Article Text

Download PDFPDF

Moral responsibility for (un)healthy behaviour
  1. Rebecca C H Brown
  1. Correspondence to Rebecca C H Brown, Department of Law, Queen Mary, University of London, Mile End Road, London E1 4NS, UK; rebecca.brown{at}


Combatting chronic, lifestyle-related disease has become a healthcare priority in the developed world. The role personal responsibility should play in healthcare provision has growing pertinence given the growing significance of individual lifestyle choices for health. Media reporting focussing on the ‘bad behaviour’ of individuals suffering lifestyle-related disease, and policies aimed at encouraging ‘responsibilisation’ in healthcare highlight the importance of understanding the scope of responsibility ascriptions in this context. Research into the social determinants of health and psychological mechanisms of health behaviour could undermine some commonly held and tacit assumptions about the moral responsibility of agents for the sorts of lifestyles they adopt. I use Philip Petit's conception of freedom as ‘fitness to be held responsible’ to consider the significance of some of this evidence for assessing the moral responsibility of agents. I propose that, in some cases, factors outside the agent's control may influence behaviour in such a way as to undermine her freedom along the three dimensions described by Pettit: freedom of action; a sense of identification with one's actions; and whether one's social position renders one vulnerable to pressure from more powerful others.

  • Allocation of Health Care Resources
  • Behaviour Modification
  • Social Aspects

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 3.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


The use of public resources to combat lifestyle-related disease sometimes comes under criticism. Interventions intended to assist the overweight or smokers are often debated in the press.1 ,2 One argument levelled at such interventions is that they reward (explicitly or otherwise) bad behaviour: individuals who fail to live healthily thereby fail to meet some obligation they have to society. Yet, many resources are directed towards helping those whose lifestyles contribute to their poor health outcomes. This can include interventions such as weight loss surgery, incentives for healthy behaviour, free sexual health testing and the provision of smoking cessation aids.3 ,4

In this paper, I explore one key assumption of this argument: that, in determining their own lifestyles, individuals act freely and are morally responsible for engaging in unhealthy behaviours. The focus for responsibility is on habitual, ‘lifestyle choices’, which contribute to the sorts of chronic disease at issue. The implications of the bad behaviour argument, as it is often expressed, are that those who fail to fulfil their obligations to live healthily are less deserving of healthcare than those who meet their obligations.

Criticisms like this in the media may be representative of public opinion, and indicate the acceptability of interventions targeting lifestyle-related disease.5 ,6 Further, personal responsibility has been increasingly incorporated into welfare state policies, particularly since the 1970s.7 There continues to be a growing trend for ‘responsibilisation’ (often related to ‘personalised healthcare’) which assumes that agents can (and should) be held morally responsible for their health outcomes.8 ,9

In this paper, I present some of the evidence from research into the causes of lifestyle-related disease, and the significance of this for ascribing moral responsibility to agents for their (health-related) actions. I use Philip Pettit's discussion of freedom as ‘fitness to be held responsible’ to consider how evidence from research into the social determinants of health and health psychology should inform our approach to considering the moral responsibility of agents for their healthy or unhealthy lifestyles.10

Non-communicable diseases and the burden on healthcare

One of the greatest challenges for healthcare providers is posed by preventable, lifestyle-related, non-communicable diseases (NCDs). These include diabetes, cardiovascular diseases, cancer, chronic respiratory diseases and mental disorders. Recently, the World Health Organization (WHO) identified NCDs as being at the ‘top of [the] world's agenda … [accounting] for over 86% of deaths and 77% of the disease burden in the WHO European Region’.11

Risk factors, such as tobacco use, poor diet and lack of physical activity contribute to one's likelihood of developing NCDs. However, despite government informational campaigns aimed at reducing risk factor exposure, unhealthy lifestyles are still common. In 2009, one quarter of adults in England were estimated to be obese (with a body mass index of 30 or more), and 21% of the adult population of Great Britain were smokers (a figure unchanged since 2007).12 ,13

Unhealthy lifestyles and NCDs are damaging both to agent welfare and the economy. The National Audit Office estimates that obesity costs the National Health Service (NHS) more than half a billion pounds a year, and probably more than two billion to the wider economy (due to lost working days, and so on.)14 Such figures are contestable but, clearly, tackling preventable deaths should be a healthcare priority.

