Series

Obesity 1
The global obesity pandemic: shaped by global drivers and
local environments
Boyd A Swinburn, Gary Sacks, Kevin D Hall, Klim McPherson, Diane T Finegood, Marjory L Moodie, Steven L Gortmaker

Lancet 2011; 378: 804–14 The simultaneous increases in obesity in almost all countries seem to be driven mainly by changes in the global food
See Editorial page 741 system, which is producing more processed, affordable, and effectively marketed food than ever before. This passive
See Comment pages 743, overconsumption of energy leading to obesity is a predictable outcome of market economies predicated on
744, and 746 consumption-based growth. The global food system drivers interact with local environmental factors to create a wide
See Perspectives page 761 variation in obesity prevalence between populations. Within populations, the interactions between environmental
This is the first in a Series of and individual factors, including genetic makeup, explain variability in body size between individuals. However, even
four papers about obesity with this individual variation, the epidemic has predictable patterns in subpopulations. In low-income countries,
WHO Collaborating Centre obesity mostly affects middle-aged adults (especially women) from wealthy, urban environments; whereas in high-
for Obesity Prevention
(B A Swinburn MD, G Sacks PhD),
income countries it affects both sexes and all ages, but is disproportionately greater in disadvantaged groups. Unlike
and Deakin Health Economics, other major causes of preventable death and disability, such as tobacco use, injuries, and infectious diseases, there are
Deakin Population Health no exemplar populations in which the obesity epidemic has been reversed by public health measures. This absence
(M L Moodie DrPH), Deakin increases the urgency for evidence-creating policy action, with a priority on reduction of the supply-side drivers.
University, Melbourne, VIC,
Australia; National Institute of
Diabetes and Digestive and Introduction
Kidney Diseases, National As UN member states prepare to gather in New York in Key messages
Institutes of Health, September, 2011, for the first High-Level Meeting of the
Washington, DC, USA • Changes in the global food system, including reductions in
(K D Hall PhD); New College,
UN General Assembly on non-communicable diseases the time-cost of food, seem to be the major drivers of the
University of Oxford, Oxford, (NCDs), the inexorable global rise of obesity will be the rise of the global obesity epidemic during the past
UK (Prof K McPherson PhD); toughest challenge that they face. Many countries can 3–4 decades, although substantial differences in national
Department of Biomedical serve as excellent exemplars for reduction of infectious
Physiology and Kinesiology, and local environments (especially sociocultural, economic,
Simon Fraser University,
diseases, injuries, and some of the risk factors for NCDs, and transport environments) produce the wide variation in
Vancouver, BC, Canada such as smoking, high cholesterol, and hypertension. obesity prevalence recorded across populations.
(Prof D T Finegood PhD); and However, no country can act as a public health exemplar • In the first half of the 20th century, increased
Department of Society, Human for reduction of obesity and type 2 diabetes. All countries
Development and Health, mechanisation and motorisation were accompanied by
Harvard School of Public
are searching for answers about how to reverse the rising corresponding decreases in food energy supply (indicative
Health, Harvard University, tide of adult and childhood obesity. of consumption), thereby keeping obesity prevalence low.
Boston, MA, USA The 2004 WHO global strategy on diet, physical activity In many high-income countries, an energy balance
(Prof S L Gortmaker PhD)
and health1 provides an excellent overall guide for societal flipping point seems to have occurred in the 1960s–70s,
Correspondence to: action. However, with few exceptions, governments have
Prof Boyd A Swinburn, WHO with an increasing food energy supply now pushing up
Collaborating Centre for Obesity
made very slow progress in the implementation of these energy intake and population weight.
Prevention, Deakin University, strategies. The food and media industries have, by • Adult obesity continues to increase almost universally,
Melbourne, VIC 3125, Australia contrast, moved rapidly by making various national2 and but in some childhood and adolescent populations the
[email protected] international3 pledges, including self-regulatory codes of epidemic seems to be flattening or even decreasing.
practice. Although independent assessment of the true • Present systems for monitoring population weight and
effect of these pledges is needed, governments also need nutrition are inadequate in almost all countries.
to meet their obligations for policy action and leadership, • Obesity is the result of people responding normally to the
which are described in several authoritative reports.1–5 obesogenic environments they find themselves in.
The aim of The Lancet’s Obesity Series is to state the Support for individuals to counteract obesogenic
case for action on obesity: what is the size and nature of environments will continue to be important, but the
the problem, what is driving its global increase, what will priority should be for policies to reverse the obesogenic
the future obesity burden be under a business-as-usual nature of these environments.
scenario, and what action is needed to reverse the • Governments have largely abdicated the responsibility for
epidemic? In this first report in the Series, we describe the addressing obesity to individuals, the private sector, and
obesity epidemic and explain the reasons for its concurrent non-governmental organisations, yet the obesity
rise across countries and the wide variation in obesity epidemic will not be reversed without government
prevalence between countries. The interaction of these leadership, regulation, and investment in programmes,
major determinants of obesity has important implications monitoring, and research.
for the action needed to reverse the epidemic.

804 www.thelancet.com Vol 378 August 27, 2011

 Series

The global rise in obesity prevalence 40 USA

The rise of the obesity epidemic seemed to begin almost England

Childhood overweight, including obese (%)

concurrently in most high-income countries in the 1970s 35 Australia
Chile
and 1980s;6 since then, most middle-income and many 30 Japan
low-income countries have joined the global surge in Brazil
obesity prevalence in adults and children.7–9 By 2008, an 25

estimated 1·46 billion adults globally were overweight 20

(body-mass index [BMI] >25 kg/m²) and 502 million
15
adults were obese (BMI >30 kg/m²).9 Furthermore, an
estimated 170 million children (aged <18 years) globally 10
were classified as overweight or obese.10 This estimate
5
includes more than 25% of all children in some countries,
more than double the proportions from the start of the 0
1974 1980 1986 1992 1998 2004
epidemic (figure 1). Year
Analysis of the patterns of the obesity epidemic in the
past four decades is limited by the absence of representative Figure 1: Estimates of percentage of childhood population overweight, including obese (with use of
International Obesity Taskforce cutoffs) in a selection of countries
data from different countries.11 Nevertheless, the pattern
Based on data from Wang and Lobstein,11 International Association for the Study of Obesity,12 and Matsushita
by which obesity prevalence rises in particular populations and colleagues.13
seems predictable. In low-income and middle-income
countries, groups of high socioeconomic status in urban
areas tend to be the first to have high obesity prevalence, Ethiopia
but the burden of obesity shifts to low socioeconomic Japan
status groups and rural areas as a country’s gross domestic China
product (GDP) increases.14,15 In Brazil, one of the few Yemen
middle-income countries with repeated cross-sectional Hong Kong
surveys of BMI, this pattern was particularly evident for Netherlands
women, with obesity rates increasing rapidly in the lowest
Mongolia
income groups.16 The highest prevalences of overweight
Australia
and obesity are in middle-age groups (45–59 years)
England
throughout this transition.12,17
Jordan
The global rise of obesity has serious health effects.
Mexico
Raised BMI is an established risk factor for diseases such
USA
as type 2 diabetes, cardiovascular diseases, and many
cancers.4,18,19 The disability attributable to obesity and its Egypt

consequences was calculated in 2004 at more than Samoa

36 million disability-adjusted life-years,18 with obesity Tonga

accounting for between 2% and 6% of total health-care 0 10 20 30 40 50 60 70 80
costs in many countries.20 NCDs are now the dominant Prevalence of obesity in adult women (%)

