A “state of the knowledge” assessment of comprehensive

 

A ‘state of the knowledge’ assessment of comprehensive interventions that address the drivers of obesity

Final Report A Rapid Assessment

Final Report

A “state of the knowledge” assessment of comprehensive interventions that address the drivers of obesity A Rapid Assessment Prepared for the National Health and Medical Research Council (NHMRC) by Tim Gill and Lesley King Adrian Bauman Philip Vita Ian Caterson Stephen Colagiuri Ruth Colagiuri Lana Hebden Sinead Boylan Debra Hector Amina Khambalia Scott Dickinson Maria Gomez The Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney December 2010

ACRONYMS LIST

4

PREFACE

5

EXECUTIVE SUMMARY

7

1.0 THE PROJECT BRIEF

11

2.0 TACKLING THE PROBLEM OF OBESITY IN AUSTRALIA

11

2.1 THE CURRENT PROBLEM OF OBESITY

11

2.2 THE DRIVERS OF OBESITY

11

2.3 THE FORESIGHT OBESITY SYSTEM MAP

13

2.4 DOMAINS OF ACTION

14

2.5 DEFINED ACTION AREAS

15

2.6 DIFFERENT TYPES OF INTERVENTIONS

15

2.7 EXPECTED OUTCOMES OF INTERVENTIONS

16

3.0 THE PROCESSES IN GENERATING THIS REPORT

18

3.1 LITERATURE SEARCH STRATEGY

18

3.2 CATEGORIES OF EVIDENCE

18

3.3 SUMMATION OF EVIDENCE

20

3.4 ASSESSMENT OF THE MERIT OF IDENTIFIED INTERVENTIONS WITHIN EACH ACTION AREA

20

3.5 ADDITIONAL CONSIDERATIONS AROUND THE MERIT OF AN INTERVENTION

21

4.0 BRIEF REVIEW OF CURRENT EVIDENCE

24

4.1 ACTION AREA - EARLY LIFE EXPOSURES AND GROWTH PATTERNS

25

4.2 ACTION AREA - ADDRESSING COMMUNITY UNDERSTANDING AND SOCIAL NORMS THROUGH MASS MEDIA

37

4.3 ACTION AREA - EXPOSURE TO MARKETING OF FOODS AND LIFESTYLES

44

4.4 ACTION AREA - IMPROVED PHYSICAL ACTIVITY, NUTRITION AND WEIGHT STATUS IN EVERYDAY SETTINGS

52

4.5 ACTION AREA - HEALTHY ACTIVE ENVIRONMENTS

66

4.6 ACTION AREA - FOOD SUPPLY

79 Page 2 of 170

4.7 ACTION AREA - FOOD ACCESS AND AVAILABILITY

87

4.8 ACTION AREA - FOOD PURCHASE AND CONSUMPTION

103

4.9 ACTION AREA - INTERVENING IN SOCIALLY DISADVANTAGED GROUPS (REMOTE, INDIGENOUS AND LOW INCOME GROUPS)

117

5.0 ASSESSMENT OF EVIDENCE REVIEWS

132

5.1 AN ASSESSMENT OF THE EVIDENCE WITHIN DEFINED ACTION AREAS

132

5.2 AN ASSESSMENT OF THE EVIDENCE ACROSS ALL ACTION AREAS

134

6.0 LESSONS FROM OTHER PUBLIC HEALTH INITIATIVES

144

6.1 TOBACCO, ALCOHOL AND OTHER PUBLIC HEALTH INITIATIVES

144

6.2 COMMUNITY CHRONIC DISEASE PREVENTION TRIALS

145

6.3 LESSONS AROUND THE USE OF REGULATION FOR BEHAVIOUR CHANGE IN PUBLIC HEALTH

146

6.4 LESSONS LEARNT ABOUT TIMEFRAMES

148

6.5 SUMMARY OF LESSONS LEARNT FROM OTHER PUBLIC HEALTH INITIATIVES

149

7.0 RECENT AUSTRALIAN INITIATIVES

151

7.1 NATIONAL PREVENTATIVE HEALTH TASKFORCE REPORT

151

7.2 ACE OBESITY AND ACE PREVENTION PROJECTS

152

7.3 PRODUCTIVITY COMMISSION REPORT. CHILDHOOD OBESITY: AN ECONOMIC PERSPECTIVE

153

8.0 INTERNATIONAL INITIATIVES TO SUPPORT THE DEVELOPMENT OF COMPREHENSIVE OBESITY PREVENTION INTERVENTIONS

155

8.1 INTERNATIONAL OBESITY TASKFORCE (IOTF) FRAMEWORKS FOR ACTION

155

8.2 EU PORGROW PROJECT

155

8.3. EU HOPE PROJECT

156

8.4 CDC'S RECOMMENDED STRATEGIES FOR OBESITY PREVENTION

157

8.5 INSTITUTE OF MEDICINE - BRIDGING THE EVIDENCE GAP

158

8.6 THE UK FORESIGHT REPORT AND PROJECT

158

9.0 DEVELOPING A COMPREHENSIVE OBESITY PREVENTION STRATEGY

161

Page 3 of 170

9.1 DEFINING APPROPRIATE OBJECTIVES

161

9.2 WHO SHOULD OBESITY PREVENTION STRATEGIES TARGET?

161

9.3 SELECTING A PORTFOLIO OF ACTIONS

163

10. CONCLUSIONS AND RECOMMENDATIONS

164

Page 4 of 170

Acronyms list ARIA ATSI BEACH BFHI BMI CHD COAG CVD DALYs EDNP FOP GI GWG HFSS INFANT IOTF NGO NHMRC NHS NICE NIP OECD PE QALY RCTs RFQ SES UK UNICEF USA WC WHO

Accessibility Remoteness Index of Australia Aboriginal and Torres Strait Islander Bettering the Evaluation and Care of Health (data set) Baby Friendly Hospital Initiative Body Mass Index Coronary Heart Disease Council of Australian Governments Cardiovascular Disease Disability adjusted life years Energy Dense Nutrient Poor Front of Pack Glycaemic Index Gestational weight gain High Fat, Salt, Sugar Infant Feeding Activity and Nutrition Trial International Obesity TaskForce Non-governmental organisation National Health and Medical Research Council National Health Survey National Institute for Health and Clinical Excellence Nutrition Information Panel Organisation for Economic Co-operation and Development Physical Education Quality Adjusted Life Year Randomised Control Trials Request for Quote Socio-economic Status United Kingdom United Nations Children’s Fund United States of America Waist circumference World Health Organisation

Page 5 of 170

Preface The problem of obesity remains one of the key public health issues in Australia, and it is now firmly on the agenda of governments, health policy makers and the community in general. It has now become a global issue; with very few countries escaping the dramatic increase in population mean body weight that Australia has experienced in recent times. Obesity is a serious medical condition associated with a wide range of debilitating, chronic and lifethreatening conditions. It imposes huge financial burdens on health care systems and the community at large. At an individual level it often results in psycho-social problems and a much reduced quality of life. As a consequence there has been increased pressure for firm and effective action to address the problem. The international published literature on obesity prevention has focussed on defining best options in terms of the level and quality of evidence of effectiveness for specific actions from controlled trials. Numerous systematic reviews to assess the efficacy of various approaches to obesity prevention have been commissioned and a range of new agencies have been established to collate and interpret this information. However, the ability of such evidence to define best practice is limited as different communities have varying cultural, economic, demographic and social characteristics that will influence the appropriate and feasible interventions. In addition, recent research together with an analysis of past public health action, has shown that a collection of disparate interventions is unlikely to be sufficient to control population weight gain and that comprehensive, integrated programs of actions will be required to deal with the complex array of factors which are driving the obesity epidemic. Australia was one of the first countries to produce an integrated national strategy for the prevention of overweight and obesity. The National Health and Medical Research Council (NHMRC) document Acting on Australia's Weight: A strategic plan for the prevention of overweight and obesity was released in 1997, yet most of its recommendations remain unexplored fifteen years later. A number of framework documents followed and the National Preventive Health Taskforce Report, National Preventative Health Strategy – the roadmap for action released in 2009 built upon this earlier work to produce a comprehensive set of recommendations to tackle obesity across a range of settings and domains. Despite the clarity and logic of these recommendations, many have not been adopted because of the perceived lack of evidence to support their potential impact. This report was commissioned by NHMRC to provide an overview of the range of evidence which is currently available to guide the development of the most appropriate, comprehensive program of action to address obesity in Australia. The Boden Institute at the University of Sydney was requested to produce a rapid overview of the current “state of knowledge” around comprehensive interventions to address obesity and provide an assessment of the implications for research and policy action on this issue in Australia. The process was completed within a brief three month period at the end of 2010 and the speed and breadth of the evidence assessment process required compromises in terms of the depth to which many issues could be explored. The Boden Institute had access to a wide range of literature reviews on obesity prevention that had been undertaken previously by health agencies in Australia and internationally, such as the Preventative Health Taskforce, ACE Obesity Group and Productivity Commission in Australia; the Foresight project and the National Institute for Clinical Excellence in the UK; and the Centers for Disease Control and Prevention in the USA. However, there have also been a number of national documents and guidelines whose release occurred after the completion of this report and thus their findings were not Page 6 of 170

considered within our assessment. These include the Blewett Report of the Review of Food Labelling Law and Policy, the revision of the NHMRC Australian Dietary Guidelines and the NHMRC Guidelines for the Management of Overweight and Obesity. In producing this report, the Boden Institute has not only drawn on existing systematic reviews and reports, but also undertaken a series of additional rapid reviews of the literature and examined evidence from outside scientific journals. Drawing together this evidence and making assessments of its merit has been challenging and required the adaption of a number of evidence assessment and summation systems. The report could not cover all possible formats and settings for obesity prevention interventions, but has adopted a framework which makes evident the breadth of action which will be required to effectively address obesity in the Australian community. It sets out a number of important findings about the limitations of current evidence assessment processes and the importance of integrating the widest range of available information to help inform policy making on obesity prevention.

Tim Gill Associate Professor and Principal Research Fellow Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders University of Sydney

Page 7 of 170

Executive Summary This report presents an overview of the evidence to support comprehensive interventions to address obesity at a population level and provides an assessment of the implications in terms of research and policy action. Obesity is a serious public health problem with over 61% of Australian adults and around 25% of Australian children being assessed as overweight or obese in the 2008 Australian Health Survey. This contributes to the significant rise in type 2 diabetes in Australia in the last decade, as well as being a major risk factor for cardiovascular disease and a number of cancers. Latest estimates put the cost of overweight and obesity at $21 billion. The drivers of obesity within the community are many and complex. A number of reports have attempted to document the wide range of behavioural, cognitive, biological and environmental factors which influence energy balance at an individual or population level. The Foresight System Map attempts to capture the complexity of these factors and the nature of their interaction which is currently driving energy accretion and disrupting individual efforts to achieve energy balance. Currently there is no region or country that has come to terms with this complex array of drivers of obesity and developed effective programs of actions to effectively address the problem of obesity. However, there is a growing body of evidence which indicates the types of interventions that will be required to be integrated into a comprehensive program of action to tackle the problem. No single intervention will be sufficient alone to allow the community to return to energy balance, but a portfolio of actions may make progressive contributions to a solution. We examined the evidence around interventions within nine key action areas in line with the domains of action from the Foresight System Map and the National Preventative Health Taskforce report - Australia: the healthiest country by 2020, National Preventative Health Strategy – the roadmap for action (NPHT, 2009). The nine action areas were: • Early life exposures and growth patterns • Addressing community understanding and social norms through mass media • Exposure to marketing of foods and lifestyles • Improved physical activity and nutrition in everyday life • Planning healthy active environments • Food supply • Food access and availability • Food purchase and consumption • Action in high risk groups Evidence was identified, collated, assessed and interpreted in a logical and organised manner utilising a system based on previous work by the Boden Institute and in line with recommendations from the US Institute of Medicine report – Bridging the Evidence Gap in Obesity Prevention (2010). We utilised a variety of systematic and intuitive search strategies to identify evidence from the published scientific literature as well as the government and NGO reports. In defining the evidence base, we tried to identify information from observational and experimental studies where suitable. However, we also included data obtained by extrapolation, deduction or logic, theoretical or accumulated understanding in line with previous evidence frameworks we have published. Page 8 of 170

The merit of different types of interventions within each action area was assessed using a composite measure based on the quality and quantity of evidence available around its effectiveness (which we defined as the “level of confidence in the evidence” together with an assessment of the potential of the intervention to contribute to achieving energy balance). This approach was based on our previous efforts to define the promise of an intervention as a product confidence in the evidence and the efficacy, reach and uptake of the intervention and is in line with the approach used in the recent World Cancer Research Fund policy report. Although the assessment of merit is based on explicit processes, there remains an element of subjectivity with this classification approach. It is not always possible to directly quantify these factors from the published literature but it is usually possible to grade interventions into high, medium and low to allow a calculation of potential impact on the achievement of energy balance. Within the scheme used within this report, a separate assessment of the potential to contribute to energy balance has been made for interventions that have a direct or an indirect effect on the achievement of energy balance. The level and quality of evidence available to assess interventions within each action area varied greatly, but it was usually sufficient to make a reasonable assessment of the relative merit of each intervention, both within and between different action areas. Reliance on traditional evidence hierarchies proved inadequate. In general, the level of well controlled experimental evidence was low providing a more important role for the breadth of evidence from smaller scale trials, observational, extrapolated and accumulated information. The value of repeated systematic reviews based on a limited number of well controlled but often poorly implemented trials is also called into question. Another major limitation of most evidence assessment schema is the inability to evaluate the combined impact of a portfolio of actions on the achievement of energy balance. This would seem to be the preferred mode of evaluation, given that many individual interventions are not designed to have a direct effect on weight status or weight-related behaviours, but rather are designed to interact with other interventions to enable or amplify behaviour change. The greatest volume of experimental evidence currently exists for interventions undertaken in everyday life settings such as school, worksites and day-care centres. However, no attempt has been made to integrate the wider body of information that could assist with judgements about the merit of interventions with this experimental data. Most reviews of this evidence show a positive impact of these types of intervention on the weight status of children and adults, with the greatest benefit shown for multi-component interventions. However, because the content and context of these interventions vary so much, and the outcomes are not always defined by the same measures, it is difficult to make clear assessments of the merit of individual approaches or to combine individual studies into systematic reviews or meta-analyses. As a consequence, many systematic reviews are unable to reach meaningful conclusions. Multi-component community-wide interventions is the area with the most consistent and promising evidence. Contrary to current perceptions, front of pack labelling and statutory restrictions on food marketing to children have the broadest base of evidence, with consistent findings to support their implementation. Menu labelling and changes to the physical environment to increase physical activity also has sufficient evidence to support a modest degree of promise. Evidence for action in these areas comes from small experimental studies, ecological assessments, observational studies, consumer research, parallel evidence and logic.

Page 9 of 170

The evidence around the merit of taxation and subsidies needs clarification. Most analyses have been conducted in the USA and although there is some observational data, most of the assessments rely on modelling of impact. Most studies demonstrate a small but potentially meaningful impact when taxes are imposed on particular products, and point to more merit in taxation of specific food products such as soft drinks. Taxation disincentives appear to have a stronger impact on behaviour than incentives through subsidies, although subsidies may need to be part of any fiscal strategy to offset equity concerns of taxation. We have no data on the price elasticity of demand around food products in Australia to allow effective modelling of taxation/subsidy impacts. Very few interventions within the action areas addressing the food supply, food access and availability and active environments were able to demonstrate a direct impact on weight status. Those studies that did were usually small scale and targeted at motivated groups. However, there is sufficient evidence that such interventions impact positively on weight related behaviours and offer additional support for the promise of these interventions from observational data, parallel evidence and program logic. Interventions to improve food and physical activity environments are important levers for addressing current inequalities in opportunities to eat well and be physically active. Interventions in these action areas have considerable synergies with actions to address climate change and other environmental concerns. There is value in identifying interventions to specifically address weight status within disadvantaged groups. Although such groups benefit from interventions which address the whole of the community, adaptations and specifically focussed implementation that addresses social or cultural complexities are often needed to ensure effectiveness with more disadvantaged groups. There have been very few, large-scale, comprehensive programs addressing the prevention of obesity that have been sustained for a long enough period to allow effective evaluation of the merit of the wide range of potential public health strategies that may be utilised. Therefore a number of commentators have sought lessons from successful actions to address other public health problems such as tobacco, alcohol and road safety, or from community-wide programs addressing other chronic diseases such as diabetes and heart disease. Whilst this has been able to enlighten discussions around potential approaches, care must be taken to interpret the nature of this parallel evidence and the context in which these actions occurred, to ensure that the learning is applied correctly to obesity prevention. Over the last two decades there has been a considerable amount of reflection and policy development around obesity prevention within Australia. National obesity prevention frameworks have been developed for children and for adults and older Australians. In addition, all jurisdictions have developed their own strategies for dealing with the problem of obesity and funded a range of major obesity prevention trials or initiatives. In 2009/10 there was three major policy or policy analysis projects: 1. The report from the National Preventative Health Taskforce; 2. The ACE Obesity and ACE Prevention reports; 3. The Productivity Commission report into childhood obesity.

These have provided an important basis for discussion around what interventions should be included in the comprehensive package of programs to address obesity. There has also been significant action at an international level which provides guidance on Page 10 of 170

comprehensive programs of action to address the drivers of obesity. Important reports from the International Obesity Taskforce, the European Union, the US Centre for Disease Control and Prevention and the Foresight Project, have all produced valuable analyses and guidelines on preventing obesity at the community-level. Constructing a comprehensive portfolio of actions to address overweight and obesity in Australia will require an ongoing process of evidence assessment and review. It is unlikely that there will ever be a definitive body of experimental evidence to guide all aspects of this process, but the collation, assessment and interpretation of all sources of evidence using an appropriate framework, such as that applied in this document, will ensure that decisions are made on the best available evidence.

Page 11 of 170

1.0 THE PROJECT BRIEF A consortium from the University of Sydney was contracted by the NHMRC Prevention and Community Health Committee to produce a rapid overview of the current “state of knowledge” around comprehensive interventions to address obesity and provide an assessment of the implications for research and policy action on this issue in Australia. The RFQ document indicated a number of specific issues on which the review and report should focus. These were further clarified and elaborated upon by discussions with NHMRC staff and meeting with the Prevention and Community Health Committee directly. It was agreed that the report should provide an assessment of multi-sectoral interventions to address obesity in Australia and internationally and the evidence base on their effectiveness. Specifically the report should focus on: • • •

An assessment of current Australian obesity prevention interventions and policies, and their evidence base An assessment of existing international policy interventions (and their relevance to Australia) An assessment of emerging or proposed population interventions/policies and their potential

The project was undertaken between September and December 2010. The rapid nature of this process and the expressed desire for broad overview rather than a comprehensive primary review of the literature mean that this report may not address all possible intervention actions reported in the literature. The breadth of this overview also limits the depth at which each possible intervention area has been explored.

2.0 TACKLING THE PROBLEM OF OBESITY IN AUSTRALIA 2.1 THE CURRENT PROBLEM OF OBESITY Obesity remains a serious health problem in Australia, and our inability to deliver a comprehensive program of action to address it effectively is likely to result in further escalation of the problem in the near future. Results from the 2007-08 National Health Survey (NHS) revealed that in 2007-08, 61.4% of the Australian population were either overweight or obese, with 42.1% of adult males and 30.9% of adult females classified as overweight, and 25.6% of males and 24% of females classified as obese (ABS, 2008). Aboriginal and Torres Strait Islander Australians are 1.9 times as likely as non-indigenous Australians to be obese. The same survey also found that 24.9% of children aged 5 – 17 years (25.8% of boys and 24% of girls) were either overweight or obese (ABS, 2008). In 2005, overweight and obese Australian adults cost the Australian economy $21 billion in direct health care and direct non-health care costs, plus an additional $35.6 billion in government subsidies (Colagiuri et al, 2010). The projected progression of the prevalence of obesity in Australia will mean that if the current rates of weight gain continue, then 42.4% of adults will be overweight and 33.9% will be obese by 2025, and only 28.1% will be an acceptable weight (Walls et al, 2010). 2.2 THE DRIVERS OF OBESITY A number of analyses have attempted to define the key determinants of obesity and there remains a degree of controversy over which factors have made the greatest contribution to the recent rise in the rates of obesity in Australia today. The most comprehensive Page 12 of 170

assessment of the situation has been undertaken by the World Health Organization in the Expert Report on Diet, Nutrition and the Prevention of Chronic Disease (WHO, 2003). This report examined the current literature and identified a range of key factors which either increase or decrease the risk of weight gain and the development of obesity (see Table 1). Table 1. Summary of the strengths of evidence on factors that might promote or protect against weight gain and obesity Evidence

Decreases risk

Increases risk

Convincing

Regular physical activity

High intake of energy-dense foods*

High dietary fibre intake

Sedentary lifestyles

Home and school environment that supports health food choices for children

Heavy marketing of energy dense foods and fast foods outlets

Probable

Promoting linear growth

Adverse social and economic conditions in developed countries (especially for women) Sugar-sweetened soft drinks and juices

Possible

Low glycaemic index foods

Large portion sizes

Breastfeeding

High proportion of food prepared outside of homes Rigid restraint/periodic disinhibition eating patterns

Insufficient

Increased eating frequency

Alcohol

* energy dense foods are high in fat/sugar and energy dilute foods are high in fibre and water such as vegetables, fruits, legumes and whole grain cereals Source: Adapted from WHO, 2003

Although the WHO review touches upon the array of behavioural and environmental influences in the aetiology of obesity at an individual and population level, it does not capture the complex interplay of these factors and how they operate at a societal level. One of the first attempts to represent the nature of the prevailing, multilayered environmental factors that influence energy balance in the modern world was the International Obesity Taskforce “Causal Web” (Kumanyika, 2001). The causal web illustrated that although food intake and energy expenditure ultimately influence energy balance, there is an array of forces that operate at the many layers of society which impact directly and indirectly upon these behaviours (Figure 1). The implications of this representation are apparent. Addressing obesity prevention will require action at many levels, and must include a focus on many of the distal factors that influence our food and activity environment.

Page 13 of 170

Figure 1 The IOTF Causal web

Source: Kumanyika, 2001

2.3 THE FORESIGHT OBESITY SYSTEM MAP Although the causal web suggests that the genesis, and thus the solution to obesity is complicated, its linear format does not clearly illustrate the “complexity” of the interactions between the various layers. The Foresight Programme of the UK Government Office for Science expanded on the “causal web” approach by utilising a systems approach to produce a complex conceptual model with 108 variables known as the “obesity systems map” (Vandenbroeck et al, 2007). The relationships between the variables are illustrated with more than 300 solid or dashed lines to indicate positive and negative influences. All the variables are interconnected and these connections give rise to feedback loops. At the core (or engine) of the map is energy balance surrounded by variables that directly or indirectly influence this key process. The foresight map approach to defining the broad-ranging drivers of obesity and the interrelationships between these factors has not met with universal approval. It has been criticised for being overly-complex to the extent that it creates a sense of confusion and despair when clarity is what is needed to address this problem. Others have also questioned how comprehensive such a map could be in that it implies that all the potential drivers of obesity are captured within the map (Finegood et al, 2010). It is probably true that very few people completely understand all the intricacies of the system it describes and that the map is not truly complete as it only reflects the perceptions of the stakeholders engaged in its development. However, it has served some useful objectives in improving our perception of the nature of the obesity problem and the approach that will be required to successfully address it. Page 14 of 170

Some of the principles that the Foresight map and process have reinforced are: • • • • •

The wide range of political, social, environmental, behavioural and physiological factors that influence individuals and societies ability to achieve energy balance and the complex multifactorial nature of the systems that give rise to obesity. The breadth of action that will be required to restore energy balance. The futility of attempting to address obesity by focussing attention just on individual behaviour change or within one domain of action. The need to consider the interaction between factors that either enable or amplify or conversely inhibit the behaviour change process required to achieve energy balance. The interaction between factors within the system is currently driving energy accretion and disrupting individual efforts to achieve energy balance.

The inescapable conclusion of such an exercise is that obesity is the result of a complex system of interrelated factors and that success in addressing the problem will only come by a program of action that attempts to address factors from across the whole of the system. A “whole of system” approach requires action: • • • • • •

Involving multiple sectors Engaging multiple agencies Includes multiple strategies Across multiple levels of jurisdiction Targeting multiple groups At multiple stages of life

2.4 DOMAINS OF ACTION A whole of system approach does not imply that all interventions must attempt to span all the elements of a system but rather that programs of action should consider the potential contribution that different elements of that system can make to the overall (system) achievement of energy balance. To help better identify appropriate intervention or leverage points within the system, the Foresight map has been broken into seven key sub-systems or themes (Figure 2). These thematic clusters allow a simplification of the process of identifying appropriate interventions to address the drivers of obesity whilst still retaining the importance of the “whole of system” approach. As a result of undertaking this system mapping, the Foresight Project was able to identify the four key domains of action to address obesity, which are listed below and set out in Figure 2. 1. Human biology - Genetics plays a part but biology is also shaped by early life experiences and the wider environment. 2. The food environment - There has also been a huge increase in the quantity of foods of high energy density. 3. The physical environment - Our lives have become increasingly sedentary, manual occupancies are in decline and most journeys are by car. 4. Culture - It is difficult to break unhealthy, but engrained, eating and activity patterns, especially when common to those around us.

Page 15 of 170

Figure 2 Domains of action within the Foresight obesity system map

Source: Vandenbroeck et al, 2007

2.5 DEFINED ACTION AREAS In order to better group the wide range of interventions which focus on influencing drivers of obesity within the key domains of the Foresight obesity systems map, we devised a set of eight key action areas. A further action area was added to bring these groupings more in line with those set out within the National Preventative Health Taskforce report - Australia: the healthiest country by 2020, National Preventative Health Strategy – the roadmap for action (NPHT, 2009). The nine action areas were: • Early life exposures and growth patterns • Addressing community understanding and social norms through mass media • Exposure to marketing of foods and lifestyles • Improved physical activity and nutrition in everyday life • Planning healthy active environments • Food supply • Food access and availability • Food purchase and consumption • Action in high risk groups

2.6 DIFFERENT TYPES OF INTERVENTIONS It is widely accepted that acting to address one single element within the obesity system highlighted by the Foresight Project or restricting action to one action area defined within the National Preventative Health Strategy is unlikely to result in a significant impact on community weight status. What is needed is a comprehensive program of action across all domains of action. Page 16 of 170

Likewise no single type of intervention is likely to be suitable to all situations or achieve a significant impact on its own. Interventions to encourage behaviour change within communities and individuals needs to be accompanied by a range of policy, structural and environmental actions which will facilitate and support behaviour change. The Foresight Project defines such actions as “enablers” or “amplifiers” (see Figure 3). Alone, such policies have little or no direct impact on obesity but may magnify the impact of other initiatives. Enablers are actions that provide the underpinning or capacity of individuals or communities to respond to behaviour change programs. Other actions or policies may amplify impact, changing norms or removing barriers, and providing incentives to change in the right direction. Current examples include the restrictions on food marketing to children and financial incentives to use active transport. Figure 3 Different types of interventions to tackle obesity

Source: Vandenbroeck et al, 2007

2.7 EXPECTED OUTCOMES OF INTERVENTIONS It is often assumed that intervening within the community to address the problems of excess weight should result in a reduction in the levels of overweight and obesity. However, Australia is currently experiencing a period of very rapid increases in the mean population BMI, resulting from a large energy surplus. Reversal of this trend will require a substantial reduction in this energy surplus, which will need to be maintained for a significant period of time. Few interventions are capable of reducing energy intake or increasing energy expenditure sufficiently, or for long enough, to achieve this effect. More appropriate outcomes would be Page 17 of 170

the prevention of weight gain or weight stability in adults, and the achievement of appropriate growth and development in children. These outcomes will only be achieved through the maintenance or re-establishment of energy balance. However, even population weight stability may be difficult to achieve in the short term, as a large proportion of Australian adults are in a state of positive energy balance, gaining weight over time. Therefore, it may be necessary to identify more sensitive short and medium-term outcomes to evaluate programs of actions to address obesity. The conceptual framework for monitoring weight and related variables presented in Figure 4 (Gill et al, 2005) can serve as a useful guide for measuring ‘lower order outcomes’ desired by preventive programs. Changes in dietary and physical activity behaviours have been shown to precede changes in weight status in adults and children, and can be detected within a timeframe of one to two years. If changes in diet, physical activity or sedentary behaviours are large enough to impact positively on improved energy balance, then these may serve as useful intermediary outcomes. Also, important ‘process evaluation’ indicators such as measured changes in policies, services, professional practices and community facilities, can be used to show that preventive programs are being implemented as planned, which is an important first step in the lead up to dietary and physical activity changes in the population. Figure 4

Source: Gill et al, 2005

Page 18 of 170

3.0 THE PROCESSES IN GENERATING THIS REPORT In line with the objectives of this report, we have attempted to identify, collate, assess and interpret information about comprehensive obesity prevention actions from a variety of sources, in a logical and organised manner. Although our processes were as methodical as possible, the relatively short timeframe has meant that the identification and collation of information has involved a mixture of systematic and ad hoc searching strategies. The assessment and interpretation of this information has elements of subjectivity but has been made against explicit and relevant criteria. 3.1 LITERATURE SEARCH STRATEGY This report is based on literature reviews, reports and policy documents produced or published between 1980 and September 2010. A series of search strategies were defined to try to capture all recent reviews evaluating interventions addressing obesity–related behaviours as well as structural and environmental changes. a. Selective literature review with a focus on: i. reviews of reviews ii. systematic reviews iii. recent publications on significant obesity prevention initiatives The first phase of this scan involved a full systematic search of the literature using defined search terms (Medical Subject Headings (MeSH) terms overweight, obesity, obese, intervention with a range of intervention types and settings) and publication selection processes. The following electronic databases were searched: Pubmed, Medline, Embase, and PsychInfo. This search yielded a significant number of returns for search terms relating to behavioural interventions of those carried out in an everyday life setting, but was less productive in identifying reviews or recent reports of single studies evaluating the impact of a structural or environment intervention on weight status or behaviour change. b. Harvest evidence from health policy and evidence assessment agencies. This strategy proved particularly rewarding and returned a number of evidence summaries and policy briefs in addition to existing systematic reviews of the literature. Sites that we utilised included: Australia: PANORG, CO-OPS, NHMRC, NPHT United Kingdom: NICE, HTA, FSA, Scottish Government Europe: WHO Europe, EU HOPE Project United States: IOM, CDC, Robert Wood Johnston Canada: Health-Evidence-Ca. International: OECD c. Scan of existing national and regional obesity prevention strategies and policies This approach yielded a list of proposed and evaluated strategies plus links to additional evidence assessments undertaken to support the policy development. d. Hand searching of quoted papers and reports. 3.2 CATEGORIES OF EVIDENCE In this project we have tried to produce an evidence base in relation to information from observational and experimental studies where suitable. However, because this approach can be extremely limited and does not necessarily provide an appropriate basis for identifying the overall impact of complex community interventions, other forms of evidence have also been incorporated into the assessments of effectiveness and feasibility. Thus, the analysis of intervention research has also considered evidence as proposed under our previous work (Gill et al, 2005) and that of the IOTF Framework (Swinburn et al, 2005). Page 19 of 170

These include: a. Observational i. Observational epidemiology Epidemiological studies that do not involve interventions, but may involve comparisons of exposed and non-exposed individuals, e.g. cross-sectional, case-control, or cohort studies. ii. Monitoring and surveillance Population-level data that are collected on a regular basis to provide time series information, e.g. mortality and morbidity rates, food supply data, car and TV ownership, birth weights and infant anthropometry. b. Experimental i. Experimental studies Intervention studies where the investigator has control over the allocations and/or timings of interventions e.g. randomised controlled trials, or non-randomised trials in individuals, settings, or whole communities. ii. Programme/policy evaluation Assessment of whether a programme or policy meets both its overall aims (outcome) and specific objectives (impacts) and how the inputs and implementation experiences resulted in those changes (process). c. Extrapolated i. Effectiveness analyses Modelled estimates of the likely effectiveness of an intervention that incorporate data or estimates of the programme efficacy, programme uptake, and (for population effectiveness) population reach. ii. Economic analyses Modelled estimates that incorporate costs (and benefits), e.g. intervention costs, costeffectiveness, or cost-utility. iii. Indirect (or assumed) evidence Information that strongly suggests that the evidence exists, e.g. a high and continued investment in food marketing is indirect evidence that there is positive (but proprietary) evidence that the food marketing increases the sales of those products and/or product categories within that target audience. d. Experience i. Parallel evidence Evidence of intervention effectiveness for another public health issue using similar strategies, e.g. the role of social marketing or policies or curriculum programmes or financial factors on changing health-related behaviours such as smoking, speeding, sun exposure, or dietary intake. It also includes evidence about the effectiveness of multiple strategies to influence behaviours in a sustainable way, e.g. health-promoting schools approach, comprehensive tobacco control programmes, or co-ordinated road toll reduction campaigns. ii. Theory and programme logic The rationale and described pathways of effect based on theory and experience, e.g. linking changes in policy to changes in behaviours and energy balance, or ascribing higher levels of certainty of effect with policy strategies like regulation and pricing compared with other strategies such as education. iii. Informed opinion The considered opinion of experts in a particular field, e.g. scientists able to peer review and interpret the scientific literature, or practitioners, stakeholders, and policy-makers able to inform judgements on implementation issues and modelling assumptions (incorporates ‘expert’ and ‘lay knowledge’).

Page 20 of 170

3.3 SUMMATION OF EVIDENCE We developed a template that allowed the integration of all the relevant information that would assist with our assessment of the merit of each intervention within the action area. The template had to be broad enough to allow the integration of information from a wide variety of sources into the evidence summary but specific enough to identify the basis of its inclusion. Although this template was developed before the release of the US Institute of Medicine report “Bridging the evidence gap in obesity prevention” (IOM, 2010) was released, we were pleased to find a great deal of similarity between this template and the one they propose for collating evidence around specific interventions. 3.4 ASSESSMENT OF THE MERIT OF IDENTIFIED INTERVENTIONS WITHIN EACH ACTION AREA The collection, collation and summary of all the relevant evidence around the effectiveness of an intervention within the evidence summaries, makes the process of assessing the merit of each intervention simpler. As discussed in section 3.2, traditional hierarchies of evidence are less relevant in defining the relative merit of specific interventions to prevent or manage obesity. Whilst studies with controlled experimental designs have strong internal validity, they are often not available and provide only one piece of the jigsaw of evidence that is available to more completely define the potential impact of an intervention. That is why we chose to utilise a composite measure based on the quality and quantity of evidence available around its effectiveness (which we defined as the “level of confidence in the evidence” together with an assessment of the potential of the intervention to contribute to achieving energy balance. This approach was based on our previous efforts to define the promise of an intervention as a product confidence in the evidence and the efficacy, reach and uptake of the intervention (Gill et al, 2005). This format is similar to the one used within the recent World Cancer Research Fund policy report (WCRF, 2009). The ability to contribute to the achievement of energy balance was seen as the most appropriate outcome to assess. This allows an assessment of potential change based on evidence of effectiveness in either weight or behaviour change and also accepts that some interventions act by enabling or reinforcing the changes achieved by other interventions. Although the assessment of merit is based on explicit processes, there remains an element of subjectivity with this classification approach. Table 2. Format for the assessment of the merit of interventions assessed Intervention

e.g. Social marketing

Level of confidence in the evidence

med

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect low

med

Page 21 of 170

3.4.1 Grading the level of confidence in the evidence The level of confidence in the evidence is a product of three separate characteristics of the available evidence around each intervention: i. quality of evidence – is there controlled experimental or high quality observational data ii. quantity of evidence – does the evidence base consist of multiple studies examining the same type of intervention iii. breadth of evidence – how many different forms of evidence are available that provide broad and consistent support for the merit of the intervention. 3.4.2 Grading the contribution of different interventions to the effective prevention and management of obesity As indicated in section 2.6, the ultimate objective of any intervention which aims to prevent weight gain is to contribute to the attainment of energy balance. In addition, successful weight loss will require the achievement and maintenance of negative energy balance for a prolonged period of time. It is unlikely that a single intervention or action (in isolation) will be sufficient to ensure a sustained return to energy balance or the creation of negative energy balance. It is more likely that several interventions in a comprehensive program of action will be required to achieve this objective. In addition, enabling or amplifying actions may not directly impact on energy balance but are more likely to contribute indirectly by enhancing or facilitating the impact of other interventions. Thus we have attempted to grade the different interventions assessed within this report in relation to their potential to make a direct or indirect contribution to the achievement of energy balance as part of an overall program of action. This grading is based on three components reported within (or extrapolated from) the literature we reviewed. These were: i. Efficacy - the size of effect (measured or modelled) ii. Reach and Uptake - the proportion of the community who would receive and be willing and able to participate in this intervention iii. Maintenance – the extent to which the program is sustained over time It is not always possible to directly quantify these factors from the published literature, but it is usually possible to grade interventions into high, medium and low to allow a calculation of potential impact on the achievement of energy balance. Within the scheme used within this report, a separate assessment of the potential to contribute to energy balance has been made for interventions that have a direct or an indirect effect on the achievement of energy balance. 3.5 ADDITIONAL CONSIDERATIONS AROUND THE MERIT OF AN INTERVENTION There are other issues (apart from evidence of effectiveness or potential) that may also inform an assessment of the merit of an intervention to achieve the desired change within the local Australian social and political environment. The ACE project refers to these as second stage filters (Carter et al, 2009) and they give an indication of the appropriateness of specific interventions within the prevailing environment as well as their potential ease of intervention locally. The application of these filters is context-dependent and thus making universal assessments of their implications for the merit of a particular intervention is inappropriate. Page 22 of 170

In our evidence summaries, we chose to consider the impact of these additional factors. a. feasibility and acceptability to all stakeholders b. relevance to Australia (transferability) c. cost considerations d. timeframes in implementation and impact e. equity issues f. other positive or negative side-effects

Page 23 of 170

References Australian Bureau of Statistics (ABS). National Health Survey 2007-08. Canberra: 2008. Cat. No. 4364.0. Carter R, Moodie M, Markwick A, Magnus A, Vos T, Swinburn B, Haby MM. Assessing costeffectiveness in obesity (ACE-obesity): an overview of the ACE approach, economic methods and cost results. BMC Public Health 2009; 9:419. Colagiuri S, Lee CMY, Colagiuri R, Magliano D, Shaw JE, Zimmet PZ, Caterson ID. The cost of overweight and obesity in Australia. Medical Journal of Australia 2010; 192(5):260-264. Economos C, Brownson RC, DeAngelis MS, Novelli P, Foerster SB, Foreman CT, Gregson J, Kumanyika SK, Pate RR. What lessons have been learned from other attempts to guide social change? Nutrition Reviews 2001; 59(3 Pt 2):S40-56. Finegood DT, Merth TD, Rutter H. Implications of the foresight obesity system map for solutions to childhood obesity. Obesity. 2010; 18 (Suppl 1): S13-16 Gill T, King L, Webb K. Best options for promoting healthy weight and preventing weight gain in NSW. Sydney: NSW Centre for Public Health Nutrition; 2005. Institute of Medicine (IOM). Bridging the evidence gap in obesity prevention: a framework to inform decision making. Washington DC: The National Academies Press; 2010. Kumanyika SK. Minisymposium on obesity: overview and some strategic considerations. Annual Review of Public Health 2001; 22:293–308. National Preventative Health Taskforce (NPHT). Australia: the healthiest country by 2020. National preventative health strategy – the roadmap for action. Canberra: Commonwealth of Australia; 2009. Swinburn B, Gill T, Kumanyika S. Obesity prevention: a proposed framework for translating evidence into action. Obesity Reviews 2005; 6:23–33. Vandenbroeck IP, Goossens J, Clemens M. Tackling obesities: future choices—building the obesity system map [internet]. Government Office for Science, UK Government’s Foresight Programme; 2007. Available from: http://www.foresight.gov.uk/Obesity/12.pdf Walls HL, Magliano DJ, Stevenson C, Backholer K, Manna HR, Shaw J, Peeters A. Project progression of the prevalence of obesity in Australia. Obesity Research and Clinical Practice 2010; 4(Suppl 1):S55. World Cancer Research Fund (WCRF). Policy and action for cancer prevention. Food, nutrition, and physical activity: a global perspective. Washington DC: AICR, WCRF; 2009. World Health Organization (WHO). Expert report on diet, nutrition and the prevention of chronic diseases: report of a joint WHO/FAO expert consultation. Geneva: WHO; 2003. Technical report series 916.

Page 24 of 170

4.0 BRIEF REVIEW OF CURRENT EVIDENCE The following section contains a detailed assessment of the merit of interventions within each of the following action areas. The evidence was collated, extracted and assessed in line with the methodology set out in section 3. 4.1 Early life exposures and growth patterns 4.2 Addressing community understanding and social norms through mass media 4.3 Exposure to marketing of foods and lifestyles 4.4 Improved physical activity and nutrition in everyday life 4.5 Planning healthy active environments 4.6 Food supply 4.7 Food access and availability 4.8 Food purchase and consumption 4.9 Action in high risk groups

Page 25 of 170

4.1 ACTION AREA - EARLY LIFE EXPOSURES AND GROWTH PATTERNS 1. Rationale for this action area • Recent research reveals that the early life period, comprising pre-conception, in utero, infancy and early childhood, is a critical period for the development of obesity. A range of biological and behavioural factors occur during this period that influence the adiposity of the child or imprint diet and physical activity behaviours which persist throughout childhood. • Although changes in our genetic makeup take generations to occur, it is possible to induce permanent changes in early life, in the way genes function (‘epigenetic’ changes) from one generation to the next. For example, there is accumulating evidence that diet before conception may influence the way genes are expressed in reproductive material and the resultant foetus. In addition, exposures such as high glucose, high fat and/or high insulin during foetal life, could be programming individuals towards obesity, even from before birth (Singhal et al, 2003). • Overweight parents have heavier babies who are more at risk of childhood obesity. In addition, mothers who have gestational diabetes have infants who are at much greater risk of obesity and type 2 diabetes later in life (Keith et al, 2006). • There is evidence from a population cohort study that weight loss pre pregnancy reduces the risk of gestational diabetes (Glazer et al, 2004). • Breastfeeding has been linked to a small decreased risk of paediatric overweight in multiple epidemiologic studies. Despite this evidence, many mothers never initiate breastfeeding and others discontinue breastfeeding earlier than needed (Armstrong & Reilly, 2002). There is some evidence that infants given formula may be less likely to consume vegetables and fruit and more likely to consume commercial infant drinks compared with infants who were breastfed (Noble & Emmett, 2006). • Parental eating habits and feeding practices also have a profound effect on a child’s early diet and risk of excessive weight gain (Fisher & Birch, 1995). There is evidence that food preferences are strongly set in the first few years of life and thus the type of foods that infants are weaned onto may be an important element in the development of obesity. • Rapid weight gain (regardless of birth-weight) in the first 12 months of life has also been linked to greater levels of excessive weight gain in childhood (Goodell et al, 2009). 2. Proposed actions to improve early diet and growth Good maternal and infant nutrition and a focus on appropriate growth have always been a focus of child health, and recent research has led to an understanding of the profound effects of under-nutrition as well as over-nutrition on later obesity and health. As a consequence, there has been little opportunity to evaluate the long term impact of interventions which seek to control maternal and infant weight gain and improve feeding practices in early life. Current evidence development has focussed on a few key issues.