Responsibility, desert and ‘bad behaviour’

Claims about bad behaviour relate to theories of desert and fairness. Contemporary philosophers have been sceptical of theories of justice based upon agents ‘getting what's deserved’.15 ,16 Yet the notion of desert persists, particularly in the media and lay expressions of opinion on policy, including healthcare.1–3 Further, support for responsibilisation in healthcare policy (often interwoven with notions of ‘personalised’ healthcare) may reflect notions of desert among more academic writers and policymakers.8 ,9 ,17–20

In this paper, I question the legitimacy of holding agents morally responsible for their (un)healthy lifestyles. Pettit equates both theories of psychological freedom and political liberty to ‘fitness to be held responsible’.10 Framing moral responsibility for one's lifestyle in this way helps to focus our assessments of responsibility on whether the environmental factors, and their impact on individual psychology, are such that they allow agents to act sufficiently freely so as to render them fit to be held responsible.

Arguments such as the bad behaviour criticism involve further claims than the mere moral responsibility of agents for their unhealthy behaviour. For instance, a principle of solidarity might be used to establish an obligation to be healthy that is not met by some agents.16 ,19 ,20 However, I shall not discuss further notions of desert, blameworthiness, solidarity and so on, as these are complex issues in their own right which fall outside the scope of this paper. I restrict myself to moral responsibility as resulting from the sort of freedom described by Pettit, and the implications of empirical evidence on health behaviour for such an analysis.

The relevant form of responsibility here seems to be merit based: where agents are deserving of some reactive attitude based on their actions. I shall not discuss the primacy or otherwise of reactive attitudes (as argued by Strawson in his seminal paper),21 but assume responsible agents must be ‘reaction-worthy’. I am explicitly excluding ‘forward-looking’ or ‘as-if’ responsibility where ascriptions are based on the positive consequences that might derive from holding agents responsible.22 As-if responsibility may discourage risky behaviour and help to avoid ‘moral hazard’ by ensuring the consequences of an action fall only on the individual performing it, regardless of whether he or she is really an apt candidate for responsibility. This may help to promote economic efficiency, by discouraging risky behaviour. As-if responsibility may also be used to develop the capacity for responsibility in children (if they are told they will be held responsible they may learn to take on the role of a responsible agent). It might also be important to treat agents as-if they are responsible if this is key to our self-perception. Responsibility, as a value, interacts with other things that we care about, and so, considerations of whether or not agents actually are morally responsible (on a merit-based account) will often need to be traded off against other reasons for treating agents as responsible.23 ,24

Merit-based responsibility ascriptions are notoriously tricky. Rawls proposes credit be awarded according to the extent to which agents ‘conscientiously strive’ to do the right thing. However, he also acknowledges that an agent's capacity to ‘strive’ may be influenced by her upbringing and social circumstances: factors out of her control and not her responsibility.25 It thus becomes more complicated than just identifying agents who ‘do right’, or even those who try to ‘do right’.

On Pettit's account of freedom as fitness to be held responsible, he identifies three key components:

  1. ‘the freedom of an action performed by an agent on this or that occasion,’

  2. ‘the freedom of the self implicit in the agent's ability to identify with the things thereby done, rather than having to look on them as a bystander,’

  3. ‘the freedom of the person involved in enjoying a social status that makes the action truly theirs, not an action produced under pressure from others.’10

It will be useful to bear these in mind when considering the influence of environmental and psychological factors in shaping the lifestyles of agents.

Socioeconomic factors and health

Research has identified a gradient between health and socioeconomic status, such that, roughly, the further down the social hierarchy an individual is, the more likely she is to experience poorer health outcomes. Differences that stem from inequalities in power, money and resources, contribute to the unequal distribution of health outcomes. These include: [M]aterial circumstances, the social environment, psychosocial factors, behaviours, and biological factors. In turn, these factors are influenced by social position, itself shaped by education, occupation, income, gender, ethnicity and race. All these influences are affected by the socio-political and cultural and social context in which they sit.26

Risk factor exposure, disease prevalence and social standing are intricately linked. Women in managerial and professional groups have a 19% rate of overweight and obesity, compared with 29% in routine and semiroutine groups.27 In men, 17% of those in professional occupations smoke compared with 31% of those in manual occupations.28 The effects of unhealthy lifestyles manifest themselves in disease prevalence and life expectancy. In one of the wealthiest areas of London, for instance, male life expectancy is 88 years; in one of the poorest parts of the same city, life expectancy is cut by 17 years, down to 71 years.26

Intuitively, it seems within one's power to avoid many of the health-affecting risk factors thus far mentioned: ultimately, each agent can still choose how she lives. It is important, therefore, to say a little about the mechanisms underlying health behaviour, as choices can, after all, be more or less freely made.

