It is by now a received truth that the world, or at least a good portion of it, is in the grip of an unprecedented obesity epidemic which threatens a health catastrophe. Speaking at an EU conference on obesity in early 2005, Professor Philip James, chair of the International Obesity Task Force, noted that obesity ‘took off in the 1980s and looks as if it was accelerating in the last five to 10 years. It’s beginning to look as if we have an exponential rise’. And Professor James is not alone in his worries. The US surgeon general has claimed that obesity, at least for Americans, is a ‘greater threat than weapons of mass destruction’, a threat which could only be solved by a ‘cultural transformation’.

The sources of this epidemic are variously identified as fondness for sloth, the overabundance of cheap food, the decline in physical activity, the inevitable corruptions accompanying postmodernity, the absence of manual labour, and the development of technology, particularly the internet. However, for the political and policy-making class, the cause of the epidemic is much more likely to be found in an increasingly fat-friendly environment, an environment engineered by the food and drink industry. Though there is a passing nod to personal responsibility in this tale of growing obesity, since obviously eating is a voluntary act, the major claim in the story is that the primary cause of obesity is to be found in the manufacturing and marketing practices of something increasingly referred to as Big Food.

Not surprisingly given this diagnosis, the prescription proposed to halt the growth of obesity focuses on government regulation and control: controlling food content and the conditions of selling food, controlling the pricing of food, and, perhaps most significantly, the marketing of food. For instance, the Food Standards Agency is considering warning labels for certain unhealthy foods, while at the same time pressuring the food industry to change the constituents of some of its most popular foods. In France, the French food standards authority, Afssa, has argued that snacking by children must be discouraged, and new regulations will provide for fines for food makers who advertise ‘junk foods’ without health warnings. In June 2005, the British Medical Association recommended a total ban on advertising ‘unhealthy’ food to children.

In November 2005, the Department of Health published an advisory document which argued, in a move reminiscent of arguments about cigarette advertising, that celebrities and cartoon characters should be prohibited from advertisements that promote ‘unhealthy’ foods to children. Ofcom, the UK’s broadcast regulator, has now implemented similar restrictions on food advertising. The publication of the EU Green Paper ‘Promoting Healthy Diets and Physical Activity’ in December 2005 also raises the question of whether involuntary restrictions on food advertising are required. This follows an April 2005 report by the European Heart Network (EHN) which concluded that children require protection from food advertising and recommended a ban on TV advertising of ‘unhealthy’ food to children.

The seemingly universal message, then, is that: obesity is a disease rather than a moral failing; that very many of us are much too fat; that our extra pounds reduce our life expectancy; that our extra pounds are largely the result of a food environment produced by the food industry; that weight loss is necessary, achievable, sustainable and associated with significant reductions in mortality; that the public health community should assume a leadership role in tackling this epidemic through a variety of largely regulatory interventions; and that failure to deal with the obesity epidemic, especially amongst children, portends an unprecedented public health disaster (approaching the Black Death, according to the director of the Centers for Disease Control and Prevention).

While we are not apologists for obesity, and believe that it represents a genuine risk for the morbidly obese, we find that the case against it is significantly, and more disturbingly, deliberately flawed. Not only are the claims about an obesity epidemic, particularly an epidemic of childhood obesity, often wildly exaggerated, the science linking weight to unfavourable mortality outcomes is frequently nonexistent or distorted. Moreover, not only is the causal account of the sources of obesity, an account which roots obesity in an environment shaped by Big Food, at odds with the best evidence, but the policy prescriptions for solving the obesity crisis are most likely ineffective, while at the same time threatening a fundamental assault on core democratic values.

Specifically, our objection to the obesity epidemic story focuses on three areas:

* The absence of scientific evidence to support the main components of the story, namely that childhood obesity is increasing at an alarming rate, that overweight and obesity increase one’s mortality risks, that the current Body Mass Index (BMI) recommendations represent population ideals, that the overweight and moderately obese should be encouraged to lose weight because such loss will improve their health, and that one of the major causes of obesity is the consumption of inappropriate foods fostered by the marketing activities of the food industry;

* The lack of substantial evidence to suggest that the significant government interventions urged to stem obesity will in fact work, and the considerable evidence which suggests that many policies advanced will be counterproductive;

* The significant risk that many of the policy proposals to prevent obesity will further erode individual liberty by enhancing the state’s power exclusively to define what constitutes a good life.

