, even though adjusting for fat intake reduced the associations between a heavily browned meat surface and colorectal cancer, it could be argued that it is inappropriate to adjust these associations for fat since this may adjust for the formation of the very mutagens that are the exposure of interest........

    The results in the present study indicate that there is an interaction (more than an additive effect) between browning of the meat surface and protein, fried meat and gravy. In support of this, mutagen formation has been shown to be related

    Page 267

    to the protein content of cooked food. Furthermore, fried meat contains carcinogenic heterocyclicamines, the formation of which depends on high cooking temperature".

Dr Barnard said that "mutagens are chemical substances which can damage DNA and can lead in some cases to cancer".

So, according to Dr Barnard, fried meat contained carcinogenic heterocyclicamines, i.e. cancer causing chemicals.

Dr Barnard referred to a U.S. National Cancer Institute booklet revised in 1991 called "Cancer of the Colon and Rectum". This contained a paragraph:

    "Colon cancer risk also appears to be related to the amount of fat in the diet. Considerable evidence links a high-fat diet to a high incidence of colon cancer. Scientists theorize that dietary fat increases the amount of bile acids (needed for digestion) in the colon. These substances may act directly to damage the bowel lining, or they may be converted to secondary bile acids, which are known to produce tumours in animals. A high-fat diet may also affect the risk of colon cancer by increasing the amount of other damaging chemicals (lipid peroxides) in the bowel, by causing changes in the cell membrane, and/or by affecting hormone regulation."

Dr Barnard said that his understanding of the "links" which appeared in official government documents describing the relationship between high fat diets and, in this case, colon cancer was that the high fat diet was suspected of making a contribution although it was not necessarily direct or firmly established as a cause.

The same booklet also referred to the mutagens produced when food, especially meat, was cooked at high temperatures and the possibility that vitamins C and E reduced the recurrence rate of colorectal polyps, which might have important implications for reducing the risk of colorectal cancer given the strong evidence that colorectal cancer developed from polyps.

The booklet appeared to be based on a National Cancer Institute database of medical reference books and recent articles.

Dr Barnard said that the observations of mutagens and carcinogens produced in the cooking of meat particularly at high temperatures were observations of laboratory animals.

Page 269

The thesis that fat caused colon cancer by promoting the translation of bile acids into secondary bile acids depended on the carcinogenic effects of secondary bile acids on animals.

Dr Barnard said that he would agree with a statement in a paper by Dr Clement Ip and others that "in reality, animal models are woefully inadequate to meet the challenge of offering definitive answers to a complicated area of research as in human nutrition and cancer aetiology", because animal models or the use animals in experimentation could not provide definitive answers, although the stress was on "definitive".

However Dr Barnard said that animal studies were consistent, and if you put all the lines of evidence together, from the animal studies to international comparisons to within country comparisons that helped to rule out genetic factors, migrant studies, cohort studies, the survival studies of cancer patients, the mechanisms particularly relating to hormones, which were no longer controversial and were quite well established, as well as the studies of vitamins and minerals, there was a very, very compelling body of evidence showing not simply that there was an association between fat and cancer, but that diets that were high in fat and low in fibre and low in vitamins and minerals played a causative role in various forms of cancer. This had led all major research bodies in the United States to recommend reductions in dietary fat. He agreed that it was probably true that dietary fibre had a role to play in the inhibition of tumour genesis independently of the amount of fat in the diet, but he repeatedly stressed that fibre and fat normally presented in an inverse ratio in a person's diet.

Dr Barnard referred to a paragraph in a book "Diet, Nutrition, and Cancer" prepared by the Committee on Diet, Nutrition, and Cancer of the Assembly of Life Sciences, National Academy of Sciences in the U.S. and published in 1982. The paragraph read:

    "The committee concluded that of all the dietary components it studied, the combined epidemiological and experimental evidence is most suggestive for a causal relationship between fat intake and the occurrence of cancer. Both epidemiological studies and experiments in animals provide convincing evidence that increasing the intake of total fat increases the incidence of cancer at certain sites, particularly the breast and colon, and, conversely,

    Page 269

    that the risk is lower with lower intakes of fat. Data from studies in animals suggest that when fat intake is low, polyunsaturated fats are more effective than saturated fats in enhancing tumorigenesis, whereas the data on humans do not permit a clear distinction to be made between the effects of different components of fat. In general, however, the evidence from epidemiological and laboratory studies is consistent."

Dr Barnard thought that "it was subsequent to this that virtually all American scientists and policy makers became quite convinced of the link and described it in the causal terms that were indicated in this document".

Another paragraph in the same book sought to deal with the contribution of diet to overall risk of cancer, saying:

    "Higginson and Muir (1979) estimated the proportion of cancers related to various aspects of the environment. They believed that precise proportions of cancer incidence could not be attributed to diet, but they did include dietary factors among the general heading "Lifestyle." They estimated that possibly 30% of cancers in men and 60% in women in the Birmingham and West Midland regions of England and Wales could be attributed to lifestyle. Wynder and Gori (1977) were more specific. On the basis of international and intranational comparisons of cancer incidence, the differences between U.S. mortality rates and the lowest reported worldwide mortality rates for each site, and results of specific case-control studies, they concluded that a little more than 40% of cancers in men and almost 60% of cancers in women in the United States could be attributed to dietary factors."

Dr Barnard said that those figures had been discussed and quoted numerous times in the research literature.

Dr Barnard thought that the Second Plaintiff's own publication "Good Food, Nutrition and McDonald's", which was produced in 1984 and 1985, simply stated what had been known for quite a period of time and was commonly accepted, when it said:

    "In recent years, doctors and nutritionists have

    Page 270

    also become aware of other important factors in healthy eating. There is a considerable amount of evidence to suggest that many of the diseases which are more common in the western, affluent world -diseases such as obesity, diabetes, high blood pressure, heart disease, stroke and some forms of cancer - are related to diet. The typical western diet is relatively low in dietary fibre (roughage) and high in fat, salt and sugar.

    Many countries have therefore published "Dietary Guidelines" - general recommendations concerning diet which are aimed at the whole population in order to prevent these diseases and to promote good health.

    These recommendations suggest that on average, people should eat more foods rich in dietary fibre (cereal foods, vegetables and fruit), and less of the foods rich in fat, sugar and salt (sodium).

    It is important to remember that although diet is an important contributory factor to these diseases, it is not the only one. Smoking, lack of exercise, an excess of alcohol, stress and many other factors also affect the incidence of the so-called diseases of affluence."

Dr Barnard would not take exception to any of that. He said that it seemed to convey more than suspicion but less than conclusive evidence that the diseases mentioned were caused by diet.

Dr Barnard's evidence in chief finished with a reference to the Executive Summary of the 1990 WHO Report on Diet, Nutrition and the Prevention of Chronic Heart Disease which said that the report issued a series of "population nutrient goals" put forward as a universal guide to the nutrient intake needed to prevent all diet related diseases and appropriate for application in all countries throughout the world. "Lower and upper intakes are set for each of the main nutrient groups, including total fat, saturated fatty acids, polyunsaturated fatty acids, protein, total carbohydrates, complex carbohydrates, and free sugars. Expressed as a proportion of total energy, this recommended "safe" range of intakes specifies the minimum intake of a nutrient needed to prevent deficiency diseases and the maximum

Page 271

intake that should not be exceeded in the interest of preventing chronic diseases. Recommended daily intakes, expressed in grams, are issued for salt, dietary fibre, dietary cholesterol, and fruits and vegetables. The zero value given as the lower limit for saturated fatty acids, dietary cholesterol, and free sugars indicates that these dietary components meet no special nutritional need and are thus not required for the prevention of any deficiency disease".

Dr Barnard interpreted this to mean that the experts responsible for the report were saying that an ideal diet from a health stand point would eliminate saturated fatty acids, dietary cholesterol and free sugars since they were not needed.

Dr Barnard did agree, towards the end of his cross-examination, that it was reasonable for the WHO report to say: "Saturated fatty acids and cholesterol are not essential nutrients and their importance relates directly to their effects in increasing blood cholesterol concentrations and promoting the development of coronary heart disease. As noted previously no lower limit to serum cholesterol has been identified below which a beneficial reduction in coronary heart disease cannot be expected so national nutrition policies should seek to minimize intake of saturated fatty acids. These fatty acids may also be specifically involved in promoting cancers, particularly of the colon and breast, although the evidence remains inconsistent. The main justification for limiting saturated fatty acids intake should therefore be the prevention of coronary heart disease".

Dr Barnard accepted that heredity, genetics, had a role to play in cancer; but he said that there were a number of uncertainties about it. He did not think that anyone believed that heredity was the prime determinant factor, except for a rather small number of cancers.

I thought, from some of his answers, that Dr Barnard was prepared to accept that if certain kinds of diet were part of the cause of heart disease or of cancer of the breast or colon the probability was that the diet in question would have to be consistently sustained for a considerable period to do any harm. He spoke of breast cancer being a chronic disease. But he also said that he found no scientific support for the suggestion that, because the dangers of high-fat, high-cholesterol, fibre-depleted foods were demonstrated in people consuming such diets on a regular basis, eating such foods intermittently was somehow safe. In his view there was no established threshold below which their consumption was known

Page 272

to be safe. He said that one might similarly assert that it is difficult to prove a health threat from the occasional use of cigarettes. That possibility would not change the fact that cigarettes clearly contributed to risk of illness and death. He said that one could facetiously argue that Russian Roulette, played with a single bullet and a six-chamber revolver, had only a 17 percent chance of inflicting any injury. This hardly made the practice a safe one. I did not find either of these examples a true or helpful analogy with diet and heart disease or cancer of the breast and bowel. Eating a certain kind of food intermittently is not going to effect the diet and the studies upon which Dr Barnard relied were concerned with diet rather than intermittent eating of certain foods.

However, Dr Barnard said that McDonald's products clearly contained significantly more fat than government guidelines and other health authorities recommended, and that even if one ate restricted amounts of McDonald's foods, they remained part of the problem, rather than part of the solution. Every time a person ate a typical McDonald's meal, they lost an opportunity to reduce the fat and cholesterol content of the diet. That constituted a significant health problem.

He also said that fatty foods tended to be habituating. Individuals consuming high-fat meals did not just erode the healthfulness of their diets at that one meal; they also increased the likelihood of continued high fat intake, due to the habituating nature of such foods, which was apparently due to both social and physiologic factors. (Birch 1992; Drewnowski 1990, 1992). By habituating Dr Barnard did not mean physically addictive, but simply that fatty foods tended to became part of the diet and stay in the diet. People tended to want to maintain a relatively constant fat intake. It was the same with salt.

So far as the level of risk from a high fat diet was concerned Dr Barnard said that if one looked at the overall effect of fibre, fat and deficiencies in vitamins and minerals, the degree to which those factors collectively influenced the risk of breast cancer was "quite large, certainly much more than threefold, because that is the type of difference that one sees in international variations".

But Dr Barnard did not give me any equivalent figure for heart disease or colon cancer.

Page 273

I was referred to an article by Dr Willett in the 30th March, 1990, edition of Nature, which included the statement: "Among non-smokers, cancers of the breast and colon are the most important malignancies in Western societies. By the age of 75 years, over eight per cent of women in the United States will develop breast cancer and about three per cent of both sexes will be diagnosed as having colon cancer". However, this was not pursued with Dr Barnard or any other witness to show a degree of risk increased by diet high in fat.

Dr Barnard agreed that whenever it was said that something in diet caused some adverse effect one had to read in the words "in some people". He said that even with smoking and cancer where nearly everyone now agreed that the evidence was very strong, one could only say that "smoking causes cancer in some people". The fact was that smoking usually did not cause cancer.

Some attempt was made to overcome the gap as to the extent of risk by Dr Barnard's Civil Evidence Act statement made on the 1st July, 1996, some months after I had ruled on the meaning of the leaflet in this part of the case.

In that statement Dr Barnard first dealt with the effect of dietary cholesterol so far as the risk of coronary heart disease was concerned.

He repeated that a Big Mac contained 83 mg of cholesterol, a cheeseburger 41 mg and a hamburger 29 mg. He quoted a paragraph from a 1988 article in an American Medical Association journal as follows: "In humans, ingestion of dietary cholesterol raises serum cholesterol, largely through its effect on low-density lipoprotein-cholesterol. Over the range of intake in usual American diets, this effect is substantial, e.g. with 300 mg of cholesterol intake per 1,000 kcal, rather than 100, serum cholesterol is on average about 6% to 7% higher, equivalent to a 12% to 14% greater risk of coronary heart disease".

Dr Barnard said that the Lipid Research Clinics Trial, one of the best-known multi-centre research studies on heart disease, showed that on average every one percent increase in the serum cholesterol level was associated with an increase in the risk of a heart attack of two percent or more. He said that dietary cholesterol also contributed to heart disease apart from its effect on serum cholesterol levels, as another article described as follows:

Page 274

    "In addition, since 1981, four prospective within-population studies have shown that dietary cholesterol intake of individuals is significantly related to their long-term CHD risk, independent of and in addition to serum cholesterol, blood pressure, and cigarette use. On average, a 200 mg/1000 kcal higher intake of cholesterol at baseline was associated with a 30% higher CHD rate (95% confidence interval, 1.1 to 1.5). Conversely, lower intakes of cholesterol were associated with significantly lower risks of CHD, and of all causes of mortality as well (sic)..... The importance of a low-dietary cholesterol intake for prevention of CHD merits emphasis."

Dr Barnard concluded that there was no reasonable doubt that dietary cholesterol increased the risk of heart disease and that foods of the type sold at McDonald's contained sufficient cholesterol to present a genuine risk.

I allowed the Defendants to put Dr Barnard's July,1996, statement in evidence although by then I had no intention of allowing any further nutritional evidence from the British witnesses. That part of the case had just closed by then, and I was unwilling to allow it to be reopened save for the admission of Dr Barnard's statement which I allowed in because his recall was not practicable and I had allowed the recal of British witnesses on both sides. So the Plaintiffs did not call evidence specifically to deal with Dr Barnard's July,1996, statement.

However, I see problems with the parts of it to which I have just referred. Firstly, the articles to which Dr Barnard referred and upon which he relied were not produced, and, as I will say later, I thought that his selections from the articles which I did see were partial. Secondly, the increases in cholesterol to which the two quotes taken by Dr Barnard refer [200 mg/1000 kcal or 300 mg rather than 100 mg] seem to be very high. For instance, I know from other evidence that a 1988 U.S. cheeseburger provided 310 kcal of energy, so the 41 mg of cholesterol in it amounts to 132 mg/1000 kcal.

Thirdly, the excerpts of the articles which Dr Barnard quoted give no indication of how long the higher intake of cholesterol has to remain in the subject's diet before the greater risks of coronary heart disease arise. The second quote

Page 275

referred to "long-term CHD risk". Dr Barnard did say that in the vast majority of individuals who reduce their fat and cholesterol intake their cholesterol levels will fall. It was a cumulative effect, and what was most important was the overall effect of a person's diet.

Fourthly, given an increase in risk there is no indication of just how great the risk actually is, although Dr Barnard said that it was estimated that over 6 million Americans had coronary heart disease in 1992.

Much the same comments apply to what Dr Barnard had to say about dietary fat and the risk of coronary heart disease in his July,1996, statement. He said that animal fats were particularly rich in saturated fatty acids (SFA). Approximately fifty percent of beef fat was saturated. He repeated his evidence that fat contributed 55% of the calories in a Big Mac, 45% of the calories of a cheeseburger, 47% of the calories of McDonald's french fries, and 39% of the calories in a hamburger. Saturated fat tended to increase serum cholesterol levels through its effect on the liver. The relationship between saturated fat intake and total cholesterol (total-C) levels was summarized in an article drafted by scientists with the National Cholesterol Education Program, published in the Journal of the American Dietetic Association in 1988:

    "... for every 1% increase in the SFA intake (on a caloric basis), plasma total-C increases approximately 2.7 mg/dL."

In Dr Barnard's view this meant that for an average person with a cholesterol level of roughly 200 mg/dL, every one percent increase in saturated fat intake would then be associated with more than a one percent rise in plasma cholesterol which, as noted above, was associated with at least a two percent increased risk of heart disease. This effect of each percentage point increase in dietary fat had to be added to the effect of the dietary cholesterol which was often part of fatty foods. Together, the effect on risk was substantial.

So far as diet and cancer risk were concerned, Dr Barnard's July,1996, statement said that cancer remained a serious and growing problem. There was no question that diet played a substantial role in cancer risk. More than 1.2 million Americans were diagnosed with cancer (excluding basal and squamous cell

Page 276

skin cancers and in situ carcinomas except bladder) annually, and more than 500,000 died of the disease, accounting for one in every five deaths in the U.S. The most common forms of cancer in the United States were prostate, breast, lung, and colorectal.

Dr Barnard repeated his view that international and case-control studies had shown a strong relationship between fat intake and breast cancer risk. He did say that most cohort (prospective) studies had not shown a relationship. This same observation was recently made by Hunter et al. in the New England Journal of Medicine. This observation was not new; rather it reflected the limitations of cohort studies. First, they had typically investigated only a narrow band of fat intake. Second, they did not generally account for fat intake during adolescence, which might be the time when fat exerted its greatest effect on breast cancer risk. Third, their follow-up periods were typically short.

It had long been apparent that a substantial portion of the risk for cancers other than breast cancer was attributable to dietary factors. Colorectal cancer, for example, was diagnosed in 152,000 Americans and killed 57,000 annually. Women and men who consumed meat frequently had 2.5 and 3.6 times the incidence of colon cancer, respectively, compared to those who consumed these products rarely or never. Dr Barnard's references again included Willett. He said that the elevated colon cancer risk from diets high in animal products was believed to be caused by the higher levels of cholesterol and bile acids, particularly secondary bile acids, that those products caused in the intestinal tract.

Dr Barnard concluded that medical consensus held that an increased consumption of high-fat, low-fibre foods affected the composition of the diet in such a way that there was a very real risk of cancer of the breast or bowel or heart disease as a result. When the dietary composition was altered in this way, it added a very real risk of obesity, which, in turn, increased the risk of breast cancer and coronary heart disease; such increased risk is established beyond any scientific doubt.

He did not, however, give me any figure, however approximate, for the risk of heart disease or of cancer of the breast or bowel, attributable to the guilty diet.

Dr Barnard wrote a book in 1993 called "Food for Life" and

Page 277

subtitled "How the New Four Food Groups Can Save Your Life". The old four food groups were meat, dairy products, grains and vegetables/fruits. The thesis of the book was that we would do much better from the standpoint of health if we based our diets on four new food groups, those being grains, legumes, vegetables and fruits. Anything else in the diet was an option. Options would include meat, dairy products, added vegetable oils, alcohol, coffee, tea and other sorts of things. Meat products, poultry, fish and diary products as well as any substantial quantity of vegetable oil were to be avoided. The book was endorsed by Dr Dean Ornish whose research Dr Barnard relied on and by Dr Colin Campbell who was another defence witness.

The book contained the statement: "All fish contain cholesterol and fat including saturated fat". It also said: "... the fish oils that were once in vogue have been found to actually encourage the production of cancer-causing free-radicals". Dr Barnard agreed that there was a body of evidence which held that the consumption of n-3 or omega-3 fatty acids in fish oils actually inhibited the promotion of malignant tumours. Dr Barnard did not see that as part of the scope of his book.

The ethical aspects of the treatment of animals were important to Dr Barnard as well his concern for human health. He also had concerns about the environmental consequences of animal agriculture.

Dr Barnard was clearly a committed campaigner for his view that we would all be healthier if we were vegetarian or vegan. He gave more detailed evidence on what he saw as a more healthy vegetarian or vegan diet but I do not propose to summarise it here. The parties accepted that I was concerned with the risks of McDonald's largely meat-based, food, rather than any benefits of a vegetarian or vegan diet, and I refused to let the Plaintiff's counsel adduce evidence about the possible risks of a vegetarian or vegan diet which was not carefully controlled and supervised.

Dr Barnard was uncompromising in his opinions on the risks of eating McDonald's food. He ended his first witness statement, typed on Physicians Committee for Responsible Medicine note paper,

Page 278

by quoting another physician who had conducted a study of heart disease, to the effect that: "When you see the golden arches, you're probably on the road to the pearly gates".

Dr Barnard was generally rather dismissive of any article which stood in the way of his firmly held views. In his view studies which found no positive correlation between a high or higher fat diet and cancer of the breast or colon, or which found a negative correlation, were really of no significance since the preponderance of investigation was for a positive correlation.

As I have already said, he thought that positive findings were always more significant than negative ones. He said this on more than one occasion, using the illustration that if you sent out twenty search parties, it did not matter that fifteen found nothing if five found what they were looking for. In the case of diet and disease the proportions of successful searchers were the other way round.

He used this illustration to meet a follow up survey of 5,485 women in the U.S., by Jones and others in 1987, which ended:

    "This study is an important addition to the examination of diet and breast cancer because of its prospective nature in which careful dietary assessment was carried out prior to ascertainment of breast cancer status. The .. survey that provided this cohort sampled from a large cross-section of the U.S. population, and thus these data do not focus primarily on the upper-socioeconomic-class women at higher risk of breast cancer who are usually the focus of breast cancer studies. These data are not consistent with the hypothesis that high dietary fat intake increases breast cancer risk. Indeed, they suggest a possible protective effect of high fat intake, but this result may be influenced by methodologic problems with the dietary assessment. These results certainly indicate the need for further exploration."

Dr Barnard did not see fit to place this paper among the very large number of references at the end of his statements.

