Views & Reviews Personal View

Educational interventions are unlikely to work because obese people aren't unhappy enough to lose weight

BMJ 2012; 345 doi: (Published 17 December 2012) Cite this as: BMJ 2012;345:e8487
  1. Paul Dolan, professor of behavioural science ,
  2. Georgios Kavetsos, research fellow
  1. 1London School of Economics, London WC2A 2AE, UK
  1. Correspondence to: G Kavetsos g.kavetsos{at}

The “obesity epidemic” is unquestionably a major public health concern because obesity increases the chance of contracting many adverse health conditions and premature death. To date, nothing has reversed the ever increasing trend in obesity. One important and interesting question is why more people do not lose weight. This is somewhat puzzling considering the adverse effects that obesity has on people’s health. But although obesity can cause health problems, it does not necessarily make people feel worse in the absence of such problems. In fact, reports suggest that obesity has little effect on subjective wellbeing—basically, happiness.

Although there are ongoing debates about the merits of subjective wellbeing measures as a guide to policy making (for example, in valuing health states),1 2 such measures are useful in assessing how people are affected by their circumstances, experiences, and behaviours.3 We are much more likely to care about something that makes us feel worse off now as opposed to in the future, which we care much less about.

Recent studies on the links between subjective wellbeing and body mass index (BMI) show that they are negatively and significantly related; this is consistent with the positive correlation between obesity and depressive symptoms.4 But statistical significance is only part of the story—the size of the estimated coefficient matters too. When we look at the size of the coefficient we see just how little BMI matters, especially when compared with other determinants of subjective wellbeing, such as unemployment, marital separation, and disability.

Data from the British household panel survey (BHPS) and the German socioeconomic panel survey (GSOEP), which are representative surveys of more than 10 000 people over about 20 years, can help us assess these effects. A unit increase in BMI has about a 15th of the effect of unemployment on subjective wellbeing. A 16 unit change in BMI has only about a fifth of the effect of marital separation.5 By contrast, the effect of being moderately disabled has about as much effect as unemployment in the BHPS, and the corresponding estimate for being seriously disabled is more than double that amount.6

These relative effects should not be that surprising. Obesity is a gradual process that occurs over a long time. It is therefore something that people do not pay that much attention to and largely get used to,7 whereas unemployment can occur relatively quickly, with much less adaptation.8

The adaptive global utility model—whereby different domains of life (such as health, work, and leisure) have relative degrees of importance that are optimally reallocated in response to changed circumstances so that overall subjective wellbeing is maintained—supports this line of argument. An empirical application of the model using the GSOEP suggests that work satisfaction becomes relatively more important to overall subjective wellbeing and health satisfaction becomes relatively less important in response to increases in BMI.9

If we want to understand and change health related behaviours we need a better understanding of how different behaviours affect subjective wellbeing. Happiness and misery are important motivators of behaviour. It seems that we have little hedonic incentive to lose weight.

This does not mean that obesity does not matter. The conditions and complications associated with obesity place an enormous and ever increasing burden on resource constrained healthcare systems. These conditions and complications also greatly affect people’s subjective wellbeing and shorten their lives.

Rather, it explains why people do not care that much about obesity in itself. It is difficult to make them care. Education and information programmes that are widely implemented in member countries of the European Union and Organisation for Economic Co-ordination and Development—especially about the long term health effects of obesity—are unlikely to be very effective. Such programmes will generally serve to widen inequalities in health because they are often more effective in those who are better educated and more informed to begin with.10

Instead, it might be more effective to bring about healthier lifestyles through bans on television advertising of unhealthy foods aimed at children, reduced energy content of foods (such as England’s New Calorie Reduction pledge set to cut five billion calories from consumers’ daily intakes), financial incentive schemes, and drawing on lessons from the behavioural sciences.11 12 Such strategies have their problems, of course, but we need to bring together a range of interventions that focus on the immediate costs and adverse effects of overeating rather than on the longer term consequences, which are rarely salient and certainly not as important as how we feel.

Such policies might help us lead longer, healthier, and happier lives—at least in the long run as we avoid the misery of the adverse health effects of obesity.


Cite this as: BMJ 2012;345:e8487


  • Thanks to Dominic King, Henry Lee, and Ivo Vlaev for comments and suggestions.

  • Competing interests: None declared.

  • Not commissioned; externally peer reviewed.