cause of preventable disease burden even in many low- Figure 2: Prevalence of obesity (body-mass index [BMI] >30 kg/m²) in adult women in a selection of countries
income countries,18,21 and obesity has overtaken tobacco as in the 2000s
the largest preventable cause of disease burden in some Data from International Association for the Study of Obesity.17
regions.22 Although the reduction in premature mortality
and morbidity from cardiovascular diseases in high- monitoring urgently needs to improve so that the
income countries during the past 40 years has been progress of the global epidemic can be tracked and
substantial, there is serious concern that the rise of obesity lessons from the experiences of different countries and
and type 2 diabetes will slow or even reverse this trend.23–25 population groups can be learnt.
The increases in overweight and obesity in adults are The available data show very wide variations in obesity
widely projected to continue to heighten the burden of prevalence globally, particularly for women (figure 2). For
obesity-related morbidity and mortality in the coming some populations (eg, China), small body-frame sizes
decades.1,26 However, encouraging reports are emerging mean that a BMI cutoff point of 30 kg/m² for obesity will
from countries such as Sweden, Switzerland, France, and underestimate the amount of over-fatness and comorbid-
Australia that overweight and obesity prevalence in some ities, compared with other populations with larger frame
childhood age groups might be flattening or even sizes (eg, Tonga).28 However, this definitional difficulty
decreasing.27 But, overall prevalence is still high. Crucially, does not account for the 100-times differences between
very few countries have adequate monitoring systems in the populations (0·7% vs 70%). Many of the reasons for
place, which is remarkable in view of the importance of the variations across populations are intuitive. For
this issue. Consequently, the frequency and standard of example, Ethiopia does not have sufficient national

www.thelancet.com Vol 378 August 27, 2011 805

 Series

wealth for obesity to have manifested itself, and countries that still have a substantial burden of
populations in Hong Kong and Jordan have had a greater undernutrition and its related diseases also have a
exposure to obesogenic food environments than do their substantial or emerging burden of overnutrition and its
counterparts in China and Yemen. However, many related NCDs. Both these conditions need to be addressed
complexities exist in understanding why some popu- together for several important reasons: fetal and infant
lations and subpopulations are more susceptible to the undernutrition followed by adult overnutrition has a
drivers of obesity than others, and how mediating factors double effect on the later burden of NCDs;33 the
affect different population groups. underlying drivers within the food system (eg, food
quality and food distribution) are often common to both
Broad economic effects on obesity disorders; and NCDs cannot be ignored even while
The most obvious environmental precondition for a efforts to reduce undernutrition continue.
population to develop obesity is sufficient wealth. The In the same way as obesity is the result of people
relation between GDP and mean BMI is positive and linear responding normally to the obesogenic environments
up to a GDP of about US$5000 per person per year; at that they find themselves in, so too do these obesogenic
greater GDP, the relation with GDP and BMI is almost environments arise because businesses and governments
flat.29 A degree of economic prosperity is thus an enabler are responding normally to the broader economic and
for obesity, but the level of prosperity does not have to be political environments that they find themselves in. A
high for obesity to manifest; in some low-income countries, central tenet of modern, market-based economies is the
such as Pacific Island nations, obesity prevalence is very benefits of economic growth; and a parallel tenet of
high.17 A return to national poverty is not a recommended business and trade is the benefits of more liberalised,
approach to reduce obesity and type 2 diabetes but, as seen less regulated global markets. Economic growth is
in Cuba and Nauru,30,31 it can have that effect. especially important for low-income countries to move
The economic transition towards increasing GDP them from poverty to economic prosperity; however, for
brings with it several other transitions: demographic high-income countries, higher levels of GDP do not bring
(younger to older population distribution, rural to urban); greater happiness and wellbeing for their citizens but do
epidemiological or health (infectious diseases to NCDs); bring greater consumption of all products.34 The
technological (low to high mechanisation and motor- technological changes that are creating cheaper and more
isation); and nutritional (traditional foods and cuisines to available food calories and the strong economic forces
more processed energy-dense foods).32 The pace of driving consumption will inevitably lead to over-
change of these transitions has increased substantially in consumption and obesity.35–37
recent decades; so many countries in transition are faced In the broader view, obesity is similar to rising
with double burdens of disease. For example, most greenhouse gases and environmental degradation as yet
another detrimental effect of individual and corporate
overconsumption.34 The pressure for market
Panel 1: Is the market failing children? liberalisation means that regulatory approaches,
A market fails when prices and the quantities bought and sold are no longer indicative of although feasible, are difficult to achieve—as exemplified
their costs and benefits to society.46 Is the market failing children? The first of four reasons in the great reluctance of policymakers to regulate
for market failure is when vulnerable individuals are not protected.47 Children are clearly a reductions in marketing of obesogenic foods and
vulnerable group that warrant societal protection, and this notion represents the strongest beverages, such as fast foods and sugar-sweetened
argument for government intervention. They are not mature, they do not have nutritional drinks, to children. There are many reasons for
knowledge, are unable to perceive the risks of their behaviour, and their choices are readily government intervention to restrict marketing to
affected by marketing.47–49 The second reason is when consumers do not have the children, including protection of the rights of children,38
information necessary to make fully informed decisions about their food selection,47, 50 as is public demand for regulations,39 and application of the
also clearly the case with children. However, generally, interventions to rectify information precautionary principle of preventive action early, even
gaps seem to have modest effects.46 The third reason for market failure is when people before absolute proof is available.40 Although obesity has
prioritise immediate gratification over potential long-term negative results, which is a been described as “a sign of commercial success but a
hallmark of childhood. The final reason relates to spill-over effects (or externalities), when market failure”41 debate exists about whether market
the costs of obesity are borne by society. Although yearly health-care costs to the taxpayer failure provides an additional argument for government
are higher for obese than for non-obese people, reduced life expectancy due to obesity intervention with respect to prevention of childhood
makes it uncertain whether the life-time social costs are actually higher.42,51 Externalities obesity41–45 (panel 1).
might arise at the family level through reduced household income or additional carer
duties.46 Thus, there is ample justification for protecting children’s health from the predatory Drivers of the obesity epidemic
effects of markets, yet almost universally, governments are failing in this responsibility. The We define a driver of the global obesity epidemic as an
charge of so-called nannyism almost inevitably arises42 in relation to regulatory environmental factor that has changed substantially during
interventions, yet for children, and even for adults, governments have a fundamental role in the past 40 years (coinciding with the upswing of the
helping to make healthy choices the easy choices.52 epidemic), is global in nature (affecting almost all countries
with enabling economic conditions), and is rapidly