The psychology of health behaviour

An extensive literature seeks to explain the psychological mechanisms underlying the social patterning of health. The ‘two systems’ psychological model proposes that, while some behaviour is influenced by values and conscious reasoning, another pathway operates which requires little or no cognitive engagement.29 ,30 It is reported that around 45% of behaviour is habitual, and not normally controlled by conscious reasoning processes.28 ,31 A large and sustained psychological effort is needed to intervene in everyday behaviours and alter habits.

This habitual nature of behaviour makes it highly susceptible to environmental influences: factors like a high density of fast-food outlets and being surrounded by others engaging in unhealthy behaviours can have a big influence on lifestyle in deprived areas.28 ,32 ,33 Moreover, individuals in deprived areas are exposed to greater levels of stress which affects the brain, enhancing the tendency for impulsive behaviour.34 Robust self-regulatory capacities are needed to alter habitual behaviours, yet development of these are hindered by poverty and deprivation early in life.35–38

Kotz and West present evidence showing that smokers from the most deprived socioeconomic groups are as likely to attempt to quit smoking (and to seek help in doing so), but are only half as likely to succeed compared with those in the highest socioeconomic groups.39 Environmental cues for unhealthy behaviours, combined with reduced cognitive capacity for self-regulation, could provide greater barriers to success for some individuals attempting to adopt healthier lifestyles.

Freedom to choose

Debates in the press raise questions relating to the moral responsibility of agents for poor health they suffer as a result of unhealthy lifestyles. Philip Pettit's account of freedom seeks to connect both freedom in the agent and political freedom (sometimes referred to as ‘free will’ and ‘political liberty’). Pettit does so by taking ‘fitness to be held responsible’ as the key indicator of freedom. Such a conception of freedom provides a useful frame for considering the extent to which agents may be considered morally responsible for their (un)healthy lifestyles: it encourages one to consider the freedom of agents in general allowed by social and environmental structures, and the mechanisms of individual psychology.

Recall Pettit's three components of freedom mentioned in the section on Responsibility, desert and ‘bad behaviour’, above.

If these three aspects of freedom are not met, an agent will not be fully free and fully fit to be held responsible. I believe the evidence from research into the social determinants of health and mechanisms of health behaviour can be informative in assessing the freedom of agents to adopt more or less healthy lifestyles. I will briefly consider each of Pettit's components of freedom in turn.

First, the action performed on any given occasion may not be fully free. The classic example of unfree, forced action is something like having a gun held to one's head. In terms of health behaviours, acting from a compulsion to smoke or eat, or an inability to muster the motivation to exercise does not appear to render an agent incapable of acting freely in the same way as the gun to the head example. However, there are different ways and extents to which an agent's psychology may come under pressure. Temptations to smoke or eat Mars bars can be strong: agents can become dependent on or addicted to such activities, which preclude them from acting in (healthier) ways which they might generally prefer.

Evidence from health psychology suggests that individuals from deprived backgrounds are less likely to develop those self-regulatory skills needed for them to intervene with habitual, impulsive behaviour. This, in combination with the environmental cues for unhealthy behaviour that poorer areas provide, means that some of those raised and living in more deprived areas could have their capacity for free action undermined to some extent.

The second aspect of freedom Pettit describes relates to identification and ownership. The question posed here is whether ‘[t]he agent cannot be detached from the action, or from the process leading to the action, in the way they may be detached from a reflex or a pathology or even an obsession or compulsion’.10 Thus, agents who act under the influence of drugs, or while asleep, will not ‘own’ their actions fully: they will not see it as ‘them’ acting at the time.

To meet the requirement for ownership of action, the agent may need to act in line with how she perceives herself, and not counter to her long-term desires and goals. Compulsive behaviours can scupper this. Consider an agent with a strong wish to be a non-smoker, who forms the intention to quit smoking and takes steps to achieve this (buys nicotine gum and patches, seeks counselling, and generally exerts every effort within her capacity to quit). If, nonetheless, she succumbs to cravings and continues to smoke it seems quite plausible that she will not feel a sense of identity and ownership over her actions, having taken every step she can to avoid them. The lack of ownership will not be so profound for the ‘reluctant smoker’ as for the sleepwalker, but it may still be said that she is alienated from her actions, and that her freedom is thus undermined.

