Obesity interventions and ethics
Professor S Holm, Cardiff Law School, Cardiff University, Law Building, Museum Avenue, Cardiff CF10 3AX, UK. E-mail: holms@Cardiff.ac.uk
The purpose of ethical reflection is to help us decide how to act in the real world, all things considered. This means ethical reflection must take into account all aspects of a proposed course of action, it must rely on good factual evidence and on an understanding of the social and legal contexts in which action has to take place.
In the present public health context, the main actors under consideration are, on the one hand, the state and doctors and others in their roles as agents of the state, and, on the other hand, the individual. The main principles or ethical values in play are respect for self-determination, the pursuit of a good life, the promotion of the common good, the obligations between parents and children and justice (1,2).
The basic problems
Two basic ethical problems recur in discussions about obesity interventions. The first is whether and when it is justifiable to intervene to promote a person’s own health or well-being, if they do not want the intervention. When can paternalism be justified?
The second is whether and when it is justifiable to negatively affect a person or people’s well-being in order to benefit others or to promote the common good. In this context, it is important to note that pure economic gain is usually not considered sufficient reason to claim that the common good is being promoted.
These two problems do not occur on a neutral background. The current nutritional and physical activity landscape is already shaped by numerous historical factors, by intentional and unintentional effects of government regulation and by the actions of commercial actors. The way society is organized is already affecting people’s well-being in different ways.
Hard and soft paternalism
A useful distinction has been made by the Finnish philosopher Heta Häyry between three different forms of paternalism: hard paternalism, soft paternalism and maternalism. Hard paternalism involves direct coercion, soft paternalism involves giving unwanted information or foreclosing some options for action and maternalism involves control by inducing a guilty conscience (3).
An example that combines soft paternalism and maternalism is the practice of telling pregnant women that they are potentially harming their babies if they drink alcohol during pregnancy, whereas the more traditional ‘if you smoke, you die’ message on cigarette packs and posters is an example of a more pure form of soft paternalism.
It is generally accepted that hard paternalism requires a stronger justification than soft paternalism. Whereas hard paternalism by definition involves the denial of self-determination, soft paternalism still allows the expression of choice. For instance, someone can be told that certain foods are unhealthy, yet they are still able to buy and eat them.
With regard to the foreclosing of options, there is a gradation from soft to hard paternalism, depending on how difficult it becomes for a person to still pursue the discouraged choice. A general practitioner who gives unsolicited advice on diet and weight loss to a patient who has only came to his surgery for a travel vaccination is engaging in soft paternalism, but may be engaging in hard paternalism if, for instance, he refuses to give the shot until his patient has begun to listen more attentively.
If smoking bans in public places had been justified in terms of their health benefits to smokers, it would have been an example of soft paternalism which gradually edged into hard paternalism as more and more areas became non-smoking. However, the official justification for smoking bans are their health benefits to non-smokers, which actually makes them examples in which the well-being of smokers has been sacrificed for the sake of the well-being of non-smokers.
Choice, rationality and information
Can we dispel some of the worries about hard paternalism by making a distinction between informed and/or rational choices that ought to be respected, and uninformed and/or irrational choices that we can legitimately interfere with because they are in some sense not the person’s ‘real’ choices (4)? Should all choices be respected?
Making such a distinction is more complicated than it might initially seem, because, although we may be able to define the extreme poles of the decision-making spectrum with some precision, it is difficult to say where on the spectrum a given decision becomes so ill informed or so irrational that we can overrule it. True, most decisions made by most people most of the time are not terribly well informed or well thought through. Nevertheless, we are usually quite happy to accept that the choices made are theirs and that we should not use any stronger means than advice or mild persuasion to alter that choice, even if we think they are making the wrong decision. Think, for example, of the parents of teenage children intent on body piercings. This holds true even if the decisions in question are significant ones like changing employment, getting married or buying a house.
If we are worried about the quality of decision-making in the context of food and exercise, and if we believe that some people do not have the necessary information to make good choices, it is much less problematic to provide them with that information than to take their choices away from them.
On the other hand, the need for consumers to have enough information to base informed choices on is one of the strongest reasons for requiring the producers, manufacturers and retailers of food to provide unbiased and easily understandable information about their products to consumers. Such requirements are not limited to information about nutrition, and include many other items that are important to choice, such as weight, country of origin, dolphin-friendliness and so on. Giving nutritional information does not necessarily produce nutritionally optimal choices. However, this is not a reason for not giving the information, as its purpose is to improve the consumer’s choice process.