Page 26 of 170

Table 3. Proposed actions to improve early diet and growth Proposed actions

Pre-conception care

Type of intervention Targeted Enabling/ behavioural amplifying 

Information sources

Improved maternal diet



Promotion of extended breastfeeding



Improving early feeding and physical activity practices Preventing early rapid weight gain



Evidence summaries, single intervention studies Systematic reviews, Cochrane reviews, guidelines, evidence summaries Numerous systematic reviews, Cochrane reviews, evidence summaries Single studies



Single studies



3. Summary of the evidence There are limited evaluations of interventions in this area as this is a relatively new area and changes may take some time to achieve. Many of the past interventions aimed at encouraging weight gain during pregnancy, ensuring high birth weight, maximising early growth and weight gain in infancy may not be appropriate in the current environment of excessive weight gain and childhood obesity in Australia. a. Pre-conception lifestyle intervention i. Basis Addressing mothers’ obesity prior to conception is likely to avoid a number of pregnancy complications and help reduce the risk of childhood obesity. However, obesity and unplanned pregnancy are associated with poorer socio-economic status and this group has poor compliance with simple preconception measures such as folic acid supplementation. ii. Evidence from trials We have been unable to identify any systematic reviews, evidence assessments or well evaluated intervention trials in this area. One Australian based case series (Galletly et al, 1996) found that obese women trying to become pregnant but experiencing infertility can achieve a statistically significant reduction in BMI through a programme that includes regular physical activity, advice about healthy eating and group support. iii. Other evidence There is some evidence of benefit from preconception lifestyle counselling and care in women with pre-gestational diabetes (Wilhoite et al, 1993). Prospective observational studies indicate that infertile, obese women who lose weight have improved spontaneous ovulation, spontaneous pregnancy rates, response to fertility therapy, and reduced miscarriage rates (Clark et al, 1995). iv. Comments There is significant potential for well-designed intervention studies of preconception care programs aimed at addressing obesity. The LIFESTYLE study in the Netherlands is examining the costs and effects of a structured lifestyle program in overweight and obese Page 27 of 170

sub-fertile women, to reduce the need for fertility treatment and improve reproductive outcomes (Mutsaerts et al, 2010). b. Improving pregnancy and maternal diet i. Basis Maternal obesity and excessive gestational weight gain are major risk factors for maternal and foetal complications and poorer health outcomes including obesity. Children born to mothers who are over-nourished or under-nourished prior to or during pregnancy are more likely to be obese during early childhood (Olstad & McCargar, 2009). Gestational weight gain (GWG) is associated with childhood obesity (Schack-Nielsen, Michaelsen et al, 2010). It is suggested that excessive GWG induces a persisting susceptibility to obesogenic environments. ii. Evidence from trials Recent systematic reviews have examined the effect of lifestyle interventions in limiting excessive weight gain during pregnancy (Ronnberg & Nilsson, 2010) or ante-natal weight control (Dodd et al, 2010) and found no consistent evidence that these interventions could lead to improved weight and health outcomes. Guelinckx et al (2008) found that two out of seven intervention trials focusing on GWG, nutrition and physical activity, reached a significant decrease in GWG. A Cochrane review of aerobic exercise interventions in pregnant women concluded that the studies were too small and methodologically flawed to make any firm conclusions, but did appear to improve or maintain fitness (Kramer & McDonald, 2006). Another Cochrane review of maternal diet found no clear benefits for high fibre diets, but there was a suggestion that low GI diets led to better health outcomes for mother and child (Tieu et al, 2008). Conversely none of these reviews found any adverse effects on the weight or health of mother or child resulting from interventions to address weight. A systematic review of dietary and/or physical activity interventions for weight management in pregnancy was conducted by Campbell et al (2008) as part of the National Institute for Health and Clinical Excellence (NICE). Five randomised control trials (RCTs), five nonrandomised trials, two case studies, fourteen observational studies and two systematic reviews were included in the review. Findings were mixed, but there is potential for effectiveness, particularly among those that are overweight or obese pre-pregnancy. No evidence on what intervention components were most effective was identified. iii. Other evidence Emerging evidence suggests that well planned lifestyle interventions can be effective in achieving improved nutrition (Guelinckx et al, 2010) and reduced prenatal weight gain (Shirazian et al, 2010). Experimental and limited clinical data suggest that the dietary ratio of n-6 to n-3 fatty acids during pregnancy is critical for early adipose tissue growth (Hauner et al, 2009).

iv. Comments Page 28 of 170

Developing guidelines to promote appropriate weight gain and healthy lifestyle in overweight and obese pregnant women remains a challenge (Guelinckx et al, 2010); and probably any programs should be more individualised (Artal et al, 2010). c. Promoting breastfeeding exclusivity and duration i. Basis There is relatively consistent evidence from longitudinal studies that breastfeeding exposure and duration are associated with small but significant reductions in the development of childhood obesity. The mechanisms by which this may happen are unknown and may relate to the establishment of better dietary self-regulatory mechanisms within the child. Evidence of the effect of breastfeeding on maternal body composition is mixed. A recent study (Hatsu et al, 2008) provided evidence that exclusive breastfeeding promotes greater weight loss in the mother than mixed feeding, even in the early postpartum period; and these authors suggest there is a need to encourage mothers to exclusively breastfeed as a means of overweight and obesity prevention. ii. Evidence from trials The evidence of interventions to promote and support breastfeeding has been reviewed extensively in two major overviews (Hector & King, 2005; Hector et al, 2010). In addition to the multiple systematic reviews contained within these overviews, the NICE report ‘Promotion of breastfeeding initiation and duration: Evidence into Practice briefing’ (Dyson et al, 2006) is a primary resource. The evidence of effectiveness is concentrated in three main areas: hospital practice; health professional education; and support of pregnant women and mothers. There is strong evidence for the effectiveness of the Baby Friendly Hospital Initiative (BFHI) as a whole and for many of the individual ‘Ten Steps’ that comprise the BFHI. The presence of a written breastfeeding policy and communication of the policy to staff, early skin-to-skin contact and not giving supplementary fluids (infant formula or glucose water) while in hospital unless medically indicated, appear to be particularly important practices in improving breastfeeding duration. There is insufficient evidence around what works best in terms of health professional education. The WHO/UNICEF training package is effective, but unlikely to be sufficient. Qualitative evidence stresses the need for education of staff providing community support services. Support in any form has been identified as a core component of programs to ensure good breastfeeding outcomes; the evidence is very strong. There is good evidence that a mixture of professional and peer support is likely to be most effective, particularly support spanning all periods, i.e. antenatal, perinatal and postnatal. The evidence for the provision of professional support alone is mixed, and is stronger for postnatal than antenatal support. There is overwhelmingly strong evidence that peer counselling is effective in improving all aspects of breastfeeding. Scaling-up experience has suggested that it must not be offered as a stand-alone intervention, and benefits from being part of an existing health professional program or initiative. Also, a single session of informal, small group and discursive breastfeeding education should be delivered in the antenatal period (including topics like the prevention of nipple pain and trauma) to improve breastfeeding outcomes in lower socioeconomic status women. Attendance at parent groups where peers are breastfeeding infants of a similar age is Page 29 of 170

effective at improving breastfeeding. Several studies have shown that targeted, peer counselling and social support, also combined with professional support, is particularly important for younger mothers. iii. Other evidence There remains a dearth of evidence in the action areas of policy initiatives, supportive environments and community action. No media campaigns were identified. There is scant evidence of the effectiveness of workplace lactation programs, although a recent study found action in this area to be effective across a large public-sector employee in the US (Balkam et al, 2011). iv. Comments In Australia the sub-groups most at risk of short duration of breastfeeding and/or not exclusively breastfeeding are younger mothers (aged less than 25 years) and those without a tertiary education. Other groups at risk are primiparous mothers and multiple births. Also, obese mothers, mothers who had a caesarean birth, mothers of pre-term and low birth weight infants, and women who smoke are at increased risk of poorer breastfeeding practices. No single intervention is likely to increase breastfeeding rates at the population level. As such, effectiveness of single intervention studies might be better assessed in terms of their effect on any of the specific enablers or barriers to breastfeeding, rather than rates of breastfeeding alone. d. Improving early infant diet and feeding practices i. Basis The early introduction of complementary (solid) foods is associated with un-healthful subsequent feeding behaviours (Grummer-Strawn et al, 2008), although it is still unclear whether this association is cause or effect (Kramer, 2010). Evidence from an Australian prospective cohort study of infants at risk of atopic conditions showed that the timing of introduction of solids significantly increased the risk of overweight and obesity (Seach et al, 2010). ii. Evidence from trials Despite the strong rationale to focus on parents’ early feeding practices as a key determinant of child food preferences, intake and self-regulatory capacity, prospective longitudinal and intervention studies are rare (Daniels et al, 2009). A recent pilot study in the US involved two nurse home visits (Paul et al, 2011). The first intervention (“soothe/sleep”) instructed parents on discriminating between hunger and other source of infant distress; soothing strategies were taught to minimise feeding for nonhunger-related fussiness and prolong sleep duration, particularly at night. The second intervention taught parents about hunger and satiety cues, the timing of introduction of solid foods and how to overcome infants’ initial rejection of health foods through repeated exposure. At one year, infants who received both interventions had a mean weight-for-length in the 33rd percentile compared with higher percentiles for those receiving only one intervention or none.

Page 30 of 170

iii. Other evidence A recent study in the UK indicated that parents are receptive to health education prior to weaning and need better support with best practice in infant feeding (Redsell et al, 2010). In particular, evidence indicates that this should focus on helping them understand the physiology of breastfeeding, how to differentiate between infant distress caused by hunger and other causes, and the timing of weaning. Some parents also require guidance about how to recognise and prepare healthy foods and facilitate physical activity for their infants. Several research trials are also underway in Australia addressing these issues. • The NOURISH RCT in Brisbane and Adelaide will provide anticipatory guidance via two modules of six fortnightly parent education and peer support group sessions, each followed by six months of regular maintenance contact. • Similarly, the INFANT (Infant Feeding Activity and Nutrition Trial) is an early childhood intervention conducted with parents over the infant’s first 4-18 months of life in Victoria (Campbell et al, 2008). This intervention also uses anticipatory guidance around particular times related to infant feeding. Early findings are encouraging. • Finally, the Healthy Beginnings Trial, in South-west Sydney, is examining the effect of a home visiting intervention involving eight home visits over two years and pro-active telephone support between visits over the first two years of life. The EMPOWER (Empowering Parents to Prevent Obesity at Weaning: Exploratory Research) programme showed that parents valued the emphasis on the listening, partnership working and shared problem-solving approach taken by the health visitor (Barlow et al, 2010). Parents identified a number of benefits including increased knowledge about the most appropriate types and amounts of food to feed their toddler, to more farreaching changes about the family as a whole, including modifications to their own diet and lifestyle. Programs of this type were perceived as more valuable than the standard help that is currently available. iv. Comments Interestingly, the most recent evidence suggests home visits by a nurse or health visitor using empathetic listening, anticipatory guidance and shared or supported problem-solving, are likely to be effective components to any approach to improve breastfeeding, weaning and early infant feeding outcomes. Similarly, a recent pilot suggests that training of health and community practitioners should focus on them being able to work more sensitively with parents of babies and pre-school children around obesity and lifestyle concerns (Rudolf et al, 2010). e. Preventing rapid weight gain in infancy i. Basis A systematic review by Ong and Loos (2006) identified 21 separate studies with data on the association between rapid infancy weight gain and subsequent obesity risk; and found that all studies reported significant positive associations. Singhal and Lucas (2004) postulate that it is not birth weight per se, but accelerated weight change from birth (catch-up growth), that pre-programs a higher risk of metabolic abnormalities later in life. ii. Evidence from trials Whilst there have been many reported interventions to improve the nutrition in low birth weight and undernourished infants, we can identify no reviews of reported trials which have Page 31 of 170

studied interventions aimed at reducing later overweight and obesity. iii. Other evidence Recently Singhal (2010) indicated that data from two prospective RCTs support a causal link between faster early weight gain and a later risk of obesity. This has important implications for the management of infants born small for gestational age. iv. Comments The link between rapid early weight gain and obesity risk has implications for the management of infants born small for gestational age, given the current focus on promoting weight gain in infant health services. Observational data from Chile indicates that programs established to help feed underweight or low birth weight infants and promote catch up growth may have contributed to the development of excessive weight gain and obesity issues within this population. Butte (2009) considered the evidence and concluded that infancy weight gain far supersedes infant feeding practices (breastfeeding) as a risk factor for childhood obesity. 4. Other initiatives proposed or implemented but not evaluated There is enormous potential to positively influence infant and later childhood weight status through intervening in early life. However, as this is a relatively new area of research and well- designed studies will take some time to reveal results, there are a range of additional suggestions that have been made. These include: • Childcare facilities’ policies to promote breastfeeding • Recommendations for restrictions on bottle feeding after weaning to reduce energy intake and growth rates. 5. Quality and nature of evidence to support action in early life The area of early life nutrition and activity and later obesity is relatively new, and thus there have been few evaluated interventions which have focused on the prevention of obesity. Thus very little of the evidence relates directly to its impact on energy balance. Instead, interventions have been assessed in terms of their impact on behaviours of even the existence of policies or practices to promote improved behaviours. Thus, there is a dearth of experimental evidence and evaluated interventions from almost all potential actions with the exception of promotion of breastfeeding. There have been some systematic reviews and Cochrane reviews examining the promotion of breastfeeding, but not in other areas. While the evidence on the effects of breastfeeding promotion on weight status is limited and relies heavily on observational data, there is strong evidence on the effectiveness of strategies to promote breastfeeding per se. There is limited evidence for approaches to managing mothers’ weight gain during pregnancy and fostering optimal weight prior to pregnancy, and scope to trial innovative approaches, given the potential benefits for the infant and the mother. The relationships between early life and later obesity still remain to be elucidated and thus many of the proposed interventions in this area are speculative; and trials begun recently will take many years to provide results in terms of prevention of childhood obesity. There is some support from animal experimental studies and strong logic and theoretical support for interventions during pre-conception, pregnancy and early life, but little firm indication of what might be the most effective approaches. Page 32 of 170

Table 4. Quality and nature of evidence to support action in early life Intervention

Level of confidence in the evidence

Pre-conception care Improved maternal diet Promotion of extended breastfeeding Improving early feeding and physical activity practices Preventing early rapid weight gain

low med high

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect med med-high low-med

low

med

low

med

6. Potential impact and sustainability This is a critical area that must be addressed for the long term success of obesity prevention strategies. However, the current interventions have not been well developed or implemented and thus their short –term impact has been small, and it may be many generations before returns are evident. The poor reach and uptake of pre-conception and maternal diet and lifestyle care programs greatly limits their potential at present; however it may be that optimal strategies have not yet been trialled. Likewise programs to prevent early rapid weight gain have great potential to reduce obesity risk, but remain relatively unexplored. Greater potential returns have been promised by pilot programs addressing early feeding and physical activity programs, but these will need to be expanded to improve their reach and uptake. Improved breastfeeding rates can make a very small impact of weight status, but because they address every mother and child, they could contribute to a reduction in childhood obesity and should be included for their other benefits on child health generally. All these targeted behaviour change programs required a constant input of resources to keep them functioning, and attempts must be made to integrate them into usual health or social care programs. 7. Other considerations a. Feasibility and acceptability to all stakeholders There is little to indicate that any of these interventions would generate resistance from different stakeholders. However, pre-conception programs face issues in recruitment of participants as many pregnancies are unplanned and these programs may not be perceived as a priority for those attempting to conceive. b. Relevance to Australia (transferability) Although many of the evaluated programs in this action area originate in the USA, the common approach to maternal and early childhood care means they are generally relevant to the Australian context. c. Cost considerations Although these behaviour change programs do not require many inputs, they are often time intensive, with one instructor servicing only a small number of individuals. Expanding on pilot programs may require considerable investment, unless programs are integrated within existing services Page 33 of 170

d. Timeframes in implementation and impact Interventions addressing in-utero and early exposures of infants are unlikely to bring immediate returns, but may improve weight status in late childhood or early adulthood. Conversely they may actually take a few generations to have any measurable benefit. e. Equity issues Low income and disadvantaged groups are likely to benefit most from such programs, but they are also the sector least likely to be attracted or recruited. Attempts need to be made to ensure that recruitment favours these groups. f.

Other positive or negative side-effects

Even without impacting on weight status, early interventions may help improve nutrition, growth and development outcomes in children and operate at a time when mothers need support and are receptive to changing behaviours.

Page 34 of 170

References Anderson AK, McDougald DM, Steiner-Asiedu M. Dietary trans fatty acid intake and maternal and infant adiposity. European Journal of Clinical Nutrition 2010; 64(11):1308-1315 Armstrong J, Reilly JJ. Breastfeeding and lowering the risk of childhood obesity. The Lancet 2002; 359:2003–2004. Artal R, Lockwood CJ, Brown HL. Weight gain recommendations in pregnancy and the obesity epidemic. Obstetrics and Gynecology 2010; 115(1):152-155 Balkam JA, Cadwell K, Fein SB. Effect of components of a workplace lactation program on breastfeeding duration among employees of a public-sector employer. Maternal and Child Health Journal 2011; 15(5):677-683 Barlow J, Whitlock S, Hanson S, Davis H, Hunt C, Kirkpatrick S, Rudolf M. Preventing obesity at weaning: parental views about the EMPOWER programme. Child: Care, health and Development 2010; 36(6):843-849 Bergmann KE, Bergmann RL, Von Kries R, Böhm O, Richter R, Dudenhausen JW, Wahn U. Early determinants of childhood overweight and adiposity in a birth cohort study: role of breastfeeding. International Journal of Obesity 2003; 27:162–172. Birch LL, Ventura AK. Preventing childhood obesity: what works? International Journal of Obesity 2009; 33:S74-S81 Bonuck KA, Huang V, Fletcher J. Inappropriate bottle use: an early risk for overweight? Literature review and pilot data for a bottle-weaning trial. Maternal and Child Nutrition 2010; 6:38-52 Butte NF. Impact of infant feeding practices on childhood obesity. Journal of Nutrition 2009; 139(2):412S-6S Campbell K, Hesketh K, Crawford D, Salmon J, Ball K, McCallum Z. The Infant Feeding Activity and Nutrition Trial (INFANT): an early intervention to prevent childhood obesity: cluster-randomised controlled trial. BMC Public Health 2008; 8:103 Clark AM, Ledger W, Galletly C, Tomlinson L, Blaney F, Wang X, Norman RJ. Weight loss results in significant improvement in pregnancy and ovulation rates in anovulatory obese women. Human Reproduction 1995; 10:2705–2712 Daniels LA, Magarey A, Battistutta D, Nicholson J, Farrell A, Davidson G, Cleghorn G. The NOURISH randomised control trial: positive feeding practices and food preferences in early childhood – a primary prevention program for childhood obesity. BMC Public Health 2009; 9:387 Dodd JM, Crowther CA, Robinson JS. Dietary and lifestyle interventions to limit weight gain during pregnancy for obese or overweight women: a systematic review. Acta Obstetricia Gynecologica Scandinavica 2008; 87(7):702-706 Dodd JM, Grivell RM, Crowther CA, Robinson JS. Antenatal interventions for overweight or obese pregnant women: a systematic review of randomised trials. BJOG: An international journal of obstetrics and gynaecology 2010; 117:1316–26. Dyson L, Renfrew M, McFadden A, McCormick F, Herbert G, Thomas J. Promotion of Page 35 of 170

breastfeeding initiation and duration: evidence into practice briefing. National Institute for Health and Clinical Excellence (NICE); 2006. Fisher JO, Birch LL. Fat preferences and fat consumption of 3-to 5-year-old children are related to parental adiposity. Journal of the American Dietetic Association 1995; 95:759-764. Galletly C, Clark A, Tomlinson L, Blaney F. Improved pregnancy rates for obese, infertile women following a group treatment program. An open pilot study. General Hospital Psychiatry 1996; 18(3):192-195. Glazer NL, Hendrickson AF, Schellenbaum GD, Mueller BA. Weight change and the risk of gestational diabetes in obese women. Epidemiology 2004; 15(6):733-737. Goodell LS, Wakefield DB, Ferris AM. Rapid weight gain during the first year of life predicts obesity in 2–3 year olds from a low-income, minority population. Journal of Community Health 2009; 34(5):370-375 Grummer-Strawn LM, Scanlon KS, Fein SB. Infant feeding and feeding transitions during the first year of life. Pediatrics 2008; 122 (Supp2):S36-42 Guelinckx I, Devlieger R, Beckers K, Vansant G. Maternal obesity: pregnancy complications, gestational weight gain and nutrition. Obesity Reviews 2008; 9(2):140-150 Guelinckx I, Devlieger R, Mullie P, Vansant G. Effect of lifestyle intervention on dietary habits, physical activity, and gestational weight gain in obese pregnant women: a randomised controlled trial. American Journal of Clinical Nutrition 2010; 91(2):373-380 Hart CN, Raynor HA, Jelalian E, Drotar D. The association of maternal food intake and infants’ and toddlers’ food intake. Child: Care, Health and Development 2010; 36(3):396-403 Hatsu IE, McDougald DM, Anderson AK. Effect of infant feeding on maternal body composition. International Breastfeeding Journal 2008; 3:18 Hauner H, Vollhardt C, Schneider KT, Zimmermann A, Schuster T, Amann-Gassner U. The impact of nutritional fatty acids during pregnancy and lactation on early human adipose tissue development. Rationale and design of the INFAT study. Annals of Nutrition and Metabolism 2009; 54(2):97-103 Hector D, Hebden L, Innes-Hughes C, King L. Update of the evidence base to support the review of the NSW Health Breastfeeding Policy: a rapid appraisal. Sydney: PANORG; 2010 Hector D, King L. Interventions to encourage and support breastfeeding. NSW Public Health Bulletin 2005; 16(3-4):56-61 Hector D, King L, Webb K. Overview of interventions to promote and support breastfeeding. NSW Centre for Public Health Nutrition and NSW Department of Health; 2004 Keith SW, Redden DT, Katzmarzyk PT, Boggiano MM, Hanlon EC, Benca RM, Ruden D, Pietrobelli A, Barger JL, Fontaine KR, Wang C, Aronne LJ, Wright SM, Baskin M, Dhurandhar NV, Lijoi MC, Grilo CM, DeLuca M, Westfall AO, Allison DB. Putative contributors to the secular increase in obesity: exploring the roads less traveled. International Journal of Obesity 2006; 30(11):1585-1594 Kramer MS, McDonald SW. Aerobic exercise for women during pregnancy. Cochrane Database of Systematic Reviews 2006; Issue 3. Art. No.: CD000180. DOI: Page 36 of 170

10.1002/14651858.CD000180.pub2 Kramer MS. Breastfeeding, complementary (solid) foods, and long-term risk of obesity. American Journal of Clinical Nutrition 2010; 91(3):500-501. Mutsaerts MA, Groen H, ter Bogt NC, Bolster JH, Land JA, Bemelmans WJ, Kuchenbecker WK, Hompes PG, Macklon NS, Stolk RP, van der Veen F, Maas JW, Klijn NF, Kaaijk EM, Oosterhuis GJ, Bouckaert PX, Schierbeek JM, van Kasteren YM, Nap AW, Broekmans FJ, Brinkhuis EA, Koks CA, Burggraaff JM, Blankhart AS, Perguin DA, Gerards MH, Mulder RJ, Gondrie ET, Mol BW, Hoek A. The LIFESTYLE study: costs and effects of a structured lifestyle program in overweight and obese subfertile women to reduce the need for fertility treatment and improve reproductive outcome. A randomised controlled trial. BMC Women’s Health 2010; 19:22 Noble S, Emmett P. Differences in weaning practice, food and nutrient intake between breast- and formula fed 4-month-old infants in England. Journal of Human Nutrition and Dietetics 2006; 19:303–313. Olstad Dl, McCargar L. Prevention of overweight and obesity in children under the age of 6 years. Applied Physiology, Nutrition and Metabolism 2009; 34:551-570 Ong KK, Loos RJ. Rapid infancy weight gain and subsequent obesity: systematic reviews and hopeful suggestions. Acta Paediatrica 2006; 95(8):904-908 Paul IM, Savage JS, Anzman SL, Beiler JS, Marini ME, Stokes JL, Birch LL. Preventing obesity during infancy: a pilot study. Obesity 2011; 19(2):353-361 Redsell SA, Atkinson P, Nathan D, Siriwardena AN, Swift JA, Glazebrook C. Parents’ beliefs about appropriate infant size, growth and feeding behaviour: implications for the prevention of childhood obesity. BMC Public Health 2010; 10(1):711 Ronnberg AK, Nilsson K. Interventions during pregnancy to reduce excessive gestational weight gain: a systematic review assessing current clinical evidence using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system. BJOG: An International Journal of Obstetrics and Gynaecology 2010; 117:1327–34. Rudolf MCJ, Hunt C, George J, Hajibagheri K, Blair MH. Development, pilot and long-term evaluation of a programme to help practitioners work more effectively with parents of babies and pre-school children to prevent childhood obesity. Child: Care, Health and Development 2010; 36(6):850-857 Schack-Nielsen L, Michaelsen KF, Gamborg M, Mortensen EL, Sørensen TI. Gestational weight gain in relation to offspring body mass index and obesity from infancy through adulthood. International Journal of Obesity 2010; 43(1):67-74 Schack-Nielsen L, Sørensen TIA, Mortensen EL, Michaelsen KF. Late introduction of complementary feeding, rather than duration of breastfeeding, may protect against adult overweight. American Journal of Clinical Nutrition 2010; 91:619-627 Seach KA, Dharmage SC, Lowe AJ, Dixon JB. Delayed introduction of solid feeding reduces child overweight and obesity at 10 years. International Journal of Obesity 2010; 34(10):14751479. Shirazian T, Monteith S, Friedman F, Rebarber A. Lifestyle modification program decreases pregnancy weight gain in obese women. American Journal of Perinatology 2010; 27(5):411Page 37 of 170

414 Singhal A. Does Weight Gain in Infancy Influence the Later Risk of Obesity? Journal of Pediatric Gastroenterology & Nutrition: 2010; 51(Suppl 3): S119–S120 Singhal A, Lucas A. Early origins of cardiovascular disease: is there a unifying hypothesis? Lancet 2004; 363(9421):1642-1645. Singhal, A, Wells J, Cole TJ, Fewtrell T, Lucas A. Programming of lean body mass: a link between birth weight, obesity and cardiovascular disease. American Journal of Clinical Nutrition 2003; 77:726-730. Thangaratinam S, Jolly K. Obesity in pregnancy: a review of reviews on the effectiveness of interventions. Brit Journal of Obstetrics and Gynaecology 2010; 117:1309–1312. Tieu J, Crowther CA, Middleton P. Dietary advice in pregnancy for preventing gestational diabetes mellitus. Cochrane Database of Systematic Reviews 2008; Issue 2. Art. No.: CD006674. DOI: 10.1002/14651858.CD006674.pub2 von Kries R, Koletzko B, Sauerwald T, von Mutius E, Barnert D, Grunert V, von Voss H. Breast feeding and obesity: cross sectional study. British Medical Journal 1999; 319:147– 150. Wen LM, Baur LA, Rissel C, Wardle K, Alperstein G, Simpson JM. Early intervention of multiple home visits to prevent childhood obesity in a disadvantaged population: a homebased randomised controlled trial (Healthy Beginnings Trial). BMC Public Health 2007; 7:76 Willhoite MB, Bennert HW Jr, Palomaki GE, Zaremba MM, Herman WH, Williams JR, Spear NH.The impact of preconception counselling on pregnancy outcomes. The experience of the Maine Diabetes in Pregnancy Program. Diabetes Care 1993; 16:450-455 Worobey J, Lopez MI, Hoffman DJ. Maternal behaviour and infant weight gain in the first year. Journal of Nutrition Education andBehavior2009; 41(3):169-175

Page 38 of 170

4.2 ACTION AREA - ADDRESSING COMMUNITY UNDERSTANDING AND SOCIAL NORMS THROUGH MASS MEDIA

1. Rationale for this action area • Providing appropriate advice and information to the population is an important first step in addressing any health concern, but knowledge and awareness are rarely enough by themselves to bring about behaviour change. • Current behaviour and willingness to change are also greatly influenced by personal attitudes, beliefs and habits as well as community social norms around undesirability of overweight, perceived health risks of weight, and appropriate diet and physical activity behaviours. • Social norms can have a strong influence over the way communities and individuals react to messages and their willingness to change behaviours. • Addressing any identified knowledge gaps and the prevailing social norms are an important element to allow people to consider behaviour change and addressing weight issues in an appropriate manner. • There are many information distribution and media approaches to effective communication around health issues. • Social marketing goes beyond communication or information provision. It is a process of steps that identifies what will motivate citizens to change behaviour, and makes them ‘offers’ approaches that will encourage change. The objective of social marketing is to change individuals’ behaviour to achieve a socially desirable goal (Donovan & Henley, 2003). • Many types of media are used for social marketing purposes including broadcast, print, electronic media and the internet (Marcus et al, 1998). 2. Proposed actions to address community understanding and social norms Although there are a variety of potential communication and marketing approaches to address any information gap and help address social norms around obesity, proposed strategies have focussed on mass media information campaigns and social marketing as the preferred intervention approaches. Table 5. Proposed actions to address community understanding and social norms Proposed actions Mass media and social marketing campaigns

Type of intervention Targeted behavioural 

Enabling/ amplifying 

Information sources Systematic reviews, evidence summaries, single evaluation studies. Parallel evidence from other behaviours e.g. smoking, alcohol

Page 39 of 170

3. Summary of the evidence Although there has been considerable investment of resources in mass media campaigns to address diet, physical activity and obesity prevention issues, they are often not evaluated well and thus we have limited assessments of their effectiveness. a. Mass media and social marketing campaigns – obesity i. Basis While it is well known that provision of information alone cannot induce behaviour change, individuals clearly need information on how to modify their behaviour and what comprises desirable behaviour if they are to be expected to engage in behaviour change. For this reason, mass media campaigns have primarily been aimed at raising awareness, providing knowledge and changing attitudes, with the aim of contributing to potential behaviour change (Noar, 2006). ii. Evidence from trials Many countries have reported conducting obesity-related mass media campaigns. A recent review identified around 50 such programs in the USA, Australia, the UK and Canada in the last two decades (Cismaru & Lavack, 2007). Campaigns included a wide variety of different components, including websites and print materials such as posters, brochures, handouts, colouring books, calendars, and handbooks, as well as radio and TV ads. In addition, some campaigns incorporate community and mass-media events, awards, rewards kits, and items such as wristbands, pedometers, step counters, and T-shirts. However, only a few studies have evaluated the impact of these campaigns in terms of their impact on weight status or weight-related behaviour change within the community. Most have evaluated mass media campaigns in terms of knowledge, awareness or attitudes shift, and others are simply process evaluations. Early community heart disease prevention trials included mass media as a major component of their strategy, but whilst these reported modest improvements in diet and exercise over 35 years, there was no significant improvement in weight (Meyer et al, 1980). The BBC media campaign “Fighting Fat- Fighting Fit” ran for a short period on radio and TV and was backed up by a set of complementary activities and resources. A total of 58% of participants completed the evaluation five months after the program, which demonstrated significant improvements in self-reported weight (Miles et al, 2001). It also showed that men and young people were difficult to engage through this strategy and may need targeted programs. iii.Other evidence In 2008, the Council of Australian Governments (COAG) launched Phase 1 of a national social marketing campaign on healthy lifestyles to contribute to a reduction in the burden of chronic disease among Australians. The ‘Measure Up’ campaign used mass media supported by other campaign channels. A key objective of the campaign was to raise awareness of waist circumference (WC) as an indicator for risk of chronic disease. In Australia, the national and NSW evaluations of the (stage 1) Measure Up campaign indicated that there was substantial awareness of the campaign and campaign messages, increased knowledge and increased proportion measuring or intending to measure waist circumference, but at this stage, no changes in people’s other weight-related behaviours (Grunseit et al, 2010).

Page 40 of 170

A recent Dutch mass media campaign which used strong warnings on the consequences of weight gain found they were able to achieve small but significant positive differences in attitudes, perceived social support, and motivation and behavioural action to prevent weight gain (Wammes et al, 2007). However, the campaign had an adverse effect on risk perception, with less people perceiving weight as a health issue of relevance to them after the campaign. There is more information on the effectiveness of mass media campaigns addressing weight-related behaviours such as diet and physical activity. - Mass media nutrition campaigns There is mixed evidence of the effectiveness of mass media campaigns targeting healthy eating. One review found that some mass media campaigns have been more successful than others due to differences in the intensity, reach, timing, and cost of the intervention (Brownson et al, 2006). Another review found that the use of mass media is effective in improving awareness, knowledge, and intentions related to healthy eating behaviour, but not as effective in changing actual behaviour (Swinburn et al, 2004). However, the success of food marketing strategies in general and the broad reach of mass media campaigns make them a promising avenue for improving healthy eating outcomes at the community level. - Mass Media Physical activity campaigns There is evidence to demonstrate that community-wide campaigns are effective in increasing physical activity, intentions to be physically active, and knowledge about exercise (Kahn et al, 2002). However, there is limited documentation on the impact of community-wide physical activity campaigns on overall trends in obesity/overweight (CDC Taskforce, 2005). iv. Comments The evaluation of most mass media campaigns occur immediately following or within a year of the campaign. This is usually too early to see any contribution made to change in community weight status. Most importantly, however, it is inappropriate to assess the effectiveness of mass media campaigns in terms of their impact on weight or other behaviour change outcomes, as their objective is to provide knowledge and shape behaviour which will allow and facilitate behaviour change. Thus they are “enabling and amplifying levers”. Mass media strategies for increasing healthy eating have only been evaluated in general or adult populations. Evidence is lacking as to the effectiveness of this strategy with children or youth. There is better evidence in relation to physical activity mass media campaigns in different groups.

4. Other initiatives proposed or implemented but not evaluated It has been suggested that social marketing approaches and mass media be applied to addressing a range of community knowledge gaps, attitudes and social norms including: • Campaigns to address social norms around weight as a health problem • Campaigns to address social bias against the obese

Page 41 of 170

5. Quality and nature of evidence to support mass media action The evidence assessing the impact of mass media campaigns on community weight status outcomes is restricted to a few social marketing campaigns. The quality of information from these evaluations is severely limited. Most mass media campaigns focus on the provision of knowledge and attempt to shape attitudes and social norms. As such, evaluating their success on the basis of their impact on weight status directly is inappropriate, particularly in the short term. Extrapolated evidence from other sources such as the impact of mass media campaigns in other public health campaigns, program logic and modelling may be a more valuable guide to their contribution to a comprehensive, whole of system approach to obesity prevention. Their effectiveness is best judged in relation to the overall medium-term impact of a program of actions on population weight status. Table 6. Quality and nature of evidence to support mass media action Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action Direct

Mass media and social marketing campaigns

med

low

Indirect med

6. Potential impact and sustainability While it is well known that provision of information alone cannot induce behaviour change, individuals clearly need information on how to modify their behaviour and what comprises desirable behaviour, if they are to be expected to engage in behaviour change. For this reason, mass media campaigns have primarily been aimed at raising awareness, providing knowledge and changing attitudes, with the aim of contributing to potential behaviour change (Noar, 2006). Mass media campaigns reach wide proportions of the population, but their uptake is often more restricted. The attitudinal and knowledge changes that are achieved are most likely to be sustained where there is a series of successive waves of media campaigns, such as has occurred in youth smoking, skin cancer prevention and fruit and vegetable campaigns. It has been suggested that social marketing programs can go beyond information and awareness-raising and influence upstream factors such as the attitudes and practices of organisations, professionals, retailers, or policy makers. This can be seen as part of their effect on community norms, and may in turn influence the acceptability of other program or policy changes. However, due to difficulties in measuring policy and environmental change, meaningful measurement of such outcomes cannot be achieved (Gordon et al, 2006). Mass media campaigns reach almost the whole of the population but their uptake is often restricted to those who respond to the message. This may limit their impact to an extent. For sustained benefit they need to be repeated and linked to a wider range of strategies.

7. Other considerations a. Feasibility and acceptability to all stakeholders Mass media campaigns are generally well accepted by most stakeholders in the community. They are greatly favoured by governments because of their visibility and the potential to Page 42 of 170

implement them relatively quickly. b. Relevance to Australia (transferability) Most of the evidence noted is directly relevant to Australia as it originates in Australia or from OECD countries with similar media systems to Australia. c. Cost considerations Mass media programs can be expensive to implement, especially where they involve television advertisements. Campaigns need to be sustained for a significant period of time and repeated to build and sustain an effect. However, they are relatively cheap when considering cost per person reached. Despite this cost, two assessments of the potential cost effectiveness of mass media campaigns addressing obesity rate it as a cost effective strategy. A study from the UK estimated the cost of such a program would be around 75 UK Pounds per person (100k in total) but that it could result in a cost saving of over 3000 pounds per Quality Adjusted Life Year (QALY) gained (Matrix Insight, 2009). The Australian ACE Prevention Program estimated the cost of a mass media program to address inactivity at around $16 Million but also rated it highly cost effective or cost saving (Cobiac et al, 2009). d. Timeframes in implementation and impact Mass media campaigns are relatively quick to organise and implement but because they only contribute to behaviour change by facilitating attitudinal or knowledge shifts, there is considerable lag between programs and weight outcomes. e. Equity issues Mass media campaigns reach everyone and often have high reach to social disadvantaged or high risk groups who watch more television. The evidence suggests that mass media campaigns can contribute to success in behaviour change with many different types of audiences, including low income and working class sectors. f. Other positive or negative side-effects Concern has been expressed about the negative influence that social marketing around weight and its health risks might have on the weight control behaviours of those susceptible to social pressures around body weight such as teenagers, and on the level of bias directed to those with a weight problem. These are issues that warrant concern but thus far there is no indication that social marketing is likely to exacerbate these problems.

Page 43 of 170

References Brownson RC, Haire-Joshu D, Luke DA. Shaping the context of health: a review of environmental and policy approaches in the prevention of chronic diseases. Annual Review of Public Health 2006; 27:341-370. CDC Task Force on Community Preventive Services. Chapter 2 in: Zaza S, Briss PA, Harris KW, editors. The Guide to Community Preventive Services. What Works to Promote Health? New York. Oxford University Press; 2005 Cismaru M, Lavack AM. Social marketing campaigns aimed at preventing and controlling obesity: a review and recommendations. International Review on Public and Non Profit Marketing 2007; 4(1/2):9-30. Cobiac LJ, Vos T, Barendregt JJ. Cost-effectiveness of interventions to promote physical activity: a modelling study. PLoS Medicine 2009; 6(7):e1000110. Donovan R, Henley N. Social marketing: principles and practice. Melbourne. IP Communications; 2003 Gordon R, McDermott L, Stead M, Angus K. The effectiveness of social marketing interventions for health improvement: what's the evidence? Public Health 2006; 120(12):1133-1139. Grunseit AC, King EL, Bauman A. Promoting waist circumference as an indicator of chronic disease: how did measure-up measure up? Obesity Research & Clinical Practice 2010; 4(Suppl1):S69 Kahn EB, Ramsey LT, Brownson RC, Heath GW, Howze EH, Powell KE, Stone EJ, Rajab MW, Corso P. The effectiveness of interventions to increase physical activity. A systematic review. American Journal of Preventive Medicine 2002; 22(4S):73-107 Marcus BH, Owen N, Forsyth LH, Cavill NA, Fridinger F. Physical activity interventions using mass media, print media, and information technology. American Journal of Preventive Medicine 1998; 15(4):362-378. Matrix Insight. Prioritising investments in preventative health. London: 2009. Matson-Koffman DM, Brownstein JN, Neiner JA, Greaney ML. A site-specific literature review of policy and environmental interventions that promote physical activity and nutrition for cardiovascular health: What works? American Journal of Health Promotion 2005; 19(3):167-193 Meyer AJ, Nash JD, McAlister AL, Maccoby N, Farquhar JW. Skills training in a cardiovascular health education campaign. Journal of Consulting and Clinical Psychology 1980; 48:129-142. Miles A, Rapoport L, Wardle J, Afuape T, Duman M. Using the mass-media to target obesity: an analysis of the characteristics and reported behaviour change of participants in the BBC's 'Fighting Fat, Fighting Fit' campaign. Health Education Research 2001; 16(3):357-372. Noar SM. A 10-year retrospective of research in health mass media campaigns: where do we go from here? Journal of Health Communication 2006; 11(1):21-42 Swinburn BA, Caterson I, Seidell JC, James WPT. Diet, nutrition and the prevention of Page 44 of 170

excess weight gain and obesity. Public Health Nutrition 2004; 7(1A):123-146. Taylor CB, Fortmann SP, Flora J, Kayman S, Barrett DC, Jatulis D, Farquhar JW. Effect of long-term community health education on body mass index. The Stanford Five-City Project. American Journal of Epidemiology 1991; 134(3):235-49 Wammes B, Oenema A, Brug J. The evaluation of a mass media campaign aimed at weight gain prevention among young Dutch adults. Obesity 2007; 15 (11):2780-2789

Page 45 of 170

4.3 ACTION AREA - EXPOSURE TO MARKETING OF FOODS AND LIFESTYLES 1. Rationale for this action area • Numerous systematic reviews have concluded that the marketing of unhealthy (or energydense nutrient-poor) foods and beverages to children negatively influences children’s eating behaviour, dietary intake and beliefs and purchase requests to their parents (Cairns et al, 2009; Hastings et al, 2003; IOM, 2006).

• Much of this literature has focused on the effects of marketing energy-dense nutrient poor (EDNP) foods on television, given that this is the main media through which children are exposed. • Despite this, it must be recognised that EDNP food marketing is evident in multiple media platforms including print, the Internet, sports sponsorship and outdoors (Kelly & Chapman, 2007; Kelly, Bochynska et al, 2008; Kelly et al, 2010). Evidence suggests advertisers have been expanding their food marketing to these other media sources for years (IOM, 2006). This, in part, may be due to increasing pressures to reduce children’s exposure from television. • International and Australian studies consistently show that most food advertising is for EDNP foods. The marketing of EDNP foods typically includes persuasive and appealing features, including cartoon and celebrity characters, jingles and themes of fun and family, thus ‘normalising’ the consumption of these foods. • Furthermore, the EDNP foods and beverages advertised to children are not consistent with national dietary guidelines and food selection guides for children. This contributes to a consumer information environment that has conflicting and confusing nutrition information, and compromises the integrity and potential efficacy of any healthy eating social marketing campaigns and consumer education messages.

2. Quality and nature of evidence to support action in early life Table 7.Proposed actions to reduce children’s exposure to marketing Proposed actions

Self-regulation of food marketing to children (Advertising or food industry led initiatives). Statutory policy to restrict food marketing to children.

Type of intervention Targeted Enabling/ behavioural amplifying 





Marketing of healthy behaviours

Consumer training in media literacy

Information sources

Reviews of the regulatory environment internationally. Independent studies evaluating selfregulatory initiatives, Reviews of the regulatory environment internationally. Impact evaluation from the UK independent regulator Ofcom. Phased approach proposed by National Preventative Health Taskforce on extent and nature of advertising for energy dense – nutrient poor foods. Some evidence of use by fruit and vegetable industry groups. Some evidence for a positive effect of marketing healthy foods.



Page 46 of 170

3. Summary of the evidence Evidence surveying regulatory environments internationally has found two key responses to addressing food marketing to children: industry led self-regulation or statutory regulation (Hawkes & Lobstein, 2010).

a. Self-regulation of food marketing i. Basis The food and advertising industries have proposed that they should act as the key regulators of food marketing to children and thus avoid the need for statutory regulation. ii. Evidence from trials Scientific studies from Australia and internationally have found self-regulation to be ineffective or inadequate in protecting children from high exposure to marketing of EDNP foods and beverages (Hebden et al, 2010a; King et al, 2010; Romero-Fernandez et al, 2009; Kunkel et al, 2009). This is due to a number of reasons: • Initiatives are voluntary for food manufacturers and services, and thus are not adopted by many companies. • Definitions used in regulatory commitments are ambiguous and lenient, and hence allow ongoing advertising of EDNP foods and variable benchmarks for evaluating compliance and impact. • Nutrient criteria used to assess the appropriateness of foods and beverages for marketing to children are extremely lenient compared with existing professional nutrient criteria (Hebden et al, 2010b). • The restrictions do not apply during times when the highest numbers of children are watching television. • Poor company compliance with their own initiatives in some cases (Romero-Fernandez et al, 2009).

iii. Other evidence Assessments of food companies’ adherence promised changes to a range of practices to improve nutrition and promote health (in addition to marketing) have shown a high proportion of companies fail to meet the commitment given on this issue (Ludwig & Nestle, 2008).

iv. Comments The Commonwealth recommended in their response to the National Preventative Health Taskforce Roadmap for Action report that government continues to monitor the impact of self-regulation before further action. The scientific evidence indicates that self-regulation is not working as a policy option in Australia. This provides a strong basis for government to take the lead on regulating food marketing to children.