806 www.thelancet.com Vol 378 August 27, 2011

 Series

transmissible (in view of the near simultaneous nature of drivers of the epidemic if they have not changed
the epidemic across countries). Some environmental sufficiently and coincidentally with the onset of the
determinants of obesity, such as the built environment, obesity epidemic. For example, if the availability and
can have important effects on behaviours;53 however, built promotion of cheap, energy-dense food increases globally
environments have not changed simultaneously and throughout several decades, the traditionally high levels
universally to become more obesogenic during the past of active transport in the Netherlands will help to
few decades. The built environment is thus unlikely to attenuate this obesogenic effect in the Dutch population;
have been a major driver of the global epidemic, although whereas the traditionally high levels of car transport in
the way in which people have responded to the built the USA will accentuate this obesogenic effect in the
environment (eg, increased traffic congestion) has changed American population. Similarly, cultural body-size
with time and might be important. The obvious possible preferences could moderate the drivers such that
drivers of the epidemic are in the food system:35 the increased food access will probably have a larger effect in
increased supply of cheap, palatable, energy-dense foods;
improved distribution systems to make food much more
accessible and convenient; and more persuasive and Panel 2: Energy balance flipping point
pervasive food marketing.54
In most high-income countries, the energy expenditure needed for daily life has
Several studies have tested the hypothesis that increases
decreased since the beginning of the 20th century because of increasing mechanisation,
in the food supply are the dominant drivers of the weight
urbanisation, motorisation, and computerisation. All else being equal, the expected result
gain in populations.55–57 Results from these investigations
of an increasingly sedentary lifestyle would have been weight gain. Therefore, why did
show that the rise in food energy supply was more than
obesity prevalence not rise substantially until the 1970s?
sufficient to explain the rise in obesity in the USA from
the 1970s,55,56 and most of the weight increase in the UK A parallel reduction of food energy intake coupled with decreasing physical activity is one
since the 1980s.57 A related hypothesis is that the policies possible explanation for the low prevalence of obesity during the first three-quarters of the
put in place in the USA and other countries to increase 20th century. Data from the US Department of Agriculture lend support to this explanation
the food supply from the 1970s led to a situation in which since they clearly show a reduction of per-person energy available in the food supply (food
the abundance of food in these countries began to push production plus imports minus exports and non-human use) from early in the 20th century
up population energy intake—a reversal of the previous until the 1960s58 (figure 3), mainly because of reduced consumption of wheat products.59,60
situation in which energy intake was pulled down by A 1948 editorial in the Journal of the American Medical Association61 attributed the fall in US
decreases in physical activity (panel 2, figure 3). wheat consumption partly to the decreased energy demands resulting from reduction of
Figure 4 shows the key drivers of the global obesity hard labour. Presciently, the same editorial warned of a potential rise in obesity prevalence if
epidemic and presents an overview framework for consumption of sugars, sweets, and fats began to increase. Indeed, the 1970s saw a striking
understanding of population-level obesity determinants rise in the quantity of refined carbohydrates and fats in the US food supply,62,63 which was
and solutions. Our framework has features in common paralleled by a sharp increase in the available calories (figure 3) and the onset of the obesity
with other frameworks for obesity determinants26,65— epidemic. As Cutler and colleagues35 note, there was “a revolution in the mass preparation
eg, the layered levels of determinants that recognise of food that is roughly comparable to the mass production revolution in manufactured
that the physiology of energy balance is proximally goods that happened a century ago,” that “lowered the time price of food consumption”.
determined by behaviours and distally by environments. The increased availability and marketing of cheap, readily available food was so great that
A further distinction is made between the obesogenic food waste has progressively increased by about 50% since the 1970s.55
drivers within the food and physical activity environ- We postulate that an energy balance flipping point has occurred in most high-income
ments, which are proximal determinants of behaviours, countries in the past century with two distinct phases: the so-called move less, stay lean
and more distal systemic drivers. Taxation regimes, phase (1910–60), characterised by decreasing physical activity levels and energy intake,
regulation of the marketplace, and social and economic and a population that remained lean; and the subsequent so-called eat more, gain weight
policies set the conditions under which businesses and phase, characterised by increasing energy intake and a concomitant rise in population
individuals operate, and many of these conditions can weight. In each of these phases, we postulate that energy balance was predominantly
have distal effects on obesity.34,66 For example, indepen- achieved by different mechanisms: decreasing energy intake (through appetite
dent of the overall wealth of a country or state, the mechanisms) to match decreasing expenditure, followed by increasing energy
higher the level of income inequality, the higher the expenditure (through increased weight and, thus, resting metabolic rate) to match
prevalence of obesity.67 These distal effects might increasing intake. Thus, decreases in physical activity in the first phase were probably able
convert to higher obesity prevalence through many to pull down energy intake, because intake was being matched to expenditure. The result
pathways, such as through psychosocial and beha- was that obesity rates did not increase during this period despite widespread uptake of
vioural effects.67 mechanisation and motorisation. The second phase seems to have been ushered in by an
Our framework (figure 4) recognises the importance of energy balance flipping point, when energy intake rose because of environmental push
environmental conditions that operate on a population to factors (ie, increasingly available, cheap, tasty, highly promoted obesogenic foods). The
accentuate or attenuate the effect that the drivers have on concomitant rise in weight was the physiological mechanism for restoration of energy
the trajectory of changes in obesity prevalence. These balance.64 Food supply data from the US lends support to this flipping point hypothesis,
moderators or modulators, although important in but it needs to be tested in other countries and with more diverse datasets.
affecting the slope of the rise of obesity, cannot be deemed

www.thelancet.com Vol 378 August 27, 2011 807

 Series

Tongan women, where large body size is a positive However, we recognise that this scheme is simplified
attribute,68 than in Japanese women, where small body and that factors such as active transport environments
size is deemed ideal.69 and body-size preferences could change sufficiently
The effects of the environmental moderators are with time to drive changes in obesity prevalence. Indeed,
shown in figure 4 as affecting the rise in obesity but are for some countries such as China, rapid urbanisation
not depicted as part of the direct driver pathways. and motorisation of the population could be judged a
major driver of the rise in obesity.70 Although the
18 000 interaction between all the environmental factors is
undoubtedly complex and multifaceted (panel 3), the
16 000 strong directional force of the systemic and environ-
mental drivers is evidenced by the continued rise of the
14 000 epidemic globally.
Food availability per person (kJ per day)

Pull phase: energy expenditure Push phase: energy intake

12 000 driving energy intake driving energy expenditure Other environmental and individual effects
The existing environments within a country (eg, the built
10 000 environment, transport systems, active recreation
opportunities, cuisines and food culture, and culture
8000 around body size) can greatly moderate or modulate the
effects of the global obesity drivers on population BMI
6000 Energy balance (figure 4). These effects can be powerful and help to
flipping point
explain much of the differences in obesity prevalence
4000 between populations. They clearly hold opportunities for
interventions to make environments less obesogenic.
2000 Interventions such as increasing the price of unhealthy
food and beverages81–83 or decreasing the price of healthy
0 foods84 have received attention in recent years, but there
1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year has been very little research into understanding and
changing the powerful sociocultural determinants of
Figure 3: Food availability for the USA, 1910–200658 food choices, physical activity, and body-size perception.85
There are two distinct phases: a decrease in food energy supply (postulated to be pulled down by reduced energy
These determinants should be a priority for research if
expenditure requirements for daily living), followed by an increase in food energy supply (postulated to be pushed
up by increasing food access). An energy balance flipping point is proposed, marking the change in how the US the high prevalence of obesity in specific ethnic groups is
population generally achieved energy balance. to be addressed.86

Environments Behaviours Physiology

Environmental moderators

Systemic Environmental Behaviour Energy

drivers drivers patterns imbalance

Policy and economic Food supply and Sociocultural, socioeconomic, High food and energy High total energy
systems enable and marketing recreation, and transport consumption with intake pushing
promote high growth environments environments which amplify associated low energy imbalance
and consumption promote high or attenuate the drivers physical activity levels
energy intake

Policy interventions Drugs, surgery

Health promotion programmes, social marketing, etc

Population effect and political difficulty

Figure 4: A framework to categorise obesity determinants and solutions

The more distal drivers are to the left and the environmental moderators that have an attenuating or accentuating effect are shown, along with some examples. The
usual interventions for environmental change are policy based, whereas health promotion programmes can affect environments and behaviours. Drugs and surgery
operate at the physiological level. The framework shows that the more upstream interventions that target the systemic drivers might have larger effects, but their
political implementation is more difficult than health promotion programmes and medical services.