Where an agent's behaviour is controlled by overwhelming desires, perhaps through addiction or compulsion, it may not count as her action, insofar as it is attributable more to the overwhelming desire than to the person as a whole.40 This is surely the case for some unhealthy behaviours (eating disorders, alcoholism and other addictions recognised as pathologies), though not all. It seems, however, given the habitual (unconscious) nature of much of this behaviour, the self-identification of an agent with her lifestyle habits may be undermined, at least to some extent. For instance, the presence of palatable foods, such as chocolate, may cause ‘unintentional’ eating in agents who have expressed an intention of avoiding high-calorie snacks. This tendency is linked to a measure of inhibitory control, and is suggested to result from ‘food choices [occurring] impulsively with little conscious awareness’.41–43

The final requirement for free action relates to social status and the absence of significant external pressure. Perhaps the classic examples of freedom being undermined in this way are people forced into prostitution or crime by the desperate circumstances they find themselves in. The relation between socioeconomic status and unhealthy behaviour may be more subtle than those cases, however, lack of resources and power have a huge influence on one's lifestyle. The pernicious impact of environment on health increases along with deprivation, so that the lower the socioeconomic status of an individual, the more likely she is to be exposed to environmental cues for unhealthy behaviour, and to lack the capacity for self-regulation needed to resist such cues. Thus, the social gradient in health is linked to this third component of freedom. Moreover, it seems that this lack of freedom of action and freedom of the self is exploited by powerful organisations to optimise their profits: recall the placement of fast-food outlets in areas of high deprivation, where residents are more susceptible to cravings for high fat, salt and sugar content foods.32 ,33

The application of criteria, such as Pettit's, for judging freedom provides us with the beginnings of a practical though imperfect way of assessing responsibility which may have relevance for healthcare provision. It should be borne in mind, however, that agents may be more or less free along each of the dimensions discussed here, and thus more or less fit to be held responsible. Such variations in freedom and responsibility will also be linked in differing ways to socioeconomic status and the gradient in health. Thus, more fine-grained analysis of the links between socioeconomic factors and health behaviour would be necessary to provide a more nuanced account of how freedom and responsibility are relevant to healthcare provision.

Concluding remarks

It has been argued that personal responsibility could play a role in healthcare provision, for example, when determining how long somebody should wait for treatment, or in ‘tie-break’ situations where medically, two people are equally in need of a treatment available only to one of them.17 ,20 The bad behaviour argument is sometimes raised to justify the feeling that those whose lifestyles contribute to their poor health status are less deserving of treatment from publicly funded organisations, such as the NHS. In general, there is also an increasing move towards responsibilisation in healthcare (and other) policy.8 ,9

In this paper, I have summarised some of the evidence from research into the social determinants of health and health psychology, and considered the relevance of this to efforts to judge moral responsibility. I have proposed that Pettit's theory of freedom as fitness to be held responsible provides a useful frame for such a discussion. I have argued that the social gradient in health can be considered as linked to freedom: those who are subject to more social deprivation are more likely to have their freedom limited across the three components of freedom that Pettit describes. As freedom is equated with fitness to be held responsible, this has relevance for policies which assume agents are morally responsible for their (un)healthy behaviour, or which seek to encourage responsibilisation.

I do not seek to argue that all those from deprived backgrounds are incapable of making robust, character-driven decisions about their lifestyles. Nor do I show that all those engaging in unhealthy behaviour are not morally responsible for their actions. In order to make a good assessment of responsibility, it would be necessary to look at the circumstances of each individual and assess to what extent she was free to act otherwise. However, this is impractical from the point of view of shaping public health policy. The social patterning of some sorts of disease does, however, suggest that the mechanisms causing disease involve less personal control than one might assume. Thus, where we are concerned with lifestyle-related diseases, such as NCDs, freedom, (and by implication, responsibility) might be undermined.

We might have other reasons for punishing bad behaviour, or for introducing more of a role for personal responsibility into healthcare: it may be more economically efficient, or it might help preserve the view of ourselves as autonomous individuals in control of our own lives. However, unless we consider a form of as-if responsibility, and not the more substantive merit-based form, this will require agents to have sufficient freedom in order to be fit to be held responsible. This is problematic given what we know about the influences on health behaviour. Aside from undermining assumptions of moral responsibility of agents with regards to their (un)healthy lifestyles, an account of freedom, such as that provided by Pettit, could indicate the sort of features that would need to be present for freedom and moral responsibility to be supported.


Thank you to Richard Ashcroft, Paul Dolan, Theresa Marteau and other members of the Centre for the Study of Incentives in Health for helpful comments on this paper.


View Abstract


  • Funding This research was funded by the Centre for the Study of Incentives in Health, from a strategic award from the Wellcome Trust Biomedical Ethics Programme (PI Marteau: 086031/Z/08/Z).

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.