Page 47 of 170

b. Statutory policy to restrict food marketing i. Basis Statutory regulation provides a promising policy option to regulate food marketing to children for the following reasons: • It has been effective in the UK (see ‘Evidence from trials’ below). • It can be credibly and legitimately based on the goals of protecting the interests of children, with no conflicts of interest. • It is more likely to provide clear objectives and a consistent and rigorous set of benchmarks for restrictions, with the potential to effectively reduce children’s exposure to marketing of EDNP foods. Evidence from trials In July 2010, the Office of Communications in the United Kingdom released their final review of statutory restrictions that were phased in between 2006 and January 2009. These restrictions banned all television advertising for foods and drinks high in fat, salt or sugar (HFSS) during ‘children’s airtime’. HFSS products were defined using a nutrient profiling tool developed by researchers from the British Heart Foundation for the Food Standards Authority (Arambepola et al, 2007; Jenkin et al, 2008). The review found that in 2009, no HFSS advertisements were broadcast during ‘children’s airtime’, illustrating advertisers’ compliance with this statutory policy. This resulted in an overall decrease in children’s exposure to HFSS advertisements of 37%, compared with 2005. (NB: children’s exposure was calculated as the number of HFSS advertisements multiplied by the number of children watching television when the advertisements were shown (Ofcom, 2010). The review also found that children also saw less advertising that used promotions and licensed characters or company owned characters (such as Ronald McDonald or Coco the Monkey) in 2009, compared with 2005. ii. Other evidence As of 2009, six countries had developed statutory policies regulating food marketing to children, with a further two countries in the process of developing their policies. In May 2010, the World Health Organization member states endorsed a set of recommendations on the marketing of foods and non-alcoholic beverages to children (resolution WHA63.14) (WHO, 2010). These recommendations state that governments should set clear definitions for policies to reduce the impact on children of marketing of foods high in saturated fats, transfatty acids, free sugars or salt. These recommendations propose that governments take leadership and act as the key stakeholders on this issue.

c. Marketing of healthy behaviours i. Basis Marketing healthy behaviours can involve advertising for healthy foods or promotion of messages which oppose unhealthy behaviours (such as soft drink consumption). Such advertising is usually the domain of governments but is sometimes used by fruit and vegetable growers and farming groups. Examples include the Baby Carrots "Eat Em' Like Junk Food" promotion, the Australian Banana Growers “No Nos and Na Nas” advertisements.

Page 48 of 170

ii. Evidence from trials There is no evidence from trials to date on the use of oppositional marketing of healthy behaviours as a strategy to address the marketing of energy-dense, nutrient poor (EDNP) foods to children. iii. Other evidence Researchers have found that when children are exposed to advertisements for healthy foods this may increase their liking and positive attitudes towards these foods (Dixon et al, 2007). However, evidence suggests that the volume of advertising for healthy foods, even with government-funded campaigns, is extremely small compared to advertising for EDNP foods (Chapman et al, 2007). iv. Comments Theoretically this strategy makes sense, given that we know advertising for unhealthy foods influences children’s food attitudes and behaviours, one presumes a similar effect may be seen for the promotion of healthy foods to children. However, while advertisements for healthy foods are desirable, this approach will not be sufficient to counteract marketing of EDNP foods, and does nothing to reduce the impact of advertisements for EDNP foods. d. Consumer training in media literacy i. Basis Media literacy could involve education to equip students in critical analysis of the content and techniques of marketing, as well as nutrition claims and label reading. These skills may be covered to some extent in existing literacy teaching; however, this is limited to school students and forms only a small part of the educational curriculum. The promotion of reduced use of mass media may also reduce exposure to food marketing; although this would be a very indirect and weak approach, particularly given the widespread extent of advertisements across settings and other media. ii. Evidence from trials We identified no evidence from trials on the efficacy of such initiatives. v. Other evidence In an Australian study, children between the ages of 8-11 were shown to have good media literacy with the ability to understand the persuasive techniques used by food advertisers on television (Mehta et al, 2010). However, these children still desired the advertised foods (particularly EDNP foods) and reported they would subsequently make purchase requests from their parents (Mehta et al, 2010). This example illustrates that the use of media literacy campaigns may be redundant, as children’s food purchasing decisions and attitudes towards foods appear to be more complex and not simply based in an inability to understand advertiser’s intentions. 4. Quality and nature of evidence to support strategies to reduce children’s exposure to food marketing There is a reasonable amount of good quality evidence addressing the issue of marketing of foods to children. While there are no controlled trials which have evaluated the impact of Page 49 of 170

changes in regulation of marketing, as this is not feasible or ethical, the evidence is impressive in that it comes from a range of sources including small scale experimental studies, ecological assessments, repeated cross sectional observational data and parallel data from other public health areas. This provides a certain amount of confidence to decision making. Table 8.Quality and nature of evidence to support action on food marketing to children Intervention

Level of confidence in the evidence

Self-regulation of food marketing Statutory policy to restrict food marketing Oppositional marketing of healthy behaviours Consumer training in media literacy

med high

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect low med med

high

low

low

low

low

5. Potential impact and sustainability Few studies have attempted to quantify the impact of restrictions on marketing for EDNP foods to children on the prevalence of overweight and obesity among children. Given that food advertising is one of many environmental factors that may impact on a child’s food intake and activity, and that the impact is also moderated by each individual child’s level of exposure to food marketing, this is a difficult task. Goris et al (2009) have estimated the contribution of unhealthy food advertising to the prevalence of overweight and obesity among Australian children to be 10 – 20%, although the study has numerous limitations related to the evidence used to inform the modelling. For example, the study used evidence regarding the effects of exposure to advertising on energy intake based on US data from 1977, combined with expert opinion. Veerman et al (2009) used a similar approach to estimate the impact of removing advertising and promotion to children aged 6-12 years in the USA. Their model predicted that reducing the exposure to zero would decrease the average BMI by 0.38 kg/m2 and lower the prevalence of obesity from 17.8% to 15.2% for boys and from 15.9% to 13.5% for girls. This is a profound effect implying that up to one in three obese children in the USA may not have been obese in the absence of TV food advertising to them. The introduction of statutory regulations should require little continued input to sustain them whilst self-regulation may require additional monitoring to test their appropriateness and effectiveness. 6. Other considerations a. Feasibility and acceptability to all stakeholders Any form of statutory regulation is going to be subject to resistance from food and advertising industries. However, given that current forms of self-regulation have had no impact on children’s exposure to food advertising in Australia (King et al, 2010), statutory Page 50 of 170

policy that includes all food and beverage advertisers is required. This may be more acceptable if government set basic parameters or frameworks for limiting exposure of children and young people to advertising and promotion, with companies able to determine how they complied within such a framework. b. Relevance to Australia (transferability) The UK statutory policy to ban marketing of foods and beverages high in fat, sugar and salt during children’s airtime is highly transferable to the Australian setting. Foods and beverages deemed as inappropriate for marketing to children could be assessed using the Food Standards Australia and New Zealand nutrient profiling tool, which was adapted from the profiling tool used in the UK for identifying foods high in fat, sugar or salt that would be banned from advertising during children’s airtime (Hebden et al, 2010b; Ofcom, 2010). Restrictions should ideally be applied during times when children comprise a large proportion of the audience. Self-regulatory policies have defined this to be 50%, however there are no time slots on Australian television where this occurs (Hebden et al, 2010a). A ban between 7:00 to 20:30 has been shown to yield the greatest reduction in unhealthy food and beverage advertisements, although data on children’s audience current viewing patterns would need to be updated to confirm time frames that would yield the greatest reduction in exposure (Kelly, King et al, 2007). c. Cost considerations The introduction of regulations (self-regulation or statutory) to restrict marketing of foods to children would be relatively inexpensive. It has been suggested that such restriction would result in considerable loss of income for television companies, advertisers and even the food companies themselves. However, an assessment of the economic impact of the UK regulations by Ofcom found the costs to be quite negligible as advertising spots were taken by other products (Ofcom, 2010). Economic modelling, to estimate the potential costeffectiveness of removing television advertising for high-fat and/or high-sugar foods during times when the highest numbers of children are watching, found this intervention to be highly cost effective, due to the low cost and large population impact (Magnus et al, 2009). d. Timeframes in implementation and impact Regulation of marketing could be introduced quickly and efficiently in Australia by modifying existing codes. Thus there might be a relatively short timeframe before this intervention could impact upon children’s behaviour and their food intake. There is no literature regarding the timeframes over which an impact on children’s weight status may be seen. e. Equity issues Given this is a population approach, equity is not an issue; in fact, this strategy may be particularly effective for more disadvantaged families. Although, one may consider those children who watch less television or access less media would benefit less from the introduction of policy to ban food marketing to children. f. Other positive or negative side-effects Banning unhealthy food marketing to children could be complemented by food labelling policy to reduce the targeting of children through food packaging e.g. characters, nutrition claims or popular personalities (Hebden et al, 2011).

Page 51 of 170

References Arambepola C, Scarborough P, Rayner M. Validating a nutrient profile model. Public Health Nutrition 2007; 11(4):371-8. Cairns G, Angus K, Hastings G. The extent, nature and effects of food promotion to children: a review of the evidence to December 2008. United Kingdom: Institute of Social Marketing, University of Stirling & Open University; 2009. Chapman K, Kelly B, King L, Flood V. Fat chance for Mr Vegie TV ads. Medical Journal of Australia 2007; 31(2):190 Dixon HG, Scully ML, Wakefield MA, White VM, Crawford DA. The effects of television advertisements for junk food versus nutritious food on children's food attitudes and preferences. Social Science & Medicine 2007; 65(7):1311-23. Goris JM, Petersen S, Stamatakis E, Veerman JL.Television food advertising and the prevalence ofchildhood overweight and obesity: a multicountry comparison. Public Health Nutrition 2009; 17:1-10. Hastings G, Stead M, McDermott L, Forsyth A, MacKintosh AM, Rayner M, Godfrey C, Caraher M, Angus K. Review of research on the effects of food promotion to children. Glasgow: University of Strathclyde, Centre for Social Marketing; 22 September 2003. Hawkes C, Lobstein T. Regulating the commercial promotion of food to children: A survey of actions worldwide. International Journal of Peadiatric Obesity 2010; [Epub ahead of print] doi:10.3109/17477166.2010.486836 Hebden L, King L, Kelly B, Chapman K, Innes-Hughes C, Gunatillaka N. Regulating the types of foods and beverages marketed to children: how useful are nutrient criteria developed by food companies? Nutrition & Dietetics 2010; 67: 258–266 (referred to as Hebden et al, 2010b) Hebden L, King L, Kelly B, Chapman K, Innes-Hughes C. A menagerie of promotional characters: promoting foods to children through food packaging. Journal of Nutrition Education and Behavior 2011; 43(5): 349-355 Hebden L, King L, Kelly B, Chapman K, Innes-Hughes C. Industry self-regulation of food marketing to children: reading the fine print. Health Promotion Journal of Australia 2010; 21(3): 229-35 (referred to as Hebden et al, 2010a) Institute of Medicine (IOM). Food marketing to children and youth: threat or opportunity? Washington DC: Food and Nutrition Board, Board on Children, Youth and Families. Institute of Medicine of the National Academies; 2006. Jenkin G, Wilson N, Hermanson N. Indentifying 'unhealthy' food advertising on television: a case study applying the UK nutrient profile model. Public Health Nutrition 2008; 12(5):61423. Kelly B, Baur L, Bauman A, King L, Chapman K, Smith B. Food and beverage company sponsorship of children’s sport – who pays? Health Promotion International 2010. [Epub ahead of print] doi:10.1093/heapro/daq061 Kelly B, Bochynska K, Kornman K, Chapman K. Internet food marketing on popular children's websites and food product websites in Australia. Public Health Nutrition 2008; 11(11):1180-7. Page 52 of 170

Kelly B, Chapman K. The extent and nature of food marketing in children’s magazines. Health Promotion International 2007; 22(4):284 - 91. Kelly B, King L, Bauman A, Smith BJ and Flood V. The effects of different regulation systems on television food advertising to children. Australian and New Zealand Journal of Public Health 2007; 31(4):340-343. King L, Hebden L, Grunseit A, Kelly B, Chapman K, Venuggopal K. Industry self regulation of television food advertising: responsible or responsive? International Journal of Pediatric Obesity 2010; [Epub ahead of print] doi:10.3109/17477166.2010.517313. Kunkel D, McKinley C, Wright P. The impact of industry self-regulation on the nutritional quality of foods advertised on television to children. A report commissioned by Children Now. University of Arizona; 2009. Ludwig DS, Nestle M. Can the food industry play a constructive role in the obesity epidemic? Journal of the American Medical Association 2008; 300(15):1808-1811. Magnus A, Haby MM, Carter R, Swinburn B. The cost-effectiveness of removing television advertising of high-fat and/or high-sugar food and beverages to Australian children. International Journal of Obesity 2009; 33(10):1094-102. Mehta K, Coveney J, Ward P, Magarey A, Spurrier N, Udell T. Australian children's views about food advertising on television. Appetite 2010; 55(1):49-55. Office of communications (Ofcom). HFSS advertising restrictions: Final Review. July 2010. Available from: http://stakeholders.ofcom.org.uk/binaries/research/tv-research/hfss-reviewfinal.pdf. Romero-Fernandez M, Royo-Bordonada MA, Rodriguez-Artalejo F. Compliance with selfregulation of television food and beverage advertising aimed at children in Spain. Public Health Nutrition 2009; First view:1-9. Veerman JL, Van Beeck EF, Barendregt JJ, Mackenbach JP. By how much would limiting TV food advertising reduce childhood obesity? European Journal of Public Health 2009; 19:365-369. World Health Organisation (WHO). Set of recommendations on the marketing of foods and non-alcoholic beverages to children. 2010. ISBN 9789241500210: Available from: http://whqlibdoc.who.int/publications/2010/9789241500210_eng.pdf.

Page 53 of 170

4.4 ACTION AREA - IMPROVED PHYSICAL ACTIVITY, NUTRITION AND WEIGHT STATUS IN EVERYDAY SETTINGS 1. Rationale for this action area • Most health behaviour change programs are organised and delivered within various settings. There is an obvious practical reason for engaging people into programs in the places where they live, work and play each day. • A settings approach to obesity prevention also provides an organising framework that facilitates the translation of theory, evidence and experience into action at all stages across the life course. It encourages cross sectoral action and provides opportunity for building the local skills and structures and fosters a greater breadth of community and participant involvement in any project. Most importantly, working within settings allows for the appropriate consideration of context when utilising evidence generated from other areas. • There is now relative consensus that obesity prevention efforts need to begin early in life, as the earlier the intervention, the greater the likelihood to effect behaviour changes. Changes in work practices and family structures within Australia have created a situation where a large proportion of young children are being cared for outside of their homes for a significant period of the day. The ABS estimates that around 35% of all children 0-4 years use formal childcare facilities, making them an obvious setting for action. Childcare centres also provide an opportunity to introduce children (and their parents) to appropriate food and physical activity behaviours that will help promote good health and growth but prevent excessive weight gain during this critical period for growth and the development of health behaviours. • Schools are an attractive and popular setting for implementing health interventions targeting children. It is compulsory for all children in Australia to attend school until at least 16 years and there are opportunities to integrate programs within the school curricula. In addition, weight-related behaviours are directly influenced by what happens within the school environment. Children spent around six hours at school each day, 40 weeks of the year. While at school they consume meals or snacks and engage in organised physical activity as well as free play. • Worksite interventions can also potentially reach a large number of people at a relatively low cost and the workplace is one of the few settings that has been successful at recruiting males into health promotion programs and can also engage members of some disadvantaged groups, who may not otherwise become engaged in programs. • We all live in communities that assist or constrain attempts to develop and promote appropriate behaviours. Local communities consist not only of a collection of families, but also commercial operators, government agencies, non-government organisations and religious groups, as well as the physical infrastructure and services available within that locality. There are a range of community factors that influence eating and physical activity and which therefore constitute potential points of intervention. • The existing health infrastructure includes services that play an important role in addressing the obesity problem, through promoting physical activity and healthy eating and by treating overweight and obesity using a variety of methods. The largest health service is primary health (medical) care, which usually deals with the treatment of individuals who have an existing weight problem.

Page 54 of 170

However, there is an opportunity to involve such services in family-focused prevention programs. The sheer numbers, distribution and contacts that general practitioners have with the community make them an important element in health service action on weight gain prevention. The Bettering the Evaluation and Care of Health (BEACH) dataset indicates that from March 2008 to April 2009, there were about 112 million general practice consultations paid for by Medicare, or an average of 5.1 visits per person (Britt et al, 2009). 2. Proposed actions to improve nutrition, physical activity and weight status in everyday settings Table 9. Proposed actions in everyday settings Proposed actions

Early Childhood Services

Type of intervention Targeted Enabling/ behavioural amplifying  

•educating providers, parents and children •modifying childcare meals and snacks, and policies for foods provided from home •promoting active play and activities •reducing television viewing at childcare

Information sources

Two systematic reviews Evidence summaries Theory and logic





Multiple systematic reviews, Cochrane reviews Meta-analyses Cost-effectiveness assessments Evidence summaries and reviews Parallel evidence Extrapolated evidence





Systematic reviews Meta-analyses Evaluation reports Evidence summaries

Whole of community programs





Narrative reviews Single study evaluations Cost effectiveness assessments Parallel evidence

Primary care health services



Schools •multi-component interventions •increase PE at schools, including afterschool programs •reduce sedentary behaviours •changes to school food environment including canteen •soft drink reduction programs •promote fruit and vegetable consumption •professional development for teachers; information for parents Worksite •informational and behavioural strategies •modify food environment •workplace redesign •physical activity challenges/programs

Systematic reviews Parallel evidence Logic

Page 55 of 170

3. Targeted behavioural and environmental change Interventions The logic and potential benefits of implementing behaviour change programs to address weight in everyday settings has meant that much of the activity within obesity prevention initiatives has been focussed within this action area. The volume of evidence resulting from the evaluations of these programs is considerable, especially for the school setting, but unfortunately it is not always consistent or clear. However, there is a degree of agreement in regard to the types of actions or issues most likely to be effective and appropriate within each setting. Whilst there has been some focus on nutrition and physical activity environments within each setting, the bulk of interventions have addressed specific nutrition or physical activity behaviours of relevance to the achievement of energy balance. a. Early childhood services i. Basis The widespread use of childcare facilities provides an opportunity to introduce children and their parents to appropriate food and physical activity behaviours that will promote good health and growth but prevent excessive weight gain. Early childhood services have good communication links with parents. ii. Evidence from trials Overall A review of three RCTs by Bond (2009) found a variety of approaches to increase physical activity. Only one study in a Latino community showed a statistically significant advantage (slower rate of increase in BMI) in the intervention group. However, trends for a decrease in BMI and weight loss favoured the intervention groups in other studies. A review by Hesketh & Campbell (2010) examined obesity prevention interventions in children between the ages of birth to five years. Most studies employed multiple modes of intervention delivery. In preschool childcare settings there were three studies reporting the intervention to be beneficial, three found no effect, and three were inconclusive. Specific focus • Educating providers, parents and children An evaluation of the “Hip-Hop to Health Jr. Program” demonstrated the value of educating preschool children and their parents about healthy eating and physical activity. One and two year evaluation showed reduced weight gain in those receiving the intervention (Fitzgibbon et al, 2005). • Modifying childcare meals and snacks A number of preschool programs such as Healthy Start (Williams et al, 2004) have been successful at modifying dietary habits of children but these have not influenced weight status. • Reducing television viewing Programs to reduce TV watching have been run through childcare and led to a reduction in TV viewing times but no impact on weight status (Dennison et al, 2004). • Providing active play The evaluation of “Munch and Move” in a small sample of NSW preschools found that the professional development program which included training on fundamental movement skills Page 56 of 170

resulted in improvements in some skills, compared to controls. iii. Other evidence Very few interventions have been specifically directed at child care facilities. However, findings from the few studies conducted to date indicate that: 1) Preschool children may not be meeting recommendations for physical activity; 2) Child-care policies and practices can greatly influence diet and physical activity levels; 3) Children in centres where parents provide food have an excessive volume of extra foods in lunchboxes (Kelly et al, 2010). iv. Comments The opportunity to positively impact on weight status through programs set within childcare facilities is relatively unexplored but offers significant potential.

b. Schools i. Basis Schools function to teach children essential knowledge and skills, including appropriate health behaviours. They are also a setting where significant food and physical activities occur and may serve as a community hub for families. ii. Evidence from trials Overall, results are inconsistent in regards to improvement in BMI across reviews; however, school-based interventions have beneficial effects on education and behavioural change. Since 1990, there have been 11 systematic reviews and five meta-analyses of school-based interventions to prevent and control obesity. The effect reported in these reviews of interventions achieving weight reduction was stated as positive in three meta-analyses, negative in four meta-analyses and two reviews, and inconclusive in 10 reviews. Although results are inconsistent in regards to improvement in BMI, school-based interventions have beneficial effects on knowledge and behavioural change. Specific focus • Multi-component interventions The strongest evidence for an effect on weight status was identified for multi-component programs, although results were still mixed. The efficacy of various components is unknown. • Increase PE at schools An increased focus and quality of PE at school led to improved levels of physical activity but no impact on weight. • Reduce sedentary behaviours Short term programs which addressed small screen use and other sedentary behaviours did report an impact on weight status. • After-school sports programs Some evidence that after-school programs associated with improved social and academic outcomes, but no evidence of improved physical activity or weight status. • Changes to school food environment including canteen Page 57 of 170

Changes in canteen menus and differential pricing produced improvements in food intake, but no associations with weight. • Soft drink reduction programs School-based programs to reduce soft drink and replace it with water were able to reduce intake and were associated with short-term improvements in weight status. • Increase fruit and vegetables Programs promoting increased fruit and vegetable consumption, including programs which involved provision of fruit, achieved a small increase in fruit intake but no impact on weight. iii. Other evidence Many programs conducted in schools focus on improving nutrition of physical activity and other health behaviours, with no specific objectives relating to energy balance or weight status. Schools have been successful sites for behaviour change in tobacco, alcohol, drug use and sexual health (Mcbride, 2003; Lynagh et al, 1997; Thomas & Perera, 2006). In addition, school-based CVD risk reduction programs such as the Child and Adolescent Trial for Cardiovascular Health (CATCH) study which did not focus on weight have been able to demonstrate improvements in key dietary and physical activity behaviours (Luepker et al, 1996). iv. Comments Most of the evaluated school-based programs included within the systematic reviews were of very short duration and may not have been sustained for a sufficient period to impact on weight, although many beneficial returns in terms of improved nutrition and physical activity were achieved. The Cochrane review of interventions for the prevention of obesity in children identified a range of features which were most associated successful interventions. These included: • Intensity and expense of intervention were not determinants of success of the intervention. • Levers of engagement and enjoyment were found to be important factors. • Interventions that used a whole school approach appeared to have the most success. • Interventions that were underpinned by a theoretical model appeared to have the most success.

c. Worksites i. Basis The structured nature and hierarchy of workplaces lend itself to integrated programs to address weight gain prevention. It is one of the few settings to engage men and low income earners. ii. Evidence from trials Overall, evidence shows that worksite programs that target nutritional and physical behaviours confer modest, positive, weight-related benefits on participants. A meta-analysis of nine RCTs by Anderson et al (2009) found positive effects with a net loss of 1.25 Kg (95% CI: -4.63, -0.96) among workers at 6-12 month follow-up. Based on pooled results from six Page 58 of 170

RCTs, net loss of BMI was 0.47 kg/m2 (95% CI: -1.02, -0.2) at 6-12 months. Similar results were observed for studies not included in meta-analysis. A review by Benedict & Arterburn (2008) found that mean weight loss and changes in BMI were usually significantly greater in intervention vs. controls. In controlled trials reporting mean change, intervention groups lost an average of 1.0 to 6.3 kg whereas changes in the control group ranged from a loss of 0.7 kg to a gain of 0.5 kg. Specific focus • Informational and behavioural strategies These multi-component programs demonstrated modest impact on weight status, especially when there was a focus on self-monitoring or they involved the family in interventions through self-learning programmes or newsletters. • Modify food environment Programs which provide healthy food and beverages at workplace facilities, e.g. in the cafeteria or vending machines, produced dietary change but no direct weight change. • Workplace redesign Workplace changes to encourage physical activity e.g. provide space for fitness or signs to encourage the use of stairs, did improve activity but not weight. iii. Comments Obese people also suffer more sickness and absences from work with around 16 million lost working days attributable to obesity-related illness in the UK in 2002 (Foresight, 2008). Achieving a healthy workforce should therefore not only result in improved health for individuals, but also bring benefits to employers and society. The hierarchy and communication systems that exist in workplaces potentially allow the dissemination of information, and the social structure of workplaces may be able to provide support and positive reinforcement for appropriate change. Environmental changes can sometimes be achieved at worksites, as food services, workspace layout, building design and exercise facilities can be amenable to change (US Surgeon General, 2001). A limitation of worksite studies is that participants are self-selected, and thus results may not be generalisable to others. Also, interventions tend to have been conducted in large whitecollar workplaces, and therefore may not be feasible or applicable for small workplaces, or for people working part-time, shift work or in casual employment.

d. Communities i. Basis Community-level interventions target communities, neighbourhoods, families, parents, couples and disadvantaged populations. They also engage government and nongovernment organisations and commercial groups into activities. ii. Evidence from trials We could identify no systematic reviews which addressed community obesity prevention programs. This may be because of the diversity of such programs and the involvement of different sectors and settings makes comparison of programs difficult. However, recent reports from large national and international community-based obesity prevention programs Page 59 of 170

are encouraging. Results from the “Be Active Eat well” program in Colac Victoria (Sanigorski et al, 2008) and the European EPODE( Ensemble Prévenonsl'obésité des Enfants) program (EPHA, 2008) are showing promising changes in weight status of the population exposed to the intervention when compared to a control population. iii. Other evidence Community-based nutrition and physical activity programs show that it is possible to deliver programs to a wide segment of the population and achieve behaviour change. Other public health programs dealing with road safety, immunisation and tobacco have successfully utilised community-based approaches to achieve behaviour change and health improvement outcomes. iv. Comment Community-based programs take significant time to develop and implement, and often programs are conducted in partnership with a number of stakeholders from a mix of organisations and significant community engagement. Thus, detecting change in weight status is likely to take many years to achieve.

e. Health services i. Basis Primary care has frequent and ongoing contact with most Australians and provides a potential opportunity to engage people into behaviour change to prevent weight gain. ii. Evidence from trials Overall Systematic reviews indicate that there is little evidence that interventions for obesity in primary care practice result in clinically significant weight loss. A review by Tsai & Wadden (2009) included seven studies. Four out of seven RCTs were primary care practice counselling. Weight change (kg) ranged from -2.3 to 1.3. The effect of intervention was statistically significant in two studies and not significant in two studies. iii. Other evidence Primary care has played a key role in the delivery of many other public health campaigns. Their role in tobacco control in providing brief advice has been valuable and they have been central to successful immunisation and alcohol campaigns (Harris, 2008). iv. Comments To help facilitate improved action on obesity, health workers will require support and ongoing knowledge and skills development, as well as supportive information and funding systems. It will be important to ensure health workers have the capacity to coordinate multi-component interventions to address the gamut of factors that are driving weight gain in the community. It is possible to achieve this via the implementation of brief advice interventions and more intensive lifestyle programs, and these can be delivered through general medical practices and /or allied health services. Furthermore, implementation can involve broadening the skills of existing clinical professionals or engaging specialist positions.

Page 60 of 170

4. Other initiatives proposed or implemented but not evaluated • School and pre-school environments: school playing fields could be retained and made accessible to communities. • School and pre-school playgrounds could be renovated to ensure suitability for play, movement, sport and outdoor education. • Pupil height and weight could be routinely recorded. • Education authorities and schools could adopt a commercial sponsorship policy that restricts the promotion of branded products including foods and beverages. • Guidelines for physical activity and food for under-5s provided by child-carers, nurseries and pre-schools could be developed, disseminated and evaluated. • Community and voluntary organisations offering children opportunities for physical activity (appropriate for culture and gender) could be supported with public funds. • Quality indicators and inspection criteria for school food policies, physical activity and health-promoting environments could be developed. Similar criteria are needed for under-5s facilities licensed by local authorities. • Quality and accreditation requirements for early childhood services and schools could include a report on how the environment promotes diet and physical activity. • Include knowledge and skills on food and physical activity in reemployment training programmes. • Guidelines for food provided in the workplace could be developed, disseminated and evaluated.

5. Quality and nature of evidence to support action in everyday settings Interventions within this action area have been subject to numerous evaluations using experimental or quasi-experimental design with weight status as an outcome. These studies have been assessed and collated into numerous systematic or Cochrane reviews and subject to meta-analyses. Despite this, there is little clarity around what interventions or elements are associated with improved weight gain prevention. Most of the evaluations have been relatively short term and the design and implementation of program has varied greatly between studies. There are more consistent findings in relation to the ability of such interventions to improve weight- related diet and physical activity behaviour, or change structural or environment factors required to support behaviour change. The evidence picture becomes clear and more consistent when information from modelling, logic, theory and experiences of other public health issues is added to the mix.

Page 61 of 170

Table 10. Quality and nature of evidence to support action in everyday settings Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect

Early Childhood Services • educating providers, parents and children

med

med

•modifying childcare meals and snacks

low

med

•reducing television viewing at childcare

low

low

Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect

Schools •multi-component interventions

high

med - high

•increase PE at schools

med

low

•reduce sedentary behaviours

med

med

•after school sports programs

low

low

•changes to school food environment including canteen

med

low

•soft drink reduction programs

med

med

•increase fruit and vegetables at school

med

low

Intervention

Level of confidence in the evidence

med

low

•modify food environment

med

med

•workplace redesign

med

med

whole of community programs

Intervention

primary care health service programs

Level of confidence in the evidence

high

Level of confidence in the evidence

med

med

low-med

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect

Worksite •informational and behavioural strategies

Intervention

med

med

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect high

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect low

Page 62 of 170

6. Potential impact The potential impact of currently utilised interventions varies greatly, depending on the setting and the type of intervention. In general, whole of community programs have demonstrated the greatest potential impact with some studies indicating a reduction in levels of overweight and obesity in study populations or a decline in the rate of population increases. However, the confidence around these findings is limited by the small number of evaluations in this area. The impact of community-based interventions is also limited by the prevalence and extent to which obesity-promoting factors external to the community or setting. Such factors include the food supply and availability, infrastructure for physical activity, food advertising and factors related to household employment, income and housing. Current primary care interventions appear to have the least impact, with little indication of efficacy and thus potential to directly impact on population weight control. The potential impact of interventions within worksites, schools and childcare settings appears to vary greatly with the nature and intensity of intervention. Although these interventions have demonstrated mixed efficacy in terms of weight outcomes, school and childcare interventions have shown consistent improvements in weight-related behaviours and structural changes and they have good reach and uptake. Workplace interventions require further examination in terms of how best to achieve significant reach. Multi-components programs offer most potential to impact positively on population weight status, whilst some single component interventions (such as after school sport) will have less impact unless combined with other complementary projects. 7. Other considerations a. Feasibility and acceptability to all stakeholders In general, settings-based programs have good resonance with most stakeholders and are often the preferred model of intervention funding by government as they utilise existing infrastructure within the settings. Staff within schools, childcare and sometimes primary care may react against the additional burden that implementing such program requires, however, as to be effective this work competes for time and resources. b. Relevance to Australia (transferability) Many of the studies assessed within this action area were undertaken within Australia or other OECD countries. Whilst education and health systems vary between countries, all elements of the programs may not be directly transferable to Australia. There is also an issue of transferability of Australian projects to different locations, as these types of interventions often are highly context dependent. c. Cost considerations and sustainability Interventions within these settings can be moderately costly to implement and evaluate. The EPODE study allocates one euro to evaluation for every dollar spent on implementation. Implementation imposes time and resource costs on existing staff or requires the development of separate infrastructure to administer and deliver the programs. The sustainability of programs within childcare, schools and worksites often relies on the contribution of engaged individuals and suffers when these leaders tire, are appointed elsewhere or the resources are diverted to other areas requiring attention. Community-based programs that engage all sectors in the planning and delivery of the program have a greater likelihood of being sustained, although there are few accounts of how resources or interventions are sustained, unless they become part of core business in a setting or organisation. Page 63 of 170

d. Timeframes in implementation and impact Interventions that focus on behaviour change alone appear to have short-timeframes for planning and implementation. In fact, some such programs have demonstrated an impact on weight status within 12 months. Although childcare centres and schools are structures to deliver training and education programs, the integration of new elements into their curricula or functions often requires considerable negotiation and planning. That is, community – based programs often take several years lead-in time, in order to engage the community, build capacity, and plan and commence activities aimed at addressing weight gain prevention. The interventions are multi-component and progressively introduced so that many years may be required before the full intervention is operating. Thus a significant lag period before such interventions might reasonably be expected to impact on weight outcomes. e. Equity issues Programs offered at sites where people live and work have a greater potential to capture men, low income and migrant groups who may not choose to participate in programs in other settings, if they are implemented in selected locations or sites. Programs offered through the settings may appear to be equally available to all people in that setting; however this is not necessarily the case. Many components of behaviour change programs require a commitment to engage and this often comes with time and financial costs. Staff with fixed rosters and 30 minutes lunchtime may not be as flexible as management in allocating time for exercise or paying extra for healthy canteen meals. Similarly, school and out of school sports programs require clothing and equipment as well as parental time which may make them less available to some. Also, more disadvantaged communities may not have the infrastructure or capacity to adopt or engage in community initiatives, as reported in evaluations of some rural communitybased chronic disease prevention programs. f. Other positive or negative side-effects There is a concern that programs run in school settings may further stigmatise children with a weight problem or may encourage inappropriate weight loss practices. A recent review from the CO-OPS Collaboration has found no evidence that this occurs (Clark et al, 2009).

Page 64 of 170

References Anderson LM, Quinn TA, Glanz K, Ramirez G, Kahwati LC, Johnson DB, Buchanan LR, Archer WR, Chattopadhyay S, Kalra GP, Katz DL, Task Force on Community Preventive Services. The effectiveness of worksite nutrition and physical activity interventions for controlling employee overweight and obesity. A systematic review. American Journal of Preventive Medicine 2009; 37(4):340-57. Benedict MA, Arterburn D. Worksite-based weight loss programs: a systematic review of recent literature. American Journal of Health Promotion 2008; 22(6):408-16. Bond M, Wyatt K, Lloyd J, Welch K, Taylor R. Systematic review of the effectiveness and cost-effectiveness of weight management schemes for the under-fives: a short report. Health Technology Assessment 2009; 13(61):1-75 Britt H, Miller GC, Charles J, Henderson J, Bayram C, Valenti L, Pan Y, Harrison C, Fahridin S, O’Halloran J. General practice activity in Australia 1999–00 to 2008–09: 10 year data tables. Canberra: AIHW; 2009. General practice series no. 26. Cat. no. GEP 26 Brown T, Summerbell C. Systematic review of school-based interventions that focus on changing dietary intake and physical activity levels to prevent childhood obesity: an update to the obesity guidance produced by the National Institute for Health and Clinical Excellence. Obesity Reviews 2009; 10(1):110-141. Carter FA, Bulik CM. Childhood obesity prevention programs: how do they affect eating pathology and other psychological measures? Psychosomatic Medicine 2008; 70(3):363371. Clark R, Waters E, Conning R, Armstrong R, Petrie R. Evidence summary: considerations regarding harm minimisation for obesity prevention policies and programs for preadolescents and adolescents. Geelong: CO-OPS Secretariat, Deakin University; 2009. Cook-Cottone C, Casey CM, Feeley TH, Baran J. A meta-analytic review of obesity prevention in the schools: 1997-2008. Psychology in the Schools 2009; 46(8):695-719. Dennison B, Russo T, Burdich P, Jenkins P. An intervention to reduce television viewing by preschool children. Archives of Pediatrics and Adolescent Medicine 2004; 158:170–176 European Public Health Alliance (EPHA). EPODE — Together Let’s Prevent Childhood Obesity [internet]. 2008. Available from: www.epha.org/a/3149. Fitzgibbon ML, Stolley MR, Schiffer L, Van Horn L, KauferChristoffel K, Dyer A. Two-year follow-up results for Hip-Hop to Health Jr.: A randomized controlled trial for overweight prevention in preschool minority children. Journal of Pediatrics 2005; 146:618–625. Foresight. Tackling Obesities: Future Choices. London: Government Office for Science; 2008. Gao Y, Griffiths S, Chan EYY. Community-based interventions to reduce overweight and obesity in China: a systematic review of the Chinese and English literature. Journal of Public Health 2008; 30(4):436-448. Gonzalez-Suarez C, Worley A, Grimmer-Somers K, Dones V. School-based interventions on childhood obesity: a meta-analysis. American Journal of Preventive Medicine 2009; 37(5):418-427. Page 65 of 170

Harris KC, Kuramoto LK, Schulzer M, Retallack JE. Effect of school-based physical activity interventions on body mass index in children: a meta-analysis. Canadian Medical Association Journal 2009; 180(7):719-726. Harris M. The role of primary health care in preventing the onset of chronic disease, with a particular focus on the lifestyle risk factors of obesity, tobacco and alcohol. Centre for Primary Health Care and Equity, UNSW; 2008 Hesketh KD, Campbell KJ. Interventions to prevent obesity in 0-5 year olds: An updated systematic review of the literature. Obesity 2010; 18(Suppl 1):S27-S35. Hudson CE. An integrative review of obesity prevention in African American children. Issues in Comprehensive Pediatric Nursing 2008; 31(4):147-170. Jaime PC, Lock K. Do school based food and nutrition policies improve diet and reduce obesity? Preventive Medicine: An International Journal Devoted to Practice and Theory 2009; 48(1):45-53. Kamath C, Vickers KS, Ehrlich A, McGovern L, Johnson J, Singhal V, Paulo R, Hettinger A, Erwin PJ, Montori VM. Behavioural interventions to prevent childhood obesity: A systematic review and metaanalyses of randomized trials. Journal of Clinical Endocrinology and Metabolism 2008; 93(12):4606-4615. Katz DL. School-based interventions for health promotion and weight control: not just waiting on the world to change. Annual Review of Public Health 2009; 30:253-272. Kelly B, Hardy L, Howlett S, King L, Farrell L, Hattersley L. Opening up Australian preschoolers’ lunchboxes. Australian and New Zealand Journal of Public Health 2010; 34:288292. Klesges LM, Dzewaltowski DA, Glasgow RE. Review of external validity reporting in childhood obesity prevention research. American Journal of Preventive Medicine 2008; 34(3):216-223. Kropski JA, Keckley PH, Jensen GL. School-based obesity prevention programs: An evidence-based review. Obesity 2008; 16(5):1009-1018. Li M, Li S, Baur LA, Huxley RR. A systematic review of school-based intervention studies for the prevention or reduction of excess weight among Chinese children and adolescents. Obesity Reviews 2008; 9(6):548-559. Luepker RV, Perry CL, McKinlay SM, Nader PR, Parcel GS, Stone EJ, Webber LS, Elder JP, Feldman HA, Johnson CC, Kelder S, Wu M. Outcomes of a field trial to improve children’s dietary patterns and physical activity. The Child and Adolescent Trial for Cardiovascular Health. CATCH collaborative group. Journal of the American Medical Association 1996; 275(10):768-76. Luttikhuis O, Hiltje B, Jansen L, Shrewsbury H, O'Malley VA, Stolk C, Summerbell RP. Interventions for treating obesity in children. Cochrane Database of Systematic Reviews 2009; 1:CD001872. Lynagh M, Schofield MJ, Sanson-Fisher RW. School health promotion programs over the past decade: a review of the smoking, alcohol and solar protection literature. Health Promotion International 1997; 12:43-61

Page 66 of 170

McBride N. A systematic review of school drug education. Health Education Research 2003; 18(6):729-742 Nowicka P, Flodmark CE. Family in paediatric obesity management: A literature review. International Journal of Pediatric Obesity 2008; 3(Suppl 1):44-50. Salmon J, Brown H, Hume C. Effects of strategies to promote children's physical activity on potential mediators. International Journal of Obesity 2009; 33(Suppl 1):S66-73. Sanigorski AM, Bell AC, Kremer PJ, Cuttler R, Swinburn BA. Reducing unhealthy weight gain in children through community capacity-building: results of a quasi-experimental intervention program, Be Active Eat Well. International Journal of Obesity 2008; 32(7):1060– 7. Seo DC, Sa J. A meta-analysis of obesity interventions among U.S. minority children. Journal of Adolescent Health 2010; 46(4):309-323. Shaya FT, Flores D, Gbarayor CM, Wang J. School-based obesity interventions: a literature review. Journal of School Health 2008; 78(4):189-196. Thomas R, Perera R. School-based programmes for preventing smoking. Cochrane Database of Systematic Reviews 2006; 3: CD001293. Available from: http://mrw.interscience.wiley.com/cochrane/clsysrev/articles/CD001293/pdf_fs.html Tsai AG, Wadden TA. Treatment of obesity in primary care practice in the United States: a systematic review. Journal of General Internal Medicine 2009; 24(9):1073-1079. US Surgeon General. The Surgeon General's call to action to prevent and decrease overweight and obesity. Rockville, Md.: Office of the Surgeon General, 2001. (Accessed October 2010, at http://www.surgeongeneral.gov/topics/obesity/calltoaction/CalltoAction.pdf.) van Wijnen LGC, Wendel-Vos GCW, Wammes BM, Bemelmans WJE. The impact of schoolbased prevention of overweight on psychosocial well-being of children. Obesity Reviews 2009; 10(3):298-312. Whitlock EP, O'Connor EA, Williams SB, Beil TL, Lutz KW. Effectiveness of weight management interventions in children: a targeted systematic review for the USPSTF. Pediatrics 2010; 125(2):e396-418. Williams C, Strobino B, Bollella M, Brotanek J. Cardiovascular risk reduction in preschool children: The “Healthy Start” Project. Journal of the American College of Nutrition 2004; 23(2):117–123. Zenzen W, Kridli S. Integrative review of school-based childhood obesity prevention programs. Journal of Pediatric Health Care 2009; 23(4):242-258.