808 www.thelancet.com Vol 378 August 27, 2011

 Series

Within a given environment, BMI differs between

individuals, which is indicative of individual moderators Panel 3: Modelling for obesity
or modulators of energy balance. For example, too little The Obesity System Map,71 introduced by the UK Foresight Programme in 2007, is the
sleep and impaired sleep quality have been shown to most comprehensive compilation so far of the determinants of obesity and their
have important effects on weight gain.87 Genetic effects inter-relations. The map clearly shows the enormous complexity of the causal relations
are one of the most highly researched determinants of relevant to obesity by linking physiology, individual behaviours, and environmental
obesity. However, apart from the identification of a few variables that stretch from local to global levels. In addition to the interdependence
rare, single genetic abnormalities, the quest to identify between factors and the importance of feedback loops, there are other characteristics of
genes related to common obesity (obesity not caused by the obesity problem that are not shown. These characteristics include the heterogeneity
an underlying condition) has been disappointing. The of the factors that need to be involved in systems change, the diversity of environments
potential epigenetic effects of behavioural and environ- where change needs to take place, the non-linear relations that govern the system’s
mental factors on genetic expression are now receiving dynamics, and the sometimes random or quantum nature of behaviour change.72 This
much attention (panel 4). complexity creates an apparently intractable or so-called wicked problem,73 and the
The role of individual choice in both the causes of and imperative to turn around the obesity epidemic clashes with its wicked nature. In aiming
solutions to obesity has great appeal because of its to establish and prioritise solutions, the challenge is to reduce the complexity of obesity
simplicity. However, this notion is in fact highly enough so that it can be understood by researchers, policy makers, and the public without
contentious and is indicative of the tension between the becoming overly simplistic. Some progress in this challenge lies in mathematical
simple and the complex that is ever present in debates modelling applied to complex health and behavioural questions.64,74,75
about obesity. Undoubtedly the final decision to
In the past 5 years, several mathematical models have been developed by use of various
consume a particular food or beverage, or to exercise or
modelling techniques.76 As with other models, they can have descriptive, explanatory, or
not, is an individual decision. However, to negotiate the
evaluative aims.77 Figure 5 shows the aims for obesity modelling at the population level.
complexity of the environment and the choices it poses,
Descriptive studies quantify the present burden and potential future trends of obesity.
many of these decisions are automatic or subconscious.
Explanatory studies analyse the causes of the rise in obesity prevalence with time and the
Cohen98 puts forward a strong argument that “excessive
variability across populations. Evaluative models assess the likely effect of interventions
food consumption occurs in ways that defy personal
to reduce future prevalence.
insight or are below individual awareness”, and provides
plausible neurophysiological mechanisms for many Mathematical models are “a set of assumptions together with implications drawn from
automatic responses to physiological, environmental, them by mathematical reasoning”.74 Compared with conceptual models, they have the
and interpersonal cues in relation to eating and physical valuable attributes of forcing theoretical precision, making assumptions transparent,
activity opportunities. Appetite physiologists label this promoting data analysis and hypothesis testing, and having many practical applications.74
response as “passive overconsumption”.99 Further Since models match reality in important ways but are far simpler than reality (they draw
increase in choice, especially through more processed attention to some aspects of reality but ignore others78), they can help us understand
products on supermarket shelves or more items on fast complex problems such as obesity and its solutions.72 Indeed, mathematical modelling has
food menus, is unlikely to reduce overconsumption and been crucial to understand and to respond to other global threats, such as climate change79
could have the opposite effect. Furthermore, cognitive and infectious diseases.80
strategies to combat overconsumption, such as weight- Irrespective of their goal, all mathematical models need a logical framework linking the
loss diets, can be successful for some individuals but pathways between each of the factors in the model, accompanied by tight specifications
are unlikely to be population solutions. Even weight-loss of the assumptions and quantitative estimates underpinning each of the links. Complex
diets that are supported by trial data, such as high frameworks, such as the Foresight Obesity System Map, attempt to incorporate most
protein or low glycaemic index diets,100 might not be individual and population-level determinants. However, substantial simplification of
suitable as solutions for global obesity because of their models can arise when only populations are included (determinants causing individual
detrimental effect on the environment (eg, high meat variability drop out), when questions are narrowed (determinants that have not changed
protein diets101) or staple food production (eg, rice). with time can be dropped for questions about the rise of obesity), or when solutions are
being modelled (factors not included in the intervention can be dropped). These methods
Approaches to address the obesity epidemic can be expected to provide the core techniques to understand and to respond to complex
Approaches to address the obesity epidemic are broadly non-communicable disorders such as obesity.51,64,75
categorised within figure 4, with a detailed discussion of
solutions presented in the report by Gortmaker and
colleagues75 in this Series. the costs of unhealthy foods). The interventions to
The effects on population trajectories of obesity motivate behavioural changes could be regarded as
prevalence are likely to fundamentally differ between an counteractions (ie, they counteract drivers of increasingly
intervention aimed at motivating behavioural changes obesogenic environments by acting on some of their
(eg, health promotion programmes, social marketing, mediators) and they might have important obesity
education) and policy interventions (in this context, prevention effects, especially in children, if applied to a
meaning enforceable actions such as laws and whole community.102,103 However, sustainability and
regulations) that reverse the environmental drivers affordability are the two major continuing challenges,
(eg, reducing the cost of healthy foods and increasing even for programmes with proven effectiveness.

www.thelancet.com Vol 378 August 27, 2011 809

 Series

Furthermore, such programmes do not address the

underlying drivers of the epidemic.
Interventions that aim to reverse obesogenic drivers
(and some of the environmental moderators) will almost
all be policy-led—mainly government policy (eg, shifting
2
agricultural polices to incorporate health outcomes,
1 banning unhealthy food marketing to children, healthy
public sector food service policies) but some could be
5 food industry policies (eg, moving product formulation
Obesity prevalence