The evidence

It is a basic requirement of good public policy that the scientific basis for disease interventions be clear and compelling. The assumptions about obesity, diet, disease and mortality, as well as the evidence, however, are unable to support the obesity epidemic story. For instance, it is now clear that overweight and even modest obesity are not associated with increased mortality risks, and that the claims that obesity represented a threat to health comparable to tobacco are unsupported by the scientific data (Flegal 2005; Gronniger 2005). To take but one example, the 2004 US Centers for Disease Control and Prevention (CDC) study, which claimed that there were nearly 400,000 annual deaths attributable to diet and physical inactivity, was discredited in April 2005 by a new study from researchers at the CDC and the National Institutes of Health which put the figure of annual deaths from overweight and obesity at just 25,814 (Flegal 2005).

It is also clear that the claims about rising childhood obesity, whether in the UK or North America, are very much exaggerated, that the levels certainly do not reach epidemic proportions, and that childhood obesity does not lead to greater health risks in adulthood (Wright 2001). For instance, as the Social Issues Research Centre pointed out, the Health Survey for England, published by the Department of Health in December 2004, found that the average weight of boys (aged 3-15) in 2003 was 31.9kg compared with 32kg in 1995. The average weight of girls was 32.4kg in 2003 and 32kg in 1995. In 2003, the average 15-year-old boy weighed 60.7kg compared with 58.8 kg in 1995. Again, from 1995-2003 there was an increase in average BMI for boys of 0.5kg and 0.6kg for girls. Clearly these changes in UK children’s BMI fail to support Professor James’s alarmist claim about an exponential increase in obesity, as the Health Survey data does not suggest an accelerating rate of increase.

Similar data is found in the United States. According to the most recent data, the prevalence of overweight and obesity in US children showed no statistically significant increase from 1999-2002 (Hedley et al. 2004). In fact, caloric intake for US children and young people has not changed significantly in recent years (Troiano et al. 2000). Similarly, in the UK, the latest National Diet and Nutrition Survey in 2000 found that since the last survey in 1983, energy intake in both boys and girls aged 4-18 had declined.

Equally unsupported is the claim about an epidemic of adult obesity. While there has been significant weight gain amongst the very heaviest segment of the population (Friedman et al. 2002), this has not been true of most of the individuals who are labelled as overweight and obese, whose weights are only slightly increased. According to Hedley et al. (2004), for instance, there has been no statistically significant increase in adult weight in the US from 1999-2002. Indeed, the abrupt change in the BMI classification of overweight from 27 to 25 resulted in millions of Americans who had previously been classified as of normal weight suddenly being described as overweight.

Even more crucially, the critical claim that there is a link between excessive food intake and childhood obesity is highly controversial. As Muecke et al. (1992) noted: ‘Though some studies support the contention that overweight children over consume food, others contradicted this widely-accepted notion. Comparisons of obese adolescents to normal peers have demonstrated comparable energy intake and nutrient distribution’. A cross-cultural review of obesity in children and adolescents in the US, France, Australia, Britain and Spain found little evidence to support the claim that overweight and obese children and adolescents consume more calories than others (Rolland-Cachera and Bellisle 2002). Indeed, the authors note that some studies find that these children may in fact consume less energy than their thinner peers.

Then too, the entire idea of ‘good’ food and ‘bad’ food, and the link between a particular diet and reduced morbidity and mortality, is very much open to question. A just-published series of studies, for instance, found that a low fat diet has little effect on reducing the risk of breast cancer, colorectal cancer or cardiovascular disease in postmenopausal women. Comparative studies looking at the fat and blood cholesterol levels across different cultures have failed to sustain the claims about a cause and effect connection between life expectancies and diets. Crete, for instance, with its ‘healthy’ Mediterranean diet has one of the lowest incidences of heart disease yet has a fat intake of 40 per cent, which is close to that of the UK. The Netherlands, which has one of the highest life expectancies in Europe, has a fat intake of 48 per cent. This compares to the Masai of East Africa who have very low levels of blood cholesterol yet derive some 66 per cent of their calories from fat. About all that might be said with reasonable certainty about the obesity-diet-mortality connection is that, as the late epidemiologist Petr Shrabanek observed, ‘people who eat, die’ (Skrabanek 1995).