Dr Barnard relied upon Dr Willett's survey of a large number of American nurses, which associated a high fat diet with colon cancer and he used it for the proposed importance of low vitamin A intake. But the fact that Dr Willett had found a negative correlation between a higher fat diet and breast cancer in his prospective cohort study of

Page 279

over 89,000 women nurses troubled him not. The reason for that in Dr Barnard's mind was that the range of fat intake was quite narrow and the level quite high, which is an understandable point. However, Dr Barnard did say that it was reasonable for Dr Willett to answer that criticism, by saying that it might not be valid because the same dietary assessment methods were adequate to detect positive association of fat intake with colon cancer in the nurses' health study.

Dr Barnard was referred to a paragraph in a March,1990, review by Dr Willett of the evidence so far. The paragraph read:

    Dietary fat and breast cancer. For the last decade, the dominant aetiological hypothesis for breast cancer has been that high fat intake, and animal fat in particular, is the primary cause of the large differences in rates between countries. This notion is based largely on two lines of evidence: striking international correlations (approximately 0.80) between per capita consumption of fat and breast cancer rates: and animal experiments in which diets high in fat increase the occurrence of mammary tumours. Subsequent evidence has not, overall, supported this hypothesis."

When asked for his comment on that statement, Dr Barnard dismissed it by saying: "Well this is a common sort of thing that Walter Willett would say".

On the other hand Dr Barnard did approve of the first part of the introductory statement to the article, saying: "Epidemiological studies of breast and colon cancers implicate diet as a causative factor ......".

In fact the sentence continued: "...... but the evidence is stronger for colon cancer, the occurrence of which may be reduced by diets with less animal fat and more fruit and vegetables".

One of the reasons for Dr Barnard's choice of the references which he particularly wished to draw to the court's attention was that, in his view, the causal links between a high fat, diet and heart disease and breast cancer and a high fat, low fibre diet and colon cancer were well established by 1991, indeed well before that.

Dr Barnard said that evidence supporting the connections

Page 280

between cancer and high fat foods, particularly meats, continued to grow, although he said that this was particularly so with prostate cancer with which I am not concerned.

Dr Barnard would not accept that experts had became less confident of a causal relationship between fat intake and the occurrence of cancer since the Committee on Diet, Nutritional, and Cancer produced its report in 1982.

Having chosen to rely on parts of the Surgeon General's report of 1988, Dr Barnard was critical of his failure, as he saw it to consider all the evidence and he was critical of him for not going far enough. Dr Barnard was not suggesting that the Surgeon General was entirely controlled by the food industry, but he regretted to say that he was certainly influenced by food producers and retailers.

Prof. Crawford also gave evidence on the relationship between diet and degenerative disease. He said that it was important to state that with the exception of mathematics there was no science which could claim to prove a hypothesis. Science operated on the basis of what is most likely to be true on the basis of the best available evidence. It was always therefore legitimate in biology to claim that there was no proof that A caused B. Moreover it would be unethical to seek to prove that dietary fat caused cancers by asking people to allow their children to be members of a test group in a trial in which the test group was to be fed, for twenty years, a diet of the highest fat and highest saturated fat content found today and then as adults, compared with children fed a diet of the lowest fat and saturated fat content. It would be unethical because there was a risk that you would do the children on the high saturated fat diet serious injury in one way or another.

Despite this difficulty Prof. Crawford said that there was unanimous agreement, worldwide from over sixty expert committees, that cardiovascular disease was closely and causally related to a diet high in saturated fats. Whilst it was accepted that other factors, such as a low intake of fresh fruit and vegetables, might be involved, the hard animal and human evidence which had stood the test of time since 1953, referred to saturated fats, typically derived from

Page 281

ruminants and hydrogenation of natural oils. There was also increasing evidence that a high intake of fat and a low intake of dietary fibre were associated with breast cancer and colon cancer. Although the same, hard experimental evidence was not so far available for those cancers, the epidemiological evidence linking western diets and those cancers was compelling.

Dealing with cancer, Prof. Crawford said that cancer was initiated by a special type of mistake in the genetic code which could be caused by radiation, virus, physical interference or a chemical component from the diet, air or surface contact. The dietary element included peroxidative damage resulting from inadequate defence of the genetic material from oxidative attack.

Once initiated, the tumour must proliferate and gradually loose the cellular markers which enabled the immune system to identify the cell as foreign. It was thought that this stage of tumour development had to be under the influence of some promoting factor or the immune system had to be compromised, otherwise the tumour cells would be destroyed by the immune system. In a third stage, tumour cells broke off from the primary tumour mass either spontaneously or as a result of physical interference and then migrated in the circulation. During this migration they had to be invisible to the immune system surveillance and find a weak spot in the endothelium lining of the blood vessels so that they could implant and develop a new tumour at a new site. Several new tumours might be seeded in this way.

The threat to life was not usually the prime tumour which could theoretically be cut out bringing an end to the problem. The real threat was usually the promotion and proliferation of cancer cells, and the important question was what was responsible for the promotion and proliferation. There was firm experimental evidence that dietary fats acted as promoters.

Prof Crawford referred to Sir Richard Doll's original thesis in the early 1950s, noting the disparity in cancer incidence from country to country and commenting that the reason was likely to be the different diets from country to country. Sir Richard's conclusion was that some 30 to 70% of cancers were related to diet, figures which were still relied upon today.

Prof. Crawford said that in the late 1970s and early 1980s the relationship between diet and cancer really became an issue and since then epidemiology had offered evidence of a very tight relationship between the 'disappearance ' of fat per caput, which reflected fat consumption in different populations, and risk of

Page 282

breast cancer, colon and prostrate cancer. Work by Carroll, in which Prof.Crawford was involved, related incidence of breast cancer in various countries to total dietary fat intake and animal fat intake, and the relationship applied in a very similar way to heart disease and cancer of the colon. The relationship did not apply to vegetal fat intake so far as breast cancer or heart disease was concerned.

Prof. Crawford referred to some of the work upon which Dr Barnard relied, for instance the increase in the incidence of heart disease, and breast, colon and prostate cancer among Japanese who had emigrated to the U.S., changing their diet when they did so, while incidence among their families who stayed in Japan remained low. He referred to Hirayama's study within Japan itself where the first cow was not slaughtered consumption until 1931.

Prof. Crawford referred to Dr Willett's studies. His summary of them was that studies by Willett and others on groups of individuals provided no evidence for a link between saturated fats and breast cancer in contrast to the epidemiology. On the other hand studies on individuals were suggestive of a link between dietary fats and colon cancer. (Willett et al., 1990). The results of such studies were not consistent. Some claimed a link with dietary fat and cancer and other claimed there was no link. Willett et al (1992) commented that while their data on individuals provided no evidence for a link between dietary fat and breast cancer, "Nevertheless, the positive association between intake of animal fat and risk of colon cancer observed in many studies provides ample reason to limit this source of energy."

Prof. Crawford said that there had been criticism of the country to country data on diet and cancer on the grounds that the data in many of the developing countries was unreliable. There might be some truth in this comment, but it needed to stated that the data had been collected by the World Health Organisation who had representatives working in such countries and had first hand knowledge of local health which would flag any serious discrepancy in reporting, in Prof. Crawford's view.

There was inconsistency in the studies of individuals. A lot of the inconsistency was due to various factors.

There was a similar situation in the early studies on heart disease and diet. The country

Page 283

to country studies provided clear evidence whereas those on individuals introduced confusion. Two people could have the same blood cholesterol for example but only one would have the heart attack. This was not surprising for two reasons.

Firstly, the time at which such studies were done was almost certainly very far distant from the time when the cancer was initiated and promoted to the point of a malignant tumour. This meant that this type of evidence was unreliable as you were probably depending on the study of someone today telling you about their conditions thirty years previously. Cancer of the breast, for instance, might require 20 to 30 years of cell doubling time (promotion) before you could feel a palpable tumour. In addition, the tumour itself had its effect on metabolism and behaviour which could distort the information obtained.

Secondly, humans were polygenetic. That is to say, with the exception of identical twins, no two people had the same genetic make up, therefore one person would be susceptible to disease A but not B, whereas another would be susceptible to B and not A. This mattered in the case of studies of relatively small numbers of individuals.

Thirdly, with studies between individuals the dietary differences were relatively small by comparison with immigration studies or studies from country to country. Studies on individuals had value, but in Prof. Crawford's opinion they are not as powerful as the country to country studies where one was dealing with whole populations of people.

So Prof. Crawford did not give studies on individuals the same weight as population studies.

Although there had been intervention trials in relation to heart disease which consistently showed that if you reduced blood cholesterol levels you lowered mortality, there had not (when Prof. Crawford first gave evidence in 1994) been intervention studies in cancer. The studies then being planned were long term follow up studies of cohort populations which had already chosen their diets, not intervention studies which would involve urging some of the sample to keep taking, or to take more of what was bad for them.

Page 284

Prof. Crawford did not personally favour the view that mutagens and other potentially cancer producing agents which were produced when you cooked meat had a role in colon cancer.

He accepted that one had to be acutely conscious of the defects of extrapolating from animals to humans.

Prof. Crawford noted the advice given by various organisations.

He referred to the U.S. Surgeon General's report.

In the interests of preventing cancer, the National Cancer Institute (USA) published (1989) dietary guidelines to reduce fat intake to 30% or less of the dietary calories, to increase fibre intake to 20-30g/day, increase the use of vegetables and fruit and minimize the consumption of salt cured, salt pickled or smoked foods.

More recently, the British Nutrition Foundation's Task Force on Unsaturated Fatty Acids (1992) concluded that: "On the basis of the evidence available from the experimental and epidemiological studies, it appears that a reduction in total fat intake may decrease the risk of developing cancer. It would also seem prudent to modify the intake of n-6 and n-3 PUFA in favour of the latter". Prof. Crawford was a member of the committee and he agreed with the conclusion although he would have put it slightly more strongly as did the Report (December 1993) of the Working Party to the Chief Medical Officer for Scotland on 'Scotland's Health'. That report commented that "The adoption of the prudent diet .... should reduce the cancer burden but it would be unrealistic to expect tangible evidence for perhaps 20 years".

The report also concluded that "evidence indicates that a diet that is low in total and saturated fat, high in plant foods ... and low in alcohol, salt-pickled, smoked and salt preserved foods is consistent with a low risk of many of the current, major cancers, including cancer of the colon, prostrate, breast, stomach, lung and oesophagus". So the Scottish report used "should" instead of the "may" in the BNF report.

Prof. Crawford valued such reports because they were produced by groups which discussed matters and tried to formulate words which were agreed across a body of expert opinion. So they

Page 285

were unbiased "whereas, with the best will in the world, individuals, I dare say including myself, will have some form of personal bias both in terms of the evidence they have individually seen through their life, and maybe their own feelings about things that really have nothing to do with the evidence".

Prof. Crawford was not, however, happy with the conclusion at paragraph 3.5.6 of COMA 41 report in 1991, that

"The Panel concluded that there is currently insufficient evidence on which to base a recommendation for a decrease in fat intakes to prevent cancer, although an increase in the consumption of fatty acid should not be encouraged. The Panel agreed that the Dietary Reference Values based on other considerations ...... were consistent with a prudent view of the current data relating dietary fat and the occurrence of cancer".

COMA reports tended to be a little behind the times, he said, because of the political implications which some recommendations generated, and in any event the purpose of that particular panel was to report on Dietary Reference Values, i.e. the amounts of energy and nutrients required in diet, not to discuss diet and cardiovascular disease or cancer.

Prof. Crawford was careful to note the roles of different fatty acids. There were two families of essential fatty acids: the n-6, or omega 6, and n-3, or omega 3, fatty acids. The n-6 fatty acids (PUFA). The n-3 fatty acids started with linoleic acid and were converted into arachidonic acid. It was a polyunsaturated fatty acids started with alphalinoleic acid and were converted to docosahexaenoic acid (DHA). Both were essential for human life. The former were required for the growth of a tumour. There was evidence that the latter, the n-3 fatty acids, typically found in fish and seafoods, had a protective effect for heart disease and cancers like cancers of the breast and colon. Meat products were very poor sources of n-3 fatty acids, particularly the DHA. On the other hand saturated fatty acids were not essential. This took Prof. Crawford to a theory which clearly enthused him; that the style of food, based on modern beef, high fat intake, was not "a natural way of eating for Homo Sapiens". Prof. Crawford expressed it in this way:

"Homo sapiens evolved over a 5 million year period.

Page 286

It is only 150 generations ago that our forebears emerged from the Stone Age in England and only 5 generations have passed since the Industrial Revolution which is when the major changes in food occurred and large scale food processing was introduced. We can be sure that these time scales are insufficient for any Darwinian change in the genetic structure of Homo sapiens to have taken place. Therefore humans today are physiologically wild and their physiology is adapted to wild foods, 'to the unsophisticated foods of Nature' ...... Meat from wild animals is low in fat and low in the saturated fat associated with contemporary animals but was (and still is) largely PUFA. The reason is simple. The PUFA are essential and used in tissue cell membranes hence are well represented in wild animals living on natural foods and enjoying ample exercise. The modern beef animal is reared for fast weight gain and is a high fat product rich in saturated fats.......

The result is today, that the modern intensively reared beef animal will be expected to have in excess of a 25% carcass fat with a 50% lean. The lean meat will also have a very high fat content, especially if it is marbled which is today a sign of excellence but in reality is the result in fat infiltration into muscle which should be considered as pathological. Generously assuming the lean is muscle it will be one fifth protein and four fifths water. Converting that to calories we multiply the 10 of protein by 4.5 which equals 45 calory units. If we do the same for fat we have to multiply the 25 by 9 to convert to calories which gives us a comparative calorific value of 225. This means that in terms of dietary energy the animal .... is providing us with 5 times the amount of dietary energy as fat compared to protein.

In the wild animal the situation was reversed. ...... the lean is about 75% associated with a carcass fat of 2-5%. Using the upper figure for fat, there is an energy value from protein of 67.5 and 45 from fat: which means more protein than fat.

The above sums illustrate the extent to which modern beef production has become distorted from the wild nature of food to which we are physiologically adapted. This distortion makes the animal experiments with high fat diets in relation to cancer promotion very relevant to Western type diets and particularly to those encouraged by McDonald's".

The human physiological adaptation to the contemporary foods was being expressed in new disease patterns.

Page 287

On the other hand fish were essentially wild animals. Asked how he would react to a suggestion that humans should disavow the consumption of fish or fish oils because the omega 3 or n-3 fatty acids to be found in them were unstable and apt to release free radicals which were dangerously carcinogenic (as propounded by Dr Barnard), Prof. Crawford said that it was not entirely bunk because one of the theories of the origin of carcinogenesis was the peroxidated theory that the DNA was attacked by free radicals and damage results in the particular twist in the molecule which results in a tumour. Experiments with marmosets had shown that they could get alopecia anaemia as a consequence of a diet high in omega 3 fatty acids. But the epidemiological and experimental evidence was that the long-chain omega 3 fatty acids that were found in fish and seafoods were protective. Prof. Crawford's reaction to the suggestion concerning dangerously carcinogenic free radicals was that he "would be surprised if that statement was made by somebody with a deep knowledge of the subject". Someone who said "avoid fish oils because they will give you cancer", "would not be talking from a good base of knowledge".

Studies like Carroll's which showed that in general countries which had high mortality from heart disease also had high mortality from breast and colon cancer, led Prof. Crawford to ask what the common denominator was, and he proposed a theory which he expressed as follows:

"If there was no connection between heart disease, colon and breast cancer it would be expected that the incidence of one disease would be randomly associated with another. That is a country with a low incidence of heart disease might have a high incidence of breast cancer. However, the distribution is not random. We consistently find that those countries with a high incidence of heart disease also have a high death rate from breast and colon cancer and vice versa.

This relationship prompts the question: what is the common denominator?

Science has not adequately addressed this question but there is a body of evidence which, if viewed laterally, provides an answer in the health of the vascular endothelium (cells that line the blood vessels).

The cause of death from heart disease is a thrombus or blood clot which blocks the flow of blood in a vital vessel feeding a part of the heart. The endothelial cells produce a substance called prostacyclin which prevents the platelets (blood clotting elements) from sticking to the vessel wall. If the endothelium is damaged the platelets adhere

Page 288

to the wall. Repair should follow underneath the cover of a mini-clot. However, persistent damage as a consequence of an unhealthy endothelium and high blood pressure due to a bad diet, toxic agents of infection, may result in the clot growing and the development of atherosclerosis. This is a simplistic view of the process which can in the end, lead to death. It is the metastases of the cancer which usually cause the real problem. The cancer cell migrating in the circulation will find it easier to implant with an unhealthy endothelium. The experimental evidence that a high fat diet promotes tumour numbers and growth parallels the evidence that a high fat diet can also promote atherosclerosis and supports this view of the common denominator between heart disease and cancer.

In considering the parallels it needs to be pointed out that the evidence on diet and cancer is less well formulated than the evidence between diet and heart disease. The data on cancer approximates, in its content, to the date on heart disease available in the 1970s. None the less, if lines are drawn laterally between the parallel evidences of heart disease and cancer, there is in my view enough evidence to make it difficult to reject the hypothesis that a high fat and high meat fat diet is similarly related to the risks of breast and colon cancer".

Prof. Crawford said that the theory of the common demoninator came to him in a flash. He had thought about it before he made his witness statement but he had never expressed it in writing before he wrote his statement. He did not know of any research group that had thought of it or had looked at the causes of cancers and heart disease side by side although several cancer research groups looked at the cause of metasteses and they would know about artherosclerosis and the unhealthy endothelium

Between giving evidence in October,1994, and returning to the witness box in June,1996, Prof. Crawford had raised his common denominator theory at two conferences. No one had shot him to bits on it. He thought that I was not necessarily right in thinking that if it had merit, researchers would immediately be keen to pursue it. They worked at the molecular biological level.

To summarise his view of the science relating to diet, heart disease and cancer, Prof. Crawford prepared a table of "the links between the heart disease and cancer parallels with dietary fat",

Page 289

as follows:

Type of Evidence				Coronary Heart Disease			Cancer				

1. Epidemiology Strong Strong

2. Migrant Data Strong Strong

3. Experiments Strong Strong

4. Data on Individuals Moderate Weak

5. Intervention Trials Moderate Not Done.

"Strong" meant an association better than 75%. "Moderate" meant between 45% and 75%. "Weak" meant between 25% and 45%. The subdivisions were "arbitrary".

Prof. Crawford's conclusion in his first witness statements, adopted in his oral evidence in October, 1994, were:

"If you accept the link between heart disease, high fat diets and particularly high saturated fat diets, then it is difficult to escape the conclusion that a similar link exists between cancer and foods rich in fat, particularly saturated fats and poor in n-3 fatty acids particularly DHA and the appropriate anti-oxidants. Several National and International organisations are in agreement that a reduction in total dietary fat and particularly saturated fat would be expected to reduce the burden of Western cancers. It would also be expected to reduce the incidence of cardiovascular disease".

What Prof. Crawford described as "the modern western diet" did not in his view cause cancer intrinsically in the sense of initiating it, although others held the view that it did; but once you had the genetic susceptibility to it in place and the cancer was initiated your diet mattered a very great deal and you could say that it was causative of the promotion of cancer. The evidence was stronger for promotion than for intrinsic cause but if a cancer was not promoted you would not die from it and that was what mattered. Hence the recommendations to reduce total fat and, in the case of the recent Scottish Report, particularly saturated fit.

Dealing with the question of risk, Prof. Crawford's concern was that the highest incidence of heart disease and cancer was in the lower socio-economic groups. In his studies with others in the East End (Doyle et al 1982), about 30% of the calorific value of food eaten by pregnant mothers came from take-away sources. This meant that McDonald's type of food was being consumed in a disproportionately high amount by those at greatest risk to the diseases of interest to the Court.

Page 290

Furthermore, the diets of school children studied in the East End suggested that a high proportion of their food was coming from convenience foods (Doyle et al 1994). The point here was that the groups at greatest risk seemed to be those most susceptible to media-type persuasion. McDonald's were encouraging the use of a style of food which was closely associated with risk of cancer and heart disease while health professionals were trying to reduce the risks.

Prof. Crawford also had a point about promoting the same kind of food in cultures where heart disease and cancers of the breast and bowel were not at present a problem, but that was rather off the ground with which I am presently concerned.

When Professor Crawford returned to the witness box in June, 1996, he said that he thought that the evidence relating diet to heart disease and cancer had hardened. Because of the exhaustive nature of the work on saturated fats, recent studies had focused on other aspects of diet such as protective agents.

There had been increasing evidence on anti-oxidants in relation to protection against blood vessel disease. This evidence had led several reporters to make strong claims for lack of dietary anti-oxidants as a cause of heart disease. Those claims need to be seen in their context of food. That is, the food sources that provided the richest sources of polyunsaturated fatty acids known to protect against heart disease were the richest in anti-oxidants.

Those protective foods were, ipso facto, low in saturated fats. The richest sources of saturated fats were the fats from ruminant, land animals. Conversely they were the poorest sources of both polyunsaturated fats and anti-oxidants. Polyunsaturated fatty acids were used to build cell membranes and for their integrity. Saturated fats and trans isomers (trans fatty acids) could interfere with this process. A major function of the anti-oxidants was to protect the polyunsaturated fats in the cell membranes. Lack of integrity of the cells lining the circulatory system encouraged the adhesion of blood cells to the walls.

The primary source of anti-oxidants was in the green leaves of all plants. That is where vitamin C, vitamin E, betacarotene and alpha-Linolenic acid, which was the parent of the omega

Page 291

3 essential fatty acids, one of the polyunsaturated fatty acids were found. Generally speaking, the seeds also contained vitamin E along with Linolenic acid. Some legumes contained alpha-Linolenic acid. Tubers could be quite rich in ascorbic acid, vitamin C.

The general recommendations which were being made were that there was a need to increase intake of fresh fruit and vegetables because of the value of anti-oxidants and the need of fibre, and because diets high in fruit and vegetables were low in the highly saturated fats and trans isomers which were perceived before as being causally related to heart disease in western types of cancer.