Page 67 of 170

4.5 ACTION AREA - HEALTHY ACTIVE ENVIRONMENTS 1. Rationale for this action area • The built environment encompasses a range of physical and social elements that make up the structure of a community and may influence obesity-related behaviours. • There are several urban form characteristics (natural and built environment) that tend to be associated with physical activity, and possibly nutrition behaviours. These include: • Mixed land use and density • Footpaths and cycle ways and facilities for physical activity • Street connectivity and design • Transport infrastructure and systems, linking residential, commercial and business areas. • A number of Australian and international studies have found that access to proximate and large public open space with attractive attributes, such as trees, water features and bird life is associated with higher levels of walking (Gebel et al, 2005). • There are various characteristics of the neighbourhood that have an impact on physical activity. For instance, the street layout and level of connectivity of streets (suburban or more traditional layout) determine route choices and distances and thereby make walking or cycling more or less practicable. • ‘Active travel or transport’ refers to physical activity undertaken as a means of transport. This can include walking and cycling. It also refers to the use of public transport, as most public transport trips require a walk or cycle trip at either end. • Recent research in New South Wales found that people who drove to work were 13% more likely to be overweight or obese than those who walked, cycled or used public transport, regardless of their income level. Additionally, the further people had to drive each day, the greater their weight increase (Wen et al, 2006). International comparisons of active transport and obesity rates yield similar findings (Basset et al, 2008). • Encouraging active forms of transport to and from school, including using public transport, is one way to increase physical activity and encourage the development of healthy commuting habits, which can be transferred into later life. To make cycling and walking more accessible, it is important that facilities for cyclists and pedestrians are further improved. Walking and cycling opportunities must be safe and serve as connections to facilities the community uses as part of daily life (e.g. schools, shops, and libraries). • Urban design features, such as the location of residential, business and commercial areas, in combination with transport services, influences people’s access to food. 2. Proposed actions to facilitate healthy, active environments Interventions within this action area focus on providing opportunities and encouragement to undertake physical activity as part of our daily lives –in terms of getting to and from work or school or improving incidental physical activity by better community design or by improving access to and quality of recreational spaces. As a consequence, many of these interventions are associated with changes to the existing community infrastructure and tend to be enabling or amplifying actions, rather than targeted behaviour-change programs, although often these occur in tandem. This also makes evaluations difficult as the objectives of these interventions are rarely associated with energy balance or weight status. Page 68 of 170

Table 11. Proposed actions to facilitate healthy, active environments Proposed actions

Type of intervention Targeted behavioural 

Enabling/ amplifying

Travelsmart- car reduction programs





Workplace organisational travel plans





Active travel to school programs

Active travel infrastructure Lockers, showers at work or school



 

Information sources

Reviews, agency reports, fact sheets Evaluation reports, expert assessments, extrapolation, modelling Observational studies, limited evaluations Reviews, Limited evaluations, observational studies Observational studies

Safe routes to school



Traffic calming



Financial or tax incentives



Evaluation studies, modelling



Observational studies

Improving access to parks and recreational facilities Park and school ground redesign Building, planning and design codes

 

Reviews, Evaluation studies, observational studies Limited evaluations, observational studies

Limited experimental studies, observation studies Program logic, modelling, theory

3. Summary of the evidence a. Active travel to school programs i. Basis Travel to school provides an excellent opportunity for children to engage in physical activity. The content of these programs varies slightly but usually contains an educational component with a focus on health and road safety, supervised walking to school programs (walking buses), cycling groups and even car-pooling. ii. Evidence from trials Many programs have been evaluated, but the outcomes mainly focus on reduced car trips. Many programs have been able to demonstrate improved participation rates over the life of the program and a small number (including one well designed Australian RCT) have shown a very small net increase in physical activity. Where evaluations have assessed weight or health measures, they have been unable to demonstrate improvements in these outcomes.

Page 69 of 170

iii. Other evidence The ACE Obesity project evaluated the cost-effectiveness of the Walking School Bus program for Australian primary school children as an obesity prevention measure. This analysis suggested that this was not a cost-effective option, but their modelling was hindered by limited availability of data and the high establishment costs of such a program. iv. Comments Variations of these programs are currently operating in many Australian states, the USA, the UK and other European countries. They are usually administered by local government authorities. Boys are more likely to participate in these programs than girls. Some baseline analyses have shown that those who already engage in active travel to school have lower fat stores and better cardiovascular fitness but this does not improve over the program. Research in this area would benefit from more consistent methodology and content of programs. b. Interventions to promote reduced use of cars (TravelSmart) i. Basis TravelSmart is a voluntary behaviour change program that encourages people to use sustainable travel modes such as public transport, walking, cycling and carpooling rather than single occupancy travel in a car. Its focus is more on environmental than health benefits. ii. Evidence from trials Although these programs have been evaluated in some depth, there has been no focus on health outcomes. An evaluation of the Perth program using repeat cross sectional community surveys indicated a significant reduction in car journeys and a shift to more active forms of travel. III. Other evidence A recent economic analysis found that although TravelSmart was unlikely to be cost-saving, it was rated a cost effective intervention to improve physical activity. iv. Comments The TravelSmart Program operates in most Australian states and a similar program exists in the UK and the USA. Implementation guides and protocols are available to support widespread implementation. The program has the active support of governments because of the potential benefits on travel infrastructure loads and road building costs. They are generally low cost programs that have some potential to improve physical activity. c. Workplace organisational travel plans i. Basis A travel plan is a package of measures produced by employers to encourage staff to use alternatives to single-occupancy car use. Such a plan could include: car sharing schemes; a commitment to improve cycling facilities; a dedicated bus service or restricted car parking allocations. It might also promote flexible working practices such as remote access and Page 70 of 170

video conferencing. These programs share many of the attributes of TravelSmart and school-based programs. ii. Evidence from trials Evaluation of these programs has focused on travel mode use and a Cochrane review of programs found that more than half reported a positive shift away from car use, but there were many problems and inconsistencies with the evaluations. The best studies revealed significant improvements in walking, cycling and public transport use. iii. Other evidence Only one study assessed health outcomes and this focussed on self-reported mental health and wellness via SF36 which showed minor improvements in some but not all of the subscales. iv. Comment Most of the reported programs were implemented in the USA or UK. This form of intervention has been under-explored within Australia and is a natural extension of TravelSmart programs. d. Provision of infrastructure cycle paths, footpaths and improved public transport i. Basis Appropriate levels of infrastructure are important to enable, support and encourage active travel. The connection between availability of such infrastructure and its use is frequently shown in qualitative assessments of public needs. Currently there is an imbalance in the infrastructure investment for car use and that for more active forms of travel. ii. Evidence from trials The effects of infrastructure interventions are rarely evaluated effectively in terms of impact on physical activity and never in terms of weight status. The most common outcome of any evaluation is frequency of use. One study evaluated the impact of using public transport to replace car use and found that it resulted in a small increase in walking and daily step counts (Wener and Evans, 2007). iii. Other evidence A range of studies from the USA (Basset et al, 2008) Netherlands (Salaens et al, 2003) and one from Victoria (Fraser and Lock, 2010) generally show that improved path networks, connectivity and separation from traffic resulted in significant improvements in use. Likewise, improvements to footpaths and development of dedicated trails to urban centres were associated with increased walking as a form of transport (Owen et al, 2004). There is significant evidence that improvements to public transport infrastructure result in increased patronage and reduced car use (Heath et al, 2006). iv. Comments These interventions have been implemented in many sites, with large programs in Denmark and the Netherlands. Most evaluations of impact are from USA, UK and WA, NSW and Victoria. Other findings from work in this area suggest that segregation from cars, level of connectivity, proximity to employment areas and reliability and repair of active travel Page 71 of 170

infrastructure, is important in influencing its use. e. Provision of lockers, change facilities and showers at worksites and school i. Basis Active travel to work or school will be encouraged if walkers and cyclers have access to changing facilities and showers to refresh before work and safe places to secure their bicycles. ii. Evidence from trials We could identify no studies that assessed impact of this intervention on measured physical activity or weight status. iii. Other evidence Two studies found that provision of facilities for bicycle parking, personal showering and locker storage at destinations resulted in small increases in rates of cycling and walking (Miklejohn and Wake, 2007; Hosking et al, 2010). iv. Comments This practice occurs in many locations. Concern over secure storage of bikes is often cited as a barrier to increased use of this means of travel to work or school. f. Development of safe routes to school i. Basis Safe routes to school programs use a variety of education, engineering and enforcement strategies that help make routes safer for children to walk and bicycle to school; as well as encouragement strategies to entice more children to walk and bike. The engineering component of this program focuses on physical changes to create safe and accessible routes to school by improving paths, creating safer crossings and slowing down traffic. ii. Evidence from trials Pre-post assessment of usage is the main form of evaluation with many, but not all studies showing increases in the number of students who report walking or cycling to school and reductions in the number of children being driven to school. iii. Other evidence There have also been demonstrated improvements in road safety behaviours. iv. Comments As many children in Australia live close to primary school, safe routes to school programs will result in improvements in neighbourhoods that may encourage increased walking and cycling in adults as well. g. Traffic calming and safety measures i. Basis Page 72 of 170

Traffic calming measures such as speed humps, traffic circles, and pedestrian refuges are utilised to slow traffic speeds and reduce flows providing greater opportunities for pedestrian and cycle traffic to mix with car traffic. ii. Evidence from trials Evaluation of these programs has not focussed on detecting improvements in measured physical activity or weight status. Evaluation relied on repeat surveys of self-reported behaviour with one study revealing 20% of participants indicating that they walked more (Morrison et al, 2004). iii. Other evidence Traffic calming reduces pedestrian injuries and has been shown to have some positive impact on the rates of walking and cycling in the calmed area, especially in children. iv. Comments Evaluated studies come mainly from UK and NZ. High levels of vehicular traffic have been associated with lower rates of physical activity in nearby areas. Vehicular traffic is often cited as a barrier to active transport. Modifying traffic patterns through calming mechanisms may be a logical way to influence physical activity levels although the effects may be greater on residential than main roads.

h. Fiscal and taxation incentives to encourage physical activity and active transport i. Basis A number of jurisdictions have introduced financial penalties, tolls or congestion taxes on those who drive cars in built up areas. Some worksites have also offered financial incentives (or offset benefits available to motorists) to those who choose not to drive and park at work. Other jurisdictions have provided tax incentives to encourage children to attend eligible physical activity programs. ii. Evidence from trials Only one study evaluated financial incentives for active transport which involved reimbursing staff that choose not to drive to work the costs of subsided worksite parking that is available to car commuters. The program resulted in a very small increase in active travel journeys in eight worksites (1%) after three years. However, an uncontrolled trial in another USA worksite saw active travel increase from 2 to 34% of the workforce as a result of a move to a new worksite and introduction of a similar scheme (Hosking et al, 2010). A recent analysis assessed the impact of the Canadian Children’s Fitness Tax Credit (CFTC) scheme which allows a tax credit of up to $500 to register a child in an eligible physical activity program (Spence et al, 2010). There was no attempt to measure the direct impact on physical activity levels or weight status. The study found that around 26% of parents claimed the tax credit but low income households were significantly less likely to report uptake of this scheme. Only 16% of parents believed the tax credit resulted in greater participation in sport by their children. iii. Other evidence Page 73 of 170

In contrast to financial incentives, congestion charges appear to have a larger impact, with evaluation of such tolls in Stockholm leading to a 25% reduction in traffic numbers and the London Tax resulting in 16% reductions in traffic volume after two years. The London charge resulted in a shift to public transport with 50-60% of journeys avoided. iv. Comments We identified evaluated programs in the USA, UK and Europe. Currently there are potential tax breaks associated with purchase and use of motor vehicles in Australia but the Fringe Benefits Tax (a tax on the benefits wage earners are given by employers) regulations inhibit a similar financial incentive for active travel. Congestion charges have also been introduced on long stay parking spaces in Melbourne by the Victorian Government but the impact has not been evaluated. i. Improved access to parks and recreational facilities i. Basis The level of access to quality parks and recreational facilities has been identified as a key predictor of level of physical activity and may explain some of the differentials in physical activity between different levels of SES. Improving access is a planning issue with responsibility at a local government level. ii. Evidence from trials There is no direct intervention research evidence to support this action. However, a Community Guide review from the US Centers for Disease Control and Prevention (CDC) concluded that efforts to increase access to places for physical activity, when combined with informational outreach, can be effective in increasing physical activity (Kahn et al, 2002). The studies reviewed by the Community Guide included interventions such as creating walking trails, building exercise facilities, and providing access to existing facilities. However, it was not possible to separate the benefits of improved access to places for physical activity from health education and services that were provided concurrently. iii. Other evidence There is a significant amount of observational data and logic to support this intervention. A comprehensive review of studies across the world indicated that access to facilities and programs for recreation near their homes, and time spent outdoors, correlated positively with increased physical activity among children and adolescents (Sallis et al, 2000). In addition, teens that live in communities that make school and recreational facilities accessible on weekends, may have lower risk for being overweight (Scott et al, 2007). A national study conducted in 2006 with 20,745 adolescents found communities with seven recreational facilities located within a five-mile radius had 32% fewer overweight teens than did communities with no facilities (Gordon-Larsen et al, 2000). iv. Comments Improved access to parks and recreational facilities has a number of social benefits including: encouraging social interaction and development of social skills, improving social networks and social capital, increasing community cohesion and pride, safer communities. j. Park and school-ground redesign i. Basis Page 74 of 170

A number of studies have shown that not all parks were equally effective in encouraging physical activity. Parks that were aesthetically pleasing, had defined activity features such as walking trails or sports facilities and were perceived as safe were associated with higher levels of overall physical activity. ii. Evidence from trials There is no clear experimental evidence to support this intervention, but a number of evaluated interventions give an indication of potential merit, although evidence is not consistent. A number of studies have examined the impact of renovations to school playgrounds and found that appropriate changes can result in higher levels of activity. A UK study conducted in 2007 redesigned playgrounds at 15 low-income schools using coloured lines to stimulate play. Objective measures of physical activity showed that students increased their physical activity by about 30 minutes per week, compared to students from 11 schools where playgrounds were not updated (Ridgers et al, 2007). Similar improvements were seen with school-ground renovations in the USA (Colabianchi et al, 2009). However renovations to play areas and recreation facilities in a public park actually resulted in decreased use in all areas except the newly added skate park (Cohen et al, 2006). iii. Other evidence A study of preschools in Stockholm, Sweden, found that children aged 4 to 6 with access to natural areas with trees, shrubbery and dirt had step counts 20% higher than children with paved play areas (Boldermann et al, 2006). iv. Comments Park and playground design clearly has an influence over usage and physical activity levels, but insufficient evidence is known regarding how to influence these positively. k. Building, planning and design codes i. Basis Communities that have a mixture of residential, commercial and public facilities all in the same area have been associated with higher levels of walking, cycling and physical activity; and essential facilities such a schools, doctors and shops are within accessible distances. The use of codes to encourage land zoning for mixed use development has been identified as a suitable approach to ensuring new residential or in-fill sites retain this balance. II. Evidence from trials There have been limited studies of interventions to improve zoning for mixed land use and the outcomes assessed are usually restricted to self-reported improvements in walking or cycling. The CDC Community Guide assessed this area and found a range of evidence to support its effectiveness with a median improvement in some aspect of physical activity of 61% (Heath et al, 2006). iii. Other evidence A review of quasi-experimental studies indicated residents from high walkability neighbourhoods (defined by higher density, greater connectivity, and more land use mix) Page 75 of 170

reported twice as many walking trips per week than residents from low walkability neighbourhoods (defined by low density, poor connectivity, and single land uses) (Saelens et al, 2003). iv. Comments Examples of suitable codes or guidance already exist in Australia. Healthy by Design: a planners’ guide to environments for active living resource was developed by the Victorian division of the Heart Foundation to provide voluntary guidance for planners and developers. Liveable Neighbourhoods is a design code introduced in Western Australia to guide the design and approval of urban development. Liveable Neighbourhoods applies to structure planning and subdivision for greenfield sites and for the redevelopment of large Brownfield and urban infill sites. 4. Other initiatives proposed or implemented but not evaluated A number of other active living interventions have been proposed in policy documents, although their impacts on physical activity or weight status have not been directly evaluated. However, their potential impact has been supported using alternative forms of evidence, such as observational data, theory or program logic and indirect or intuitive evidence. The strength of this evidence requires further assessment. These initiatives include: • Programs to promote social norms around active travel • Training in cycling skills • School staging areas and drop off zones a distance from main entrance • Improved public transport routing • Increase employment opportunities closer to where people live • Creating less car dependent suburbs • Bike rental facilities • Secure bike parking at train stations • Mandated urban planning provisions that account for health impact 5. Quality and nature of evidence to support action in active transport Evidence has been drawn from systematic reviews, Cochrane reviews and primary studies of mixed quality where evaluation relies heavily on self-report. The impact of active travel interventions are difficult to evaluate using tight experimental designs and thus evidence of a positive impact often comes from other forms of evidence. Infrastructure interventions are usually justified on the basis of ecological associations or analysis of factors associated with physical activity or weight status. Repeat cross-sectional surveys on active transport participation provide some indication of impact, but do not allow for attribution to the intervention itself. As a consequence, most systematic reviews are dominated by reports of smaller behavioural change targeted at motivated groups programs.

Page 76 of 170

Table 12. Quality and nature of evidence to support action in active transport Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action Direct Indirect low

Active travel to school

med

TravelSmart

med

low

Organisational travel plans

low

low

Active travel infrastructure

low

low-med

Worksite change facilities

low

low

Safe routes to school

med

low

Traffic calming

med

low-med

Fiscal incentives

low

low

Improved access to parks and recreational facilities

low

med

Park and school ground redesign

med

low-med

Building, planning and design codes

low

med

med

6. Potential impact and sustainability Evidence from the most promising and best evaluated interventions suggest that active travel programs could increase walking as a mode of transport by around 15-30 minutes per week in the general population, although more targeted programs produce a greater effect. The demonstrated impact of these programs on cycling was negligible or non-existent. There was no clear indication of the likely impact of these programs on energy balance or weight status, although this amount of additional activity is likely to have only a modest effect. The CDC assessment of community-scale urban design and land use policies suggest that they may increase physical activity by 61%, making this a valuable contribution to the level of incidental physical activity and possibly energy balance. These schemes have the potential to reach large sections of the community if fully implemented. The impact of improved access and design of recreational spaces is unknown and difficult to separate from other promotional activities that often occur concomitantly. However, because these facilities are not utilised on a daily basis by large sections of the community, their impact may be small. Most of these interventions result in change to community infrastructure that requires most of the resource investment at the point of establishment. As a consequence, such interventions are easier to sustain. 7. Other considerations a. Feasibility and acceptability to all stakeholders

Page 77 of 170

The promotion of active living is already an agreed objective of most governments and has strong community support. The only major vested opposition comes from the automotive industries, who oppose the diversion of funds from road infrastructure. There is strong convergence of active travel and environmental initiatives which brings additional impetus. b. Relevance to Australia (transferability) Most of the implemented and evaluated programs come from developed economies with similar government and community structures such as Australia or have been already implemented in parts of Australia. Evidence on active travel efficacy from some smaller, compact and older European cities may require caution in application to Australian cities. Parks and school playground usage may also vary between countries but the validity of approaches is likely to be consistent across countries. c. Cost considerations Active living interventions requiring infrastructure development or improvements will be initially expensive to implement but have multiple benefits and are part of the usual planning processes and do not necessarily generate additional costs. Reaching large sections of the community with behaviour change programs such as TravelSmart is also expensive, but potential returns make them worthwhile and cost-effective. d. Timeframes in implementation and impact Infrastructure projects have medium to long timeframes to complete and often require considerable time to impact on the desired outcome. Improving active travel is likely to be slow and incremental. e. Equity issues Some but not all studies show that men are more likely to participate and benefit from active transport interventions. More men and boys cycle to work or school, and a Victorian study suggested that women respond better to segregated bike and walk trails. There is no indication of social disparities in the use of programs or infrastructure, with studies showing programs work just as well in low SES areas. However, access to parks and public open spaces has a social gradient in Australia and investments in improving major public parks often favour more established and wealthier suburbs. f. Other positive or negative side-effects There are numerous health benefits from improved levels of physical activity, and adults and children who engage in more active travel have been shown to have improved fitness and reduced risk for a number of chronic conditions. There has been some suggestion that children who walk or cycle to school may be less active at other times, however studies that have assessed this issue show the reverse to be true in boys. There are considerable synergies with actions to address climate change and other environmental concerns (Fraser & Lock, 2010).

Page 78 of 170

References Active Living Research Program. Active Transportation: making the link from transportation to physical activity and obesity. Princeton NJ: Robert Wood Johnston Foundation; 2009 Alexander LM, Inchley J, Todd J, Currie D, Cooper AR, Currie C. The broader impact of walking to school among adolescents: seven day accelerometry based study. British Medical Journal 2005; 331(7524):1061–1062 Bassett DR, Pucher J, Buehler R, Thompson DL, Crouter SE. Walking, cycling, and obesity rates in Europe, North America and Australia. Journal of Physical Activity and Health 2008; 5:795-814 Boldemann C, Blennow M, Dal H, Martensson F, Raustorp A, Yuen K, Wester U. Impact of preschool environment upon children’s physical activity and sun exposure. Preventive Medicine 2006; 42(4):301-308. Cobiac LJ, Vos T, Barendregt JJ. Cost-effectiveness of interventions to promote physical activity: a modelling study. PLoS Medicine 2008; 6(7):p.e1000110. Cohen D, Ashwood J, Scott MM, Overton A, Evenson KR, Staten LK, Porter D, McKenzie TL, Catellier D. Public parks and physical activity among adolescent girls. Pediatrics 2006; 118(5):e1381–e1389. Colabianchi N, Kinsella A, Coulton C, Moore SM. Utilization and physical activity levels at renovated and unrenovated school playgrounds.Preventive Medicine 2009; 48:140–143 Cooper AR, Andersen LB, Wedderkopp N, Page AS, Froberg K. Physical activity levels of children who walk, cycle or are driven to school. American Journal of Preventive Medicine 2005; 29(3):179–184 Cooper AR, Page AS, Foster LJ, Qahwaji D. Commuting to school: are children who walk more physically active? American Journal of Preventive Medicine 2003; 25(4):273-76 Fraser SDS, Lock K. Cycling for transport and public health: a systematic review of the effect of the environment on cycling. European Journal of Public Health 2010; doi:10.1093/eurpub/ckq145 Fulton JE, Shisler JL, Yore MM, Caspersen CJ. Active transportation to school: findings from a national survey. Research Quarterly for Exercise and Sport 2005; 76(3):352–357 Gebel K, King L, Bauman A, Vita P, Rigby A, Capon A. Creating healthy environments: a review of links between the physical environment, physical activity and obesity. Sydney: NSW Health Department and NSW Centre for Overweight and Obesity; 2005 Gordon-Larsen P, McMurray RG, Popkin BM. Determinants of adolescent physical activity and inactivity patterns. Pediatrics 2000; 105:E83 Heath GW, Brownson RC, Kruger J, Miles R, Powell KE, Ramsey L. The effectiveness of urban design and land use and transport policies and practices to increase physical activity: a systematic review. Journal of Physical Activity and Health 2006; 3(Suppl 1):S55-76

Page 79 of 170

Hosking J, Macmillan A, Connor J, Bullen C, Ameratunga S. Organisational travel plans for improving health. Cochrane Database of Systematic Reviews 2010; 3:CD005575 Independent Social Research. Impacts of better use transport interventions: review of the evaluation evidence base. London: 2009 Kahn EB, Ramsey LT, Brownson RC, Heath GW, Howze EH, Powell KE, Stone EJ, Rajab MW, Corso P. The effectiveness of interventions to increase physical activity. A systematic review. American Journal of Preventive Medicine 2002; 22(4 Suppl):73-107 McCormack G, Giles-Corti B, Lange A, Smith T, Martin K, Pikora TJ. An update of recent evidence of the relationship between objective and self-reported measures of the physical environment and physical activity behaviours. Journal of Science and Medicine in Sport 2004; 7(Suppl 1):81-92. Miklejohn D, Wake D. A tale of two cities: workplace travel plan programs in Melbourne and Perth. Available in: 30th Australasian Transport Research Forum, Department of Infrastructure; 2007 Morrison D, Thomson H and Petticrew M. Evaluation of the Health Effects of a neighbourhood Traffic Calming Scheme. Journal of Epidemiology and Community Health, 2004; 58(10): 837–840 Ogilvie D, Egan M, Hamilton V, Petticrew M. Promoting walking and cycling as an alternative to using cars: systematic review. British Medical Journal 2004; 329:763 Owen N, Humpel N, Leslie E, Bauman A, Sallis JF. Understanding environmental influences on walking: review and research agenda. American Journal of Preventive Medicine 2004; 27(1): 67–76. Ridgers ND, Stratton G, Fairclough SJ, Twisk JW. Long-term effects of a playground markings and physical structures on children’s recess physical activity levels. Preventive Medicine 2007; 44(5):393-397 Saelens BE, Sallis JF, Frank LD. Environmental correlates of walking and cycling: findings from the transportation, urban design, and planning literatures. Annals of Behavioral Medicine 2003; 25:80-91 Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of physical activity of children and adolescents. Medicine and Science in Sports and Exercise 2000; 32:963-75 Scott MM, Cohen DA, Evenson KR, Elder J, Catellier D, Ashwood JS, Overton A. Weekend schoolyard accessibility, physical activity, and obesity: the Trial of Activity in Adolescent Girls (TAAG) study. Preventive Medicine 2007; 44(5):398-403 Spence JC, Holt NL, Dutove JK, Carson V. Uptake and effectiveness of the Children’s Fitness Tax Credit in Canada: the rich get richer. BMC Public Health 2010; 10:356 Wen LM, Orr N, Millett C, Rissel C. Driving to work and 18. Overweight and obesity: finding from the 2003 New South Wales Health Survey, Australia. International Journal of Obesity 2006; 30(5):782-6 Wener RE and Evans GW. A morning stroll: levels of physical activity in car and mass transit commuting. Environment and Behaviour. 2007; 39 (1): 62–74 Page 80 of 170

4.6 ACTION AREA - FOOD SUPPLY 1. Rationale for this action area • There is widespread acceptance that the current food supply is not conducive to the maintenance of energy balance. The widely available processed foods have levels of transfats, saturated fats, salt and sugar well above those recommended for good health and weight control and provide excess kilojoules. • Reformulating processed foods to reduce levels of fat, salt, sugar and kilojoules offers considerable scope to yield substantial health benefits, without requiring behavioural change at the population level. • Encouraging results have been achieved in the UK in some aspects of food reformulation, where the Government is working with the food industry and retailers to reduce salt levels. In British adults, the population intake of salt has fallen by approximately 10%, which could potentially save more than 6,000 lives a year (He & McGregor, 2008). • The current Food and Health Dialogue is a valuable initiative to drive reforms in this area, but more support must be provided to increase the range of food products covered to accelerate improvements in the nutritional profile of processed foods. • Short-term experimental studies have shown that larger portions of energy-dense food are associated with increased energy intake (Ello-Martin et al, 2005). Specifically, they have shown that both children and adults consume more at a single eating episode when offered larger portions of energy-dense food. • Conversely, a larger portion of a food with low energy density, such as a salad served at the start of a meal, can reduce overall energy intake at the meal (Rolls et al, 2007). • Studies from Denmark (Matthiessen et al, 2003) and the USA (Neilsen & Popkin, 2003) indicate that: - the portion sizes of many (but not all) commercial energy-dense foods and beverages and fast-food meals seem to have increased over time, particularly over the last ten years; - the number of super-sized food items available in grocery stores and supermarkets seems to have increased substantially; and - conventional and fast-food restaurants serve larger so-called value meals and offer allyou-can-eat buffets in the competition for customers.

2. Proposed actions to improve access and availability of appropriate foods Whilst there are many potential options to improve the quality of food supply, action in this area has focussed on three key approaches which address the problem from slightly different perspectives. The first involves improving the availability of healthy products through the development of new healthier processed products. The second requires the reformulation of existing products, to make small improvements to their nutritional quality which would result in a smaller but broader impact on the food supply. Lastly, a reduction in the portion size of commonly purchased, single use, packaged foods and beverages has also been promoted as potential approach. Surprisingly there has been little attention to strategies to directly reduce the number of unhealthy snack foods and beverages currently available.

Page 81 of 170

Table 13. Proposed actions to improve access and availability of appropriate foods Proposed actions

Type of intervention Targeted behavioural

Reported activities

Enabling/ amplifying

Development of new healthier, less energy-dense products



Evaluation studies from UK

Reformulation of existing products to reduce energy, saturated fat, salt sugar and remove trans fats



A range of regions including Australia, Europe and the USA are currently negotiating the reformulation of key food products to improve nutritional profile.



Systematic review, Action reported in Europe, Australia. Voluntary initiatives lead by food industry

Portion control



3. Summary of the evidence This is a difficult action area in which to identify direct evidence of the effectiveness of any of these proposed approaches, as interventions are often developed on an ad hoc and opportunistic basis, and rarely evaluated for their impact or outcome. a. Development of new healthier, less energy-dense products i. Basis Providing healthier, lower energy options to consumers will make it easier (possible) to choose more appropriate food products in all categories and reduce their risk of overconsumption. ii. Evidence from trials We could not identify any studies evaluating the effectiveness of this strategy with any health-related outcome. Participation in a nutrition signposting system in the Netherlands resulted in the development of a range of new products with less salt, saturated fat, sugar and more fibre, but no improvement in calorie content (Vyth et al, 2010). ii. Additional evidence There is some logic behind this approach and the development of low fat diary has been credited with assisting with reductions in the level of CVD in countries such as Finland, the USA and Australia (Buttriss, 2005); although this is being challenged in some recent assessments (German et al, 2009). The proliferation of low fat products in the past two decades has done little to address the excess energy intake and may in fact have been a contributor to passive over-consumption of kilojoules. Experience from alcohol reduction programs suggest that without complementary action, the provision of additional products which still compete with unhealthier, more familiar alternatives is unlikely to contribute to sustained behaviour change.

Page 82 of 170

iv. Comments Although the production of new healthier, lower calorie alternatives to add to the food supply may be the preferred approach of food companies attempting to diversity product lines whilst maintaining market share, a more effective option may come from a reduction in the number of unhealthy food products currently on sale. It has been established that increased variety of food products is associated with increased intake and thus potential to contribute to weight gain. Restricting the availability of existing food products may result in a greater reduction in calorie intake than offering more options. b. Reformulation of existing food products to improve their nutritional profile i. Basis Making minor changes to the formulation of current food products may reduce the fat, sugar, salt or kilojoules content or improve fibre and fruit and vegetable content. This does not require consumers to make a considered decision to swap to alternative food products, and thus may make a small but useful contribution to energy balance. These reformulations could be voluntary or achieved through regulation (as has been the case with trans-fat removal in certain countries). ii. Evidence from trials We could identify no reports that directly evaluated the impact of product reformulations on immediate health outcomes or weight. On a few occasions a direct attempt has been made to measure the extent of the reduction in the nutrient of concern after a program of product reformulation. A limited number of studies have attempted to measure the impact of product reformulation on the population intake of key nutrients. For example the UK Food Standard Authority have measured the change in population salt intake (via urinary sodium) pre and post reformulation and demonstrated a reduction in average population sodium intake levels. This has not yet been correlated to health outcomes such as blood pressure or incidence of coronary heart disease and stroke (UK Food Standards Agency, 2009). Studies of the impact of nutrition signposting systems in the Netherlands (Vyth et al, 2010) and Australia (Fear et al, 2004) revealed significant improvements through reformulation in products achieving accreditation, resulting in significant reductions in salt, sugar and calories and improvements in fibre content. iii. Other evidence Most of the data on benefits on weight and health from food product reformulation come from extrapolations and modelling. A number of agencies and organisations have made estimates of the likely extent of change to have resulted from product reformulation initiatives they have introduced, or they have extrapolated on the basis of what could be achieved if targets for product change were achieved. Estimates from the UK suggest that a reduction in saturated fat consumption in line with national guidelines could prevent 3,500 premature deaths a year (The Strategy Unit, 2007). Finland has also had a long standing program of product reformulation, with a special focus on salt reduction in the food supply. The outcomes have been evaluated more broadly. A reduction of fat intake and an increase in potassium intake via the use of reduced-sodium, potassium and magnesium-enriched salt and an increased consumption of fruit and vegetables have also contributed to the fall in cardiovascular disease (Karppanen & Mervaala, 2006). Page 83 of 170

c. Portion control i. Basis Strategies to address large portion sizes could include actions by manufacturers (including caterers) to reduce the size of single portion food and beverage products or the education of consumers about appropriate portion size. ii. Evidence from trials Reducing package size or portion of prepared foods Although reducing portion or pack sizes is often promoted as a possible strategy for addressing excessive energy intake, only a few studies have attempted to evaluate its impact. We could find only one systematic review which assessed a few small experimental studies (Steenhuis & Vermeer, 2009). In short term studies where portions of prepared foods were restricted, there was a subsequent reduction in energy intake. One short term study showed that participants who were provided with snack foods in smaller portions consumed around a third less snack food than those given usual size packages, but only in the first week of a two week trial (Stroebele et al, 2009). Educating consumers Three short-term intervention studies addressed consumers' portion size estimation skills, or education about appropriate portion sizes, and found portion size labelling or portion size information ineffective in decreasing energy intake (Steenhuis & Vermeer, 2009). iii. Other evidence Meal replacement programs effectively operate through fixed portion sizes and can serve as a model of the impact of portion control on weight. A range of clinical trials have demonstrated the efficacy of such an approach at achieving weight control, and although their effectiveness in real life is still unproven, weight loss is higher in the group treated with meal replacements than in the standard intervention (Heymsfield et al, 2003). A recent study of fast food companies’ response to calls to reduce portion size found no significant improvement in portion sizes of key fast food products between 1998 and 2006 despite calls from public health bodies to do so. The authors suggest that portion size reduction will not be achieved by voluntary efforts of the companies alone (Young & Nestle, 2007). iv. Comments Dietary surveys provide no clear evidence that portion size is an important determinant of the risk of becoming obese. However, portion size is difficult to judge since the dietary assessment in many large cohorts relies on food frequency records rather than food diaries with individual estimates of portion size. Some food manufacturers already produce multipacks composed of smaller portion sized products than their standard individually-sold versions. But the value of such an approach is questionable as ‘presence drives consumption’, i.e. the frequency of consumption of products sold in multipacks may increase simply because they are in the home. Consumer research from the Netherlands suggest that consumers prefer an increase in the range of portions rather than a reduction in existing package sizes (Vermeer et al, 2010). Page 84 of 170

A USA study by O'Dougherty et al (2006) showed that a third of fast-food restaurant patrons favoured a law requiring restaurants to change their pricing strategies and offer lower prices for smaller portions, instead of more value for money for larger portions. 4. Other initiatives proposed or implemented but not evaluated • Reduce the number of high energy-dense snack foods on the market • Decrease portion sizes served in restaurants and of packaged food in grocery stores incrementally, without knowledge of consumers • Serve larger portions of healthy, low energy-dense foods • Use commercially packaged meals that are controlled for portion size and energy density • Food labels with useful and clear information about portion size • Attractive pricing strategies to promote smaller portions • Pricing of foods based on unit weight and not bulk discount; that is volumetric pricing as used to control alcohol consumption in Scandinavia and other countries. 5. Quality and nature of evidence to support action in food supply Most interventions within this action area would be defined as enabling or amplifying action areas and thus there is very little direct experimental evidence on which to base assessments of their effectiveness. In general, the information used to support action in this area has been observational data, ecological assessment and extrapolation from other areas of action. Where evaluation of the impact of interventions has been undertaken, the outcome measures are often not related to energy balance or weight-related nutrition behaviours. Many of the estimates of effectiveness are based on modelling under ideal conditions. Table 14. Quality and nature of evidence to support action in food supply Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action

Development of new healthier, less energy-dense products

med

Direct

Indirect low

Reformulation of existing products to reduce energy, saturated fat, salt sugar and remove trans fats Portion control

med

med

med

low

med

Page 85 of 170

6. Potential impact and sustainability Without clear indications of the efficacy of interventions in terms of their effect on weight related behaviours or weight status, it is difficult to determine their likely impact on energy balance. The impact on energy balance of the introduction of new healthier products is difficult to estimate as the effect will depend on their acceptance and uptake. Previous experience has suggested that this is often low. Modelling around the impact of product reformulation suggests that its impact on calorie intake could be substantial, as its reach and uptake are universal and for popular foods, will continue to have this impact every day. However, the modelling around these changes tends to be based on ideal conditions and assumes that there is no compensation or product shifting. Despite this, logic suggests that product reformulation is likely to have a moderate impact on calorie intake. Likewise logic dictates that the impact of enforced portion control could be significant, particularly if the impact seen in short-term trials (where reduced portions lead to a reduction in snack food intake by 1/3) could be sustained. However, this also requires consumers not to compensate by consuming extra servings of the reduced portion. The benefits of these interventions in the food supply are that once instituted, they should be reasonably well sustained. However, the longevity of new food products and alternative portion sized products in the market is variable and dependent upon many factors. History suggests that most will fail despite their merit. This is another rationale for reducing unhealthy rather than increasing healthy options. 7. Other considerations a. Feasibility and acceptability to all stakeholders Actions addressing product development or reformulation are more acceptable to the food industry as they are already a part of their business processes. Some progress to improve the food supply has already been achieved through industry initiatives with minor coaxing from governments. Any attempt to restrict or reduce the number of snack food products is unlikely to be acceptable to the food industry and may produce some resistance from consumers. Governments view food reformulation as an area on which they can facilitate some cooperation between industry and public health agencies without direct investment of resources. b. Relevance to Australia (transferability) Most of the experiences in restructuring the processed food supply are directly applicable to Australia, as the industry and markets operate in much the same way across the world and food product composition is similar. c. Cost considerations The cost implications of reformulating foods, developing new products or even producing new packaging sizes, is not inconsiderable. At present, almost all this cost is borne by the food industry which is a factor in their reluctance to take on these actions. However, the risk to profits from these strategies is less than strategies which focus on reducing unhealthy food products or making them less affordable or available.

Page 86 of 170

d. Timeframes in implementation and impact Most of these actions require some time to enable innovation or to allow a staged approach to product changes. However, once instituted their impact on intake of energy is immediate and lasting. e. Equity issues The introduction of healthier food options is usually associated with price premiums which make such products less affordable or attractive to low income earners. Similarly, unless portion size reduction is accompanied by commensurate reduction in price, such options will be less attractive to those on a tight food budget. Product reformulation appears not to impact on equity issues. f. Other positive or negative side-effects The introduction of new healthier food options could result in misunderstanding by consumers that these products can be consumed in larger portions or more frequently – leading to overconsumption of food energy. This situation arose with the introduction of low fat options in the 1980s.

Page 87 of 170

References Buttriss J. A public health approach to cardiovascular disease risk reduction. Chapter 13 in: Stanner S, editor. Cardiovascular disease: diet, nutrition and emerging risk factors. London: British Nutrition Foundation; 2005 Ello-Martin JA, Ledikwe JH, Rolls BJ. The influence of food portion size and energy density on energy intake: implications for weight management. American Journal of Clinical Nutrition 2005; 82(1):236S– 241S Fear T, Gibbons C, Anderson S. The Heart Foundation's 'Tick' Program. Driving innovation for a healthier food supply. Food Australia 2004; 56(12):599-600 German JB, Gibson RA, Krauss RM, Nestel P, Lamarche B, van Staveren WA, Steijns JM, de Groot LCPGM, Lock AL, Destaillats F. A reappraisal of the impact of dairy foods and milk fat on cardiovascular disease risk. European Journal of Nutrition 2009; 48 (4):191–203 He FJ, MacGregor GA. Salt intake and cardiovascular disease Nephrology Dialysis Transplantation 2008; 23(11):3382-3385 Heymsfield SB, van Mierlo CA, van der Knaap HC, Heo M, Frier HI. Weight management using a meal replacement strategy: meta and pooling analysis from six studies. International Journal of Obesity 2003; 27:537–549. Karppanen H, Mervaala E. Sodium intake and hypertension. Progress in Cardiovascular Diseases 2006; 49:59-75 Ledikwe JH, Ello-MartinJA, Roll BJ. Portion sizes and the obesity epidemic. Journal of Nutrition 2005; 135:905–909. Matthiessen J, Fagt S, Biltoft-Jensen A, Beck AM, Ovesen L. Size makes a difference. Public Health Nutrition 2003; 6:65–72. Nielsen SJ, Popkin BM. Patterns and Trends in Food Portion Sizes, 1977-1998. Journal of the American Medical Association 2003; 289:450-453 O'Dougherty M, Harnack LJ, French SA, Story M, Oakes JM, Jeffery RW: Nutrition labelling and value size pricing at fast-food restaurants: A consumer perspective. American Journal of Health Promotion 2006; 20(4):247-250. Rolls BJ, Roe LS, Meengs JS. The effect of large portion sizes on energy intake is sustained for 11 days. Obesity 2007; 15:1535-1543. Rolls BJ, Roe LS, Meengs JS. Salad and satiety: energy density and portion size of a firstcourse salad affect energy intake at lunch. Journal of American Dietetic Association 2004; 104:1570–1576. Steenhuis IHM, Vermeer WM. Portion size: review and framework for interventions. International Journal of Behavioural Nutrition and Physical Activity 2009; 6:58 Stroebele N, Ogden LG, Hill JO. Do calorie-controlled portion sizes of snacks reduce energy intake? Appetite 2009; 52(3):793–796. The Strategy Unit. Food Matters: towards a strategy for the 21st century. UK Cabinet Office; 2007. Available from: Page 88 of 170

www.cabinetoffice.gov.uk/~/media/assets/www.cabinetoffice.gov.uk/strategy/food/food_matt ers1%20pdf.ashx. UK Food Standards Agency. UK Salt reduction initiatives. 2009, London FSA. Accessed in October 2010 from http://www.food.gov.uk/multimedia/pdfs/saltreductioninitiatives.pdf Vermeer WM, Steenhuis IHM, Seidell JC. Portion size: a qualitative study of consumers’ attitudes toward point-of-purchase interventions aimed at portion size. Health Education Research 2010; 25(1):109-120 Vyth EL, Steenhuis IHM, Roodenburg AJC, Brug J, Seidell JC. Front-of-pack nutrition label stimulates healthier product development: a quantitative analysis. International Journal of Behavioral Nutrition and Physical Activity 2010; 7:65 Young LR, Nestle M. Portion Sizes and obesity: responses of fast-food companies. Journal of Public Health Policy 2007; 28:238–248.