Equally unsupported are the recommendations, based on the assumed connections between overweight and modest obesity and premature mortality, for population-wide weight loss. For example, the US National Heart, Lung and Blood Institute advises overweight individuals, regardless of their current weight, to lose 10 per cent of their weight. These recommendations not only fail to take into account the extensive evidence that sustained weight loss is extraordinarily difficult for very many people, but ignore the evidence of an association between weight loss and increased mortality. They effectively counsel a weight loss programme for the entire population, which not only has no scientifically established benefit but also carries substantial risks for many individuals. Indeed, it might be argued that such unfounded and reckless health advice constitutes malpractice.

The evidence supporting the claim that advertising is the cause of fat children and adults also looks highly suspect. UK ad-spend on food and drink has actually been falling in real terms for the last five years. It now constitutes 18 per cent of all TV advertisements, compared to 34 per cent in 1982. Across the Atlantic, according to the Federal Trade Commission, advertising during children’s TV programming has declined by 34 per cent in recent years. Indeed, despite claims about the influence of TV food advertising aimed at children, UK children watch less television daily than any other group in the population.

Again, there does not appear to be compelling evidence that even heavily advertised foods like breakfast cereals and fizzy drinks are making children obese. Epidemiological studies do not support the claim that dietary fat causes obesity in kids (Ebbeling et al. 2002). Nor are there any clinical trials that validate the theory that an increase in carbohydrate consumption through things like breakfast cereals, sweets and fizzy drinks leads to overeating and obesity. Equally importantly, the scientific literature contains numerous studies that count against the claim that such heavily advertised carbohydrates lead to fatter kids. A Harvard study, for instance, last year looked at 14,000 school children and found that however junk food was defined, with or without fizzy drinks, it did not lead to obesity in children (Field et al. 2004).

Then too there are extensive and significant econometric studies which have examined the connection between food advertising and the size of the food market, both in Europe and the US (Duffy et al. 1999). These studies have found, for the most part, that while food advertising may influence the consumption of food brands, it does not increase either total food consumption or food category consumption. These findings constitute a near fatal flaw in the claims that advertising causes childhood obesity, since to hold advertising responsible for obesity they would need to show in a convincing fashion that it influences not simply brands, but either entire diets or at least parts of diets.

Finally, the suggestion that advertising restrictions or bans, voluntary or involuntary, might reduce childhood obesity is contradicted by some real-world evidence. Since 1980, the Canadian province of Quebec has banned all food advertising to children, yet childhood obesity rates and consumption of so-called unhealthy food are not substantially different there from other parts of Canada. Then too, for the last decade Sweden has also had a food advertising ban for children, but this has not resulted in significant reductions in childhood obesity or marked differences in obesity rates compared with other European countries (Ashton 2004). As even the World Health Organization’s review of food advertising to children conceded, there is a ‘lack of objective research into the effects of regulation on dietary patterns and longer term health’ (Hawkes 2004).

Efficacy

The second problem with the obesity epidemic story centres on the question of efficacy, namely, will the measures being suggested actually work? Or, just as crucially, will they work without unforeseen counter-productive effects? One of the more unfortunate consequences of the exaggerated claims about an obesity epidemic is that such talk, both in the media and amongst politicians and policymakers, has produced an unwarranted sense of alarm and urgency that works against the careful consideration not only of what the evidence of harm really shows, but where that evidence actually points in terms of solutions. Interventions to ‘solve’ the obesity epidemic are increasingly accorded the privileged position of self-evident truths, rather than being subjected to a rigorous process of examination which asks whether there is a fit between the problem and the proposed remedy, and which leads to winnowing out the misguided, the ineffective, the wasteful and the positively harmful.

For instance, will the solutions to obesity really change not only the eating habits of individuals but of entire populations so that they will be, if not thin, then at least less fat? The numbers do not look promising when out of every 100 people who diet, only four are able to maintain their post-diet weight. As the authors of one study concluded: ‘Obesity must be recognized as a chronic condition for which no cure can reasonably be expected’ (Serdula et al. 1999).

Several studies that have looked at dieting - whether through commercial programmes or through self-help, diet types (low carbohydrate, high protein, high fat vs. low calorie, high carbohydrate, low fat) and weight loss counselling - have concluded that attempts at weight loss are largely unsuccessful, even in highly controlled situations. Describing the results of one study that compared low carbohydrate and low fat diets, the researchers concluded that ‘Adherence was poor and attrition was high in both groups’ (Foster et al. 2003). More worrisome yet are the health risks which accompany dieting. Several studies have reported that weight loss is associated with increased mortality (Nilsson et al. 2002; Wedick et al. 2002).