In Prof. Crawford's view the evidence on anti-oxidants, whilst broadening understanding of the pathogenesis of vascular disease, also strengthened the view of saturated fats as causal. The new evidence on anti-oxidants placed the discussion into the context of foods and meals rather than the ingredients.

The type of meal that McDonald's served in the 1980s was rich in the atherogenic aspects and poor in the protective aspects of the diet.

Prof. Crawford said that the work on cancer still lagged behind that of heart disease but the experimental evidence was similarly implicating fats.

Prof. Crawford stood by his previous statement which offered an explanation for the coincidence of heart disease and cancer in the same populations, namely that cancer and heart disease had the common principle of adhesion to the walls of the circulatory systems which in the one case caused thrombosis and in the other the spread of cancer. Adhesion was known to be enhanced by saturated fats.

At the First International Conference on Fat and Oil Consumption in Health and Disease, in Washington in 1995, experimental evidence was discussed at the conference, showing that omega 3 [otherwise n-3] polyunsaturated fatty acids protected against metastasis of experimental breast cancer. They also protected against coronary thrombosis.

Prof. Crawford tried to help quantify the risk of heart

Page 292

disease from eating McDonald's food.

He did a calculation on a British Heart Foundation figure of 169,514 deaths from heart disease in the U.K. in 1990, and research commissioned by the Second Plaintiff (Taylor Nelson Eating Out Monitor) which Prof. Crawford took to show that 1,905,000 people ate at McDonald's in the U.K. once a week, and 425,000 several times a week or more often, and studies by Ancel keys supported by experimental data, to the effect that dietary saturated fats had a 70% role to play in heart disease which was a multifactorial process.

He also said that the incidence of chronic illness like heart disease and cancer in the lower socio-economic groups in the U.K. was 43% which was twice that of the upper groups. One in three in the lower socio-economic groups would have a heart attack or stroke before the age of retirement. His own data from the East End of London was that that high risk sector could obtain up to 30% of their dietary energy from "take-away" sources like McDonald's and that was likely to apply to other inner city areas in the U.K.

From these figures Prof. Crawford's calculation went as follows:

"Of the 1,905,000 eating once weekly at McDonald's that meal is likely to contribute a 7th of the week's saturated fat intake and a minimum of a 14th. In the lower socio-economic groups one in three men will have a heart attack or stroke before they reach the age of retirement. To get an idea of the contribution to risk one can use the factorial calculation of the proportions which gives at 70% of risk a figure of 63,500 heart attacks and 31,750 at a minimum. If saturated fats and trans-isomers only contribute to 35% of the risk the figures are 31,750 and 15,875 respectively. These figures pertain to those of the 1,905,000 who eat once a week at McDonald's before they reach retirement, age and are only intended as a guide to assess the contribution to the sum total of an individual's saturated fat load".

Prof. Crawford said that the one seventh of the week's saturated fat intake presumed eating one meal a day, which was "saturated fat". The one fourteenth presumed two a day, which were "full of saturated fats".

So, taking the assumption of one saturated fat meal a day,

Page 293

of which one a week was McDonald's, Prof. Crawford had divided 1,905,000 by 7 before taking 70% for dietary responsibility. He then divided again by 3 for the one third of those in the lower socio-economic groups who had heart attacks. This gave him the figure of 63,500. An assumption of two saturated fat meals a day, of which one a week was McDonald's, gave 31,750.

Prof. Crawford was not saying that a contribution of one seventh of a week's saturated fat intake would kill 63,500 of McDonald's 1.9 million weekly users. He was saying that if McDonald's meals contributed one seventh of their saturated fat load regularly throughout the relevant part of their lives, which was from childhood through to middle age, "it might contribute very strongly to their heart attack or their stroke".

Prof. Crawford's calculations heaped assumption upon assumption and had no logic to them. Prof. Crawford did say that a vast amount of money and several years research investigating relevant risk factors would have to be spent to calculate with any precision the risk of heart disease from eating McDonald's saturated fat meals.

Returning to his common-demoninator theory, Prof. Crawford accepted that in order to metastasise, cancer cells had to break off from the tumour. They could then spread through the body either in the bloodstream or through the lymphatic system. Migrating tumour cells had adhesion molecules, called integrins, which seemed to assist in the process of implanting the tumour cell in endothelium. Prof. Crawford did not have any evidence that the presence of the integrins or their ability to attach themselves to the endothelium had any relationship with a high fat diet or high serum cholesterol. Moreover the major sites for metastasised tumours were the liver, the lungs and bone, and one did not find atheromas in those areas of the body except in patients who had essential hypertension.

I have already mentioned Dr Colin Campbell, or rather Prof. Campbell as he is Jacob Gould Schurman Professor of Nutritional Biochemistry at Cornell University.

For nearly forty years Prof. Campbell's principal scientific interest has been the effects of nutrition status on long term health and particularly on the causation of cancer. He has carried out original research and he has been a member of various U.S. and international advisory committees and panels.

Page 294

Prof. Campbell was a member of the committee responsible for the book "Diet, Nutrition, and Cancer" published in 1982, to which Dr Barnard referred. He is presently co-chairman of an international panel of scientific experts who have been researching dietary and nutritional policies concerning the dietary prevention of cancer worldwide.

Since 1983, Prof. Campbell has been Director of a multinational responsible for nationwide surveys of diet, lifestyle and mortality in rural China. Professor Crawford has worked on the project, as a world authority on the analysis of blood and urine samples and dietary fat and fatty acids in particular.

Prof. Campbell's summary of the project's findings was set out in a letter dated July,1994, to Mr Peter Cox, another defence witness, in the following terms:

    "The range of fat intake in the breast cancer analysis was 6-24% of calories. This is a particularly interesting range because it is below what most individuals in Western countries use, thus it considers what diet-disease relationships may exist when diets low in fat, and more importantly, high in plant matter are consumed. We found a positive statistically significant association between fat intake and breast cancer mortality rates. In short, this (and several other analyses of these data) suggest that, even though breast cancer is much less common and fat intake is much lower in rural China, the higher the fat intake (in China), the higher the breast cancer rates (in China). Of course, I am certain that the higher breast cancer rates are not due solely to dietary fat, but to a broad constellation of dietary factors associated with these levels of dietary fat."

The fat intake average in rural China was almost 14.4% of calories compared with 37% in the U.S. The average daily fibre intake in China was 33g. In the U.S. it was 11g. If I understand Prof. Campbell's evidence and a 1994 article which he wrote with Dr Chen Junshi correctly, their China survey concerned itself with areas of rural China, excluding urban dwellers and herdspeople of north-western China.

Page 295

Prof.Campbell dealt with the suggestion, made by Prof. Willett among others, that his China findings were confounded and so deprived of weight or validity by "industrialisation" in that cancer rates were higher in more industrial parts of China itself. We wrote:

    "As to confounding by "industrialisation", this is a red herring with little meaning. In virtually all societies, as industrialisation becomes more significant (that is, when people have more disposable income), one of the first changes to occur is an increase in the intakes of total fat and foods of animal origin (generally simultaneously). I do not know of any significant evidence which suggests that industrialisation-related non-dietary factors can account for this dietary association, especially when there is outstanding empirical and theoretical evidence which shows that the association (of dietary fat and breast cancer) which we observe is biologically plausible (especially when taking into consideration the full constellation of dietary factors and mechanisms of diet and causation)."

Prof. Campbell said that by "industrialisation-related non dietary factors" he meant primarily pollution and stress, and also lack of exercise. The "outstanding empirical and theoretical evidence" included studies, particularly cross-national studies, which had shown an association between breast cancer as well as other cancers and fat and studies which addressed the biological plausibility of the way in which fat worked. "The full constellation of dietary factors and mechanisms of disease causation" referred to other factors present in or missing from certain kinds of food, and which also contributed to disease prevalence and disease causation. Examples were protein and caloric density or energy density to which dietary fat also contributed. Missing factors included hundreds of different kinds of anti-oxidants which were being discovered and hundreds of different kinds of dietary fibre factions which were generally missing from high fat diets.

In a further letter dated August, 1995, also incorporated in this evidence, Professor Campbell said that not only was the range of dietary fat in the China survey below most Western levels, the

Page 296

level of fibre was well above them, and the lower the dietary fat and the higher the dietary fibre, the lower the mortality rates for a wide range of cancers and cardiovascular diseases. His China investigation strongly supported the proposition that a high fat, low fibre diet was causal in the development of these diseases.

Professor Campbell's conclusion was that the findings suggested that even small additions of foods of animal origin to an otherwise all plant diet caused the occurrence of these diseases.

The meaning as I had ruled it to be by the time that Prof. Campbell gave evidence on 4th June,1996, was put to him. He had not analysed the specific nutritional content of McDonald's food, but assuming that typical McDonald's meals were high in fat, sugar and animal products and salt, (sodium), and low in fibre, vitamins and minerals, a diet of that nature would lead to a very real risk that one would suffer cancer of the breast or bowel or heart disease, as a result, in his opinion. Prof. Campbell held this view because of the large variety of mechanisms with biological plausibility and the relationship between the intakes of those foods and nutrients and the prevalence of those diseases in human populations to which he had referred in his letters. Prof. Campbell said that the comprehensive plausibility which was now being obtained together with the numerous associations which had been published many times over "clearly establishes a very, very strong and convincing association of causality. So, I would suggest it does, in fact, raise risk substantially, especially for people who are particularly vulnerable, for various and sundry reasons, to these kinds of diseases."

By people who were vulnerable, Prof. Campbell primarily meant people who were genetically susceptible but vulnerability was also, perhaps, related to other events and experiences in life. Individuals were presented with varying degrees of risk.

In the neighbourhood of 60 to 75% of premature deaths in countries such as the U.K. and the U.S. were diet related chronic diseases. Prof. Campbell considered that any deaths from these diseases occurring before the age of eighty-five or ninety were somewhat premature. Although some people would say seventy-five or eighty Prof. Campbell thought that there was a general consensus "that we ought to be able to enjoy a fairly healthy life without these diseases up until the ages of 80 to 85 or 90 or so, if in fact we did things properly."

Page 297

Prof. Campbell frequently referred to "these diseases" which I took to mean chronic, degenerative diseases, although he sometimes made it clear when talking of "these diseases" that he was, for instance, speaking of heart disease and cancer or heart disease alone.

Prof. Campbell said that colleagues were discovering that treatment with low fat diets was capable of reversing chronic disease, but there was no forewarning of this suggestion and it was not pursued further.

Asked about the effects of eating two or three "high fat/low fibre etcetera" meals in a week, Prof. Campbell referred to the China survey and said that "the closer we get to an all plant based diet, low in fat, high in fibre, the lower are these diseases across the board."

He would say that eating a high fat/low fibre etcetera meal twice a week "definitely raises risk, particularly of course for those individuals who are most vulnerable, as I pointed out before. So, there is definite risk there for doing that, to say nothing of the fact that using such food, even rather modestly like that, tends to encourage people to, if you will, consume still more as time passes"

Prof. Campbell did not expand on the extent of the raised risk of eating two such meals. Later in his evidence he described a diet that was high in fat, salt, and sugar and low in dietary fibre and also high in animal products, high in animal protein which went along with high animal foods in general, and low in antioxidant nutrients, as "a high risk diet."

Prof. Campbell described the China project as the most comprehensive project on diet and disease ever undertaken. The surveys were done in China because cancers and various other diseases exhibited exceptional geographic localisations so it made sense to examine those local regions to determine the responsible dietary and lifestyle factors. There had been two major surveys. In the first, in 1983, 367 items of information were collected from each of 20,000 people in 6,500 families. In the second in 1989 and 1990, about 700 items of information were collected from each of 40,000 people in 10,200 families. 1.2 million death certificates were examined. There was enough data to interpret for at least twenty years.

Page 298

Prof. Campbell's summary of the findings as he interpreted them so far was that dietary pattens in China were strikingly different from Western countries, the major difference being the consumption of foods of animal origin. Animal protein intake, for example, was tenfold greater on average in the U.S. than in China. Although the biology of the diet and disease relationship was infinitely complex and was easily misunderstood when interpreted in a reductionist manner, the main nutritional conclusion from the study was the finding that the greater the consumption of a variety of good quality plant-based foods, the lower the risk of those diseases which are commonly found in Western countries, e.g. cancers, cardiovascular diseases, diabetes. Based on these and other data, he hypothesized that 80 to 90 percent of all such diseases could be prevented before the age of 90 years. By a "reductionist manner" Prof. Campbell meant looking at individual components of food and individual chemicals and their so-called mechanisms of action, rather than trying to get a better feel of how all the individual items worked together.

The optimum lifetime blood cholesterol concentration might be as low as 100 to 125 mg per decilitre compared to an average concentration of 210mg/dL in the U.S. The same dietary factors which increased blood cholesterol concentrations among Americans (at much higher ranges) also increased cholesterol at the lower concentrations of Chinese. These included, for example, increased intakes of dietary fat and legumes. Moreover, the lower the blood cholesterol, the lower the risk of various cancers; there was no evidence of a cholesterol threshold below which further decreases in disease would not occur. Those two facts were quite remarkable, in that they suggested that almost any consumption of animal-based fat (higher in fat, lower in fibre) might increase blood cholesterol (among many other biochemical changes) from a very low level, to be followed by a significant increase in the prevalence of degenerative diseases. Many other analyses of those same data for individual diet-disease relationships support the interpretation.

The evidence, both in the U.S. where most studies were done with people whose cholesterol levels ranged between 160 to about 320 mg/dL, and in the China Project where the range of cholesterol was about 90mg/dL country average to a high of 170 mg/dL country average, was that there was quite a linear relationship between determined cholesterol and various cancers and heart disease. In the U.S. there was a linear relationship between cholesterol level and

Page 299

prevalence of those diseases. At lower levels in China, the higher the intake of fat and of animal protein in particular, in meat and diary products, and the lower the intake of fibre and legumes, the higher the cholesterol level and the diseases started to occur, from near nothing to something significant.

The Chinese consumed more total calories (per unit of body weight), yet they had far less obesity than Americans, probably accounted for both by greater physical activity and greater consumption of low fat, plant based diet. Breast cancer was low in China, but nonetheless greater with the consumption of the typical Western diet (high in animal-based foods and fat and low in plant-based foods) which encouraged body growth rates to be too rapid, and sexual maturation to occur too early. Prof. Campbell referred to the studies showing that earlier sexual maturation, in particular amongst young girls, which was associated with a high risk of breast cancer later in life. This was a very consistent finding. What we were now learning was that sexual maturation tended to occur earlier in some societies where young girls were pushed, with rice diets early in life, to grow as fast as they could. We tended to do that in Western societies.

Prof. Campbell said that the conclusions to be drawn from his and other studies suggested that it should be possible significantly to reduce diseases typically found either in Western industrialised countries or in developing countries. He had considerable confidence in this. He said that if all countries were to use the dietary and lifestyle recommendations emerging from this study, it was estimated that the reduction of disease burden would result in an enormous saving in health care costs.

He stood by a statement made in the 1994 article which he had written with Chen Junshi:

    "There appears to be no threshold of plant-food enrichment or minimisation of fat intake beyond which further disease prevention does not occur. These findings suggest that even small intakes of food of animal origin are associated with significant increases in plasma cholesterol concentrations, which are associated in turn with significant increases in chronic degenerative disease mortality rates."

Page 300

Professor Campbell said that the present dietary guidelines, mostly focused on the effects of dietary fat when reduced to 30 percent of calories, were not likely to yield the disease prevention benefits originally inferred. That figure had been based on estimates of what consumers might be willing to accept. The reduction should be greater to yield disease prevention benefits.

The fact that in certain Mediterranean countries high fat diets were associated with a low rate of heart disease and some cancers did not mean that a high fat diet was not so bad after all. The fat consumed was "basically plant juice......olive juice." Virgin olive oil included a lot of very good factors. Mono-unsaturates were present in olive oil. It was probably the best kind of oil. Those Mediterranean societies had a significant reduction of heart disease and cancer rates compared with northern Europe. But whereas they consumed plant material at a level similar to that of rural China, fat intake in rural China was lower and their heart disease and cancer rates were lower. So there was a hypothesis in that high levels of intake of olive oil, even, might have some adverse effects. Later Prof. Campbell said that there was a highly significant inverse relationship between mono-unstuatrate fatty acid and oleic acid intake (two components of olive oil) and coronary heart disease.

Prof. Campbell was critical of a paper "Cohort Studies of Fat Intake and the Risk of Breast Cancer - A Pooled Analysis" by Hunter and others including Prof. Willett, which had appeared only recently in the 8th February,1996, edition of The New England Journal of Medicine, which was relied upon by Dr S. J. Arnott, a consultant in Radiotherapy and Oncology, called by the Plaintiffs.

The thrust of the article is summarised in the Abstract as follows:

    "Background. Experiments in animals, international correlation comparisons, and case-control studies support an association between dietary fat intake and the incident of breast cancer. Most cohort studies do not corroborate the association, but they have been criticised for involving small numbers of cases, homogeneous fat intake, and measurement errors in estimates of fat intake. Methods. We identified seven prospective studies in

    Page 301

    fourcountries that met specific criteria and analysed the primary data in a standardised manner. Pooled estimates of the relation of fat intake to the risk of breast cancer were calculated, and data from study-specific validation studies were used to adjust the results of measurement error.

    Results. Information about 4980 cases from studies included 337,819 women was available. When women in the higher quintile of energy-adjusted total fat intake were compared with women in the lower quintile, the multivariate pooled relative risk of breast cancer was 1.05 (95 percent confidence interval, 0.94 to 1.16). Relative risks for saturated, monounsaturated, and polyunsaturated fat and for cholesterol, considered individually, were also close to unity. There was little overall association between the percentage of energy intake from fat and the risk of breast cancer, even among women whose energy intake from fat was less than 200 percent. Correcting for error in the measurements of nutrient intake did not materially alter these findings.

    Conclusions. We found no evidence of a positive association between total dietary fat intake and the risk of breast cancer. There was no reduction in risk even among women whose energy intake from fat was less than 20 percent of total energy intake. In the context of the Western lifestyle, lowering the total intake of fat in midlife is unlikely to reduce the risk of breast cancer substantially."

    The article itself concluded:

    "The possibility that aspects of diet during childhood or adolescence, including energy intake and total fat intake, may be associated with the risk of breast cancer decades later cannot be ruled out on the basis of the results of prospective studies of adult women. Nonetheless it appears unlikely that a reduction in total fat consumption by middle-aged and older women will substantially reduce their risk of breast cancer."

Prof. Campbell said that when assessing relative risks in searching for the effects of a specific factor such as dietary fat, investigators adjusted for so-called confounding variables such as age at menarche, menopausal status, parity, age at birth

Page 302

of first child, body-mass index, height, education , history of benign breast disease, history of breast cancer in a sister, oral contraceptive use, fibre intake, alcohol intake and energy intake; and in Prof. Campbell's view in so doing they were taking away some of the effects that would otherwise be expected. A number of the confounding variables could be diet-related, and removing them would tend to minimise the effect that was otherwise there so that it was not seen. Also dietary habits were particularly important for those who were genetically susceptible.

Prof. Campbell said that the conclusions of the study could be questioned on two grounds.

Firstly, although the range of fat intake went somewhat lower than had previously been examined, that is below 20% of calories, that lowest group included a fairly small number of individuals.

Secondly, the chief problem was that the examination focused specifically on just dietary fat. The subjects decreased dietary fat alone and kept the rest of their diet more or less the same. People who were asked to decrease their fat intake tended simply to decrease their use of added fat by using lean cuts of meat and low fat diary products but not really changing their intake of animal-based foods. So on theoretical grounds, Prof. Campbell would not expect to see a decrease in breast cancer. There was no evidence in the paper to show what the nutrient intake of the people on the so-called low-fat diets were.

The other recent report heavily relied upon by the Plaintiffs was a review of scientific evidence on diet and cancer prepared for the Food and Drink Federation by Dr M.J. Hill, Chairman of the European Cancer Prevention Organisation, published in that Organisations 1995 Journal and submitted to the COMA Working Group on Diet and Cancer, whose report is currently expected. The introduction to the report began with the statement that "there is a general consensus that diet is an important aetiological component in carcinogenesis, with some cancers showing an association with affluence, e.g. those of colon, breast, prostate, ovary and endometrium......." It went on to say: A common feature of dietary advice is that it is often presented as a 'single issue' matter. The introduction said:

"A common factor in our assessment of the perceived relationship between

Page 303

diet and cancer must be the variable and often poor quality of the data." The report reviewed the type of data, reviewed the data on the relationship between overall cancer risk and specific food groups or nutrients and finally summarised the data and drew conclusions with particular reference to the advice to be given to the COMA Working Group.

Having made an analysis of the currently available scientific literature on diet and cancer, the overall conclusion of Dr Hill's report was that "only two recommendations (to the public) might be justifiable in the light of current knowledge:

1.Maintain a healthy body weight: many cancers as well as many other diseases with a high mortality risk, are associated with overweight. Avoidance of obesity may decrease the risk of cancer at a number of sites and is unlikely to increase the risk of cancer at any site. Obesity can be avoided by increased exercise or by control of energy intake; more help should be given in encouraging increased exercise, especially in young people, since maintenance of a healthy body weight should ideally be started from an early age.

2.Eat plenty of raw and preserved fruit, salads and vegetables: there is evidence to suggest that these may be protective against a wide range of cancers and are not positively related to any, although much of the evidence is obtained from observational studies."

Prof. Campbell said that he had no problem with those two statements save that they were really much too conservative. Professor Hill, as Prof. Campbell thought him to be, had not taken into consideration a lot of things that were being discovered in the China Project and elsewhere.