Targeting high-risk groups
What are the ethical issues raised by interventions that target identified high-risk groups? First, there are potential problems of social stigmatization and changed self-perception if the targeting results in individuals being identified as belonging to a high-risk group. Insofar as obesity is already a stigmatized condition, being identified as being at risk of obesity may also amount to or produce stigma. This problem may be obviated to some degree if the targeting is performed by self-identification, but such a strategy will often lead to very inefficient targeting.
Second, there are potential problems of justice. The reason for targeting high-risk groups is usually cost-effectiveness. A given expenditure of resources will result in more obesity being prevented than if we had chosen to employ a ‘broader’ strategy. But do individuals at similar high risk, who are outside the identified high risk groups, not have the same claim on our preventive resources? A basic principle of formal justice, originally outlined by Aristotle is that equal cases should be treated equally, and this indicates that it is at least potentially problematic to focus prevention only on high-risk groups, unless such groups can be so precisely defined, identified and targeted that all high-risk individuals are included.
A different problem occurs in health promotions that target the whole population. In order to achieve a reduction in the prevalence of obesity, it may be necessary to achieve behaviour change far beyond simply the obese. It might be necessary to shift the whole weight curve downwards (or shift the whole energy intake or physical activity curve). Thus, all those who are overweight, or simply perceive themselves to be so, will have to change their lifestyles in order to reduce the number of obese people. Many of those who are made to feel concerned and are induced to change their lifestyle might never have become obese or have had any significant health problems related to their weight. This means that we are affecting some individuals negatively in order to benefit others or to create healthcare savings at the societal level. This can be justified if the bad thing that is averted (obesity and its health effects) clearly outweigh the negative effects experienced by the non-obese, but if this is truly the case, it needs to be substantiated.
Choice and responsibility
It is a commonplace that people are responsible for the foreseeable consequences of their choices and actions. This seems to indicate that the obese are responsible for being obese and for the health and other consequences of their obesity. There are, however, a number of complications in making a claim of personal responsibility.
The first is that it is not clear to what degree lifestyle is actually a matter of conscious choice. There are at least large elements of socialization and social construction involved in acquiring and maintaining a given lifestyle, and the possibility to break with or significantly alter one’s lifestyle is significantly influenced by a range of socioeconomic factors.
The second is that the lifestyle in question may contribute significantly to the person’s sense of well-being. They may simply like to live the way they live and may accept future negative effects as a reasonable trade-off for their current pleasure and well-being. We may well think that they have either miscalculated the trade-off over time, or that they could get the same amount of well-being now by changing their lifestyle. But whether a given individual will have a better quality of life as a burger-eating smoker or a vegetarian fitness enthusiast is probably not a question that can be objectively decided by a third party.
The final problem is that, even if we could demonstrate that the obese had chosen to be obese, and are therefore responsible for their own obesity and its consequences, that still would not settle the question of whether we should hold them responsible and treat them differently. Unless we treat all similar, non-obesity-related instances of negative health effects of personal choice in the same way, the obese would have a strong claim that they are being treated unjustly.
We should also note that there is an inherent tension between a focus on personal choice and responsibility and the justification for large-scale interventions that affect some non-obese individuals negatively (e.g. by inducing unnecessary lifestyle changes). If the obese are personally responsible for their condition, it becomes doubtful whether others should sacrifice their interests to benefit the obese. For these reasons, it is ethically problematic to pursue a policy primarily emphasizing choice and responsibility.
This raises further the question of whether it is a condition, in expecting personal responsibility, that society does its part. Can we legitimately expect members of society to show responsibility if society does not shoulder its responsibilities or is not perceived to be shouldering its responsibilities? Is there a requirement of ‘reciprocity of responsibility’? It may, for instance, be argued that, if society could reduce obesity by prohibiting specific marketing strategies, but declines to do so, it is hypocritical to put much emphasis on personal responsibility.
Many arguments in the debate about obesity try to determine exactly who has responsibility, but it is important to remember that it is not a zero-sum game. There is more than enough responsibility to go round for all parties; the state, the food industry and the individual. We have no reason to believe that only one of these parties is responsible, and even less reason to believe that only one party should be held responsible.
