ArticlePDF Available

Policy options for responding to obesity: UK national report of the PorGrow project

Authors:
  • Science Policy Research Unit University of Sussex
  • UK Health Security Agency
Policy options for
responding to obesity:
UK national report of the
PorGrow project
Tim Lobstein, Erik Millstone, Miriam
Jacobs, Andy Stirling and Lisa Mohebati
SPRU – Science and Technology Policy Research
University of Sussex
August 2006
2
Table of Contents
PAGE
TABLES ..................................................................................................................................................4
FIGURES ................................................................................................................................................5
ACRONYMS ..........................................................................................................................................7
EXECUTIVE SUMMARY....................................................................................................................8
1 EPIDEMIC OF OBESITY.........................................................................................................13
1.1 OBESITY DEFINITIONS ..........................................................................................................13
1.2 SOURCES OF DATA IN THE UK .............................................................................................. 14
1.2.1 Prevalence data for the UK – adults...............................................................................15
1.2.2 Prevalence data for children in the UK – children.........................................................21
1.3 SUMMARY OF MAIN POINTS ..................................................................................................24
2 ESTIMATED COSTS OF OBESITY .......................................................................................25
2.1 HEALTH CONSEQUENCES OF OVERWEIGHT AND OBESITY .....................................................25
2.2 HEALTH CARE COSTS............................................................................................................ 28
2.3 OTHER ECONOMIC COSTS .....................................................................................................29
2.4 SUMMARY OF MAIN POINTS ..................................................................................................31
3 TRENDS IN FOOD CONSUMPTION AND PHYSICAL ACTIVITY ................................. 33
3.1 SOURCES OF DATA................................................................................................................ 33
3.2 FOOD CONSUMPTION ............................................................................................................34
3.3 PHYSICAL ACTIVITY .............................................................................................................40
3.4 SUMMARY OF MAIN POINTS ..................................................................................................44
4 POLICY-MAKING INSTITUTIONAL STRUCTURES........................................................46
4.1 PUBLIC SERVICE AGREEMENTS ............................................................................................48
4.2 LOCAL DELIVERY PLANS .....................................................................................................48
4.3 OVERVIEW ...........................................................................................................................51
4.4 SUMMARY OF MAIN POINTS ..................................................................................................53
5 POLICY DEBATES AND INITIATIVES IN THE UK .......................................................... 54
5.1 POLICY DEBATES.................................................................................................................. 54
5.2 INITIATIVES ..........................................................................................................................57
5.3 MARKETING CONTROLS ....................................................................................................... 58
5.4 SCHOOL FOOD STANDARDS .................................................................................................. 61
5.5 POLICY INITIATIVES BY NON-GOVERNMENTAL BODIES ........................................................62
5.6 NATIONAL NGOS WITH OBESITY CONCERNS........................................................................ 63
5.7 SUMMARY OF MAIN POINTS ..................................................................................................63
6 MULTI-CRITERIA MAPPING: A METHODOLOGY.........................................................65
6.1 INTRODUCTION TO MCM ..................................................................................................... 65
6.2 ELICITATION FRAMEWORK...................................................................................................67
6.2.1 Recruitment of Participants and Scoping.......................................................................67
6.2.2 The MCM Interview........................................................................................................68
6.3 METHODS OF ANALYSIS .......................................................................................................70
6.4 SUMMARY OF MAIN POINTS ..................................................................................................72
7 STAKEHOLDERS AND THEIR PERSPECTIVES...............................................................74
7.1 DECIDING WHICH STAKEHOLDERS OR PARTICIPANTS TO INCLUDE........................................74
7.2 PERSPECTIVES ......................................................................................................................76
7.3 UK PARTICIPANTS................................................................................................................ 77
7.4 SUMMARY OF MAIN POINTS ..................................................................................................78
8 OPTIONS FOR ADDRESSING OBESITY .............................................................................79
8.1 INTRODUCTION.....................................................................................................................79
3
8.2 SCOPE OF PROCESS AND DEFINITION OF OPTIONS ................................................................ 79
8.2.1 Core Options...................................................................................................................80
8.2.2 Discretionary options.....................................................................................................82
8.3 GROUPING THE OPTIONS INTO CLUSTERS.............................................................................. 85
8.4 UK RESULTS ........................................................................................................................86
8.4.1 Engagement with Predefined Options............................................................................ 86
8.4.2 Engagement with additional options ..............................................................................88
8.5 OVERVIEW ...........................................................................................................................88
8.6 SUMMARY OF MAIN POINTS ..................................................................................................89
9 DEVELOPING CRITERIA....................................................................................................... 91
9.1 INTRODUCTION.....................................................................................................................91
9.2 APPRAISAL CRITERIA............................................................................................................93
9.3 NUANCES IN THE USE OF CRITERIA .......................................................................................95
9.4 ISSUES ..................................................................................................................................97
9.5 WEIGHTINGS ........................................................................................................................99
9.6 OVERVIEW OF CRITERIA AND ISSUES ................................................................................. 105
9.7 SUMMARY OF MAIN POINTS ................................................................................................106
10 APPRAISING OPTION PERFORMANCE (SCORING)..................................................... 108
10.1 ELICITING SCORES FOR OPTIONS......................................................................................... 108
10.2 APPRAISAL OF OPTIONS ......................................................................................................108
10.2.1 Caveats concerning within- and across-option comparisons ..................................108
10.2.2 Core options.............................................................................................................122
10.2.3 Discretionary options...............................................................................................136
10.3 DIVERSITY IN THE APPRAISAL OF OPTIONS ......................................................................... 152
10.4 SUMMARY OF MAIN POINTS ................................................................................................170
11 MAPPING OPTION PERFORMANCE ................................................................................172
11.1 INTRODUCTION................................................................................................................... 172
11.2 THE OVERALL PICTURE....................................................................................................... 172
11.3 ANALYSIS OF OPTION RANKS WITHIN PERSPECTIVES .......................................................... 183
11.3.1 Perspective A. Public interest, non-governmental organisations............................184
11.3.2 Perspective B: Food chain, large industrial and commercial organisations...........186
11.3.3 Perspective C: Small food and fitness commercial organisations...........................187
11.3.4 Perspective D: Large non-food industrial and commercial organisations.............. 187
11.3.5 Perspective E: Policy-makers.................................................................................188
11.3.6 Perspective F: Public providers .............................................................................191
11.3.7 Perspective G: Public health specialists.................................................................192
11.3.8 Additional options scored.......................................................................................194
11.4 PATTERNS OF CONSENSUS AND DIVERSITY .........................................................................200
11.4.1 Optimistic and pessimistic conditions for appraisal scores.....................................201
11.4.2 Conditionalities........................................................................................................202
11.4.3 Divergence within clusters of options......................................................................203
11.5 OVERVIEW OF OPTION RANKS ............................................................................................ 203
11.5.1 Most popular options...............................................................................................207
11.5.2 Least popular options ..............................................................................................207
11.6 MAIN SUMMARY POINTS.................................................................................................... 208
12 EVALUATION .........................................................................................................................211
12.1 EVALUATION OF PROCESS AND RESULTS ........................................................................... 211
12.2 CRITICAL REFLECTIONS ..................................................................................................... 212
12.3 IMPLICATIONS FOR POLICY ................................................................................................212
12.4 MAIN SUMMARY POINTS..................................................................................................... 216
APPENDIX 1: INSTITUTIONAL LOCATIONS OF INTERVIEWEES.....................................218
REFERENCES ................................................................................................................................... 219
4
TABLES
Table 1-1. Categories of adiposity according to BMI (adults) ....................................14
Table 1-2. Scotland 1995 and 1998: adults aged 16-64..............................................17
Table 2-1. Estimated increased risk for obese people developing obesity-related
diseases (non-obese = 1.0) ........................................................................................... 26
Table 2-2. Estimates of the direct costs of obesity to national health services........... 28
Table 2-3. The estimated health service costs for treating obesity and consequential
diseases, England 1998 and 2002 ................................................................................29
Table 2-4. Estimated days of certified sickness attributable to obesity, England 1998
......................................................................................................................................30
Table 2-5. The estimated work-related costs of obesity and consequential diseases,
England 1998 and 2002 ...............................................................................................30
Table 3-1. Household expenditure on food and drink, indicating the proportion spent
outside the home, 1980-2004.......................................................................................38
Table 3-2. Household expenditure on food, showing a decline 1984-2004 especially
among lower income households................................................................................. 38
Table 3-3. Typical price (in pence) per 100 kcal (2000 prices), changes in price
relative to general food price index 1992-2000, and consumption levels by high and
low income households, 2000...................................................................................... 39
Table 4-1. Department of Health guidance for Local Delivery Plans .........................49
Table 5-1. Stated government action to meet the child obesity targets .......................57
Table 7-1. Participants grouped into Perspectives for analytical purposes .................76
Table 7-2. Participants interviewed in the UK.............................................................77
Table 8-1. Grouping of options into clusters. ..............................................................85
Table 8-2. Options selected and rejected for scoring................................................... 86
Table 8-3. Additional options scored by participants. ................................................. 88
Table 9-1. Number of criteria selected by participants interviewed in the UK ...........92
Table 9-2. Criteria chosen by participants for policy appraisal................................... 93
Table 9-3. Grouping of criteria into issues ..................................................................97
Table 9-4. Issues according to the various perspectives .............................................. 98
Table 11-1. Options grouped into clusters.................................................................172
Table 11-2. UK participants grouped into Perspectives. ...........................................183
Table 11-3. Additional options proposed by UK participants ...................................194
Table 11-4. Upper 'optimistic' rank scores overall and after adjustment for excluding
specified participants .................................................................................................205
FIGURES
Figure 1-1. Levels of overweight and obesity according to age group (men) in 2003)
......................................................................................................................................15
Figure 1-2. Levels of overweight and obesity according to age group (women in 2003)