Page 89 of 170

4.7 ACTION AREA - FOOD ACCESS AND AVAILABILITY 1. Rationale for this action area • Eating healthier foods, especially more fresh fruits and vegetables, helps to reduce the risk of obesity and chronic disease (Temple, 2000; He et al, 2004). • Families and children from low-income communities and racial/ethnic minority backgrounds are less likely to have diets that meet nutrition guidelines for good health and are more likely to be obese. Inequalities in access to stores that stock healthy foods may contribute to these disparities. • The majority of studies that have examined the relationship between store access and dietary intake find that better access to a supermarket or large grocery store is associated with healthier food intakes (Larson et al, 2009). • The relationship between the availability of restaurants (both fast food and full-service) and dietary intake has also been studied. In general, these studies have found that greater availability of fast food restaurants and lower prices of fast food restaurant items are related to poorer diet. • Access to full service restaurants shows either no relationship or a positive relationship with healthy dietary intake. • Supermarkets are now the dominant setting for purchase of most foods. Studies for the USA suggest that obesity prevalence increased and fruit and vegetable consumption decreased with increasing distance to supermarket in metropolitan areas, but not in nonmetropolitan areas. Data from the UK does not support such an association. • Larger grocery shops generally have greater availability, lower costs and better-quality fresh produce than smaller grocery stores. Nevertheless, some small specialist stores (such as greengrocers and market stalls) appear to offer cheaper prices for fruits and vegetables than supermarkets (White et al, 2004). • Research consistently demonstrates that car ownership and use of a car to buy food is socioeconomically patterned and that this is a key determinant of access to food stores (Shepherd et al, 1996; Guy, 1996; Bromley & Thomas, 1993; Guy, 1985; Guy et al, 2004; Mooney, 1990; Donkin et al, 2000; Caraher et al, 1998; Lang & Caraher, 1998; Robinson et al, 2000; Clarke et al, 2002; Barratt, 1997; Lang, 1995; Guy & David, 2004). • Carrying shopping, as well as the problems of storage, remain important barriers to accessing supermarkets by specific socio-demographic groups, including older people, people without cars and in poor housing (Bromley & Thomas, 1993; Guy, 1985; Guy et al, 2004; Donkin et al, 2000; Caraher et al, 1998; Lang & Caraher, 1998; Lang, 1995; Guy & David, 2004). • While studies in the UK and Australia have shown relatively good supply of affordable healthy food in disadvantaged urban areas, a higher density of fast food outlets has also been consistently shown in poor neighbourhoods (Cummins & Macintyre, 2006). • Aboriginal people and Torres Strait Islanders face difficulties in accessing an adequate and healthy diet, particularly in remote areas of Australia. Only a very limited variety of food is offered in remote community stores, relative to larger rural towns and urban centres. Perishable items such as dairy foods, fruit and vegetables are frequently in short supply. Page 90 of 170

2. Proposed actions to improve access and availability of appropriate foods A number of options have been proposed to improve the access and availability to more nutritious foods, such as fruit and vegetables other fresh produce. A small number of options have also been proposed to address opportunities to reduce access and availability of food products that are more likely to contribute to weight gain, such as fast foods, snack foods and sweetened drinks. Table 15. Proposed actions to improve access and availability of appropriate foods Proposed actions

Type of intervention Targeted behavioural

Information sources

Enabling/ amplifying

Provide new food retail outlets



Systematic reviews, Robert Wood Johnson Foundation evidence summaries, single studies

Improve the quality of foods stocked at retail outlets



Observational reports from the USA

Improve the quality of options at prepared food outlets



Small trials in the USA



Reviews and evidence summaries

Increase sale of produce directly from growers



Farmers markets assessments from USA

Improve the affordability of appropriate low energy foods



European experiences, modelling studies, parallel data from other intervention areas

Improve the transport connections to quality retail food outlets



Limited observational

Reduce number and access to fast food outlets



Ecological, logic and extrapolated information

Improve distribution of fresh produce to regional/ remote locations



Limited evaluation observational data from Europe and Australia



Strong experimental data from USA

Improve access to plain drinking water



Food voucher schemes



Reduce availability of energy dense, nutrient poor foods in the home



Ecological associations, EUHOPE assessment

3. Summary of the evidence Few studies have evaluated strategies for improving access to healthy, affordable foods and reducing access to high-calorie, low-nutrient foods. Several strategies and actions have been proposed to attract supermarkets to underserved neighbourhoods, improve the availability of healthy foods such as fruits, vegetables and whole grain products, and reduce access to energy-dense foods in fast food establishments and restaurants.

Page 91 of 170

a. Open new supermarkets in poorly served areas i. Basis The availability of particular foods in a neighbourhood is associated with the reported consumption of such foods. Introducing new supermarkets to poorly served areas increases access to a wider variety of healthy food, allowing residents to improve their dietary consumption. The potential benefits of greater local fruit and vegetable availability may be especially pronounced for more disadvantaged households. ii. Evidence from trials Two UK studies (Leeds, Glasgow) have evaluated the impact of opening a new supermarket in an underserved location (Wrigley, 2003; Cummins et al, 2005). Results showed that shopping behaviour changed, but had only small impact on markers such as improved fruit and vegetable purchases (increased 1/3 serve) which disappeared when controlled for secular change. iii. Other evidence Studies in the USA showed that in metropolitan areas, the odds of obesity increased and the odds of consuming fruit and vegetables five times or more per day decreased as distance to supermarkets increased. This was not the case, however, in non-metropolitan areas (Michimi & Wimberly, 2010). Very little work has focussed on strategies to attract supermarkets into underserved neighbourhoods. A survey of city planners in 32 communities in the USA identified a range of options to attract supermarkets to underserved areas. However, only three cities reported successfully implementing systematic efforts to establish new supermarkets (Pothukuchi, 2005). iv. Comments Data from the USA shows that obesity among black and white Americans was associated with lower numbers of supermarkets and higher numbers of convenience stores in census tracts. However, there was no evidence that individuals shopped within their own census tracts, and these results may be confounded by the socioeconomic characteristics of neighbourhoods. b. Improve the quality of food in small stores in poorly served areas i. Basis In disadvantaged areas with distant supermarkets, small stores may act as a substitute for sourcing groceries and consequently may have an important role to play in influencing consumption. Stores in more deprived areas generally have low quality fresh produce compared to stores in more affluent areas. Food quality may influence food purchase and consumption and help partially explain neighbourhood differences in food consumption patterns. ii. Evidence from trials A number of small poorly controlled studies in the USA have examined the impact of reducing unhealthy and improving healthy food options in convenience stores (Ayala et al, 2009; Gittelsohn, 2009). One trial was able to demonstrate a one serve increase in fruit and vegetable consumption (over the controls), following the introduction of a range of fresh produce packs into selected stores. Page 92 of 170

iii. Other evidence The Remote Indigenous Stores and Takeaways (RIST) Project was established in 2005 by the SA, WA, NT, QLD, NSW and Australian Government Health Departments to improve access to a healthy food supply for Aboriginals and Torres Strait Islanders in remote communities. The project aims to establish and improve standards for ‘healthy' remote stores, and resources include checklists on healthy food supply, maximising the shelf-life of fruit and vegetables and a freight improvement toolkit. The RIST framework could be used to improve food quality in poorly served areas, as well as remote areas.

c. Increase access through farmers’ markets i. Basis Farmers’ markets are increasingly popular in Australia with at least 75 farmers’ markets in NSW alone. They have the potential to increase fruit and vegetable consumption by increasing availability in socially disadvantaged areas with poor access to healthy foods. ii. Evidence from trials A recent review of the nutritional impact of farmers’ markets showed that shopping at farmers’ markets generally increased consumption of fruit and vegetables (McCormack et al, 2010). However, the studies included in the review assessed the impact of the markets on fruit and vegetable consumption among participants involved in monetary incentive programs. To date, no studies have assessed the effects of a farmers’ market on diet without the use of incentives. iii. Other evidence Two studies in the USA have demonstrated that the Farmers’ Market Nutrition Programs for elders and low-income women in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) may lead to improvements in intake of fruits and vegetables (Johnson et al, 2004; Anderson et al, 2001). Community gardens and garden-based nutrition intervention programs may also have the potential to promote increased fruit and vegetable intake (Alaimo et al, 2008) and may increase willingness to taste fruits and vegetables among youth (Robinson-O’Brien et al, 2009). iv. Comments Low income programs in the USA (such as WIC) which have demonstrated benefits in intake of fruit and vegetables through farmers’ markets rely on the provision of vouchers for purchase of specified foods and its applicability in Australia may be limited. d. Reduce access to fast foods and vending machines i. Basis The rapid supply and demand for fast food is concerning, as those who consume large amounts of fast food have less healthy nutritional profiles (Bowman & Vinyard, 2004; French et al, 2001; French et al, 2000) and have a higher risk of adverse health outcomes such as weight gain (Prentice & Jebb, 2003; Binkley et al, 2000) and diabetes (Pereira et al, 2005). Reducing access to fast food and vending machines may reduce the likelihood of consuming energy-dense nutrient-poor foods. Page 93 of 170

ii. Evidence from trials While there is sufficient evidence for limiting fast food consumption, there is no evidence evaluating reduced access to fast food. Most of the literature regarding the impact of vending machines on health focuses on beverage consumption among school children. Saying this, there is limited evidence to suggest a relationship between school vending machines and health. A UK study found no link between consumption of confectionery purchased from vending machines and ‘poor’ dietary practice or ‘undesirable’ lifestyle habits (New & Livingstone, 2003). A study in the USA found that banning sweetened beverages in vending machines in a West Virginia school district did not significantly change the amount of beverages purchased at vending machines or in stores outside the school (Spangler, 2006). Forshee et al (2005) conducted a risk analysis of the relationship of beverage consumption from vending machines and BMI. The risk assessment showed no impact on BMI by removing soft drink consumption in school. More recently, an evaluation of the policy reducing access to school vending machines in the USA indicated that limiting access to soft drinks at school may not reduce children’s intake, as they may obtain soft drinks from alternate outlets (Fletcher et al, 2010). III. Other evidence Zoning and land use policies that regulate fast food restaurants may affect consumption (Ashe et al, 2003; Paquin, 2008). South Los Angeles has introduced regulations to prevent the establishment of any further fast food outlets in the district and will attempt to attract healthier options. However, there have been no evaluations of this or other measures to inhibit access to unhealthy foods. Arguments against limiting prevalence of fast food restaurants include the argument that access to healthier food alone will not prevent overeating and it may restrict competition (Lawrence & Gostin, 2007). Many school policies in Australia have adopted a traffic light classification system in which foods are categorized as green, amber or red according to their nutritional value (the red category differs from state to state, however it generally includes foods/drinks that are high in fat, sugar and sodium). Some states, including New South Wales, Queensland and Victoria have banned red category foods/drinks from being sold in school vending machines. Some argue that removal of vending machines may prove inefficient if habits do not change at home (Lawrence & Gostin, 2007). e. Improved quality of foods available at catering outlets i. Basis Catering outlets are among the most important and promising venues for environmental, policy, and pricing initiatives to increase intake of healthy food. Unfortunately however, away-from-home foods typically contain more fat and saturated fat and less fibre, calcium and iron than foods prepared at home (Lin et al, 1999). ii. Evidence from trials Although examples of increasing availability of healthy choices in restaurants have been reported (such as providing more types of fruit and vegetables as well as healthier preparations of fruit and vegetable menu options including non-fried choices and fruit and vegetables without high-fat sauces), there is limited evidence assessing the impact of improved quality on diet-related health outcomes. Page 94 of 170

iii. Comments The importance of providing healthy foods at catering outlets is highlighted by the fact that almost 67% of the 53 WHO European Region countries have formally acknowledged the catering sector as a stakeholder in national nutrition policies. Strategies are mainly directed towards labelling of foods/meals, training of health and catering staff, and advertising. Resources for healthy catering polices are provided by the Australian Heart Foundation and Dietitians Association of Australia.

f. Improved access to plain drinking water i. Basis Sugar-sweetened beverage intake is considered an important contributing factor to obesity in childhood (Ludwig et al, 2001; Vartanian et al, 2007; Malik et al, 2006; James & Kerr, 2005). Sugary drinks such as soft drinks have been banned from schools throughout Australia; however these restrictions could be implemented across other institutions e.g. hospitals, sporting venues etc. Saying this, it is difficult to directly influence access to soft drinks as they are widely available. An alternative strategy would be to improve access to other beverages such as water. ii. Evidence from trials A New Zealand intervention study found that children from intervention schools (provided with cooled water filters) reported consuming fewer carbonated beverages, fruit juice/drinks and total sweet drinks compared to controls (Taylor et al, 2007). However, the differences were mainly due to the increased consumption of soft drinks among the control group. Water consumption did not differ significantly between groups post-intervention and BMI was only reduced in those who were not overweight at baseline. It may be that simply increasing access to drinking water is not enough and that drinking water should be accompanied by a reduction in sugary drinks. In fact, replacing sugar-sweetened beverages with water was associated with reductions in total energy intake for children and adolescents (Wang et al, 2009). Installing water fountains in public places and facilities can increase water intake and prevent and reduce overweight and obesity (Muckelbauer et al, 2009). iii. Other information In Sweden it is compulsory to provide access to free water in all venues where food is served, and in certain states in Australia, it is mandatory to serve cold tap water either free of charge or at a reasonable price if the restaurant is licensed to serve alcohol (Department of Racing Gaming & Liquor, 2009; NSW Government Legislation Liquor Regulation, 2008) g. Transport or retailer subsidies to improve cost of fresh produce in remote regions. i. Basis

Providing access to fresh produce is costly in remote regions due to the energy consumed for transport and storage over long distances. In Australia, the price of food has been found to be related to remoteness, as measured by the Accessibility Remoteness Index of Australia (ARIA) (Lee et al, 2002). Page 95 of 170

ii. Evidence from trials An evaluation of the Canadian Food Mail Program (Madore, 2007), which subsidises the cost of transporting nutritious perishable foods to isolated communities, found that increasing the freight subsidy from 30 to 80 cents per kilogram for healthy products like fruits, vegetables and dairy as part of a pilot project in three communities resulted in the purchase of these products doubling. Recently this subsidy has shifted from transport to wholesale/retail support. iii. Other evidence Australia already has a remote area transport subsidy scheme but its impact on price and consumption of foods is not monitored. h. Food voucher schemes i. Basis Low-income individuals are more likely to consume nutrient-poor foods leading to disparities in health such as higher rates of hypertension, diabetes, cardiovascular disease, obesity, and dental disease. Providing monetary incentives to such individuals could encourage consumption of healthier food, thereby helping to reduce the large gap in health inequalities which exist between different socioeconomic groups. ii. Evidence from trials Interventions that provide coupons redeemable for healthier foods and bonuses tied to the purchase of healthier foods increase purchase and consumption of healthier foods in diverse populations, including university students, recipients of services from the Supplemental Nutrition Program for Women, Infants, and Children (WIC), and low-income seniors (Anderson et al, 2001; Jeffery et al, 1994; Cincirpini, 1984). For example, one communitybased intervention indicated that WIC recipients who received weekly $10 vouchers for fresh produce, increased their consumption of fruits and vegetables compared with a control group, and sustained the increase six months after the intervention (Herman et al, 2008). iii. Other evidence As mentioned above, low income programs in the USA (e.g. WIC) have demonstrated benefits through the provision of vouchers; however the applicability in Australia may be limited.

i. Mix public housing into well served neighbourhoods i. Basis Housing is one of the social determinants of health (Herman et al, 2008; CDC, 2005).Residential neighbourhoods are associated with the current well-being of residents. This is especially true for children - those who grow up in poorly served neighbourhoods fare substantially worse on a wide range of outcomes compared to those who grow up in more affluent neighbourhoods (Brooks-Gunn et al, 1993). An alternative to improving food availability in poorer areas is to move low income earners into well served areas.

Page 96 of 170

ii. Evidence from trials The Moving to Opportunity for Fair Housing Demonstration Program in the USA was able to demonstrate that those who were relocated from high poverty areas into better neighbourhoods significantly reduced their risk of developing obesity when compared to those who remained in the same housing or shifted to housing in other low income areas (Katz et al, 2007). iii. Other evidence Balancing the social mix of residents from across different housing tenures and income levels is a common theme of contemporary housing and planning policies in Australia. Concerns have been raised about the ability of the tenure diversification approach to benefit existing socio-economically disadvantaged residents. Arthurson (2001) notes that through the process of attracting more affluent residents to the area, the inequality of current residents can be added to rather than ameliorated. 4. Other initiatives proposed or implemented but not evaluated • Reduce access to fast food restaurants; • Improve transport options to supermarkets; • Local food production; • Establish cooperative grocery stores; • Connect growers with neighbourhood convenience stores, community centres, health care clinics and religious organizations; • Develop community gardens; • Establish mobile stores to deliver healthful products to a drop spot or doorsteps; • Adopt land use and zoning policies that restrict fast food establishments near school grounds and public playgrounds; • Implement local ordinances to restrict mobile vending of calorie-dense, nutrient-poor foods near schools and public playgrounds; • Implement zoning designed to limit the density of fast food establishments in residential communities; • Adopt building codes to require access to and maintenance of fresh drinking water fountains (e.g. public restroom codes); • Require that plain water be available in local government-operated and administered outdoor areas and other public places and facilities; and • Realign bus routes or provide other transportation, such as mobile community vans or shuttles to ensure that residents can access supermarkets or grocery stores easily and affordably through public transportation.

5. Quality and nature of evidence to support action in food access Most of the evidence in this action area is indirect and relates to behavioural outcome such as purchasing behaviour rather than specific weight-related outcomes. This is not surprising given that most interventions assessed within this action area are enabling or amplifying actions. There is very little experimental evidence or quasi-experimental studies to assess the impact of interventions proposed within this action area. Where there has been evaluation studies conducted they have been small, isolated and often difficult to interpret. The outcomes of most assessments have been process oriented or behaviours rather than specific weight-related measures. However, there is strong support for action in this area from theory, logic as well as program logic; especially as such Page 97 of 170

interventions are aimed at reducing inequities. Much of the work comes from the USA and it is likely that the findings are influenced by the structures and social programs that are unique to the USA. Table 16. Quality and nature of evidence to support action in food access Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action

Open supermarkets in poorly served areas

med

Direct

Indirect low

Improve quality of food in small stores in poorly served areas

low

med

Increase access through farmers markets

med

low

Reduce access to fast foods and vending machines

low

low

Improved quality of foods available at catering outlets

low

low

Improved access to plain drinking water

med

Transport/retailer subsidies in remote regions

low

Food vouchers

high

Mix public housing

low

low

low-med med

med low

6. Potential impact and sustainability The impact of many of the interventions to address food access and availability appears to be small at the whole of population level, but are likely to be much greater within disadvantaged communities. This may reflect the reasonably good quality of access and availability to healthy foods that many Australians already experience. However, such findings should not disqualify such interventions from consideration in a comprehensive package of action as a means of addressing high risk groups within the community. The only studies that have attempted to quantify the impact of interventions on weight status (such as the removal of soft drink vending machines) have demonstrated no or minor effect. The use of vouchers within specific disadvantaged groups and improving the quality of food in convenience stores shows the greatest promise in terms of positive dietary change, but this is difficult to quantify on the basis of existing evidence. Unevaluated interventions such as providing community transport or improving transport connections to food shops may prove to be more effective and cheaper alternatives, but unfortunately we currently have no information about their impact. Interventions to improve access and availability of quality food are likely to be relatively easy to sustain but unlike many other environmental interventions they require the continued input of resources and time to keep them operational. Page 98 of 170

As more and more individuals are eating outside the home, improving the quality of food sold at catering outlets is imperative and could have a substantial impact on diet. Unfortunately, to date, there has been little evaluation of the influence this strategy may have on diet. While farmers’ markets are increasingly popular in Australia and have great potential for increasing consumption of fresh, healthy food, they must be well regulated to ensure a variety of healthy food. 7. Other considerations a. Feasibility and acceptability to all stakeholders There may be moderate levels of support for many of the interventions aimed at improving access and availability as they can be viewed as a practical way of dealing with issues of inequality. However, as many of the interventions only provide benefit to disadvantaged sectors of society, the general population may be less supportive of the allocation of funds to serve just these groups. In addition, some of the interventions require investment of time and manpower of private businesses to improve the quality of foods provided. This may not be achieved without some financial incentives. b. Relevance to Australia (transferability) Issues such as the development of new supermarkets in disadvantaged areas, the use of food vouchers and the promotion of farmers’ markets all demonstrated significant impact in the USA, but have not shown similar returns in other areas. It is likely that the different social and support structures in the USA and reliance on food vouchers, which must be used at defined venues rather than untied welfare support, influence the success of these programs, but make them less immediately relevant to Australia. Currently there is no firm data to indicate that access to supermarkets is a problem in the Australian context. c. Cost considerations Much of the cost of improving food access and availability is currently borne by the private sector and even though initial costs for infrastructure such as supermarkets is high, the high return makes it feasible in urban areas but not in rural and remote areas. Australia is very urbanised, and providing transport and other subsidies to support food access in remote areas can be expensive per head of person served. Food vouchers also account for a considerable part of the total costs of the WIC program in the USA but could be issued in replacement of untied welfare support. d. Timeframes in implementation and impact Improving food access through changes within existing structures and services could be quite rapid and even the establishment of new supermarkets is relatively fast. Bigger time issues revolve around gaining wide stakeholder support. Once established, these new facilities or services would have a fairly rapid impact on food behaviours and intake. e. Equity issues As previously indicated many of these interventions are directly targeted at addressing current inequities in food access and availability rather than improving these factors for all Australians. Although Australia is highly urbanised, approximately one-third of the population still live outside the big cities, and a small proportion live in remote areas. Investments may need to be made within this action area to prevent worsening inequities even though they are hard to justify on a cost-effectiveness basis. Page 99 of 170

f. Other positive or negative side-effects Many Australian farms are struggling to survive, and increasing the distribution of farmers’ markets may not only help the farmer via increased profitability, recognition and social inclusion, but also the public through improved social cohesion, wellbeing and potentially health. Farmers’ markets have great potential in Australia to ‘re-connect’ people to food – a connection which has been almost entirely lost due to influences from transnational companies and aggressive marketing of energy-dense, nutrient poor foods. The introduction of supermarkets into a poorly served area may introduce healthy competition and as a result, already existing smaller stores may increase availability of affordable, healthy produce.

Page 100 of 170

References Alaimo K, Packnett E, Miles RA, Kruger DJ. Fruit and vegetable intake among urban community gardeners. Journal of Nutrition Education and Behavior 2008; 40(2):94-101. Anderson JV, Bybee DI, Brown RM, McLean DF, Garcia EM, Breer ML, Schillo BA. Five a day fruit and vegetable intervention improves consumption in a low income population. Journal of the American Dietetic Association 2001; 101(2):195-202. Arthurson, K. Achieving social justice in estate regeneration: the impact of physical image construction housing studies. Housing Studies 2001; 16(6):807-26. Ashe M, Jernigan D, Kline R, Galaz R. Land use planning and the control of alcohol, tobacco, firearms, and fast food restaurants. American Journal of Public Health 2003; 93(9):1404-1408 Ayala GX, Baquero B, Linnan L, Laraia BA, Bloom P. Working with tiendas to promote healthy eating. Presentation at the IOM/NRC workshop on the public health effects of food deserts. Washington DC: January 27 2009. Barratt J. The cost and availability of healthy food choices in Southern Derbyshire. Journal of Human Nutrition and Dietetics 1997; 10:63–69. Binkley JK, Eales J, Jekanowski M. The relation between dietary change and rising US obesity. International Journal of Obesity Related Metabolic Disorders 2000; 24(8):1032-9. Bowman SA, Vinyard BT. Fast food consumption of U.S. adults: impact on energy and nutrient intakes and overweight status. Journal of the American College of Nutrition 2004; 23(2):163-8. Bromley RDF, Thomas CJ. The retail revolution, the carless shopper and disadvantage. Transactions of the Institute of British Geographers 1993; 18:222–236. Brooks-Gunn J, Duncan GJ, Klebanov PK, Sealand N. Do neighbourhoods influence child and adolescent development? American Journal of Sociology 1993; 99(2):353-95. Caraher M, Dixon P, Lang T, Carr-Hill R. Access to healthy foods: part I. Barriers to accessing healthy foods: differentials by gender, social class, income and mode of transport. Health Education Journal 1998; 57:191–201. Centers for Disease Control and Prevention (CDC). Social Determinants of Health [internet]. 2005. Available from:http://www.cdc.gov/socialdeterminants/ Cincirpini P. Changing food selections in a public cafeteria: an applied behavioural analysis. Behavioural Modification 1984; 8:520–39. Clarke G, Eyre H, Guy C. Deriving indicators of access to food retail provision in British cities: studies of Cardiff, Leeds and Bradford. Urban Studies 2002; 39:2041–2060. Cummins S, Macintyre S. Food environments and obesity- neighbourhood or nation? International Journal of Epidemiology 2006; 35(1):100-4 Cummins S, Petticrew M, Higgins C, Findlay A, Sparks L. Large scale food retailing as an intervention for diet and health: quasi-experimental evaluation of a natural experiment. Journal of Epidemiology and Community Health 2005; 59(12):1035-1040. Page 101 of 170

Department of Racing Gaming & Liquor. Provision of free drinking water [internet]. 2009. Available from:http://www.rgl.wa.gov.au/ResourceFiles/Policies/Drinking_Water.pdf Donkin AJ, Dowler EA, Stevenson SJ, Turner SA. Mapping access to food in a deprived area: the development of price and availability indices. Public Health Nutrition 2000; 3:31-38. Fletcher JM, Frisvold D, Tefft N. Taxing soft drinks and restricting access to vending machines to curb child obesity health affairs 2010; 29(5):1059-1066 Forshee RA, Storey ML, Ginevan ME. A risk analysis model of the relationship between beverage consumption from school vending machines and risk of adolescent overweight. Risk analysis 2005; 25(5) French SA, Story M, Neumark-Sztainer D, Fulkerson JA, Hannan P. Fast food restaurant use among adolescents: associations with nutrient intake, food choices and behavioural and psychosocial variables. International Journal of Obesity Related Metabolic Disorders 2001; 25(12):1823-33. French SA, Harnack L, Jeffery RW. Fast food restaurant use among women in the Pound of Prevention study: dietary and demographic correlates. International Journal of Obesity Related Metabolic Disorders 2000; 24(10):1353–1359. Gittelsohn, J. Overview of Efforts to Change the Food Environment. Presentation at the IOM/NRC Workshop on the Public Health Effects of Food Deserts. Washington DC: January 27, 2009 Guy CM, Clarke G, Eyre H. Food retail change and the growth of food deserts: a case study of Cardiff. International Journal of Retail and Distribution Management 2004; 32:72–88. Guy CM, David G. Measuring physical access to ‘healthy foods’ in areas of social deprivation: a case study in Cardiff. International Journal of Consumer Studies 2004; 28:222–234. Guy CM. Corporate strategies in food retailing and their local impacts.A case study in Cardiff. Environment and Planning A 1996; 28:1575–1602. Guy CM. The food and grocery shopping behaviour of disadvantaged consumers: some results from the Cardiff. Transactions of the Institute of British Geographers 1985; 10:181– 190. He K, Hu FB, Colditz GA, Manson JE, Willett WC, Liu S. Changes in intake of fruits and vegetables in relation to risk of obesity and weight gain among middle-aged women International Journal of Obesity 2004; 28:1569–1574. Herman DR, Harrison GG, Afifi AA, Jenks E. Effect of a targeted subsidy on intake of fruits and vegetables among low-income women in the Special Supplemental Nutrition Program for Women, Infants, and Children. American Journal of Public Health 2008; 98:98–105. James J, Kerr D. Prevention of childhood obesity by reducing soft drinks. International Journal of Obesity 2005; 29:S54–S57 Jeffery RW, French SA, Raether C, Baxter JE. An environmental intervention to increase fruit and salad purchases in a cafeteria. Preventive Medicine 1994; 23:788–92. Johnson D, Beaudoin S, Smith LT, Beresford SA, LoGerfo JP. Increasing fruit and vegetable Page 102 of 170

intake in homebound elders: The Seattle Senior Farmers’ Market Nutrition Pilot Program. Preventing Chronic Disease 2004; 1(1):A03 Katz LF, Kling JR, Leibman JB. Experimental analysis of neighbourhood effects. Econometrica 2007; 75(1):83-119. Lang T, Caraher M. Access to healthy foods: part II. Food poverty and shopping deserts: what are the implications for health promotion policy and practice? Health Education Journal 1998; 57:202–211. Lang T. Food and poverty: beyond blaming the poor. Primary Health Care 1995; 5:10–16. Larson NI, Story MT, Nelson MC. Neighbourhood environments: disparities in access to healthy foods in the U.S. American Journal of Preventive Medicine 2009; 36(1):74-81. Lawrence O, Gostin JD. Law as a tool to facilitate healthier lifestyles and prevent obesity Journal of the American Medical Association 2007; 297:87-90. Lee AJ, Darcy AM, Leonard D, Groos AD, Stubbs CO, Lowson SK, Dunn SM, Coyne T, Riley MD. Food availability, cost disparity and improvement in relation to accessibility and remoteness in Queensland. Australian and New Zealand Journal of Public Health 2002; 26(3):266-272 Lin BW, Frazao E, Guthrie J. Away-from-home foods increasingly important to quality of American diet. Food & Rural Economics Division, Economic Research Service, US Department of Agriculture; 1999. Agriculture Information Bulletin No. 749. Ludwig DS, Peterson KE, Gortmaker SL. Relation between consumption of sugarsweetened drinks and childhood obesity: a prospective, observational analysis. The Lancet 2001; 357:505-8. Madore O. The impact of economic instruments that promote healthy eating, encourage physical activity and combat obesity: literature review. In Brief, Parliamentary Information and Research Service, Library of Parliament Canada; 2007. Available from: www.parl.gc.ca/information/library/PRBpubs/prb0634-e.htm Malik VS, Schulze MB, Hu FB. Intake of sugar-sweetened beverages and weight gain: a systematic review. American Journal of Clinical Nutrition 2006; 84(2):274-288. McCormack LA, Laska MN, Larson NI, Story M. Review of the nutritional implications of farmers' markets and community gardens: a call for evaluation and research efforts. Journal of the American Dietetic Association 2010; 110(3):399-408 Michimi A, Wimberly MC. Associations of supermarket accessibility with obesity and fruit and vegetable consumption in the conterminous United States. International Journal of Health Geographics 2010; 9:49 Mooney C. Cost and availability of healthy food choices in a London health district.Journal of Human Nutrition and Dietetics 1990; 3:111–120. Muckelbauer R, Libuda L, Clausen K, Toschke AM, Reinehr T, Kersting M. Promotion and provision of drinking water in schools for overweight prevention: randomized, controlled cluster trial. Pediatrics 2009; 123(4):e661-e667

Page 103 of 170

New South Wales Government Legislation. Liquor regulation- Drinking water to be available free of charge where liquor served [internet]. 2008. Available from: http://www.legislation.nsw.gov.au/maintop/search/inforce

New SA, Livingstone MBE. An investigation of the association between vending machine confectionery purchase frequency by schoolchildren in the UK and other dietary and lifestyle factors. Public Health Nutrition 2003; 6:497-504 Paquin S. Le zonage et la classification des usages de restauration et de commerces d’alimentation: une mesure d'urbanisme pour diminuer l'épidémie d'obésité? Canadian Journal of Urban Research 2008; 17(1):48-62. Pereira MA, Kartashov AI, Ebbeling CB, Van Horn L, Slattery ML, Jacobs DR Jnr, Ludwig DS. Fast-food habits, weight gain, and insulin resistance (the CARDIA study): 15-year prospective analysis. The Lancet 2005; 365(9453):36-42 Pothukuchi K. Attracting supermarkets to inner-city neighbourhoods: Economic development outside the box. Economic Development Quarterly 2005; 19(3):232–244 Prentice AM, Jebb SA. Fast foods, energy density and obesity: a possible mechanistic link Obesity Reviews 2003; 4(4): 187–194, Raphael D. Introduction to the social determinants of health. In Raphael D editor. Social Determinants of Health: Canadian Perspectives. 2nd ed. Toronto: Canadian Scholars' Press; 2008. p. 2-19 Robinson N, Caraher M, Lang T. Access to shops: the views of low-income shoppers. Health Education Journal 2000; 59:121–136. Robinson-O'Brien R, Story M, Heim S. Impact of garden-based youth nutrition intervention programs: a review. Journal of the American Dietetic Association 2009; 109(2):273-80 Shepherd R, Paisley CM, Sparks P, Anderson AS, Eley S, Lean MEJ. Constraints on dietary choice: the role of income. Nutrition and Food Science 1996; 5:19–21. Spangler JAL. Beverage vending purchasing patterns and attitudes in southwest Virginia high school students. Thesis, Virginia Polytechnic Institute and State University; 2006. Taylor RW, McAuley KA, Barbezat W, Strong A, Williams SM, Mann JI. APPLE Project: 2-y findings of a community-based obesity prevention program in primary school–age children. American Journal of Clinical Nutrition 2007; 86(3):735-742 Temple NJ. Antioxidants and disease: more questions than answers. Nutrition Research 2000; 20:449–59. Vartanian LR, Schwart MB, Brownell KD. Effects of soft drink consumption on nutrition and health: a systematic review and meta-analysis. American Journal of Public Health 2007; 97(4):667-675 Wang YC, Ludwig DS, Sonnevill K, Gormaker SL. Impact of change in sweetened caloric beverage consumption on energy intake among children and adolescents. Archives of Pediatrics Adolescent Medicine 2009; 163(4):336-343. White M, Bunting J, Raybould S, Adamson A, Williams E, Mathers J. Do food deserts exist? A multi-level, geographical analysis of the relationship between retail food access, socioPage 104 of 170

economic position and dietary intake. Final Report to the Food Standards Agency, UK. Newcastle upon Tyne: Newcastle University; 2004 Wrigley N. Deprivation, diet, and retail-food access: findings from the Leeds 'food deserts' study. Environment and Planning 2003; 35:151-188.

Page 105 of 170

4.8 ACTION AREA - FOOD PURCHASE AND CONSUMPTION 1. Rationale for this action area • Consumers make a number of important decisions regarding food products at the point of purchase or point of consumption. As a consequence, there is considerable scope to influence their behaviours at this time. • Australia has regulations that require all packaged food to provide certain nutrition information in the form of a Nutrition Information Panel (NIP). Recent research suggests consumers find NIPs confusing and difficult to interpret, and that they place little reliance on them. There is support from consumers for a simpler, clear representation of the nutritional value of food products (Cowburn & Stockley, 2005; Mhurchu & Gorton, 2007). • Australians spend around one third of household food expenditure on food consumed away from the home and in 2009, 3.7 billion meals were served by commercial foodservice outlets (BIS Shrapnel Pty Ltd, 2009). • Many people, even nutrition professionals, underestimate the number of calories in fastfood meals and when people eat out, they tend to consume more calories and fat, and fewer vegetables, fruits, and fibre (Rudd Center for Food Policy and Obesity, 2008). • Economic theory suggests that food prices affect food intake. Therefore, manipulating the price of food is likely to have both short- and long-term consequences for body weight (Goldman et al, 2009). • Some researchers suggest that the relative cheapness of high energy dense foods is a major driver of their consumption by less affluent consumers (Drewnowski, 2004). Conversely, the high relative cost of low energy dense foods, such as fruit and vegetables, is seen as a barrier to their increased consumption. 2. Proposed actions to improve access and availability of appropriate foods Because of the critical influence that the point of purchase decision has on food consumption, there have been a number of proposed strategies to address these issues and help prevent weight gain. However, very few of these interventions have been subjected to detailed assessment or evaluation. The main interventions for which there is evidence to assess their potential effectiveness at influencing energy balance, relate to food labelling and pricing of food or meals. Table 17. Proposed actions to improve access and availability of appropriate foods Proposed actions

Type of intervention Targeted behavioural

Menu labelling for takeaway and prepared food outlets

Enabling/ amplifying 

Limited observational assessments





Systematic reviews, evidence summaries from FSA, CDC, RWJ, WHO and several NGOs Some experimental studies





Nutrition profiling Front of pack labelling for packaged foods

Information sources

Evaluations from USA, evidence summaries

Page 106 of 170

3. Summary of the evidence This is a difficult action area in which to identify direct evidence of the effectiveness of the proposed approaches, as interventions are often implemented on an ad hoc and opportunistic basis, and rarely evaluated for their impact or outcome. Nutrition profiling is the science of classifying foods according to their nutritional composition and it underlies all forms of intervention in this action area. One use of nutrient profiling is to classify foods as healthy or less healthy by applying specific scientific based thresholds for nutrients of public health concern, such as total energy, saturated fat, sodium and added sugar. In Australia, the Food Standards Australia and New Zealand Nutrient profiling tool has been proposed for use in determining the appropriateness of foods to carry health claims (FSANZ, 2010). Nutrient profiling has potential to be used to inform the use of Front of Pack (FOP) labelling schemes or symbols, although there is little evidence regarding the effectiveness of their use. To be an effective system, nutrient profiling needs to account for both positive and negative nutrients, assessed per 100g or per 100mL (Arambepola et al, 2007; Drewnowski et al, 2009). A nutrient profiling model developed by Rayner and colleagues has been used in the UK to determine which foods are appropriate for marketing to children during children’s airtime (Ofcom, 2010). Food industries have also developed their own nutrient profiling systems, for example the Nestle Nutrition Profiling System (Nestlé, 2009). A limitation of any nutrient profiling system is that the models are based on single nutrients. This promotes the idea that processed and packaged foods are comparable to whole foods, such as fruits, vegetables, bread and cereals, low fat dairy and lean meats, as long as they meet specific nutrient thresholds.

a. Front of pack nutrition labelling i. Basis Given that we know Australian consumers have a poor understanding of using Nutrition Information Panels to make healthy food choices (Cowburn & Stockley, 2005; Mhurchu & Gorton, 2007), FOP labelling schemes provide a simplified consumer-friendly means of communicating a product’s nutritional composition to the consumer. Many different formats have been developed and tested, however they generally follow two key formats. The first is the traffic light scheme, whereby nutrient thresholds are applied for a subset of nutrients such as saturated fat, sodium total fat and total energy, to assess whether the content of that nutrient is ‘green’ (low), ‘amber’ (moderate) or ‘red’ (high). The second is a per cent daily amount scheme which presents the amount of a nutrient as a percentage of the daily recommended amount, e.g. if a product contains 800mg sodium, for an Australian adult this would be 35% of the upper level recommended intake. ii. Evidence from trials No trials of the impact of FOP labelling on community weight status were identified. Evaluations have been restricted to uncontrolled studies examining the impact of labelling on purchase behaviour.

Page 107 of 170

A small trial of traffic light labelling system in Australian consumers was shown to have no effect on purchasing of healthier products over a four week period (Sacks et al, 2009). This may be due to other factors that we know influence food purchasing behaviour such as taste, brand loyalty and family preferences (Grunert ,Wills et al, 2010). iii. Other evidence While evidence shows consumers prefer %RDI schemes, their ability to use these to select healthier foods is poor (Kelly et al, 2009; BMRB Social Research, 2009). Alternatively, the most effective FOP scheme has been a traffic light colour scheme that includes text: ‘high’, ‘medium’ and ‘low’ for a small number of nutrients (Kelly et al, 2009; BMRB Social Research, 2010). Consumers prefer the inclusion of saturated fat, total fat, sugars and sodium on FOP schemes (Kelly et al, 2009; Grunert, Fernández-Celemín et al, 2010). The inclusion of energy appears to be less valuable (BMRB Social Research, 2010). The use of FOP labelling to make purchase decisions has been estimated at 17-27% among consumers in the UK; however they are more likely to be used by individuals with existing nutrition knowledge and an interest in healthy eating (Grunert, Wills et al, 2010). Researchers have shown that the use of uniform FOP labelling can have indirect impacts on food supply, and lead to re-formulation or development of healthier products (Vyth et al, 2010). iv. Comments Researchers have found that consumers find the use of different FOP schemes on food packages, such as has occurred in the UK, to be confusing. It may be that the operation of competing schemes serves the industry better than consumers, and actually undermines consumer trust in such FOP schemes (BMRB Social Research, 2009; BMRB Social Research, 2010). Therefore, if a FOP labelling system was to be adopted, it must be universal across all food retail and service settings and across all product categories. One universal scheme would assist in educating consumers about nutrients and foods. A review of food labelling policy in Australia and New Zealand is currently being conducted by Dr Neal Blewett for the Australia and New Zealand Food Regulation Ministerial Council and is due to be finished in December 2010.

b. Point of purchase menu labelling i. Basis The prevalence of obesity has been associated with an increased consumption of energydense nutrient-poor foods outside the home (Bowman & Vinyard, 2004; Kant & Graubard, 2004). At the same time there have been increases in portion size, and ongoing underestimating of the calorie content of these foods (Variyam & Golan, 2002; Young & Nestle, 2002). ii. Evidence from trials While menu labelling has great potential to influence the energy consumed through takeaway and other food outlets, the evaluations of such programs have produced mixed results, with most of the studies of poor methodological quality. No studies have assessed the impact on weight status or used valid dietary methods to assess the impact of menu labelling on overall diet, although as previously indicated, this may be seen as an inappropriate outcome to assess especially in the short term. Most studies have examined recognition and use of the information, or the impact of providing menu labelling on intent to purchase foods and beverages. In general, the impact of menu labelling was modest and varied Page 108 of 170

across demographic groups, with the majority of studies showing some positive impact on intent or behaviour (NHFA, 2010; Clegg et al for BMRB Social Research, 2009). Preliminary evaluations of the major menu labelling initiatives in the USA, such as New York City, have been mixed, with initial studies indicating increased awareness but little behaviour change, and later studies suggesting small but significant reductions in calorie selection after labelling (RWJF, 2009; Elbel et al, 2009). Apart from fast food outlets, several studies have examined the impact of providing nutrition information on purchasing behaviour among patrons in worksite and university cafeterias and in restaurants. An assessment of 21 studies with varying levels of menu labelling found a positive impact in two thirds of those studies (Harnack & French, 2008). In addition, some experimental studies have examined the difference in menu orders made by participants who were provided with either a menu with nutrition labelling or one without. A study conducted in adolescents indicated that when exposed to menu labelling they modified their food choices, resulting in a substantial reduction in energy levels in the ordered meal (less 250 calories). However, two similar studies in adults produced variable results, with males increasing their calorie selection in one study (Cleg et al for BMRB Social Research, 2009). iii. Other evidence A health impact assessment using a simulation model was used to estimate potential impact of menu calorie-labelling on weight gain. Assuming that 10% of patrons at large-chain restaurants (20 or more facilities) would order reduced-calorie meals, with an average reduction of 100 calories/meal, 40.6% of the 6.75 million pound average annual weight gain in the country population (>5yrs old) would be averted (Kuo et al, 2009). Assessments of the impact of menu labelling on the quality of food products available after labelling suggest that there has been some product reformulation (Jargon, 2010) or changes in menu combinations (Harnack et al, 2008) to reduce calorie content on meals offered. iv. Comments Five states in the USA have enacted menu labelling legislation - California, Maine, Massachusetts, Oregon, and New Jersey. New York City (NYC) has schemes in operation or approved. The NSW and Victorian governments have both announced that menu labelling for kilojoules will be required in chain restaurants from 2011. c. Point of purchase supermarket signage i. Basis The idea to use supermarket shelf signage or (point-of-purchase signage) has been around for at least a decade, and supermarket shelf-tags have been used in many parts of the world to communicate nutrition information. ii. Effectiveness A range of studies have evaluated the impact of shelf-signage schemes on food selection behaviours. Although the study periods were long (1-2 years after introduction of the scheme) (Lang et al, 2000; Sutherland et al, 2010; Gittelsohn et al, 2009), outcomes assessed were restricted to purchase intentions or behaviours. An evaluation of the Guiding Stars Shelf-signage Scheme one year after its introduction in a small number of supermarkets in one region of Detroit, USA found that 28% of consumers Page 109 of 170

reported being aware of the signs, although of these only 56% reported using them to guide their food choices (Sutherland et al, 2010). Those who had been screened for chronic disease risk factors were more likely to use the signs to assist their purchasing decisions. Another tiered sign system was implemented across 168 stores in two US states and after two years a small but significant increase in the purchase of highest rated products (from 24.5% to 25.9%) was found (Sutherland et al, 2010). A third analysis measured changes in store sales compared to controls in a low-income, predominantly African American populated area (Gittelsohn et al, 2009). They found that the intervention stores were significantly more likely to report sales of the healthier products based on the presence of shelf-signage. A recent study in a USA college used point of purchase tagging to identify healthier options, and then used computerised sales data to assess its impact on purchases. It found promising but non-significant increases in purchases of the range of tagged items (Dumanovsky et al, 2010). iii. Other evidence There is some evidence that the use of these schemes can reduce market share of less healthy food and beverage products, and increase market share of healthier products (Schucker et al, 1992). This effect could help promote product reformulation and development. Including nutrition information on the shelf with pricing information may increase consumers’ use of nutrition information in purchasing decisions, although research is needed to confirm this. A barrier to introducing shelf-signage may be gaining support from the retail industry (O'Loughlin et al, 1996).

d. Taxes on energy dense, nutrition poor snack foods i. Basis Energy-dense, nutrient-poor foods are consistently cheaper in terms of energy content for a given price than less energy-dense and often more nutrient-rich foods (Drewnowski, 2004; Finkelstein et al, 2005). Taxing less healthy foods could create a financial incentive for consumers to avoid them. ii. Evidence from trials A range of reviews have been conducted on the effects of fiscal strategies to address consumption and weight status, although only a small number of the articles assessed refer to empirical studies and instead rely heavily on modelling. A recent review of 24 studies found that food taxes and subsidies generally influenced consumption in the desired direction in high-income countries, and that imposing substantial taxes on fattening foods may improve health outcomes such as body weight and chronic disease risk (Thow et al, 2010). Another review focussed purely on empirical studies and found some evidence for small effects on body weight that may be greater in low SES groups (Andreyeva et al, 2010). A recent experimental study, which was not included as part of the aforementioned review, showed stronger support for taxes than for subsidies as a means of reducing consumption of less healthy foods and increasing consumption of healthier alternatives (Epstein et al, 2010). Another study attempted to evaluate the impact of a snack tax in the US state of Maine, using obesity data obtained from the Behavioural Risk Factor Surveillance Survey but found no correlation, although this may be because the follow-up period was short (Powell & Chaloupka, 2009). A review of price reduction experimental studies found that they were effective at increasing intake of low fat snacks in the short-term, with price reductions of 10%, 25% and 50% on Page 110 of 170

lower fat snacks resulting in increased sales of 9%, 39% and 93% respectively, compared with usual price conditions (Oaks, 2005). iii. Other evidence USA State governments already target sales taxes of soft drinks and selected snack foods. Of importance to policymakers, recent surveys show that the public is willing to pay increased taxes, if the funds generated are used to address childhood obesity. Data on the effects of soft drink taxes is discussed below. iv. Comments Most assessments generally support current recommendations that taxes and subsidies should be included as part of a comprehensive strategy to prevent obesity. We have no measures of price elasticity of demand for food items within Australia and generally rely on data from the USA. A recent review indicated that price elasticity for foods and non-alcoholic beverages ranged from 0.27 to 0.81 (absolute values), with food away from home, soft drinks, juice, and meats being most responsive to price changes (0.7– 0.8), although there were variations in estimates (Freedman & Connors, 2010). Although demand for food is relatively inelastic, the power of small price changes, especially applied to foods most responsive to such changes, should not be underestimated given that their effects accumulate across a population. Recently the Danish government imposed tax increases of 25% on ice cream, chocolate, and sweets and will also increase taxes on soft drinks, tobacco, and alcohol products in a bid to reduce the burden on public health services and to tackle obesity, heart disease, and other illnesses. Part of the rationale is that higher taxes on foods will stimulate innovation to produce healthier products which avoid this impost. e. Specific soft drink (soda) tax i. Basis The consumption of sugar-sweetened beverages has been linked to risks for obesity, diabetes, and heart disease (Vartanian et al, 2007; Malik et al, 2006; Fung et al, 2009). They are a discrete class of beverage, with no significant contribution to nutrition and are often consumed in large amounts by groups at high risk for obesity. A specific soft drink tax has been proposed as a means of reducing the intake of these beverages (Brownell et al, 2009). ii. Evidence from trials Reviews and studies that have examined the impact of imposing a soft drink tax report mix findings. In 2006, Kim and colleagues found that US states without a soft drink (or snack food tax) were more than four times likely than other states with tax to undergo a high relative increase in obesity prevalence (Kim & Kawachi, 2006). Since 2006, the impact appears to have had little or no significant impact on obesity prevalence or BMI. Powell and colleagues showed no significant associations between state-level soft drink taxes and adolescents’ BMI (Powell et al, 2009). Fletcher and colleagues found that the impact of changes in US soft drink taxation from 1990-2006 had a statistically significant impact on behaviour and weight, however the magnitude of the effect was small (Fletcher et al, 2009). By assessing change in weight status from repeated cross-sectional surveys after the introduction of state-level taxes, they found that an increase in the soft drink tax rate of 1 Page 111 of 170

percentage point leads to a decrease in BMI of 0.003 points and a decrease in obesity and overweight of 0.01 and 0.02 percentage points, respectively. A year later, the same author found little difference in obesity prevalence in US states with a soft drink tax compared to those states without a tax (Fletcher et al, 2010). The same year, Sturm and colleagues also found that taxing soft drinks did not substantially affect overall levels of soft drink consumption or obesity rates (Sturm et al, 2010). However, certain subgroups may be more sensitive to these taxes than others e.g. those already overweight, lowincome, African-American. iii. Other evidence A lower level of increase in obesity prevalence was seen in those US states that had an existing soft drink tax, whilst those states that have repealed a soft drink tax were found to be at high risk of experiencing a high relative increase in obesity (French, 2003). As of January 2009, 33 states taxed the sale of soft drinks at an average rate of 5.2%. It has been suggested however that a multi-pronged approach is needed to reduce soft drink consumption among youths (Powell et al, 2009). iv. Comments In addition to any impact on obesity prevalence or BMI, it has been suggested that a greater impact could come from the dedication of the revenues they generate for other obesity prevention efforts and health related programs (Brownell et al, 2009; Sturm et al, 2010; Brownell & Frieden, 2009). f. Subsidies for fruit and vegetables i. Basis Economic theory suggests that if the price of a particular food decreases, purchase of that food will increase, and vice versa. Therefore, decreasing the price of fruit and vegetables may increase their consumption, particularly among low-income individuals. ii. Evidence from trials A recent review suggests that food subsidies can influence consumption in high-income countries and should be included as part of a comprehensive strategy to prevent obesity (including food taxes) (Thow et al, 2010). However, an experimental study not included as part of that review found that reducing prices of healthier foods significantly increased the amount of energy purchased (Epstein et al, 2010). On the other hand, they also found that, when prices of unhealthy foods were increased, there was a significant decrease in energy purchased. These results suggest that a subsidy for fruit and vegetables is unlikely to positively influence rates of obesity. iii. Other evidence As reported in the Food Supply action area, targeted subsidies for fruit and vegetables to low income groups through the WIC program in the USA, was successful in increasing their intake. Free fruit programs have also been shown to impact positively on children’s fruit consumption (Smed et al, 2007). In addition, providing transport subsidies to improve availability and reduce the cost of fresh fruit and vegetables in remote areas has been identified as a potentially effective strategy.

