Front-of-Pack Nutrition Rating Systems and Symbols in Canada

by

Teri Elyse Emrich RD, MPH

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Department of Nutritional Sciences University of Toronto

© Copyright by Teri Emrich 2014

ii

Front-of-Pack Nutrition Rating Systems and Symbols in Canada

Teri Elyse Emrich Rd, MPH

Doctor of Philosophy

Department of Nutritional Sciences University of Toronto

2014

Abstract

Front-of-pack (FOP) nutrition rating systems and symbols provide simplified information on the

nutritional content of foods. Despite concerns that the presence of multiple FOP systems may be

confusing and misleading to consumers, the government has rejected recommendations to

implement a single, standardized FOP system claiming that the current unregulated marketplace

is already meeting the needs of Canadians. The overall objective of this work was to generate

Canadian evidence to determine the legitimacy of the government’s position.

The specific objectives were to evaluate the extent to which foods meeting the criteria of

voluntary FOP systems are identified by the system’s symbol, to compare the nutritional quality

of foods with and without FOP systems, to describe Canadians’ attitudes towards FOP systems,

and to determine experimentally the most liked and understood FOP system by consumers. We

generated evidence through three studies.

Studies 1-2 employed a 2010 database of Canadian packaged food labels. Compared to products

without FOPS, products with FOPS were not being used to market foods consistently lower in

calories, saturated fat, sodium, and sugar in 10 food categories and 60 subcategories.

Significantly many more products qualified to carry two major Canadian FOPS than actually

displayed these FOPS on their label.

Study 3 used a national survey panel to test Canadians preferences for four FOPS. 86% of

consumers supported the implementation of a single, standardized FOPS. Nutrient-specific

systems (a traffic light and a Nutrition Facts table based FOPS) were preferred and rated more

consumer friendly than other systems.

iii

In summary, FOPS are not being used to promote the healthiest products in the Canadian

marketplace, and could be misleading consumers. Despite the government’s position not to

adopt a single, standardized FOPS, Canadian consumers would support such an initiative and

would prefer a nutrient specific FOPS.

iv

Acknowledgments

I would like to thank my supervisor, Dr. Mary L’Abbe, for four years of continued support and

mentorship. Dr. L’Abbe continually encouraged me to take advantage of whatever opportunities

came my way, be they academic, professional, or personal, and was instrumental in opening the

doors that led to so many of those opportunities. Over the course of my PhD, I have worked

harder and learned more than during any other period of my life, but Dr. L’Abbe always

reminded me to make time for fun and to celebrate every accomplishment – no matter how

minor.

My thesis committee members, Dr. Joanna Cohen, Dr. Marco DiBuono, and Dr. Wendy Lou,

have always been available to provide thoughtful input into my work, bringing in fresh

perspectives that made the research much richer. I am grateful to these three people for investing

their valuable time and energy in my development as a professional and a researcher. I am also

to grateful to Dr. David Jenkins and Dr. Sara Kirk who served as my external examiners for my

Departmental and Senate defenses, respectively. They both challenged me in ways that made me

see my work from a new perspective. Laura Pasut has been my professional mentor for the later

years of my PhD. Her insights have opened my eyes to the possibilities ahead of me and have

helped to focus my sights on life after my PhD.

In the last four years I have been so fortunate to work with a wonderful and diverse group of

fellow graduate students and trainees who were readily available to lend a helping hand

whenever needed. Dr. Joanne Arcand, has made a significant contribution to the training of all

the students in Dr. L’Abbe’s research group, myself included. I have learned so much from her

about the discipline and dedication required of academic research. I am also grateful for the

contribution of Alyssa Schermel to the development of the Food Label Information Program,

without which much of my research would not have been possible. Christina Wong and Mary

Scourboutakos have been my constant companions on this journey. They commiserated with

me, provided listening ears, and made me feel that I was never alone in this. Ying Qi, of Dr.

Lou’s research group, made a significant contribution my work and exhibited endless patience

when working with me on my analysis, for which I am extremely grateful. Julio Mendoza was

instrumental in my work with the Canadian Consumer Monitor, and his management of this

consumer panel facilitated my survey work.

v

My family was my biggest cheerleader over the course of my graduate program and encouraged

me to persevere and to keep moving forward. I thank my parents, Margo and Walter, my sisters

Katie and Andi, and my nephews Jack and Henry – they were my escape from my academic

world and have brought my life tremendous joy. My grandparents, Jack and Anne Disher, I have

never felt that anyone is more proud of my accomplishments and me than these two. A phone

call with them could help restore my confidence in myself even on my toughest day.

Finally, to my wonderful partner in life Aaron, you have been my biggest supporter on this

journey. My pursuit of this dream meant that many aspects of our life together were put on hold,

but you never complained and always saw the bright side. You bore the brunt of my stress and

anxiety and were there to talk me down whenever I needed you. I am so happy that you came

through this with me and I am excited to finally begin the next chapter of our life together.

I have been fortunate to receive support for my research and training from several organizations.

The Canadian Institutes of Health Research provided funding through a Banting and Best

Doctoral Scholarship, a Population Intervention for Chronic Disease Prevention: A Pan-

Canadian Program Training Fellowship, and a Public Health Policy Training Fellowship. In

particular, the Population Intervention for Chronic Disease Prevention contributed to my

research and training far beyond the funding provided with its most significant contributions

coming through the training and fellowship opportunities it provided. The Advanced Foods and

Materials Network also provided funding early in my training.

vi

Table of Contents

Acknowledgments iv

Table of Contents vi

List of Tables ix

List of Figures x

List of Abbreviations xi

List of Appendices xii

Chapter 1 1

1 Introduction 1

Chapter 2 3

2 Background and literature review 3

2.1 Diet related chronic disease 3

2.2 An overview of nutrition information on food labels 4

2.2.1 Nutrition labelling 4

2.2.2 Nutrition marketing 6

2.3 An overview of FOP nutrition rating systems and symbols 11

2.3.1 History of FOP symbols 12

2.3.2 FOP nutrition rating system and symbol categories 13

2.3.3 Nutrient profiling 17

2.4 FOP systems and the nutritional quality of foods 20

2.4.1 FOP systems and reformulation 20

2.4.2 The nutrient profiles of foods with and without FOP systems 22

2.5 Influence of FOP systems on perceptions of nutritional quality 24

2.6 Is there a need for a single, standardized FOP system? 27

2.6.1 The ideal FOP system for use with consumers 28

2.7 Summary 35

Chapter 3 36

3 Scope and Hypotheses of Thesis 36

3.1 Scope and Objectives 36

vii

3.2 Specific hypotheses 36

3.3 Preview of Chapters 4-6 37

Chapter 4 39

4 Food products qualifying for and carrying front-of-pack symbols: a cross-

sectional study examining a manufacturer led and a non-profit organization led

program 39

4.1 Abstract 40

4.2 Background 42

4.3 Methods 45

4.4 Results 50

4.5 Discussion 54

4.6 Conclusions 57

Chapter 5 59

5 Front-of-pack symbols are not a reliable indicator of products with healthier

nutrient profiles 59

5.1 Abstract 60

5.2 Introduction 61

5.3 Methods 65

5.3.1 Data analysis 66

5.4 Results 66

5.4.1 Differences between products with and without FOP symbols by food category 67

5.4.2 Differences between products with and without FOP symbols by food subcategory 67

5.4.3 Comparison of different FOP symbol types 68

5.5 Discussion 72

5.5.1 Limitations 73

5.5.2 Conclusions 74

Chapter 6 75

6 Consumer perceptions of the Nutrition Facts table and Front-of-Pack nutrition

rating systems 75

6.1 Abstract 76

6.2 Introduction 77

6.3 Materials and Methods 78

viii

6.3.1 Participants 78

6.3.2 Survey Design and Procedure 79

6.3.3 Statistical Analyses 80

6.4 Results 82

6.4.1 Participants 82

6.4.2 Consumer friendliness of FOP systems 82

6.4.3 Effect of NFt on consumer friendliness of FOPs 82

6.4.4 Conceptual Understanding 82

6.4.5 Effect of FOP on perceived overall healthiness and nutrient content 83

6.4.6 Consumer preferences for FOP systems 84

6.5 Discussion 93

6.5.1 Conclusion 95

Chapter 7 96

7 Discussion 96

7.1 General Discussion 96

7.1.1 Evaluation of existing FOP system policy 97

7.1.2 Evaluation of potential FOP system policy 100

7.2 Future directions 101

7.2.1 Policy 101

7.2.2 Research 109

Chapter 8 112

8 Summary and Conclusions 112

References 114

Appendices 128

Appendix A 129

Appendix B 134

Appendix C 136

Copyright Acknowledgements 145

ix

List of Tables

CHAPTER 4

Table 4-1 Nutrient criteria for crackers to qualify for Health Check™ and Sensible Solutions™ symbols

CHAPTER 5

Table 5-1 Comparison of the nutrient content per reference amount of products with and without front-of-pack nutrition rating symbols*

CHAPTER 6

Table 6-1 Demographic characteristics of survey respondents (n=3029)

Table 6-2 Mean scores of consumer friendliness by FOP system and NFt

Table 6-3 Mean scores of statements describing the NFt and FOP systems

Table 6-4 Mean healthiness, calorie and nutrient content scores of a frozen meal and breakfast cereal by NFt and FOP

CHAPTER 7

Table 7-1 US Food and Drug Administration and US Department of Agriculture nutrient profiling criteria for “healthy” claims

x

List of Figures

CHAPTER 2

Figure 2-1 Nutrition information on prepackaged foods in Canada

Figure 2-2 Examples of different front-of-pack symbol types

Figure 2-3 Conceptual model of FOP system use

CHAPTER 4

Figure 4-1 Front-of-pack symbols evaluated in the present study

Figure 4-2 Proportion of food products that qualified for, compared with the proportion of food products that carried, the different front-of-pack symbols

CHAPTER 5

Figure 5-1 Examples of different front-of-pack symbol types

Figure 5-2 Difference in calorie, saturated fat, sodium, and sugar content per reference amount between products with and without front-of-pack symbols by symbol type

CHAPTER 6

Figure 6-1 Front-of-Pack Nutrition Rating Systems evaluated in the randomized mock package experiment.

xi

List of Abbreviations

ANCOVA Analysis of Covariance

CCM Canadian Consumer Monitor

FLIP Food Label Information Program

FOP Front-of-Pack

NFt Nutrition Facts table

US United States

xii

List of Appendices

Appendix A Proportion of food products that qualified for Health Check™ compared to the proportion of food products that carried the system's symbol by subcategory (N=7503)

Appendix B Proportion of food products that qualified for Sensible Solutions™ compared to the proportion of food products that carried the systems symbol by subcategory (N=3009)

Appendix C Comparison of the nutrient content per reference amount of products with and without front-of-pack nutrition rating symbol nutrition marketing by product subcategory

1

Chapter 1

1 Introduction

Globally, diet-related chronic disease – including obesity, diabetes, cardiovascular disease,

cancer, osteoporosis, and dental diseases – are on the rise (1). Within Canada, three of these

conditions – cardiovascular disease, cancer, and type 2 diabetes – are among the leading causes

of death (2). In order to prevent these diet-related chronic diseases, the World Health

Organization recommends that healthy populations limit their intake of saturated fat, trans fat,

cholesterol, sugar, sodium, and total energy while ensuring adequate intakes of fat,

polyunsaturated fat, carbohydrate, protein, fibre, and vitamins and minerals (1,3). Similarly,

2007’s Eating Well with Canada’s Food Guide recommends Canadians limit foods and

beverages high in calories, fat, sugar or salt, as well as trans fat while meeting their needs for

vitamins, minerals and other nutrients (4).