......................................................................................................................................16
Figure 1-3. Trends in the levels of obesity and overweight 1980-2003 (men)............16
Figure 1-4. Trends in the levels of obesity and overweight 1980-2003 (women).......17
Figure 1-5. Obesity (BMI 30) among adults in different black and ethnic minority
groups, England 1999 ..................................................................................................18
Figure 1-6. Obesity (BMI 30) among adults according to the employment status of
the head-of-household, England 2003 .........................................................................19
Figure 1-7. Obesity (BMI 30) among adults according to annual household income,
adjusted for family size, England 2003........................................................................20
Figure 1-8. Obesity (BMI >= 30) among adults according to the employment status of
the head-of-household, England 1996 .........................................................................20
Figure 1-9. Overweight and obesity by age, 2001-2002 (boys) ..................................21
Figure 1-10 Overweight and obesity by age, 2001-2002 (girls)..................................22
Figure 1-11. Overweight and obesity trends in boys and girls aged 5-10 years,
England, 1974-2003.....................................................................................................23
Figure 1-12. Obesity trends across household income levels (quintiles), England
2001-2002 ....................................................................................................................24
Figure 2-1. The relationship between increasing body mass (BMI) and relative risk of
early death (Women only, USA database)................................................................... 27
Figure 3-1. Increasing food energy supplied to consumers .........................................34
Figure 3-2. Changes in per capita daily supplies of fats, sugars and protein, 1971-2001
......................................................................................................................................36
Figure 3-3. Decreasing food energy in household purchases ......................................37
Figure 3-4. Proportion of children in Wales exercising for at least 4 hours per week,
1986-2000 ....................................................................................................................42
Figure 3-5. Trends in transport to school, 1985/6-2002 ..............................................43
Figure 3-6. Leisure-time broadcast television watching, 1983-1987 and 1997-2001 .44
Figure 4-1. The delivery chain for preventing child obesity .......................................52
Figure 9-1. Weight extrema for Public Interest NGOs -- Perspective A...................100
Figure 9-2. Weight extrema for Food Chain Large Industrial and Commercial
Organisations -- Perspective B...................................................................................101
Figure 9-3. Weight extrema for Small Food and Fitness Commercial Organisations --
Perspective C .............................................................................................................102
Figure 9-4. Weight extrema for Large Non-Food Industrial and Commercial
Organisations -- Perspective D ..................................................................................102
Figure 9-5. Weight extrema for Government Policy Makers -- Perspective E..........103
Figure 9-6. Weight extrema for Public Providers -- Perspective F............................104
Figure 9-7. Weight extrema for Public Health Specialists -- Perspective G .............104
Figure 9-8. Weight extrema for all participants......................................................... 105
Figure 10-1. Perspective A: Public interest NGOs ....................................................110
Figure 10-2. Perspective B: Food chain commercial organisations ..........................112
Figure 10-3. Perspective C: Small food and fitness commercial organisations ........114
6
Figure 10-4. Perspective D: Non-food commercial organisations.............................115
Figure 10-5. Perspective E: Policy-makers................................................................ 117
Figure 10-6. Perspective F: Public service providers ................................................118
Figure 10-7. Perspective G: Public health professionals ...........................................120
Figure 10-8. Lower and upper ranks for option 1......................................................124
Figure 10-9. Lower and upper ranks for option 3......................................................128
Figure 10-10. Lower and upper ranks for option 5....................................................132
Figure 10-11. Lower and upper ranks for option 6....................................................134
Figure 10-12. Lower and upper ranks for option 9....................................................138
Figure 10-13. Lower and upper ranks for option 15..................................................144
Figure 10-14. Lower and upper ranks for option 16..................................................146
Figure 10-15. Lower and upper ranks for option 17..................................................148
Figure 10-16. Lower and upper ranks for option 19..................................................150
Figure 10-17. Lower and upper ranks for option 20..................................................151
Figure 10-18. Issue I: Positive societal benefits ........................................................ 154
Figure 10-19. Issue II: Additional health benefits ..................................................... 155
Figure 10-20. Issue III: Efficacy in addressing obesity ............................................. 156
Figure 10-21. Scores for 'efficacy' by perspectives ...................................................157
Figure 10-22. Issue IV: Economic costs to the public sector ....................................161
Figure 10-23. Issue V: Economic costs to individuals ..............................................162
Figure 10-24. Issue VI: Economic costs to the commercial sector............................163
Figure 10-25. Issue VIII: Practical feasibility............................................................164
Figure 10-26. Scores for 'practical feasibility' by Perspectives .................................165
Figure 10-27. Issue IX: Social acceptability.............................................................. 166
Figure 10-28. Scores for 'social acceptability' by perspective...................................167
Figure 11-1. Average ranks for all participants combined ........................................173
Figure 11-2. Average ranks for all participants in Perspective A: Public Interest NGOs
....................................................................................................................................184
Figure 11-3. Individual participants in Category 19: Public interest NGOs..............185
Figure 11-4. Average ranks for all participants in Perspective B: Food Chain
Commercial Organisations.........................................................................................186
Figure 11-5. Ranks for the participant in Perspective C: Small Food and Fitness
Companies..................................................................................................................187
Figure 11-6. Average ranks for all participants in Perspective D: Large Non-Food
Commercial Organisations.........................................................................................188
Figure 11-7. Average ranks for all participants in Perspective E: Policy-Makers.....189
Figure 11-8. Individual participants within Perspective E: Policy-makers ...............190
Figure 11-9. Average ranks for all participants in Perspective F: Public Service
Providers ....................................................................................................................191
Figure 11-10. Average ranks for all participants in Perspective G: Public Health
Specialists ..................................................................................................................192
Figure 11-11. Extreme ranks and conditionality for Perspective G: Public health
specialists ................................................................................................................... 193
Figure 11-12. Rank means for participants with additional options..........................196
Figure 11-13. Extremes of ranks for each option (all participants) ........................... 201
Acronyms
ASA Advertising Standards Authority
ASO Association for the Study of Obesity
BMI Body Mass Index (kg/m2)
BOGOF buy-one-get-one-free
CAP Common Agricultural Policy
CHD Coronary Heart Disease
DCMS Department for Culture, Media and Sport
DEFRA Department for Environment, Food and Rural Affairs
DfES Department for Education and Skills
DALYs Disability-Adjusted Life Years
DoH Department of Health
EU European Union
FAO Food and Agriculture Organization
FAU Food Advertising Unit
FDF Food and Drink Federation
FSA Food Standards Agency
GDAs Guideline Daily Amounts
GP General Practitioner
HFSS (food and drink) high in fat, salt or sugar
HM Treasury Her Majesty’s Treasury
HNR Human Nutrition Research
HSE Health Survey for England
LAAs Local Area Agreements
LDPs Local Delivery Plans
LEA Local Education Authority
LSHTM London School of Hygiene and Tropical Medicine
MCM Multi-Criteria Mapping
MRC Medical Research Council
NAO National Audit Office
NGOs Non-governmental organisations
NHS National Health Service
NIHCE National Institute for Health and Clinical Excellence
OECD Organisation for Economic Co-operation and Development
Ofcom Office of Communications
OFSTED Office for Standards in Education (England)
PCTs Primary Care Trusts
PE Physical Education
PorGrow Policy Options for Responding to the Growing Challenge of Obesity
Research Project
PSA Public Service Agreement
QUIDs Quantitative Ingredient Declarations
SHAs Strategic Health Authorities
UK United Kingdom
UN United Nations
UNICEF United Nations Children’s Fund
US or USA United States or United States of America
VAT value added tax
WHO World Health Organization
Executive Summary
Obesity is an acute and rapidly growing threat to public health, reaching epidemic
proportions not only in industrialised but also in developing countries. Excess weight
is estimated to be responsible for nearly 300 000 deaths annually in the European
Union, that is, nearly 1 in 12 of all deaths recorded, by contributing to cardiovascular
disease and cancer. Obesity and related diseases are among the most unevenly
distributed health conditions, and there is a trend towards an increase in differences
between social classes. The epidemic now consumes at least 8% of overall health care
budgets and creates a major economic burden through loss of productivity and
income, as well as, substantial indirect financial costs and intangible social costs, such
as underachievement in school and discrimination at work.
Unhealthy diets and physical inactivity are the main contributors to the rapidly
growing epidemic of obesity. The reasons for the adoption of unhealthy diets and
sedentary lifestyles are societal and, therefore, substantial changes of strategy are
required. Traditional ways of preventing and treating obesity have almost invariably
focused on changing the behaviour of individuals, an approach that has proven
inadequate, bearing in mind the escalating trend in obesity worldwide.
There is now a broader consensus that reversing current obesity trends will require a
multifaceted public health policy approach. There will need to be a better balance,
than has so far been the case, between individual and population-wide approaches and
between education-based initiatives and multi-sectoral and cross-sectoral
environmental interventions. National and international actions and collaborations
will be required to counteract this epidemic, to which national and local governments
and international organisations need to respond. Addressing obesity is a specific
priority of the EU’s Public Health Action Programme for 2003–2008. Identifying,
developing, supporting and implementing effective strategies to slow and then reverse
current trends in the incidence of obesity are community responsibilities. A wide
range of different kinds of interventions could be introduced to influence different
aspects of the production and consumption of foods and levels of physical activity. A
concerted effort will be necessary to ensure that the current obesity epidemic does not
exacerbate the growing epidemic of cardiovascular disease and type 2 diabetes in the
developed and developing world nor reverse the favourable trends in cardiovascular
risk being seen in many affluent countries.
It would be unrealistic to expect one single uniform set of policies to be appropriate
for both genders, for all age and social groups, and for all European countries. The
aims of the PorGrow project were, therefore, the exploration and analysis of the
perspectives of key stakeholders towards a range of different options that might be
introduced to respond to the challenge of obesity, and a cross-national comparison of
those perspectives between nine participating member states.
A novel and powerful technique called Multi Criteria Mapping (MCM) has been used
to provide an integrative and comparative data collection and analysis methodology
with which the different perspectives of key stakeholders in nine EU Member States
on a broad range of possible types of interventions can be analysed.
This report deals with findings at the national level in the UK. A report on cross-
national findings between nine participating member states will follow.
9
The PorGrow project conducted a systematic process to identify key public policy
options that might have a bearing on how to respond to the rising trend in the
incidence of obesity in the UK. Using the MCM method, quantitative and qualitative
data were gathered from representatives of 21 types of organisations representing
relevant stakeholder interest groups.
During structured interviews, participants were asked to compare the performance of
seven ‘core’ policy options, and up to 14 ‘discretionary’ options, and invited to
introduce their own ‘additional’ options. To appraise those options, participants
defined criteria, that is, the factors that they will take into account when evaluating the
options. Participants judged the performance of each chosen option against each of
their criteria; they assigned a score for every option under each criterion using a linear
ordinal scale. The only constraint of the scale was that the higher the score the more
optimistic was the appraisal of the performance of that option. Participants were
invited to score each option using each criterion by reference to both optimistic and
pessimistic assumptions, and to make those assumptions explicit. As a final step,
participants weighted the criteria in order to reflect their relative importance. Using a
simple formula, the scores under each criterion are multiplied by the criteria
weightings to produce overall relative pessimistic and optimistic rankings for all the
options.
MCM is not a procedure that, in this context, could generate a recipe of proven
efficacy to solve the obesity problem, either in general or over a particular time scale,
but it does nonetheless provide a tool with which the challenge of obesity can sensibly
and effectively be approached. The data gathered for this study, when analysed in
context of debates about public health policy in the UK, collectively indicate that
there is a broad consensus that in order to reverse the trend in the incidence of obesity
it will be necessary to implement a portfolio of several different kinds of measures,
and that the design and implementation of those individual measures will need to be
co-ordinated and integrated.
The data indicate that there is a broad consensus that it will be necessary to
implement an interconnected portfolio of measures relating to improved health
education, improved opportunities and incentives for physical activity, more and
better information and changes to the supply and demand of food and beverages.