4
towards healthier compositions, self-regulation of
marketing to children). Policy-led solutions that apply to
environments and affect the whole population have
3
several strengths compared with health education and
promotion programmes.104 They tend to be sustainable,
affect the whole population (including those who are
difficult to reach), become systemic (affect default
behaviours), and reverse some of the environmental
drivers. The degree of political difficulty for imple-
mentation of policy and regulatory interventions is
Time (decades)
typically much higher than that for programme-based
Figure 5: A schematic diagram of the major uses of modelling of population prevalence of obesity and education-based interventions105 (figure 4). Reasons
The black line is the trajectory of increases in obesity prevalence with time, with the present burden (1) and for this reluctance to enact affordable and cost-effective
projected future burdens (2) being descriptive uses of modelling. Explanatory uses of modelling include explaining policies include the powerful lobby force of the food (and
the rise in obesity with time (3) and the differences in prevalence rates (4). Evaluative uses include assessment of allied) industries against government regulation of the
the potential for solutions to reduce the future burden of obesity (5).
food market106,107 and public reluctance to change
environments to which they have become accustomed
(such as car access and cheap parking in cities, and high
Panel 4: How much can genetics explain of the obesity epidemic? fat and sugar food choices in canteens). Nevertheless, the
One useful way to think about the relation of genes with obesity was expressed by experience with trans-fatty acids in Denmark, where
George Bray when he said, “the genetic background loads the gun, but the legislation was introduced to restrict their use in food
environment pulls the trigger”.88 The rise of obesity prevalence throughout the past production, is an example of a cost-effective government
few decades clearly cannot be accounted for by population genetic changes. The food policy that was successfully enacted for population
heritability of body-mass index (BMI) is often cited as 40–70%,89 yet large health benefit.108
genome-wide association studies to identify common single nucleotide Policy interventions for obesity can only be realistically
polymorphisms (SNPs) associated with BMI have been unable to explain more than a directed at the environment (making healthy choices
small proportion (<2%) of BMI variability.90,91 Heritability is often misinterpreted as easier) rather than the individual (compelling them to
being the proportion of BMI variance caused by genetics; whereas heritability studies, take the healthy choices). Unlike other public health
rather than estimating cause, estimate the proportion of genetic variance and non- issues for which enforceable policies can directly require
genetic variance that explains BMI variance in a given study population. The study specific behaviours (such as wearing a seat belt or not
populations are usually based on twin cohorts which, even if reared apart, live within smoking in offices), there are no regulations that will
very narrow environmental variances (rather than the wide variance of environments require people to eat, or not eat, certain foods and to
across countries shown in figure 2), which results in high heritability estimates. These exercise,66 with the possible exception of a few rules
estimates seriously understate the environmental effects and thus provide no operating in school environments. For this reason,
indication of the potential to reduce obesity via environmental change.92 Part of the obesity prevention policies do not proscribe particular
missing heritability of obesity might be the result of small individual contributions of eating and physical activity behaviours and are thus
many SNPs that genome-wide association studies have been underpowered to detect, much less intrusive of human liberties than many
or some rare variants that are not identified by common SNPs. Much of the variability policies already in place to control other public health
between individuals in bodyweight might be attributable to gene–environment or problems. The major strategies available to directly affect
gene–behaviour interactions,93 including interactions in the intrauterine behaviours aim to increase the motivation to make
environment.94 An intriguing possibility mediating the gene–environment interaction healthy choices, and include social marketing, health
is the potential contribution of epigenetic mechanisms that modify the expression of education, and health promotion programmes.
genes. For example, DNA methylation patterns can be affected by maternal diet; these
epigenetic modifications can persist for decades,95,96 and possibly be inherited by future Implications
generations.97 Epigenetic contributions to obesity will need to be addressed through In this report, we have provided an overview of the
minimisation of the environmental triggers, rather than manipulation of the genetic global obesity epidemic, describing the size and nature
guns, in the first instance. of the problem, discussing its drivers, and mapping out
key contributing factors. We draw attention to the food

810 www.thelancet.com Vol 378 August 27, 2011

 Series

system, operating through the energy intake side of the However, recognition in low-income and middle-income
energy balance equation, as the dominant driver of the countries that the growing contribution of obesity to its
rise in obesity, although many other environmental and burden of malnutrition is a threat to national development
individual factors modulate the effect of the drivers on remains low. Some governments are making promising
obesity prevalence in populations and obesity presence moves to convert the rhetoric into action. For example, in
in individuals. The highly complex global, national, and the mid to late 2000s, the Government in England
local food systems, although feeding the world’s developed a series of initiatives such as the Foresight
population, are nevertheless falling short by both report,26 the cross government strategy for obesity,118
promoting overnutrition (overweight or obesity) in an restrictions on unhealthy food marketing to children,119
estimated 2 billion adults and children9,10 and not and a yearly national child measurement programme.118
reaching the further 1 billion people who are Some other countries, notably Brazil, have made
undernourished.109 Both these forms of malnutrition substantial steps in national monitoring programmes,
are major contributors to preventable population restricting marketing to children, and improving school
disease burden and are thus substantial barriers to food; and the US White House Task Force on Childhood
meeting the Millennium Development Goals.110 With Obesity120 holds much promise to bring about change,
the global population expected to grow to 9 billion because of its high-level political commitment.
people by 2050,111 and with increased disruption to food International leadership from the UN and its agencies,
production from climate change—such as increased and more national leadership from countries will be
droughts, floods, hurricanes, and global tempera- essential to solve this great challenge of the global obesity
tures101—the global food system will be placed under epidemic, which is why the UN High Level Meeting on
great stress which will probably result in more unequal NCDs in September, 2011, is so central to global
distribution of calories. Policies that address the food population health and development.
system drivers of obesity will therefore be an important Contributors
part of the much broader imperative to re-orient food All authors jointly formulated the major ideas, and read and approved
systems towards health and development, particularly the final version of the report. BAS led the writing of the report.
GS assisted in writing the report, and collated and included comments
in vulnerable populations. from all authors. KDH co-drafted the sections about genetics and the
As argued in this report and elsewhere,34 the economic food supply and commented on the remainder of the paper. MLM
priorities and policies that promote consumption-based co-drafted the section about market failure and commented on the
growth, and the regulatory policies that promote market remainder of the report. KM, DTF, and SLG provided comments on
the report.
and trade liberalisation have produced many benefits
but are now increasingly regarded as contributing to Conflicts of interest
We declare that we have no conflicts of interest.
the global crises of overconsumption in general. Obesity
is but one of these crises, as the private sector becomes Acknowledgments
This work was done under auspices of the Collaborative Obesity
ever more effective in its exploitation of basic human Modeling Network as part of the Envision Project, supported by the
biological drives, desires, and weaknesses. Solutions to National Collaborative on Childhood Obesity Research, which
obesity and to improve health and development cannot coordinates childhood obesity research across the National Institutes of
be based on the existing framework (consumption- Health (NIH), Centers for Disease Control and Prevention (CDC),
Department of Agriculture, and the Robert Wood Johnson Foundation
driven growth creating financially-defined prosperity) (RWJF). This work was supported in part by grants from RWJF (grant
because this approach has helped to create the numbers 260639, 61468 and 66284), CDC (U48/DP00064-00S1 and
difficulties in the first place. Governments and 1U48DP001946), including the Nutrition and Obesity Policy, Research
and Evaluation Network, and the Office of Behavioral and Social
international organisations such as the UN need to
Sciences Research of NIH. This work is solely the responsibility of the
provide global leadership on these issues and not authors and does not represent official views of the Centers for Disease
abdicate them to the private sector.112,113 Actions that Control and Prevention or any of the other funders.
reduce the economic incentives for over consumption References
in general (eg, a price on carbon and subsidies on 1 WHO. Global strategy on diet, physical activity and health. Geneva:
public transport) can have benefits for both human and World Health Organization, 2004. http://www.who.int/
dietphysicalactivity/en/ (accessed June 23, 2011).
environmental health.34 Moreover, a new framework is 2 Peeler CL, Kolish ED, Enright M, Burke C. The children’s food &
needed that is based on broadly defined sustainable beverage advertising initiative in action. A report on compliance
economics for prosperity (including economic, social, and implementation during 2009. Arlington: Council of Better
Business Bureaus, 2010.
health, and environmental outcomes).114 3 European Union. EU Pledge 2009 Monitoring Report,
Four decades after the onset of the global obesity September 09. Accenture Media Services, 2009.
epidemic, the awareness of the threat of obesity to 4 World Cancer Research Fund and American Institute for Cancer
Research. food, nutrition, physical activity, and the prevention of
population health and wellbeing can be seen in the cancer: a global perspective, 2007. Washington, DC: American
plethora of national reports and strategic plans from Institute for Cancer Research, 2007.
many countries, yet their conversion to action remains 5 Koplan J, Liverman C, Kraak V, eds. Preventing childhood obesity:
health in the balance. Washington, DC: Institute of Medicine of the
largely unrealised (Australia has had three such National Academies, The National Academies Press, 2005.
unfulfilled national taskforce plans since 1997115–117).