In order to make healthy food choices, in line with the above-mentioned recommendations, the

WHO suggests that “consumers require accurate, standardized and comprehensible information

on the content of food items” (3). To that end, Canada has been actively engaged in providing

consumers with information on the nutritional content of foods through regulated nutrition

labelling since 2003 (5). However, research is beginning to show that the nutrition labelling

regulations have not been entirely effective in helping Canadians make healthier choices. While

the majority of consumers report reading food product labels at least sometimes, actual use of

food labels while shopping has been found to be much lower than reported in international

studies (6-9). In addition, while consumers commonly report understanding nutrition labels,

actual understanding is more moderate than reported (7-11). Many Canadian consumers are

confused about the technical and numerical portions of nutrition labelling, while others struggle

to put this information in the context of a healthy diet (12). In some cases, food label

understanding is hampered by consumers’ lack of knowledge of nutrient requirements and

reference amounts (10). Consumer confusion and uncertainty about one or more elements of the

nutrition label impedes their ability to use the label to guide their food choices. Indeed, experts

have raised concerns that the current labels are too complex and require too much time to

interpret (13). As a result, the House of Commons Standing Committee on Health, as well as

government working groups, have called for the development and implementation of a simplified

2

front-of-pack (FOP) nutrition rating system or symbol (13,14). Internationally, several

governments have already begun to pursue such a strategy (15). Both the Australian and United

Kingdom governments have introduced standardized, voluntary FOP systems with government-

developed nutrient profiling criteria (16,17). Australia has developed a health star FOP system

while the United Kingdom has developed a traffic light FOP system.

In contrast, the current lack of specific regulations for FOP nutrition rating systems in Canada

has led to the proliferation of a number of different FOP systems, each with their own criteria

(13,18,19). As a result, concern has been raised that the proliferation of competing FOP systems

may lead to greater confusion and mistrust among consumers trying to choose food products on

the basis of the nutrition information available on food packages (13,18-22). Furthermore,

existing FOP systems have been the subject of criticism for using nutrition criteria that aren’t

stringent enough to exclude unhealthy choices, and as a result, do not guide consumers’ choices

towards truly healthy products. In light of calls for the development and implementation of a

simplified FOP nutrition rating system for Canada, the potential contribution of FOP systems to

supporting consumers’ ability to identify healthier choices should be studied. The overall

purpose of this research is to explore the healthfulness of foods designated by FOP nutrition

rating systems in Canada and the consumer friendliness of different FOP symbols.

3

Chapter 2

2 Background and literature review

2.1 Diet related chronic disease

Worldwide, chronic non-communicable diseases are the leading cause of death, and deaths due

to chronic disease are on the rise (23). Four chronic diseases – cardiovascular diseases, diabetes,

cancer, and chronic respiratory diseases – cause 63% of deaths globally (23); while in Canada,

these same four diseases account for approximately 75% of all deaths (24). Overall, the

economic burden of chronic disease in Canada is estimated to exceed $93 billion per year (24).

The burden of chronic non-communicable diseases in Canada, and internationally, has led to

calls to address the common risk factors that contribute to chronic disease (2,3,25).

A common modifiable risk factor shared by many chronic diseases, including obesity, diabetes,

cardiovascular diseases, cancer, dental diseases, and osteoporosis, is an unhealthy diet (1). In

order to prevent these diet-related chronic diseases, the World Health Organization recommends

that healthy populations limit their intake of saturated fat, trans fat, cholesterol, sugar, sodium,

and total energy, while ensuring adequate intakes of fat, polyunsaturated fat, carbohydrate,

protein, and fibre (1,3). Similarly, 2007’s Eating Well with Canada’s Food Guide recommends

Canadians limit foods and beverages high in calories, fat, sugar or salt, as well as trans fat while

meeting their needs for vitamins, minerals, and other nutrients (4).

It is widely accepted that a reduction in deaths due to chronic disease can be achieved, in part,

through investments in population interventions to promote healthy diets, such as the eating

patterns proposed by the World Health Organization and Eating Well with Canada’s Food Guide

(23). Among the main interventions proposed to promote healthy diets are food labelling and

restrictions on the marketing of unhealthy food products (25).

4

2.2 An overview of nutrition information on food labels

2.2.1 Nutrition labelling

Food labels are a means through which product information is communicated to consumers (26).

According to the Canadian Food Inspection Agency, a food label serves three primary functions:

1. It provides basic product information;

2. It provides health, safety, and nutrition information; and,

3. It acts as a vehicle for food marketing, promotion and advertising.

To date, food labelling interventions that promote healthy diets to reduce the risk of chronic

disease have focused on the food label’s second function. According to the World Health

Organization, “consumers require accurate, standardized and comprehensible information on the

content of food items in order to make healthy choices” (3). To provide consumers with this

nutrition information, the World Health Organization suggests governments require that

information on key nutritional aspects be provided on food labels – what is commonly known as

nutrition labelling. The Codex Alimentarius Commission recommends that mandatory nutrition

labelling be presented in a tabular format that displays information on nutrients including:

energy, protein, available carbohydrate, fat and saturated fat, sodium, and total sugars (27).

In keeping with the recommendation of the World Health Organization, mandatory nutrition

labels are now imminent or in place in over 20 countries world wide and voluntary nutrition

labelling standards are in place in at least 10 more (15). Within Canada, mandatory nutrition

labelling standards were passed in 2003, 15 years after voluntary nutrition labelling guidelines

were first introduced (5). Mandatory nutrition labelling came into full effect in 2007 and

required the declaration of energy and 13 core nutrients per serving in a Nutrition Facts table

(NFt) on almost all prepackaged foods with a few exceptions.

2.2.1.1 Nutrition labelling: Use and understanding

Since mandatory nutrition labelling came into full effect in 2007, the number of Canadians who

report often or always using the NFt to inform first time purchases has been on the rise (28-30).

The most recent data show that 74% of Canadians are often or always using the NFt for first-time

purchases (30). This finding is consistent with self-reported data from other countries with

nutrition labels, that also reported high levels of use (31-33). However, effective use of the NFt

5

in food choice requires understanding, and research on consumers’ abilities to understand

nutrition labelling is mixed. A few studies have found that consumers, and consumers with low

levels of health literacy in particular, perform poorly when their food label knowledge is tested

(34-36); others have found that, when tested, a large proportion of consumers can successfully

use the nutrition label to find needed information (29,30,35,37-39).

The quantitative portions of nutrition labels, including the percent daily value and amount of

food, appear to be a barrier to many consumers’ understanding and interpretation

(11,29,30,34,35,38-53). Numerous studies have found the percent daily value – the proportion of

daily requirements for a given nutrient provided in a serving of the food – to be a difficult

concept for consumers (11,29,30,34,39-42,45,48-51,54). Indeed, in 2012 only 60% of Canadians

reported feeling confident in their ability to use the percent daily value when choosing foods –

this figure fell to 46% among consumers with less than a high school education (30).

Furthermore, fewer than half of all consumers can determine what is ‘a little’ or ‘a lot’ of a given

nutrient using the percent daily value (29,30). With respect to using nutrition label information

on the amount of food, numerous studies have found consumers struggle to compare two foods

with different serving sizes (29,30,39,46,48), as well as to determine the amount of different

nutrients found in servings larger or smaller than the amount of food depicted in the nutrition

label (35,38,39,44,51-54). For example, one study of Canadian parents found that only 16.4% of

parents could estimate the number of calories in an entire bottle of Cola using a NFt that gave the

nutrition information per serving (52). Consumers with the lowest levels of education appear to

struggle most with nutrition label numeracy (29,30,38,55). In addition to the numeracy

challenges, consumers have also reported difficulty with nutrition label terminology, which is

perceived by some as too technical and advanced (10).

Difficulties understanding and interpreting nutrition labels, like those noted above, have led

expert groups, such as Canada’s House of Commons Standing Committee on Health, to call for a

simplified nutrition labelling scheme (13,56-58). Indeed, research has shown that simplified

labels promote more accurate evaluations of foods by consumers (55,59,60); and, several studies

have reported a benefit to simplified nutrition labels that employ graphics and symbols

(10,33,60-65). Further, research has suggested that consumers may be better served by nutrition

information on the front of food packages rather than on the back, as is the current practice

(10,63,66-68).

6

2.2.2 Nutrition marketing

As noted in the previous section, food labels can function as a vehicle for food marketing,

promotion, and advertising (26). Nutrition marketing has been defined as “any marketing … of

food and beverages using health or nutrition information beyond minimum requirements”

(69,70). In the case of food labels, the NFt constitutes the minimum requirements and any

additional health and nutrition information provided beyond the NFt would therefore be

considered nutrition marketing. Within Canada, voluntary health and nutrition information on

food packages falls under two umbrellas, nutrient content claims and health claims (Figure 2-1).

Figure 2- 1. Nutrition information on prepackaged foods in Ca nada

Nutrition information on Canadian food

labels

Mandatory

List of ingredientsNutrition Facts

tableNutrient content

claims

Disease risk reduction and

therapeutic claimsFunction claims

Nutrient function claims

Strain-caims

1. Nutrition information on prepackaged foods in Ca nada

Voluntary

Nutrient content claims

Health claims

Function claims

Probiotic claims

-specific caims

Non-strain-specific claims

General health claims

Body weight claims

Third-party endorsements and

logosHeart symbols

7

Heart symbolsGuidance for

healthy eating

8

2.2.2.1 Nutrient content claims

Nutrient content claims describe the amount of a nutrient in a food and fall under Canada’s Food

and Drug Regulations (5). The Regulations define a closed list of nutrient content claims with

specific compositional criteria in place for their use. For example, to carry the claim “free of

sodium or salt” a food must contain less than 5 mg of sodium per reference amount and serving

of a stated size, or less than 5 mg of sodium per serving of a stated size if the food is a

prepackaged meal. Additional examples of nutrient content claims include “low in fat”, “a

source of calcium”, and “zero trans fat”, and are listed in a Table following section B.01.513 of

the Food and Drug Regulations.

2.2.2.2 Health claims

Health claims describe (or imply) a relationship between eating a food, or a food ingredient, and

health, and are subject to varying degrees of regulation in Canada (26). Health claims have been

classed into three main categories (Figure 2-1). The first category is disease risk reduction and

therapeutic claims, which fall under Canada’s Food and Drug Regulations. Disease risk

reduction claims describe a link between a reduced risk of developing a diet-related disease, and

the consumption of a given food, or food constituent, as part of a healthy overall dietary pattern.

Presently, five disease risk reduction claims – with prescribed wording – are permitted under the

Regulations (71). For example, “A healthy diet with adequate calcium and vitamin D, and

regular physical activity, help to achieve strong bones and may reduce the risk of osteoporosis”

(26). Therapeutic claims are about treating or easing disease or health conditions, or about

repairing or altering body functions. Presently, six therapeutic claims are permitted (71). For

example, “16 g (2 tablespoons) of ground flaxseed supplies 40% of the daily amount shown to

help lower cholesterol” (72). Prior to permitting the use of a disease-risk reduction or

therapeutic claim on food labels and advertisements, Health Canada completes a review of the

scientific evidence for the claim (71). Only claims substantiated by the evidence are permitted.