Similar questions of efficacy surround the policies advanced to counter childhood obesity, including changing school food, removing vending machines with ‘bad’ food and increasing physical education. For instance, in a very large school-based obesity prevention programme - Child Adolescent Trial for Cardiovascular Health (CATCH) - run in the US and involving 50 schools in four states, there were no statistically significant changes in the children’s blood pressure, BMI or cholesterol levels (Luepker 1996). Such policy failures are hardly surprising given that several recent studies have shown that adult attempts to control children’s eating patterns results, not only in children eating more, but in the increased likelihood of body image problems and eating disorders (Birch et al. 1991).

There is also the substantial worry about unintended outcomes, as some experts suggest that the most significant cause of eating disorders is a food and diet obsessed environment. As nutritionist Francis Berg (2004) has observed of rising rates of eating disorders: ‘Many specialists…are convinced that the current high rates of eating disorders in the US are the inevitable result of 60-80 million adults dieting, losing weight, rebounding, and learning to be chronic dieters’.

In fact, when one takes into account the revised figures of the relatively few lives lost prematurely due to obesity and weighs them against the very significant health costs associated both with dieting and eating disorders, then a compelling case can be made that damage to health from attempting to lose weight is far greater than the health consequences of overweight and obesity. Indeed, perhaps rather than a campaign against obesity it is a war on thinness that is required.

Consequences: individual autonomy and defining the good life

Scientific evidence and efficacy are but quibbles, however, beside the third problem with the war on obesity, namely its enormous potential for eroding individual liberty through employing the engine of the state, particularly its propaganda and regulatory powers, to once again define and enforce a single vision of what constitutes a good life.

In a democratic society there is considerable unease with the idea that the government has a role in structuring citizens’ souls through telling them what to think, believe or be. This is because there is a fundamental conflict between the core democratic values of autonomy and respect - values which imply that the government is a creation of its citizens - and any form of social engineering, with its counter assumption that citizens are legitimately the creation of the government. The war on obesity proposes to ignore, indeed to erase, this unease since its key assumption is that the state has both the right and the scientific expertise to define what constitutes healthy living. It proposes that this state judgement is inherently superior to the individual’s judgement that fat is, if not good, at least personally tolerable. And it contends that the individual has an obligation to order his life according to the state’s judgement about health, and that it might justifiably force him to conform if he demurs.

This claim that state-enforced healthy living is scientifically mandated is both fundamentally fraudulent and morally illegitimate. Consider the typical argument against, say, eating large amounts of fast food. The champion of the campaign against obesity will claim that it is a scientific fact that if you stop eating large amounts of fast food you will live longer. Therefore, he will say, you should stop eating so much fast food. But this will only work if another premise is added to the argument, namely, if you value living longer more than you value eating fast food, then you should stop eating so much fast food.

But as soon as this premise is added, the scientific character of the war on fat is exposed for what it is - a semantic trick that attempts to conceal the value-laden and ideological nature of the undertaking. Although it might well be science that suggests that one could live longer if one eats less fast food (although the evidence on this is very meagre), it is not science that tells me that I ought to value living longer more than eating fast food.

This does not make injunctions about obesity unworthy of attention; it merely suggests that they are not the pronouncements of science so much as views about the merits of a particular way of living. And this means that they must be justified like every other bit of moral philosophy about the good life through careful argument, not by dogmatic faux scientific pronouncements and the force of law.

At best, then, the scientific foundations of the war on fat extend only to its claims about the causes of disease, and these, we argue, are highly contradictory. When those in favour of getting the state involved in the obesity business begin to talk about what to do with this information, they cease to speak as scientists. This, in turn, has enormous implications for public policy about obesity. When the critics of obesity tell us that we must all be thinner they must tell us - and this they have never done - not only why a life of say 70 years packed full of the self-chosen pleasures of fast food and chocolate is in some sense inferior to a life of 73 years without those pleasures, but why the state is justified in intervening to amend my choice of the former in preference to the latter.

This does not mean that a life of 70 years crammed with fast food and chocolate is necessarily better than 73 abstemious years. What is does suggest is that these are not scientific choices that the promoters of the war on obesity can make for the rest of us so much as individual moral choices about the kind of life we want, a core instance of what it means to act autonomously in a democratic society. What the war on obesity proposes to do under the guise of ‘scientific decision-making’ or ‘evidence-based medicine’ is to replace my decision about the trade-offs between fast food and a risk of a possibly shorter life with the government’s calculus of what is good for me. Rather than allowing individuals to make their own health decisions about fat and obesity we are being told to put our faith in a health establishment and bureaucracy which consistently misrepresents scientific findings, issues dietary guidelines that are sloppy and in many instances purely arbitrary, and proposes solutions that are without intellectual, moral or practical rigour. At its core, the war on obesity presumes a nursery-nation full not of rational and self-governing adults, but of docile infants too uncertain of their own values and how best to realize them to be left to make their own way in the world.