It is generally accepted that many instances of obesity can be traced to early childhood or possibly even to the intrauterine environment. A potential, and on the face of it, very attractive intervention strategy is therefore to aim for optimal nutrition in this period. This would probably involve both intensive dietary advice to all parents, more intensive monitoring of weight and body shape of all children, and targeted intervention in families where monitoring reveals problems. Such an intervention could be seen as an extension of current antenatal and health visiting initiatives.
The main ethical arguments for the intervention would be that it has good consequences for the child and for society, and that parents already have an ethical and legal obligation to promote the ‘best interests of the child’ (in law the obligation is for the child’s interests to be paramount). A general problem with the ‘best interests of the child’ argument is that we do and must allow parents to act against the best interests of the child, or to act on their conception of the best interest in many contexts. We do not think it is wrong for young parents to save for pensions, even if spending the money now on their child could benefit the child. Furthermore, if there is more than one child in the family, it will sometimes be impossible to promote the best interests of both children at the same time when these interests are in conflict (5).
It is therefore unclear whether parents actually have a strong obligation to promote optimal nutrition in their children, or just a strong obligation to promote ‘good enough’ nutrition. Achieving optimal nutrition for the child would, in the prenatal period, physically require that the mother changes her nutritional pattern, and may later require that the whole family begins to eat differently to the way they did before they had the child. This may well be beneficial for everyone, but it is not obvious that there’s a strong obligation to do this.
The scepticism about obligation is reinforced by two further factors. First is the observation that dietary advice is not constant and that it is not all based on good evidence. ‘Optimal nutrition’ therefore actually means ‘nutrition currently believed to be optimal’.
Second is the fact is that not all overweight children go on to become obese adults. Targeting risk groups in childhood therefore also necessarily leads to some families being the target of unnecessary interventions.
Promoting body image
Another set of questions arises in interventions aimed at promoting a specific body image or range of body images (i) over the strength of the evidence base for promoting this specific body image; (ii) over the possible side effects (e.g. an increase in anorexia or other eating problems) and (iii) over the risk of stigmatization of those with body shapes deemed not ideal.
In one way, discussing possible future interventions in this area is strange because it is quite obvious that our culture and many other cultures already promote a specific body image, not through public health campaigns but through ordinary advertising (6). Based on our experiences, we can justifiably claim that it is difficult to promote one body shape as good without implying that other shapes are bad, and it is unclear whether it is possible to prevent people from linking bad body shape to personal and moral badness. This is particularly the case when obesity is already stigmatized.
It will be evident from the discussion so far, that two interlocking discussions underlie many of the ethical controversies in the obesity intervention area.
The first is a discussion concerning the correct analysis of value. Are all values subjective or personal, or are there objective values? Is health good for everyone whether or not they value it, or is it only good for those who choose to pursue it? The second is a discussion concerning the degree to which society or the state is justified in interfering in personal choices, and the legitimacy of different kinds of interventions.
Although these two discussions are conceptually distinct, they are linked because it may be easier to justify paternalistic intervention if there are truly objective values.
However, even if there are objective values, there are probably more than one (e.g. health is not the only value). This means that there might be a conflict of values and that they will have to be balanced or prioritized. It is unclear that health will always come out on top in when conflicting values are weighed up.
This means that policies involving hard paternalism will always be difficult to justify. However, even the most ardent libertarians can support many policies based on soft paternalism, especially policies where choice is not curtailed, but choosing healthier options is made easier, for instance, by making the healthy choice the default. Such policies could be labelled ‘libertarian paternalist’ (7) and are unobjectionable as long as they are based on solid evidence that the healthy choice is really the healthiest option. Such evidence does not necessarily have to amount to ‘scientific proof’– a notion that is controversial in itself – but it has to be more than just deduction from theory. All interventions have costs, both ethical and economic/opportunity costs, and it is only when we have some evidence for their effectiveness that we get a handle on whether the costs outweigh the benefits.
Conflict of Interest Statement
No conflict of interest was declared.
I gratefully acknowledge many discussions with colleagues in the Cardiff Institute of Society, Health and Ethics and the stimulus provided by my participation in the EUROBESE project sponsored by the European Commission, DG-Research.