Further, to use a given disease risk reduction claim or therapeutic claim on its label, a food must

meet prescribed conditions related to energy and/or other nutrients relevant to the claim (26).

The second category of health claims is function claims (26). Function claims relate the positive

contribution of eating a food, or food constituent, to normal body functions and biological

activities (e.g. health, maintaining physiological functioning, physical or mental performance).

9

These claims cannot refer to the treatment or prevention of any disease or of the symptoms of a

disease. While function claims must comply with subsection 5.1 of Canada’s Food and Drugs

act, that forbids “false, misleading, or deceptive” product labelling (73), there are presently no

specific regulations governing their use (26). However, the Canadian Food Inspection Agency’s

Guide to Food Labelling and Advertising has set guidelines for their use. “Consumption of 1

cup (250 ml) of green tea increases antioxidant capacity in the blood” is an example of a function

claim. Nutrient function and probiotic claims are subsets of function claims. Nutrient function

claims describe the generally accepted role of energy or nutrients in maintaining health, or in

normal growth and development. Unlike more general function claims, nutrient function claims

are provided for under the Food and Drug Regulations; the Regulations set out minimum

requirements for the food’s content of protein, or specific vitamins or minerals relevant to the

claim. “DHA, an omega-3 fatty acid, supports the normal physical development of the brain,

eyes and nerves primarily in children under two years of age” is an example of a nutrient

function claim. Probiotic claims are statements about the health benefits or effects of specific

probiotic strains (strain-specific claims) or the nature of probiotics (non-strain specific claims).

Any probiotic claim that is therapeutic in nature would be subject to the Food and Drug

Regulations, however there are presently no such claims approved for use in Canada. Aside

from that, there are no specific regulations for the use of probiotic claims, however guidelines

have been described in Health Canada’s The Use of Probiotic Microorganisms in Food (74).

These guidelines include a requirement that the food contain at least the minimum amount of

probiotic required to achieve the claimed effect. “Provides live microorganisms that naturally

form part of the gut flora” is an example of a probiotic claim (26).

The final category of health claims is general health claims (26). General health claims are

broadly defined as claims that “promote health through healthy eating or that provide dietary

guidance”. Such claims do not discuss health effects or conditions, or disease and are generally

not subject to any specific regulations. However, they are subject to subsection 5.1 of the Food

and Drugs Act and as such may not be “false, misleading, or deceptive” (73). Further, guidance

for their use is given in the Canadian Food Inspection Agency’s Guide to Food Labelling and

Advertising (26). The Guide advises on general health claims that relate to body weight, third-

party endorsements and logos, heart symbols, and guidance for healthy eating. For instance,

general health claims may state that a food is intended for use in maintaining body weight (e.g.

10

“As part of healthy eating, this food may assist in achieving and maintaining a healthy body

weight because it is lower in energy”), provided the food’s label meets prescribed conditions.

Health professionals or organizations, individuals and groups may give third-party endorsements

or logos to products, so long as they do not mislead or confuse consumers as to the advantages of

the food. According to the Guide, to avoid such confusion a third-party endorsement or logo

should adhere to the following principles:

1. Does not give the impression that a single food or brand of food is "healthier" than, or

nutritionally superior to, other foods not bearing the third party's name, statement, logo,

symbol, seal of approval or other proprietary mark. …

2. Does not give the impression that the food is a treatment, preventative or cure for

disease. A third party's name, statement, logo, etc. must not suggest that a food may

prevent, cure or treat a disease … Such a suggestion is false and specifically prohibited

by the Food and Drugs Act (26).

General health claims in the form of heart symbols are generally not acceptable, but they can be

used in conjunction with third-party endorsements and logos or with disease-risk reduction

claims related to heart health (26).

2.2.2.3 Nutrition marketing: Use and Understanding

Surveys of packaged foods sold in North American have found that nutrition marketing

messages are common on food labels (70,75-78). Recent estimates place nutrition marketing

information on 48.1% of Canadian (78), 44.3-71.0% of American (70,75), and 14% of Australian

packaged food labels (79). In addition to widespread use by manufacturers, a small but

significant proportion of consumers appear to be seeking out nutrition marketing information on

food labels. In 2008, 21% of Canadian label readers looked for nutrient content claims and 16%

looked for health claims (6). European estimates are even higher, with 59% of consumers

reportedly ‘often’ or ‘always’ reading claims (80). Moreover, a recent ethnographic study found

that Canadian consumers made use of front-of-package nutrition marketing information, like

nutrient content claims, more often than the NFt (81).

Widespread use of nutrition marketing information is of interest because nutrition marketing

could potentially influence food purchases, which may impact consumption patterns and finally

chronic disease risk (69,70). The presence of nutrition marketing information on food labels has

11

been shown to influence consumers’ perceptions of food products’ healthiness and nutritional

quality (80,82-89). In fact, the influence of nutrition and health claims on consumers’

perceptions of healthiness and nutritional quality have been found to extend beyond the

nutritional and health elements identified in the claim – what is known as the ‘halo’ effect

(82,83,86,88,89). For example, consumers rated the same bread lower in calories when they saw

a nutrition marketing claim stating that the product was low in carbohydrates than when they saw

no such claim (86). In addition, consumers perceived a cereal as nutritionally superior, as well

as more beneficial to digestive health in the presence of a fibre claim compared to the same

cereal without a fibre claim; they also believed that their likelihood of disease risk and weight

gain was lower (88). Similarly, consumers rated a soup as overall healthier, and more beneficial

for hypertension, high cholesterol, heart problems, constipation and diabetes when either a

nutrient-content, function, or disease-risk reduction claim related to sodium was present; despite

the fact that all soups had the same NFt and the claims only mentioned a relationship between

sodium and blood pressure (89). This is problematic because, as previously noted, in order to

carry nutrient-content, disease-risk reduction, therapeutic, nutrient function, and probiotic claims,

foods need only meet compositional criteria related, or relevant to, the nutrient or food

component that is the subject of the claim (26,74,90). Further, with respect to disease risk

reduction and therapeutic claims, the relationship between the nutrient or food components that

are the subject of the claim and health have only been substantiated for the disease or condition

cited in the claim (71). Therefore it would appear that these nutrition marketing claims might be

giving consumers erroneous impressions about the nutritional and health qualities of the products

that carry them. This is problematic given that the presence of nutrition marketing claims, and

health claims in particular, have led consumers to report greater intentions to purchase a product

(80,83,84,89,91); however, it is unclear if greater purchase intentions are translated into actual

in-store purchases.

2.3 An overview of FOP nutrition rating systems and symbols

Yet another form on nutrition marketing found on food labels, not yet discussed, is front-of-pack

(FOP) nutrition rating systems and symbols. FOP systems and symbols summarize key

nutritional aspects and characteristics of food products and are typically found on a product’s

principal display panel, but may be elsewhere on the food label (19). FOP systems and symbols

12

often integrate characteristics of traditional nutrition labels, nutrient-content claims, and even

disease-risk reduction, therapeutic, and function claims.

2.3.1 History of FOP symbols

FOP nutrition rating systems and symbols date back to the 1980s, when the American Heart

Association initiated its Heart Check symbol (1987) and the Swedish National Food

Administration introduced its Nordic Keyhole (1989) (19). Canada’s first FOP symbol, the

Heart and Stroke Foundation of Canada’s Health Check™, was introduced in 1999 to help

consumers identify healthier food choices. Typically, manufacturers receive the right to use

third-party symbols by paying a program fee (92). Although non-profit health organizations and

governments developed the earliest entrants into the FOP symbol marketplace, industry quickly

followed suit with food manufacturers, retailers, industry and non-industry associations, and non-

industry experts all introducing their own symbols in the years that followed (13,19,22,92,93).

For instance, in 2005 PepsiCo expanded its Smart Spot symbol into Canada and was quickly

followed by a number of other proprietary FOP systems developed by food manufacturers (e.g.

Kraft Sensible Solutions, President’s Choice Blue Menu, Kellogg’s Get-the-Facts, etc.) to help

consumers identify their “healthier” products.

In the only systematic study of the use of FOP symbols on food packages completed to date, our

research group at the University of Toronto identified 158 unique FOP symbols in use on

packaged food labels in Canada in 2010-2011 (78). Overall, one or more FOP symbols appeared

on 18.9% of food products. The number of FOP symbols in use internationally has not been

comprehensively studied. However, reports have described 8-30 different FOP nutrition rating

systems or symbols in use internationally (19,92-96). Further, the overall prevalence of FOP

systems on food products in markets outside of Canada has not been studied. Point-of-purchase

research emanating from the Netherlands, United Kingdom, and Australia has found that some

FOP systems can be found on up to 18% of products purchased by consumers (97,98). However,

the proportion of products carrying a FOP system in these studies may not reflect the prevalence

of such products in the marketplace, but rather reflect the popularity of products bearing FOP

systems.

13

2.3.2 FOP nutrition rating system and symbol categories

In 2010, the United States (US) Institute of Medicine classified the myriad FOP systems and

symbols in use internationally into three categories (19):

1. Nutrient-specific systems;

2. Summary indicator systems; and,

3. Food group information systems.

The sections that follow describe these three categories and examples of each FOP type are

found in Figure 2-2.

Figure 2-2: Examples of different front- of

Nutrient Specific Systems

Nutrient- specific systems that display the amount of calorie s and select nutrients per serving

The Grocery Manufacturers Association and the Food Marketing Institute’s Facts Up Front

Nutrient- specific systems based on claim criteria

General Mills’ Goodness Corner

of-pack symbol types

Nutrient Specific Systems

specific systems that display the amount of calorie s

The Grocery Manufacturers Association and the Food

specific systems based on claim criteria

Summary Indicator Systems

Heart and Stroke Foundation of Canada’s Health Check™

Kraft’s Sensible Solutions™

14

Heart and Stroke Foundation of

Food Group Information

Systems

Whole Grain Council’s Whole Grain Stamp

15

2.3.2.1 Nutrient-specific systems Nutrient-specific systems include 1) systems that display the amount per serving of calories or

select nutrients from the NFt on the FOP, and 2) symbols based on nutrient-content or health

claim criteria (19) and appear on 4.9% of packaged Canadian food products (78). Nutrient-

specific systems are based in the regulations for nutrition labelling and claims (19). Systems that

display the amount per serving of calories and/or select nutrient on the FOP are closely related to

mandatory back-of-pack nutrition labels, such as the NFt. European researchers sometimes refer

to these systems as “banding” or “nutrient signposting” (93). These systems are designed to be

comprehensive and appear on as many foods as possible. In addition to displaying the amount of

calories and select nutrients per serving, these systems may also display percent daily value

information (19). Systems displaying percent daily value information are known as percent-

daily value systems, guideline daily amount systems, or daily intake guide systems depending on

the country, however for simplicity they will be referred to hereafter as daily value systems.