Page 112 of 170

iv. Comments Assessments of various tax and subsidy scenarios in Denmark have indicated that provision of subsidies for more nutritional foods, such as fruit and vegetables, may be necessary to overcome some of the potential detrimental effects of taxing high fat/high sugar foods. The mix of tax and subsidies should be carefully designed to shift consumption away from undesirable foods towards desirable ones (Kim & Kawachi, 2006). It must also be considered that whilst tax impositions are universally passed onto consumers, subsidies are often kept to improve profits and do not result in the same level of price reductions. 4. Quality and nature of evidence to support action on food access and availability There is a surprising amount of evidence around proposed interventions in this action area, although the quality of that evidence is mixed. Controlled trials of environmental, structural or fiscal policy change are extremely difficult to initiate at the community or population level and it is difficult to assess the direct impact of these interventions from observational studies because of the large number of potential confounders. However, investigators have attempted to overcome these problems by establishing small trials in a controlled but artificial environment, in order to assess the potential impact of a policy or environmental change if it were to achieve the full level of implementation. For example, researchers have artificially lowered the price of low fat snacks and food products in a workplace canteen or created dummy menus which contain the type of labelling proposed for fast food restaurants and assessed the response of consumers. In addition, there has been substantial use of modelling with real data inputs and quantitative calculations to assess the likely impact of policy changes on behaviour and weight status. This balance of evidence, together with theory, logic and additional observational data, allows a reasonable assessment of the potential impact of most interventions in this action area. Table 18. Quality and nature of evidence to support action on food access and availability Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action

Front of pack nutrition labelling

high

Direct med

Indirect high

Point of purchase menu labelling

med

low

med

Point of purchase supermarket signage

med

low

low-med

Taxes on energy dense, nutrient poor snack foods

med

med

med

Subsidies for fruit and vegetables

med

Soft drink tax

high

low low-med

med

Page 113 of 170

5. Potential impact and sustainability The evidence suggests that the direct impact of interventions in this action area on weightrelated behaviour and weight status is likely to be small. However, because these interventions often reach the vast majority of the population, and once implemented are usually sustained, the overall population impact is likely to be significant. However, some of the interventions (such as menu labelling) may take considerable time and may need to be combined with other strategies to make a measurable contribution to energy balance. Most of these interventions can be sustained with little further investment of resources post launch. In fact food taxes collected or subsidy schemes can be directed to ensure longevity of the programs. 6. Other considerations a. Feasibility and acceptability to all stakeholders Whilst many of these proposed interventions might be easy to implement, with little immediate reaction from all stakeholders, there are some which have already created controversy, debate and resistance. Increasing tax is likely to receive a poor reception from consumers and industry as no one likes price rises, although some of this concern could be mitigated by concomitant subsidies. Governments are also reticent to increase taxes or make tax regimes more complex and difficult to administer. Front of pack labelling and menu labelling have evoked resistance from the food industry, as they see them as a threat to their profitability and market share, or fear that more demanding government interventions might follow if they accede to these initiatives. Regarding labelling, the food industry will tend to favour their own internally developed schemes. However, there is good evidence that consumers need consistent labelling or signage schemes that they understand, in order for them to influence healthier food purchasing behaviours. b. Relevance to Australia (transferability) Very little of this work, with the exception of front of pack nutrition labelling interventions, has been undertaken in Australia, with most originating in the USA or Europe. Different tax systems, existing regulatory requirements and cost structures, may limit the direct transferability of overseas schemes to Australia. Many modelling exercises could be repeated using Australian conditions, but these may also require information about how local consumers behave and in particular the price elasticity of demand for key foods. c. Cost considerations All these initiatives will come at a cost and some of this cost will be borne by tax payers and some by private industry. However, the vast majority of these costs come with the establishment of the program, and once this is funded, less resources will be required to sustain them. Although industry may express concern over the cost of implementing front of pack or menu labelling, there is a clear indication that it is a cost they can bear because they have already implemented their own preferred schemes prior to being required to do so. Costs however, would be associated with providing essential consumer education to accompany the scheme. Regarding fiscal strategies, food taxation could help fund food subsidisation. d. Timeframes in implementation and impact Many of the fiscal strategies are likely to be quick to implement and may have immediate effects. However, there is some evidence to suggest that labelling and signage schemes Page 114 of 170

may take many years before a significant number of consumers regularly use them to purchase healthier food items. This is due to other factors that can influence purchasing behaviour such as taste, brand loyalty and family preferences. e. Equity issues There is some indication that snack food taxes, soft drink taxes or taxes based on individual nutrients will impact more on consumers on low incomes and will require some consideration of how this may be offset. Making food subsidisation coincide with food taxation may help ensure the disproportionate burden of food taxation on low-income individuals is reduced. Provided that any labelling and signage schemes are understood by lower-literacy consumers (such as the traffic light scheme), there should be no equity concerns with the use of these strategies. f. Other positive or negative side-effects As discussed, one study found that decreasing the price of healthy produce had a negative impact as it encouraged purchasing more calories. This would suggest that any subsidisation should coincide with a taxation of energy-dense nutrient-poor foods. Apart from any direct effect that soft drink taxes may have on intake, the money generated may also fund other obesity prevention strategies or health programs.

Page 115 of 170

References Andreyeva T, Long MW, Brownell KD. The impact of food prices on consumption: a systematic review of research on the price elasticity of demand for food. American Journal of Public Health 2010; 100:216-222 Arambepola C, Scarborough P, Rayner M. Validating a nutrient profile model. Public Health Nutrition 2007; 11(4):371-8. Bere E, Veierød M, BjellandM, Klepp K-I. Free school fruit—sustained effect 1 year later. Health Education Research 2007; 21(2):268-75 BIS Shrapnel Pty Ltd. Fast Food in Australia. Sydney, NSW; 2009. BMRB Social Research. Citizens' forums on food: Front of Pack (FoP) Nutrition Labelling. 2010. BMRB Social Research. Comprehension and use of UK nutrition signpost labelling schemes. 2009. Bowman SA, Vinyard BT. Fast food consumption of U.S. adults: impact on energy and nutrient intakes and overweight status. Journal of the American College of Nutrition 2004; 23(2):163-8. Brownell KD, Farley T, Willett WC, Popkin BM, Chaloupka FJ, Thompson JW, Ludwig DS. The public health and economic benefits of taxing sugar-sweetened beverages. New England Journal of Medicine 2009; 361(16):1599-605. Brownell KD, Frieden TR. Ounces of prevention-the public policy case for taxes on sugared beverages. New England Journal of Medicine 2009; 360(18):1805-8. Clegg S, Jordan E, Slade Z for BMRB Social Research. An evaluation of the provision of calorie information by catering outlets. London: Food Standards Agency; 2009. Cowburn G, Stockley L. Consumer understanding and use of nutrition labelling: a systematic review. Public Health Nutrition 2005; 8(1):21-8. Drewnowski A, Maillot M, Darmon N. Should nutrient profiles be based on 100g, 100kcal or serving size? European Journal of Clinical Nutrition 2009; 63(7):898-904. Drewnowski A. Obesity and the food environment: dietary energy density and diet costs. American Journal of Preventive Medicine 2004; 27(3 Suppl 1):154-62. Dumanovsky T, Huang CY, Bassett MT, Silver LD. Consumer awareness of fast-food calorie information in New York City after implementation of a menu labelling regulation. American Journal of Public Health 2010; 100:2520-2525 Elbel B, Kersh R, Brescoll VL, Dixon LB. Calorie labelling and food choices: A first look at the effects on low-income people in New York City. Health Affairs (Millwood) 2009; DOI: 10.1377/hlthaff.28.6.w1110 Epstein LH, Dearing KK, Roba LG, Finkelstein E. The influence of taxes and subsidies on energy purchased in an experimental purchasing study. PsycholScience 2010; 21(3):406-14. Finkelstein EA, Ruhm CJ, Kosa KM. Economic causes and consequences of obesity. Annual Page 116 of 170

Reviews of Public Health 2005; 26:239-57. Fletcher JM, Frisvold D, Tefft N. Can soft drink taxes reduce population weight? Contemporary Economic Policy 2009; 28(1):23-35. Fletcher JM, Frisvold D, Tefft N. Taxing soft drinks and restricting access to vending machines to curb child obesity. Health Affairs 2010; 29(5):1059-66. Food Standards Australia and New Zealand (FSANZ). Calculation method for determining foods eligible to make health claims: Nutrient profiling calculator. [cited 2010 September 1]; Available from: http://www.foodstandards.gov.au/_srcfiles/method%20to%20determine%20eligibility%20_fin al_.pdf. Freedman MR, Connors RA. Point-of-purchase nutrition information influences foodpurchasing behaviours of college students: A pilot study. Journal of the American Dietetic Association 2010; 110:1222-1226. French SA. Pricing effects on food choices. Journal of Nutrition 2003; 133(3):841-843 Fung TT, Malik V, Rexrode KM, Manson JE, Willett WC, Hu FB. Sweetened beverage consumption and risk of coronary heart disease in women. The American Journal of Clinical Nutrition 2009; 89(4):1037-42. Gittelsohn J, Song H-J, Suratkar S, Kumar MB, Henry EG, Sharma S, Mattingly M, Anliker JA. An urban food store intervention positively affects food-related psychosocial variables and food behaviours. Health Education and Behaviour 2009; [Epub ahead of print]. Goldman D, Lakdawalla D, Zheng Y. Food prices and the dynamics of body weight. Cambridge, MA: National Bureau of Economic Research; 2009. Grunert K, Fernández-Celemín L, Wills J, Storcksdieck genannt Bonsmann S, Nureeva L. Use and understanding of nutrition information on food labels in six European countries. Journal of Public Health 2010; 18 (3): 261 Grunert KG, Wills JM, Fernández-Celemín L. Nutrition knowledge, and use and understanding of nutrition information on food labels among consumers in the UK. Appetite 2010; 55(2):177-89. Harnack L, French S, Oakes J, Story MT, Jeffery RW, Rydell SA. Effects of calorie labelling and value size pricing on fast food meal choices: From an experimental trial. International Journal of Behavioural Nutrition and Physical Activity 2008; 5:63. Harnack LJ, French SA. Effect of point-of-purchase calorie labelling on restaurant and cafeteria food choices: A review of the literature. International Journal of Behavioural Nutrition and Physical Activity 2008; 5:51-56. Jargon J. Restaurants begin to count calories. The Wall Street Journal [internet]. 2010 January 22. Available from: http://online.wsj.com/article/SB10001424052748704381604575005530811257728.html Kant AK, Graubard BI. Eating out in America, 1987-2000: trends and nutritional correlates. Preventive Medicine 2004; 38(2):243-9. Kelly B, Hughes C, Chapman K, Louie J, Dixon H, King L. On behalf of a collaboration of Page 117 of 170

public health and consumer research groups. Front-of-pack food labelling: traffic light labelling gets the green light. Sydney: Cancer Council; 2008. Kelly B, Hughes C, Chapman K, Louie JCY, Dixon H, Crawford J, King L, Daube M, Slevin T. Consumer testing of the acceptability and effectiveness of front-of-pack food labelling systems for the Australian grocery market. Health Promotion International 2009; 24(2):120129. Kim D, Kawachi I. Food taxation and pricing strategies to "thin out" the obesity epidemic. American Journal of Preventive Medicine 2006; 30(5):430-437. Kuo T, Jarosz CJ, Simon P, Fielding JE. Menu labelling as a potential strategy for combating the obesity epidemic: a health impact assessment. American Journal of Public Health 2009; 99(9):1680-6. Lang JE, Mercer N, Tran D, Mosca L. Use of a supermarket shelf-labelling program to educate a predominately minority community about foods that promote heart health. Journal of the American Dietetic Association 2000; 100(7):804-8 Malik VS, Schulze MB, Hu FB. Intake of sugar-sweetened beverages and weight gain: a systematic review. American Journal of Clinical Nutrition 2006; 84(2):274-88 Mhurchu C, Gorton D. Nutrition labels and claims in New Zealand and Australia: a review of use and understanding. Australian and New Zealand Journal of Public Health 2007; 31:105 112. National Heart Foundation of Australia (NHFA). Rapid review of the evidence: The need for nutrition labelling on menus. NHFA; 2010 Nestlé.The Nestlé nutritional profiling system, its product categories and sets of criteria. [PDF Document] Vevey, Switzerland: Nestec Ltd; 2009 [cited 2009 December 17]; Available from: http://www.research.nestle.com/NR/rdonlyres/D70A5C47-0BC8-4B02-A65D7CDB846497D8/0/NutritionalProfilingTableFINAL.pdf Oaks B. An evaluation of the snack tax on the obesity rate of Maine. San Marcos: Texas State University, Department of Political Science; 2005 Office of communications (Ofcom). HFSS advertising restrictions: Final Review [internet]. July 2010: Available from: http://stakeholders.ofcom.org.uk/binaries/research/tvresearch/hfss-review-final.pdf. O'Loughlin J, Ledoux J, Barnett T, Paradis G. La Commande du Coeur ("Shop for Your Heart"): a point-of-choice nutrition education campaign in a low-income urban neighbourhood. American Journal of Health Promotion 1996; 10(3):175-8. Powell LM, Chaloupka FJ. Food prices and obesity: evidence and policy implications for taxes and subsidies. Milbank Q 2009; 87(1):229-57 Powell LM, Chriqui J, Chaloupka FJ. Associations between state-level soda taxes and adolescent body mass index. The Journal of Adolescent Health: official publication of the Society for Adolescent Medicine 2009; 45(3):S57-S63 Robert Wood Johns Foundation (RWJF). Menu labelling: does providing nutrition information at the point of purchase affect consumer behaviour? A research synthesis. RWJF; June 2009 Page 118 of 170

Rudd Center for Food Policy and Obesity. Menu labelling in chain restaurants: opportunities for public policy. USA: Yale University; 2008. Sacks G, Rayner M, Swinburn B. Impact of front-of-pack ‘traffic-light’ nutrition labelling on consumer food purchases in the UK. Health Promotion International 2009; 24(4):344-352. Schucker RE, Levy AS, Tenney JE, Mathews O. Nutrition shelf-labelling and consumer purchase behaviour. Journal of Nutrition Education 1992; 24:75-81. Smed S, Jensen DJ, Denver S. Socio-economic characteristics and the effect of taxation as a health policy instrument. Food Policy 2007; 32:624e39. Sturm R, Powell LM, Chriqui JF, Chaloupka FJ. Soda taxes, soft drink consumption, and children's body mass index. Health Affairs (Millwood) 2010; 29(5):1052-8. Sutherland LA, Kaley LA, Fischer L. Guiding Stars: The effect of a nutrition navigation program on consumer purchases at the supermarket. American Journal of Clinical Nutrition 2010; 91(4):1090S-4S. Thow AM, Jan S, Leeder S, Swinburn B. The effect of fiscal policy on diet, obesity and chronic disease: a systematic review. Bulletin of the World Health Organisation 2010; 88(8):609-14. Variyam JN, Golan E. New health information is reshaping food choices. Economic Research Service Food Reviews 2002; 25(1):13-8. Vartanian LR, Schwart MB, Brownell KD. Effects of soft drink consumption on nutrition and health: a systematic review and meta-analysis. American Journal of Public Health 2007; 97(4):667-75. Vyth E, Steenhuis I, Roodenburg A, Brug J, Seidell J. Front-of-pack nutrition label stimulates healthier product development: a quantitative analysis. International Journal of Behavioral, Nutrition and Physical Activity 2010; 7(1):65. Young LR, Nestle M. The contribution of expanding portion sizes to the US obesity epidemic. American Journal of Public Health 2002; 92:246-249.

Page 119 of 170

4.9 ACTION AREA - INTERVENING IN SOCIALLY DISADVANTAGED GROUPS (REMOTE, INDIGENOUS AND LOW INCOME GROUPS)

1. Rationale for this action area • Aboriginal and Torres Strait Islander (ATSI) communities are 1.2 times more likely to be overweight or obese (2004-05), compared with Australian population averages (ABS, 2008). Australians from low income households are more likely to be obese, compared with those from high income households (21% versus 16%); and those living in outer regional areas of Australia are more likely to be overweight or obese, compared with Australians from major cities (ABS, 2004-05). • Furthermore, 26% of Indigenous Australians reside in remote or very remote areas of Australia, compared with 2.3% for the Australian population average (ABS, 2010). • Healthy foods in remote areas of Australia have been shown to cost more and be more limited in availability (Burns et al, 2004; Harrison et al, 2007; Lee et al, 2002; Government of Queensland, 2006). • An assessment of 132 ATSI residences across Australia showed only 6% had adequate facilities to store, prepare and cook meals (Torzillo et al, 2008). • Disadvantage groups tend to view their health differently to the average Australian, and consider health a lower priority. • Lower income groups have been reported to be less responsive to community and schoolbased health promotion efforts (Robertson et al, 2007), hence require tailored or targeted intervention strategies. Note that while population based obesity prevention initiatives may also reach disadvantaged population groups, this summary focuses on targeted interventions aimed at preventing overweight and obesity among ATSI populations, those residing in remote areas of Australia and those with low household income.

Page 120 of 170

2. Proposed actions to address socially disadvantaged groups Table 19. Proposedactions to prevent overweight and obesity among socially disadvantaged groups (i.e. ATSI, Remote, and low-income groups) Proposed actions

Type of intervention Targeted Enabling/ behavioural amplifying 

Food vouchers and subsides on healthy food Cooking and budgeting skills Community healthy weight programs Community food gardens or growers markets Free school breakfast/lunch/fruit/veg programs

Information sources

Healthy Food Basket Surveys from NSW, QLD, NT, TAS, WA and VIC. Literature regarding efficacy of taxes/subsidies.

   

Norwegian Free Fruit at School program, and the Looma free school breakfast and lunch scheme for ATSI children.

3. Summary of the evidence Evaluations of strategies or programs for disadvantaged groups are largely based on uptake and participation in strategies, or process evaluation to improve the program. Many of the strategies for disadvantages groups are aimed at increasing access to fresh healthy foods. To support these structural changes, other strategies including education and promotions of healthy eating and physical activity (e.g. cooking classes, budgeting skills, exercise groups) need to be conducted in parallel. a. Food vouchers and subsidises for healthy foods i. Basis People in remote or socially disadvantaged areas may experience reduced access to healthy and affordable food (Burns et al, 2004). Given that large numbers of Aboriginal and Torres Strait Islanders reside in remote communities, this also places the Indigenous population at an increased risk. ii. Evidence from trials International initiatives The US Women, Infants and Children (WIC) program provides supplemental nutritious foods, nutrition education and counselling at WIC clinics, and screening and referrals to other health, welfare and social services (Food and Nutrition Service US Dept of Agriculture, 2005). The WIC program and its many components have undergone considerable evaluation with most reviews indicating significant success against its objectives (Food and Nutrition Service US Dept of Agriculture, 2010). Until recently the food package of WIC has focussed on ensuring nutritional adequacy of mothers’ and children’s diets through improving access and availability of healthy foods but recently it has also focussed on obesity and chronic disease prevention. Evaluations of impacts in these areas are currently being undertaken. However, the benefits that WIC delivers in improved birth outcomes, pregnancy status and early nutrition are likely Page 121 of 170

to have significant effect on later health and obesity in the children. Evaluation of the WIC pregnancy and postpartum program found those enrolled delivered infants with higher mean birth weights in a subsequent pregnancy than women who received WIC prenatally only; had better breastfeeding rates, and had lower risk of maternal obesity at the onset of the subsequent pregnancy (Caan et al, 1987). In the USA, food stamps to subside the cost of healthy foods for low-income families have been proven to be effective in increasing purchasing of healthier foods with significant improvements in fruit and vegetable intake (Lin & Guthrie, 2007). Ideally, fiscal strategies on taxation of less healthy foods and subsidies for healthier foods should occur concurrently to minimise the disproportionate impact of food taxes on low-income groups (see Food Purchase and Consumption Action section). The Convenience Store program in North East England is aimed at improving the availability and sales of fruit and vegetables in low income areas (Kuipers, 2010). Local grocery stores in low income areas in the North East of England have used the funding to improve store equipment and staff training in storing, selling and minimising waste of fruit and vegetables. Fruit and vegetables are also promoted to the community within stores by educating staff on the nutritional benefits of fruit and vegetables to pass on to customers and offering free local cookery classes. The program commenced in November 2008 and initial impact assessment shows sales of fruit and vegetables have increased by between 30 - 300% in every participating store. Supporting groups to run their own food co-operatives have been implemented through the Edinburgh Community Food Initiative and the Community Food Co-operative programme, yet only the latter has been evaluated (Kuipers, 2010). The evaluation found people in the most deprived areas of Wales who bought from the food co-operative were eating more fruit and vegetables, had improved social connectedness and perceived improvement to their health. National initiatives Increasing the variety and access to healthy foods was achieved through local policy in grocery stores of the Arnhem Land community in the Northern Territory (Lee et al, 1996). The policy increased consumer purchasing of the healthier foods on offer, although this was more common within community grocery stores that actively promoted the healthier options (Lee et al, 1996). A more recent example is the Remote Indigenous Stores and Takeaway (RIST) project which aims to improve access to healthy food in remote Indigenous community grocery and takeaway stores. Project resources include guidelines, marketing ideas and optimal storage tips for healthy food in remote community stores, and a toolkit to improve the freight transport of healthy foods to remote stores. Results from a six-month pilot of the project in seven remote communities found the project effectively increased purchasing of fruit and vegetables (as in the case of the Arnhem Land project) particularly in those communities where the policy is supported and endorsed by the community (Northern Territory Government, 2008). Each Australian state and territory currently has their own implementation strategy for this project and Queensland Health is funding a state-wide evaluation of the project implementation and resources. iii. Other evidence Subsidised fruit and vegetables have also been provided through the Fighting Disease with Fruit program whereby families pay $5 and receive $40 worth of fruit and vegetables, with the remaining $35 subsidised by the Aboriginal Medical Service (Jones & Smith, 2007). The Page 122 of 170

program included serving children fruit at school daily and this action resulted in improved vitamin C status among children with existing vitamin C deficiency (Jones & Smith, 2007). iv. Comments Providing subsidies to encourage the consumption of increased fruit and vegetables is a common strategy but it is rarely evaluated to detect changes in behaviour. Many of the schemes are tied to farm surpluses or production programs. b. Budgeting and cooking skills i. Basis Cooking skills are a fundamental part of dietary education, particularly for disadvantaged groups who may lack the skills or knowledge to prepare and cook healthy meals. For example, among a sample of 462 Brisbane households, the main person who prepares food was more likely to report a lack of confidence to cook vegetables if they had a lower household income; lack of confidence was also associated with being less likely to purchase a greater variety of vegetables on a regular basis (Winkler & Turrell, 2010). ii. Evidence from trials Budgeting for grocery shopping has been incorporated into an existing course for people in financial debt in the Netherlands, through the Healthy Food Doesn’t Have to Cost Much project (Kuipers, 2010). During group meetings, topics included how to deal with debts, purchasing and budgeting, supermarket tour, saving money and healthy and inexpensive food. The intervention directly resulted in reduced saturated fat consumption and increased fruit juice consumption, but no effect of fruit and vegetable intake. The supermarket tours offer a practical activity that is preferred by low-SES groups. iii. Other evidence FOODcent is a program that has been used for low-income groups to develop skills in food budgeting, preparation, safety and cooking, and has been adapted for ATSI communities (North Metropolitan Area Health Service, 2006). This program has not been formally evaluated in terms of impact on dietary intake and nutrition status, although researchers have shown positive changes in self-reported dietary, cooking and shopping behaviours among a sample of largely low-income families (Foley & Pollard, 1998). This program uses a train-the-trainer approach and the program benefits if those delivering the program are also of low-income status. The Cooking Classes for Diabetes Program consisted of 18 weekly sessions for the ATSI community and was run in Western Sydney between 2002 and the end of 2006 (AHMRC, 2009). Sessions had participants cook meals and conduct discussions on healthy eating and food safety. Participants received a cookbook containing successful recipes from the sessions and recipes contributed by participants. Changing family eating behaviours and the high cost of fresh foods were identified as barriers by participants. A contributor to the success of the program reported by participants was running the programs through the Aboriginal Medical Service (AMS). Other similar cooking class programs have been reported to improve participants’ food security through developing cooking, shopping and budgeting skills as well as reducing social isolation (Thorpe & Brown, 2008). Cooking Your Way to Health runs activities for 7-11 year olds in 14 different primary schools in Ealing Borough, England (Kuipers, 2010). Primary evaluation showed the clubs were sustainable, and effective in improving knowledge of how food can affect health and helping Page 123 of 170

people improve their diet, taking into account barriers including socio-economic factors, knowledge and skills. iv. Comments Food cost surveys around Australia have reported an increasing cost of healthy foods such as fruit and vegetables, and a significantly increased cost of healthy foods in remote areas of Australia (Innes-Hughes et al, 2010). This disparity works against key nutrition messages of national social marketing campaigns such as Go For 2 & 5 or Measure Up and contributes to healthy inequities experienced by those living in remote areas of Australia, where a greater proportion of ATSI persons reside. c. Community gardens i. Basis Community gardens permit community members to develop the skills and knowledge to grow and harvest their own fresh produce, especially if conducted with an education component. ii. Evidence from trials A community garden project Growing in Confidence was set up in two sites in North West Ireland, providing information based two hour sessions that run over eight months to teach disadvantages people about growing, preparing and consuming their own fresh food (fruit, vegetables, herbs etc.) (Kuipers, 2010). As a result of attending the program, fruit and vegetable consumption increased, and 86% of participants reported eating more healthily, while 83% reported growing their own fruit and vegetables. iii. Other evidence Community gardens have had positive results among Aboriginal communities including selfreported improvements in access to fruit and vegetables, as well as horticultural skill development, employment opportunities, self-esteem and social interaction (Thorpe & Browne, 2008). iv. Comments Apart from potentially increasing user’s consumption of fresh foods, community gardens also foster social bonds that may positively influence psychosocial health (Teig et al, 2009). A survey of community gardens in New York, USA found they were four times more likely to occur in low-income neighbourhoods, where the main reason for participation in community gardens was having access to fresh foods (Armstrong, 2000).

d. Community healthy weight programs i. Basis Numerous researchers have reported a key to improving the health status of the Australian Indigenous community or low income communities is by working with these communities and having them identify their needs and contribute to developing healthy weight programs for themselves. Developing inter-sectoral partnerships with key community stakeholders is also important to support and sustain community based programs. Page 124 of 170

ii. Evidence from trials International studies NRG-Doit is a Dutch program specifically targeting adolescents with lower socio-economic and educational level (Singh et al, 2009a). The program involved education and environmental changes to provide physical activity sessions outside school hours and changes in and around school cafeterias. The lifestyle behaviours targeted by the intervention included consumption of sugar-containing beverages and high-energy snacks, physical activity and screen-viewing. After 20 months, the intervention was successful in preventing unfavourable increases in skin fold thickness measures and in reducing sedentary screen viewing time for boys. At 12 months, students reported a significantly lower intake of sugar-sweetened beverages. Both teachers and students rated this program positively and most teachers reported they would implement the program again in the future (Singh et al, 2009b). Using sportspersons to promote physical activity and healthy eating among low income communities has been shown to be effective among children from low income families in the Netherlands through the Scoring for Health program. Children sign a lifestyle contract with one of the footballers’, meanwhile physical activity and healthy nutrition are promoted through participating schools. As a result, participating children increased their compliance with physical activity recommendations and improved their nutrition. There was a decrease in number of children drinking soft drinks every day, and eating more than three snacks a day and a significant increase in fruit and vegetable consumption. The Footballer’s popularity with children is a likely factor in the program’s success. National studies The Healthy Weight Program ran over 50 programs with ATSI groups in Queensland between 1996-99 (Dunn & Dewis, 2001). A series of programs were conducted by trained facilitators and involved label reading, being active, low fat cooking, budgeting and healthy eating information. Follow-up measures after eight weeks for weight, and waist and hip circumference were collected for 91 of the 260 participants, and found most of these participants lost weight (66% females; 77% males) and decreased waist measurements (69% of participants). However, limitations included large drop-out rates, facilitators found participants lacked motivation, lacked transport to get to sessions and had difficulty finding time to run the programs after work. Physical activity sessions run through schools for rural and remote children, with or without an education component, have been shown to be successful in reducing Body Mass Index (BMI), although multi-faceted and multi-component interventions to address the social determinants of obesity are encouraged (Clark et al, 2009). A further example of community based interventions where the success was at least in part due to community ownership and action is the Looma Healthy Lifestyle program for the ATSI community in Looma, QLD (Rowley et al, 2000). The program also involved a number of key stakeholders (Aboriginal Health Workers, grocery store managers, and local council and community groups). While the intervention began as a diabetes Program, engaging the community to further develop the program and implement activities within their own community occurred along with structural changes such as appointing a Sport and Recreation officer and a community member as a store manager with mandate to improve food quality, ensuring the availability of wholemeal breads and flours and margarines in grocery stores (Rowley et al, 2000). The program was successful in lowering the prevalence of hypercholesterolemia, but did not influence the increasing prevalence of diabetes or obesity (McDermott et al, 2000). Page 125 of 170

The Gutbuster program for males in the Torres Strait was successful after 12 months in reducing body weight, waist circumference and total fat mass (Egger et al, 1999). The program was designed to be adapted, implemented and delivered by participating communities, although this program identified that the use of an outside source of health information was better received than a known community member who acted as a trained ‘expert’ (Egger et al, 1999). The “lifestyle” Diseases and Diabetes Management and Care Program was introduced in four remote communities in the far north of Western Australia (WA), of which many were of low SES status, in an attempt to prevent chronic diseases such as obesity and cardiovascular disease through community-based lifestyle modifications (Gracey et al, 2006). A key feature of this program was that it was driven and supported by the Aboriginal communities. The program resulted in increased knowledge about nutrition, exercise and disease, altered attitudes and behaviours toward diet and exercise, improved weight control and physiological markers. In one of the communities in the Kimberly region, after seven months 49% of participants had lost weight, waist circumference decreased in 60% of diabetics, 54% had improved HbA1c results and 59% had lowered total plasma cholesterol levels (Gracey et al, 2006). The Walkabout Together Program in Townsville, QLD, was established due to concerns over the disproportionately high prevalence of obesity and diabetes among the Townsville Indigenous community (Heath et al, 2006). The program involved a baseline assessment with general practitioner (GP) and regular follow-ups with health workers, a GP and dietician. After 12 months the mean weight of participants (n=126) had decreased by 1.9kg and waist circumference by 2cm. iii. Other evidence The Healthy Beginnings Trial aims to prevent the early onset of childhood overweight and obesity through home visits that promote healthy feeding behaviours, physical activity, less sedentary behaviour and enhanced parent-child interaction. The intervention involves eight home visits from a specially trained community nurse, along with telephone support and is currently being conducted with new mothers in the most socially and economically disadvantaged areas of Sydney, Australia (Wen et al, 2007). The program is currently being evaluated. A randomised trial specifically targeting low-income groups in the UK, assessed the effect of receiving a behavioural or nutrition counselling to increase fruit and vegetable consumption (Steptoe et al, 2003). Behavioural counselling increased mean intake of fruit and vegetables more than nutrition counselling, however the sample size was very small (n=40) and the participants at baseline consumed more fruit and vegetables than what is usually found among lower income groups. The Expanded Food and Nutrition Education Program conducted in the US aims to assist low-income youth and families to acquire practical food knowledge, skills, attitudes and behaviour change (including money management and getting the most from health assistance programs). The program is delivered by volunteers with or without some form of nutrition training. The program has been shown to influence a range of food practices (including food budgeting, food safety and food preparation). Cost–benefit analysis suggests that for every US $1 invested in the program, between US $2.48 to $10.64 in reduced healthcare costs can be expected (Rajopal et al, 2002).

Page 126 of 170

iv. Comments Studies in Denmark and Finland have shown pregnancy may be a life-stage when lifestyle interventions to prevent weight gain are likely to be successful because women are highly motivated to provide the best care for their infant (Kinnunen et al, 2007; Polley et al, 2002). The WAVES activity program evolved from an existing women’s support group and classes were held at a hydrotherapy pool at a community health centre. The women in the classes organised Aboriginal walking classes as a flow on effect from the WAVES program. The WAVES program has received a number of awards, acknowledging its success. The effectiveness of school-based programs in low-income areas may be increased by linking them to out-of-school activities through the family and community, and focusing on a health promoting environment in the school (Robertson et al, 2007).

e. Free school breakfast/lunch/fruit/vegetable programs i. Basis Provision of free food at school and other venues is seen as a certain means of exposing children to fresh foods as well as assisting people to limit the number of high energy snacks. ii. Evidence from trials The Norwegian Free Fruit at School programme has been proven effective and sustainable over three years in increasing children’s fruit and vegetable intake (Bere et al, 2007). In this program children are provided with a piece of fruit or a vegetable at school, at no cost to the school or parents. The program was also effective for low-income families and those with low baseline habitual fruit and vegetable intake. iii. Other evidence Food provision was another component of the Looma Healthy Lifestyle program. School canteens provided healthy school lunches (wholemeal salad rolls with cold meat or cheese, fresh fruit and fruit salad, yoghurt and fruit juice) and a low cost breakfast (cereal, sultanas, milk, and a piece of fruit and fruit juice) for children (O’Dea, 2005). Together these two meals provided most of the recommended daily intake for children at the time. iv. Comments While school nutrition policies that include breakfasts, vending machines, snacks and meal services have not been shown to prevent weight gain, they can be effective in improving dietary patterns in disadvantaged school-children and adolescents (Robertson et al, 2007). Sustainability will depend on community support for the provision of foods to their children at school. 4. Other initiatives proposed or implemented but not evaluated • • • •

Developing school fund-raising programs focused around cooking and fresh produce in disadvantaged areas Special sports competitions in indigenous communities Use of sportspeople as opinion leaders on healthy diet and physical activity Training of local weight counsellors. Page 127 of 170

5. Quality and nature of evidence to support action within socially disadvantaged groups This is an important area where there have been limited evaluated programs at this time. As a consequence, the amount and quality around most of these interventions is low. However, many programs currently being implemented are evaluating their impact using experimental designs. Table 20. Quality and nature of evidence to support action within socially disadvantaged groups Intervention

Level of confidence in the evidence

Potential contribution to the achievement of energy balance as part of a program of action* Direct

Subsidising healthy food

med-high

Cooking and budgeting skills

Indirect high

med

Community healthy weight med med programs Community food gardens or low growers markets Free school breakfast/ med low lunch/fruit/veg programs * within disadvantaged communities specifically rather than general population

low-med

low low-med

6. Potential impact and sustainability Interventions in disadvantaged groups are more likely to be successful if operated at a local community level, and national programs, mass media and social marketing campaigns are less likely to effectively engage disadvantaged groups (Kuipers, 2010). A key to the success and ongoing sustainability of many of the community initiatives for disadvantaged communities is the involvement of community members from development and implementation, to the evaluation (Kuipers, 2010). Inter-sectoral partnerships between key stakeholders in obesity prevention may help support these community initiatives locally. The available literature indicates that education alone is not sustainable and structural changes to the physical and political environments of these communities should be implemented to complement educational activities or behavioural interventions. The adoption of local ATSI trained facilitators to deliver the interventions for this community has also been identifies by many researchers as important (COO, 2005). To ensure programs do not proceed too quickly within the community, a dedicated Aboriginal Health Worker in the community is essential for programs to achieve desired results (COO, 2005). Creating and supporting ATSI positions in nutrition and health within local communities has been identified as a key area for action, although this may require ongoing support for the position and professional development (Thorpe & Browne, 2008). Healthy weight programs for ATSI communities have been shown to be successful in improving dietary intake, physical activity and are more sustainable when the community themselves control and maintain ownership of the intervention program (Rowley, 2000; COO, 2005; Lee et al, 1995). In the case of Lee 1995, their program with the Minjilang Page 128 of 170

community of the Northern Territory was implemented at the request of the community themselves (Lee et al, 1995). This program also showed sustained weight loss over 12 months. Sustained weight loss among ATSI populations has only been shown in the Gutbuster program for Torres Strait males (Egger et al, 1999), and the ATSI program at Minjilang (Lee et al, 1995). Most programs have shown improvements in self-reported dietary intake or activity and increased purchasing of healthy foods as measured by store-turnover (COO, 2005). Focusing physical activity interventions on or around sport seems to be particularly useful for men. Screening and feedback of the results can act as a stimulus to the community for action. Also, there was generally a good response to being asked questions /interviewed about needs – people like to know someone else is interested in them.

7. Other considerations A limitation to the literature in this field is that many initiatives have not been well evaluated (COO, 2005). Others have reported the barriers to conducting more rigorous evaluations of Indigenous nutrition and physical activity programs to include the perceived inappropriateness of nutrition surveys, lack of community interest in participating as a control group, difficulty following-up participants, and lack of evaluation expertise among Aboriginal Health Workers (Thorpe & Browne, 2008). a. Feasibility and acceptability to all stakeholders There are always some issues around ownership and participation that need to be considered when working with indigenous and disadvantaged groups. Failure to plan and implement sensitively may cause resistance from within disadvantaged groups. b. Relevance to Australia (transferability) The main source of evidence around food vouchers and subsidises is the US WIC program. Currently the structure does not exist in Australia to drive the establishment and maintenance of similar programs. Most of the evidence for community gardens and food cooperatives has come from Ireland where these initiatives are popular. This may be due to a culture of supporting locally grown produce that may not be so apparent in Australia. c. Cost considerations Schemes to subsidise the cost of healthy foods in lower income areas could be achieved through taxations on energy-dense, nutrient-poor foods such as soft drinks. d. Timeframes in implementation and impact Most healthy weight programs for ATSI groups only have short periods of evaluation where a weight loss may have been achieved for the first two or three months. e. Equity issues These strategies have been established to specially deal with existing inequalities. f.

Other positive or negative side-effects.

Initiatives to address community health issues such as community gardens, cooking programs and school breakfast programs, aid the development of community cohesion and self-determination, especially when initiated from within the community. Page 129 of 170

8. Key recommendations for policy or practice for obesity prevention for disadvantaged groups in Australia National Preventative Health Taskforce, 2009  Support ongoing research on effective strategies to address social determinants of obesity in Indigenous and low-income communities.  Develop tailored approaches and services to reach Indigenous and low-income groups, particularly through partnerships with local governments that focus on obesity-promoting environments and mobilise programs in schools and other community settings. NSW Centre for Overweight and Obesity, 2005 (ABTSI communities specifically):  Successful interventions occur when they are a response to the expressed needs of the community rather than a response to the health agency’s agenda.  Community involvement from the beginning leads to empowerment. A flexible and relaxed approach is necessary.  Community relevance, acceptance and accessibility are paramount.  Community ownership and management of the programs are generally seen as particularly important.

