Obesity and eating disorders
Prof A Hill, Academic Unit of Psychology and Behaviour, University of Leeds, School of Medicine, 15 Hyde Terrace, Leeds, West Yorkshire LS2 9LT, UK. E-mail: A.J.Hill@leeds.ac.uk
In the main, obesity and eating disorders are regarded as separate problems managed by different professions using different theoretical models and clinical approaches. Obesity is the traditional province of biological sciences, medicine and nutrition, while eating disorders are attended to by mental health professionals, psychiatry and psychology. Their expertise and approaches are grossly under-represented in obesity. Obesity researchers and specialists go to conferences organized by national and international obesity associations and supported by commercial interests, at which there is a minority of content on eating disorders. Eating-disorder specialists have their own professional groups and community but rarely attract substantial commercial sponsorship for their meetings. This division is also mirrored in the widely differing public (and media) attitudes to the disorders – indifference or blame in the case of obesity, and sympathy and awe in the case of eating disorders.
Antipathy between the two is regularly reinforced by the media in its simplistic presentation of scientific achievement and its obsession with image and celebrity. Women in the media spotlight are alternately ridiculed for weight gain or weight loss, with accompanying commentaries either berating them on their lack of self-control or praising them for their dogged pursuit of thinness. It is necessary to recognize the long-standing nature of the division between these clinical problems when considering the issues raised below.
The relationship between obesity and eating disorders
Obesity and binge-eating disorder (BED)
Compulsive over-eating in obesity and binge-eating in the absence of compensatory behaviours have been in the clinical literature for some 50 years. Only in the 1990s did BED enter the formal classification of diseases, and even now it has research rather than diagnostic status. BED is characterized by recurrent episodes of uncontrolled over-eating without the compensatory vomiting or laxative abuse that is defining of bulimia nervosa. The majority of obese individuals do not have BED, but the risk of presenting with BED increases with increasing obesity. Prevalence estimates (mostly from the USA) suggest BED affects 2–5% of obese community samples and 30% of these seeking weight-loss treatment (1). People with BED also have greater psychological morbidity. Conceptualizations of BED have argued that it is distinguishable from bulimia nervosa but that it is not a useful subtype of obesity (2). The evidence suggests that treatments directed at binge-eating can be successful but have little effect on weight loss. Weight-loss treatments are equally (modestly) effective in obese people with or without BED, but those with an affective disorder (more likely in BED) do less well. There is speculation that being a risk factor for weight gain, BED may have contributed to the increase in obesity (3).
Obesity as a risk for the development of an eating disorder
Childhood or parental obesity has been shown to be a specific risk factor for the development of bulimia nervosa but not anorexia nervosa (4). How this risk is manifest is more speculative. It may be that a family context of weight-concern and weight-control dialogue sensitizes a child to its appearance and promotes dissatisfaction. In addition, the normalization of weight control may encourage an overweight child or adolescent to attempt dietary restriction, a behaviour that strongly reinforces weight and shape concerns (5).
Weight control or dieting
Argument about the benefits and dangers of dieting exemplifies the tension between the two areas. For eating disorders, severe dieting is a risk, a defining feature, and a maintaining behaviour. For obesity, dieting is a potential solution to a major public health problem. The debate and possible options for resolution are summarized in an important paper by Brownell and Rodin, ‘The Dieting Maelstrom’ (6). Many of the issues raised have not been resolved. Two will be briefly highlighted here.
First, the development of an effective weight-loss agent is likely to be met with considerable opposition by those working in eating disorders, especially if pharmacological in nature, as there will be a natural concern that such an agent might be abused by those of normal weight and by those with eating disorders in particular. Second, there are many different weight-loss strategies available to the consumer, some that are considered healthy and others that are not. Research in teenagers shows that those overweight are more likely to use extreme and unhealthy behaviours (vomiting, using laxatives or diuretics), to binge-eat and also to have poor psychological well-being (7). Unhealthy weight control is also associated with other types of risk-taking behaviour (drugs, sex).
There are four major issues.
How does obesity operate as a risk for eating disorders?
Some of the possible personal and family dynamics are described above. There is an emerging literature on the family transmission of weight concerns and dieting to children but it is very difficult to distinguish processes. There is a tendency to neglect the powerful social narrative that praises thinness and derides fatness, which affects the lives both of children and their parents. We need more information on this before acting aggressively to prevent obesity. Parallels with the campaigns to reduce tobacco use by stigmatizing smokers have been suggested for obesity. The potential to inflame weight and shape concerns in teenagers and younger is an obvious danger.