Daily value systems have been found on up to 67% of packaged food labels in some countries

outside of Canada (99,100). Systems that display the amount per serving of calories or select

nutrients may also include colours (traffic lights) or words that interpret for the consumer if the

specific nutrient amounts found in the product are “high”, “medium”, or “low” (19). In 2008-

2009, traffic light systems were found on ~3% of packaged food labels in the United Kingdom –

the country where they originated (100). An example of a system that displays the amount of

calories and select nutrients per serving is the US Grocery Manufacturers Association and Food

Marketing Institute’s Facts Up Front (101). Facts Up Front displays on a food label’s principal

display panel the amount of calories, fat, sodium, and sugar, plus two additional nutrients, per

serving of a food. When applicable, Facts Up Front also provides percent daily value

information.

Symbols based on claims would be considered nutrient-content or health claims (depending on

the claim they portray) in addition to being FOP symbols. As a result, they would be subject to

the same regulations described in section 1.2.2.1 or 1.2.2.2, requiring them to meet the thresholds

established per reference amount and serving of a stated size for the nutrient-content or health

claim depicted in the symbol. Multiple symbols based on claim criteria may be placed on a

single product depending on which nutrients are included in the system (19). An example of

such a system used in Canada is General Mill’s Goodness Corner which awards foods multiple

16

symbols based on their compliance with nutrient content claims such as “low fat”, “excellent

source of iron”, and “high fibre” (94).

2.3.2.2 Summary indicator systems

Summary indicator systems, also known as signpost or health logos, give no specific nutrient

content information to consumers, but instead use a single symbol, icon, or score to provide

summary information about the nutrient content of a food (19). In Europe, summary indicator

systems are known as “point of purchase” systems or “integrative approaches” (93). FOP

systems within this category are typically only applied to ‘healthy’ or ‘healthier’ choices within a

product category. This type of symbol has been identified on 7.5% of packaged foods sold in

Canada (78), and fewer 2% of products in Europe (100). Summary indicator systems attempt to

assess the overall healthiness of a food using either thresholds or algorithms (19). Threshold-

based systems typically establish maximum levels for nutrients to limit, and minimum levels for

nutrients or food components to encourage, to judge whether a product qualifies for a summary

indicator symbol (21,22). Algorithm-based systems award points for the presence of nutrients or

food components to encourage and subtract them for the presence of nutrients to limit to arrive at

a final score that is used as the summary indicator symbol (19,21). Summary indicator systems

are considered to be general health claims, and as such are not subject to any specific regulations.

However, as noted in section 1.2.2.2, they may not be “false, misleading, or deceptive” (73) and

they should follow all relevant sections of The Guide to Food Labelling and Advertising, such as

the guidance on the use of third-party endorsements and logos (26). An example of a summary

indicator system based on a threshold model is the Heart and Stroke Foundation of Canada’s

Health Check™ (102). The Health Check™ system awards a single symbol to the products of

participating manufacturers that meet Health Check™’s category-specific threshold criteria

established by Registered Dietitians based on Eating Well with Canada’s Food Guide and

nutrient-content claim criteria. An example of a summary indicator system based on an

algorithm model is NuVal ®(103). Using an algorithm that takes over 30 nutrients and food

components into account, NuVal ® gives food a single score from 1-100. One limitation of

summary indicator systems is that consumers cannot tell at the point-of-purchase the role

different nutrients and food components played in the food’s evaluation, and for some systems

(e.g. NuVal ®) full details of the scoring system are not publically available (19).

17

2.3.2.3 Food group information systems

Food group information systems include symbols that indicate that a food group (such a

vegetables and fruit) or a food ingredient important to the diet (such as whole grains) is present

in a food product (19). Some of the food group symbols reference the presence of a serving (or

partial serving) of a food group from Eating Well with Canada’s Food Guide (using Food Guide

standards of what constitutes a serving). This type of symbol appears on 3.5% of packaged

foods sold in Canada (78). An example of a food group information symbol is the Whole Grain

Council’s Whole Grain stamp (104).

In addition to the three categories of FOP symbols described by the Institute of Medicine, our

research group identified yet another category in our study on the use of FOP symbols on 2010-

2011 food packages: hybrid systems (78). Hybrid systems combined features of two or more of

the categories previously described and appeared on 7.0% of all packaged foods products in

Canada.

2.3.3 Nutrient profiling

Underpinning each FOP nutrition rating system, regardless of category, is a nutrient-profiling

scheme (19,93). Nutrient profiling is the classification of foods based on their nutritional

composition (95,105), however in the absence of specific regulations for their use, FOP symbols

in Canada do not adhere to any standardized nutrient profiling criteria (13). While most FOP

symbols share an overarching intent to help consumers quickly compare foods, identify

nutritious choices, and determine if a food meets their nutrient needs, they each have their own

purpose and target-audience (19). In addition, the nutrient profiles underlying different FOP

systems may vary in their: 1) application of nutritional criteria; 2) selection of nutrients and other

food components included; 3) amount of food; 4) type of model; and, 5) criteria’s basis

(19,20,93-95,105,106). Indeed, the US Institute of Medicine has stated that no two FOP symbols

share the same underlying nutrient profile (19).

Application of the criteria refers to whether or not the system applies criteria universally (across-

the-board) or by product category (19,22,93,95,105,106). Category specific criteria recognize

naturally occurring differences in nutrient profiles by product category and allows consumers to

identify healthier selections within a product category (19,22,107). However, a category-based

system requires the development of food categories, which is not always evident (19). Universal

18

criteria apply the same criteria to all foods, and allow consumers to compare foods across all

categories. Universal criteria also allow FOP systems to use existing government regulations for

nutritional labelling and nutrient content claims as the basis for the criteria (19). Summary

indicator systems tend to use category specific criteria over universal criteria; whereas nutrient-

specific and food group information systems tend to apply their criteria universally. Regardless

of whether a FOP system’s nutrient profile is category specific or universal, one study has found

that consumers believe that all FOP systems compare products across all product categories.

Feunekes et al (63) reported that 58% of European consumers thought that FOP systems

compared products across the board, while only 27% thought they compared products within

categories.

There is tremendous variety in the nutrients and food components included in the nutrient

profiling schemes that are used from one FOP system to the next. Reviews have identified FOP

systems based on as few as one or two nutrients and/or food components to more than 30

(19,92,93,103,106,108). Nutrients incorporated in one or more FOP system included energy,

total fat, saturated fat, trans fat, cholesterol, omega-3 fats, carbohydrate, starch, added sugar,

fibre, whole grains, protein, sodium, and vitamins and minerals (vitamin A and C, folate, iron,

calcium). Most nutrient profiles include a combination of nutrients to limit (“negative”

nutrients) and nutrients and/or food components to encourage (“positive nutrients”), although

some systems focus on just one or the other (19,92,105,106). Nutrients to limit have been

defined variously as calories, fats and oils (total fat, saturated fat, trans fat, cholesterol), added

sugars, and sodium (19,22,105). Nutrients and food components to encourage have been defined

variously as vitamins and minerals (such as calcium, potassium, magnesium, and vitamins A, C,

and E), fibre, fruits and vegetables, whole grains, and low-fat dairy. Depending on the system,

the nutrients included in the profile may be based on existing dietary guidance from government

or other authoritative bodies, or a scientific advisory panel put together by the system proprietor

(19). The nutrients and/or food components included in a FOP system should be those that have

been acknowledged as important in consensus documents developed by authoritative bodies,

however, many systems include nutrients and food components for which there is insufficient

evidence to support their inclusion. In 2011, the US Institute of Medicine concluded that

saturated and trans fat, sodium, and added sugars are the nutrients of greatest public health

concern, and therefore should be included in the nutrient profile of FOP systems (19,109).

19

Nutrition labelling in Canada is based on reference amounts and servings of a stated size (26). A

reference amount is a regulated amount of a type of food typically eaten by an individual in one

sitting and is the basis for nutrient-content and health claims in Canada. A serving of a stated

size is the amount of food forming the basis for the nutrition information found in the NFt.

Manufacturers are allowed to select from a range of serving sizes – which encompass the

reference amount – for use in the NFt depending on the product. FOP systems may be based on

reference amounts, serving size, or other amounts of food such as 100 g, 100 ml, or 100 calories,

or some combination of these (19,20,95,106). Stockley’s (93) review of FOP symbols in

Europe, Australia, and North America found that most schemes developed by public sector

interests (like health organizations, governments, and nutrition councils) used a standardized

amount, like per 100 g, as their base, whereas commercial interests (like manufacturers and

retailers) appeared to use both per serving and per standardized amounts equally.

Two types of nutrient profiling models are used in FOP systems: thresholds and algorithms

(108). As previously described, threshold criteria typically establish maximum levels for

nutrients to limit, and minimum levels for nutrients or food components to encourage (21,22),

whereas algorithm models award points for the presence of nutrients or food components to

encourage and subtract them for the presence of nutrients to limit to arrive at a final score

(19,21). Nutrient-specific systems based on claims criteria, or that add interpretive elements

(such as colours or words) to information on the amount of calories or select nutrients per

serving, use thresholds, as do food group information systems (19). Summary indicator systems

may use either thresholds or algorithms. Reviews of FOP nutrition rating systems have revealed

that threshold criteria are the dominant model for nutrient profiling (19,93,105).

Finally, the basis for the numbers used in nutrient profiling criteria differs from one FOP system

to the next. A number of FOP systems base their criteria on government nutrition labelling

regulations, dietary guidance from government or other authoritative bodies, or on the opinions

of scientific advisory panels (19,93). However, for many FOP systems used in Canada and

abroad the basis of their nutrient profiling criteria is not apparent (92,93,105). This is especially

true for commercial FOP systems.

Presently there is no single approach to nutrient profiling that is generally accepted for

classifying foods for FOP systems or other purposes. Moreover, the World Health Organization

20

and other experts have raised concerns that multiple different nutrient profiling models, as are

seen in existing FOP systems, can lead to consumer confusion and mistrust (13,56,92,94,95).

Further, the nutrient profiling criteria underlying many popular FOP systems have been criticized

as being too relaxed (i.e. highly processed foods high in sugar and low in whole grains may

qualify for symbols) (56). For example, the industry developed Smart Choices summary

indicator system came under scrutiny when it was shown that products such as Froot Loops and

Cocoa Krispies qualified for this symbol on the basis of the system’s nutrient-profiling criteria

(56,110). Further, the nutrient profiling criteria of the majority of FOP systems is undocumented

in the literature, and fewer still have been empirically evaluated or validated (92,105), making it

difficult to determine if these systems are truly directing consumers to healthier choices. As a

result, there have been calls from a public interest group for the development of minimum

standards to ensure consistency across voluntary FOP symbols (109). In 2014, Health Canada

released a nutrient profiling system to classify foods from the four food groups into four tiers:

“Foods in line with Canada’s Food Guide guidance” (tiers 1 and 2); “Foods partially in line with

Canada’s Food Guide guidance” (tier 3); and, “Foods not in line with Canada’s Food Guide

guidance” (tier 4) (111). While this system has the potential to be applied as a minimum

standard for FOP symbols, at present Health Canada describes these criteria as exclusively a

surveillance tool.