On the other hand, there is also a broad consensus that technical ‘fixes’ such as the
increased use in processed food products of synthetic sweeteners or fat
substitutes, or physical activity monitoring devices (such as pedometers), have no
significant part to play in addressing the problem.
There is a clear perception amongst a majority of UK interviewees that the Common
Agricultural Policy (CAP) may be contributing to elevated intakes of some unhealthy
foods, but there was little confidence that the CAP will be reformed in ways directly
beneficial to public health in the short- to medium-terms. There was, moreover, a
recognition that allowing the price of sugar in the EU to converge with world market
prices might well result in higher levels of sugar in processed foods and beverages,
not lower ones.
There was a very broad consensus that there will be a need for public policy-makers
to coordinate different elements of the overall portfolio of policies on obesity, but
there was little support for the creation of a new governmental institution with
10
responsibility for obesity policy. That proposal, often disparaged in terms of a ‘fat
controller’, received only modest levels of support and high levels of scepticism.
Rather interviewees expect the Department of Health to take the lead.
In respect of the group of options intended to improve levels of knowledge and
understanding about food, diet, health and fitness, there was broad support for
improved education of both school-aged children and the general adult population, but
with repeated emphasis that education on its own will not be sufficient, and that it will
need to be accompanied by improvements in the quantity and quality of information
available to citizens. Better food and drink labelling is very widely seen as essential,
even by stakeholders in the food industry. Improved labelling information will need to
indicate both clearly and directly relevant qualities of food and drink products. While
there is evidently some resistance to a traffic light labelling system in some parts of
the food industry, a system of that sort is widely and consistently supported by other
stakeholders and perspectives. Several influential industrial groups have proposed
that instead of a traffic light labelling system, it would be preferable to adopt one that
provided indications of the percentages of ‘guideline daily amounts’; indeed some
labels of those types are already in use.
Curiously, some of labels providing percentage Guideline Daily Amounts (GDAs),
are also colour-coded, but not using the Red, Amber and Green of the classic traffic
light, as proposed by the Food Standards Agency (FSA). (For example, a Tesco
Chicken salad sandwich, purchased in May 2006, provided quantitative estimates of
calories, sugar, fat, saturates and salt in the product, along side indications of those
figures of percentages of GDAs. The information on calories was in black against a
background coloured light blue, sugar was black against a pink background, fats and
saturates were against a light green background, while salt was against an orange
background.) Under those circumstances there may be a risk that differing colour
schemes may contribute to confusing consumers.
None of our interviewees, other than those in the food industry, saw traffic lights as
inconsistent with, or a competitor to, labelling with percentage GDAs. There is
nothing in our evidence to suggest that a combination of the two would not be at least
as helpful as, and perhaps more helpful than, either on its own; the Tesco label
referred to above indicates that colour-coding and numerical estimates can
comfortably co-exist, but also that colour-coding needs to be standardised and
understood.
There is broad support for the argument that changes will be needed in both patterns
of food consumption and levels of physical activity. A few representatives of the food
industry have argued that obesity is entirely a consequence of declining levels of
physical activity, and unrelated to food or drink consumption, but that group
repeatedly declined to participate in this project. There are high levels of support for
providing improved facilities for physical activity and recreation and for consequent
changes to transport and planning policies. The under-use of school and college
gymnasia, swimming pools and sports facilities, after the end of the school day and
during school holidays, was widely seen as wasteful.
There are high levels of support for restrictions on the marketing and advertising of
certain categories of food and drink especially to children and young people; and
while advertising industry representatives were amongst those least enthusiastic for
11
that option, a large majority of other stakeholders, in all the other perspectives
including companies in the food chain, did not share their antipathy to such controls.
The interviews were conducted before the publication in April 2006 of proposals from
the Office of Communications (Ofcom) that suggested some limited restrictions on
food and drink advertising on television to children, but nothing that was said during
the interviews on the issue of controls on advertising suggested that anyone
anticipated the kinds of proposals that Ofcom issued, and so our data provide no direct
comments on those proposals. On the other hand, the interviewees who argued in
favour of restricting advertising never implied that they favoured limiting the ban to
programmes aimed solely at children, with the continuance of the transmission of
such adverts during general family viewing programmed during evenings and up to
the 9 pm ‘watershed’. A large majority indicated that they favoured a more
comprehensive ban on the advertising and marketing of unhealthy foods and drinks,
especially to children and young people.
Economic instruments, such as taxes on obesogenic foods and/or subsidies on
relatively healthy foods were not widely supported. Those from the industrial and
commercial perspectives criticised those measures for distorting markets, while those
interviewees representing public health non-governmental organisations (NGOs),
public policy-makers, public service providers and public health professionals were
concerned about the disproportionate adverse impact of such taxes on low income
groups. Those who argued that the relative prices of healthy foods had a significant
impact upon patterns of purchasing and consumption proposed rather that minimum
wage rates and benefit levels should be adjusted to take account of the costs of buying
healthy diets rather than by providing subsidies to all shoppers. Several stakeholders
also argued in favour of providing schools, hospitals and prisons with increased
resources to enable them to provide healthier meals.
Overall, the general consensus among the wide range of stakeholders interviewed, in
the context of changing conditions in the UK, is that to address the challenge of
obesity, policy-makers should consider implementing policies in the following areas,
with special focus on:
1. Improving levels of knowledge and understanding about food, diet, health and
fitness:
a. Addressing food and health in the school curriculum;
b. Improving health education offered to the general public.
c. Further training for health professionals in obesity prevention and
diagnosing and counselling those at risk of obesity.
d. More research into the effective prevention and treatment of obesity.
2. Increasing opportunities and incentives for physical activity.
a. Improving planning and transport policies so that more physical
activity can be incorporated into daily life.
b. Enabling schools and communities to have access to adequate sports
facilities.
12
3. Improving the quantity and quality of information available to citizens
concerning food, diet, physical activity, fitness and health:
a. Implementing controls on the advertising and promotion of food and
drink products, especially those targeted at children.
b. Requiring mandatory nutritional information labelling for all processed
food, for example using energy density traffic light system.
4. Modifying patterns of demand and supply for food stuffs and beverages:
a. Establishing controls on the provision and sale of fatty snacks,
confectionery and sweet drinks in public institutions such as schools
and hospitals.
There will be a need, moreover, to maintain and improve monitoring of Body Mass
Index (BMI, kg/m2) levels in the population, and especially amongst particularly
vulnerable groups. There also needs to be improved monitoring of changing patterns
of food consumption and levels of physical activity. If the trends change in the
desired direction, and at an acceptable rate, then it may be appropriate to maintain
those policies in place. To the extent, however, that the trends either fail to move in
the desired direction or shift only very slightly and slowly, then policy-makers will
need to introduce further measures in one, several or all of those four categories.
Comparative, cross-national studies of the relative effectiveness of different national
policy approaches will also be useful, as long as they take into account the relevant
cultural similarities and differences between countries.
There is nothing in all the documents studied or the data gathered for this study to
suggest that the problem of obesity cannot be overcome; even though almost no-one
thinks that the problem can be easily or simply resolved. Resolve to solve it will
however be indispensable.
13
1 Epidemic of obesity
1.1 Obesity definitions
Obesity is normally defined as an abnormal or excessive fat accumulated in adipose
tissue to the extent that health may be impaired, although the distribution of fat affects
the risks of disease and the kinds of disease that result from obesity.1 The distribution
of the fat affects the risk associated with obesity, with excess fat in the abdominal area
leading to the highest risk of subsequent degenerative disease. Although obesity was
included in the 6th International Classification of Diseases in 1948 it has only recently
begun to be more widely regarded as a disease in its own right, and not simply a risk
factor for other disorders and diseases. (WHO op cit) This may be because of the
multiple problems that arise as a consequence of obesity, including direct effects on
bones and joints and direct psychological consequences in terms of reduced self-
esteem.
Various methods for measuring adiposity can be used in order to classify or diagnose
obesity: direct measures include magnetic resonance imaging, X-ray absorptiometry,
hydro-densitometry and bio-electrical impedance. These require sophisticated
equipment, complex calibration and the use of fully-trained staff and are impractical
for population survey purposes. Indirect measures of adiposity include body weight
adjusted for height, waist circumference, the ratio of waist-to-hip circumference and
skin-fold thickness measures. These are relatively easy to obtain using simple
equipment and less qualified staff.
The most common method used to classify obesity is the Body Mass Index, a measure
of weight (in kilograms) divided by height (in metres) squared. This is only an
approximation of adiposity, as persons with high levels of lean (muscle) mass will
also have relatively high BMI scores. It does not indicate the location of the adiposity,
whereas waist circumference, or the ratio of waist circumference to hip circumference
(the waist-hip ratio) and skin fold measures do provide some indication of the location
of the adiposity. BMI, however, is by far the most widely used measure in
anthropometric surveys of obesity and will generally be the method for defining
obesity in the data given in this report.
The World Health Organization (WHO) adopted the definition that a BMI of 30 in
adults should be taken as the threshold for obesity. Because questions were raised
about the universal applicability of this criterion, the WHO convened an expert group
that recommended differential BMI cut off points for Asian populations. Though the
main definition has been widely accepted and has been used in official UK statistics
since 1980, the differential BMI cut off points for Asian populations have yet to be
adopted in the assessment of relevant minority populations in the UK.
In addition, more severe levels of obesity have been defined, as has a less severe
definition, described as ‘overweight’. These are shown in Table 1-1.
14
Table 1-1. Categories of adiposity according to BMI (adults)
DESCRIPTION BMI (kg/m2)
Underweight under 18.5
Normal range 18.5-24.99
Overweight 25 or more
Obese 30 or more
sub-classifications:
overweight pre-obese
moderately obese
severely obese
very severely obese
25-29.99
30-34.99
35-39.99
40 or more
Source: adapted from WHO 20001
Note that there can be confusion about the use of the word ‘overweight’. It may refer
to all persons with a BMI of 25 or more, or it may mean only those persons with a
BMI between 25 and 29.99 (sometimes this is referred to as ‘overweight non-obese’
or ‘pre-obese’).
The BMI classifications given in Table 1-1 apply to adults. They cannot be applied to
children because the relationship between weight and height is changing throughout
childhood as a result of normal growth patterns. Several alternative approaches have
been defined for measuring children: one uses age-based charts for weight, height and
BMI based on a reference population, and defines excessive adiposity as a BMI more
than two standard deviations above the reference population mean. A second
approach also uses a reference population, but defines obesity as a BMI above the 95th
centile on a chart showing the BMI distribution of the population. The reference
population used in either of these two definitions may be a local population (e.g. data
from English children in 1990 used by the UK Department of Health) or a population
recommended for international comparison such as that used by the WHO, consisting
of large sample of children in the USA from birth to 20 years of age. A new WHO
international child growth standard was released in April of 2006, based on the
growth of breast-fed children from a number of countries around the world, but this is
currently available only from birth to six years of age.