    The final two paragraphs of the report read:

    "There are theoretical and practical reasons for suggesting that dietary advice such as this should be targeted particularly at the young.

    As the results emerge from current intervention studies, it is likely that it will be reasonable to advise the general public to eat decreased amounts of total fat and saturated fat, as is consistent with the specific U.K. Government Health of the Nation White Paper targets, and to eat greater amounts of cereals

    Page 304

    and complex carbohydrates. That point has probably already been reached for groups of people at high risk of cancer (such as first-degree relatives of patients with cancer of the breast, colorectum or endometrium) in whom the balance of risk and benefit is tilted towards benefit."

Prof. Campbell was tested on his conclusions. He accepted that when epidemiologists spoke of risk factors they meant factors of diet or lifestyle, which had been identified in the aetiology of the disease in question.

Risk factors for breast cancer included early age at menarche which was probably in large part diet related.

Age at first pregnancy was a risk factor: the earlier the better. It was probably not diet related.

Parity (number of children) was a risk factor up to a point: the more the better. It was not diet related save in so far as food balance might allow no more children.

Age at menopause was a risk factor: the earlier the better. It was diet related as was age at menarche. High fat, low fibre diets tended to lengthen the total reproductive period by causing an earlier age of onset of menarche in addition to changing menopause.

Breast feeding was a beneficial factor. That was not diet related for the mother.

Oral contraceptives created a very modest risk. They were, obviously not diet related.

Body mass was a risk factor and diet related.

Smoking was a risk factor for cancer. Less women smoked in China than the U.S.

Alcohol was a risk factor for cancer and was usually included in diet. It was a substantial conveyor of calories. Consumption was very low among women in China. In most places it was almost non-existent.

All these factors were taken into account.

Page 305

Lack of exercise was a risk factor for cancers to some modest extent, both for physiological effects and because it might lead to obesity. This was taken into account, but cycling to work in China was not taken into account, which Prof. Campbell accepted as a valid point.

Prof. Campbell accepted that one should look at individual risk factors and also at the whole picture. His 1994 article contained the paragraph:"......fat intake in rural China is exceptionally low when compared with Western experience. In addition to its correlation with plasma cholesterol, fat intake was also correlated weakly but significantly (P 0.05 meant that there was a less than 1 in 20 chance that the result arose by chance) with breast cancer. However, Chinese women, when compared with British women for example, exhibited later age at menarche, earlier age at menopause, greater parity, earlier age at first birth, and lower circulating concentrations of estrogen, all of which favour a reduction in breast cancer risk."

Prof. Campbell took the view that all those factors favoured a reduction in breast cancer, but age at menarche, early age of menopause and circulating levels of oestrogen were influenced by increasing levels of dietary fat. Greater parity and earlier age at first birth were social factors, but of less significance than the other three.

The article went on to say:

    "Intakes of 14 complex carbohydrate and fibre fractions were obtained in this study to determine whether particular fibre fractions were associated with particular diseases especially cancers of the large bowel. Average dietary fibre intake in China was about three times higher than average U.S. intake with one country mean being as high as 77g/d. So far we have prepared only a brief report of these data. However based on an overview of the univariate correlations, colon and rectal cancer mortality rates were consistently inversely correlated with all fibre and complex carbohydrate fractions except for pectin, which showed no correlation. These relationships although consistent, appeared rather weak because only thamnose-containing complex carbohydrate intakes reached statistical significance for cancer of the colon. (r-033 P <0.55). It appears then that within

    Page 306

    the range of 7-77g fibre/d, where mean intakes of 29 of the 65 countries were above the upper U.S. recommendation limit of 30-35g fibre/d there is evidence of a weak inverse relationship between cancer of the large bowel and the intake of multiple complex carbohydrate and dietary fibre fractions."

Prof. Campbell had no reason to revise the view expressed there. So, if a protective effect of eating plants was to be found it must be elsewhere in the various nutrients which made up the plant; but not exclusively because the inverse relationship between fibre intake and colon cancer (a beneficial effect) was significant although weak.

Prof. Campbell was asked about a 1994 paper "Diet and breast cancer in Shanghai and Tianjin, China" by Yuan and others whose data indicated "a strong protective effect against breast cancer development with intake of foods rich in fibre, vitamin C and carotene." The results were also "compatible with dietary fat having a modest positive effect on breast cancer within the range of exposure experienced by women in China."

Prof. Campbell had not found anything wrong with the methodology of the study or the statistics. The authors were qualified in pathology, molecular biology and clinical medicine, but none were nutritionists. The findings were more or less in line with Prof. Campbell's. The authors were saying that their findings were compatible with Hunter's and Willett's "null hypothesis"; but this was after adjustments for factors which ought not to be subtracted because they were affected by diet.

Prof. Campbell agreed with my suggestion that the watershed between his view and the view of some others was whether it was a valid exercise to adjust for "confounding factors" which, he said, were to some extent diet-related. Another distinction was whether one looked at fat levels alone in diet or, as Prof. Campbell said one should, all the other things going on which tended to characterise high and low fat diets, although fat itself also contributed to cancer of the breast.

There was a general consensus of opinion, Prof Campbell said, that fat in the diet was a cause of breast cancer but the strength of the association had been debated quite vigorously.

In Prof. Campbell's view the intake of fat had a causative effect for breast and colon cancer; and if you ate

Page 307

less fat and so ate more vegetables and got more energy from complicated carbohydrates this had a protective effect. There was what Prof.Campbell called an "aggregate effect." Prof. Willett had taken a position almost by himself in the field in saying that dietary fat was not causative of breast cancer.

In Prof. Campbell's view the 1991 COMA 41 conclusion that there was currently insufficient evidence on which to base a recommendation for a decrease in fat intake to prevent cancer and its adoption of a maximum of 35% of energy to be derived from dietary fat were far too timid. It was probably the most timid of any country he knew.

The statement in the 1990 WHO report that the evidence linking a high intake of total fat with the development of a number of cancers could not be considered to be sufficiently strong to be termed causal, merely meant that not everyone on the committee responsible for the report agreed that the evidence was sufficiently strong.

Prof. Campbell did say that reducing fat intake to about 30% of calories consumed was not going to reduce heart disease and cancers much. For substantial reductions, one had to go lower than 30%. Above 30% energy from fat the relationship between fat and disease appeared to cease to be linear. It was much less significant.

Professor Campbell ended his evidence by saying that there was a causal relationship between overall diet, of which fat was a part and to which it was a contributor, and breast cancer and colon cancer. The same was especially so with regard to animal products.

The chances of a high fat diet, including animal products, not being causal in cancer of the breast and cancer of the bowel were one in a thousand. The effect was that of the high intake of fat itself and the tendency to displace the consumption of foods of plant origin. That presumed genetic susceptibility. Prof. Campbell was absolutely convinced that we would find a cluster of genes for every simple disease known to human kind. But it was not the genes that actually established risk. It was the factors which allowed the expression of those genes which led to disease and those factors were largely dietary in origin. What led to disease was the long period between the genetic susceptibility and the disease itself, and that is where diet

Page 308

came into play. Everyone had the genetic susceptibility to varying degrees. There was a widely known 1980 study of Doll and Peto, analysing a large number of different studies which concluded that not more than 2 or 3% of all cancers were genetically based and a recent study of identical twins found that they did not get the same cancers. We all had a genetic propensity to develop cancer. Some people have a greater genetic propensity than others. But the overweaning influence was diet, and so far as cancers of the breast and colon were concerned the overweaning influence was diet high in fats of animal origin.

Ms Brophy, to who I have already referred , said that "a poor diet" which was low in fibre, fresh fruit and vegetables and high in fat and animal products had been scientifically linked with cancers, particularly bowel cancer and heart disease. She referred to Willett for the link between bowel cancer and meat consumption. She referred to a more recent,1990, Doll paper estimating that 30 to 70% of all cancers were linked to diet. The consensus today, approved by the Health Education Authority was that around 35% of cancers were linked to diet.

Ms Brophy went on to say that "in other words a change of diet would reduce cancers by that amount." Sugar promoted tooth decay and obesity, and eating too much salt could lead to high blood pressure which in turn caused heart disease and strokes.

In Ms Brophy's view there was a very real risk that you would suffer cancer of the breast or bowel or heart disease if your diet was high in fat, sugar, animal products and salt, and low in fibre, vitamins and minerals.

Towards the end of the evidence on the topic of nutrition the Defendants put in two Civil Evidence Act statements relating to high fat diet and heart disease.

The first statement was made by Dr Michael Miller M.D., the Director of the Centre for Prevention Cardiology at the University of Maryland School of Medicine. A study which he and colleagues had carried out involved putting twenty healthy men and women on a diet high in total saturated fat for a month. Fat provided about 40% of total caloric intake. Then the subjects were given specially concocted milkshakes, and blood fats (triglycerides), cholesterol and other chemistries were measured in blood samples taken over a period of ten hours. The same exercise was then carried out after the subjects had been on what

Page 309

Dr Miller described as a low fat diet (less than 30% of total caloric intake from fat). It was found that the body's ability to clear blood fat was significantly impaired after the high fat diet phase, so more fat remained in the bloodstream. A potentially important implication of persistent amounts of fat in the bloodstream was the it tended to be incorporated by scavenger cells known as macrophages. As they continued to engulf the blood fats, macrophages become larger and might eventually deposit on the lining of blood vessels including the coronary arteries which supplied blood to the heart. So Dr Miller concluded that persistent amounts of blood fat might accelerate hardening of the arteries (atherosclerosis) and lead to a heart attack, stroke or death. Food products that contained excess amounts of saturated fats might therefore be important contributors to this process. Such products included hamburgers, cheeseburgers, deep fried foods and ice cream, all of which were readily available in fast-food outlets such as McDonald's.

This sounds interesting work, but I note that Dr Miller, as a serious medical researcher, spoke in terms of possibilities so far as accelerating atherosclerosis was concerned, and he gave no estimate of the time or degree of acceleration involved.

The second statement was made by Prof. Robert A. Vogel M.D., Head of the Division of Cardiology at the same medical school. He and colleagues had carried out a study into the effects of eating a high fat (50 gram) McDonald's breakfast on endothelial functions which was thought to be the initial factor in the development of coronary heart disease.

The study was undertaken the better to elucidate the mechanism by which a high-fat diet led to the development of coronary and other arteriosclerosis. One clearly established mechanism was that high-fat meals increased the body's serum cholesterol. This led to the development of atherosclerosis. Prof. Vogel's data showed that an additional direct pathway existed as well. In contrast to a low-fat breakfast (cereal, skim milk), the high fat diet led to a 50% reduction in endothelial function as manifested by a reduction in high blood flow dilatability (flow-mediated vasoactivity) within two hours of eating. In addition to initiating the atherosclerotic process, this mechanism might also be responsible in part for the plaque rupture which

Page 309

cause myocardial infarction. His group had not obtained direct evidence for the latter, however. But their work did in his view confirm the clearly established relationship between a high-fat diet and coronary heart disease, which he described as "the number one source of mortality in westernised countries."

Prof. Vogel sent an abstract summarising his group's study. It ended with the sentence: "These data suggest high-fat meals transiently impair endothelial function independent of long-term effects on serum cholesterol."

Mr Geoffrey Cannon gave evidence in October, 1994, and June, 1996.

Mr Cannon is Director of Science of the World Cancer Research Fund. He is responsible for administrating the research grants which the Fund gives for laboratory and epidemiological work. He is director of the Fund's Diet and Cancer Project which brings together eminent scientists to report on the role of food and nutrition in the prevention of cancer.

Mr Cannon is also Chairman of the National Food Alliance which is the representative body for all national organisations in the U.K., outside government and industry, concerned with food and health policy and nutrition policy. He is a member of the Government's National Nutrition Task Force.

Mr Cannon is not a scientist himself. Although his degree included physiology he did not consider that relevant. He was put forward as an expert on the role of diet and cancer by virtue of his very wide and detailed reading on the subject and his discussions with experts and the very considerable thought which he had given to the subject. He described his expertise as being the translation of science into public policy.

Mr Cannon's evidence had three main prongs. Firstly, he said that there was an established consensus among the scientific community worldwide to the effect that the diet typically eaten in industrialised countries such as the U.K. and the U.S.A. was an important cause of major disease including cardiovascular disease and some cancers.

Secondly, Mr Cannon said that the consensus was that the U.K. diet was unhealthy largely because it contained too much fat, saturated fats, sugar, and salt and was correspondingly inadequate in bread and cereal foods, vegetables and fruit, and that all those aspects of the diet increased the

Page 311

risk of cardiovascular disease and some cancers.

Mr Cannon's own view accorded with what he saw as the consensus, largely guided by it.

Mr Cannon's evidence expanded on these points and his written and oral evidence incorporated references to a very large number of articles and reports. I do not propose to summarise Mr Cannon's evidence on those points here. It was based on the reports of others which I can read for myself and it reached conclusions on questions which I must decide for myself on the evidence of the scientific experts who were called on either side. Mr Cannon expressed the view that opinions of the individual scientists called by the Plaintiffs and Defendants to give evidence were not material to this case, except in as much as they explained the current state of scientific judgment: I should be looking at what Mr Cannon saw as the common agreement of current scientific judgment and finding accordingly. I cannot accept that. It would amount to an abrogation of my duty to make my own findings on the issues which I have set out on the basis of the evidence which I have heard and read.

I have reread all Mr Cannon's evidence, but what he said in the witness box amounted, very courteously, to substituting his own analytical judgment for mine, without having heard the witnesses whom I heard.

Prof. Wheelock was the first of the Plaintiffs' expert witnesses on the third issue. I will take his evidence relatively shortly because the Plaintiffs relied principally on Prof. Donald Naismith so far as coronary heart disease was concerned and Dr Sidney Arnott as far as cancers of the breast and bowel were concerned. Prof. Wheelock gave some evidence upon which the Defendants relied.

It was Prof. Wheelock's view that diet did play an important role in the development of "the so-called degenerative diseases which included coronary heart disease. The Plaintiffs admitted that there was a considerable amount of evidence of a causal relationship between a diet high in saturated fat, and sodium, and obesity, high blood pressure and heart disease; and at the time when Prof. Wheelock gave evidence in July, 1994, the Defendants took the view that the was the end of the contest so far as heart disease was concerned.

Page 312

Prof. Wheelock gave his own description of the main categories of research into the causes of disease. In his view epidemiological studies could provide associations between diet or characteristics of diet and the incidence of disease but they did not demonstrate cause and effect. Before one could sensibly postulate cause and effect one would ideally need to understand the mechanism which explained in detail the link between the diet and changes which occurred in the body and the final disease state. Experimental and clinical studies made a contribution to that end.

If all three kinds of studies, epidemiological, experimental and clinical, pointed in the same direction over a period of time one could probably, safely say that there was cause and effect, but one must not underestimate the very considerable difficulties in coming to definite conclusions.

Prof. Wheelock thought that there was a high level of probability of a causal relationship between intake of saturated fat and cardiovascular disease. It was above the level of proof at which an ordinary person ought to "watch it". But there was a low level of evidence associating intake of dietary cholesterol with health problems. The Americans were more concerned with that. It was thought that there was a relationship between the consumption of fat and a raised serum cholesterol (concentration in the blood). Some saturated fatty acids were known to increase the concentration of cholesterol on the blood. It was generally accepted that polyunsaturated fats and monounsaturates fats would reduce it. In general terms the higher the concentration of cholesterol in the blood, the greater the risk of developing atherosclerosis and of consequent death from heart disease.

He agreed with what was said about dietary fat in the Second Plaintiff's Good Food, Nutrition at McDonald's in March,1995, to the effect that: "Total fats, saturated fats and cholesterol in the diet are all thought to be linked to incidence of coronary heart disease, but the importance of dietary cholesterol has been largely discounted. The effects of the total amount of fats, particularly saturated fats, on increasing blood cholesterol is generally believed to be important. Hence, current advice is to reduce the total intake of fat in the diet but also to replace some of the saturated fats with unsaturated fats from vegetable oils and fatty fish."

He also agreed with the dietary advice given in the October, 1995, Good Food,

Page 313

Nutrition at McDonald's to which I have already referred.

He accepted that in populations where the proportion of saturated fat in the diet was high the risk of heart disease would be high, but when it came to cancer of the breast and cancer of the bowel the picture was more complicated

He actually went as far as to agree with Mr Morris's suggestion that he was "pretty certain.... that heart disease comes from high saturated fat content". He would "go along with that". But one had to consider whether a person smoked, the amount of exercise they took and, arguably, the amount of stress they were subjected to.

So far as cancer was concerned "we may be pretty sure that diet is a factor in the development of ...... the different cancers.... but being able to pin point precisely what it is in the diet which is responsible for that is another matter altogether."

Prof. Wheelock was sceptical of the proposed relationship between consumption of fat and cancer of the breast or cancer of the colon. There was some evidence suggestive of a relationship, but there was extreme difficulty in trying to establish a causal relationship of cause and effect and one would not be justified in concluding that there was a causal relationship. There was no reliable, credible, scientific literature which convincingly proposed such a relationship, in his view.

When papers or reports spoke of risk of disease from certain intakes, it was a statistical risk. Prof. Wheelock thought that the COMA 41 conclusions at para 3.5.6. that there was currently insufficient evidence on which to base a recommendation for a decrease in fat intakes to prevent cancer was an excellent summing up of the current position in July,1994.

Since the COMA report in 1991 things had moved on so far as salt was concerned. There was a growing awareness of the links between dietary sodium and hypertension and the incidence of strokes and more recent evidence was in favour of a causal relationship between them.

COMA 41 preferred to deal with dietary non-starch polysaccharides (NSPs)

Page 314

rather than "fibre" because is so far as there was any evidence to suggest benefit or protective agency for fibre in diet it appeared that the insoluble fibres were more effective than soluble. It was fair to say that NSP rich foods might be a barrier to intake of energy and useful in the prevention and treatment of obesity. Someone who ate little NSP might well be someone who ate a lot of fat.

So far as large bowel disease, including cancer of the large bowel or colon was concerned, he could not improve on the COMA 41 statement that : "It is not currently possible to identify NSP as a major dietary factor in the aetiology of these diseases. At best the aetiology may be ascribed to the consumption of a type of diet characterised by low starch, low NSP, high fat. Better epidemiological studies are needed."

Prof. Wheelock said that it was a fair line of reasoning that one of the proposed benefits of a diet relatively high in fibre was that increased bulk in the stool might remove bile acids in greater quantities and more quickly, because there was no doubt that if the amount of fibre in the diet was increased, then the transit time was reduced. The contents would be gone more quickly and that could possibly reduce any toxic materials more quickly before they did any damage. However, it would be difficult to say this had been established as a medical fact. There was convincing evidence for increasing foods that contained a fair amount of dietary fibre, such as fruit and vegetable and wholemeal bread and whole grained rice, but whether or not the benefits actually came from the increase in fibre or from other constituents present in those foods, such as the antioxidants that you got in fresh fruit and vegetables, was difficult at this stage (July,1994) to say.

Prof. Wheelock said that there was great difficulty in working out any relationship between dietary components and a disease such as cancer which could take twenty, thirty or forty years to develop and there were lots of other factors which could complicate the matter.

Finally, Prof. Wheelock still agreed with an article he had written in late 1991 which among other matters said that diet was the primary factor in the development of most of the degenerative diseases in many industrialised countries. The predominant diseases now were degenerative, such as heart disease and certain cancers. Research had established the causes of those diseases

Page 315

and now there was convincing evidence that diet was a key factor in their development. The most effective change in diet was to reduce the proportion of fat in the diet. High fat diets facilitated the consumption of energy. In relation to "cause of heart disease", when the risk was high, the diet was characterised by a high fat content, especially saturated fat, and by relatively low amounts of dietary fibre.

In relation to cancers it was now generally agreed that the incidence of all the common cancers in humans was determined by potentially controllable external factors, because people in different parts of the world suffered from different kinds of cancers, depending on their habits, diet and customs, rather than on their ethnic origins.

We had now reached the point where we can be very confident that diet was the primary factor in the development of most of the degenerative diseases that are the major public health problem in many industrialised countries.

I have already referred to Prof. Wheelock's hypothetical advice about the risks of eating a McDonald's meal every day unless one was very careful balancing one's diet.

Prof. Donald Naismith was Professor of Nutrition at Kings College, London, until about three years ago. He is now Emeritus Professor. He had been on the main Committee on Medical Aspects of Food Policy (COMA).

He said that cardiovascular disease arises from a combination of atherosclerosis (raised plaques on the linings of the arteries that result from the accumulation of modified lipids, including cholesterol, and other degenerative changes) and thrombosis (the formation of blood clots). Atherosclerosis narrows the artery and reduces the blood flow through it. Thrombosis formation can block the artery completely and lead to acute myocardial infarction (a heart attack). Both processes were believed, in part, to be influenced by diet.

Prof. Naismith said that the documented evidence implicating diet in the development of cardiovascular disease was vast and diverse. In addition to the kinds of investigation which had been carried out in relation to cancer, claims about the causation or protective effects of food or nutrients had been verified by diet modification (Nutritional Intervention Studies)

Page 316

in the case of cardiovascular disease, since the various risk factors (blood cholesterol level, blood pressure, body weight) could be detected in a relatively short time.

The nutritional aspects of cardiovascular disease had been reviewed by the Cardiovascular Review Group of the Committee on Medical Aspects of Food Policy and Prof. Naismith's comments on diet and heart disease were largely based on the COMA 46 report in 1994, which he regarded as authoritative.

Prof. Naismith thought that the COMA 46 recommendations were made by a body of scientists who had looked very carefully at a great deal of evidence and were making recommendation in the interests of public health. They were not saying we had conclusive evidence that eating a diet which was high in fatty saturated acids was going to cause heart disease. What they were saying was that the evidence seemed to be pointing in that direction, so if recommendations were made there was a reasonable chance that the incidence of cardiovascular disease would be reduced.