2.4 FOP systems and the nutritional quality of foods

2.4.1 FOP systems and reformulation

FOP systems have been proposed as a potential public health intervention to improve population

level dietary intakes (19); one way in which FOP systems are anticipated to do so is by

stimulating manufacturers to reformulate their products in a more healthful way in order to meet

the FOP system’s nutrient profiling criteria. The current evidence, although limited in both

quantity and quality, suggests that the nutrient-profiles underlying selected summary indicator

systems are in fact stringent enough to stimulate healthier product reformulations by

manufacturers (112-115). Researchers in Australia and New Zealand, Canada, and the

Netherlands have found that summary indicator systems have been successful in stimulating

participating manufacturers to lower their foods’ sodium, saturated fat, trans fat, and/or calorie

21

content and/or increase their fibre content, across a range of product categories, or formulate new

products in order to meet the system’s nutrient profile. For example, in order to qualify for the

Choices logo, participating companies in the Netherlands were able to reformulate processed

meat products and reduce their saturated fat content from 3.09 g to 1.74 g and their sodium

content from 1018 mg to 835 mg per serving (114). Similarly, Canadian manufacturers

participating in the Health Check™ program were able to achieve a reduction of 80-150 mg of

sodium per reference amount in reformulated products in categories ranging from dinners and

entrees to deli meats (115). Although these findings are promising, because the FOP systems in

question are voluntary, and sometimes require the manufacturer to pay a licensing fee for their

use, it is unclear whether competitors not participating in these programs undertook similar

reformulations over the same period of time. Furthermore, the products of manufacturer’s not

participating in these programs have not been evaluated against the systems’ nutrient profiling

criteria. As a result, it is unclear how many of said products would qualify for these FOP

systems as is, versus require reformulation should they choose to participate in the program.

Thus the potential for summary indicator systems to stimulate reformulation by prospective

program participants is unclear.

Only one study to date has examined the potential impact of nutrient-specific systems on

reformulation. A recent study analyzed the nutritional quality of breakfast cereals over a period

of six years (2004-2010), during which time a voluntary daily value system – Australia’s Daily

Intake Guide – was introduced (2006) (116). The study compared the nutrient content of cereals

year over year as well as the nutrient content of cereals with and without the Daily Intake Guide

FOP system. Of the 67 cereals available as the same product in 2004 and 2010, 60 had changed

their formulation; however, this did not result in significant changes in their nutrient content,

aside from a small increase in the protein content of cereals without the FOP system, leading the

authors to conclude that nutrient-specific systems, like the Daily Intake Guide, that provide

information on calorie and nutrient content per serving on the FOP without interpretive

information do not appear to promote product reformulation. Similar data on reformulation is

unavailable for other nutrient-specific systems (those based on claims or with interpretive

elements) and food group information systems.

22

2.4.2 The nutrient profiles of foods with and without FOP systems

Experts in Canada have raised concern that consumers may, by default, perceive foods without

FOP symbols as less healthy than foods with symbols (78,81), despite the fact that there no

evidence to suggest that foods with symbols have lower levels of nutrients of public health

concern. In the current Canadian environment where multiple proprietary and third-party FOP

symbols are in use, not all products on the grocery shelves have been evaluated using the

available FOP systems. Therefore, many products may qualify for one or more FOP symbols but

may not actually use these symbols on their labels, making it difficult for consumers to use these

symbols to compare products (19). For example, all products carrying the Heart and Stroke

Foundation of Canada’s Health Check™ symbol meet the Health Check™ criteria, but not all

products that meet the Health Check™ criteria carry the symbol. As a result, consumers cannot

be sure if a product does not carry the Health Check™ symbol because it did not meet the

system’s nutrient profiling criteria or simply because its manufacturer is not participating in the

program. Moreover, manufacturer developed FOP symbols may only be carried by qualifying

products of the manufacturer who developed the system in question, and as a result consumers

cannot be certain products without a symbol are in fact of a lower nutritional standard or if that

manufacturer simply does not have its own proprietary symbol (92). However, just how many

products in the Canadian marketplace have been excluded from carrying specific FOP systems

for reasons unrelated to their nutritional composition has not been evaluated. As a result, it is

unclear if Canadians can rely on specific licensed or proprietary FOP symbols to identify

products of superior nutritional quality.

As previously noted, the nutrient profiling criteria behind the FOP symbols found on food labels

differ from one system to the next. However, the details of these nutrient profiles are invisible to

the consumer at the point of purchase (117). In environments where a single FOP system has

been universally applied (i.e. all products sold in store are evaluated against the same nutrient

profiling criteria and symbols are awarded to all qualifying products), one study has shown that

consumers can rely on FOP symbols as an indicator of products with lower levels of negative

nutrients and higher levels of positive nutrients. Sutherland et al. (118) found that breakfast

cereals bearing the Guiding Stars summary indicator system, which was universally applied to

all products sold at a Northeastern US grocery chain, had less sugar and sodium, and more fibre

than cereals without Guiding Stars. However, more research is needed to determine if their

23

conclusion can be applied to other universally applied FOP systems. Moreover, the current

Canadian grocery marketplace is not characterized by the use of a single FOP system,

universally applied to all products, but rather the use of multiple FOP systems with distinct

nutrient profiling criteria. Lack of specific regulations governing the use of these FOP systems

means that there is no minimum nutrient profile that foods must meet to carry a FOP symbol.

Without minimum nutrient profiling standards, products that qualify for one FOP symbol may

not qualify for a different FOP symbol. For instance, products that have failed to meet the

nutrient profiling criteria for a third-party FOP symbol, like the Heart and Stroke Foundation’s

Health Check™, may qualify for a symbol developed by their own manufacturer whose nutrient

profiling criteria is less stringent (92). Indeed, different FOP systems do not show perfect

agreement in which foods can qualify for a symbol. A European study that applied the nutrient

profiling criteria of three FOP systems to the same foods found only moderate agreement (41-

60% identical ratings) between how some of the FOP systems classified the products (108). As a

result, it is unclear if consumers can rely on the mere presence of any FOP symbol as an

indicator of products that meet a higher nutritional standard.

In 2011, the US Institute of Medicine stated that in order to best promote the health of the public,

FOP systems must take into account nutrients and nutritional components of greatest relevance

to chronic disease risk (109). This led them to conclude that calories, saturated and trans fat,

sodium, and added sugar should be considered in the nutrient profiles of FOP systems. Despite

this, most countries – Canada included – allow products to carry nutrient-specific symbols

without considering their overall composition with respect to calories and these four nutrients of

public health concern (119). Indeed, nutrient-specific systems based on claim criteria have

caused concern that some products qualifying for a nutrient-content or health claim for one

nutrient may not be “low” in other nutrients that should be limited (19,108). Similarly, food

group information systems typically only consider a single food group or ingredient and not

overall nutrient content (19). Further, although summary indicator systems consider multiple

nutrients and food components in their nutrient profiling criteria, the elements considered differ

from one system to the next (19,92,93,106,108), and do not necessarily include calories and the

nutrients flagged by the Institute of Medicine as being important to public health. As a result, it

is presently unclear if FOP symbols in general can be relied upon as a guide to foods with more

24

favourable levels of calories and nutrients related to chronic disease risk (i.e. saturated fat, trans

fat, sodium, and added sugar).

2.5 Influence of FOP systems on perceptions of nutritional quality

Despite the fact that the overall healthiness and nutrient content of foods with FOP symbols is

unclear, evidence suggests that consumers perceive products with certain categories of FOP

symbols as being overall healthier, as well as having superior nutrient profiles (120-124). With

respect to nutrient-specific systems, one study has demonstrated that even the presence of a FOP

system that simply provides information on the content of calories and select nutrients per

serving increases consumers’ product healthfulness perceptions and purchase intentions over

conditions where the same product is shown without this nutrient-specific system (124).

Moreover, research has show that when an element of interpretation, such as traffic light colours,

is added to a nutrient-specific system that displays the amount of calories and selected nutrient

per serving, consumers’ perceptions of healthfulness and nutrient content are influenced

favourably (123). In a 2011 study, consumers were randomized to view a frozen meal with or

without a traffic light nutrient-specific system that rated the food’s sodium and cholesterol

content as high (red), calorie and sugar content as medium (amber), and fat and saturated fat

content as low (green) (123). Consumers exposed to the traffic light rated the frozen meal as

significantly lower in calories and the nutrients given green or yellow traffic lights. These

favourable evaluations even extended to nutrients not included in the traffic light symbol –

consumers exposed to the traffic light also rated the meal as being lower in trans fat and

carbohydrates. Ratings of the meal’s content of nutrients given red traffic lights did not differ

between the two conditions, however consumers who saw the traffic light system expressed

greater intentions to purchase the product.

With respect to nutrient-specific symbols based on claims criteria, given that these symbols are

forms of nutrient-content and health claims, it follows that these symbols most likely exert the

same effect on consumers’ perceptions of the healthiness and nutrient content previously

demonstrated for claims made on food labels. As previously described in section 1.2.2.3,

consumers perceive products as healthier and of superior nutritional quality when they carry

nutrition marketing messages such as nutrient-content and health claims (80,82-89), and that

25

these perceptions extend beyond just the nutrient that is the subject of the claim (i.e. the ‘halo’

effect) (82,83,86,88,89). Indeed, experts have raised concerns that FOP systems that only

consider a single nutrient in their nutrient profiles (as with some nutrient-specific symbols based

on claims criteria) risk oversimplifying complex nutrition messages (92).

With respect to summary indicator systems, several studies have found that consumers perceive

foods as healthier overall when they carry a summary indicator system compared to when they

do not, even when this perception is not merited by the food’s nutrient content (120-123). For

instance, participants in a crossover experiment who were exposed to the same chocolate mousse

cake on two occasions, once with a summary indicator symbol (Choices) and once without a

symbol, rated the cake as significantly less unhealthy when the symbol was present (122). In

addition, Andrews et al. (123) found that consumer’s randomized to view a frozen meal with a

summary indicator symbol (Smart Choices) rated the meal as significantly healthier than

consumers randomized to see the frozen meal with no symbol. Further, consumers who saw the

symbol in this study also rated the meal as lower in calories, fat, saturated fat, trans fat, sodium,

and sugar than consumers who saw the meal without the symbol. Moreover, consumers who

saw the symbol perceived regular consumption of the meal as less likely to contribute to weight

gain and coronary heart disease risk, and expressed greater intentions to purchase the product.

However, the presence of a mandatory nutrition label alongside the meal with the Smart Choices

symbol increased consumers’ accuracy in evaluating nutrient levels. Another study of the Smart

Choices summary indicator symbol failed to detect any significant differences in consumers’

perceptions of overall healthfulness, and sugar and vitamin content of a breakfast cereal when

the Smart Choices symbol was present versus when it was absent, nor did it lead to self-reported

greater intentions to purchase the product (125). However, it is worth noting that 92% of

participants reported looking at the mandatory Nutrition Facts panel of this cereal (125), and a

moderating effect of the mandatory nutrition label on healthiness evaluations of products with

the Smart Choices symbol was observed in the study by Andrews et al. (123).

Finally, evidence suggests that the effect of summary indicator systems on perceptions of

healthiness and nutrient content is stronger than that of nutrient-specific systems. Andrews et al.

(123) found that while the presence of either a summary indicator symbol (Smart Choices) or a

nutrient specific system (traffic lights) favourably influenced consumers’ perceptions of a frozen

26

meal over the absence of such systems, summary indicator systems influenced favourable

perceptions to a greater extent.

To our knowledge, the impact of food group information systems on consumers’ perceptions of

healthiness and nutrient content has not been studied. Further, at the time this thesis research

was carried out, the impact of different nutrient-specific and summary indicator systems on

Canadians’ perceptions of the healthiness and nutrient content of different foods had not been

evaluated.