A third approach is to take populations of children where data are available from
infancy to adulthood and plot the BMI centile curves back from adulthood through
childhood, equivalent to adult BMIs of 25 and 30. This provides a series of cut-offs
linked to the adult definition, adjusted for age and gender. Cut-off values on the basis
of six combined surveys in Asia, Europe, and North and South America have been
published by Cole et al2 and are recommended by the International Obesity Task
Force for comparison of child overweight and obesity prevalence statistics across
different populations.
1.2 Sources of data in the UK
According to the Department of Health there are two principle sources of official data
on obesity in England: (a) the Health Survey for England, from which it is possible to
estimate the prevalence of obesity (ie the percentage of the population who are obese)
15
as defined by BMI, and (b) Hospital Episode Statistics, which give the number of
finished consultant episodes for all adults and children admitted to hospital with a
diagnosis of obesity.3 The Health Survey for England monitored obesity levels in
adults in 1980, 1987 and annually since 1993. Data for children are available from
surveys in 1974, 1984 and annually since 1994.
In Scotland, three recent surveys included anthropometric data, undertaken in 1995,
1998 and 2004. The results of the 2004 survey have not yet been published. In
Northern Ireland three Health and Social Wellbeing surveys have been conducted, in
1997, 2001 and 2005. Results from the 2005 survey have not yet been published, and
the 2001 survey did not include anthropometric data. In Wales, provisional results are
available from the Welsh Health Survey 2003-2004.
Additional information is available from non-official sources, including further
analysis of official data and smaller-scale surveys reported in the scientific journals.
1.2.1 Prevalence data for the UK – adults
Recent surveys in England suggest that obesity affects around 24% of adult men and
26% of women, and overweight (pre-obese) affects a further 44% of men and 34% of
women.4 However, these figures for all adults belie the much higher obesity
prevalence that occurs in older age groups: for those aged between 45 and 74 years
old, obesity affects nearly 30% of both men and women (see Figures 1-1 & 1-2). In
terms of the life course, over 75% of men and 57% of women are already overweight
or obese by the age of 35, and more than 70% of adults can expect to become
overweight or obese by the time they reach 55, on present figures. The prevalence
levels are slightly lower among adults over 75 years old, possibly because obesity-
related chronic diseases reduce life expectancy.
Figure 1-1. Levels of overweight and obesity according to age group (men) in
2003)
Overweight and obesity, men, 2003
0
10
20
30
40
50
60
70
80
90
16-
24 25-
34 35-
44 45-
54 55-
64 65-
74 75+ all
age group
prevalence (%)
overweight
obese
Source: Health Survey for England 20034
16
Figure 1-2. Levels of overweight and obesity according to age group (women in
2003)
Overweight and obesity, women, 2003
0
10
20
30
40
50
60
70
80
90
16-
24 25-
34 35-
44 45-
54 55-
64 65-
74 75+ all
age group
prevalence (%)
overweight
obese
Source: Health Survey for England 20034
Trends in prevalence over the last two decades show a gradual increase in the
proportion of all adults who are either overweight or obese: in particular the
prevalence of obesity has risen in the period 1980-2003 from 8% to 23% in women,
and 6% to 23% in men (see figures).
Figure 1-3. Trends in the levels of obesity and overweight 1980-2003 (men)
Trends, men 1980-2003
0
10
20
30
40
50
60
70
80
1980
1987
1993
1995
1997
1999
2001
2003
prevalence (%)
overweight
obese
Source: Health Survey for England 20034 and earlier surveys
17
Figure 1-4. Trends in the levels of obesity and overweight 1980-2003 (women)
Trends, women 1980-2003
0
10
20
30
40
50
60
70
80
1980
1987
1993
1995
1997
1999
2001
2003
prevalence (%)
overweight
obese
Source: Health Survey for England 20034 and earlier surveys
Data for Scotland indicate slightly lower levels of adult obesity compared to England
at the same period. Scottish data also shows the highest levels of obesity among older
people, with obesity levels in men aged 45-54 of 29%, and women aged 55-64 of
32%.
Table 1-2. Scotland 1995 and 1998: adults aged 16-64.
Men Women
1995 1998 1995 1998
Overweight and obese 56 61 47 52
of which obese 16 19 17 21
Source: Scottish Executive, 20005
Official reports from Northern Ireland indicate that in 1997 some 17% of adult men
and 20% of adult women were obese.6 It is likely that these data were based on self-
reported statements of height and weight, which are notorious for underestimating the
true prevalence of overweight and obesity.7
A Welsh Assembly report in 2002 noted: “More than 50% of adults in Wales are
overweight and 17% are obese. The prevalence of obesity has doubled in
approximately ten years.”8 A summary of the results of the Welsh Health Survey
2003-2004 indicated that 54% of Welsh adults were overweight or obese, of whom
17% were obese, but these data were based on self-reported measures.9
Socio-economic status and ethnicity data are also available for some population
groups surveyed in the 1999 Health Survey for England. Odds ratios for obesity
indicated that men from the South Asian and Chinese communities were less likely to
be obese (0.6 Indian, 0.8 Pakistani, 0.4 Bangladeshi, and 0.4 Chinese), while Irish
18
men were more likely (1.3) to be obese than men in general. Among women Black
Caribbean and Pakistani women were more likely to be obese (1.8 and 1.5), while
Chinese and Bangladeshi women were much less likely to be obese (0.2 and 0.4).10
It should be noted that central adiposity as measured by waist-hip ratio showed
different results: women from all the minority ethnic groups were more likely than the
general population to have a raised waist-hip ratio (odds ratios ranging from 1.3 in
Irish women to 3.2 in Bangladeshi women); men showed more diversity, with Black
Caribbean men the least likely to have a raised waist-hip ratio (0.6), and Indian and
Pakistani men the most likely to (1.5 and 1.4).
Figure 1-5. Obesity (BMI 30) among adults in different black and ethnic
minority groups, England 1999
Obesity levels in population groups
0 5 10 15 20 25 30 35
Black Caribbean
Indian
Pakistani
Bangladeshi
Chinese
Irish
General population
Prevalence (%)
Women
Men
Source: Health Survey for England 199911
Additional information is available on socio-economic status and obesity prevalence.
In the 2003 Health Survey for England a detailed analysis of socio-economic status
was undertaken. The first analysed obesity levels in adults according to the
employment status of the head of the household or main income earner (see Figure 1-
6). This status is not based on skill levels, but on different forms of employment
relations. Women’s obesity prevalence was lower in managerial and professional
households (i.e. those that supervised or managed others and those who had a
significant level of autonomy in their occupation, at 18.7%) and in intermediate
households (i.e. those who had a moderate degree of autonomy in their occupation,
also at 18.7%) than in lower supervisory and technical households (i.e. those who had
a limited degree of autonomy in supervision or in the performance of technical tasks,
at 28.8%) and households whose reference person was in a semi-routine or routine
occupation (i.e. those in a direct exchange of money for effort with the least amount
of employee discretion, at 29.1%). The prevalence of very severe obesity (BMI > 40
kg/m2) was 1.6% among women in managerial and professional households, but 4.1%
among women in semi-routine or routine households. The prevalence of overweight
19
including obesity was also lower among women in managerial and professional
households (around 50%) than among those in lower supervisory and technical and in
semi-routine and routine households (around 60%). These results were not
appreciably changed by age standardisation.
For men the difference between employment categories was less marked. Obesity
prevalence was 20.9% in men in managerial and professional households and 19.7%
in intermediate households, but in the other three groups it was higher: 26.7% (small
employers and own account workers), 24.9% (lower supervisory and technical) and
21.6% (routine and semi-routine). Age standardisation reduced the magnitude of the
difference between the first two groups and the last three, but it remained significant.
Using an alternative measure of socio-economic status – annual household income,
adjusted for the number of person in the household – showed a very similar pattern:
women showed a sharp increase in prevalence of obesity in the lower income groups
compared with the higher ones while men showed highest obesity levels among the
middle of the income range (Figure 1-7).
Figure 1-6. Obesity (BMI 30) among adults according to the employment status
of the head-of-household, England 2003
Obesity and employment status
0 5 10 15 20 25 30 35
Semi-routine &
routine
Low er supervisory
& technical
Small employers &
self-employed
Intermediate
Managerial &
professional
Prevalence (%)
Women
Men
Source: Health Survey for England 20034
20
Figure 1-7. Obesity (BMI 30) among adults according to annual household
income, adjusted for family size, England 2003
Obesity and income level
0 5 10 15 20 25 30 35
Lowest
4th
3rd
2nd
Highest
Prevalence (%)
Women
Men
Source: Health Survey for England 20034
Trends have changed little over a decade. A previous survey of socio-economic status
(the 1996 Health Survey for England12) showed very similar results to the 2003
survey – women showed a strong social class gradient while men did not – except that
the earlier survey found lower obesity levels overall (see Figure 1-7).
Figure 1-8. Obesity (BMI >= 30) among adults according to the employment
status of the head-of-household, England 1996
Source: Health Survey for England 199612
Obesity and employment status 1996
0 5 10 15 20 25 30 35
V (low SES)
IV
III (manual)
III (non-manual)
II
I (high SES)
Prevalence (%)
Women
Men
21
Lastly, the same pattern was found in the Scottish Health Survey 1998: Of four
categories of employment status, from highest to lowest, men showed obesity
prevalence rates of 18%, 24%, 21% and 20% respectively, while women showed
obesity prevalence rates of 18%, 20%, 26% and 26% respectively.
1.2.2 Prevalence data for children in the UK – children
As noted earlier, data for childhood overweight and obesity can be reported according
to different classification criteria: in this section the international classification system
(IOTF) is used unless otherwise indicated.
The most extensive data comes from the series of Health Surveys for England, with
only small amounts of data available for the devolved regions.13 The most recent data
available (2001-2002) shows overweight and obesity levels (combined) for boys to
average 21.8% and for girls to average 27.5% between the ages of 2 and 15 years.
Obesity alone is typically around 5% for boys and 7% for girls. Prevalence levels of
overweight and obesity tend to be higher during the years leading up to puberty and
during puberty for both genders, but at all ages the levels exceed 16% for boys and
21% for girls.
Figure 1-9. Overweight and obesity by age, 2001-2002 (boys)
Boys aged 2-15
0
5
10
15
20
25
30
23456789101112131415 7
to
10
13
to
15
all
2-
15
age (years)
Prevalence (%)
overweight
obese
Source: Health Survey for England 200213
22
Figure 1-10 Overweight and obesity by age, 2001-2002 (girls)
Girls aged 2-15
0
5
10
15
20
25
30
35
2 3 4 5 6 7 8 9 10 11 12 13 14 15 7
to
10
13
to
15
all
2-
15
age (years)
Prevalence (%)
overweight
obese
Source: Health Survey for England 200213
In April 2006, the Department of Health added a new table of data to its existing web-
based Health Survey, with which trends in childhood obesity could be identified.14
Those data included estimates of rates of overweight and obesity in children in
England for 2003 and 2004, and providing annual trend data from 1995 onwards.