In this sense, the war on obesity is about far more than fat: it is about the relentless attempt by both the health establishment and the state to order who we are under the benign idea of health. Stripped to its essentials, the war against obesity is really a war on what the health establishment conceives to be illegitimate pleasure. The proposal to impose, for instance, a sin tax on certain books, plays, ideas or associations in order to change people’s behaviour and to improve their mental health would not survive a moment of serious consideration. Yet the same policy applied to food appears to many to be uncontroversial precisely because it is ‘only’ about health.

The lasting legacy of a war on obesity will be both a much fatter government and a much thinner citizenry. The government will be fatter due to its expanded power to shape inappropriately the lives of its citizens and the citizenry will be thinner in its capacity for choice, self-government and personal responsibility. We should prefer a society of the fat and the free to one of the lean who have surrendered to the government the right to decide what constitutes a good life.

Patrick Basham and John Luik are co-authors of Diet Nation: Exposing the Obesity Crusade.

Ashton, D. (2004). ‘Food advertising and childhood obesity’. Journal of the Royal Society of Medicine 97: 51-2.

Berg, F. (2004). Underage and Overweight: America’s Childhood Obesity Crisis. New York, Hatherleigh Press.

Birch, L. et al. (1991). ‘The variability of young children’s energy intake’. New England Journal of Medicine 324: 232-35.

Duffy, M. et al. (1999). ‘The influence of advertising on the pattern of food consumption in the UK’. Journal of Advertising 18: 131-68.

Ebbeling, C. et al. (2002). ‘Childhood obesity; public health crisis, common sense cure’. Lancet 360: 473-82.

Field, A. et al. (2004). ‘Snack food intake does not predict weight change among children and adolescents’. International Journal of Obesity 28: 1210-16.

Flegal, K. (2005). ‘Excess deaths associated with underweight, overweight and obesity’. Journal of the American Medical Association 293: 1861-67.

Foster, G. et al. (2003). ‘A randomized trial of a low-carbohydrate diet for obesity’. New England Journal of Medicine 348: 2074-81.

Freedman, D. et al. (2002). ‘Trends and correlates of class 3 obesity in the United States from 1990 through 2000’. Journal of the American Medical Association 288: 1758-61.

Gronniger, J. (2005). ‘A semi-parametric analysis of the body mass index’s relationship to mortality’. American Journal of Public Health 95: 1-8.

Hawkes, C. (2004). Marketing Food to Children: The Global Regulatory Environment. Geneva, WHO.

Hedley, A. et al. (2004). ‘Prevalence of overweight and obesity among US children, adolescents, and adults, 1999-2002’. Journal of the American Medical Association 291: 2847-50.

Luepker, R. (1996). ‘Outcomes of a field trial to improve children’s dietary patterns and physical activity’. Journal of the American Medical Association 275: 768-76.

Muecke, L. et al. (1992). ‘Is childhood obesity associated with high-fat foods and low physical activity?’. Journal of School Health 62: 19-23.

Nilsson, P. et al. (2002). ‘The enigma of increased non-cancer mortality after weight loss in healthy men who are overweight or obese’. Journal of Internal Medicine 252: 70-78.

Rolland-Cachera, M. and F. Bellisle (2002). ‘Nutrition’ in W. Burniat et al. (eds) Child and Adolescent Obesity: Causes and Consequences, Prevention and Management. Cambridge, Cambridge University Press.

Serdula, M. et al. (1999). ‘Prevalence of attempting weight loss and strategies for controlling weight’. Journal of the American Medical Association 282: 1353-58.

Skrabanek, P. (1995). ‘Fat heads’. National Review. 1 May.

Troiano, R. et al. (2000). ‘Energy and fat intakes of children and adolescents in the US 2000’. American Journal of Clinical Nutrition 72: 1345-53.

Wedick, N. et al. (2002). ‘The relationship between weight loss and all-cause mortality in older men and women with and without diabetes mellitus’. Journal of the American Geriatrics Society 50: 1810-1815.

Wright, C. et al. (2001). ‘Implications of childhood obesity for adult health’. British Medical Journal 323: 1280-84.