While numerous studies have demonstrated that FOP systems influence consumers’ perceptions

of healthiness and nutrient content, and purchase intentions, there is little research actually

demonstrating an effect of FOP symbols on actual purchases and consumption (126). A study of

the Guiding Stars system found that the sales of breakfast cereals with this summary indictor

symbol increased following the system’s introduction in a chain of grocery stores in the

Northeastern US (118). Similarly, the presence of a traffic light system increased the

consumption of foods with “green’ traffic lights and decreased the consumption of foods with

“red” traffic lights in a cafeteria setting (127). In contrast, two studies that examined the effect

of nutrient-specific traffic light systems on product sales found no impact of these symbols on

the sale of ‘healthier’ and ‘less-healthy’ products (128,129). However, it is worth noting that

both these studies were relatively short in duration and only examined the retailer’s own-brand

products from a handful of food categories. Research has found little to no effect of the use of

FOP systems on dietary intakes of individual consumers (64,121). For instance, Borgmeier and

Westenhoefer (64) found five different FOP systems had no impact on energy and nutrient

content of one-day menus selected by German consumers. However, Reid et al (121) found that

self-reported users the Health Check logo consumed 3.5% (p<0.05) fewer calories from fat than

non-users, however it is unclear which behaviour drives the other (130). Similarly, while some

have suggested that the presence of a FOP system on a food label may influence consumers to

consume more of the product than they would if the FOP system was not present, research to

date has not supported this hypothesis (122). While research on the impact of FOP symbols on

product purchases and consumption is limited, Canadian consumers do report looking for these

symbols on food labels. 23% of Canadian label readers report looking for better choice slogans,

symbols or logos [FOP systems] in general on food labels (6), while 23% report occasionally or

regularly looking for the Health Check™ symbol specifically (121). Furthermore, a study out of

27

the Netherlands found that self-reported intentions to purchase products with a summary

indicator symbol, the Choices logo, translated into a larger number of products with the symbol

being purchased at the supermarket (98), suggesting the potential positive influence of FOP

symbols on actual product purchases.

2.6 Is there a need for a single, standardized FOP system?

In section 1.2.1.1 simplified nutrition labels, on the front of food packages, were proposed as a

solution to some of the challenges associated with using the mandatory nutrition label, the NFt,

to choose foods consistent with a healthy diet. FOP systems have been proposed as this

simplified nutrition label (19). However, concerns have arisen that the current FOP system

environment, with multiple co-existing symbols, may be more of a hindrance than help. As

previously noted, experts worry that the presence of multiple FOP systems with different

underlying nutrient profiling criteria may be misleading consumers (13,19,92,94), and there is

also concern that the inconsistent appearance and application of various FOP systems may be

confusing consumers and inhibiting their ability to compare and select foods consistent with a

healthy diet. In fact, following a comprehensive review of the literature on FOP systems, the

Institute of Medicine concluded that multiple systems could confuse the consumer and cause

problems in consumers’ ability to interpret the nutrition information the systems provide (109).

Research conducted in the United Kingdom has found that consumers have difficulties

comparing different types of FOP systems, particularly when the systems shared no common

elements such as gram weights or percent daily values (131). These difficulties were

exacerbated when consumers where asked to compare two products with different FOP systems

where the healthier option was not obvious. Comparing products using different FOP systems

was time consuming for the study participants. While consumers were willing to devote the time

necessary to compare products in this way in the study environment, many indicated that they

would be unlikely to do so in a real-life shopping situation and would look to other information

on the packaging to make their decisions. Qualitative research has found that when comparing

foods with traffic light and daily value systems, consumers sometimes misinterpret the

background colour, such as blue, of monochromatic daily value systems as indicators of low

nutrient levels (132). This is of particular interest as, here in Canada, many manufacturers apply

blue colour schemes in the packaging of the “healthy” lines of their private label brand products

(e.g. President’s Choice Blue Menu or Compliment’s Balance).

28

Concerns about consumer confusion surrounding the use of multiple FOP systems in the same

marketplace has led to expert calls for the adoption of a single, standardized FOP system

(13,56,109). In 2007, the Parliament of Canada’s Standing Committee on Health recommended

that the federal government “implement a mandatory, standardized simple front of package

labelling requirement on pre-packaged foods for easy identification of nutritional value” (13).

Despite this, the Canadian government has taken a hands-off approach to FOP systems, stating

that “it’s great that organizations … have developed systems to help give Canadians even more

information about the food they purchase” and that they do not intend to pursue a single,

standardized FOP nutrition rating system (133).

Beyond the opinions of experts, consumers in many countries have also expressed desire for a

single, standardized FOP system for use on all packaged foods (10,65,131). Confronted with

multiple FOP systems on food products, consumers in the United Kingdom have questioned why

a consistent FOP labelling system has not been adopted by industry (131,132), while, in

Australia, 90% of participants in an intercept survey favoured consistent FOP labelling across all

food products, perceiving it as easier to understand (65). However, to our knowledge,

consumers’ attitudes towards the adoption of a single, standardized FOP system has not been

examined in Canada.

2.6.1 The ideal FOP system for use with consumers

Despite calls for the adoption of a single, standardized FOP system for use in Canada and in

other countries, the ideal FOP system for use with consumers is not clear. Grunert and Wills

(68) have proposed a theoretical model to explain the steps to successful food label use that can

be applied to the context of FOP systems (Figure 2-3). According to the model, exposure to a

FOP system on the food label is the first pre-requite for use of a FOP system in decision-making,

and the chance of exposure is increased if consumers actually search for the system on the label.

If the FOP system is to be used, perception (a.k.a. noticing) of the system and its information

must result from the exposure. What follows perception is liking, or preferences, for the label

and understanding of the label, which can both impact on the use of a FOP system. Grunert and

Wills (68) hypothesize that three basic considerations guide consumers liking of FOP systems: 1)

simplicity; 2) knowledge of what the system stands for and how it was arrived at; and, 3) the

extent to which they feel coerced. Understanding of a FOP system can be of two types:

29

conceptual (objective) understanding and substantive (subjective) understanding (68,134,135).

Conceptual understanding refers to a consumer’s ability to understand the intended meaning

behind a FOP system, while substantive understanding refers to a consumer’s ability to interpret

the information on the FOP system and the extent to which they believe they have interpreted it

correctly. Most of the steps in this model will be influenced by a number of factors, including

the consumer’s interest in, and knowledge of, nutrition issues, the consumer’s demographic

characteristics, and the FOP systems format (68).

Figure 2-3 Conceptual model of FOP system use (adap ted from Grunert and Wills

(68))

2.6.1.1 Consumers’ preferred FOP system

Only a limited number of studies have examined consumer preferences for different FOP

systems (126). In studies comparing consumer preferences for summary indicator versus

nutrient-specific systems, consumers appear to favour nutrient-specific systems (33,62,63,136-

138). A 2007 survey of an ethnically diverse sample of consumers in New Zealand found that

34% of respondents expressed a preference for a nutrient-specific multiple traffic light system,

compared to 24% who preferred a summary indicator single traffic light system, and 13% who

Search

Exposure

Perception

Liking

Understanding

Subjective

Objective

Use

Influences:

Interest

Knowledge

Demographics

FOP system format

30

preferred a nutrient-specific daily value system (33). Likewise, a survey of a large cohort of

French consumers found that 59% of respondents preferred a multiple traffic light system to four

summary indicator alternatives (138). The respondents also perceived the multiple traffic light

system as more likeable and attractive. In addition, Feunekes et al. (63) found consumers from

four European countries perceived a multiple traffic light system as more likeable than four

summary indicator options, and in yet another similar survey (reported on in the same paper),

found that consumers rated a nutrient-specific daily value system as more likeable than three

summary indicator options. Finally, Belgian consumers preferred a nutrient-specific daily value

system to a summary indicator system in a self-administered questionnaire (137). Similarly, in

their review of both the scientific and grey literature on European consumers’ response to

nutrition information on food labels, Grunert and Wills (68) concluded that summary indicators

were the least liked system. Preferences for summary indicator systems that are common in

Canada, including the Heart and Stroke Foundation’s Health Check™ symbol and common

manufacturer developed systems like Pepsi’s Smart Spot and Kraft’s Sensible Solutions, have

not been evaluated against nutrient-specific systems.

In studies that directly compared the two most common nutrient-specific systems – multiple

traffic light systems and daily value systems, – reports of which system is preferred have been

conflicting, but appear to favour multiple traffic lights (33,137,139,140). A qualitative study

conducted with parents in New Zealand reported that consumers perceived traffic light systems

as preferable to use while shopping than daily value systems (139). Equally, as previously noted

a multiple traffic light system was preferred by more respondents than was a daily value system

in a survey of an ethnically diverse group of New Zealanders (33). Further, an Australian study

found consumers rated a traffic light system higher than a daily value system when asked to rate

the systems along several dimensions of consumer friendliness (140). In contrast are the results

of Moser et al. (137), who conducted two similar surveys with populations in Germany and

Belgium that assessed preferences for a daily value system used in both countries compared to a

multiple traffic light system. While German consumers expressed a strong preference for the

traffic light system, Belgian consumers strongly preferred the daily value system. German

consumers rated the traffic light system higher than the daily value system along consumer

friendliness measures such as utility, and design; Belgian consumers rated the daily value system

higher than the traffic light system on along consumer friendliness measures such as clarity,

31

utility, and attractiveness. Daily value systems currently appear on some food labels in Canada

(92,94), however, traffic light systems do not. Therefore it is unclear if Canadian consumers

would display similar preferences, like those noted above, for unfamiliar traffic light systems

versus more familiar daily value systems.

The evidence suggests that perceived credibility or trustworthiness play a role in consumers’

preferences for FOP systems. Feunekes et al. (63) found that in addition to rating a traffic light

system higher on a liking scale than summary indicator systems, consumers also rated the traffic

light system higher with respect to credibility. Similarly, German consumers rated their

preferred traffic light system higher in trustworthiness than their less preferred daily value

system (137).

The results of Moser et al.’s (137) surveys of German and Belgian consumers made clear that

preferences for FOP systems may vary by country, with respondents from these two European

countries having completely opposing preferences for traffic light versus daily value systems.

The results of Feunekes et al’s (63) survey lend further support to this finding: in their survey of

four European countries a significant interaction between FOP system format (six formats were

tested) and country emerged for consumers’ ratings of the likeability and credibility of the FOP

systems. In addition, another European study that tested eight different FOP calorie labelling

systems with consumers in Germany, the Netherlands, France, and the United Kingdom found

differences in consumers evaluations of different FOP systems, as well as in the order of

preferred FOP systems, between countries (10). In addition to differences in preferences at the

country level, there is also evidence to suggest that preferences for different FOP systems differ

by ethnicity (33,62). Gorton et al. found differences in preferences for nutrient-specific versus

summary indicator systems among certain ethnic groups (33). To our knowledge, consumer

preferences for various FOP systems have not been tested in Canada. Given that country-

specific differences appear in consumers’ preferences for various FOP systems, the collection of

Canadian specific data appears warranted.