Figures for Scotland obtained in the Scottish Health Survey 1998 show average BMI
values for children to be higher than for their English counterparts in the same year in
all age groups (2-6, 6-11 and 12-15) and for both genders. The prevalence levels for
overweight and obesity are not given. Children aged 4-11 showed higher prevalence
levels for overweight and obesity in Scotland compared with England in surveys
undertaken in 1984 and 1994.
In Northern Ireland, a survey of children aged 4-5 years conducted in 2002-2003
found 6% of children obese, and a survey of children 2001-2002 in primary and
secondary schools also fond obesity prevalence of 6%, a figure that was slightly
higher than that found among children in the Health Survey for England conducted
the same year.15
Trends across time are available for younger children aged between five and ten years
old. These show a rapid rise in overweight and obesity (combined) during the 1990s
which has continued to the latest available survey (2002-2003). In the last two
decades, the prevalence of overweight (including obesity) has virtually trebled, and
the prevalence of obesity alone has more than trebled.
23
Figure 1-11. Overweight and obesity trends in boys and girls aged 5-10 years,
England, 1974-2003
Children: trends 1974-2003
0
5
10
15
20
25
30
35
1970 1980 1990 2000
Prevalence (%)
Boys overweight (inc obese) Girls overweight (inc obese)
Boys obese Girls obese
Source: Health Survey for England 20034 and Chinn and Rona 200116
Ethnic and socio-economic differences have been studied in children in the 1999
Health Survey for England.17 Income trends were apparent in child obesity prevalence
rates (see Figure 1-12) with girls from lower income households especially likely to
be obese. Household income differences for overweight, non-obese children were not
significant.
Afro-Caribbean girls were more likely to be overweight than girls in the general
population, and these girls along with Pakistani girls were more likely to be obese.
Indian and Pakistani boys were more likely to be overweight. There were no
significant differences in the prevalence of overweight and obesity in children from
different social classes.
A smaller-scale survey of over 2,000 adolescents aged 11-14 in East London reported
high levels of overweight and obesity among all ethnic groups examined (white
British, Indian and Bangladeshi), with the highest levels among Indian males (36%
overweight) and black African females (40% overweight).18 No association between
BMI and measures of socioeconomic status were found although the group as a whole
was relatively deprived compared with the UK population (37% from homes with no
employed parent, 48% from homes eligible for social benefits such as free school
meals).
24
Figure 1-12. Obesity trends across household income levels (quintiles), England
2001-2002
Child obesity and income level
0246810
Lowest
4th
3rd
2nd
Highest
Prevalence (%)
Girls
Boys
Source: Health Survey for England 20034
1.3 Summary of main points
The prevalence of overweight and obesity in the U.K. has increased over the
past 2 decades for both adults and children, especially in relation to obesity
(BMI 30 kg/m2).
Over 75% of adult men and 57% of adult women are estimated to be
overweight or obese by age 35 in England, with lower numbers reported for
Scotland, Northern Ireland and Wales (although data from estimates for the
latter two countries are derived from self-reports).
Among children aged 2-15 years, the prevalence of overweight and obesity is
estimated at a minimum of 16% for boys and 20% for girls at any age group.
The prevalence of obesity among children has more than trebled over the past
two decades.
The Health Survey for England has indicated differentials in terms of the
prevalence of obesity across ethnicity and social and economic status for both
adults and children, especially among females. Women and girls of Afro-
Caribbean and Pakistani backgrounds show a higher prevalence of obesity
than the population in general, as do women and girls of lower social and
economic status (lower employment status or household income).
25
2 Estimated costs of obesity
This section will consider two aspects of the costs of obesity: the health costs – i.e. the
ill health associated with obesity – and the financial costs, which are largely a
consequence of the health costs.
2.1 Health consequences of overweight and obesity
The World Health Organization has estimated the relative importance of different risk
factors for disease in their impact on the total burden of disease experienced by
develop countries, measured in terms of the number of lost years of healthy life
(Disability-Adjusted Life Years – DALYs). The top ten risk factors, responsible for
about 60% of total DALYs, are (in decreasing order of magnitude) tobacco smoking,
raised blood pressure, alcohol consumption, raised blood cholesterol, overweight, low
fruit and vegetable intake, physical inactivity, illicit drug taking, unsafe sex and iron
deficiency.19 This list mixes some causes and consequences: for example, excess
bodyweight is closely linked to raised blood pressure and raised cholesterol levels.
Overweight and its consequences together exceed the burden of ill health linked to
tobacco, and, with dietary inadequacies and physical inactivity, accounts for over 20%
of the total disease burden (in DALYs) in developed economies.
Several reviews have identified the major health consequences of obesity, which
range from life-threatening diseases and premature death to non-fatal but disabling
complaints which have an adverse effect on the quality of life.20 Overweight and
obese individuals are at increased risk for life-threatening diseases which are among
the principal causes of death in England, including:
Coronary heart disease
Hypertension and stroke
Certain cancers (endometrial, breast, colon, oesophagus and kidney)
Type 2 diabetes
In addition, non-life-threatening problems associated with overweight and obesity are
also prevalent in the UK, including:
Orthopaedic disorders and osteoarthritis
Gall stones
Sleep apnoea and asthma
Menstrual abnormalities
Polycystic ovary syndrome
Liver steatosis
Low self-esteem, low achievement and depression
26
Table 2-1. Estimated increased risk for obese people developing obesity-related
diseases (non-obese = 1.0)
Disease
Relative risk for
women Relative risk for
men
Type 2 Diabetes 12.7 5.2
Myocardial Infarction 3.2 1.5
Angina 1.8 1.8
Hypertension 4.2 2.6
Stroke 1.3 1.3
Cancer of the Colon 2.7 3.0
Ovarian Cancer 1.7 -
Gall Bladder Diseases 1.8 1.8
Osteoarthritis 1.4 1.9
Source NAO 200121
In addition to the associated illness, the UK National Audit Office estimated that over
30,000 deaths in England were attributable to obesity in 1998, approximately six per
cent of all deaths in that year.19 This figure has since been revised upwards, to 34,100
deaths for 2002, equivalent to 6.8% of all deaths.22 This compares to about 10 per cent
of all deaths due to smoking, and less than one per cent from road accidents. In total,
this amounted to 275,000 lost years of life – on average, each person whose death
could be attributed to obesity lost nine years of life, with some 9,000 of the deaths
related to obesity occurring before state retirement age (1998 figures).
For young adults, the risk of an earlier death for someone with a BMI of 30 is about
50 per cent higher than that for someone with a BMI in the range 20 to 25, and with a
BMI of 35 the risk is more than doubled NAO.19
27
Figure 2-1. The relationship between increasing body mass (BMI) and relative
risk of early death (Women only, USA database)
Source: NAO 200121, citing Manson et al 199523
The average loss of life attributable to obesity has been estimated to be over two years
on current UK life expectancy statistics and is expected to rise to over five years as
healthy life expectancies increase faster for normal weight than for obese people.
76.5
82.7
77.4
2.6 years
5.3 years
Age
Projected male
life expectanc
y
at birth
Projected male life expectancy at
birth if obesity/overweight trends
continue
73.9
70
75
80
85
2003 2009 2015 2021 2027 2033 2039 2045 2051
Year
Source: UK
Department of Health 200524
Figure 2-2. Projected life expectancy for men: trends over the next 50 years
28
2.2 Health care costs
Illness associated with obesity gives rise to costs to the UK’s national health services
(NHS). Direct costs of obesity have arisen mostly from medical consultations, drugs
and treatments of diseases attributable to obesity. Using 1998 figures for England, the
National Audit Office estimated that the costs of obesity amounted to £480 million, or
about 1.5 per cent of NHS expenditure in that year. Of this total, only £9.5 million
was spent on obesity itself and consisted mostly of the cost of consultations with
general practitioners because of a lack of obesity treatment options available at the
time. The bulk of the cost arose from treating conditions caused by obesity. The major
cost drivers are hypertension, coronary heart disease, and Type 2 diabetes, which
together accounted for £386 million. Osteoarthritis and stroke accounted for a further
£52 million of costs.
These estimates are acknowledged to be low compared to the findings of studies
undertaken overseas: in countries where the prevalence of obesity is similar to that in
England, the direct costs of obesity have been estimated to lie between two and six
per cent of national health care budgets.20 If this range applied in England, the direct
costs to the NHS of treatment for obesity and its consequences would have been
between £0.7 and £2.1 billion in 1998. Subsequent calculations suggest that, with the
rising prevalence of obesity and better evidence of the links between obesity and
various medical conditions, the condition could account for upwards of 2.5% of health
service costs.20
Table 2-2. Estimates of the direct costs of obesity to national health services
Prevalence of obesity
(BMI>30)
Country
Year of
estimate
Proportion of
total healthcare
expenditure due
to obesity At time of
estimate Latest
available
USA 1999 8.5% 30.5% 30.5%
USA 2000 4.8% 30.5% 30.5%
Netherlands 1981-89 4% 5.0% 10.3%
Canada 1997 2.4% 14.0% 13.9%
Portugal 1996 3.5% 11.5% 14.0%
Australia 1989/90 >2% 10.8% 22.0%
England 1998 1.5% 19.0% 23.5%
France 1992 1.5% 6.5% 9.0%
Source: House of Commons 200420
For the UK, updated figures for health service costs were provided for the
parliamentary enquiry into obesity in 2003-2004. These are as follows:
29
Table 2-3. The estimated health service costs for treating obesity and
consequential diseases, England 1998 and 2002
1998
£ million
2002
£ million
GP consultations 6.8 12 – 15
Ordinary admissions 1.3 1.9
Day cases 0.1 0.1
Outpatient attendances 0.5 0.5 - 0.7
Prescriptions 0.8 13.3
Total cost of treating obesity 9.5 45.8 - 49.0
GP consultations 44.9 90 – 105
Ordinary admissions 120.7 210 – 250
Day cases 5.2 10 – 15
Outpatient attendances 51.9 60 – 90
Prescriptions 247.2 575 – 625
Total cost of treating the consequences
of obesity 469.9 945 - 1,075
Source: House of Commons 200420
2.3 Other economic costs
The premature death of members of the economically active workforce leads to lost
productivity. In England, UK National Audit Office estimated that, in 1998, obesity
caused a loss of over 40,000 years of working life in the pre-retirement working
population.
Besides these losses, obesity was also estimated to lead to over 18 million days of
sickness absence (1998 figures). This figure is based on the days of medically
certified sickness absence attributable to obesity and its consequences. This figure is
likely to be an underestimate as it excludes both self-certified and uncertified sickness
absence, and takes no account of sickness due to diseases for which the proportion of
cases attributable to obesity cannot be quantified. For example, back pain associated
with obesity is excluded from the estimates, as there was no figure for relative risk on
which to make the calculations. Back pain is one of the most common causes of
sickness absence and its inclusion could increase the estimate significantly.