www.thelancet.com Vol 378 August 27, 2011 811

 Series

6 Sassi F, Devaux M, Cecchini M, Rusticelli E. The obesity epidemic: 28 Rush E, Plank L, Chandu V, et al. Body size, body composition,
analysis of past and projected future trends in selected OECD and fat distribution: a comparison of young New Zealand men of
countries. Paris: Organisation for Economic Co-operation and European, Pacific Island, and Asian Indian ethnicities. NZ Med J
Development (OECD), Directorate for Employment, Labour And 2004; 117: U1203.
Social Affairs, Health Committee, 2009. 29 Lobstein T, Leach R. Foresight. Tackling obesities: future choices—
7 WHO. Reducing risks, promoting healthy life. Geneva: World international comparisons of obesity trends, determinants and
Health Organization, 2005. responses—evidence review, adults. London: Government Office
8 WHO. Global strategy on diet, physical activity and health: for Science, 2007.
a framework to monitor and evaluate implementation. Geneva: 30 Keke K, Phongsavan P, Dan L, et al, for Nauru Ministry of Health,
World Health Organization, 2008. WHO, and the Centre for Physical Activity and Health, University
9 Finucane MM, Stevens GA, Cowan MJ, et al. National, regional, of Sydney. Nauru NCD Risk Factors STEPS Report. Suva, Fiji:
and global trends in body-mass index since 1980: systematic World Health Organisation, 2007.
analysis of health examination surveys and epidemiological studies 31 Franco M, Orduñez P, Caballero B, et al. Impact of energy intake,
with 960 country-years and 9·1 million participants. Lancet 2011; physical activity, and population-wide weight loss on cardiovascular
377: 557–67. disease and diabetes mortality in Cuba, 1980–2005. Am J Epidemiol
10 Lobstein T, Baur L, Uauy R, for the IASO International Obesity 2007; 166: 1374–80.
TaskForce. Obesity in children and young people: a crisis in public 32 Popkin B. The nutrition transition and its health implications
health. Obes Rev 2004; 5 (suppl 1): 4–104. in lower-income countries. Public Health Nutr 1998; 1: 5–21.
11 Wang Y, Lobstein T. Worldwide trends in childhood overweight and 33 Barker DJ, Winter PD, Osmond C, Margetts B, Simmonds SJ.
obesity. Int J Pediatr Obes 2006; 1: 11–25. Weight in infancy and death from ischaemic heart disease.
12 International Association for the Study of Obesity, International Lancet 1989; 334: 577–80.
Obesity Taskforce. Global childhood overweight. http://www.iaso. 34 Egger G, Swinburn B. Planet obesity: how we’re eating ourselves
org/site_media/uploads/Global_Childhood_Overweight_ and the planet to death. Sydney: Allen and Unwin, 2010.
April_2011.pdf (accessed June 23, 2011). 35 Cutler DM, Glaeser EL, Shapiro JM. Why have Americans become
13 Matsushita Y, Yoshiike N, Kaneda F, Yoshita K, Takimoto H. Trends more obese? J Econ Perspect 2003; 17: 93–118.
in childhood obesity in Japan over the last 25 years from the 36 Philipson TJ, Posner RA. The long-run growth in obesity as
national nutrition survey. Obes Res 2004; 12: 205–14. a function of technological change. Perspect Biol Med 2003;
14 Monteiro CA, Moura EC, Conde WL, Popkin BM. Socioeconomic 46 (suppl 3): S87–107.
status and obesity in adult populations of developing countries: 37 Finkelstein EA, Ruhm CJ, Kosa KM. Economic causes and
a review. Bull World Health Organ 2004; 82: 940–46. consequences of obesity. Annu Rev Public Health 2005; 26: 239–57.
15 Mendez MA, Monteiro CA, Popkin BM. Overweight exceeds 38 Swinburn B, Sacks G, Lobstein T, et al. The ‘Sydney Principles’
underweight among women in most developing countries. for reducing the commercial promotion of foods and beverages
Am J Clin Nutr 2005; 81: 714–21. to children. Public Health Nutr 2008; 11: 881–86.
16 Monteiro CA, Conde WL, Popkin BM. Income-specific trends in 39 Morley B, Chapman K, Mehta K, King L, Swinburn B,
obesity in Brazil: 1975–2003. Am J Public Health 2007; 97: 1808–12. Wakefield M. Parental awareness and attitudes about food
17 International Association for the Study of Obesity, International advertising to children on Australian television.
Obesity Taskforce. Global obesity prevalence in adults. http://www. Aust NZ J Public Health 2008; 32: 341–47.
iaso.org/site_media/uploads/Prevalence_of_Adult_Obesity_ 40 Tickner J, Raffensperger C, Myers N. The precautionary principle
May_2011_New.pdf. accessed 23/6/2011 (accessed June 23, 2011). in action: a handbook. Windsor, ND: Science and Environmental
18 Ezzati M, Lopez A, Rodgers AD, Murray CJL, eds. Comparative Health Network, 1999.
quantification of health risks: global and regional burden of disease 41 Moodie R, Swinburn B, Richardson J, Somaini B. Childhood
attributable to selected major risk factors. Geneva: World Health obesity—a sign of commercial success, but a market failure.
Organization, 2004. Int J Pediatr Obes 2006; 1: 133–38.
19 Australian Institute of Health and Welfare (AIHW) and National 42 McCormick B, Stone I, for the Corporate Analytical Team.
Heart Foundation of Australia. The relationship between Economic costs of obesity and the case for government
overweight, obesity and cardiovascular disease. Canberra: AIHW intervention. Obes Rev 2007; 8 (suppl 1): 161–64.
(Cardiovascular Disease Series No. 23), 2004. 43 Jan S, Mooney GH. Childhood obesity, values and the market.
20 WHO. Prevention and control of noncommunicable diseases: Int J Pediatr Obes 2006; 1: 131–32.
implementation of the global strategy. Sixty-first World Health 44 Anand P, Gray A. Obesity as market failure: could a ‘deliberative
Assembly, Resolution WHA61.14. Geneva: World Health economy’ overcome the problems of paternalism? Kyklos 2009;
Organisation, 2007. 62: 182–90.
21 Lopez A, Mathers C, Ezzati M, Jamison D, Murray C. 45 Kuchler F, Golan E. Is there a role for government in reducing
Global burden of disease and risk factors. Washington: The World the prevalence of overweight and obesity? Choices 2004; Fall: 41–45.
Bank, 2006. http://www.dcp2.org/pubs/GBD (accessed
46 Freebairn J. Taxation and obesity? Aust Econ Rev 2010; 43: 54–62.
June 23, 2011).
47 Mazzocchi M, Traill W, Shogren J. Fat economics: nutrition, health
22 Hoad V, Somerford P, Katzenellenbogen J. High body mass index
and economic policy. Oxford: Oxford University Press, 2009.
overtakes tobacco as the leading independent risk factor
contributing to disease burden in Western Australia. 48 McCarthy M. The economics of obesity. Lancet 2004; 364: 2169–70.
Aust NZ J Public Health 2010; 34: 214–15. 49 Institute of Medicine. Food marketing to children and youth. Threat
23 Stewart ST, Cutler DM, Rosen AB. Forecasting the effects of obesity or opportunity? Washington: National Academy of Sciences, 2006.
and smoking on U.S. life expectancy. N Engl J Med 2009; 50 Sassi F. Obesity and the Economics of Prevention. Fit not fat. Paris:
361: 2252–60. Organisation for Economic Co-operation and Development, 2010.
24 Olshansky SJ, Passaro DJ, Hershow RC, et al. A potential decline in 51 Wang Y, McPherson K, Marsh T, Gortmaker SL, Brown M. Health
life expectancy in the United States in the 21st century. N Engl J Med and economic burden of the projected obesity trends in the US
2005; 352: 1138–45. and the UK. Lancet 2011; 378: 815–25.
25 Ford E, Ajani U, Croft J, Critchley J, Labarthe D, Kottke T. 52 WHO. Ottawa Charter for Health Promotion. Ottawa: First
Explaining the decrease in U.S. deaths from coronary disease, International Conference on Health Promotion; 1986. Report No:
1980–2000. N Engl J Med 2007; 356: 2388–98. WHO/HPR/HEP/95.1.
26 Butland B, Jebb S, Kopelman P, et al. Foresight. Tackling obesities: 53 Frank LD, Andresen MA, Schmid TL. Obesity relationships with
future choices—project report. London: Government Office for community design, physical activity, and time spent in cars.
Science, 2007. Am J Prev Med 2004; 27: 87–96.
27 Rokholm B, Baker J, Sorensen T. The levelling off of the obesity 54 Kitchen P, Brignell J, Li T, Spickett-Jones G. The emergence
epidemic since the year 1999—a review of evidence and of IMC: a theoretical perspective. J Advertising Res 2004;
perspectives. Obes Rev 2010; 11: 835–46. March: 19–30.