How important is dieting as a risk factor for eating disorders?
The risk associated with dieting was estimated in an influential study by Patton et al. In a cohort of Australian school girls, those who were moderate dieters were five times more likely and severe dieters 18 times more likely to develop an eating disorder (bulimic syndrome only) than their non-dieting peers (8). Similarly, in a 5-year study of US adolescents, those girls dieting at recruitment were at twice the risk of engaging in extreme weight-control behaviours and reporting an eating disorder at follow-up (9).
But is dieting a fulcrum for eating disorders or merely a behaviour indicative of something else? Reviewing the risk literature, Stice concluded that dieting is better seen as a marker for chronic over-eating and that factors such as thin-ideal internalization and body dissatisfaction are better evidenced (10). He especially noted the need for more prospective and experimental research in this area.
Is dieting a risk for obesity?
The idea that ‘dieting makes you fat’ has been resurrected by recent research (11). Some of the most provocative (for example, Neumark-Sztainer et al. (9), Field et al. (12), Stice et al. (13)) shows that teenage girls who diet are more likely to gain weight and become obese than their non-dieting peers. Again, it is unclear whether dieting is causal in this relationship or whether it is a largely unsuccessful strategy turned to by people who have a tendency to over-eat and put on weight. There is increasing evidence that properly run and evaluated child obesity interventions reduce weight and eating pathology(see below). This makes work on improving the conceptualization of dieting an even higher priority.
The UK situation
Research data from the USA dominate this overview. Fairburn’s work on risk factors and treatment (4) is the exception to this. We know very little, for example, about the prevalence of BED and its relationship with obesity in UK samples. Similarly, there is a need for a better understanding of teenage dieting, its association with overweight, risk for eating disorders, and association with ethnicity and socioeconomic status in the UK.
It is not obvious where major beneficial developments are taking place other than those described in this overview. There is great uncertainty as to how to respond to the steady increase in child and adult obesity. Presenting the change in prevalence as an epidemic or pandemic is not helpful. It is unlikely to have motivated many of the general public to further change their behaviour. It has not helped those of us working in research or clinical practice as despite increasing coverage, there has been no substantial release of resource. Nor has it been of benefit to people working in eating disorders. Rather, the response has been to revert to portraying obesity as lifestyle failure and to hold the obese personally to blame for their failure of will power.
Addressing eating disorders – are they relevant to managing obesity?
The National Institute for Clinical Excellence guidelines on the treatment of eating disorders were published in 2004. There is very little evidence for the management of anorexia nervosa. Cognitive behaviour therapy (CBT) is the dominant evidence-based approach for bulimia nervosa and BED, although there is also evidence for the effectiveness of interpersonal psychotherapy. These are not brief treatments, however, and are therefore relatively resource-intensive; the usual duration of CBT is 16–20 sessions over 4–5 months. The use of a self-help programme is recommended as a first step for both bulimia nervosa and BED. Although far less resource-demanding, this recommendation is based on less-than-overwhelming evidence.
With regard to self-help, obesity has a much longer history than eating disorders. The varied approaches differ from those applied to eating disorders, which are psycho-educational- and CBT-oriented. Self-help in eating disorders is often delivered by a guide, usually from a non-specialist health background, who can offer support, feedback and advice. Self-help in obesity should not be dismissed. There is an acknowledgement that much of what we know about the limited success of mainstream treatments derives from a selective, non-representative sample of treatment-seekers in US facilities. A large sector of the population successfully manage their weight in the community and rarely come within the view of researchers.
Fairburn has recently adapted his eating disorders CBT approach to suit obesity. The outcomes are not yet published. The approach is to enable weight loss but in particular to target weight-loss maintenance, the phase following weight loss. The tendency of many (but not all) obese people to regain weight after treatment cessation is notorious. Fairburn’s outline approach is a one-to-one approach that follows a therapeutic style similar to that in eating disorders but over an extended period of time (14). The implication is that effective and maintained weight loss takes time and requires a great deal of therapeutic effort.