30
Table 2-4. Estimated days of certified sickness attributable to obesity, England
1998
Reason for work
absence Days
Obesity 418,000
Type 2 Diabetes 5,960,000
Hypertension 5,160,000
Angina Pectoris 2,390,000
Myocardial Infarction 1,230,000
Cancers* 970,000
Osteoarthritis 950,000
Gout 530,000
Stroke 440,000
Gallstones 20,000
Total 18,068,000
Source: NAO 200119
* Endometrial cancer, colon cancer, rectal cancer, ovarian cancer and prostate cancer
combined
Table 2-5. The estimated work-related costs of obesity and consequential
diseases, England 1998 and 2002
NAO 1998
£ million
2002
£ million
Lost earnings due to
attributable mortality 827.8 1,050-1,150
Lost earnings due to
attributable sickness 1,321.7 1,300-1,450
Total indirect costs 2,149.5 2,350-2,600
Source: House of Commons 200420
Combining the direct health costs and the work-related economic costs gives a figure
of over £2.6bn in 1998, rising to between £3.3bn and £3.7bn in 2002.
These estimates are based on England, not the whole of the UK. England is home to
84% of the UK population, and on the assumption that the costs can be extrapolated in
direct proportion, the costs for the UK (in 2002) were £3.9bn – £ 4.4bn.
Furthermore, the costs were estimated for obese people, not for those overweight but
not obese. Overweight leads to a raised risk of disease to a lesser extent than obesity,
typically around a quarter of the raised risk of premature death compared with obese
people, but the numbers of overweight people in the population are typically two to
31
three times the number of obese people. It would not be unreasonable to add a further
£1bn to the figure for 2002 for the UK.
Furthermore the figures presented here refer only to adult obesity. The costs of
childhood obesity and overweight have not been considered. It is impossible to put a
figure on the lost productivity – due to the need for care by a parent or for the child’s
own lost earning due to lost school education. Medical costs for child obesity have not
been estimated, except in one US study where ailments related to obesity (primarily
asthma, sleep apnoea, diabetes and gall bladder problems) accounted for some 1.7%
of US hospital costs. An equivalent figure for the UK, taking account of lower obesity
prevalence and lower health care costs, may add another £0.5bn to the total above,
including both the health care and the employment costs.
These figures were calculated using 2002 information. Updating these figures to 2006
prices would give an estimated total for the costs of obesity in the UK of around £6bn.
These figures should be raised further due to the cost of the social and psychological
effects of being obese. Excess bodyweight is linked to a lower likelihood of finding a
marriage partner, of finding work and of being promoted. Overweight people are
likely to be on lower earnings (perhaps reducing the lost-days-of-work costs) but are
more likely to suffer low self-esteem and depression. Psychiatric problems, especially
depression, are the largest single cause of lost days of healthy life in developed
economies, and besides the human costs they are a major cost to the health services
and a cause of lost productivity.
Lastly, there are the costs or benefits (depending on perspective) of the weight-loss
industry. At any one time some 12 million people in Britain are attempting to lose
weight. In 2001 the UK market for slimming products was estimated to be worth
some £5.2 billion (€8.2bn) and was forecast to reach £6.6 billion by 2006.25
2.4 Summary of main points
The health consequences of obesity range from life-threatening diseases and
premature death to potentially disabling conditions affecting quality of life.
In 2002 it was estimated that approximately 34,100 deaths were attributable to
obesity, or 6.8% of all deaths ( as compared to 10% for smoking or 1% for
road accidents).
The average loss of life attributable to obesity was estimated to be over two
years in 2003 and is expected to rise to over five years as healthy life
expectancies increase faster for normal weight than for obese people
The direct cost of treating obesity in England was estimated £480 million in
1998, or about 1.5 per cent of National Health Service expenditure in that
year, with the bulk of the costs arising from treating conditions caused by
obesity. Updated figures for health service costs in the UK were provided for
the parliamentary enquiry into obesity in 2003-2004, and we around twice the
1998 estimates (£991 million - £1124 million), and it was suggested that the
condition could currently account for upwards of 2.5% of health service costs.
The indirect cost of obesity related to lost earnings because of death or illness
attributable to obesity have been estimated at £2.1 billion in 1998 and £2.35 -
£2.6 billion in 2002.
32
Combining the direct health costs and the work-related economic costs
estimated for England in 2002 (£3.3 - £3.7 billion) leads to figures which can
be used to estimate, via direct extrapolation, the total cost of obesity in the UK
as a whole in 2002: £3.9 – £4.4 billion.
When adjustments are made to include costs associated with overweight pre-
obese adults (+ £1 billion) and childhood overweight and obesity (+ £0.5
billion) estimates of the cost of obesity in 2006 are around £6 billion.
3 Trends in food consumption and physical activity
3.1 Sources of data
There are seven main sources of information on food consumption and physical
activity patterns, as follows:
1. Trends in national supply data, based on agricultural production, imports,
exports and changes in stock levels. From these data estimates of food supplied to
the domestic market can be obtained. Estimates are also made of wastage and
processing losses to arrive at a figure for food ‘moving into consumption’ within the
UK. The figures rely on estimates supplied by national governments to bodies such as
the UN Food and Agriculture Organization and the Organisation for Economic Co-
operation and Development (OECD), and given the different methods which may be
used by countries to arrive at their data, the data are unreliable for making close
comparisons between countries, but do have validity for analysing trends within a
country. The data are supplied as total quantities and quantities per unit population,
and as such do not take account of changes in population structure (e.g. more older
people, changes in inequalities). Of special interest in these datasets are trends in per
capita food energy supply (caloric content) and trends in specific nutrients such as
fats, oils and sugars.
2. Figures for commercial markets and sales figures and their trends over time.
The availability of market surveys may be subject to commercial restrictions. Market
data are valuable in identifying specific sub-sectors of the population – e.g. sales to
children, or sales of children’s products – and shifting patterns of consumption within
specific sectors – e.g. a rise in low calorie beverages within the soft drinks sector.
Also of interest are trends in investment in food supply and manufacturing, and trends
in agricultural support policies that may influence investment programmes. Trends in
advertising, retailing and catering – such as the rise in fast food outlets – may also be
helpful in analysing the ‘upstream’ factors affecting food choices.
3. Trends in household purchases, shopping or food basket surveys. Most
countries take sample surveys of shopping behaviour in order to make economic
forecasts, to monitor inflationary trends and to estimate changing market demand.
Surveys may be limited to price issues (the cost of common items) or may include
quantitative food data indicating the amount of specific types of food being
purchased. Surveys usually need to be interpreted in the light of the household
structure – number of adults, children of various ages etc. The studies may be
valuable for indicating socio-economic differences in purchasing patterns and dietary
quality.
4. Dietary surveys of samples of individuals. These are generally expensive to
mount, especially if they are to be representative of the whole population, and are
unlikely to be undertaken sufficiently frequently to be able to offer more than very
rudimentary trend data. Dietary surveys are also prone to reporting error: they rely on
recall of recent food consumption or on diaries being kept by the individuals
concerned.
34
5. Data on food security may also be relevant: these are traditionally measured in
terms of national capacity (e.g. the production of staple foods as a proportion of total
consumption, or the amount of staple foods held in storage as a proportion of daily
national consumption) but there are also measures of household food security, such as
the amount of total household income that is spent on obtaining food. The relative
prices of foods (relative to other foods or to overall expenditure) may indicate
economic incentives to consume.
6. Also of interest are data on food accessibility and availability: this is especially
important to groups of people with low household food security. Accessibility (is the
food available within a reasonable distance/time) and availability (is the actual food
wanted available at the locations people have access to) are obviously linked to
affordability – with food effectively unavailable if it is priced above a family’s
budget, and inaccessible if the family cannot afford the necessary transport and time
to gain access.
7. Physical activity patterns are far less well recorded than are food supply and
consumption patterns. Surveys using self-reported data of habitual activity have been
undertaken but otherwise most data is indirect – such as measures of community
cycling activity, sales of bicycles, or membership of fitness facilities. Surveys of
physical inactivity are also sparse: indirect measures include TV watching, video
watching and computer use, including gaming and online gaming, and sales of games
machines.
3.2 Food consumption
Food balance sheet figures from the UN Food and Agriculture Organization (FAO)
indicate a trend within the UK towards greater food energy consumption per head
since the 1970s.
Figure 3-1. Increasing food energy supplied to consumers
Per capita food energy supply
3000
3100
3200
3300
3400
3500
1971 1976 1981 1986 1991 1996 2001
kcal/person/day
Sources: (i) FAO Food balance sheets, three year averages 26;
(ii) National Statistics, UK population estimates (latest adjustments Oct 2004) 27
35
The FAO figures indicate an increase in calorific energy supplied by food of some
150-300 kcal/person/day. For an average adult an increase in food intake of 100kcal
per day could lead to a gain in bodyweight of 4kg over a year, assuming no increase
in energy expenditure and that 0.45 kg (or 1 pound) of fat represents an energy
reserve of about 4000 kcal28. For the average man this represents about one BMI unit,
and for a woman about 1.3 BMI units. Thus an average daily surplus of 100kcal in
energy balance for an adult with a BMI of 22 kg/m2 could lead to him or her
becoming overweight in 2-3 years and obese in 6-8 years. Snack foods containing 100
kcal are easily found (a 330ml can of soft drink typically contains 120kcal, a 30g bag
of potato snacks 150kcal, and a 50g portion of chocolate 250kcal), whereas increases
in energy expenditure require a greater level of effort to obtain.
There is no clear pattern to identify the source of the increasing calorific intake over
the last two decades: FAO data for supplies moving into consumption show declining
sugar supplies, and small rises in the quantities of fats and proteins over the same
period (Figure 3-2). However, the figure for sugars refers primarily to sugar obtained
from cane, beet and corn and ignores sugars contained in fruits and fruit products, and
foods sweetened with fruit products, such as apple concentrate. Also of note is that
gram per gram, fat has over twice the amount of calories than either sugar or protein
(9 kcal/g vs. 4 kcal/g). As such, an increase in the amount of fat in the diet raises its
energy density to a greater extent than comparable increases of either sugar or protein.
Higher energy density has been associated with higher energy intakes and higher
prevalence of obesity in adults29, and as such may be the most salient nutrient behind
increased caloric consumption. Examination of the food balance sheets suggests that
vegetable oils have been the major source of increasing fats in the food supply: in the
early 1970s the supplies of fats were around 37g from vegetable sources and 56g from
animal sources, per person per day; while by the early 2000s the figures were 44g and
57g respectively.