Finally, research has shown that consumers perceive FOP systems as a valuable addition to

standardized back-of-pack nutrition labelling (10). However, despite the fact that back-of-pack

nutrition labels are mandatory in many countries (15), including Canada (5), only one study to

date has examined preferences for FOP systems in relation to a mandatory nutrition label.

32

Gorton et al. (33) found that 34% of consumers preferred a traffic light system to the mandatory

back-of-pack nutrition label, compared to 26% who preferred the mandatory label. Further, we

are unaware on any study that has tested consumer preferences for different FOP systems in

situations both with and without a standard back-of-pack nutrition label.

2.6.1.2 Understanding

Self-reported understanding of FOP systems was relatively high in most studies, with little

variation in understanding between systems (63,68,132,134,135,138). For example, consumers

in the French survey perceived most FOP systems as easy to understand, with 93% of consumers

or more responding that a variety of summary indicator systems and a multiple traffic light

systems were easy to understand (138). Consumers in the survey of four European countries

scored multiple traffic light systems as easier to understand than summary indicator systems

(63). Comparisons of self-reported understanding of the two common nutrient-specific systems,

traffic light systems and daily value systems, tended to favour the traffic lights (139,140).

However, while consumers in the German survey, discussed previously, rated a traffic light

system as easier to understand than a daily value system, German survey participants from a

another similar study rated the daily value system as easier to understand than the traffic light

system (137). Differences in reported understanding of various FOP systems appear at the

country level, as well as between consumers with different degrees of nutrition knowledge,

interest in healthy eating, label use, and socioeconomic status (63,134,135,137). Self-reported

understanding, however, should be interpreted with caution as research has shown that found that

self-reported understanding of an FOP system does not necessarily translate into better

performance on tests of substantive understanding (63,132).

2.6.1.2.1 Conceptual Understanding

Conceptual understanding (defined in section 2.6.1) of summary indicator systems has not been

directly tested. However, there is evidence to suggest that many consumers attribute the wrong

meaning to them. An early study of the New Zealand National Heart Foundation Heart Tick

symbol found that 43% of consumers thought that to help prevent heart disease you should only

eat foods with the Heart Tick symbol (120). Further, a study of the long-running Swedish

summary indicator symbol, the Green Keyhole, found at least a minimal understanding of the

symbol among consumers (141). When asked to define its meaning 53% of men and 76% of

33

women could, as a minimum, state that the symbol meant a ‘good food’ or ‘healthy food

product’.

With respect to nutrient-specific systems, qualitative research suggests that conceptual

understanding of traffic light systems may be low in some consumers (132). Consumers who

reported they were unfamiliar with such systems thought that the traffic light colours were a

system design element rather than an interpretation of the foods nutrient levels, or that the colour

was related to the nutrient (e.g. fats are always coloured red). Similarly, some misunderstood

daily value systems, with many consumers having a poor understanding of what the percent daily

value means; some consumers interpreted it as the percentage of nutrient in the product. Further,

a study that directly tested conceptual understanding of these two styles of two nutrient-specific

systems found only moderate levels of conceptual understanding among consumers in the United

Kingdom (135). A study that directly tested conceptual understanding of daily value systems in

six European countries found low levels of conceptual understanding with substantial variation

between countries (134). The conceptual understanding of different FOP system styles in

Canadians has not been tested.

2.6.1.2.2 Substantive Understanding

Studies of consumers’ substantive understanding of FOP systems have been inconsistent in their

findings (33,63,65,132,138). A handful of studies in Europe, the United Kingdom, and Australia

have used experimental designs to test the substantive understanding of summary indicator

systems and nutrient-specific systems with consumers (63-65,132,135). Most of these studies

exposed participants to food product pairs, or sets of three, carrying FOP systems and had

consumers identify the healthier product. Most studies that have tested nutrient-based systems

against one another have found either no differences between traffic light systems and daily

value systems, or that traffic light systems are better at helping consumers identify healthier

choices than are daily value systems (64,65,132). For instance, Kelly et al. (65) found 81% of

participants were able to correctly identify the healthier of two products using a traffic light

system, compared to 64% using a monochromatic daily value system. The only conflicting study

found a daily value system was marginally better than the traffic light system at helping

consumers identify the healthier products (88.0% versus 86.6%) (135).

34

The results of studies that tested a combination of summary indicator systems and nutrient-based

systems have been less consistent. For instance, in one study Feunekes et al. (63) found that

graded summary indicator systems (which gave foods one to five stars or smileys) and multiple

traffic light systems were significantly better than score based summary indicator systems at

helping consumer differentiate products based on healthfulness (P<0.01). Furthermore, the

summary indicator systems in general performed most consistently in helping consumers

differentiate between a range of product categories (P<0.01). However, a later study by the same

researchers found virtually no significant differences between summary indicator systems and a

nutrient-based system in helping consumers differentiate between healthy and less healthy

product variants (63). Furthermore, Borgmeier and Westenhoefer (64) found that nutrient-based

systems were superior to summary indicator systems in helping German consumers identify

healthier choices in pair-wise food comparisons (p<0.01).

Other studies have asked consumers to rate the overall healthiness and nutrient content of single

foods to assess subjective understanding of FOP symbols (33,65,132,135,138). For example,

Mejean et al. exposed consumers to three soup variants without a back-of-pack nutrition label

and asked them to respond true or false to five statements related to healthiness and nutritional

value of these soups (138). Consumers demonstrated higher levels of subjective understanding

of summary indicator systems with 58-63% of respondents scoring 4/5 or 5/5 on the true/false

quiz with these systems compared to 53% of respondents achieving the same result with a

multiple traffic light system. Similarly, Gorton et al. (33) exposed consumers to one of three

FOP systems, or a back-of-pack nutrition label, on a package of crackers and asked them

whether they considered the food ‘healthy’ or ‘not healthy. A summary indicator single traffic

light system resulted in the most correct responses (83%), followed by a multiple traffic light

system (80%), and lastly a daily value system (49%). 54% of consumers responded correctly

using the mandatory nutrition label. Studies that compared substantive understanding of

nutrient-specific systems have found that traffic light systems perform better than daily value

systems (65,132). An Australian study found consumers were more accurate in the identification

of nutrient levels when they used traffic light systems compared to when daily value systems

were used (65). Further, two studies have found that consumers could evaluate the overall

healthiness and nutrient content of a single product with a higher degree of accuracy using either

traffic light or daily value systems (132,135). One of these studies also found that substantive

35

understanding of both traffic light and daily value systems could be increased by the presence of

text describing nutrient levels as “high”, “medium”, or “low” (132).

2.7 Summary

There is myriad nutrition information, both mandatory (such as the NFt) and voluntary (such as

nutrition marketing information) found on Canadian food labels. Evidence suggests that

mandatory nutrition information is both under utilized and poorly understood by consumers

while the large amount of voluntary nutrition information may be confusing and misleading

consumers. FOP nutrition rating systems and symbols are a form of voluntary nutrition

information on food labels that has received a lot of attention in recent years. Since their

introduction in the late 1980s, FOP symbols have proliferated in Canada and internationally,

such that there are now a multitude of symbols, each with their own unique appearance and

underlying nutrient profiling criteria. While the intention of FOP systems is to help consumers

make healthier choices, presently, it is unclear if consumers can rely on FOP symbols to identify

products that meet higher nutritional standards. Despite this, consumers appear to perceive

products with FOP symbols as healthier than those without. As a result, experts worry that the

presence of multiple FOP systems in the marketplace may be confusing and misleading

consumers, prompting calls for a single, standardized system to be used on all food products.

Notwithstanding these calls, the ideal FOP system for use with consumers has not been

identified, and Canadian consumers have not been studied to identify their preferred FOP format.

Moreover, Canadians attitudes toward the adoption of a single, standardized FOP system have

not been examined.

36

Chapter 3

3 Scope and Hypotheses of Thesis

3.1 Scope and Objectives

This thesis was designed to generate food supply and consumer data to support the evolution of

Canadian food labelling policy as it relates to FOP systems. In the current Canadian

environment, there is a lack of specific regulations for FOP systems. Presently, it is unclear if

FOP systems are being used to promote foods that meet higher nutritional standards due to a lack

of information on the nutritional quality of foods with FOP systems compared to those without.

Such evidence is needed to evaluate the efficacy of Canada’s current hands-off approach to FOP

system regulation. Moreover, although a Canadian expert committee has called for the adoption

of a single, standardized FOP system to support consumers in making food choices consistent

with a healthy diet (13), there is no evidence to suggest that Canadian consumers would support

such an initiative. Further there is a lack of Canadian data on consumers’ liking and

understanding of different FOP system options to support the identification of an appropriate

single, standardized system should this avenue be pursued. Thus the overall objectives of this

work were to:

1. Assess the extent to which certain voluntary FOP symbols identify all foods whose

nutritional composition meets the system’s nutrient profiling criteria.

2. Compare the calorie, saturated and trans fat, sodium, and sugar content of foods with

FOP symbols to foods without symbols by system type.

3. Describe Canadian’s attitudes to the regulation of FOP systems.

4. Determine experimentally which FOP system is most liked and understood by Canadians.

3.2 Specific hypotheses

We hypothesized that:

1. More foods would meet the nutrient profiling criteria of certain voluntary FOP systems

than would actually carry these symbols on their food label.

37

2. In the absence of minimum, standardized nutrient profiling criteria for foods carrying

FOP symbols on their food labels, foods with FOP symbols versus those without would

contain similar levels of calories, saturated and trans fat, sodium, and sugar.

3. Canadian consumers would support the adoption of a single, standardized FOP system.

4. Canadian consumers would prefer nutrient-specific FOP systems to summary indicator

systems.

5. Canadians’ conceptual understanding of various FOP systems would be low.

3.3 Preview of Chapters 4-6

Objective 1 was examined in Chapter 4 using a 2010-2011 national database of Canadian

packaged food labels known as the Food Label Information Program (FLIP). In this

investigation, we applied the nutrient profiling criteria of two popular summary indicator

systems, a third-party symbol and a manufacturer developed symbol, to the foods in the FLIP

database to determine the proportion of products that would qualify to carry these systems’

symbols. The proportion of products qualifying to carry one of these two symbols was compared

to the proportion of products actually carrying these symbols.

Objective 2 was examined in Chapter 5 by expanding our work with the FLIP database. In this

study we systematically scanned all products in FLIP for the presence of FOP symbol nutrition

marketing, and categorized the FOP symbols used on the label by type (nutrient-specific systems

based on claims criteria, summary indicator systems, food group information systems) as per the

definitions put forward by the US Institute of Medicine. We then compared the calorie, saturated

and trans fat, sodium, and sugar content of foods with and without this form of nutrition

marketing to determine if FOP symbols are a reliable indicator of products with healthier nutrient

profiles.

Objectives 3 and 4 were examined in Chapter 6 using an online, national consumer survey.

Objective 3 was addressed through survey questions designed to discern Canadians’ attitudes

toward the use and regulation of FOP systems with Canada. Objective 4 was addressed through

a randomized mock package experiment that tested five FOP conditions (no FOP systems, two

summary indicator systems, two nutrient-specific systems) on two mock food packages with or

without the mandatory NFt. Consumer friendliness was examined with respect to perceived

38

liking, helpfulness, credibility, understanding, and influence on purchase decisions using a 5-

point likert-type scale.

Together these three studies provide data on the quality of current FOP nutrition rating system

environment, the effectiveness of Canada’s current policies on FOP systems, and suggest future

directions for FOP systems policy in Canada.