Prof. Naismith said that the cause of vascular disease was multifactorial. Diet was one factor. There was a general ethnic factor. Some people were predisposed to having cardiovascular disease. Smoking was an important factor. It probably had a greater effect on the development of cardiovascular disease than diet. Prof. Naismith was very pleased to see that the question of physical activity had been addressed because this was an important factor, given that the instance of obesity had increased in man in the last ten years with no change in the composition of the diet. Being overweight was a very important factor. There were other dietary factors apart from fat and saturated fat such as fibre content of the diet, interest in antioxidants, and in nutrients which promoted the degenerative aspects of atherosclerosis and, more recently, folic acid which is one of the B vitamins had come into the picture as a factor involved. It was important, therefore to recognise that when assessing the contribution of various dietary components to the development of cardiovascular disease, one was dealing with degrees of risk, not certainties. In the preface of COMA 46 it was stated that "The review group regard as wholly unrealistic the view that no recommendation can be made until there is absolute certainty", and later "the aim of prevention is to reduce the probability of subsequent disease."

Page 317

The major physiological characteristics that affected the likelihood of future occurrence of disease were raised blood cholesterol (hypercholesterolaemia), raised blood pressure (hypertension), an increased tendency for the blood to clot (thrombosis) and obesity.

Dealing first with hypertension, Prof Naismith said that high blood pressure promoted the formation of atheromatous lesions. Thus coronary heart disease risk rose with blood pressure. Although cross-population studies showed that dietary sodium (salt) intake was related to blood pressure, this had never been convincingly confirmed within a single population. Intervention studies on people with elevated blood pressure had shown a weak positive correlation between the reduction in salt intake and blood pressure, the fall being greater in older people, and in those with more severe hypertension, but even in the latter group the influence was variable. Any affect on young normotensive subjects was disputed. With a normal cardiovascular system a change in salt intake had no appreciable effect. If you had cardiovascular disease a reduction might prove beneficial but this very much related to one's genetics. Prof. Naismith agreed with the conclusions in COMA 46: "Sodium intake appears to be an important determinant of blood pressure in the population as a whole at least partly by influencing the rise in blood pressure with age. Although the quantitative degree of sensitivity to salt is still unclear, it appears to be greater with increasing age and higher initial blood pressure, and contributes to the quantitatively more effects of obesity and alcohol intake. A diet lower in common salt and higher in potassium would be expected to result in lower blood pressure and to a smaller rise in BP with age."

Prof. Naismith said that the average intake of sodium in the UK greatly exceeds the physiological requirement, and so a reduction in salt intake was recommended by the COMA Review Group.

Since much of the salt consumed was added at table at most meals every day, an adjustment in intake at that stage would seem more practicable and effective than the avoidance of foods thought to be high in salt (eg. a McDonald's meal) that were eaten only occasionally. A flat teaspoon of salt was probably about four grams. (4000mg).

The concern was mostly with cerebrovascular disease rather

Page 318

than cardiovascular disease.

So far as hypercholesterolaemia was concerned, Prof. Naismith said that there was extensive evidence indicating that high levels of serum cholesterol were predictive of a high risk of cardiovascular disease. Both epidemiological and clinical evidence showed that the quantity and quality of dietary fat were fundamental and modifiable determinants of plasma cholesterol.

About thirty years ago it was discovered that fats rich in saturated fatty acids (mainly of animal origin) raised plasma cholesterol whereas those rich in polyunsaturated fatty acids (mainly from plant sources) had a lowering effect. During the past three decades the picture had changed very little, although a great deal more was now known about biochemical mechanisms and the properties of individual fatty acids. More emphasis was now placed on the composition of the dietary fat than on total fat intake; In countries such as Greece and Spain total fat intakes were just as high as in the U.K., but the experience of cardiovascular disease was little over one third of the reported in the U.K.

More than 60% of the fatty acid that made up the fat of dairy or beef cattle either had no effect at all on blood cholesterol levels or actually reduced it. One had to look at specific components of saturated fat which might or might not have an effect on blood cholesterol.

The results of dietary intervention studies on at risk populations (middle-aged men, or patients with diagnosed heart disease) had been disappointing. The modifications tested in almost every randomized controlled trial had not been adequate effectively to reduce cholesterol levels. In a secondary intervention trial carried out by the MRC Epidemiology Unit in Wales on 2000 men who had already had a heart attack there was no significant reduction in mortality in those given advice to reduce the intake of saturated fats. There was, however, a 30% reduction in deaths among those who had been advised to eat oily fish at least twice every week. (Burr et al.,1989). The beneficial effects of oily fish were attributed to their content of long-chain polyunsaturated fatty acids which were believed to reduce thrombosis formation. Prof. Naismith agreed with statements in COMA 46 that a 1% lower plasma cholesterol translated into a 2 or 3% lower risk of coronary heart disease and that in fourteen trials over four years duration a 10%

Page 319

reduction in plasma cholesterol led to a 10% reduction in risk. The effect on dietary cholesterol was less important than that of fatty acids.

It was prudent, however, to recommend a reduction in total fat intake in the adult population from around 38% to 35% of the total food energy, and to make adjustment in early adult life. A shift from animal to vegetable fats was also proposed. Not only was it likely that a substantial proportion of the population would benefit from a lowering of the blood cholesterol level, but the rising incidence of obesity in the adult population might also be halted or reversed.

In Prof. Naismith's opinion the small proportion of fat derived from the occasional McDonald's meal (a mixture of animal and vegetable fats) would have little effect on a total fat intake or on fatty acid composition, and no appreciable effect on the blood cholesterol level and thus on the risk of developing cardiovascular disease.

So far as obesity was concerned, the incidence of cardiovascular disease was high in obese man and women, particularly in those under fifty years of age. There was a gradual increase in risk with rising Body Mass Index which was a measure of the degree of adiposity. People who were obese usually suffered from hypertension and an elevated plasma cholesterol level, both of which declined when weight was lost.

Obesity arose from an imbalance between food energy intake and energy expenditure (basal, metabolic cost plus physical activity) over a period of years. It also had a genetic component, but whether this manifested itself in physiological or behavioural differences for lean people was unclear.

It had been suggested that sugar, because of its attractive taste, encouraged excessive energy consumption, but evidence had accumulated that both children and adults who derived a high proportion of their energy from sugar tended to be leaner, not fatter, than the average (Naismith et al.1995).

It was now thought that a diet high in fat facilitated "passive overconsumption" of energy leading to weight-gain, particularly when coupled with low level of physical activity.

The energy imbalance was subtle and the disease insidious.

Page 320

The difference in energy content between one pint of whole milk and one pint of semi-skimmed milk consumed daily would lead to the accretion of loss of 13 pounds (5.9kg) of body fat in one year.

The proposed reduction of the average fat intake to 35% of total energy would be expected to reverse the trend in weight gain and in the associated risk factor (hypertension and hypercholesterolaemia).

In relation to fibre and cardiovascular disease, Prof. Naismith said that although the notion that a dietary component (fibre that was not digested in the small intestine) could affect the blood cholesterol concentration, might seem improbable, short term experiments on human subjects in which some foods rich in insoluble fibre (beans and oats) were fed to volunteers showed a modest lowering of the LDL-cholesterol concentration. However the amounts used were substantial compared with usual dietary intakes, and the effect was not shared with most forms of insoluble fibre.

A number of prospective studies had revealed an inverse relationship between dietary fibre intake (from cereals and from fruit and vegetables) and the incidence of coronary heart disease. A high fibre diet was characterised by its high content of antioxidant nutrients and lower than average content of saturated fat.

The two year intervention study conducted on 2000 men who had recently experienced a heart attack to which reference was made earlier, found that no significant reduction in mortality from coronary heart disease in those who increased their cereal fibre intake (Burr, 1989).

Prof. Naismith said that there was no clear evidence that a high fibre diet prevented overweight or obesity. There was a concept that because fibre was bulky this would somehow fill the stomach of the consumer and he would be less inclined to eat more food but unfortunately it did not work that way. If people were prepared to eat really large amounts of vegetables it could conceivably have an effect.

Having said all this the relationship between fat ingestion and the development of coronary heart disease was far from clear, largely because of many confounding variables, dietary, behavioural and environmental, and the claim that McDonald's food promoted it was not supported by any scientific evidence. He thought that there were

Page 321

a number of questions to be asked about the very short statements and abstracts of Dr Miller's and Prof. Vogel's work and results and he did not find them significant to the role of fat in the causation of cardiovascular disease.

Prof. Naismith ended his evidence by saying that one must consider diet and other aspects of lifestyle as a whole, and not look at individual components of the diet because one could be very misled by that. Exercise which Prof. Naismith mentioned on a number of occasions, was a very important aspect of lifestyle and it had been addressed for the first time in the COMA 46 report.

The Plaintiffs' principal witness on the subject of diet and cancers of the breast and bowel, was Dr Sidney Arnott.

Dr Arnott has been Consultant in Radiotherapy and Oncology at St Bartholomew's Hospital, London, for about twelve years now. For about ten years before that he was Senior Lecturer in Clinical Oncology at the University of Edinburgh. The remit of that particular appointment was to conduct research into a variety of treatments of cancer, but his work did have a clinical component, treating patients. His work at Barts is more treatment orientated. The radiotherapy involves treatment by X-rays and by the insertion of radioactive substances in tumours. His work has always involved a necessary concern for the aetiology of cancers, that is the science of the causes of cancer, which could take a variety of different forms including genetic predisposition, biological causes such as viral infections, chemical induction and physical causes such as exposure to radiation.

Dr Arnott has particularly written on the subject of colo-rectal cancer and its pathology, aetiology and treatment. He has examined medical undergraduates and postgraduates and when he first gave evidence in July,1994, he was an examiner for the Royal College of Surgeons of Edinburgh. He said that if he was examining Dr Barnard on the aetiology of cancers he would have to fail him for failing to give a balanced presentation of the evidence for and against diet and components of diet as causes of cancers. He produced only one side of the evidence.

Dr Arnott started his evidence by explaining that there are

Page 322

many different kinds of cancers. Cancer of the breast and colon are adenocarcinomas, cancers of covering tissue: the skin in the case of the breast which is essentially a modification of the skin, and the epithelium in the case of the colon (large bowel) or rectum.

Dr Arnott said that a lot of the mechanism by which a cancer (a tumour) developed were poorly understood, but it would appear that there was a process of initiation which implied some change in the genetic mechanism within the cell arising perhaps from radiation exposure or genetic predisposition. It was now thought that there were oncogenes but there were also inhibitors. The balance kept us healthy unless the inhibition was lost for some reason. This stage, the change in the cell, was probably irreversible but it did not influence one's life expectancy unless there was the next stage of promotion.

Promotion of the changed cell seemed to be related not only to chemical exposure, but perhaps viral infections or other biological factors including genetic predisposition. But it was not an irreversible process and the factor influencing promotion and therefore moving the cell further along the line to the development of a cancer or tumour could be inhibited or halted by the removal of the particular exposure which was promoting progress or by exposure to inhibitors which might reverse the process.

After the stage of promotion came the stage of progression, at which stage change within the cell enabled it to adopt a malignant potential with unlimited growth and the ability to divide and invade surrounding tissue, and the ability to metastasise and escape from the primary site and spread either in the lymph system or the blood stream to other parts of the body, setting up secondary cancers.

The whole process might take many years to develop. For instance some of those who suffered the atomic bomb at Hiroshima received a substantial exposure to radiation but tumours sometimes did not develop until as long as twenty or thirty years later.

The aetiology of cancer was an extremely important field because in spite of many developments over many years there were major difficulties in the curability of established cancer. If the development of cancer could be prevented that would be a

Page 323

tremendous boon and would make a major contribution to world health.

However, there were difficulties with the aetiology of cancers. One of the problems was that the direction of a lot of research was stimulated by epidemiological studies; they could rarely, accurately specify a cause and effect relationship. What they did was raise an hypothesis which stimulated research to see if cause and effect could be accurately identified.

There were difficulties with all the forms of epidemiological studies so far as cancers were concerned, in Dr Arnott's view.

So far as population studies of different countries were concerned, there were many reasons why there might be differences between different peoples, and diagnoses and cancer registers in Third World countries were very poor.

Armstrong and Doll stressed the weakness of international correlation and cautioned that they should be taken only as suggestions for further research and not as evidence of causation or as bases for preventative action.

Population studies were often based on food balance tables which measured the total national production of a substance such as fat, including imports, and then tried to make an allowance for wastage and exports, in order to calculate the consumption in a particular country. It was a very crude analysis.

Moreover when you got down to populations where the intake of energy from fat was low, for instance less than 25%, other aspects of their lifestyle might be important. It was complex.

With case control studies one took a healthy group of people who did not have the disease in question and compared them with people who had the disease and when you looked to see what happened to them you might see a lower incidence in the numbers of the group which was healthy when you started to study them. Again people's recall of diet was very poor and "portions" of food meant different things to different people.

In Dr Arnott's view, cohort or prospective studies were the most reliable or the least fraught with difficulties. Some could be properly criticised for narrow differences in, for instance, fat intake; and for shortness of time span; but dietary details were collected before the development of cancer. Questions about food intakes in the

Page 324

last twenty-four hours, for instance, were regularly asked, and the possible effects of the disease on diet and on dietary recall were therefore avoided. Diet was difficult to recall at the best of times and people who developed cancer did, consciously and subconsciously, modify their diets and this could influence their recall making them think it was bad beforehand.

There were problems with animal studies. The physiology of animals was different. The animals themselves were specially bred to make them more susceptible. They were subjected to specific stimuli in artificial circumstances and given very large doses of a suspected substance. This often brought quick results which could not be directly transposed to humans; but they did sometimes provide pointers.

So far as studies on human beings were concerned, it was not ethical to carry out the same experiments as with animals. Intervention Studies were taking place but it was very difficult to get people to make major changes in their diet. Inhibitors had been looked at, but there were no answers so far.

Dr Arnott did not know of any epidemiological work which had established the cause of cancer of the breast or cancer of the bowel. In his view this situation remained extremely controversial. The studies suggested that a great deal more research was needed.

Dr Arnott was taken through a number of papers: many of them the same as those which Dr Barnard referred to. I will refer to his comments on some of them, which particularly reflected or introduced his own views, dealing with cancer of the breast first, then colon cancer, before dealing with developments between the middle of 1994, when he first gave evidence, and May,1996, when he returned to the witness box.

Dr Arnott referred to Dr Kinlen's 1987 paper as an introduction to his views on the epidemiological risk factors for breast cancer. I have set out the summary when relating Dr Barnard's evidence. Dr Arnott thought that dietary fat as a cause of breast was a popular theory at that time but he agreed with Dr Kinlen that the evidence

Page 325

for it had been exaggerated. The important epidemiological factors in relation to the incidence of breast cancer were those which Dr Kinlen mentioned, for example age at menarche (the younger, the greater the risk), and age at birth of first child (increased risk if over age of thirty). Parity was also important. If one had more children there was a significant reduction of risk.

Menarche probably occurred at an early age in people who were well fed compared with people who were suffering from nutritional deprivation. Having children late in life was often a western may of life. So was having less children.

Later age of menopause which was found mainly in western countries was associated with an increased risk of the development of breast cancer.

The association between obesity and breast cancer was secondary in the sense that women who were obese and who had breast cancer had a poorer outlook. The theory was that oestrogen-like compounds might be produced in adipose tissues and change the hormonal environment of the particular patients body. Breast cancer was a hormonally influenced cancer. That is why the time when it most frequently developed was about the time of the menopause.

One of the most predominant features in breast cancer was family history; if a blood relative had breast cancer that significantly increased the risk.

Dr Kinlen reviewed a number of the studies up to the time of his paper, including some of those upon which Dr Barnard relied, and in Dr Arnott's view the results showed the lack of consistency in all of the findings in the published work, and that one could not say that there was a relationship between fat intake and breast cancer.

Dr Arnott did, however, say that weight for weight fat was the greatest source of energy or a readily available source of energy, and that Dr Kinlen's statement that there was much evidence for the role of "overnutrition" in breast cancer aetiology and that the more calorigenic nature of fat than either carbohydrate or protein suggested a connection, was a fair statement.

It was this way of resolving the differences between the international correlational and individual-based studies,

Page 326

suggested by Dr Kinlen, upon which Dr Barnard in part relied.

But Dr Kinlen went on to say that: "At the population level the international data do not allow fat to be distinguished adequately from other features of wealth, while at the individual level we are ignorant of whether the effects of calories from fat differ from (calories) from carbohydrate and protein. Until such a difference is found, it seems preferable to regard excess calories from any source as the important factor".

Dr Arnott thought that although that was said in 1987, it was a very reasonable proposition and the state of knowledge had not progressed since then, and that we still did not know whether total calorie intake was a cause of breast cancer.

Dr Arnott thought that Dr Kinlen's paper was a very good summary of the evidence, even as we understood it today: "There are accepted risk factors for breast cancer which we know about and these seem to be the predominant factors. What the influence of diet is remains unclear. Certainly, there is no clear evidence that fat is responsible for the development of breast cancer. There is suggestive evidence that it may be overnutrition in a general sense with the calories coming from other sources than fat".

Dr Arnott was referred to the 1987 paper by Jones and others which followed up 5,485 women. The paper acknowledged methodological problems. For instance the women were over twenty-five and the distribution of fat intake was heavily shifted towards relatively high intakes. But the report did show that the women who tended to agree to take part in such surveys tended to be in the upper socio-economic classes which were at higher risk because they often subjected themselves to risk factors like low numbers of children and having a first child at a late age. Moreover Dr Arnott agreed with its conclusion that there was a long way to go.

Vatten and Kvinnsland in 1990 found a positive association between height and the incidence of breast cancer. The women concerned had been subjected to nutritional deficiencies before, at the time of and just after menarche, leading to the hypothesis that the change in diet and better feeding after the war occurred at a sensitive time and also led to people of greater height.

Dr Arnott referred with approval to Dr Kinlen's 1991 paper,

Page 327

parts of which I have set out in relating Dr Barnard's evidence. He said that the protective effect of green vegetables suggested by certain studies, which Dr Kinlen said required confirmation, had not been obtained since 1991.

The 1992 paper of Dr Willett and others, which Dr Barnard criticised for its high and narrow range of fat intakes, was called "Dietary Fat and Fibre in Relation to Risk of Breast Cancer: An 8-Year Follow-up". The summary of the paper was:

"Objective - To address the hypotheses that dietary fat increases and fibre decreases the risk of breast cancer.

Design - Prospective cohort study with dietary assessment at baseline, using a validated, self-administered food frequency questionnaire.

Setting/Participants - 89494 women in the Nurses' Health Study who were 34 through 59 years of age in 1980 and who were followed up for 8 years (95% complete).

Results - 1439 incident cases of breast cancer were diagnosed, including 774 among postmenopausal women. After adjustment for age, established risk factors, and total energy intake, we observed no evidence of any positive association between total fat intake and breast cancer incidence (relative risks [RRs] for increasing quintiles of fat intake were 1.0, 0.85, 0.96, 0.91, and 0.90: 95% confidence interval for highest vs lowest quintile, 0.77 to 1.07). Among postmenopausal women alone, corresponding RRs were 1.0, 0.89, 1.00, 0.95, and 0.91. Comparing extreme deciles of total fat intake (49% vs 29% of total energy intake), the RR was 0.86 (95% confidence interval, 0.67 to 1.08). A similar absence of any positive association was observed without adjustment for energy intake: for tumours less than 2 cm as well as 2 cm or greater in diameter: for saturated, monounsaturated, and polyunsaturated fat: and after excluding the first 4 years of follow-up. Also, we found no suggestion of any positive association when using a more detailed and precise dietary questionnaire completed in 1984 (666 subsequent cases), even when women consuming less than 25% of energy from fat were used as the comparison group. No suggestion of a protective effect of dietary fibre was observed (RRs for increasing quintiles were 1.0, 0.95, 0.93, 1.02, and 1.02).

Conclusions - These data provide evidence against both an

Page 328

adverse influence of fat intake and a protective effect of fibre consumption by middle-aged women on breast cancer incidence over 8 years ......"

This was referred to as a demonstration of the conflict in the epidemiological studies on fat intake and breast cancer. In fact while the consumption of dietary fat had fallen significantly over the last twenty or thirty years in the U.S., the incidence of breast cancer had increased by some 40% over the same period.

The final sentence of the 1992 paper of Willett and others read: "Nevertheless, the positive association between intake of animal fat and risk of colon cancer observed in many studies provides ample reason to limit this source of energy."

Dr Arnott said that Dr Willett's study of American nurses to a large extent set the suggestion that a large intake of animal fat gave rise to the risk of colon cancer, running. But the sentence quoted was not mainstream medical thinking. Most thinking medical people would say that we do not know the answer to fat and colon cancer because studies again produced conflicting evidence regarding the relationship between dietary fat and the development of colon cancer. The answer was by no means clear.

Dr Arnott referred to a 1978 paper by Reddy, Wynder and others which looked into a high risk population in Metropolitan New York and a low risk population in rural Kuopio, Finland. The dietary fat intake was the same although it was mostly from meat in New York and from dairy products in Kuopio. The paper said: "The data suggest that one of the factors contributing to the low risk of large bowel cancer in Finland, in spite of high dietary intake of fat, appears to be the fact that high dietary fibre leads to an increase in stool bulk, thus diluting bile acids which have promoting activity."

Dr Arnott said that in animal experiments it had been shown that, when applied to tumours induced by another chemical, bile acids would promote the development of the tumour, but it was not known whether that applied to the human situation. The concentration of the chemicals applied to the animals was greater than to humans.