36
Figure 3-2. Changes in per capita daily supplies of fats, sugars and protein, 1971-
2001
Major nutrient supplies moving into consumption
70
80
90
100
110
120
130
140
150
1971 1976 1981 1986 1991 1996 2001
grams/person/day
protein sugars fat
Sources: (i) FAO Food balance sheets, three year averages 26;
(ii) National Statistics, UK population estimates (latest adjustments Oct 2004) 27
Household food supply figures tend to suggest a fall in total energy consumed over
the same period. The data collected in household food surveys relies on a member of
the household keeping a weekly record of all foods entering the purchases for home.
From 1993, a smaller sample of households provides a record of all meals, snacks and
drinks consumed outside the home. Although these records provides useful details for
comparative purposes (e.g. between sub-populations or across time) they are less
reliable in terms of quantitative analysis, in part because no attempt is made to
account for uneaten foods and in part because self-reported consumption, especially
of foods eaten outside the home, is likely to be unreliable and below actual
consumption levels.
37
Figure 3-3. Decreasing food energy in household purchases
1500
1700
1900
2100
2300
2500
2700
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
kcal/person/day
Includes soft drinks, alcoholic beverages and confectionary brought into the home.
Excludes all foods and drinks eaten outside the home.
Source: Defra National Food Surveys 30, 31
Since 1993, the survey has included estimates of food and drink eaten outside the
home. In the period 1994-1998 the estimates ranged between 240 kcal/person/day to
265 kcal/person/day.31 In 2002-3 this figure dropped to 210 kcal/person/day and in
2003-4 to 205 kcal/person/day.31 (Of these daily figures, alcoholic drinks consumed
outside the home accounted for about 30kcal/person.) These low figures, and the
downward trend, are surprising and are probably inaccurate, given that the
expenditure on foods eaten outside the home has risen substantially as a proportion of
total food expenditure.
38
Table 3-1. Household expenditure on food and drink, indicating the proportion
spent outside the home, 1980-2004
Total food and
drink
£
Outside home
£
% outside
home
1980 22.70 3.90 17%
1985 32.70 5.80 18%
1990 41.67 8.68 21%
1995 52.88 14.38 27%
2000 61.90 20.50 33%
2004 79.60 26.00 33%
Expenditure per household per week. Includes alcoholic drinks.
Source: Family Expenditure Surveys 1984 32, Family Spending 2003-4 33
A further cause of concern about the use of self-reported dietary and food purchase
records is their discrepancy from industry figures for sales of foodstuffs. For example,
the household food survey reports average weekly purchases of confectionery to be
138 grams per person, plus a further 20-25 grams per person eaten out of the home,
for the year 1996. The same year an industry market survey reported UK sales of
confectionery equivalent to 250 grams per person per week, suggesting rates of
underestimation as high as 40%. Similarly, in 1996, soft drinks purchases reported in
the household survey were around 1.6 litres/person/week plus a further 0.3 litres
consumed away from home. In that year industry sales in the UK reached 3.3
litres/person/week.
From an economic point of view, increasing food consumption may be expected at a
time when food prices fall relative to income. In the UK, the retail price index for
food items has risen less quickly than has the index as a whole since the mid-1970s.
Household income has risen more rapidly than the retail index, and hence income
spent on food items has fallen as a percentage of all expenditure.
Table 3-2. Household expenditure on food, showing a decline 1984-2004
especially among lower income households
1984 2003-4
all
households lowest
20% highest
20% all
households lowest
20% highest
20%
Food
expenditure £31.43 £14.50 £49.43 £64.90 £31.30 £103.50
All expenditure £151.92 £50.44 £291.41 £356.20 £142.25 £637.10
Proportion 21% 29% 17% 18% 22% 16%
Expenditure per household per week. Excluding alcoholic beverages.
Source: Family Expenditure Survey 198432, Family Spending 2003-433
39
The reduction in relative food prices needs to be considered in greater detail. While
prices for fresh produce has tended to increase at rates above-inflation rates, mass-
produced processed foods have not increased so rapidly, making their purchase
increasingly attractive to shoppers, perhaps especially to lower-income householders.
Furthermore, mass-produced products tend to be calorie-rich compared with fresh
produce, encouraging a ‘cheap calorie’ culture for households on tight budgets.
Table 3-3. Typical price (in pence) per 100 kcal (2000 prices), changes in price
relative to general food price index 1992-2000, and consumption levels by high
and low income households, 2000
Typical
price per
100 kcal
(pence)
Change
in price %
RPI
Amount
purchased high
income
(g/person/week)
Amount
purchased low
income
(g/person/week)
Sugar 2 82 57 116
Margarine 2 84 13 22
Other oils and fats 3 111 136 178
White bread 3 84 175 375
Wholemeal bread 4 90 94 65
Frozen potato products 7 103 158 224
Biscuits and cakes 7 103 232 254
Processed fruit products 7 100 64 54
Full fat milk 7 77 479 895
Low fat milk 10 78 1136 1057
Confectionery 11 104 66 62
Soft drinks, not diet 12 95 511 703
Frozen fish products 21 100 29 41
Fruit juices 21 91 483 211
Meat products 22 106 633 723
Carcase meat 23 102 216 248
Fresh fruit 26 106 895 573
Fresh fish 51 135 45 25
Other fresh vegetables 62 108 593 391
Fresh green vegetables 83 122 264 190
Soft drinks, diet >100 94 465 279
Household income high=>£725/week, low=<£180/week
Sources: (i) Food price data from National Food Surveys 30
(ii) RPI data from National Statistics 34
(iii) Household food purchases from National Food Survey 2000 35
40
3.3 Physical activity
Physical activity is not easy to measure and most surveys have relied on self-reported
estimates of time spent in various forms of activity and inactivity. A questionnaire
survey of over 1000 adults in Great Britain, conducted for Eurobarometer in 2002,
found that around 60% of the respondents had taken no vigorous physical activity in
the previous seven days.36 Around 15% had taken vigorous physical activity more
than three times in that period. Some 45% of respondents had not undertaken
moderate physical activity (e.g. cycling or carrying light loads, but excluding
walking). About 18% said they had not undertaken as much as a 10-minute walk in
the last week. Around one in three respondents did not agree with the proposal that
‘The area where I live offers me many opportunities to be physically active’.
In terms of inactivity, the Eurobarometer report found that British adult respondents
said they were typically inactive (sitting at a desk, reading, chatting, viewing
television) for less than 3 hours a day, or 3 to 5.3 hours a day, or more than 5.3 hours
a day in almost equal numbers. According to the UK 2000 Time Use Survey37 British
adults on average spend 8.7 hours asleep each day, 2.5 hours watching television, 1.4
hours eating and drinking and 1.8 hours listening to music, reading, chatting or
‘resting’. Children aged 8-15 spent 10.2 hours asleep, 2.3 hours watching TV, 1.1
hours eating and 1.2 hours on the other inactive behaviours.
Using a different survey methodology, figures for television watching suggest an
average adult spends 3.6 hours per day watching television38 (averaged across 1997-
2001) dropping to 3.2h among higher income households and rising to 4h among
lower income households. The figures are also strongly age-dependent, rising from
3.1h at age 16-34 to 4.9h at age 55+. These surveys show children to be watching
about 2.6h per day.
Trends in physical activity and inactivity over time have not been consistently
measured. The Eurobarometer survey was the first of its kind so no previous
comparable figures are available. The Health Survey for England 2003 reported
comparisons of activity across 1997-2003 by age group and sex, based on self-
reported frequency of undertaking the current recommendations for cardiovascular
health of ‘at least 30 minutes of moderate-intensity activity at least five days per
week’, as shown in the table below.
41
Table 3-5. Percentage of adults achieving cardiovascular physical activity
recommendations
Age
Men
16-24
% 25-34
% 35-44
% 45-54
% 55-64
% 65-74
% 75+
%
Total
%
1997 49 41 37 32 23 12 7 32
1998 53 45 41 34 30 14 6 34
2003 53 44 41 37 32 17 8 35
Women
1997 26 26 29 24 19 8 5 21
1998 28 28 28 25 18 9 3 21
2003 30 29 30 30 23 13 3 24
Source: HSE 200339
Self-reported activity among children indicates that significant numbers are achieving
the recommended minimum levels of at least 30 minutes moderately intense activity
on at least five days per week. The survey of Health Behaviour in School-aged
Children40 included England, Scottish and Welsh samples. Some 29% of girls and
48% of boys in England, 23% of girls and 38% of boys in Scotland, and 18% of girls
and 39% of boys in Wales are meeting the recommended activity levels.41 (Children
were aged 11-15 and sampled in 2001-02.)
The Welsh Assembly Technical Report on Young People42 reported the percentage of
children undertaking at least four hours of physical activity per week across 14 years
1986-2000. School-based physical activity was not included. The results suggest no
specific increase or decrease in out-of-school activity levels, and a similar lack of
secular trends was found in terms of the proportion of children exercising on at least
four occasions (time limit unspecified) outside of school.
42
Figure 3-4. Proportion of children in Wales exercising for at least 4 hours per
week, 1986-2000
Source: Welsh Assembly 200242
A survey of participation in out-of-school sporting activities has shown a trend
towards increased numbers of children becoming active in sport, over the period
1994-1999,43 rising from 36% of children to 45%, but the figure declined to 42% in
2002.44 School curricula in Britain are widely reported as having reduced the number
of hours of physical exercise in the timetable since the 1980s, but there is some doubt
as to whether this has a significant effect on children’s overall activity levels. A
comparison of the measured activity (using accelerometers) of children from three
schools with very different timetabled exercise levels suggests that children
compensate: out-of-school activity levels for children in schools with low levels of
timetabled activity fully compensated for the lost exercise.45
Active transport (walking, cycling etc) has not been monitored. Sales of bicycles are
reportedly high, but cycle use may not reflect cycle ownership. For children, there is
evidence that the proportion walking or cycling to school has fallen over the last two
decades for both primary and secondary level school children. In both age groups the
proportion being driven to school by car has approximately doubled, although the
total is still less than half of primary children and a quarter of secondary children.
0
10
20
30
40
50
1986 1988 1990 19921994 1996 1998 2000
Percentage
boys girls
43
Figure 3-5. Trends in transport to school, 1985/6-2002
1992/94 2002
1985/86 1992/94 2002
1985/86
Walk
Car
Bus
Bicycle
Other
Age 5-10 Age 11-16
0
20
40
60
80
100
Proportion using different forms of transport
Source: Transport Trends Summary, Department for Transport, 2004 46
Trends in television watching indicate no significant increase in children’s viewing
hours since the early 1980s, when children aged 12-15 were recorded as watching
2.5h per day. This rose to 3.2h in the late 1980s, with the advent of Channel 4 and
increased daytime and breakfast broadcasting, but fell back to 2.6h and 2.7h for four-
year-olds and 15-year-olds respectively in the early 1990s.47 By the late 1990s it
remained at 2.6h per day.38
44
Figure 3-6. Leisure-time broadcast television watching, 1983-1987 and 1997-2001
0
1
2
3
4
5
6
age 4-15 16-34 35-64 65+
hours/person/day
1983 1984 1985 1986 1987
1997 1998 1999 2000 2001
Source: Social Trends 1985 48, 1989 49; BARB 2002 38
3.4 Summary of main points
From 1971 to 2001, food balance sheets indicate an increase in the number of
calories available to UK consumers from 3240 kcal/person/day to 3400
kcal/person/day.