812 www.thelancet.com Vol 378 August 27, 2011

 Series

55 Hall KD, Guo J, Dore M, Chow CC. The progressive increase 81 Sacks G, Veerman JL, Moodie M, Swinburn B. ‘Traffic-light’
of food waste in America and its environmental impact. PLoS ONE nutrition labelling and ‘junk-food’ tax: a modelled comparison
2009; 4: e7940. of cost-effectiveness for obesity prevention. Int J Obes (Lond) 2010;
56 Swinburn B, Sacks G, Ravussin E. Increased food energy supply published online Nov 16. DOI:10.1038/ijo.2010.228.
is more than sufficient to explain the US epidemic of obesity. 82 Mytton O, Gray A, Rayner M, Rutter H. Could targeted food taxes
Am J Clin Nutr 2009; 90: 1453–56. improve health? J Epidemiol Community Health 2007; 61: 689–94.
57 Scarborough P, Burg MR, Foster C, et al. Increased energy intake 83 Brownell KD, Frieden TR. Ounces of prevention—the public policy
entirely accounts for increase in body weight in women but not in case for taxes on sugared beverages. N Engl J Med 2009;
men in the UK between 1986 and 2000. Br J Nutr 2011; 360: 1805–08.
105: 1399–404. 84 Ni Mhurchu C, Blakely T, Jiang Y, Eyles HC, Rodgers A. Effects
58 Economic Research Service, US Department of Agriculture. of price discounts and tailored nutrition education on supermarket
Food availability (per capita) data system. Washington, DC: US purchases: a randomized controlled trial. Am J Clin Nutr 2010;
Department of Agriculture. http://www.ers.usda.gov/Data/ 91: 736–47.
FoodConsumption (accessed June 23, 2011). 85 McCabe MP, Ricciardelli LA. Parent, peer and media influences on
59 Gerrior S, Bente L, Hiza H. Nutrient content of the U.S. Food body image and strategies to both increase and decrease body size
Supply. 1909–2000. Washington, DC: US Department of among adolescent boys and girls. Adolescence 2001; 36: 225–40.
Agriculture, Center for Nutrition Policy and Promotion, 2004. 86 Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ,
60 Putnam J. Major trends in the U.S. food supply, 1909–99. Food Rev Flegal KM. Prevalence of overweight and obesity in the United
2000; 23: 8–15. States, 1999–2004. JAMA 2006; 295: 1549–55.
61 Anonymous. Nutritional contributions of wheat. JAMA 1948; 87 Chaput JP, Despres JP, Bouchard C, Tremblay A. The association
138: 972–73. between sleep duration and weight gain in adults: a 6-year
62 Putnam J, Allshouse J, Kantor L. US per capita food supply trends: prospective study from the Quebec Family Study. Sleep 2008;
more calories, refined carbohydrates, and fats. Food Rev 2002; 31: 517–23.
25: 2–15. 88 Bray GA. The epidemic of obesity and changes in food intake:
63 Putnam J KL, Allshouse J. Per capita food supply trends: progress the Fluoride Hypothesis. Physiol Behav 2004; 82: 115–21.
towards dietary guidelines. Food Rev 2000; 23: 2–14. 89 Comuzzie AG, Allison DB. The search for human obesity genes.
64 Hall KD, Sacks G, Chandramohan D, et al. Quantifying the effect of Science 1998; 280: 1374–77.
energy imbalance on bodyweight change. Lancet 2011; 378: 826–37. 90 Bogardus C. Missing heritability and GWAS utility.
65 Kumanyika S, Jeffery RW, Morabia A, Ritenbaugh C, Antipatis VJ, for Obesity (Silver Spring) 2009; 17: 209–10.
the Public Health Approaches to the Prevention of Obesity Working 91 Hebebrand J, Volckmar AL, Knoll N, Hinney A. Chipping away
Group of the International Obesity Taskforce. Obesity prevention: the the ‘missing heritability’: GIANT steps forward in the molecular
case for action. Int J Obes Relat Metab Disord 2002; 26: 425–36. elucidation of obesity—but still lots to go. Obes Facts 2010; 3: 294–303.
66 Sacks G, Swinburn B, Lawrence M. Obesity Policy Action 92 Jencks C. Heredity, environment, and public policy reconsidered.
framework and analysis grids for a comprehensive policy approach Am Sociol Rev 1980; 45: 723–36.
to reducing obesity. Obes Rev 2009; 10: 76–86. 93 Bouchard C. Gene-environment interactions in the etiology
67 Wilkinson RG, Pickett KE. Income inequality and population of obesity: defining the fundamentals. Obesity (Silver Spring) 2008;
health: a review and explanation of the evidence. Soc Sci Med 2006; 16 (suppl 3): S5–10.
62: 1768–84. 94 Yajnik CS. Fetal programming of diabetes: still so much to learn!
68 Mavoa HM, McCabe M. Sociocultural factors relating to Tongans’ Diabetes Care 2010; 33: 1146–48.
and Indigenous Fijians’ patterns of eating, physical activity and 95 Ahmed F. Epigenetics: Tales of adversity. Nature 2010; 468: S20.
body size. Asia Pac J Clin Nutr 2008; 17: 375–84. 96 Stein AD, Kahn HS, Rundle A, Zybert PA, van der Pal–de Bruin K,
69 Hayashi F, Takimoto H, Yoshita K, Yoshiike N. Perceived body Lumey L. Anthropometric measures in middle age after exposure to
size and desire for thinness of young Japanese women: a famine during gestation: evidence from the Dutch famine.
population-based survey. Br J Nutr 2006; 96: 1154–62. Am J Clin Nutr 2007; 85: 869–76.
70 Bell AC, Ge K, Popkin BM. The road to obesity or the path to 97 Lange UC, Schneider R. What an epigenome remembers. Bioessays
prevention: motorized transportation and obesity in China. Obes Res 2010; 32: 659–68.
2002; 10: 277–83. 98 Cohen DA. Neurophysiological pathways to obesity: below
71 Vandenbroeck IP, Goossens J, Clemens M. Foresight. Tackling awareness and beyond individual control. Diabetes 2008; 57: 1768–73.
obesities: future choices—building the Obesity System Map. 99 Blundell JE, King NA. Over-consumption as a cause of weight gain:
London: Government Office for Science, 2007. behavioural-physiological interactions in the control of food intake
72 Finegood DT. The complex systems science of obesity. In: (appetite). Ciba Found Symp 1996; 201: 138–54.
Cawley JH, ed. The Oxford handbook of the social science 100 Larsen TM, Dalskov SM, van Baak M, et al. Diets with high or low
of obesity. Oxford, UK: Oxford University Press, 2011: 208–36. protein content and glycemic index for weight-loss maintenance.
73 Rittel H, Webber M. Dilemmas in a general theory of planning. N Engl J Med 2010; 363: 2102–13.
Policy Sci 1973; 4: 155–69. 101 McMichael AJ, Powles JW, Butler CD, Uauy R. Food, livestock
74 Rodgers JL. The epistemology of mathematical and statistical production, energy, climate change, and health. Lancet 2007;
modeling: a quiet methodological revolution. Am Psychol 2010; 370: 1253–63.
65: 1–12. 102 Sanigorski AM, Bell AC, Kremer PJ, Cuttler R, Swinburn BA.
75 Gortmaker SL, Levy D, Carter R, et al. Changing the future Reducing unhealthy weight gain in children through community
of obesity: science, policy, and action. Lancet 2011; 378: 838–47. capacity-building: results of a quasi-experimental intervention
76 Levy DT, Mabry PL, Wang YC, et al. Simulation models of obesity: program, Be Active Eat Well. Int J Obes (Lond) 2008; 32: 1060–67.
a review of the literature and implications for research and policy. 103 de Silva-Sanigorski AM, Bell AC, Kremer P, et al. Reducing obesity
Obes Rev 2010; 12: 387–94. in early childhood: results from Romp & Chomp, an Australian
77 Jacobs P, Rapoport J. Economics of health and medical care. community-wide intervention program. Am J Clin Nutr 2010;
5th edn. Sudbury: Jones and Bartlett Publishers, 2004. 91: 831–40.
78 Pearl J. Causality: Models, reasoning, and inference. Cambridge: 104 Swinburn B, Egger G. Analyzing and influencing obesogenic
Cambridge University Press, 2000. environments. In: Bray G, Bouchard C, eds. Handbook of obesity:
79 McMichael A, Campbell-Lendrum D, Corvalan C, et al, eds. Climate clinical applications. 3rd edn. New York: Informa Health Care,
change and human health: risks and responses. Geneva: World 2008: 177–93.
Health Organization, 2003. 105 James P, Rigby N. Developing the political climate for action. In:
80 Weiss RA, McMichael AJ. Social and environmental risk factors Waters E, Swinburn B, Seidell J, Uauy R, eds. Preventing childhood
in the emergence of infectious diseases. Nat Med 2004; obesity: evidence, policy and practice. Oxford, UK: Blackwell,
10 (suppl 12): S70–76. 2010: 212–19.