A final point to note is the difference in user support services and associated public awareness efforts. The Eating Disorders Association is a very successful group working on behalf of people with eating disorders and their families and carers. It has played an important role in increasing public awareness about eating disorders and in lobbying for improved services. There is no comparable established organization for obesity. User involvement and support have been relatively invisible in the UK, as in most other countries. This is unlikely to remain the case. A challenge will be for any such organization to operate independently of commercial weight-loss interests.
Managing obesity – implications for eating disorders
One of the most frequently voiced concerns, especially by health professionals working with overweight and obese children, is that active engagement in weight loss may result in some children developing an eating disorder. This has been hard to counter as there have been relatively few long-term child obesity treatments reported in the literature. The best known have been the family based programmes run by Epstein in the USA (15). Some of these children have been followed for 10 years and more. While there have been occasional cases of eating disorders presenting in children who went through these programmes, it is claimed that the rates are similar to those found in the community. Butryn and Brownell reviewed the literature in 2005 and found five studies relevant to this issue (16). Their conclusion was that professionally administered weight-loss programmes for children and adolescents did not increase symptoms of eating disorders. Indeed, they were associated with significant improvements in psychological well-being (see, for example, Walker et al. (17)). There are two caveats attached to this conclusion. First, there are relatively few professionally administered programmes in the UK relative to the number of children and adolescents trying to lose weight. Weight control, properly and expertly managed, is not a risk for eating disorders but, without this organization and control, it may be a risk. Second, it is likely that some overweight children and adolescents are more at risk of eating disorders than others. More attention is needed to identify what these other risks might be and how to screen for them.
The possibility of launching a negative health campaign on behalf of obesity to mobilize the public to change their behaviour has already been raised. Any action that further stigmatizes obesity and fat people is likely to inflame child and adolescent weight concerns and risks increasing eating disorders (18).
Future joint working
Both obesity and eating disorders would benefit from a reduction in the negativity that obesity attracts. Women at risk of eating disorders would not judge their appearance so critically if the perceived consequences of not being thin were not so great. The view that obese people should have a negative view of their weight in order to motivate them to lose weight is misplaced. Most people find it extremely difficult to make large and long-lasting changes to their weight. High levels of dissatisfaction do not guarantee success. Rather, they can add to a personal sense of failure and futility in trying to lose weight. Reducing the negative portrayal of obesity (aesthetically and in terms of health) may paradoxically allow more people to achieve the modest weight losses (∼10%) that health professionals are recommending. Personal weight-loss goals that are too great will normally end in failure, leading people eventually to give up. It is not surprising to find in surveys of weight-loss behaviour that those who are overweight are trying most intensively but those who are obese are less involved.
Improving public understanding of the complexities of weight regulation and weight loss would have enormous benefit. It is interesting that a psycho-educational approach is at the heart of many eating disorder treatments but is overlooked in obesity management. Rather, there is a simplistic message about energy balance normally summarized by the four-word phrase, ‘Eat less, exercise more’. If the solution to obesity was that simple, there would not be the levels of obesity seen today. Health professionals, Government and the media need to work together to provide realistic, consistent and helpful information. Part of the solution may lie in better training health professionals in the complexity of weight management.
Of considerable concern is the influence of the weight-loss industry. Quick-fix products and diet plans are usually little more than quackery but promise consumers what they want to hear: fast and effortless permanent weight loss. The damage is huge. Failure to lose weight leads customers to blame themselves rather than the ineffectiveness of the product. A history of failures leads to giving up all together. Unlike pharmaceuticals, diet products do not have to prove their effectiveness. They rely on personal or celebrity testimony to encourage people to spend their money – and, ultimately, give up hope. There is scope for regulation of wild claims.
It is important to consider common preventative approaches for obesity and eating disorders. The possibilities of putting together school-based approaches have been considered (for example, O’Dea (18), Neumark-Sztainer (19)). One focus might be on promoting self-esteem in children and adolescents. This may have additional benefit for other health issues such as self-harm and conduct disorder. Alternatively, environments that promote and skill young people in healthy eating would be of mutual benefit to obesity and eating disorders. Follow-up analysis of the US Planet Health obesity prevention intervention aimed at early high-school years showed a significant reduction in the number of girls using extreme weight-loss practices compared with those in control schools (20). Further evidence of the long-term benefits of such interventions is needed.
Conflict of Interest Statement
No conflict of interest was declared.