The UN Food and Agriculture Organization data show declining sugar
supplies (although the use of fruit-based sweeteners is not accounted for), and
small rises in the quantities of fats (mainly of vegetable origin) and proteins
over the same period,
On the other hand, household food surveys report a decrease in the number of
calories entering the home (from about 2540 kcal/person/day in 1974 to 2090
kcal/person/day in 2002) as well as those consumed outside of the home (from
240-260 kcal/person/day in 1994-8 to 205 kcal/person/day in 2003-4). This
seems highly unlikely since: 1) The proportion of total food and drink
expenditure spent outside the home has risen from 17% in 1980 to 33% in
2004; and 2) Industry sales figures from 1996 suggested that self-reported
diaries and food purchase records from that year were likely underestimating
the consumption of confectionery and soft-drinks by as much as 35-50%. The
relative price of food has decreased over the past 2 decades, with a lower
proportion of average household incomes being spent on food (from 21% to
18%). This has been mostly driven by energy-dense, mass produced products
rather than by fresh produce.
Lower-income households were reported as purchasing larger quantities of
cheaper, higher energy-dense food per capita than higher-income households
in 2000, including such food items as sugar, oils and fats, white bread, frozen
potato products and full fat milk.
45
Physical activity levels have been found to be remarkably low in the U.K. with
60% of UK adults interviewed in the Eurobarometer 2002 survey not having
participated in any vigorous physical activity in the week prior to the
interview, and nearly one fifth (18%) not even having taken a 10 minute walk
in the previous 7 days.
The average percentage of adults achieving cardiovascular recommendations
for physical activity remains low, but has shown modest increases from 1997
to 2003 (from 32 to 35% of men and 21 to 24% of women).
The average adult in the UK spends an estimated 3.6 hours/day watching
television.
Only one-fifth to one-third of girls and one-third to one-half of boys aged 11-
15 sampled around the UK in 2001-2002 participated in at least 30 minutes of
moderately intense activity on 5 days per week.
Though there was a modest increase was reported in the number of children
becoming active in sport from 1994 to 1999 (from 36% to 45%) it was
followed by a slight decrease in 2002 (to 42%).
There is some indication that children compensate for lower levels of activity
in school with higher levels of activity outside of school in the UK.
The proportion of children walking or cycling to school has fallen between
1985/6 and 2002, and the number being driven has doubled in the same period.
No significant increases have been evident in the number of hours children
spend watching television in the UK from the early 1980s to 2001, with
estimates remaining around 2.6 hours of television per day.
46
4 Policy-making institutional structures
Primary responsibility for enacting health policy in the UK resides with the
Department of Health (DoH) and the corresponding devolved departments for Wales,
Scotland and Northern Ireland. Those departments are advised by a series of advisory
bodies and by a Chief Medical Officer. The DoH is led by the Secretary of State for
Health, who is a member of the Prime Minister’s cabinet. The Secretary of State for
Health is supported by a team on junior ministers who are responsible for specific
areas of health policy, including public health, health service delivery, standards, and
specific medical services.
The Department of Health lists 39 advisory bodies on specific topics ranging from
acupuncture to vaccines and xeno-transplantation. There is no advisory body
specifically responsible for advising on obesity or obesity-related diseases, but there is
a stakeholder forum on nutrition (the Nutrition Forum) and an expert panel on
nutrition (the Scientific Advisory Committee on Nutrition). The latter replaced an
extensive network of advisory groups under the Committee on Medical Aspects of
Food Policy, in March 2000.
Food policy is also a responsibility of the Department of the Environment, Food and
Rural Affairs (DEFRA) and the Food Standards Agency (FSA). DEFRA is primarily
responsible for food supply policies such as agricultural policy, as well as
environmental issues and rural development. The UK, as a member of the European
Union, participates in the Common Agriculture Policy and common policies in
fisheries and in trade of agricultural and food products with non-EU members. A
review of food and farming policies was undertaken in 2001-2002, but considerations
of public health were tangential to its agenda.50
The Food Standards Agency, which describes itself as ‘an independent Government
department’, was established in 2000 in part as a response to the crisis in public
confidence over food safety following the epidemic of Bovine Spongiform
Encephalopathy and various other food safety concerns in the 1990s. The Agency is
not specifically answerable to a government department but reports to Parliament
through Health ministers and through the relevant devolved administrations. The
Agency is primarily advisory to government: it does not have legislative authority;
although it is responsible for regulatory policy-making in relation to food safety. It
does however have rights of inspection and control, primarily for food safety
practices. It is free to publish the advice it issues. The Agency’s strategic plan for
2005-2010 has as its key aims:
to continue to reduce food-borne disease
to reduce further the risks to consumers from chemical contamination
including radiological contamination of food
to make it easier for all consumers to choose a healthy diet, and thereby
improve quality of life by reducing diet-related disease
to enable consumers to make informed choices.
47
Several bodies deal with policies in relation to physical activity, but primary
responsibility for sport comes under the government’s Department for Culture, Media
and Sport (DCMS). Among its many areas of activity, the Department is responsible
for television regulation, advertising and sports promotion. Its four strategic priorities
are stated as:
Children and Young People : Further enhance access to culture and sport for
children and give them the opportunity to develop their talents to the full and
enjoy the benefits of participation
Communities: Increase and broaden the impact of culture and sport, to enrich
individual lives, strengthen communities and improve the places where people
live, now and for future generations
Economy: Maximise the contribution that the tourism, creative and leisure
industries can make to the economy
Delivery: Modernise delivery by ensuring our sponsored bodies are efficient
and work with others to meet the cultural and sporting needs of individuals
and communities.
The DCMS sponsors 39 non-departmental public bodies, including many leading
museums, arts and architecture councils and two bodies specifically related to sports
promotion: UK Sport and Sport England. UK Sport is responsible for managing and
distributing public investment (£29 million annually) and is a statutory distributor of
funds raised by the National Lottery (9.2% of sport allocation). Sport England is also
a distributor of public funds, emphasising its role ‘creating opportunities for people to
start in sport, stay in sport and succeed in sport’ and states that it is ‘responsible for
delivering the Government’s sporting objectives’.
The DCMS, acting jointly with the Department of Trade and Industry, oversees the
Office of Communications (Ofcom), which describes itself as ‘the regulator and
competition authority for the UK communications industries, with responsibilities
across television, radio, telecommunications and wireless communications services’.
Ofcom is responsible for making recommendations to government on the regulation
of advertising through these media.
The Department for Education and Skills (DfES) is responsible for food and
physical activity policies in schools. In 2005, a School Food Trust was set up in
England (funded by the DfES and the Lottery Fund) with the aim of promoting the
health and education of children and improving the quality of food supplied and
consumed in schools. Its specific targets include ensuring that lunch and non-lunch
foods served in schools meet appropriate standards, increasing the uptake of school
meals and improving food education. The DfES also provides resources and
professional development for teachers on issues related to health and nutrition
education, as well as the writing and implementation of a whole school food policy
guide. In England there are plans to monitor adherence to the new school food
standards to be implemented in 2006 through the Office for Standards in Education
(OFSTED), an independent department responsible for the inspection and regulation
of childcare, schools, colleges, children’s services, teacher training and youth work.
48
4.1 Public Service Agreements
Health policies, and many other government policies, are increasingly being codified
in the form of Public Service Agreements – documents outlining targets and
responsibilities for actions between government departments or between government
and local authority services. One specific, target-led Public Service Agreement (PSA)
relates directly to obesity, and is to be delivered by the Department of Health in
collaboration with the Department for Education and Skills (DfES) and the
Department for Culture, Media and Sport (DCMS), and in principle other
departments. The PSA requires the responsible departments To halt the year-on-year
increase in obesity among children under 11 by 2010 in the context of a broader
strategy to tackle obesity in the population as a whole.51
Besides this obesity-specific PSA target, there are two targets aimed at increasing
population physical activity levels. The first, which is under the responsibility of the
DCMS, aims: By 2008, to increase the take up of sporting opportunities by people
aged 16+ from priority groups.
The second, under the joint responsibility of the DCMS and the DfES, aims to:
Enhance take-up by 5-16 year-olds of a minimum of 2 hours a week of high quality
PE and sport by 75% by 2006 and 85% by 2008 in England.
In addition, there are targets for local delivery through schools:
All schools to have active travel plans by 2010
All schools to be in School Sports Partnerships by 2006
These targets have emerged on the basis of a series of public health papers addressing
related issues, starting with a White Paper entitled Choosing Health52 and followed by
Delivering Choosing Health53, Choosing a Better Diet54 and Choosing Activity.55
4.2 Local Delivery Plans
Underpinning the PSA targets set out in the White Paper Choosing Health is a set of
data indicators, which will be the basis for monitoring and performance management
from the Department of Health, through Strategic Health Authorities, to Primary Care
Trusts (PCTs – family health bodies at local level). PCTs are the principle bodies
responsible for delivery of targets. Local planning for Choosing Health is to be
accomplished through Local Delivery Plans, which are already used to deliver other
health services that require co-ordination of agencies at local level.
Local Delivery Plans (LDPs) should be developed in close consultation with local
authority partners and other key stakeholders, acting together in Local Strategic
Partnerships, who are responsible for ensuring that an LDP is aligned with other local
community strategies.
Local Strategic Partnerships have been developed in recognition of the need to ensure
consistency of policies across different departments. The Partnerships involve the
participation of other local service providers, in both statutory and voluntary sectors.
Local municipal and rural authorities, responsible for planning, education, social
services, trading standards, environmental health, and economic development, and
local voluntary bodies generating charitable and community activities, are expected to
49
work together with health authorities in the completion of LDPs. There have been few
comprehensive counter-obesity plans developed to date, although, as noted above,
such plans are expected as part of the Public Service Agreement under the public
health White Paper. Regional health bodies (Strategic Health Authorities, SHAs) will
monitor progress, and will identify ‘Spearhead PCTs’ which will act as pilot schemes
expected to make faster progress than other PCTs.
LDPs will have two main elements: contributions to national targets and locally
agreed targets. Contributions to national targets will be in the form of three-year
trajectories that PCTs will agree with their SHA and the SHA will, in turn, agree a
trajectory with the Department for Health. A PCT’s local targets will be agreed and
monitored by its SHA, and will conform to the principles set out in a guidance
document National Standards, Local Action.56 Local targets will be set in partnership
with local authorities.
Under the delivery plans, local authorities have a role in co-delivering the targets set
out in the White Paper and, with PCTs, are expected to lead implementation, </