www.thelancet.com Vol 378 August 27, 2011 813

 Series

106 Corporate Europe Observatory. A red light for consumer 115 National Health and Medical Research Council (NHMRC).
information: the food industry’s €1-billion campaign to block health Economic issues in the prevention and treatment of overweight
warnings on food. Brussels: Corporate Europe Observatory, 2010. and obesity. Acting on Australia’s weight: a strategic plan for the
107 Brownell KD, Warner KE. The perils of ignoring history: big tobacco prevention of overweight and obesity. Canberra: NHMRC, 1997.
played dirty and millions died. How similar is Big Food? Milbank Q 116 National Preventative Health Taskforce. Australia: the healthiest
2009; 87: 259–94. country by 2020—National Preventative Health Strategy—Overview.
108 Stender S, Dyerberg J, Astrup A. High levels of industrially Canberra: Commonwealth of Australia, 2009. http://www.health.
produced trans fat in popular fast foods. N Engl J Med 2006; gov.au/internet/preventativehealth/publishing.nsf/Content/
354: 1650–52. nphs-overview-toc (accessed June 23, 2011).
109 Food and Agriculture Organization of the United Nations. Global 117 National Obesity Taskforce. Healthy Weight 2008 Australia’s Future.
hunger declining, but still unacceptably high. Geneva, Switzerland: Canberra: Department of Health and Ageing, 2003. http://www.
Food and Agriculture Organization of the United Nations, health.gov.au/internet/healthyactive/publishing.nsf/content/
Economic and Social Development Department, 2010. http://www. healthy_weight08.pdf/$File/healthy_weight08.pdf (accessed
fao.org/docrep/012/al390e/al390e00.pdf (accessed June 23, 2011). June 23, 2011).
110 Geneau R, Stuckler D, Stachenko S, et al. Raising the priority of 118 Cross-Government Obesity Unit, Department of Health and
preventing chronic diseases: a political process. Lancet 2010; Department of Children, Schools and Families. Healthy weight,
376: 1689–98. healthy lives: a cross-government strategy for England. London:
111 Department of Economic and Social Affairs, Population Division. HM Government, 2008. http://webarchive.nationalarchives.gov.
World population prospects: the 2008 revision. Population uk/20100407220245/http://www.dh.gov.uk/prod_consum_dh/
newsletter. Geneva: United Nations Secretariat, 2009. http://www. groups/dh_digitalassets/documents/digitalasset/dh_084024.pdf
un.org/esa/population/publications/popnews/Newsltr_87.pdf (accessed June 23, 2011).
(accessed June 23, 2011). 119 United Kingdom Office of Communications. HFSS Advertising
112 Yach D, Khan M, Bradley D, Hargrove R, Kehoe S, Mensah G. Restrictions - final review. London: Ofcom, 2010. http://
The role and challenges of the food industry in addressing chronic stakeholders.ofcom.org.uk/binaries/research/tv-research/
disease. Global Health 2010; 6: 10. hfss-review-final.pdf (accessed June 23, 2011).
113 Ludwig DS, Nestle M. Can the food industry play a constructive role 120 White House Task Force on Childhood Obesity. Solving the
in the obesity epidemic? JAMA 2008; 300: 1808–11. problem of childhood obesity within a generation. White House
Task Force on Childhood Obesity, Report to the President.
114 Jackson T. Prosperity without growth? The transition to a
Washington DC, 2010. http://www.letsmove.gov/sites/letsmove.
sustainable economy. London: The Sustainable Development
gov/files/TaskForce_on_Childhood_Obesity_May2010_FullReport.
Commission, 2009. http://www.sd-commission.org.uk/data/files/
pdf (accessed June 23, 2011).
publications/prosperity_without_growth_report.pdf (accessed
June 23, 2011).

814 www.thelancet.com Vol 378 August 27, 2011