Dr Arnott himself thought that the dilution was of bile acids which might have promoting

Page 329

activity. The bowel acids and other factors produced in the bowel were responsible for the digestion of food and then subsequently were acted upon by bacteria which normally existed in our bowels. These were the substances which had been thought to be possibly related to the development of tumours. Diluting these or encouraging the passage of the faeces through the bowel more rapidly allowed them to act with the bowel lining only for a relatively short period of time, or reduced the concentration of them acting against the bowel lining.

Dr Arnott approved and agreed with what Dr Basil Morson and others wrote in a chapter called Malignant Epithelial Tumours, of a book called Gastro-intestinal Pathology. Dr Morson was an accepted world figure in bowel pathology. He wrote as follows:

"Aetiology and pathogenesis

    Multiple factors are involved in the aetiology of colorectal cancer just as multiple steps are implicated in its pathogenesis. It is probable that both environmental and constitutional factors are interwoven within the aetiology but that the precise composition of events varies from one individual to another. Thus neither one factor nor one set of factors will underlie the aetiology of all cases of colorectal cancer. Cancer is ultimately a disorder at the level of the gene, though the behaviour of a malignant cell will depend also upon the mode of gene expression; this may be influenced by environmental factors. The initiated state may in some instances be constitutional, perhaps due to the inheritance of a transforming or potentially transforming gene (oncogene) or the inherited loss of a growth controlling gene ......


    The considerable geographical variation in the incidence of colorectal cancer signals the importance of environmental factors in the aetiology of the disease. This variation does not appear to be related to racial differences as migrants from low- to high-risk areas acquire the pattern of incidence appropriate to their adopted country. The attention

    Page 330

    of epidemiologists and experimentalists has focused mainly upon diet. Blame has been placed upon the excess consumption of meat and animal fat, refined carbohydrate or beer at the expense of fibre, vegetables and trace elements. It is likely that the various dietary hypotheses are inter-related and that a summation of factors is ultimately responsible for the provision of a carcinogenic microenvironment ......

    Faecal bile acid levels have been measured in high and low-risk populations. This epidemiological approach has been utilized by several groups and strongly supports the role for bile acids in large bowel carcinogenesis. It has been suggested that increased consumption of meat and hence animal fat would account for the high levels of faecal bowel acids detected in high-risk populations. There is a close relationship between per capita meat intake and colorectal cancer incidence in various populations. Seventh Day Adventists who consume little or no meat show a relatively low rate of colorectal cancer as compared to their fellow Americans. However, a survey of dietary variations within Seventh Day Adventists showed no relationship between meat consumption and large bowel cancer. New Zealand Maoris enjoy a high intake of fat and protein, yet the incidence of colorectal cancer is low in this group. It is relevant also that dietary manipulation has failed to influence the bacterial flora of the large intestine. Case control studies (in which each patient with colorectal cancer is matched to a 'normal' control) have usually failed to support a link between either meat intake or faecal bile acid levels and colorectal cancer. It is important that patients with colorectal cancer are carefully selected for such studies. Patients with cancer may have modified their diet and the presence of liver metastases could influence bile acid metabolism. It has been argued that the failure of case control studies to support dietary and/or the faecal bile acid

    Page 331

    hypothesis may reflect the confounding influences of other important factors (e.g. genetic) in the aetiology of bowel cancer ...... the dilutional effect of a bulkier stool remains an attractively simple hypothesis, although experimental models have produced conflicting findings ......

    The promoting effects of faecal bile acids are thought to be due to cytotoxic injury leading to regeneration with accompanying increased cell turnover."

These are but extracts from what Dr Morson wrote. Dr Arnott said that what Dr Morson was saying was that there were these theories which had been put forward to try to explain why there might be a relationship between bile acids, fat in the diet and the possible causation of colorectal cancer. But we could not go any further forward and we did not have the evidence to say there was a direct causal relationship. Those were purely theories which had been put forward to try to explain what some people had suggested was an effect.

Dr Arnott said that so far as increased consumption of vegetables was concerned one could not go further than saying that it might reduce the risk of fatal colon cancer, although it might be the strongest dietary relationship with colon cancer. Some studies had not found a benefit so far as breast cancer was concerned. Dr Arnott said that it was difficult to ascertain whether the method by which meat was cooked was a factor. He said that there was no doubt that when food was cooked, molecular changes in the constituents of the food were produced. It was a factor that had to be taken into consideration but there were no consistent findings in the studies.

As with breast cancer, if a blood relative had developed cancer of the colon or rectum, that increased the risk of developing that cancer: two to four times in the case of cancer of the colon or rectum. That led to the feeling that genetics played a very important part in the genesis of colon cancer, but the genetic aspects of cancer were very complex.

Dr Arnott thought the conclusion of COMA 41, to the effect that there was currently insufficient evidence on which to base a recommendation for a decrease in fat intakes to prevent cancer, although an increase in the consumption of any fatty acids should be discouraged, was very fair bearing in mind the current level of knowledge.

Dr Arnott gave a very low weight to the validity of a proposed causal relationship between a diet high in fat and breast cancer, a slightly higher but still low weight to a diet

Page 332

high in energy and breast cancer and colorectal cancer and to fat and colon cancer, and slightly higher still, but still quite low weight to the consumption of large amounts of plant fibre as a protective agent in relation to breast cancer and colon cancer. He said that we were seeing some evidence to suggest that those factors might be involved but we did not have the scientific evidence to show that they were involved. They remained theories; not proven fact.

Most certainly almost all cancers had a long gestation period, and people's diets changed, and the diseases tended to be multi-factorial.

Speaking of the various studies generally Dr Arnott said that there was a general problem in medicine that it was "always nice to report something positive whereas, in fact, in reality, reporting a negative is equally as important". One had to have consistency of results and the results must be reproducible from study to study before one could say that a definite relationship between diet or a component of diet and cancer was accepted as medically proven. There was no consistency and there was no reproducibility in the studies.

Both towards the end of his evidence in September, 1994, and when he returned in May,1996, Dr Arnott said that in the 1970s it looked as if a similar relationship to that between smoking and lung cancer had been found between dietary fat and cancer of the breast and bowel, but in the mid to late 1980s as studies had been done to try to confirm the relationship the picture had became less clear. Between 1994 and 1996 the strength of belief in a direct relationship between diet and cancers of the breast and bowel had became less strong.

Dr Arnott referred to the February,1996, paper by Hunter and others which involved the collaboration of several of the world's leading units in the field of epidemiology and to which Dr Barnard referred, in order to pass it off, in his July,1996, statement. The abstract to the article read:

    "Background. Experiments in animals, international correlation comparisons, and case-control studies support an association between dietary fat intake and the incidence of breast cancer. Most cohort studies do not

    Page 333

    corroborate the association, but they have been criticized for involving small numbers of cases, homogeneous fat intake, and measurement errors in estimates of fat intake.

    Methods. We identified seven prospective studies in four countries that met specific criteria and analyzed the primary data in a standardized manner. Pooled estimates of the relation of fat intake to the risk of breast cancer were calculated, and data from study-specific validation studies were used to adjust the results for measurement error.

    Results. Information about 4980 cases from studies including 337,819 women was available. When women in the highest quintile of energy-adjusted total fat intake were compared with women in the lowest quintile, the multivariate pooled relative risk of breast cancer was 1.05 (95 percent confidence interval, 0.94 to 1.16). Relative risks for saturated, monounsaturated, and polyunsaturated fat and for cholesterol, considered individually, were also close to unity. There was little overall association between the percentage of energy intake from fat and the risk of breast cancer, even among women whose energy intake from fat was less than 20 percent. Correcting for error in the measurement of nutrient intake did not materially alter these findings.

    Conclusions. We found no evidence of a positive association between total dietary fat intake and the risk of breast cancer. There was no reduction in risk even among women whose energy intake from fat was less than 20 percent of total energy intake. In the context of the Western lifestyle, lowering the total intake of fat in midlife is unlikely to reduce the risk of breast cancer substantially."

Dr Arnott said that studies had previously suggested that a relationship between dietary fat and breast cancer was most marked in post menopausal women. But in this case their results were similar to the entire population.

There was a marginally significant relationship between cholesterol intakes and breast cancer incidence, but this was seen in only one group of women in the nurses health study in the U.S.

Page 334

and not in any of the other studies.

Dr Arnott also referred to Dr Hill's review of the scientific evidence published in 1995. Dr Arnott said that it was a major review of all the sources of evidence relating diet to development of cancer and critically evaluated the quality of such evidence. It clearly pointed out that there was increasing realisation of the inaccuracy of international epidemiological surveys and that the current attitude to such studies was that they are only able to generate hypotheses which required further testing in more accurate studies. The conclusion was drawn that at the present time only by evaluation in prospective studies and possibly intervention trials could one fully assess the relationship between diet and cancer.

The known shortcomings of case control studies were described. Dr Hill also described the limitations of the prospective cohort studies which were difficult to organise and required large amounts of time and money. With regard to these studies a further limitation was pointed out that a long follow-up period was needed in order to avoid the study simply measuring the aetiology of late stage events.

With regard to the use of animal experiments, whilst it was accepted that the study of carcinogenesis in animal models had provided invaluable information since investigators had been allowed to control for virtually all potentially confounding factors so that any one substance could be examined in isolation, it was pointed out that the very complexity of human carcinogenesis could not be taken into account when setting up animal models.

Dr Arnott said that this comprehensive review quite clearly demonstrated the uncertainty which existed at present concerning the relationship between diet and the development of cancer of the breast or cancer of the bowel.

Dr Arnott said that one of the few positive conclusions reached in Dr Hill's major review was that fruit and vegetables did appear to protect against carcinogenesis at a wide range of sites.

Patients frequently asked him: "What do I do about my diet? What modifications can I make?" So he tried to read the evidence that was published to give patients a reasonable judgment about what they might do to improve

Page 335

their chances of having successful treatment, and obviously being cured of their disease. He tried to be honest with them and said that at the moment, the answer was not known, but the thing which appeared to come across most strongly was that a good intake of fruit and vegetables was important and that the other factor which seemed to be important was that they should try to take some exercise. He said that question marks regarding fat in the diet had been raised, and certainly we knew that people who were obese did seem to have a worse outlook following the treatment of their cancers than people who were thin. One of the easiest ways to become obese was to eat too much fat, because of the relative amount of calories that fat contained relative to the size of the meal you took. So he said that it was only reasonable, if you were trying to avoid becoming obese, to try to cut down on your fat intake.

Mr Morris put to Dr Arnott a quote from the leaflet complained of that: "A diet high in fat, sugar, animal products and salt and low in fibre, vitamins and minerals is linked with cancer of the breast and bowel and heart disease".

Dr Arnott thought that was a reasonable thing to say to the public; and at the time (September, 1994) the Defendants appeared to treat Dr Arnott's answer as the end of the matter on this part of the case. It was not the end of the matter because that single sentence was unspecific about the nature of the link and it did not reflect the meaning of the whole of the relevant parts of the leaflet.

Dr Arnott's last word on diet and cancer of the breast and cancer of the bowel was that whereas he would say that if someone continued to smoke cigarettes in any substantial numbers he would create a very real risk for himself of getting cancer of the lung, he could not say someone taking a diet high in fat and low in fibre created a very real risk of getting cancer of the breast or cancer of the colon (bowel) because he did not think the evidence was there.

Dr Arnott did say that he was convinced by the evidence associating a high intake of fat, particularly saturated fat, with coronary heart disease. He would say that there was "a good relationship between coronary artery disease and fat intake". Saturated fat had an influence on the amount of cholesterol in the blood and that was a factor in atherosclerosis.

Page 336

Also the presence of fats themselves within the blood led to deposition of fats on blood vessels - atherosclerosis. But there was no such simple mechanical explanation available in relation cancer of the breast or cancer of the colon.

Dr Arnott did not think that there was any scientific basis for Prof. Crawford's theory of the common denominator between heart disease and cancer. Two common sites of cancer and metastases were the liver and lungs, but atheroma, i.e. fat deposited in the middle of the wall of the blood vessels causing changes in the membrane overlying that part of the wall, was never found in the arteries supplying the liver and was found in the lungs only when pre existing pulmonary hypertension existed, which was uncommon.

There was no evidence that he knew of that dietary fat increased the stickiness of cancer cells. One difficulty with cancer was that cells could break off and spread to other parts of the body, so increasing their stickiness would be beneficial in terms of localising the cancer to a particular site. It did not make any sense to Dr Arnott as an oncologist that cancer cells circulating in blood which had more fat in it would themselves became stickier and stick to the walls of blood vessels or the linings of the lymphatic system. The suggested mechanism broached on a topic which taxed medical science as to why certain organs should be preferential sites of secondary deposits and how secondary deposits arose. There had been a lot of research on the topic and it had not come up with any answers.

The Plaintiffs' final expert on diet and degenerative disease was Prof. Harry Keen, Emeritus Professor of Human Metabolism and Honorary Consultant Physician and Late Director of the Unit for Metabolic Medicine, United Medical and Dental School of Guys and St Thomas's Hospital.

Prof. Keen was primarily called to give evidence on diet and diabetes but he has considerable learning and experience in the role of diet in the genesis of coronary heart disease, cerebro-vascular disease and human disease generally. he has served on WHO and COMA committees.

Prof. Keen said that there were strong correlations between high fat intake and total food energy consumption and between them and obesity. High average fat intakes were

Page 337

also associated with low natural dietary fibre consumption and with more prosperous economies in which relative physical inactivity was prevalent.

Prof. Keen said that fat was a ready source of energy which might, therefore, contribute to obesity if eaten in excess quantities. He said that there was no real distinction in this respect between animal fat and vegetable fat: one would receive nine calories per gram of fat, whatever the fat was.

Prof. Keen would not entirely agree with a statement in the WHO Report that: "Excessive dietary fat is more readily stored, and fibre-rich complex carbohydrates are much bulkier and tend to limit energy intake." He would agree that it was less energy demanding to store excess fat than excess carbohydrate. There was no difficulty in storing excess carbohydrate. It was just more energy costly to do, not enormously so, but you had to turn it into fat, then store it as fat. The WHO statement rather implied that in some way the fat slipped into place rather more readily than carbohydrate slipped into place as fat. He thought that there was absolutely no evidence for that. What there was evidence for was that if you ate excess carbohydrate and it was stored as fat, it cost a little more in terms of energy to do so. So perhaps the net effect, calorie for calorie, was rather less. He agreed that fibre-rich complex carbohydrates were much bulkier, but he did not know what the evidence was for saying that they tended to limit energy intake. A person eating them had to eat more and you had to use more chewing energy to get them down, but a lot of people seemed to get plump on high fibre complex carbohydrate foods. It certainly did not preclude people putting on additional weight. It might make it more tedious.

Since obesity was a risk factor for non-insulin-dependant diabetes mellitus with which Prof. Keen had spent so much of his life, he must have come across a lot of fat people, and he went on to say that on the whole fat people did not eat any more than thin people and they sometimes ate less. The composition of their diets so far as fat and carbohydrate contents were concerned did not differ from thin people. They took roughly the same proportions of fat, carbohydrate and protein in their diet. It might seem reasonable for people who ate more fat to be fatter and people who ate more carbohydrate to be thinner but it was not what he found. Presumably the difference was in their physiological make-up and they were eating too much of something, fat or carbohydrate, or both, for theirs.

In Prof. Keen's view dietary fat might be partly responsible

Page 338

for the problem of obesity but it was just an index for a whole lifestyle, a whole complex of things which went along with the consumption of fat, like owning a car, taking less exercise and eating more of everything. You might cut fat, but you would not necessarily change your lifestyle.

Prof. Keen said the evidence that a high intake of sodium was a cause of hypertension was very indirect. It was based on the Intersalt Study sponsored by the WHO. Sodium excretion which was taken to be an index of sodium intake, over twenty-four hours was measured in many cultural groups and that was related to the rise in blood pressure with age, not to blood pressure itself. It was found that in some but not all cultures the rate of rise of blood pressure with age increased, so there appeared to be an overall relationship. If you moved a whole population up, more people would fall into the hypertensive category, so to that extent the number of people with hypertension would depend on the average salt intake and the higher that average salt intake, the more the people who would fall into the hypertensive group; but it was very difficult - Prof.Keen would say impossible - to identify which individual was going to suffer hypertension.

Prof. Keen quoted with approval the 1989 report of a COMA Panel on Dietary Sugars and Human Disease, of which he was Chairman, to the effect that current consumption of sugars particularly sucrose played no direct causal role in the development of cardiovascular (atherosclerotic coronary, peripheral or cobra-vascular) disease or essential hypertension. Sugar in its complex form as starches was a major source of food energy. An excess of sugar, like an excess of any source of food energy would contribute to obesity. Sucrose had certain adverse effects. It was causally related to dental caries. But so far as any other major effects on health, other than over-consumption through obesity were concerned, the evidence was not convincing at all.

A conclusion of the WHO Report was put to Prof, Keen that:

"Dietary factors are not known to influence the development of a wide range of chronic diseases, e.g., coronary heart disease, various cancers, hypertension, cerebrovascular disease, and diabetes. The conditions are the commonest cause of premature death in developed countries and they impose major burdens on society. On current projections,

Page 339

cardiovascular diseases and cancer will emerge, or be established, as substantial health problems in virtually every country in the world by the year 2000."

Prof. Keen would express the matter rather differently. He thought that the word "influence" meant that if you changed the dietary factors you would change the development of the chronic disease. There was little evidence for that. It might be so, but it had not been demonstrated. It would be fair to say that dietary factors and the diseases were linked but that the links might be indirect which was a very important proviso.

Prof. Keen did say that so far as coronary heart disease was concerned one should try to limit the proportions of one's diet as fat to 35% or 30% calories. There were two good reasons for reducing fat in the diet. It would help to prevent obesity if one was moving that way, and it would help to protect the cardiovascular system.

Finally, Prof. Keen said that if one was in weight equilibrium as a person whose weight was at a desirable level, taking 30 to 35% of calories as fat and the rest as carbohydrate and protein would be perfectly acceptable.

Finally, so far as the third issue is concerned, I note that "The Health of the Nation" targets for lowering the average percentage of food energy derived by the population from saturated fatty acids and from total fat, to which I have already referred, appeared in the section of the White Paper, which was devoted to reducing the level of ill-health and death caused by coronary heart disease and stroke, and the risk factors associated with them. The paper said:

"It is generally accepted that the main risk factors for CHD and stroke are: cigarette smoking, raised plasma cholesterol, raised blood pressure and lack of physical activity.

Plasma cholesterol is particularly important in setting the baseline level of CHD risk. Raised blood pressure is the most important risk factor for stroke. Other predictors of risk, such as socioeconomic factors, influences in early life and stress, are less well understood.

All the main risk factors can be influenced by changes in behaviour. For those who smoke, stopping smoking is the single most effective means of reducing the risk of CHD and stroke.

Page 340

Excessive dietary intake of saturated fatty acids results in raised plasma cholesterol levels. Excessive consumption of alcohol, and sodium (mainly as sodium chloride - common salt) contribute to raised blood pressure. Obesity contributes to both raised plasma cholesterol levels and raised blood pressure. Obesity results from a dietary energy intake chronically in excess of energy expenditure and is thus related to both diet and physical activity. Physical activity also acts directly to reduce the risk of CHD and stroke. Beneficial effects on CHD or stroke should therefore result from: stopping smoking, reducing consumption of saturated fatty acids and sodium, reducing alcohol consumption and increasing physical activity.

Because the factors can be influenced, much heart disease and stroke is potentially preventable. People of all ages can reduce their risk of CHD and stroke. Even late in life, changes in lifestyle can give considerable benefits."

The paper said that Coronary Heart Disease (CHD) accounted for about 26% of deaths in England in 1991. It was both the single largest cause of death, and the single main cause of premature death. Approximately 12% of all deaths in 1991 resulted from stroke.

The section on cancers began by saying that after coronary heart disease, cancers were the most common cause of death in England in 1991, accounting for about 25% of deaths in 1991. It said that there were many types of cancers and that understanding of the causes of them varied greatly, as did current ability to prevent, treat and cure them.

Most of the section on cancer was devoted to smoking and to lung cancer, but it did say that more women died from breast cancer in England than from any other form of cancer. It quoted the "Ten Commandments" from the European Code Against Cancer. The first commandment was to stop smoking. The fifth was: "Frequently eat fresh fruit, and vegetables, and cereals with a high fibre content." The sixth was: "Avoid becoming overweight and limit your intake of fatty foods."

It was part of the Second Plaintiff's case that it supported "The Health of the Nation."

Before reaching my conclusions on whether it has been proved that a diet high in fat (including animal fat), animal products and salt (sodium) and low in fibre leads to a real risk of heart disease or cancer of the

Page 341

breast or cancer of the bowel, I must give my assessment of the main expert witnesses.

Dr Barnard had a wide knowledge of the copious literature on diet and disease, and no doubt his medical training gave him some additional insight into the research, but I found him an unsatisfactory witness. I agree with Dr Arnott's criticism that Dr Barnard failed to give a balanced presentation of the evidence for and against diet and components of diet as causes of cancers. Dr Barnard was far too adept at stressing what accorded with his long-held views and dismissing what did not. He showed no perception of the difficulties which clearly face the genuinely open-minded researcher in this field. He took an over optimistic view of what some reports, like the U.S. Surgeon General's for instance, meant rather than noting what they actually said. Even then he overstated his case, casting blame on individual meals or at least intermittent meals or diet and likening them to the one bullet in Russian roulette. I thought that his example of some search parties finding what they were looking for, thereby enabling one to ignore not only those which found nothing, but also by inference at least, those which found evidence pointing in another direction, was seriously flawed and demonstrated his own flawed approach. His reference to the pearly gates, even if meant light-heartedly, was a serious blow to his credit as a genuine, expert witness, in my judgment. I would not accept his account of the effect of papers to which he referred unless those papers were available to be checked in full.