Page 130 of 170

References Aboriginal Health and Medical Research Council of NSW (AHMRC). 10 out of 10 deadly health stories: Fruit and vegetable program and market garden. AMS Grafton: Australian Department of Health and Ageing; 2009. Available from: http://www.ahmrc.org.au/Downloads/Publication/10outof10healthstories.pdf. Armstrong D. A survey of community gardens in upstate New York: implications for health promotion and community development. Health & Place 2000; 6(4):319-27. Australian Bureau of Statistics (ABS). Australian Social Trends. Canberra: 2010. Cat. no. 4102.0. Viewed 2010 December 8. Available from: http://www.abs.gov.au/ausstats/[email protected]/mf/4102.0 Australian Bureau of Statistics (ABS). Overweight and Obesity in Adults, Australia. Canberra; 2004-05. Cat. no. 4719.0. Viewed 2010 December 8. Available from: http://www.abs.gov.au/ausstats/[email protected]/Latestproducts/4719.0Main%20Features3200405?opendocument&tabname=Summary&prodno=4719.0&issue=2004-05&num=&view=. Australian Bureau of Statistics (ABS). The Health and Welfare of Australia's Aboriginal and Torres Strait Islander Peoples, Australia. Canberra; 2008. Cat.no. 4704.0. 2010 December 8. Available from: http://www.abs.gov.au/AUSSTATS/[email protected]/39433889d406eeb9ca2570610019e9a5/4D56 2E9436FD1BA7CA2574390014BCE3?opendocument. Bere E, Veierød M, Bjelland M, Klepp KI. Free school fruit—sustained effect 1 year later. Health Education Research 2007; 21(2):268-75. Besharov DJ, Germanis P. Rethinking WIC: an evaluation of the Women, Infants, and Children Program. Washington DC: AEI Press; 2001 Burgess CP, Johnston FH, Berry HL, McDonnell J, Yibarbuk D, Gunabarra C, Mileran A, Bailie RS. Healthy country, healthy people: the relationship between Indigenous health status and “caring for country”. Medical Journal of Australia 2009; 190(10):567-72. Burns C, Gibbon P, Boak R, Baudinette S, Dunbar JA. Food cost and availability in a rural setting in Australia. Rural and Remote Health 2004; 4(4):311. Caan B, Horgen DM, Margen S, King JC, Jewell NP. Benefits associated with WIC supplemental feeding during the interpregnancy interval. American Journal of Clinical Nutrition 1987; 45:29-41 Centre for Overweight and Obesity (COO). A literature review of the evidence for interventions to address overweight and obesity in adults and older Australians (with special reference to people living in rural and remote Australia and Aboriginal and Torres Strait Islanders) undertaken for the Department of Health and Ageing National Obesity Taskforce. Canberra, ACT: University of Sydney; 2005. Clark R, Conning R, Waters E, Armstrong R, Petrie R. Evidence Summary: remote and rural issues in the prevention of obesity for pre-adolescents and adolescents. Geelong, NSW: Deakin University; 2009. Dunn S, Dewis E. Health Weight Program Evaluation, Queensland 1996 - 1999. Aboriginal and Islander Health Worker Journal 2001; 25(4):26-8.

Page 131 of 170

Egger G, Fisher G, Piers S, Bedford K, Morseau G, Sabasio S, Taipim B, Bani G, Assan M, Mills, P. Abdominal obesity reduction in Indigenous men. International Journal of Obesity 1999; 23(6):564-9. Foley RM, Pollard CM. Food Cent$— implementing and evaluating a nutrition education project focusing on value for money. Australian and New Zealand Journal of Public Health. 1998; 22(4):494-501. Food and Nutrition Service. US Department of Agriculture. About WIC [internet]. [updated 2005 February 11; cited 2010 March]. Available from: http://www.fns.usda.gov/wic/aboutwic/. Glasson C, Chapman K, James E. Fruit and vegetables should be targeted separately in health promotion programmes: differences in consumption levels, barriers, knowledge and stages of readiness for change. Public Health Nutrition 2010; [Epub ahead of print]: doi: 10.1017/S1368980010001643 Government of Queensland. The 2006 Healthy Food Basket Survey. Brisbane: Government of Queensland; 2006 Gracey M, Bridge E, Martin D, Jones T, Spargo RM, Shephard M, Davis EA. An Aboriginaldriven program to prevent, control and manage nutrition-related "lifestyle" diseases including diabetes. Asia Pacific Journal of Clinical Nutrition 2006; 15(2):178. Harrison MS, Coyne T, Lee AJ, Leonard D, Lowson S, Groos A, Ashton BA. The increasing cost of the basic foods required to promote health in Queensland. Medical Journal of Australia 2007; 186(1):9-14. Heath D, Longstreet D, Malouf P, Panaretto K. Walkabout Together: a lifestyle intervention program developed for Townsville's overweight Indigenous people. In Touch newsletter of the Public Health Association of Australia 2006; 23:18-9. Horne PJ, Tapper K, Lowe CF, Hardman CA, Jackson MC, Woolner J. Increasing children's fruit and vegetable consumption: a peer-modelling and rewards-based intervention. European Journal of Clinical Nutrition 2004; 58(12):1649-60. Innes-Hughes C, Bowers K, King L, Chapman K, Eden B. Food security: the what, how, why and where to of food security in NSW. Discussion Paper. Sydney: PANORG, Heart Foundation NSW and Cancer Council; 2010. Jones R, Smith F. Fighting disease with fruit. Australian Family Physician 2007; 36(10):8634. Kinnunen TI, Pasanen M, Aittasalo M, Fogelholm M, Hilakivi-Clarke L, Weiderpass E, Luoto R. Preventing excessive weight gain during pregnancy - a controlled trial in primary health care. European Journal of Clinical Nutrition 2007; 61(7):884-91. Kuipers Y. Focusing on obesity through a health equity lens: a collection of innovative approaches and promising practices by European and international health promotion bodies to counteract obesity and improve health equity. EuroHealthNet; 2010. Lee AJ, Bonson AP, Yarmirr D, O'Dea K, Mathews JD. Sustainability of a successful health and nutrition program in a remote Aboriginal community. Medical Journal of Australia 1995; 162(12):632-5. Lee AJ, Darcy AM, Leonard D, Groos AD, Stubbs CO, Lowson SK. Food availability, cost Page 132 of 170

disparity and improvement in relation to accessibility and remoteness in Queensland Australia New Zealand Journal Public Health 2002; 26(3):266-72. Lee AJ, Hobson V, Katarski L. Review of the nutrition policy of the Arnhem Land Progress Association. Australian and New Zealand Journal of Public Health 1996; 20(5):538-44. Lin BH, Guthrie JF. How do low-income households respond to food prices? Can food stamps do more to improve food choices? An economic perspective. Washington DC: Department of Agriculture, Economic Research Service; 2007. Lorraine A, Cummins G, Pickett L. Fun, food and fitness for Noongars. In 6th National Rural Health Conference: Canberra, ACT; 2001. McDermott R, O’Dea K, Rowley KG, Knight S, Burgess P. Beneficial impact of the homelands movement on health outcomes in central Australian Aborigines. Australian and New Zealand Journal of Public Health 1998; 22(6):653-8. McDermott R, Rowley KG, Lee AJ, Knight S, O'Dea K. Increase in prevalence of obesity and diabetes and decrease in plasma cholesterol in a central Australian Aboriginal community. Medical Journal of Australia 2000; 172(10):480-4. National Preventative Health Taskforce (NPHT). Australia: the healthiest country by 2020. National preventative health strategy - the roadmap for action. Canberra, ACT: Commonwealth of Australia; 2009. North Metropolitan Area Health Service. FOODcents for Aboriginal and Torres Strait Islander people in WA program [internet]. [updated 2006 January; cited 2010 December]. Available from: http://www.healthyfuture.health.wa.gov.au/Health_topics/ASTIFOOD/ATSI_FC_Manual.pdf Northern Territory Government. The Remote Indigenous Stores and Takeaways (RIST) project report: Northern Territory Government; 2008. O’Dea K. Preventable chronic diseases among Indigenous Australians: the need for a comprehensive national approach. Heart, Lung and Circulation 2005; 14(3):167-71. Polley BA, Wing RR, Sims CJ. Randomized controlled trial to prevent excessive weight gain in pregnant women. International Journal of Obesity 2002; 26: 1494-502. Rajopal R, Cox RJ, Lambur M, Lewis EC. Cost-benefit analysis indicates the positive economic benefits of the Expanded Food and Nutrition Education Program related to chronic disease prevention. Journal of Nutrition Education and Behavior 2002; 34:26-37. Robertson A, Lobstein T, Knai C. Obesity and socio-economic groups in Europe: evidence review and implications for action. 2007. Rowley KG, Daniel M, Skinner K, Skinner M, White GA, O’Dea K. Effectiveness of a community-directed 'healthy lifestyle' program in a remote Australian Aboriginal community. Australian and New Zealand Journal of Public Health 2000; 24(2):136-44. Rowley KG, O’Dea K, Anderson I, McDermott R, Saraswati K, Tilmouth R, Roberts I, Fitz J, Wang Z, Jenkins A, Best JD, Wang Z, Brown A. Lower than expected morbidity and mortality for an Australian Aboriginal population: 10-year follow-up in a decentralised community. Medical Journal of Australia 2008; 188(5):283-7.

Page 133 of 170

Singh AS, Chin A, Paw MJM, Brug J, van Mechelen W. Dutch obesity intervention in teenagers: effectiveness of a school-based program on body composition and behaviour. Archives of Pediatrics and Adolescents Medicine 2009; 163(4):309-17. (referred to as Singh et al, 2009a) Singh AS, Chin A, Paw MJM, Brug J, van Mechelen W. Process evaluation of a schoolbased weight gain prevention program: the Dutch Obesity Intervention in Teenagers (DOiT). Health Education Research 2009; 24(5):772-7. (referred to as Singh et al, 2009b) Smith J. Wadja Warriors Football Team's Healthy Weight Program. Aboriginal and Islander Health Worker Journal 2002; 26(4):13-5. Steptoe A, Perkins-Porras L, McKay C, Rink E, Hilton S, Cappuccio FP. Behavioural counselling to increase consumption of fruit and vegetables in low income adults: randomised trial. British Medical Journal 2003; 326:855. Teig E, Amulya J, Bardwell L, Buchenau M, Marshall JA, Litt JS. Collective efficacy in Denver, Colorado: strengthening neighbourhoods and health through community gardens. Health & Place 2009; 15(4):1115-22. Thorpe S, Browne J. Closing the nutrition and physical activity gap in Victoria: Victorian Aboriginal nutrition and physical activity strategy. Policy proposals for promoting healthy eating and physical activity among indigenous Victorians 2009-2014. Melbourne; 2008. Torzillo PJ, Pholeros P, Rainow S, Barker G, Sowerbutts T, Short T, Irvine A. The state of health hardware in Aboriginal communities in rural and remote Australia. Australian and New Zealand Journal of Public Health 2008; 32(1):7-11. Wen L, Baur L, Rissel C, Wardle K, Alperstein G, Simpson J. Early intervention of multiple home visits to prevent childhood obesity in a disadvantaged population: a home-based randomised controlled trial (Healthy Beginnings Trial). BMC Public Health 2007; 7(1):76. Winkler E, Turrell G. Confidence to cook vegetables and the buying habits of Australian households. Journal of the American Dietetic Association 2010; 110(5 Suppl 1):S52-S61.

Page 134 of 170

5.0 ASSESSMENT OF EVIDENCE REVIEWS The detailed scan of the evidence in section 4 utilised a range of sources of evidence to assess the potential contribution of specific interventions to the achievement and maintenance of energy balance in communities. Our starting point was the assessment of traditional systematic reviews of evidence published in the scientific literature. Whilst such data was available in some action areas, no such systematic assessment of the evidence was available to help judge the merit of interventions in all action areas. In general, the evidence from a traditional review of the literature was very sketchy and limited. This is in part a consequence of tight definitions applied to the type and rigour of information that might be considered acceptable for inclusion in traditional evidence assessments such as Cochrane reviews or other systematic reviews. More valuable information was obtained from a scan of the grey literature. Many countries have established bodies which collate and assess evidence around public health and health promotion issues and produce reports to inform policy development. These bodies such as the Centers for Disease Control and Prevention in the USA and National Institute for Health and Clinical Excellence in the UK are able to take a less restrictive view of what can be included as evidence and how it should be assessed and interpreted, which allows them to integrate a wider body of knowledge. In addition, a range of NGOs or quasi government agencies associated with advocacy around public health also collate information and undertake assessments of the evidence to support action in certain areas to prevent obesity. Integrating the evidence from a wide range of sources, which are presented in a range of formats with varying levels of scientific rigour is a challenge. We developed a template and evidence assessment system which was based on our previous work (Gill et al, 2005) together with the work of the International Obesity Taskforce (Swinburn et al, 2005) and the World Cancer Research Fund (2009). This allowed us to assess the comparative merit of different interventions within each action area and also review the whole body of evidence across all intervention areas to guide better formulation of comprehensive plans of action to prevent obesity. 5.1 AN ASSESSMENT OF THE EVIDENCE WITHIN DEFINED ACTION AREAS To effectively address obesity it is important to address all the drivers of obesity through a comprehensive program of action across all the potential domains of action. We utilised the Foresight map of the drivers of obesity and the domains of action or clusters of drivers identified through the foresight process. By taking this approach, we are able to develop guidance of what specific interventions may be more effective at addressing the drivers of obesity within each domain of action and thus enable a more whole of system response to be developed. The level and quality of evidence within each action area varied greatly but there was still sufficient information from a range of sources on which to assess the promise of various interventions. Table 21 sets out a summary of the interventions assessed as showing the most promise within each action area, and also lists the amount and type of evidence under both the traditional NHMRC evidence assessment scheme and our broader, more integrative definition of evidence. The table also lists any other considerations which were included in the appraisal of the promise of the nominated interventions. Only the most promising interventions within each action area have been included, but no attempt has been made to rank the relative merit of the interventions across each action area.

Page 135 of 170

Table 21. Summary of evidence to support action from evidence reviews Action area

Interventions with most promise

Level of traditional evidence

1. Early life exposures and growth patterns

Improved diet

low

Total body of consistent evidence low-med

2 mass media

Mass media and social marketing campaigns

low

low

3. Exposure to marketing of foods and lifestyles

Statutory policy to restrict food marketing

low

high

4. Improved physical activity and nutrition in every day life

Multi-component interventions in a. communities b. schools c. early childcare d. worksites

low-med med low low-med

high med low-med med

5. Planning healthy active environments

-Improved access to parks and recreational areas - Building, planning and design codes

low

med

low

med

6. Food supply

Reformulation of existing products

low

med

7. Food access and availability

- Improve quality of food in small stores in poorly served areas - Improved access to plain drinking water - Front of pack labelling - Menu labelling - Soft drink tax

low

med

low

med

low

high

low low

med med

- Food vouchers and subsidies for fresh produce

low - med

high

8. Food purchase and consumption

9. Action in high risk groups

maternal

Comments

-Small evidence base but enormous potential - Small costs -Important equity issue - Promise not high when implemented alone -Important element of overall package of actions -Good reach, low costs - Resistance from some sectors - Low cots - Short timeframe - No indication of cost to industry - Good resonance with most stakeholders - Issues around sustainability - Short to med timeframes - Moderate cost - Lack of intervention studies - Large change in PA needed to impact on energy balance - Interventions expensive but sustainable - Long time frames - Synergy with environmental concerns - Acceptable to most stakeholders - Medium timeframe - No equity issues - Low cost options - Reduces inequities - Sustainable - Short timeframes

- Strong resistance from certain sectors - Some concern over transferability of evidence - Some equity issues - Cost implications - Concerns on sustainability - Some issues with transferability of evidence - Short time frames

Page 136 of 170

5.2 AN ASSESSMENT OF THE EVIDENCE ACROSS ALL ACTION AREAS Segmenting all the information and assessing effectiveness of interventions within nine defined action areas allowed us to more efficiently process and report on the large volume of evidence we had collected. It also enabled us to make comparative assessments of the potential of specific interventions within each action area to help achieve and maintain energy balance. However, experience and evidence from past public health interventions indicate the importance of developing a comprehensive suite of actions which involve interventions from across many sectors to effectively tackle the prevention of weight gain. To inform the construction of such a portfolio of actions it is important to assess the findings from the evidence summaries as a whole, to get a clearer picture of what they mean for comprehensive, whole of system approaches to the prevention of obesity. Such information is also invaluable for identifying research issues and gaps. Finding 1 The level and quality of experimental evidence (as assesses by NHMRC evidence grades) is limited across all action areas, and so the acceptance or rejection of proposed interventions needs to be based on an assessment of the total body of information available to gauge its potential contribution to the achievement and maintenance of energy balance. No intervention should be automatically ruled in or out until this is undertaken. 1. Level of "traditional” NHMRC evidence The most striking finding of the comprehensive scans of the evidence set out within the evidence summaries is the limited extent of high quality, consistent empirical or experimental evidence that exists to support action on most of the interventions evaluated or proposed strategies in each action area. On reflection, however, this should not be surprising, as there has only been a relatively short history of action to prevent obesity and most of the interventions are what might be termed “first generation” which have been developed and implemented without reference to the learned experiences of previous interventions. However, there is a common perception that within obesity prevention there is a clear evidence dichotomy separating interventions into two classes; those that clearly have empirical evidence to support the merit of their implementation and those interventions that must be rejected because there is no empirical evidence to support their introduction. The evidence summaries do not support the existence of such dichotomy based on the level and quality of empirical evidence, and thus no intervention should be ruled out on this basis. All interventions need to be assessed against the total body of the information that is available, to gauge their level of promise or potential to contribute to the achievement and maintenance of energy balance. Finding 2 The evaluation of obesity interventions have not been undertaken in a consistent or structured fashion, making an assessment of the merit of a specific intervention difficult and the integration of results from several intervention studies through systematic reviews or meta-analyses of limited value. There is a need to set consistent evaluation timeframes, define an agreed hierarchy of outcomes and use the same measurement for changes in these outcomes. 2. Consistency in evaluations The greatest volume of experimental evidence currently exists for interventions undertaken in everyday life settings such as school, worksites and day-care centres. In the school setting alone, there have been 11 systematic reviews and 5 meta-analyses of interventions to address obesity since 1990, although many of these reviews have been based on the same studies. Despite this level of evidence, assessment activity and some high quality Page 137 of 170

evaluation research designs, the findings in terms of weight-related outcomes have been mixed. Part of the reason for this lack of clarity in findings has been the large variations in the way that interventions have been implemented, evaluated and reported. Firstly, the intensity of interventions varies considerably, yet this is not always adequately considered in comparing effects. Secondly, evaluation measures and timing have been inconsistent. When weight status has been an outcome, it has been reported in a multitude of incompatible measures. Some studies have measured net change in kilograms or weight lost, others have used changes in BMI units or BMI z scores, others have only reported waist circumference, and others have used other anthropometric measures such as skinfold thickness. Timeframes also varied, with some studies evaluating effects after only 6 months and most studies limited to a 12 month follow-up period. Experience from past public health campaigns suggest these timeframe may be too short to see an effective change in weight status and thus there should be standardised program outcomes matched to the timeframe of the evaluation. Many evaluations were able to demonstrate changes in weight-related behaviours or at least structural change in the intended direction. Although this does not indicate the efficacy of the intervention in terms of energy balance, it is an indication of the promise of such an approach and can be used as part of the broader body of evidence. Finding 3 There has been an over-reliance on systematic reviews to define the level of evidence to support specific obesity prevention initiatives. More effort needs to be invested in gathering a more complete body of information which can be used to guide decisions around the potential of an intervention to contribute to the achievement and maintenance of energy balance. 3. The role of systematic reviews Another important reason for the variable results from individual obesity prevention initiatives in everyday settings is that the content of the interventions varies greatly. Even in interventions that have focussed on a single strategy, the approach to achieving changes has often been different across studies, in terms of duration and number of contacts for example, making each study unique and difficult to combine within a systematic review or meta-analysis. This problem is magnified because of the strong influence of context on the outcome of any behavioural change intervention. The reliance on systematic reviews as the tool to collate and assess the quality and consistency of evidence around specific interventions to address obesity may have been counter-productive, as it has led to the focus on a very narrow range of outcomes to assess the success of an intervention and assumes that very different interventions in very different circumstances are comparable. Thus, this approach excludes or neglects the value of building a broad base of evidence from multiple sources of information which would allow a better classification of the potential of an intervention to contribute to the achievement and maintenance of energy balance alone, or more importantly, as part of a program of action. Finding 4 Evaluations of obesity prevention initiatives have focussed on defining “what” works. However, the achievement of a defined outcome is a product of both the content of the intervention together with the quality of the implementation process. More effort needs to be devoted to defining how to deliver specific interventions within the context of different communities, rather than just defining what to do. 4. The importance of both process and content. As a result of current evidence assessment schemes, the important influence of context and process has frequently been overlooked in identifying effective interventions that should be Page 138 of 170

promoted as part of an obesity prevention strategy. Under such schemes, there is an implicit assumption that similar interventions are expected to produce the same outcome when implemented in a wide range of communities or settings, and failure to do so classifies such an intervention as ineffective. In addition, these approaches assume that the success or failure of an intervention relies only on the content of the intervention and not on the quality of the development and implementation process. It is likely that some of the variation found within the evidence summaries for the same intervention implemented in different settings results from the implementation of that intervention in settings with fundamental differences or because the quality of the process of implementation was variable. For example, the evidence summaries show that the redesign of parks and playgrounds has produced a mixed response, with some interventions showing a significant positive impact on physical activity but in other settings this intervention has produced no improvement or a reduction in activity. Current assessments of the evidence around this intervention suggest that because of the mixed results, it cannot be recommended as an effective strategy. Attention to the process of implementation or appropriateness of the chosen intervention may reveal a different interpretation and so assessments of the detail and quality of implementation needs to be integrated into the evidence assessment process. While it can be argued that the interpretation of findings is an integral part of existing evidence review systems, often this information is ‘lost’, as the focus of most reviews is on collating and comparing outcomes with reference to the quality of the evaluation design only. Finding 5 An alternative evidence assessment scheme needs to be developed which is based on an assessment of the total body of information and better defines the potential of an intervention or program of action to contribute to the achievement and maintenance of energy balance, when implemented as part of a comprehensive program of action. Such a system should help define research approaches that are appropriate to generate this total body of evidence. 5. The need for an alternative evidence assessment scheme Whilst it is difficult to find an evidence assessment scheme that is capable of integrating all the relevant information into the decision making process, current systems which focus on evidence collected from tightly controlled experimental studies and which do not allow the consideration of less rigorous (but often more relevant) information sources may not best serve the needs of policy makers. The strengths and limitations of different study designs and their relevance to the generation of evidence of effectiveness in obesity prevention and other health promotion interventions has been discussed by various groups (Heller & Page, 2002; Kroke et al, 2004; Glasgow et al, 1999). There is a growing consensus that evidence assessment schemes need to be broadened to accommodate a wider range of information that may be of value in helping make assessments about the merit of different interventions or packages of interventions (see section 8.5 IOM Bridging the Evidence Gap). This is particularly relevant in a situation where it has not been possible, or is inappropriate or unethical to gather the type of evidence that current clinical evidence decision models require. Indeed some commentators have suggested that we may need to forego the notion of certainty that is implicit in current evidence assessment schemes, and instead assess interventions (or more appropriately comprehensive programs of action) in terms of their promise or potential to contribute to the achievement and maintenance of energy balance (Swinburn et al, 2005). From our assessment of the evidence reviews undertaken for this project we propose that an expanded evidence assessment scheme should include: • Large scale controlled experimental (or quasi-experimental) studies with defined Page 139 of 170

outcomes, where possible • Process evaluations of the development, implementation, reach and usage of the components of large scale interventions • A series of small scale experimental studies in controlled conditions to answer specific questions about likely behavioural responses to a specified set of circumstances. The conditions can be varied to determine how this influences the impact of the intervention on behaviour. • Consumer research to assess attitudes and beliefs about certain behaviours or proposed interventions which might help predict behaviour or clarify understanding. • Collection and collation of a wide range of ancillary observational data including - ecological assessment of different levels of exposure to obesity-promoting factors - repeated cross sectional assessments of weight status and weight-related behaviour at a population level  monitoring or surveillance observational studies (before and after naturally-occurring interventions) • Modelling using existing data and various explicit assumptions • Application of logic models and theory to test robustness of assumptions and projections regarding program implementation, reach and extent of change Finding 6 Very few packages of action have been evaluated for their combined impact on energy balance, although some recent, whole of community programs such as EPODE are the exception. This would seem to be the preferred mode of evaluation, given that many individual interventions are not designed to have a direct effect on weight status or weightrelated behaviours, but rather are designed to interact with other interventions to enable of amplify behaviour change. Such an approach is of particular relevance to the evaluation of interventions that are primarily enabling of amplifying actions. They should be judged on their potential to contribute as part of a program of action to the achievement and maintenance of energy balance 6. Evaluation of total packages of action The evidence summaries within this project take a more expansive approach to the assessment of the potential of specific interventions than traditional review processes. Our collation and presentation of information has been undertaken in an objective manner and the interpretation of this information has been made using specific criteria, although it is difficult to avoid some subjectivity in this process. Unfortunately, the nature of experimental design means that very rarely are programs of action involving multiple intervention evaluated as a whole. There are reported evaluations of multi-component interventions, but rarely is a package of separate interventions included within the one evaluation. The exceptions are some recent whole of community interventions, such as the EPODE program and the “Be active- Eat well” program in Colac, where these initiatives have been evaluated as a whole. This has great benefit, in that it is often the package of interventions acting in a complementary and interactive manner, rather than the sum of the individual intervention components, that are responsible for the outcomes achieved. This may explain why these programs are some of the few evaluated programs demonstrating significant impacts on population weight status. The limitation of such an approach is that it may not allow an assessment of which components are effective in achieving change. Such an approach is of increased relevance when the interventions are primarily enabling or amplifying actions, as these are unlikely to have any large, independent impact upon weight status, especially within the short-term, as their chief mode of action is to support and improve other interventions. Their potential to contribute as part of a program of action to the achievement and maintenance of energy balance is a more useful metric than any independent effects. Page 140 of 170

Finding 7 The greatest volume of experimental evidence currently exists for interventions undertaken in everyday life settings such as school, worksites and day-care centres. However, no attempt has been made to integrate the wider body of information that could assist with judgements about the merit of interventions with this experimental data. Most reviews of this evidence show a positive impact of these types of intervention on the weight status of children and adults with the greatest benefit shown for multi-component interventions. However, because the content and context of these interventions vary so much, and the outcomes are not always defined by the same measures, it is difficult to make clear assessments of the merit of individual approaches or to combine individual studies into systematic reviews or meta-analyses. As a consequence, many systematic reviews are unable to reach meaningful conclusions. Multi-component community-wide interventions is the area with the most consistent and promising evidence. 7. Everyday settings Although there are a number of inconsistencies and gaps highlighted by the evidence summaries, they do provide an appropriate basis for making some value judgements about the merits of various interventions. The large amount of experimental data collected as part of the evaluation of interventions within the everyday life settings such as school, worksites and day-care centres should allow greater clarity around the merit of such interventions. Unfortunately, the inconsistent findings from these studies limit their immediate interpretation. In addition, because of the volume of experimental data within this action area, less effort has been applied to harvesting information from other sources which could help in making assessments of their potential contribution to the achievement and maintenance of energy balance. Despite these inconsistencies, there is still sufficient information to indicate that intervening in these settings brings positive behavioural change and some likely impact on weight status. More research needs to be applied to determine which type of interventions, together with the most appropriate implementation processes, will bring the greatest reward. Most individuals only spend a proportion of their time within these settings, which may reduce the potential of actions within these settings to impact on weight status, unless combined with complementary approaches. For example, children attend school for approximately five hours each day, for five days each week for around 40 weeks each year. Although this is a substantial amount of time, very little of it is discretionary, in terms of increasing time spent in activity and only around 12% of food is consumed during this period. It is therefore not surprising that interventions with multiple components tended to do better in terms of their impact on weight. This finding was reported in terms of schools, worksites and especially within community-based programs. Finding 8 Contrary to current perceptions, front of pack labelling and statutory restrictions on food marketing to children have the broadest base of evidence, with consistent findings to support their implementation. Menu labelling and changes to the physical environment to increase physical activity also has sufficient evidence to support a modest degree of promise. Evidence for action in these areas comes from small experimental studies, ecological assessments, observational studies, consumer research, parallel evidence and logic. 8. Interventions with a broad base of evidence Less experimental evidence was found to support interventions in other action areas. However, whilst this creates certain limitations, it has led to the collection and analysis of a more extensive base of information on which to judge the merit of these interventions. The evidence summaries show that interventions aimed at achieving some structural changes, Page 141 of 170

such as regulation of TV food advertising, front of pack labelling, menu labelling and changes to the physical environment to increase physical activity, have a broad base of evidence; and this in fact illustrates our proposed, expanded evidence assessment scheme. In these action areas, a series of small scale experimental studies under controlled conditions have been conducted to answer specific questions around consumer responses to marketing restrictions, front of pack labelling and menu labelling. In addition, substantial consumer research has been undertaken to gauge consumer attitudes, beliefs and knowledge to these issues. There have also been ecological assessments around exposure of children to TV advertising of foods and the collection of observational data pre and post changes to TV advertising restrictions in the UK and menu labelling laws in the USA. This has also been supplemented with modelling of various scenarios and assessment of the logic and theory behind such interventions. Overall this has created a base of information that provides relatively strong support for front of pack labelling, restrictions to food marketing and modest support for changes to the physical environment to increase physical activity and menu labelling. Finding 9 The evidence around the merit of taxation and subsidies needs clarification. Most analyses have been conducted in the USA and although there is some observational data, most of the assessments rely on modelling of impact. Most studies demonstrate a small but potentially meaningful impact when taxes are imposed on particular products, and point to more merit in taxation of specific food products such as soft drinks. Taxation disincentives appear to have a stronger impact on behaviour than incentives through subsidies, although subsidies may need to be part of any fiscal strategy to offset equity concerns of taxation. We have no data on the price elasticity of demand around food products in Australia to allow effective modelling of taxation/subsidy impacts. 9. Evidence around fiscal strategies There has also been an equally impressive range of actions undertaken to help define the merit of fiscal strategies in the prevention of obesity. Such an intervention is not open to evaluation by a traditional controlled trial design, and so much effort has been directed into small experimental studies, modelling and opportunistic assessment of tax regime changes. There have been a limited number of studies directly assessing the effect of fiscal strategies on physical activity. Incentive schemes to encourage walking and cycling to work, or tax deductions for sports programs have not had a significant impact on levels of activity, although congestion charges seem to have had a greater impact on reducing car use in the districts in which they have been imposed. The imposition of food taxes or subsidies has been studied in more depth, and was found to generally influence consumption in the right direction although a large percentage of these studies are from the USA and many relied on modelling rather than other indicators of change. The imposition of taxes on the basis of nutrient content (so called “fat taxes”) appear to be more difficult to implement and sustain and so most tax regimes have identified specific food groupings as the basis of taxation or subsidies. Most legislators have chosen to restrict such taxes to general categories of high energy dense, nutrient poor snack foods or beverages or to soft drinks alone, although Denmark has introduced taxes that target a wider range of specific foods products. There is reasonable support for the merit of such taxes, although modelling the impact of such taxes using estimated price elasticity of demand for these food categories tends to give more promising results than ecological assessments of varying tax regimes, or observational data examining changes after the introduction of a new tax. The observed impact of taxes has usually been small, but potentially meaningful as it applies Page 142 of 170

to the whole of the population, with universal reach and uptake. There are indications that such approaches are more likely to have a positive impact within lower income groups with less discretionary income spent on food. The evidence around the use of subsides is less clear, with short term experimental interventions demonstrating a strong impact of price subsidies on the selection of low fat items from a canteen. However, other experimental studies which assessed various combinations of taxation and subsidy suggested that taxation drives consumption of the taxed food down, but that subsidies may result in increased intake of the subsidised food without leading to reductions in undesirable foods and thus contributing to overconsumption of energy. This effect may be magnified by the finding that taxation costs are almost universally passed on, but subsides are often partly absorbed by manufacturers or vendors, resulting in a smaller price reduction. Targeted food subsidies appear to have a strong beneficial effect when provided in the form of food vouchers to low income groups, and it is likely that any taxation strategy will need to include subsidies to limit the differential impact across income groups. The biggest problem in assessing the results of fiscal strategies imposed in other countries is determining their relevance or applicability to Australia. Different economic and social systems will respond differently to these measures, and currently we have no clear understanding of the price elasticity of demand for key food products for Australian consumers. Finding 10 Very few interventions within the action areas addressing the food supply, food access and availability and active environments were able to demonstrate a direct impact on weight status. Those studies that did were usually small scale and targeted at motivated groups. However, there is sufficient evidence that such interventions impact positively on weight related behaviours and additional support for the promise of these interventions from observational data, parallel evidence and program logic. Interventions to improve food and physical activity environments are important levers for addressing current inequalities in opportunities to eat well and be physically active. Interventions in these actions areas have considerable synergies with actions to address climate change and other environmental concerns 10. Supportive food and physical active environments Many years ago, the Ottawa Charter for Health Promotion highlighted the important role of environments in influencing health behaviours and the necessity of addressing these environments to facilitate appropriate behaviour change. This is now a well-accepted principle of action, which has even more relevance in the prevention of obesity, where structural, economic and environmental changes in recent years is seen as one of the key drivers of obesity. However, identifying the specific interventions which will help create more supportive food and physical activity environments is not straight-forward. There is little indication that developing new healthier food products, without removing unhealthy versions will lead to improvements in food purchases and consumption. There is more support from observational studies of the merit of reformulation of existing food products to make them healthier; but such data indicate the overall effect on energy balance is likely to be small. The modelling and extrapolations around product reformulations provide very positive estimates of the potential impact on energy and nutrient intake; but these are best case scenarios and may over-estimate impact. Extrapolations of the potential impact of controlling portion sizes from small experimental studies has similar problems, and observation data indicate that people make adjustments by increasing the number of small portion items consumed. The lack of outcome data related to weight status Page 143 of 170

associated with these interventions makes obtaining clear estimates of their potential impact on energy balance difficult; but there is very strong logic and theory support for such approaches. Access to healthy, affordable foods and reducing access to high-calorie, low-nutrient foods is another area where logic and theory indicate there is substantial opportunity for health gain; but there has been very little systematic assessment of interventions in this domain. Evidence is available from a wide array of studies, but it is generally of low quality and indirectly based on behavioural outcomes such as purchasing behaviour rather than specific weight-related outcomes. There are also differences between countries in the associations between measures of food access with diet and weight outcomes. For example, the availability of supermarkets in poorly served areas within the USA is strongly associated with poorer diet and weight status; but these associations are much weaker in other countries. This is likely to reflect contextual factors, such as urban planning and transport factors. Despite these limitations, the overall assessment indicates that many interventions in this area offer promise in contributing to improved energy balance. In addition, such interventions are likely to make an important contribution to reducing inequalities in access and availability that have been associated with inequalities in rates of obesity. Likewise, the design and provision of environments and transport systems that promote active living offer enormous potential to contribute to energy balance, but there remains a lack of clarity around what interventions are most valuable in achieving improved physical activity and weight status. The physical activity literature is replete with cross-sectional studies examining the physical environment and showing consistent correlates with physical activity behaviours, but the paucity of intervention or longitudinal evidence makes the existing evidence base much less clear in terms of contributing to weight loss. The evidence summaries revealed a range of observational data which supports the potential merit of active transport initiatives aimed at increasing walking or cycling and reducing car use, improved attention to urban design and improved access to recreational and open space. However their likely impact on increased physical activity has been estimated as quite small, resulting in some cases in less than 30 minutes additional activity over a week. This is insufficient alone to make a major contribution to the achievement of energy balance, but may not be seen as insignificant in combination with a broad package of actions to address obesity. There is also considerable synergy between interventions aimed at reduced car use and more efficient use of public transport and active recreation and the need to reduce use of fossil fuels and other environmental concerns. Finding 11 There is value in identifying interventions to specifically address weight status within disadvantaged groups. Although such groups benefit from interventions which address the whole of the community, adaptations and specifically focussed implementation that addresses social or cultural complexities are often needed to ensure effectiveness with more disadvantaged groups 11. Interventions in disadvantaged groups Addressing many of the structural and environmental factors associated with the current obesity-promoting environment may also help address inequalities in weight status; yet there is also value in identifying specific interventions to more closely target disadvantaged groups within the Australian community. Of particular interest is identifying interventions that improve access to affordable, healthy foods and opportunities for activity for low income groups and indigenous communities. Although most of the interventions identified within the evidence summaries can be applied at a population level, it is through implementation to meet the special needs of disadvantaged groups that they can have substantial impact to Page 144 of 170

redress inequities. This is particularly relevant to both urban and rural indigenous communities, who have many of the worst markers of disadvantage in Australia and have social constructions of health behaviour that may not match the broader community. The evidence is still building in this area but targeted implementation of subsidies, skill development and capacity building programs are demonstrating merit.

Page 145 of 170

References Gill T, King L, Webb K. Best options for promoting healthy weight and preventing weight gain in NSW. Sydney: NSW Centre for Public Health Nutrition; 2005. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health promotion interventions: the RE-AIM framework. American Journal of Public Health 1999; 89:1322– 1327. Heller RF, Page J. A population perspective to evidence based medicine: ‘evidence for population health’. Journal of Epidemiology and Community Health 2002; 56:45–47. Kroke A, Boeing H, Rossnagel K, Willich SN. History of the concept of ‘levels of evidence’ and their current status in relation to primary prevention through lifestyle interventions. Public Health Nutrition 2004; 7:279–284. Swinburn B, Gill T, Kumanyika S. Obesity prevention: a proposed framework for translating evidence into action. Obesity Reviews 2005; 6:23–33

World Cancer Research Fund (WCRF). Policy and action for cancer prevention. Food, nutrition, and physical activity: a global perspective. Washington DC: AICR, WCRF; 2009.