Prof. Crawford was a serious expert and an engaging man, but he was clearly a considerable enthusiast for his own theories which he saw in simple terms in a very difficult area of conflicting research. This left me wary. I was not concerned by his theory about the meat of domestic animals when compared with that of the wild animals which our distant ancestors fed on, in part; that ultimately boiled down to the difference in various kinds and quantities of fat in the animals' meat, and I see no reason to doubt Prof. Crawford's expertise on that. I am not, however, prepared to accept his theory of a common denominator between dietary fat, particularly saturated fat, as a cause of coronary heart disease and as a cause of cancer. It is pure theory, unsupported by any kind of investigation, in an area, medicine and particularly

Page 342

oncology, in which Prof. Crawford is not an expert, in my view. It did not make sense to Dr Arnott who is an expert in medicine and particularly oncology, and I am surprised that it has not been picked up with enthusiasm if it has any possible, apparent merit, despite Prof. Crawford's own explanation of why this might be. Indeed I am surprised that Prof. Crawford has not taken the theory farther than he has with his evidence in court and his mention of it at two conferences, if he sees it as a serious, new hypothesis in an area where any possible advance is very important.

Moreover, I could see no sound basis for Prof. Crawford's view that since the science of diet and cancer is where the science of diet and heart disease was many years ago, the former will come to the same conclusion as the latter in due course.

I thought that his arithmetical calculations, attempting to quantify the risk of heart attack or stroke from eating at Mcdonald's, were bizarre for a scientist.

Prof. Campbell is a distinguished man, but I was wary of what I saw as his personal enthusiasm for the benefits of a diet completely free of animal products and the harm done by even small additions of foods of animal origin to an otherwise all plant diet. He endorsed Dr Barnard's 1993 book "Food for Life". He is perfectly entitled to his personal views, of course, and they may have stemmed from his scientific enquiries, but they made it difficult for me to be confident that the views which he expressed as an expert witness were soundly based on what he had found, particularly in China, in an area where there is so much room for interpretative judgment.

Prof. Campbell acknowledged genetics and other non-dietary life events so far as increased susceptibility to degenerative disease was concerned, but he seemed unwilling to accept them as real risk factors in their own right. His view that any death from degenerative disease before the age of eighty-five or ninety was premature put him on another planet from me. I thought that, for such a distinguished scientist in such an important field, he had a rather scattergun approach to his evidence on diet as a cause of "these diseases". Like Dr Barnard he seemed quite unable to attach any weight to material which might cast doubt on his own committed views.

I thought that Prof. Wheelock, Dr Arnott, Prof. Naismith and Prof. Keen were all balanced witnesses so far as the relationship between diet and heart disease and cancers of the breast and bowel was concerned. Dr Arnott, Prof. Naismith and Prof. Keen had no axe to grind or committed view one way or the other.

Page 343

Neither did Prof. Wheelock on this topic, in my view, although he was attacked for being in the pay of the Second Plaintiff. I thought that his evidence on diet and heart disease was particularly candid, and he made no bones about his view that diet was a factor in the development of the cancers.

All four witnesses showed a scientific approach to what had been proved and what was merely plausible hypothesis and to the difficulties in the way of proving that diet or particular elements of diet caused or significantly increased the risk of a particular disease. I thought that Dr Arnott who has devoted most of his professional life to enquiry into the causes of various cancers and to their treatment was the most impressive witness so far as diet and cancer was concerned. He made a strongly favourable impression. He was also the most naturally reserved when it came to the question of what was necessary for proof. I doubt that mere balance of probabilities is enough for his literally and metaphorically clinical mind when it comes to ascribing a cause or causes for an illness, especially when he is giving evidence in public. I believe that he was quite the opposite of Dr Barnard, Prof. Crawford and Prof. Campbell in this respect. I do not, however, consider that he was oversceptical so far as diet and the particular components of diet in the aetiology of disease were concerned, as his evidence on high dietary fat and heart disease showed.

As a general approach in this highly contentions area, I value the opinions of Prof. Wheelock, Dr Arnott, Prof. Naismith and Prof. Keen more than those of Dr Barnard, Prof. Crawford and Prof. Campbell, and I value the opinions of Dr Arnott above all other witnesses so far as the question of diet in the aetiology of cancer in concerned.

Has, then, a diet high in fat (including saturated fat) and animal products and salt (sodium), and low in fibre, been proved to lead to a very real, that is serious, risk of heart disease, cancer of the breast or cancer of the bowel?

There was a degree of agreement so far as heart disease is concerned.

The Plaintiffs admitted that there was a considerable amount of evidence of a causal relationship between a diet high in saturated fat and sodium, and obesity, high blood pressure and heart disease.

Prof. Wheelock went further in saying that there was a high level of probability of a causal relationship between intake of saturated fat and cardiovascular disease.

Page 344

In my view it is inevitable that a U.S. or U.K. diet which is high in both total fat and animal products will be high in saturated fat. High total fat and high animal product content and high saturated fat content generally go hand in hand in the diet in those countries, unlike countries where the high total fat level is largely due to olive oil.

Dr Arnott was convinced by the evidence which associated a high intake of fat, particularly saturated fat, with coronary heart disease.

Prof. Keen was more reserved, but he did see good reasons, related to cardiovascular disease, for reducing fat in the diet to 35% or 30% of calories.

Prof. Naismith was also more reserved, but this seemed to be largely because of the disappointing results of intervention studies on patients who already had a diagnosis of heart disease, where reducing saturated fat intake failed to reduce serum cholesterol levels. This did not seem to me to contradict the evidence that fats rich in saturated fatty acids, mainly of animal origin, raised plasma cholesterol and that high overall plasma cholesterol was predictive of a high risk of cardiovascular disease.

There was a real measure of agreement that saturated fat in the diet influenced serum cholesterol levels which in turn influenced atherosclerosis and also that fats in the blood led to atherosclerosis. To that extent a simple mechanical explanation (as Dr Arnott described it) for a high level of dietary saturated fat increasing the risk of heart disease, was shown.

A diet low in fibre will often mean diet which is low in fruit, vegetables and cereals, whether or not it goes with a diet high in fat (including saturated fat) and animal products, so there is the additional factor that the protective effect of fruit, vegetables and cereals, for which there is growing evidence, perhaps because of their levels of protective anti-oxidants, may well be limited.

Page 245

Then there is the factor that a diet high in fat is palatable and calorigenic and facilitates what Prof. Naismith called "passive over consumption" of energy, leading to weight gain with its increasing risk of heart disease for those who become increasingly overweight. Prof. Keen spoke of the strong correlation between high fat intake and total food energy consumption and obesity, although he saw the matter in less simple terms than Dr Barnard.

To all this must be added the common ground of a very high incidence of heart disease in the U.S. and the U.K., killing marginally more people in the U.K. than all cancers (including lung cancer) put together, according to The Health of the Nation.

Putting all these factors together my own judgment is that a diet high in fat (including saturated fat) and animal products, and low in fibre, sustained over very many years, probably does lead to a very real, that is serious, risk of heart disease in due course.

This conclusion does not help the Defendants to justify the meaning and message of this part of the leaflet because of my finding that it is not true to say that eating McDonald's food, more than just occasionally, might well make your diet high in fat and animal products and low in fibre. It does mean, in my judgment, that the small proportion of McDonald's customers who eat McDonald's food several times a week will take the very real risk of heart disease if they continue to do so throughout their lives, encouraged by the Plaintiffs' advertising to which I will come.

There is evidence that a diet high in salt (sodium) increases the risk of hypertension and therefore cardiovascular disease, but more significantly cerebrovascular disease, in some older or already hypertensive people; but in my view the evidence goes no further than that at this time.

I was not convinced by the evidence in relation to dietary cholesterol, but that is beside the point because the leaflet does not accuse dietary cholesterol.

Sugar, and vitamins and minerals, passed from the scene with my finding that McDonald's food was not generally high in sugar and that it could not rightly be described as low in vitamins or minerals.

Page 346

Turning to diet and cancer of the breast, diet has clearly been identified as one of a number of possible risk factors. The difficulty lies in demonstrating that diet is a risk factor in its own right as opposed to a component in a way of life which is associated with an increased risk of cancer of the breast. An even greater difficulty lies in demonstrating that any particular component of diet is a risk factor in its own right.

The most positive evidence of a direct, causal association between diet, and in particular a diet high in fat including saturated fat and animal products, and low in fibre, and an increased incidence of cancer of the breast, comes from the findings of a number of population studies. This evidence is suggestive of a causal connection between diet, and in particular diet high in fat including saturated fat and animal products, and low in fibre, and a sufficient number of cases of breast cancer to raise the incidence of breast cancer across the population to a statistically appreciable degree.

However, it seems to me to be a false step to go straight from those population studies to the confident conclusion that diet, and particularly diet high in fat including saturated fat and animal products, and low in fibre, causes cancer of the breast or even that it leads to a very real risk of cancer of the breast in a significant proportion of the population, let alone in any particular reader of the leaflet complained of.

There is the general difficulty of extrapolating the extent of risk or increased risk from the findings of population studies to a particular person. Then there are a number of particular difficulties so far as the aetiology of cancer of the breast is concerned.

There is a doubt about the reliability of some of the information which has gone into population studies, particularly so far as the precise amounts of various ingredients of diet are concerned.

Other kinds of studies have produced conflicting results. I bear in mind that the differences in fat intake in studies which have found no positive correlation between fat intake and the incidence of breast cancer is comparatively small and that at higher levels of fat intake the difference in effect may fall off, if Prof. Campbell is correct; but in my view the conflict can not be completely ignored. There are obvious contradictions like the fact that consumption

Page 347

of dietary fat has fallen in the U.S. in the last twenty or thirty years whereas the incidence of breast cancer has increased. This may be due to the time which it takes for the disease to develop, but it is an anomaly.

Perhaps most importantly there is the difficulty presented by the fact that cancer of the breast is a multifactorial disease. That is one point upon which all the experts appear to agree. The result is that the presence of one factor confounds the weight to be attached to another. My education in medical statistics in the course of the trial was extremely limited, but I was left with the strong feeling that none of the allowances made for confounding factors in the epidemiological studies could be said to be so reliable that one could have every confidence in the conclusions to which the various authors came. It is an extremely difficult field because there are so many factors, for instance heredity, frank diet which may vary over the long period which breast cancer takes to develop, or at what may be crucial times of life, body size and weight, age of menarche, age of menopause, the length of time between the two, age at first pregnancy, parity, amount of exercise taken quality of nutrition and overall energy intake. Not only are there so many factors, it is difficult to quantify the degree of effect of diet on those factors which are susceptible to being affected by diet and it is difficult to distribute responsibility between those which are susceptible to being affected by diet and those which are not, because they are purely constitutional or social, for instance.

Honest experts clearly disagree on this matters. My protracted consideration of the evidence which I heard demonstrates that much at least.

Furthermore, although apparently plausible mechanisms for the causative effect of high fat including high saturated fat intake, particularly, and low fibre intake, have been proposed, they have not been confirmed to anything like the level of smoking and lung cancer, or even high fat, including saturated fat, and raised serum cholesterol and heart disease, for instance; and it is difficult to see them being confirmed in the immediate future when animal experiments may be misleading and human experiments or intervention studies face ethical and practical difficulties.

To me, Dr Kinlen's 1987 paper suggesting that the associations with fat intake may be indirect and reflect the role of excess calories or "overnutrition" in breast cancer

Page 348

aetiology reads very sensibly. His 1991 paper carried much the same message. The growing support for a high proportion of fruit, vegetables and cereals in diet may be consistent with this as may the benefits of exercise. Dr Kinlen's suggestion may provide an explanation for Prof. Campbell's findings, even though per capita calorie intake was high in the subjects of his survey.

But even if it is accepted that a high animal fat diet contributes readily to "overnutrition", and even if overnutrition is a front runner as a cause of cancer of the breast, it does not follow that it is the likely cause, and at the end of the day I have come to the conclusion that the state of the evidence is too unclear for it to be proved, on balance of probabilities, that a diet high in fat,including saturated fat, and animal products, and low in fibre, even if sustained for very many years, leads to a significantly increased risk of cancer of the breast, let alone to a very real, that is serious or substantial, risk. It is possible that it increases the risk to some extent, but that is as far as the evidence takes me.

Turning to diet and cancer of the bowel or colon, many of the same difficulties apply as bear on breast cancer. It is multifactorial and heredity is the strongest risk factor.

The evidence is stronger, in some respects, for a diet high in fat, including saturated fat, and animal products and low in fibre being a cause of cancer of the bowel.

There is not the same degree of conflict between population studies and case control or prospective studies. Dr Willett's conclusions from his large survey of U.S. nurses highlights this.

There is an attraction in any theory that diet is a causative factor in cancer of the bowel, because so much of what one eats passes through the bowel, and it would appear that the attraction is not superficial, because there are plausible mechanisms relating to the aggressive and protective qualities of bile acids and bowel chemicals, although I feel unable to find that browned meat is a risk factor for bowel cancer, when the evidence relies upon observation of laboratory animals. None of the witnesses apart from Dr Barnard supported the theory.

A number of papers have supported a protective effect of high fibre intake, including the New York-Kuopio paper where the dietary fat intake was the

Page 349

same in both groups and large came from animals, although New York's came mostly from meat and Kuopio's mostly from dairy products. There are plausible mechanisms for the protective effect of high fibre intake, although they remain unproved, and I accept that diets high in animal fat generally tend to be low in fibre or, at least, not high in it.

The evidence for the protective effect of vegetables was described more favourably by Dr Arnott in relation to cancer of the bowel.

Nevertheless the overall picture still remains unclear, in my view.

In my judgment it has not been proved to the required standard of balance of probability, although it is strongly arguable, that even a lifelong diet which is high in fat, including saturated fat, and animal products, and low in fibre, significantly increases the risk of cancer of the bowel. In any event, the evidence fell short of showing that any increased risk amounted to a very real, that is serious or substantial, risk. It is strongly possible that a sustained diet which is high in fat, including saturated fat, and animal products, and low in fibre, increases the risk of cancer of the bowel, but that is as far as the evidence takes me.

In any event, as I have already said, it is not true to say that eating McDonald's food, albeit more than just occasionally, might well make your diet high in fat, animal products and salt (sodium), and it follows from all this that McDonald's food is not very unhealthy as stated in the leaflet.

The fifth issue in this part of the case, is whether McDonald's (including the First and Second Plaintiffs) knowingly deceive customers by claiming that their food is a nutritious part of any diet, despite knowing it to be very unhealthy as stated in the leaflet. With the failure to justify the allegation that the First and Second Plaintiff's food is very unhealthy in the way described in the leaflet, a large part of the sting has gone from the allegation that they deceive customers by claiming that their food is a useful and nutritious part of any diet. But this last defamatory charge in the meaning of the whole of this part of the leaflet could in my view by partially justified if any substantial deception in McDonald's nutritional information could be proved.

Page 350

The Defendants' case principally relied upon a promotion by the First Plaintiff in the U.S. in 1987 and in various nutritional guides produced by the Second Plaintiff in the U.K. between 1984 and 1985 and 1989.

The Defendants' case was that both Plaintiffs put out material which gave nutritional information and advice but whose dominant purpose was to paint McDonald's food as "nutritious" and of real benefit to the diet, in order to persuade the public to eat more of it, however much they ate already; and that this was misleading and deceived the public.

In relation to the U.S. promotion in 1987, the Defendants called Mr Stephen Gardner who is Dean of Clinical Education and visiting Professor of Law at the Southern Methodist University School in Dallas. He is a licensed attorney. He has a particular expertise in the law and practice of advertising and marketing of the nutritional values of food products. Between November,1984, and December,1991, he was an Assistant Attorney General for the State of Texas in charge of the Dallas regional office, and he had a specific assignment in the consumer protection division. Much of his work involved collaboration with Attorney Generals of other states, New York and California in particular.

Large parts of the evidence of Mr Gardner and of Mr John Horwitz the First Plaintiff's Assistant General Counsel, were devoted to the questions of whether the First Plaintiff was reluctant to issue stores with booklets giving nutritional information when the Attorneys General wanted them, whether it stole a march on the Attorneys General and other companies in a press release claiming credit for the issue of nutrition brochures, and whether it was forced to bring the 1987 advertising campaign to an end prematurely because it was shown to be misleading. I did not find this evidence which was largely inconclusive helpful in deciding whether the campaign was deceptive in fact.

The fact is that the campaign was launched in January,1987. The Attorneys General perceived it to be misleading. The First Plaintiff denied that it was misleading, contending that it gave good nutritional advice and information and said, truly, that McDonald's food could be part of a healthy diet.

Page 351

Both Mr Gardner and Mr Horwitz expressed views on whether the advertising was misleading. Their evidence brought out the arguments but I have copies of the advertisements and I can judge that essential question for myself.

I accept Mr Gardner's evidence that by the mid 1980s American consumers, or a significant proportion of them, wished to eat food that was healthy or nutritious in the sense that it would be of benefit to them and their diets, and that in particular they wished to reduce what they would see as deleterious elements in their diets such as fat, saturated fat, sodium and cholesterol in particular. They needed information about the nutritional content of food choices in order to do this.

The 1987 advertising campaign consisted of six two-page print advertisements inserted repeatedly in various national magazines. Both pages were inserted at first: later only the first page was inserted. A December,1986, letter from Mr Edward Rensi, the First Plaintiff's U.S. President and Chief Operations Officer, to the "McDonald's Family" meaning everyone working with and for the McDonald's system, makes it clear that the purpose of the campaign was to tell McDonald's customers and the public at large that McDonald's food was "wholesome", "nutritious" and "healthful" and that "the McDonald's menu plays an active part in a well-balanced nutritious diet."

Mr Horwitz said that it was intended that it should climb on the back of the success of the brochure of nutritional information which McDonald's had issued not long before, and that it should make people more aware of the brochure.

Mr Gardner and fellow officers took the view that the intent and impact of the campaign was to deceive consumers into believing that McDonald's food was "healthful and wholesome". Mr Gardner said that the advertisements repeatedly used the word "nutritious" in respect of McDonald's food. He thought that to an American consumer "nutritious" food meant "healthy food, to wit food that does not contain significant amounts of the negative nutrients such as fat, cholesterol and sodium."

An advertisement concerning salt said on the first page: "We're careful about the salt in the foods we buy." The second page said that: "Our sodium is down across the menu." Mr Gardner said this was not true because the same advertisement listed four

Page 352

products (regular fries, regular cheeseburger, six-piece McNuggets and vanilla milkshake) none of which had their sodium content lowered in the past year. McDonald's expressed the view that the advertisement was true because items with lower sodium content were spread across the menu. In my judgment it was misleading because it would lead consumers of the popular regular fries, regular cheeseburger, six-piece McNuggets and vanilla milkshake to believe that their salt content was down too.

An advertisement praising "real milk" in McDonald's shakes said that they contained "wholesome milk, natural sweeteners, a fluid ounce of flavouring and stabilisers for consistency. And that's all." There were "no artificial preservatives". In fact McDonald's own ingredient booklet showed that the milkshakes contained artificial flavour, sodium benzoate and sodium hexametaphosphate which are two chemical preservatives.

McDonald's argued that the milkshakest contained flavouring and that it was the flavouring which contained a small amount of sodium benzoate, a preservative. Sodium hexametaphosphate was a stabiliser, not a preservative.

I am unable to conclude that this part of this advertisement was misleading to any significant degree if at all.

It was alleged that an advertisement about "cholesterol and McDonald's good food" emphasised the relatively low cholesterol content of the regular hamburger, but it did not mention the saturated fat content which was alleged by the Attorneys General to be much more relevant to those with cause for concern about heart disease, as I have just found.

It is clear from an advertising agency letter that in November,1986, the Assistant Director of the University of Toronto Medical School was consulted about "the Nutrition Campaign". Apparently he thought that the campaign was "well done, informative, interesting, and timely", but he raised concern about the cholesterol advertisement. He did not believe that fat content was adequately addressed. His view was that cholesterol content should not be divorced from percentage of calories from fat. Health professionals recommended that no more than 30-35% of calories should come from fat and this was not addressed.

The text of the "cholesterol advertisement" did not in fact

Page 353

draw attention to the cholesterol content of the regular hamburger in the text. The text said that there was general consensus among nutrition experts that you should consume no more than 100mg of cholesterol for every 1,000 calories of food, not to exceed 300mg per day. It went on to say that a regular order of fries contained "only 9mg of cholesterol, only 4.6g saturated fat, and a total of only 220 calories", before going as to speak of the 100% cholesterol-free shortening used to cook some products. It said: "Avoiding too much fat and cholesterol doesn't mean "never eat" certain foods. It's the total amount of fats and cholesterol in your diet that matters". It then referred to a chart of nutritional information of the kind "available for all our menu items to help you when ordering at McDonald's". This gave the 9mg of cholesterol for 220 calories in regular fries and it also gave 29mg of cholesterol for 263 calories in a regular hamburger (which is in fact 110g per 1000 calories), 47g for 453 calories in a Filet-O-Fish Sandwich (104g per 1000 calories) and 31g for 352 calories in a vanilla milk shake (88g per 1000 calories). All of those figures fall within or near to the recommended 100g per 1000 calories.

The same chart gave figures for total fat, but not saturated fat, and they were given in grams not percentages of energy (calories). Recommendations for intake of fat and saturated fat were not set out. If, but only if, one knew that one gram of fat produced nine calories could one work out that the total fat in the regular fries was 37% of calories (12g:81 calories out of 220). In the regular hamburger it was 37%. In the Filet-O-Fish Sandwich it was 54% and in the vanilla milk shake it was 20%.

One could work out the percentage of energy from saturated fat in the fries because the saturated fat content of 4.6g was given in the text. So the percentage of energy was 18.8% (4.6 x 9 = 41.4 calories out of 220). But one would have to do the sums. The saturated fat content was not given for the regular hamburger, the Filet-O-Fish Sandwich or the vanilla milk shake.

In fact more detailed nutritional information in "McDonald's Food The Facts" of April,1987, shows 11.3g of total fat including 4.43g of saturated fatty acids for the 263 calories in the regular hamburger (38.6% and 15%). Saturated fat percentage of energy for the Filet-O-Fish Sandwich was 11.5% (5.56g x 9 = 50 calories out of 435).

So the fries and regular hamburger were a bit above the 30%

Page 354

for total fat which had been recommended by NACNE in this country and which was recommended by the WHO three years later. The Filet-O-Fish was well above. The fries and regular hamburger were high in saturated fat, but the Fillet-O-Fish Sandwich only marginally so.

I cannot find U.S. Government recommendations prevailing at 1987. One of the advertisements refers to the U.S. Departments of Agriculture and Health and Human Services Dietary Guidelines for Americans, but the U.S. Surgeon General's Report on Nutrition and health says that those Guidelines made recommendations, including "avoid too much fat, saturated fat, and cholesterol", in general terms. The Surgeon General's Report itself in 1988 recommended most people to "reduce consumption of fat (especially saturated fat) and cholesterol, at a time when the mean daily intake of fat was 36% of energy for men and 37% for women and the saturated fatty acid figure was 13% for both. Prof. Crawford told me that the National Cancer Institute's 1989 U.S.A. dietary guidelines included advice to reduce total fat intake to 30% or less of calories. I do not recall being taken through any recommendations of specific percentages, prevailing in the U.S.in 1987 but in my view despite this and despite the advertisement's reference to total amounts of fat in the diet being what mattered, the cholesterol advertisement must have left a lot of readers with the impression that McDonald's food met dietary recommendations, when in my view it made it more difficult rather than easier to avoid the Government's guideline to "avoid too much fat, saturated fat". A "leaner beef" advertisement certainly did that in my view, saying that McDonald's used beef which was 25% lower in fat than the kind of beef which most people bought. Only if you knew how to do the sums could you calculate from the chart in the advertisement that the McD.L.T. had total fat which was 58% of calories, and that a Quarter-Pounder with Cheese and a Big Mac had 55%. Their saturated fat contents were not given.

Various of the advertisements commended balance from eating a variety of foods", with "more about your balanced diet and McDonald's good food", "more about what we're all about and McDonald's good food", and "more about calcium and McDonald's good food", and "leaner" hamburger beef with no more than 22.5% fat. One illustration had different McDonald's food products balanced one on top of another. In my view that implied that you could eat a balanced diet at McDonald's. Associated text actually said that variety in the McDonald's menu meant that the customer

Page 355

could balance what he ordered. It went on to tell him to balance his McDonald's meal with other meals which he ate, a message repeated in the advertisements, but to my eye that message got rather lost in the claim for the variety and balance in McDonald's products. What I accept to be good nutritional advice and information was mixed with compliments to McDonald's food. The tables in some of the advertisements gave quantities of nutrients in grams for a variety of products. As in the cholesterol advertisement they included fat but not saturated fat. One referred to there being U.S. government guidelines for a healthful diet, the most important being to eat a variety of foods. It said that McDonald's measured up to these dietary guidelines by serving many variations within the basic food groups. "We serve meat, fish, poultry and eggs; dairy products and grains".

I was shown an 18th March,1986, confidential memorandum from an advertising and communications agency working for McDonald's which suggested "talking 'moderation and balance'" because McDonald's could not really address or defend nutrition. "We don't sell nutrition and people don't come to McDonald's for nutrition". The campaign did talk moderation and balance, but it also, in my view, tried to sell nutrition and to get people to come to McDonald's for nutrition. The overall impact of the advertisements together was quite clearly to give the consumer the impression that he would be doing himself a good turn, so far as his health and nutrition were concerned, by eating at McDonald's. The First Plaintiff must have known that and seen it as a selling point in the current mood of interest in healthy eating.

In my view it needed a lot of gall to paint McDonald's food in such a beneficial light, to refer to salt and cholesterol recommendations in precise figures, to refer to recommendations to "avoid too much fat [and] saturated fat" but not to tell the consumer how to relate the total fat figures given in the charts to the recommendations for total fat and not to give saturated fat contents at all, save in respect of regular fries, thereby, in my view, completely sidelining the soundest criticism of McDonald's food, that a significant number of its menu items are high in fat and saturated fat. The fact that, as Mr Horwitz said, charts in stores gave full nutritional breakdowns, did not heal the partiality of the advertising campaign.

Mr Horwitz said that any claim that McDonald's acted with an

Page 356

intent to deceive was false and totally unwarranted, but in my view anyone who knew something about nutritional issues at the time, including the perceived benefits of keeping down dietary fat, saturated fat, cholesterol and sodium, must have appreciated that the campaign stressed the lowered sodium and cholesterol in some items without drawing attention to the continuing high fat and saturated fat. It may be said that "that's advertising", but the advertisements purported to provide a public service by giving nutritional advice and information. Mr Horwitz did not purport to be an expert on nutrition. In fact none of the First and Second Plaintiffs' employees called to give evidence, were.

The Second Plaintiff's publications, upon which the Defendants particularly relied on the fifth issue, were commented on by Mr Peter Cox. Mr Cox's central core expertise, as he put it, is in marketing. He founded and managed an advertising agency before becoming the first Chief Executive of the Vegetarian Society of the United Kingdom. He is now a writer and researcher with a special interest in the marketing of food, particularly meat products, to the general public. He frequently lectures to specialist audiences on the deceptive marketing of meat products. He has produced a number of books, including the 1994 "Realeat Encyclopedia of Vegetarian Living".

Mr Cox's stand was that purveyors of food like the Plaintiffs were under no obligation, save for some labelling and hygiene regulations, to give nutritional advice, but if they did so as the Plaintiffs had done, they should tell the whole truth. In his view the Plaintiffs' literature was frequently very one-sided, stressing the positive elements of their products while omitting or minimising the negative aspects. They practised unethical marketing by wrapping their marketing up as nutritional information and concern about the health of their customers. It was deceptive.

I was happy to hear Mr Cox's view of matters, but most of his points stood or fell according to the view I took of whether McDonald's food was unhealthy according to the meaning which I have found, measuring the answer to that question against the terms of the McDonald's leaflets to which Mr Cox drew my attention.

The Second Plaintiff's booklet "McDonald's Food: The Facts" produced in July,1989, which Dr Lobstein said he had some difficulty getting hold of when it came out, started with "A Note

Page 357

to Our Valued Customers" which contained the sentences:

    "Our food is of the highest quality; our ingredients prove it. At McDonald's we believe food should do you good as well as taste good. That is why we have based our menu on good nutritious foods".

But, said Mr Cox: "Healthy, good nutritious foods will be quite low-in-fat, specially low in saturated fat. They will be from fresh products, and they will be predominantly of non-animal origin: in other words, lots of vegetables, lots of fruit, particularly with particular reference to green leafy vegetables. They are especially healthy-promoting. ...... Those are exactly the sort of foods that are conspicuously not on McDonald's menu items".

A later page of the same booklet was headed "What Nutrients Do We Need". It had a table of "Macronutrients" which included "Fats" divided into "Saturated Fatty Acids (Saturates) Unsaturated Fatty Acids (Macronutrients and Polyunsaturated)", all in large print, which were said to be needed for (and then in small print) energy and as an essential component of membrane and other tissues and as a carrier for fat soluble vitamins A, D, E and K. There was a note in equally small print that "saturates tend to raise the blood cholesterol level while polyunsaturates lower it", and that "too high an intake of saturated fat may increase the risk of heart disease".

A later page headed "What Is A Healthy Diet?" referred to "Balance" with the advice to eat the right amount of energy, eat less fat (particularly saturated fat), eat more fibre and watch intake of sodium (salt) and sugar.

Page 358

Charts at the back of the booklet had figures for the fat and saturated fat content of menu items, but the fat and saturated fats were expressed in grams which Mr Cox thought to be pretty meaningless to the ordinary reader. They would all seem to be small amounts. The conventional, more widely adopted way of giving information concerning fat and saturated fat content was as a percentage of energy (k calories or k joules). If one gave the percentages they could be compared with dietary recommendations. The charts gave the k calories and k joules of each menu item but one had to do some sums to work out the percentages. Also if the percentages were given it was easier to compare different foods.

Mr Cox next complained about the Second Plaintiff's "The Facts about Eating at McDonald's", produced in February, 1985, which said: "At McDonald's we believe food should do you good as well as taste good" and "Whatever you choose to eat at McDonald's you can always be sure of great taste, nutritional value and quality". A page headed "What you get out of it" had a chart of menu items with figures for energy, carbohydrate, fat, protein and sodium/salt, but no mention of saturated fat at all. Nor did it give fibre content. The facing page said that "a glance at the chart will tell you what's in a McDonald's menu item". A page headed "Golden rules for healthy eating" spoke of "Variety" and said that "A McDonald's menu offers you a wide choice" which Mr Cox said it did not, and included oils and fats, sweets and chocolates among the variety of food groups. I note that the same page did advise to keep an eye on how much fat, sugar and salt one ate. It also said: "Naturally, to maintain a balanced diet you should eat plenty of cereals, fresh vegetables and fruit". This was followed by: "But McDonald's menu contains ingredients from all the food groups".

The Second Plaintiff's "McDonald's Nutrition Guide" of July, 1985, had the same page of "Golden rules for healthy eating" and the same "What you get out of it" chart, with no saturated fat or fibre column, and the "glance at the chart" comment. So the reader, Mr Cox said, would think that the chart told him all he needed to know.

Of course all those publications were some years ago; but McDonald's "Nutrition: A Question of Balance", revised January, 1994, said that: "At McDonald's we have a responsibility to help our customers eat a healthy balanced diet .... The McDonald's view is that food should do you good as well as taste good .... We will only serve our customers food of the highest standards of quality, nutrition, hygiene and food safety .... hamburgers which are 100% pure lean beef .... To help all our customers eat healthily, we are constantly making our menu even more nutritious. Our first aim is to reduce the amount of fat in same menu items, whilst not affecting the quality or the taste".

The text ended with: "Every time you eat at McDonald's you will be eating good, nutritious food.

Variety is the order of the day. If you eat a Big Mac and French Fries regularly, you

Page 359

should ensure that at other times you eat different types of foods so that your diet is balanced and you are eating plenty of minerals, vitamins, protein and fibre. Think about variety when planning your meals.

The accompanying table will help you to select the best McDonald's meal for yourself".

The table had figures for energy in k calories and k joules and protein carbohydrate, fat, sodium and fibre all in grams. The fat column had two lines of figures, the first of which was obviously total fat. The second was "Of which Saturates". This heading is in print so small that I did not notice it at first, although anyone who knew about nutrition would realise that a second column of smaller figures would refer to saturates, in my view.

A larger 1993 fold-out "McDonald's Nutrition Information" had saturated fat, again just in grams, but more easily legible.

Dr Barnard said that McDonald's advertising, presumably in the U.S., had been misleading regarding the fat content of its food. The First Plaintiff announced the arrival of the McLean Deluxe burger by claiming that it was 91 percent fat-free. McDonald's was reporting the fat content by weight. But when dietitians or scientists measure the fat content of foods, they are not interested in the percentage by weight, because water content strongly affects the measurements. The relevant measurement is the percentage of calories from fat. Using that standard, the McLean Deluxe burger pattie itself was 49 percent fat, due to its main ingredient being ground beef. The bun and pickle and toppings diluted the burger down to 29 percent fat which was lower than the regular burger, but the 91 percent figure was misleading.

Prof. Wheelock was asked what "nutritious" meant, as used by McDonald's literature to describe McDonald's food. At first he said that it meant something containing nutrients. Then he accepted that most people would think it meant something of nutritional benefit: something which would do them some good.

Both Prof Wheelock and Dr Arnott commented on the accuracy of parts of the Plaintiffs' publications, to the effect that the nutritional advice given was fair and sound. For instance Dr Arnott thought that the Second Plaintiffs

Page 360

golden rule for healthy eating advising variety and moderation were good general advice. Prof Wheelock's reaction to a section of the First Plaintiff's Annual Report on Food and Nutrition, which recommended the basic principles of Balance, Variety and Moderation, was that he could not put it better himself.

In my view the overall impact of the Plaintiffs' publications in the U.K. and the U.S., to which I have referred, to create for the ordinary reader the impression that McDonald's food was positively good for him, not just in the sense of giving him some needed energy intake or some protein, some fibre, vitamins and minerals, but in the broader sense of being a positively useful contribution to a healthy diet. Why else should McDonald's go to such promotional lengths to trumpet dietary advice while describing their food in complimentary terms, including "nutritious" which must impute something more than "containing nutrients" which is true of every food.

Both Plaintiffs must have known that this would be the impact of the material to which I have referred. Certainly Mr Rensi said so in his December,1986, letter. I am not prepared to say that there was not a real element of public benefit in the provision of nutritional information to which I have referred. But I consider that the way in which it was presented in the publications to which I have referred demonstrated its main purpose to be marketing, it pretended to a nutritional benefit which its food, high in fat and saturated fat, and sodium and low in fibre at the relevant time, did not match.

The Defendants also relied on some of McDonald's television advertising. Although it came well after the publication of the leaflet, which was relied upon, it might be thought to show a consistency of purpose.

In some Ronald McDonald TV advertisements he appears with a Rock band called "Ronald and the Nutrients". Mr Green, the First Plaintiif's Senior Vice President of Marketing, said that the point of television advertisements featuring Ronald and the Nutrients was to tell children what the basic food groups were. It was not to associate McDonald's with nutritious food in children's minds. I can not accept this. In one advertisement there is a flashing sign in the background, reading "nutrients." Protein sings: "You need a lot of me to keep you growing", as he bangs a drum with "McDonald's" painted on its side. The advertisement was clearly a vivid message associating McDonald's with healthy food.

Page 361

The most substantial point made under this head was that McDonald's regularly associated itself with sporting figures and sporting events in its advertising, thereby, it was suggested, giving a false impression that eating McDonald's food would encourage good health, fitness, strength and sporting ability.

A number of advertisements in the U.S.A featured Michael Jordan, a famous, fast, tall, strong, well-coordinated basketball player. Mr Green said that Michael Jordan had been associated with McDonald's in the U.S. for eight to ten years, almost since he started playing basketball professionally. He was shown eating a particular McDonald's concoction which, Mr Green said, he had come up with himself. It was suggested that it was a particularly unhealthy concoction, but there was little specific evidence about that.

Miss Dibb, the Defendants' main expert on children's advertising, thought that the advertisements featuring Michael Jordan, "an elite athlete", would associate the product with sporting achievement and that the association was misleading.

Mr Green said that the point of featuring Michael Jordan was merely to further McDonald's image by associating itself with a very popular person. It was not done in order to create the impression that eating McDonald's food led to physical or sporting excellence.

The same comment applied to advertisements like "Olympic Hopefuls" and McDonald's Olympic sponsorship. Mr Green said that the object of this was to associate McDonald's with the Olympic spirit, not to attempt to gain a healthy image.

Mr Green said that McDonald's had associations or alliances with a number of sporting bodies. Customers enjoyed sports in the U.S. and it showed that McDonald's enjoyed and had the same likes and values as they did. It was not trying to tell people that McDonald's food was healthy food.

On the other hand Mr Hawkes, the Second Plaintiff's Chief Marketing Officer, appeared to accept that part of the purpose of advertising which associated McDonald's with the World Cup was to associate McDonald's with fitness and vigour. One of the aims was that people might think that McDonald's food was healthy.

In my view Mr Hawkes showed greater candour than Mr Green

Page 362

in this respect. No doubt there is a large element of reflected glory and topicality a "feel-good factor",involved in McDonald's associating itself with sporting stars and major sporting events, but I have no doubt that an equally important aim was and is to associate McDonald's food with healthy, physical excellence, so far as customers are concerned, although this might well be lost on children. Certainly, that was my first reaction to the advertisements in question.

I was also referred to a printed advertisement featuring a character called Slugger who shows off his biceps in praise of the meat group and the suggestion that two servings a day can make it easier to do things like climb higher and ride your bike further, for which there appeared to be no scientific basis.

In my view McDonald's officers have always thought of their food as healthy and this is what nutritious means to them. That was the theme of Mr Rensi's December,1996, letter introducing the 1987 campaign to the McDonald's Family. Mr Beavers said that McDonald's food was nutritious and promoted as such, and I think he meant that it did you positive good. It was argued for the Plaintiffs that there was no such thing as a healthy or unhealthy food: only diets could be healthy or unhealthy.

But over the period with which I am concerned, it became no longer good enough, in my view, to describe food products as positively beneficial just because they were made with natural ingredients. I was shown an extract from the U.S. Food and Drug Administration's magazine "Consumer" of September,1994, which said: "A 'healthy' food is low in fat and saturated fat, contains limited amounts of cholesterol and sodium, and is a good source of one or more of six important nutrients, under federal regulations issued last May". Mr Green said that the regulations were going into effect over a period running into 1997. They would affect all advertising of food as "healthy". In my view what has now apparently become U.S. law reflects what must have been good sense as long ago as the mid 1980s. The various advertisements, promotions and booklets to which I have referred have pretended to a positive nutritional benefit which McDonald's food, high in fat and saturated fat, animal products and sodium, and at one time low in fibre, did not match.

To this extent the last charge in this part of the meaning of the leaflet complained of has been partially justified.

Page 363

My eventual conclusions and findings in this part of the case are accordingly as follows.

At the material time of publication of the leaflet between September,1987, and September,1990, McDonald's food was high in fat (including saturated fat) and salt (sodium) and animal products and it has continued to be so. It was low in fibre at the material time of publication of the leaflet, but it has not been proved to be so now. It has not been shown that McDonald's food generally is high in sugar, although some individual times are. It has not been shown that McDonald's food is low in vitamins or minerals.

It is not true to say that eating McDonald's food albeit more than just occasionally, might well make your diet high in fat, animal products and salt (sodium), let alone sugar, or low in fibre, let alone vitamins and minerals. Such a statement is not justified. It is not true in substance and in fact because it is only true (so far as fat, animal products, salt and fibre are concerned) in relation to a small proportion of people who eat McDonald's food several times a week. The leaflet does not say that if you eat McDonald's food several times a week it might well make your diet high in fat, animal products and salt (sodium), and low in fibre. It leads the reader to believe that this is to be expected from anything more than just the occasional McDonald's meal.

It follows that it cannot be right to say that eating McDonald's food will bring the very real risk that you will suffer cancer of the breast or bowel or heart disease as a result of making your diet high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, even if such a diet carries such a risk.

Nevertheless I have gone on to consider whether it has been proved that a diet high in fat (including saturated fat), animal products and salt (sodium), and low in fibre leads to a very real, that is serious or substantial, risk of heart disease or cancer of the breast or cancer of the bowel.

In my judgment a diet high in fat (including saturated fat) and animal products, and low in fibre, sustained over very many years, probably does lead to a very real risk of heart disease in due course.

Page 364

This conclusion does not help the Defendants to justify the meaning and message of this part of the leaflet because of my finding that it is not true to say that eating McDonald's food more than just occasionally might well make your diet high in fat and animal products and low in fibre. It does mean, in my judgment, that the small proportion of McDonald's customers who eat McDonald's food several times a week will take the very real risk of heart disease if they continue to do so throughout their lives, encouraged by the Plaintiffs' advertising to which I will come.

It has not been proved, on balance of probabilities, that a diet high in fat, including saturated fat, and animal products, and low in fibre, even if sustained over very many years, leads to a very real risk of cancer of the breast in due course, although it is possible that it increases the risk to some extent.

It has not been proved, on balance of probabilities, that a diet high in fat, including saturated fat, and animal products, and low in fibre, even if sustained over very many years, leads to a very real risk of cancer of the bowel in due course, although it is strongly possible that it increases the risk to some extent.

It follows that McDonald's food is not very unhealthy as stated in the leaflet.

However, I do find that various of the First and Second Plaintiffs' advertisements, promotions and booklets have pretended to a positive nutritional benefit which McDonald's food, high in fat and saturated fat and animal products and sodium, and at one time low in fibre, did not match.

Save in this last respect, the defamatory meaning and message of the leaflet to the effect that McDonald's food is very unhealthy because it is high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, and because eating it (more than just occasionally) may well make your diet high in fat, sugar, animal products and salt (sodium), and low in fibre, vitamins and minerals, with the very real (that is serious or substantial) risk that you will suffer

Page 365

cancer of the breast or bowel or heart disease as a result, and that McDonald's know this but they do not make it clear, and that they still sell the food and they deceive customers by claiming that their food is a useful and nutritious part of any diet, is not justified.

Had the leaflet said that in order to do his or her best to keep any risk of suffering heart disease or cancer of the breast or cancer of the bowel to a minimum the reader would be well advised to keep consumption of fat, particularly saturated fat, and animal products down and consumption of fruit, vegetables and cereals up, should take a fair amount of exercise and, above all, avoid smoking, and had it gone on to say that if the reader ate McDonald's meals several times a week indefinitely, without taking a lot of care with the rest of his or her meals, there would be a very real risk of heart disease in due course and that it was possible that there would be some increase in the risk of cancer of the breast and cancer of the bowel also, and had it concluded that the reader would, therefore, be well advised to ignore any McDonald's advertising which might appear to encourage him or her to eat McDonald's meals several times a week, then, in my view, McDonald's (the Plaintiffs) would have had no just cause for complaint. But that was not the message which the leaflet gave. Its actual message of a very real risk of heart disease or cancer of the breast or cancer of the bowel simply from eating at McDonald's more than just occasionally for any period, and of a cover up of that fact, is not justified.

Page 366

[ previous page ] [ next page ]