Page 146 of 170

6.0 LESSONS FROM OTHER PUBLIC HEALTH INITIATIVES There have been very few, large-scale, comprehensive programs addressing the prevention of obesity that have been sustained for a long enough period to allow effective evaluation of the merit of the wide range of potential public health strategies that may be utilised. Therefore a number of commentators have sought lessons from successful actions to address other public health problems or from community-wide programs addressing other chronic diseases. Whilst this has been able to enlighten discussions around potential approaches, care must be taken in interpreting the nature of this parallel evidence and the context in which these actions occurred, to ensure that the learning is applied correctly to obesity prevention. 6.1 TOBACCO, ALCOHOL AND OTHER PUBLIC HEALTH INITIATIVES Tobacco control is often held up as a model for what can be achieved by concerted public health action and a number of analyses of the success of action on smoking have been published. Although there are clear difference between tobacco and diet and physical activity, there are sufficient similarities to benefit from lessons learnt from tobacco control programs. The CDC report on Best Practices for Comprehensive Tobacco Control Programs identified that successful tobacco efforts were based on a combination of: 1. comprehensive and coordinated programmatic and policy interventions designed to influence social norms, systems, and networks; and 2. effective, sustained, and adequately funded health communication campaigns that delivered strategic, culturally appropriate, high-impact messages (CDC, 2007). Such programmes have been successful in achieving large shifts in community attitudes to smoking and the creation of an environment where not smoking is as easy (and often easier) than smoking. There is direct relevance of these findings to the design of programs of action to address obesity. Central to the success of tobacco control has been strong advocacy from all sectors of society including the media. This has been important in shaping community attitudes and later in influencing and giving confidence to policy makers to implement change in the face of strong resistance from the tobacco industry. A great deal of emphasis has been placed on the influence of regulatory change (such as smoking bans in public places) on changes in smoking behaviour (Mercer et al, 2003). However, in most cases these regulations have only been implemented after the majority of the community (including smokers themselves) has accepted them as valuable. Major environmental interventions need a clear justification and broad community support to be successful. Well-conducted health research with a consistent outcome is often a key factor in shaping community attitudes. This is important in how regulation would best be applied to the prevention of obesity and is discussed in another section. A recent review of the factors successful in reducing alcohol consumption found that price disincentives and limiting access through age regulations, availability and service hours of outlets and servers training have been the most effective measures to reduce consumption. Restriction of advertising was moderately successful but education alone had very limited success (Gilmore, 2007). A useful analysis of the lessons from a range of successful attempts at social change, including increasing breastfeeding rates, seat belt use, and recycling was undertaken by the US Institute of Medicine (Economos et al, 2001). This analysis revealed that keys to effective change include recognition that there was a crisis; establishing major economic implications associated with the crisis; having a strong science base including research, data, and evidence for action; identified leaders who could work for their cause by utilising Page 147 of 170

their knowledge, competence, talents, skills, and even cunningness and charisma; developing coalitions to move the agenda forward; a strategic, integrated media advocacy campaign; involvement of the government at the state level to write and implement legislation, and at the local level to implement change; mass communication that included consistent positive messages supported by scientific consensus and was repeated in a variety of venues; policy and environmental changes that promoted healthy behaviours; and a plan that included many components that worked synergistically (Economos et al, 2001). 6.2 COMMUNITY CHRONIC DISEASE PREVENTION TRIALS Considerable experience and evidence of benefit can be drawn from the large- scale community cardiovascular disease (CVD) and diabetes prevention trials, which have included weight as an intermediary outcome. Such trials have demonstrated that it may be possible to prevent weight gain, if not reduce weight at a population level, and thus can provide useful information about effective strategies to address obesity. 6.2.1 Community CVD prevention trials The results of the early large-scale community CVD prevention trials, such as the Stanford Three Community and Five Community studies, as well as the Minnesota Heart Health Program, had limited impact on weight status and reinforced the difficulty of preventing weight gain in the community. However, later programs, such as the Pawtucket Heart Health Program, were able to make a modest impact on weight gain in the intervention community. These programs demonstrated the large time lag (5-10 years) that can be expected between the implementation of a truly community-wide program and the extent of behavioural change likely to be required to impact upon the weight status of the community. An analysis of the trials by Jeffery (1995) suggests that unless weight is the primary outcome of the intervention, it is unlikely that sufficient focus will be placed on achieving the level of change required to impact on energy balance and community weight status. North Karelia CHD prevention program The North Karelia Community CHD prevention program requires special attention because of its longevity and the extraordinary reductions in CVD events that it has been able to achieve (Pietenen et al, 1996). The improvements in CHD risk factors over the past 25 years of the North Karelia project have been well documented with dramatic reductions in blood pressure, blood cholesterol and rates of smoking and consequent rates of premature mortality from heart disease. In addition fat intake decreased, average fruit and vegetable intake increased and markers of physical activity, although less impressive, also showed some improvement (Vartiaitnen et al, 1994). However, over this period the level of obesity continued to rise within North Karelia and other regions of Finland suggesting that these strategies had been less successful in influencing body weight than other risk factors although the lessons of successful social change should still apply to obesity prevention strategies. The North Karelia Project was initiated in 1972 and was based on the same theoretical framework and contained the same degree of monitoring and evaluation as US-based community CHD prevention programmes. However, there were a number of important features which set this project apart from other contemporary programmes. The North Karelia Project arose as a direct consequence of a community petition for action on an issue they perceived to be important and was commenced only after a comprehensive community analysis and a period of community consultation. As a consequence there was strong community involvement in the planning, management and implementation of activities. Instead of establishing costly new structures, the North Karelia project was based on simple and practical activities which utilised existing community resources where possible, and Page 148 of 170

were managed by lay workers and voluntary agencies. Health workers provided strong leadership and support, and these activities were integrated into their normal work rather than being an additional research exercise. As a result, the North Karelia Project was far more sustainable than other such interventions and many of the original activities continue today, having been further developed as a result of continual evaluation. Analyses of changes in dietary and physical activity behaviour across the duration of the North Karelia project indicate that changes in health-related behaviour are often slow to occur. Even simple messages such as switching from full fat to low fat milk require many years of re-enforcement before being adopted by a sizeable proportion of the community. Often the more highly educated persons are capable of making such changes earlier than less well-resourced members of the community (Pietenen et al, 1996). This is another important reason why the sustainability of health promotion programmes is crucial to enabling health behaviour change to occur over lengthy time periods. 6.2.2 Diabetes prevention trials Recent large community-based trials examining the progression to diabetes in persons identified as glucose intolerant have produced significant reductions in diabetes rates by attention to exercise and diet, with small, but significant, weight losses of around 3-4 kg on average. Although these trials focussed on high risk individuals there are still important lessons for the primary prevention of obesity. The trials have been criticised for the resources required to achieve the reductions by lifestyle interventions, but the clear lesson is that initial intensity of the intervention is a very important element in achieving outcomes with high risk groups (Bray, 2009). Most programs had an intense initial intervention which was sustained by regular follow-up over time. In addition, a clear lesson from these trails was that small lifestyle changes lead to significant reductions in risk in those who are able to sustain this change (Cardoso & Hu, 2007). Obesity prevention programs may also benefit from a focus on modest changes in behaviour sustained over a significant period of time. 6.2.3 Large scale nutrition and physical activity programs A recent World Health Organization review examined the effectives of nutrition and physical activity interventions in the prevention of non-communicable disease and found that although most interventions were only evaluated in the short-term there were some clear lessons from the most successful programs (WHO, 2009). They found interventions that were multicomponent and adapted to the local context were the most successful, and that culturally and environmentally appropriate interventions were far more likely to be implemented and sustained. Also, interventions were more likely to succeed if part of an integrated program of actions, rather than an isolated action. Interventions to increase incidental activity through modification of the physical environment, improvement of facilities and policy reform offer the strongest promise for improving the level of physical activity in the community. Furthermore, interventions that use the existing social structures of a community, such as schools or the weekly meetings of older adults, reduce barriers to implementation. Implicit in all successful interventions was the participation of the stakeholders throughout the process and the use of community or religious leaders. 6.3 LESSONS AROUND THE USE OF REGULATION FOR BEHAVIOUR CHANGE IN PUBLIC HEALTH The apparent success of regulatory actions on the behaviour of the community in tobacco cessation, drink driving and alcohol consumption, as well as improved road safety through seat belt and cycle helmet use, has led to great enthusiasm for the application of regulation to addressing food and physical activity behaviour change in obesity prevention. Numerous reviews have highlighted the limitations of education alone and identified the importance of Page 149 of 170

regulatory change as a central component in the success of past public health campaigns, but closer inspection reveals that its use may not always be warranted or successful. Certain conditions must be met before regulatory change can be effectively implemented, achieve acceptance and have the desired effect on behaviour. This is particularly important in obesity prevention where there is much resistance to the concept of regulation from vested interests and therefore strong public support is required for regulatory change to succeed. Behaviour change theories suggest that a change in knowledge can help change attitudes that then drive behaviour change and this has been used as the basis of many public education programs which have been instituted with limited success. In their assessment of the introduction of workplace smoking ban, Borland et al (1994) observe that regulatory changes can drive behaviour change and precede attitude shifts but that positive attitudes to regulatory and other environmental changes are usually a critical pre-requisite if such changes are to be accepted. They argue that the sequence in changes in attitudes and behaviour depends upon whether the behaviour affected is seen to be under strong selfregulatory control or more at the mercy of external regulatory forces (see figure 6.1) Figure 5 – The potential for regulation to drive behaviour change

Source: Borland et al, 1994

For example a behaviour change such as salt reduction has low self-regulatory change because most salt is added into food by processors and thus regulatory change would have a higher level of acceptability and may even be necessary to achieve this behaviour. A similar situation may apply to limiting children’s demands for heavily advertised foods which also has a low level of self-regulatory control as it is driven by the high level of inappropriate food advertising. Thus regulatory and environmental change would be more acceptable and may even be a necessary precursor to behaviour change and this behaviour change may help change or reinforce changed attitudes and social norms. However, behaviour such as soft drink consumption in adults, which is perceived to be under high self-regulatory control, Page 150 of 170

may not receive the same level of support for regulatory or fiscal intervention and will require a change in attitudes to change behaviour. Of course, these assessments of the order of influence on behaviour do not preclude the use of regulatory interventions in behaviours which are perceived to have strong self-regulatory control, but it makes them less appealing without a program of action to gain public support. As children inherently have low self-regulatory control over most dietary and physical activity behaviours, regulatory and environmental intervention addressing these behaviours in young children may harness more support. 6.4 LESSONS LEARNT ABOUT TIMEFRAMES There are several mechanisms by which timeframes have the potential to influence our interpretation of information gathered from the evaluation of public health interventions. Some of these issues came through clearly in the evidence summaries presented in this report and other have been identified through an assessment of the evaluation of past comprehensive public health interventions addressing other issues. 6.4.1 Implementation and evaluation issues Comprehensive obesity prevention strategies, especially those involving multiple interventions, over multiple settings and involving multiple agencies, take a considerable amount of time to plan, engage all stakeholders and fully implement. Many programs set a three year time frame for this process to bring them into line with major funding bodies. However, the process frequently takes much longer than this as much of the preliminary stages of planning and implementation are subject to unexpected delays and full engagement of stakeholders in a slow process. In addition, implementation is usually a staged process with elements of the program being added in a sequential manner. As a consequence, many programs find that when they undertake their first period of evaluation after three years, that much of the program may only have been operating for 12 months or less. Thus what is termed a three year evaluation is in effect evaluating a program that has been operational for less than 12 months. Expecting to detect changes in weight status within this period or even significant behaviour is ambitious. 6.4.2 Differential in uptake Many large scale behaviour change programs have identified a clear differential in uptake of interventions with better educated or resourced sections of the community being the early adopters. These early adopters will show changes in behaviours and measured outcomes well before the less well-resourced groups have considered their options for change. Thus evaluation strategies should seek to identify and measure any differentials in uptake, and programs need to be sustained for a significantly long period of time to ensure the more disadvantaged sections of the community have a chance to benefit. 6.4.3 Evolution of programs Mistakes, omissions and inefficiencies are an inevitable feature of the primary implementation of a new intervention. Thus many innovative obesity prevention programs are evaluated as ineffective and are never repeated because the quality of implementation was lacking. The effectiveness of this intervention will grow if the lessons and experiences are integrated into the planning and development of future versions. Unfortunately because of the newness of obesity prevention action, most programs are still “first generation” and cannot be expected to yield the same returns as more mature programs of action associated with other public health initiatives such as tobacco control. Page 151 of 170

6.4.4 Exposure and generational effect Not only do programs improve with time but the level of continuous or repeated exposure that individuals or communities have to an intervention is likely to influence the impact it has on behaviour and weight status. Thus interventions adopted now may need to wait until the current generation of children are adults to fully reap the benefits of the intervention. This is particularly true with interventions in early life which influence not only in-utero environment and immediate health status of the child, but also influence the genetic expression or epigenetic inheritance of that child. This in turn may influence the weight status of the child as an adult and the in-utero environment for any offspring of the child.

6.5 SUMMARY OF LESSONS LEARNT FROM OTHER PUBLIC HEALTH INITIATIVES There are a range of findings that are relatively consistent across a range of other public initiatives that identify their utility in guiding public health initiatives to address obesity. These include: 1. Mass media campaigns are important in changing social norms and improving the knowledge and attitudes of the community. 2. Education is not enough to change weight-related behaviours. There is a need to deal with societal and environmental factors which support or inhibit behaviour change. 3. Strong support and advocacy from all sections of society, including the media is important for the long term success of public health initiatives aimed at behaviour change. In addition major environmental interventions need a clear justification to be successful. 4. Programs need to reach all sections of the community - not just the motivated. 5. Obesity prevention programmes need to focus directly on changing weight-related behaviours. 6. A high level of community involvement at all stages is necessary to maintain impetus. 7. A sound theoretical base provides a framework for action and evaluation. 8. Changes in health-related behaviours can be slow. 9. Sustainability of health promotion programmes is crucial to allow behaviour change over time. 10. Programmes should allow the integration of existing initiatives. 11. A comprehensive range of strategies, carried out over multiple settings, is required to have an impact on behaviour 12. Small behaviour or weight changes can have a population impact provided they are achieved by a significantly large proportion of the population. 13. There is a need for regular monitoring and improvement. 14. Health behaviour change programs need to be appropriately resourced. 15. Public support may need to be harnessed before considered regulatory interventions.

Page 152 of 170

References Borland R, Owen N, Hill DJ and Chapman S. Regulatory innovations, behaviour and health: implications of research on workplace smoking bans. International review of Health Psychology 1994; 3: 167-85 Bray GA. Preventing diabetes: Lessons learned from the Diabetes Prevention Program and its follow-up. Obesity and Weight Management 2009, 5(6):273-276. Cardoso MA, Hu FB. Nutritional interventions and primary prevention of type 2 diabetes. Current Nutrition & Food Science 2007; 3:47-53 Center for Disease Control and Prevention (CDC). Best practices for comprehensive tobacco control programs-October 2007. Atlanta, GA: CDC; 2007 Economos C, Brownson RC, DeAngelis MS, Foerster SB, Foreman CT, Gregson J, Kumanyika SK, Pate RR. What lessons have been learned from other attempts to guide social change? Nutrition Reviews 2001; 59(3 Pt 2):S40-56. Gilmore I. What lessons can be learned from alcohol control for combating the growing prevalence of obesity? Obesity Reviews 2007; 8(Suppl. 1):157–160 Kumanyika S, Jeffery RW, Morabia A, Ritenbaugh C, Antipatis VJ, Public Health Approaches to the Prevention of Obesity (PHAPO) Working Group of the International Obesity Task Force (IOTF). Obesity prevention: The case for action. International Journal of Obesity Related Metabolic Disorders 2002; 26(3):425-36. Mercer SL, Green LW, Rosenthal AC, Husten CG, Khan LK, Dietz WH. Possible lessons from the tobacco experience for obesity control. American Journal of Clinical Nutrition 2003; 77:1073S-1082S Pietinen P, Vartiainen E, Männisto S. Trends in body mass index and obesity among adults in Finland from 1972 to 1992. International Journal of Obesity 1996; 20:114-20. Vartiaitnen E, Puska P, Jousilahti P, Korhonen HJ, Tuomilehto J, Nissinen A. Twenty-year trends in coronary heart risk factors in North Karelia and in other areas of Finland. International Journal of Epidemiology 1994; 23:495-504. World Health Organization (WHO). Interventions on diet and physical activity: what works. Summary report. Geneva: WHO; 2009.

Page 153 of 170

7.0 RECENT AUSTRALIAN INITIATIVES Over the last two decades there has been a considerable amount of reflection and policy development around obesity prevention within Australia. National obesity prevention frameworks have been developed for children and for adults and older Australians. In addition, all jurisdictions have developed their own strategies for dealing with the problem of obesity and funded a range of major obesity prevention trials or initiatives. A recent report from the Productivity Commission sets out a detailed listing of these initiatives in an appendix (Crowle & Turner, 2010). In 2009/10 there have been three major policy or policy analysis projects which have provided an important basis for discussion around what interventions should be included in the comprehensive package of programs to address obesity. 7.1 NATIONAL PREVENTATIVE HEALTH TASKFORCE REPORT The Preventative Health Taskforce was established in 2007 to provide evidence-based advice to government and health providers on preventative health programs and strategies, and support the development of a National Preventative Health Strategy. In October 2008, the Taskforce released a discussion paper, Australia: The Healthiest Country by 2020 and three associated technical papers on obesity, tobacco and alcohol. These documents formed the basis for conducting consultations and calling for public submissions. The Taskforce received 397 public submissions to the report and held 40 consultations with almost 1,000 stakeholders. Participants included a mix of people representing industry, public health, professional groups, consumer groups, government, and other nongovernment organisations. Following this review and consultation process, the Taskforce released its final report in September 2009, titled Australia: the Healthiest Country by 2020. The Taskforce put forward 136 recommendations and 35 areas for action, tackling obesity, tobacco and alcohol as key drivers of chronic disease. There were 10 defined action areas in obesity. 1. Drive environmental changes throughout the community that increase levels of physical activity and reduce sedentary behaviour 2. Drive change within the food supply to increase the availability and demand for healthier food products, and decrease the availability and demand for unhealthy food products 3. Embed physical activity and healthy eating in everyday life 4. Encourage people to improve their levels of physical activity and healthy eating through comprehensive and effective social marketing 5. Reduce exposure of children and others to marketing, advertising, promotion and sponsorship of energy-dense nutrient-poor foods and beverages 6. Strengthen, skill and support primary healthcare and public health workforce to support people in making healthy choices 7. Address maternal and child health, enhancing early life and growth patterns 8. Support low-income communities to improve their levels of physical activity and healthy eating 9. Reduce obesity prevalence and burden among Indigenous Australians 10. Build the evidence base, monitor and evaluate effectiveness of actions The Taskforce report was well received by most expert commentators and was praised for its comprehensive and coherent program of action it recommended. Page 154 of 170

However, when the Government published their response to the Taskforce recommendations (Taking Preventative Action - A Response to Australia: The Healthiest Country by 2020) it proposed immediate action on only a few of the recommendations on obesity and rejected the merit of a number of interventions which had strong public support. Responsibility for further action on the Taskforce report has now been handed to the yet to established National Preventive Health Agency.

7.2 ACE OBESITY AND ACE PREVENTION PROJECTS The ACE Obesity report (2006) and the ACE Prevention report (2010) take a comparative approach to examining cost-effectiveness using a methodology that has been developed and tested over a range of previous projects. The ACE Obesity project examined 13 interventions selected by a panel and determined their likely population impact in reducing childhood and adolescent obesity together with the projected costs of each intervention based on estimates from current programs. The ACE Prevention project followed a similar process in examining 123 chronic disease prevention interventions, although only 27 are covered within the report. The ACE process involves utilising data from the efficacy of an intervention and modelling to predict the health benefits as body mass index (BMI) units saved and disability-adjusted life years (DALYs) saved. The process is capable of modelling from measured changes in behaviour in childhood to derive an outcome in terms of BMI change post intervention and projecting the impacts on DALYs over the child's lifetime (on the assumption that changes in BMI were maintained into adulthood). This methodology has its limitations but enables valid comparison of the potential impact of interventions, although comparisons must take into account the strength of the evidence used. A comparison of the 13 selected interventions in ACE obesity revealed that the likely health benefit varied considerably, as did the strength of the evidence from which that health benefit was calculated. The greatest health benefit is likely to be achieved by the 'Reduction of TV advertising of high fat and/or high sugar foods and drinks to children', 'Laparoscopic adjustable gastric banding' and the 'multi-faceted school-based programme with an active physical education component' interventions. In these times of increasing health care demands and finite resources, cost effectiveness analysis has become an important tool in the setting of health priorities. The generation of evidence that provides guidance on the expected health returns for the resources invested allows a comparison and ranking of interventions but there are a range of other factors that will also influence the merit or promise of various obesity interventions. The ACE process attempts to deal with some of these factors by requiring all interventions to be assessed against a set of “second stage filters’ which address issues such as feasibility, acceptability, transferability equity etc. A limitation of the ACE Obesity process is that it examined each intervention in isolation and neglected the interaction and synergies expected between interventions and with other system factors. The ACE prevention project uses special pathway analysis to take account of overlapping effects although it is not able to take account of the synergistic effects that arise from multiple complementary effects on the one behaviour. In addition the ACE Prevention project included modelling of packages of intervention to prevent diabetes in addition to single intervention.

Page 155 of 170

7.3 PRODUCTIVITY COMMISSION REPORT. CHILDHOOD OBESITY: AN ECONOMIC PERSPECTIVE This recent report adopts a behavioural economics perspective in appraising the merit of 27 selected child and student-focused prevention programs conducted in Australia, whose aims include promoting exercise, improving diet and cutting time spent in front of television and computer screens. The report concludes that, on balance, the evidence to date shows that many interventions to prevent childhood obesity show promise, but that the complex nature of the problem indicates that policies need to be carefully designed to maximise costeffectiveness and that any programs need to continue to gather evidence in order to refine Government responses. In terms of their economic focus, they argue that market incentives to provide information about the cause and prevention of obesity are weak, creating a role for government, but currently that role has not been clearly articulated or justified by the evidence. While this report has assimilated a significant volume of material, particularly on current programs in Australia, the methods, interpretation and conclusions of the report have some limitations. Most importantly, the authors have not considered some fundamental population health concepts, such as the effects of interventions on important intermediate behaviours, the potential population impact, possible delayed and synergistic effects of interventions, and the impact of changes in physical activity and diet on other chronic disease risk factors. The report considers each intervention as a single action, potentially equivalent, whether it is policy, a small intervention that would generally be implemented in combination with other actions, or a comprehensive setting-based intervention. Unlike the ACE report, this report does not take account of the potential population impact, which is the product of population reach and size of effect. Thus they argue that the regulation of food marketing would produce a small size of effect and thus is unlikely to have any significant impact on weight status ignoring the fact that the population reach and thus total population impact is large. In addition, the report considers all interventions in terms of their direct and immediate effects on weight status. Most interventions will not have direct effects on weight, but have direct effects on contributing and antecedent behaviours and/or attitudes which have indirect and/or delayed effects on weight. Some indirect effects enable behaviour change, or make it more likely that behaviour change will occur, leading in turn to changes in weight. If immediate behaviour change is expected, along with short term changes in weight, then population level interventions will continue to provide disappointing evidence of effectiveness. By considering each intervention as a single, stand-alone strategy, the report does not consider any delayed effects, or synergistic effects arising from the impact of multiple complementary interventions, or cumulative effects of multiple interventions. This is a fundamental omission, particularly when compared to the more sophisticated model and approach developed in the UK by the Foresight program.

Page 156 of 170

References ACE-Obesity. Assessing cost-effectiveness of obesity interventions in children and adolescents: Summary of Results. Melbourne, Victoria: Victorian Government Department of Human Services; 2006 Commonwealth of Australia. Taking preventative action – A response to Australia: The healthiest country by 2020 – The report of the National Preventative Health Taskforce Canberra: Commonwealth of Australia; 2010 Crowle J, Turner E. Childhood Obesity: An economic perspective. Productivity Commission Staff Working Paper. Australian Government Productivity Commission; 2010 National Preventative Health Taskforce (NPHT). Australia: the healthiest country by 2020. Technical Report No 1. Obesity in Australia: a need for urgent action. Canberra: Commonwealth of Australia; 2009 National Preventative Health Taskforce (NPHT). Australia: the healthiest country by 2020. National Preventative Health Strategy – the roadmap for action. Canberra: Commonwealth of Australia; 2009 Vos T, Carter R, Barendregt J, Mihalopoulos C, Veerman L, Magnus A, Cobiac L, Bertram M, Wallace AL, ACE-Prevention Team. Assessing cost-effectiveness in prevention (ACEPrevention): Final report. Brisbane & Melbourne: University of Queensland & Deakin University; 2010

Page 157 of 170

8.0 INTERNATIONAL INITIATIVES TO SUPPORT THE DEVELOPMENT OF COMPREHENSIVE OBESITY PREVENTION INTERVENTIONS It is possible to identify a considerable number of initiatives, both within Australia and internationally, which set out recommendations or policy frameworks for effective action on obesity. However, at present it is difficult to identify any single country or jurisdiction that has developed, implemented and evaluated a truly comprehensive obesity prevention strategy. However, in recent years there have been a number of projects that provide valuable guidance on appropriate approaches to help ensure action on obesity is coherent, consistent and comprehensive. Some of these are listed below

8.1 INTERNATIONAL OBESITY TASKFORCE (IOTF) FRAMEWORKS FOR ACTION The IOTF have provided a number of policy and planning tools with the objective of supporting and encouraging improved action on the prevention and management of obesity. The Sydney principles for reducing the commercial promotion of foods and beverages to children (Swinburn et al, 2008) were developed to support and guide the formation of an International Code on Food and Beverage Marketing to Children. The set of seven principles provide a framework for action on changing food and beverage marketing practices that target children. The Principles state that actions to reduce marketing to children should: (i) support the rights of children; (ii) afford substantial protection to children; (iii) be statutory in nature; (iv) take a wide definition of commercial promotions; (v) guarantee commercial-free childhood settings; (vi) include cross-border media; and (vii) be evaluated, monitored and enforced. The IOTF framework for translating evidence into action (Swinburn et al, 2005) is defined by five key policy and programme issues that form the basis of the framework. These are: (i) building a case for action on obesity; (ii) identifying contributing factors and points of intervention; (iii) defining the opportunities for action; (iv) evaluating potential interventions; and (v) selecting a portfolio of specific policies, programmes, and actions. Each issue has a different set of evidence requirements and analytical outputs to support policy and programme decision-making.

8.2 EU PORGROW PROJECT The EU funded PorGrow project aimed to identify potential policy initiatives to tackle the growing rates of obesity in the countries of the EU (Lobstein & Millstone, 2006). It did this by interviewing farmers and food manufacturers, retailers, caterers and advertisers, teachers, sports and physical activity organisations, public health experts, advocacy groups, consumer representatives and others, in each of nine EU member states: The results reveal a broad consensus of opinion that a portfolio of measures will be needed to slow and then reverse the rising trend in the incidence of obesity, supported by a general acceptance that the costs of the various policy options are less important than their social and health benefits, efficacy, acceptability and practical feasibility. The proposed portfolio included: • Educational initiatives, supported by improved access to information, healthier foods, and opportunities for physical activity. • Mandatory and improved nutrition labelling and controls on marketing terms were considered more feasible and socially acceptable than controls on advertising, but controls on advertising were considered as potentially more effective than other informational options in tackling obesity. Page 158 of 170

• There was widespread antipathy to fiscal interventions, such as taxes on 'unhealthy' foods or subsidies on 'healthy' ones. Controls on food composition were considered effective in tackling obesity, and were widely considered to be both feasible and acceptable. • High levels of additional social and health benefits were anticipated from changes in transport and planning policies, but the costs to the public sector were considered high and the implementation difficult and long-term. Improved provision of and access to sports and physical recreational facilities were highly regarded. • 'Technological' solutions, such as increasing the use of artificial sweeteners and fat substitutes, the use of pedometers and the use of medication for weight control, were widely considered ineffective and unacceptable for tackling the obesity epidemic. • Reform of the Common Agricultural Policy, from a health perspective, was considered socially desirable and acceptable but costly and difficult to implement.

8.3. EU HOPE PROJECT The HOPE project (Health promotion through obesity prevention across Europe: an integrated analysis to support European health policy) was a three year project established to examine the evidence base and create a set of scenarios to evaluate strategies and potential interventions directed at the key determinants of obesity in the European Union European policies on Obesity. The HOPE project brought together a network of experts involved in research and policy in both western and Eastern Europe. One stream within this project examined existing policy initiatives with Europe and other aligned countries to determine common approaches, level of implementation and evaluation and identify areas for improvement. At present the results of the project are only available in draft or summary formats (van Rijs, 2010). One of the clearest points to emerge from the report is that there is no shortage of Policies – the project identified 120 obesity, nutrition and physical activity policies from 31 different European states. However, they also identified a shortage of dedicated financial and human resources, and poorly sustained and politically backed implementation for these policies. Environmental change strategies were included frequently in these polices. Common components which featured often were safe transport routes, land use planning, development of community sports facilities, and catering (often in schools). A few policies specifically mentioned introducing measures to address marketing and advertising of foods to children, and a similar number proposed actions to encourage food reformulation by manufacturers. Economic measures were identified infrequently, and were often proposals rather than commitment to definite action. Relatively few evaluations of obesity prevention policies were identified, although information on policy implementation and impact had been collected for all of the Scandinavian countries, Germany, Slovenia and the UK. Most evaluations relied on data from national surveys with little attempt to assess the effect of the different policy components (e.g. legislative policies, government guidance, information campaigns, financial support of relevant local infrastructures, and many others). National surveillance systems of diet, physical activity, and prevalence of obesity were identified as invaluable in tracking changes over time. More recently there have been attempts to develop measures/indicators of environmental change that could impact on diet and physical activity patterns with the intention of including such measures in future national surveys.

Page 159 of 170

The HOPE review made a number of recommendations in relations to improving the effectiveness of policy actions to address obesity within the EU. These included: • Where there are existing obesity prevention polices, these should be assessed to determine the value of additional elements which would enable a more coherent and integrated approach. • Policies should be flexible enough to accommodate new research findings as they come on-stream. • Research funding bodies within countries and from the EC should give a higher priority to research that assesses the effectiveness of interventions which can form components of a comprehensive obesity prevention strategy, using a variety of different methodologies. • A comparative analysis of policies and policy led interventions should be undertaken in countries with different levels and rates of growth of obesity to help assess the ‘quality’ of policies. • Carrying out some in-depth case studies exploring the links between integrated policies which have supported behavioural change and obesity outcomes. • All member states should identify and allocate dedicated financial and human resources to enable implementation and evaluation of obesity prevention policies. This needs to be accompanied by a long-term, high-level, cross party, commitment to take forward agreed actions to tackle obesity. 8.4 CDC'S RECOMMENDED STRATEGIES FOR OBESITY PREVENTION In 2009, the CDC released its first comprehensive set of recommended strategies and measures to help communities tackle the problem of obesity through environmental change and policies that promote healthy eating and physical activity (Kahn et al, 2009). Based on an assessment of the evidence, this report recommends that communities should do the following: • Increase availability of healthier food and beverage choices in public service venues. • Improve availability of affordable healthier food and beverage choices in public service venues. • Improve geographic availability of supermarkets in underserved areas. • Provide incentives to food retailers to locate in and/or offer healthier food and beverage choices in underserved areas. • Improve availability of mechanisms for purchasing foods from farms. • Provide incentives for the production, distribution, and procurement of foods from local farms. • Restrict availability of less healthy foods and beverages in public service venues. • Institute smaller portion size options in public service venues. • Limit advertisements of less healthy foods and beverages. • Discourage consumption of sugar-sweetened beverages. • Increase support for breastfeeding. • Require physical education in schools. • Increase the amount of physical activity in physical education programs in schools. • Increase opportunities for extracurricular physical activity. • Reduce screen time in public service venues. • Improve access to outdoor recreational facilities. • Enhance infrastructure supporting bicycling. • Enhance infrastructure supporting walking. • Support locating schools within easy walking distance of residential areas. • Improve access to public transportation. • Zone for mixed-use development. • Enhance personal safety in areas where persons are or could be physically active. • Enhance traffic safety in areas where persons are or could be physically active. • Participate in community coalitions or partnerships to address obesity Page 160 of 170

8.5 INSTITUTE OF MEDICINE - BRIDGING THE EVIDENCE GAP Bridging the Evidence Gap (IOM, 2010) is a new report from the Institute of Medicine which sets out a proposed framework for evidence-informed decision-making in obesity prevention efforts. While focused on obesity prevention, this framework could guide more general efforts to assess and use scientific evidence in complex, multifactorial public health challenges. The obesity prevention framework focuses on approaches for assessing policyand community-level interventions designed to influence food, eating, and physical activity environments. It is guided by the need for a systems approach that explicitly takes into account the social contexts in which decisions are made and the multiple interacting determinants of policy and community action. The report is based around the use of the L.E.A.D. framework, short for Locate evidence, Evaluate it, Assemble it, and Inform Decisions. Decision makers, their intermediaries, and researchers can apply the L.E.A.D. framework and its innovative process for generating, identifying, evaluating, and assembling evidence to inform the decisions that must be made about obesity prevention. The report is currently published in draft format and responses to its content are being sought. 8.6 THE UK FORESIGHT REPORT AND PROJECT The UK Foresight Project ‘Tackling Obesities – Future Choices’ was commissioned by the Government Office for Science and Innovation in July 2005. Over 300 experts from a range of disciplines and stakeholders from within, and beyond Government, have reviewed the evidence base and used a variety of techniques to examine the question: ‘How can we deliver a sustainable response to obesity over the next 40 years?’ New research was also commissioned to fill in some of the gaps and cast new light on the determinants of obesity and its likely prevalence in the future. The Foresight project report ‘Tackling Obesities – Future Choices’ was published in October 2007. The report has used the scientific evidence base from across a wide range of disciplines in order to identify the broad range of factors that influence obesity; create a shared understanding of the relationships between key factors influencing levels of obesity and their relative importance; build on this evidence to identify effective interventions; and analyse how future levels of obesity might change and identify the most effective future responses. The UK Foresight Report has identified a number of fundamental considerations in designing a strategy for obesity or chronic disease prevention. These include: • No single intervention is sufficient or pre-eminent. Multiple layers of change are required, at organisational, population and policymaker levels. These interact with each other, so that population changes may trigger political support for policy initiatives, while organisational and environmental changes can enable behaviour changes amongst community members. Individual and organisational changes are based on social motivations and drivers, and these are legitimate points for government action. The initial aim is for social and individual priorities to favour healthy behaviour change, or to make it more likely. • The system map confirms that a diverse set of responses is required to address obesity, with changes designed to amplify and enable other changes, and then accelerate and accumulate over time. • The generational nature of obesity means that a portfolio of interventions for adults and children is optimal. • A large scale and range of environmental, organisational and behavioural change is Page 161 of 170

required. • Government responses will vary according to values along two key dimensions: (i) focus on individual or social responsibility; and (ii) emphasis given to anticipatory or repair interventions. Models with higher levels of social responsibility and anticipatory action produce most change, most quickly. The Foresight project recommends an approach of ‘practice-based evidence’, which involves generating evidence through ongoing implementation studies. This involves immediate action, applying promising interventions, and progressively redressing the current evidence gaps. In 2009, the National Centre for Clinical Excellence in the UK commissioned a four year project to examine a whole-system approach to prevention of obesity based on the Foresight model.

Page 162 of 170

References Butland B, Jebb S, Kopelman P, McPherson K, Thomas S, Mardell J, Parry V. Foresight. Tackling Obesities: Future Choices- Project Report 2nd edition. Government Office for Science, UK. 2007 Institute of Medicine (IOM). Bridging the Evidence Gap in Obesity Prevention: A framework to inform decision making. Washington DC: The National Academies Press; 2010 Kahn LK, Sobush K, Keener D, Goodman K, Lowry A, Kakietek J, Zaro S. Recommended community strategies and measurements to prevent obesity in the United States. CDC Morbidity and Mortality Weekly Report 2009; 58(No. RR-7): 1-29 Lobstein T, Millstone E. Policy options for responding to obesity: Summary report of the ECfunded project to map the view of stakeholders involved in tackling obesity – the PorGrow project. University of Sussex; 2006 Swinburn B, Gill T, Kumanyika S. Obesity prevention: a proposed framework for translating evidence into action. Obesity Reviews 2005; 6:23-33 Swinburn B, Sacks G, Lobstein T, Rigby N, Baur LA, Brownell KD, Gill T, Seidell J, Kumanyika S, International Obesity Taskforce Working Group on Marketing to Children. The 'Sydney Principles' for reducing the commercial promotion of foods and beverages to children. Public Health Nutrition 2008; 11(9):881-6 Van Rijs M. An introduction to the HOPE project. Viewed on 2010 November 1. Available from: www.hopeproject.eu

Page 163 of 170

9.0 DEVELOPING A COMPREHENSIVE OBESITY PREVENTION STRATEGY Evidence reviews and experience from previous public health initiatives are a crucial step that informs the development of a comprehensive program of action to address public health problems. However, the findings from such reviews do not define the structure and content of these strategies. The National Public Health Partnership's Planning Framework for Public Health Practice (NPHP, 2000) sets out a more structured process of planning which integrates this knowledge and results in the development of a portfolio of action that provides the best mix of interventions to effectively address community weight issues. There is a clear need for national and state leadership and policy development on the prevention of obesity, however, as a large component of any portfolio of action will be implemented and evaluated at the local level, the selection of the best mix of interventions will need to be developed at that level, albeit within a larger regional, state or national framework. 9.1 DEFINING APPROPRIATE OBJECTIVES Despite the difficulty of achieving total energy balance, weight stability and a reduction in the level of overweight and obesity in the community, it is important that objectives of any program or intervention addressing this problem must be clearly related to achieving energy balance and preventing weight gain. This does not mean that there should be an excessive focus on weight and weight loss, but rather that an intervention be planned to achieve sufficient impact on dietary intake and/or energy expenditure to influence energy balance and weight status. Imprecise objectives relating to healthy eating or active living are not an accurate reflection of the changes sought. Weight gain prevention is not achieved by merely putting a range of interventions focussing on nutrition and physical activity into a healthy weight program 9.2 WHO SHOULD OBESITY PREVENTION STRATEGIES TARGET? Deciding where to invest limited time and resources in obesity prevention is a difficult task but finite health resources make this a necessity. The WHO has identified three distinct but equally valid and complementary levels of obesity prevention (Figure 6). The specific ‘targeted’ approach directed at very high-risk individuals with existing weight problems is represented by the core of the figure, the ‘selective’ approach directed at individuals and groups with above average risk is represented by the middle layer, and the broader universal or population-wide prevention approach is represented by the outer layer. This replaces the more traditional classification of disease prevention (primary, secondary, and tertiary), which can be confusing when applied to a complex multifactorial condition such as obesity.

Page 164 of 170

Figure 6. Levels of obesity prevention intervention. Universal prevention is the domain of public health, whereas selective and targeted prevention are predominantly dealt with in community and health care service settings. Community settings include schools, colleges, worksites, community centres, and shopping outlets.

Adapted from Gill, 1997 9.2.1 Whole community (universal) Overweight and obesity are public health problems of relevance to the whole community and require strategies that focus on population-wide change rather than attempting to address individuals or small groups in isolation of the community in which they live. An effective population strategy needs to both improve population knowledge about obesity and its management and reduce the exposure of the community to obesity-promoting factors in the environment. Action at a population level requires coordination at a central level and the investment of resources to be maintained over a long period of time to achieve population change. Family focus There are numerous reasons why children should be a major focus of any obesity prevention strategy. There is strong evidence that a high proportion of overweight or obese children will become obese adults. Childhood obesity also has immediate effects on health, and weightrelated conditions are becoming more prevalent and their effect more pronounced as the rates of childhood obesity increase. However, children grow rapidly and increase the level of lean body mass as they age, and so reducing or keeping fat mass constant allows the normalisation of weight over time. Thus, childhood (particularly younger children) is a period during which prevention efforts have a higher chance of success. However, children also have little direct control over the environment in which they live. Parents and other caregivers mostly control decisions regarding the food available and the opportunities for activity. In addition, the behaviours of parents and other siblings have a profound effect on the diet and physical activity behaviours of children. For this reason, it is preferable to focus childhood obesity prevention efforts on the family environment rather than directly on children. 9.2.2 High-risk groups (selective) There are a number of groups that appear to be at higher risk of developing overweight and obesity. These groups warrant special attention and include the following: • Those with a family history of weight problems • Socially disadvantaged and isolated communities • Indigenous communities • Certain ethnic groups • Smokers who have recently quit smoking • Those that have recently lost weight Page 165 of 170

In addition, there are certain times in a person’s life when the person is more prone to weight gain. These age groups could be considered for selective prevention interventions. These times include the following: • Prenatal • Adiposity rebound (5–7 years) • Adolescence • Early adulthood • Pregnancy • Menopause 9.2.3 Those with an existing weight problem (targeted) In developing weight gain prevention strategies, it is important not to neglect those with an existing weight problem who could benefit from more intensive efforts to help prevent further weight gain.

9.3 SELECTING A PORTFOLIO OF ACTIONS Even after potential interventions have been assessed for their potential to help achieve and maintain energy balance, there a number of decisions that must be made to achieve the best mix of actions in the intervention portfolio. These can be addressed in a logical and sequential manner but in the end the composition of a local portfolio will be greatly influence by community structures, services and capacities. Some issues to consider include: a. Choose the most promising interventions from within each action area b. Choose interventions from across all action areas – this may require the addition of interventions with less certainty of evidence by high potential to influence energy balance c. Within action areas choose a mix of focussed behaviour change strategies and enabling and amplifying actions d. Focus on universal actions but include more targeted interventions to address high risk groups. e. Try to combine complementary interventions to maximise the intensity and impact It is very unlikely that all the selected interventions within a portfolio will be implemented simultaneously. Implementation is usually a staged process with different groups and sectors involved in the implementation of a number of single or multi-component interventions individually or as part of a broader community approach. Therefore a wellconstructed portfolio will have the greatest capacity to effectively address the key drivers of obesity when implemented in its entirety over time. To achieve this, it is important to ensure that the process of implementation follows best practice and interventions are maintained for a sufficiently long period.

Page 166 of 170

References Gill TP. Key issues in the prevention of obesity. British Medical Bulletin 1997; 53(2):359–388 National Public Health Partnership (NPHP). A planning framework for public health practice. Melbourne: 2000

Page 167 of 170

10. CONCLUSIONS AND RECOMMENDATIONS The problem of obesity has continued to worsen over the past few decades in Australia and has reached a point where six out of every ten Australian adults and one in every four children has been classified as having a weight problem. The prevalence of obesity in children may be stabilising now but there continues to be rapid rises in the rates of obesity in young adults. As a consequence, the health, social and economic consequences of obesityrelated disease (especially type 2 diabetes) will continue to rise over the coming decades unless serious effort is directed at reducing weight gain in the community. The genesis of obesity is complex and the drivers of energy imbalance and weight gain operate at many levels. This has been well illustrated by the development of the Foresight Project Obesity System Map which has highlighted the huge array of factors that influence controls on energy balance and their interconnectivity. It has also indicated the futility of attempting to address obesity by selectively addressing only a few of these issues. Obesity is a “whole of system” problem that requires a “whole of system” solution. This project undertook a rapid review of the existing literature to assess and assemble the body of evidence that supports actions within various domains of action within the obesity system and came to the following conclusions: • The level and quality of experimental evidence (as assesses by NHMRC evidence grades) is limited across all action areas and so the acceptance or rejection of proposed interventions needs to be based on an assessment of the total body of information available to gauge its potential contribution to the achievement and maintenance of energy balance. No intervention should be automatically ruled in or out until this is undertaken. • The evaluation of obesity interventions have not been undertaken in a consistent or structured fashion, making an assessment of the merit of a specific intervention difficult and the integration of results from several intervention studies through systematic reviews or meta-analyses of limited value. There is a need to set consistent evaluation timeframes, define an agreed hierarchy of outcomes and use the same measurement for changes in these outcomes. • There has been an over-reliance on systematic reviews to define the level of evidence to support specific obesity prevention initiatives. More effort needs to be invested in gathering a more complete body of information which can be used to guide decisions around the potential of an intervention to contribute to the achievement and maintenance of energy balance. • Evaluations of obesity prevention initiatives have focussed on defining “what” works. However, the achievement of a defined outcome is a product of both the content of the intervention together with the quality of the implementation process. More effort needs to be devoted to defining how to deliver specific interventions within the context of different communities, rather than just defining what to do. • An alternative evidence assessment scheme needs to be developed which is based on an assessment of the total body of information and better defines the potential of an intervention or program of action to contribute to the achievement and maintenance of energy balance, when implemented as part of a comprehensive program of action. Such a system should help define research approaches that are appropriate to generate this total body of evidence.

Page 168 of 170

• Very few packages of action have been evaluated for their combined impact on energy balance, although some recent, whole of community programs such as EPODE are the exception. This would seem to be the preferred mode of evaluation, given that many individual interventions are not designed to have a direct effect on weight status or weightrelated behaviours, but rather are designed to interact with other interventions to enable of amplify behaviour change. Such an approach is of particular relevance to the evaluation of interventions that are primarily enabling or amplifying actions. They should be judged on their potential to contribute as part of a program of action to the achievement and maintenance of energy balance. • The greatest volume of experimental evidence currently exists for interventions undertaken in everyday life settings such as school, worksites and day-care centres. However, no attempt has been made to integrate the wider body of information that could assist with judgements about the merit of interventions with this experimental data. Most reviews of this evidence show a positive impact of these types of intervention on the weight status of children and adults with the greatest benefit shown for multi-component interventions. However, because the content and context of these interventions vary so much, and the outcomes are not always defined by the same measures, it is difficult to make clear assessments of the merit of individual approaches or to combine individual studies into systematic reviews or meta-analyses. As a consequence many systematic reviews are unable to reach meaningful conclusions. Multi-component, community-wide interventions is the area with the most consistent and promising evidence. • Contrary to current perceptions, front of pack labelling and statutory restrictions on food marketing to children have the broadest base of evidence, with consistent findings to support their implementation. Menu labelling and changes to the physical environment to increase physical activity also has sufficient evidence to support a modest degree of promise. Evidence for action in these areas comes from small experimental studies, ecological assessments, observational studies, consumer research, parallel evidence and logic. • The evidence around the merit of taxation and subsidies needs clarification. Most analyses have been conducted in the USA and although there is some observational data, most of the assessments rely on modelling of impact. Most studies demonstrate a small but potentially meaningful impact when taxes are imposed on particular products, and point to more merit in taxation of specific food products such as soft drinks. Taxation disincentives appear to have a stronger impact on behaviour than incentives through subsidies, although subsidies may need to be part of any fiscal strategy to offset equity concerns of taxation. • We have no data on the price elasticity of demand around food products in Australia to allow effective modelling of taxation/subsidy impacts. • Very few interventions within the action areas addressing the food supply, food access and availability and active environments, were able to demonstrate a direct impact on weight status. Those studies that did were usually small scale and targeted at motivated groups. However, there is sufficient evidence that such interventions impact positively on weightrelated behaviours, and additional support for the promise of these interventions from observational data, parallel evidence and program logic. • Interventions to improve food and physical activity environments are important levers for addressing current inequalities in opportunities to eat well and be physically active. Interventions in these actions areas have considerable synergies with actions to address climate change and other environmental concerns.

Page 169 of 170

• There is value in identifying interventions to specifically address weight status within disadvantaged groups. Although such groups benefit from interventions which address the whole of the community, adaptations and specifically focussed implementation that addresses social or cultural complexities are often needed to ensure effectiveness with more disadvantaged groups. Lessons can be taken from successful actions to address other public health problems such as tobacco, alcohol and road safety or from community-wide programs addressing other chronic diseases such as diabetes and heart disease. Whilst this has been able to enlighten discussions around potential approaches, care must be taken to interpret the nature of this parallel evidence and the context in which these actions occurred to ensure that the learning is applied correctly to obesity prevention. Over the last two decades there has been a considerable amount of reflection and policy development around obesity prevention within Australia. National obesity prevention frameworks have been developed for children and for adults and older Australians. In addition all jurisdictions have developed their own strategies for dealing with the problem of obesity and funded a range of major obesity prevention trials or initiatives. Recent academic or policy analysis projects have provided an important basis for discussion around what interventions should be included in the comprehensive package of programs to address obesity. However, each of these analyses has failed to come to terms with the issues surrounding “whole of system” approaches to obesity prevention. There has also been significant action at an international level which provides guidance on comprehensive programs of action to address the drivers of obesity. Important reports from the International Obesity TaskForce, the European Union, the US Centre for Disease Control and Prevention and the Foresight Project have all produced valuable analyses and guidelines on preventing obesity at the community-level. Constructing a comprehensive portfolio of actions to address overweight and obesity in Australia will require an ongoing process of evidence assessment and review. It is unlikely that there will ever be a definitive body of experimental evidence to guide all aspects of this process, but the collation, assessment and interpretation of all sources of evidence using an appropriate framework such as that applied in this document, will ensure action is not deferred whilst ensuring that decisions are made on the best available evidence.

Page 170 of 170