39

Chapter 4

4 Food products qualifying for and carrying front-of-pack symbols: a cross-sectional study examining a manufacturer led and a non-profit organization led program

This manuscript has been published: Emrich TE, Cohen JE, Lou WY, L’Abbe MR. Food

products qualifying for and carrying front-of-pack symbols: a cross-sectional study examining a

manufacturer led and a non-profit organization led program. BMC Public Health [Internet]. 2013

Sept [cited 2014 Jan 6]; 13(846):[about 9 pp.]. Available from:

http://www.biomedcentral.com/1471-2458/13/846

This study addressed objective #1 of my thesis, to:

• assess the extent to which certain voluntary FOP symbols identify all foods whose

nutritional composition meets the system’s nutrient profiling criteria.

Student’s contribution:

The original idea for this study and its design were mine. I was involved in the development of

the FLIP database as part of the L’Abbe lab group, including data collection at grocery retailers,

as well as data entry and validation. I independently reviewed each food label in FLIP for the

presence of FOP systems (using the definitions described in the manuscript) and classified each

FOP system identified by type (with validation from a second independent reviewer). For this

study, I independently classified all FLIP products into the appropriate Health Check™ and

Sensible Solutions™ food categories and applied the nutrient criteria for each of these FOP

systems. I designed the analysis of the data with the help of Dr. Wendy Lou and her graduate

student, Ying Qi, and carried out the analyses independently. I completed the original

interpretation of the data and independently prepared a draft of the manuscript prior to engaging

my co-authors in reviewing and revising the manuscript that was published in BMC Public

Health.

40

4.1 Abstract

Background: Concern has been raised that the coexistence of multiple front-of-pack (FOP)

nutrition rating systems in a marketplace may mislead consumers into believing that a specific

food with a FOP is ‘healthier’ than foods without the symbol. Eleven summary indicator FOP

systems are in use in Canada, including one non-profit developed system, the Heart and Stroke

Foundation’s Health Check™, and ten manufacturer-developed systems, like Kraft’s Sensible

Solutions™. This study evaluated FOP’s potential to mislead consumers by comparing the

number of products qualifying to carry a given FOP symbol to the number of products that

actually carry the symbol.

Methods: The nutritional criteria for the Health Check™ and the Sensible Solutions™ systems

were applied to a 2010–2011 Canadian national database of packaged food products. The

proportion of foods qualifying for a given FOP system was compared to the proportion carrying

the symbol using McNemar’s test.

Results: Criteria were available to categorize 7503 and 3009 of the 10,487 foods in the database

under Health Check™ and Sensible Solutions™, respectively. Overall 45% of the foods

belonging to a Health Check™ category qualified for Health Check™ symbol, while only 7.5%

of the foods carried the symbol. Up to 79.1% of the foods belonging to a Sensible Solutions™,

category qualified for Sensible Solutions’s™ symbol while only 4.1% of the foods carried the

symbol. The level of agreement between products qualifying for and carrying FOP systems was

poor to moderate in the majority of food categories for both systems. More than 75% of the

products in 24 of the 85 Health Check™ subcategories and 9 of 11 Sensible Solution™

categories/subcategories qualified for their respective symbols based on their nutritional

composition.

Conclusions: FOP systems as they are currently applied are not, in most instances, a useful

guide to identifying healthier food products in the supermarket as many more products qualify

for these systems than the number of products actually displaying these symbols on FOP, and the

level of agreement between qualifying and carrying products is poor to moderate. The adoption

of a single, standardized FOP system would assure consumers that all products meeting certain

nutritional standards are designated by the symbol.

41

Keywords: Nutrition labelling, Front-of-pack nutrition rating systems, Nutrient criteria

42

4.2 Background

The World Health Organization has stated “consumers require accurate, standardized and

comprehensible information on the content of food items in order to make healthy choices” (3).

To that end, mandatory nutrition labels have been adopted in more than 20 countries, including

the European Union member states, Mexico, and China, and voluntary nutrition labels have been

adopted in at least 11 more (15). In Canada, regulations mandating nutrition labelling on most

packaged foods were adopted 2003 in response to mounting evidence of the contribution of diet

to chronic disease (5). The Canadian Nutrition Facts table reports the amount of calories, fat,

saturated and trans fat, cholesterol, sodium, carbohydrate, fibre, sugar, protein, vitamin A,

vitamin C, calcium, and iron per serving of a food and is similar to the Nutrition Information and

Nutrition Facts panels used in countries such as the US, United Kingdom, Australia, and New

Zealand. At the same time the Nutrition Facts table was adopted, Canada updated regulations for

the use of nutrient content claims and established rules for the use of diet-related health claims

on food products. Canada is just one of many countries, including Japan, China, Australia and

New Zealand, European Union member states, and the US, permitting some form of nutrient or

health claims on food labels (142).

Not included in Canada’s 2003 regulatory revisions was another form of food label nutrition

information, front-of-pack (FOP) nutrition rating systems and symbols. FOP systems provide

simplified information about the nutritional characteristics of a food and have been in use

internationally since American Heart Association first launched its Heart Guide initiative (1987)

and Sweden’s National Food Administration created its Keyhole symbol (1989) (19). Despite

being used internationally for more than 25 years, few specific regulations are in place governing

their use, although standardized FOP systems are being considered in several countries (15). In

Canada a mandatory FOP system is not presently being considered, and the only regulatory

requirement currently governing the use of FOP systems is that they not be “false, misleading, or

deceptive” (26). To minimize the potential for misrepresentation, the Canadian Food Inspection

Agency has issued additional guidance that FOP systems should not give the impression “that a

single food or brand of food is “healthier” than … other foods not bearing the [FOP symbol]”

(26).

43

Since the introduction of the Heart Guide and Keyhole programs, the number of FOP systems in

the marketplace internationally has multiplied (15,19). Each of these FOP systems has their own

unique symbol and nutritional criteria to identify qualifying products. One hundred fifty-eight

unique FOP systems have been identified in the Canadian marketplace, including 11 summary

indicator systems that use a single symbol on products that meet the system’s criteria (78). Of the

summary indicator systems, there was only one third-party, non-profit developed system, the

Heart and Stroke Foundation of Canada’s Health Check™ symbol (Figure 4-1). The Health

Check™ symbol can be placed on qualifying products from any manufacturer (provided the

manufacturer has paid into the program) (143). The remaining 10 systems were manufacturer or

industry-developed and their symbols were placed exclusively on qualifying products of the

proprietary manufacturer (7). Examples of manufacturer-developed summary indicator systems

in use in Canada include Kraft’s Sensible Solutions™ (Figure 4-1), Lassonde’s Health

Signature®, Old Dutch Foods’ Snack Wise™, Pepsi’s Smart Spot™.

Many summary indicator systems are based on nutrient thresholds that establish maximum levels

for nutrients to limit and minimum levels for nutrients or food components to encourage and

often use different thresholds for different food categories (144,145). The number of food

categories covered by each FOP system ranges from as few as one (Snack Wise™ is only

applied to snacks) to as many as 85 (Health Check™ criteria were available for 85 sub-categories

found within 6 major categories at the time data were collected) (145,146). Appendix A lists the

Health Check™ food categories [see Appendix A]. Sensible Solutions™ has developed criteria

for the most food categories of all the manufacturer-developed summary indicator systems, with

criteria available for eight major food categories and five additional subcategories (94,144). An

Appendix lists the Sensible Solutions™ food categories [see Appendix B]. FOP systems in

Canada included both absolute and relative nutrient thresholds (92,94). Absolute thresholds

establish minimum and maximum levels for nutrients and food components, whereas relative

thresholds are set relative to the nutrient content of an appropriate reference product. In the latter

case, products can qualify for a symbol by being lower in a nutrient like calories, fat, saturated

fat, sugar, or sodium, than the reference; thus, symbols can potentially appear on foods high in

saturated or trans fat, sugar, or sodium but low in the nutrient of interest. The nutrient criteria of

summary indicator systems, and manufacturer-developed criteria in particular, whether absolute

or relative, have been accused of not being stringent enough (56).

44

As FOP systems have multiplied, so too have concerns that these systems may be confusing and

misleading to consumers (13,19,92,94). In the most comprehensive review of FOP systems to

date, the Institute of Medicine concluded that the coexistence of multiple FOP systems, with

different nutritional criteria, make it difficult for consumers to interpret nutrition information and

to compare products (109). Expert reports have raised further concerns about the role of FOP

systems in hindering product comparisons, worrying that nutritious foods not part of a FOP

program may, by default, be perceived as less healthy (92,94). This is of concern, as many foods

may not carry FOP symbols for reasons unrelated to their nutritional value. However, just how

many foods are being excluded from carrying a specific FOP symbol for reasons unrelated to

nutritional composition has not been examined.

This study assesses the proportion of Canadian grocery products that qualify for a Health

Check™ or a Sensible Solutions™ symbol based on their nutritional composition compared to

the proportion of products that actually carry these symbols in order to evaluate the potential of

the current FOP systems to mislead consumers.

Figure 4-1. Front-of-pack symbols evaluated in the present study.

A

B

(A) Heart and Stroke Foundation Health Check™ and (B) Kraft Sensible Solutions™.

45

4.3 Methods The FOP systems examined in this research were the Heart and Stroke Foundation’s Health

Check™ and Kraft’s Sensible Solutions™. Health Check™ was chosen as it is the only non-

profit, third-party summary indicator FOP system identified to date in Canadian reports

(78,92,94). Sensible Solutions™ was chosen because it is the manufacturer-developed FOP

system with nutrient criteria established for the largest number of food categories (94).

The nutrient criteria used to determine if a product qualified for the Health Check™ or Sensible

Solutions™ symbol were obtained from the systems’ proprietors (144,145). The Health Check™

criteria are based on levels of total fat, saturated and trans fat, sodium, carbohydrates, fibre,

sugar, protein, and vitamins and minerals and the presence of fruit and vegetables and whole

grains. Sensible Solutions™ criteria are based on the same nutrients and food components as

Health Check™ (145), but also include criteria related to calories, cholesterol, added sugars,

serving size, and functional nutritional benefits (144). Both FOP systems use threshold criteria

but differ, 1) at what level the thresholds have been set, and, 2) with respect to the application of

relative thresholds. In addition to allowing products to qualify for their symbol by meeting

absolute thresholds, Sensible Solutions™ also allows some products to qualify for their symbol

using relative threshold criteria (provided that it passes a review by Kraft’s Nutrition

Department) (144). Examples of Health Check™ and Sensible Solution™ criteria for crackers

are found in Table 4-1 and the remaining criteria are publicly available online from Health

Check™ (http://www.healthcheck.org) and Kraft (http://www.kraftcanada.com).

Data for this study were drawn from the FLIP, a national database of food label information

developed at the University of Toronto (78). The FLIP includes the food label information from

10,487 national and private label grocery products in 23 food categories collected throughout

2010–2011 from the three largest grocery retailers in Canada (Loblaw Inc, Sobeys Inc, and

Metro Inc) and one major western Canadian grocery retailer (Safeway). As previously described

by Schermel et al. (78), by systematically scanning the grocery store shelves we aimed to collect

every food product with a Nutrition Facts table within each of the 23 categories, including all

available national and private label brands, but excluding seasonal products (e.g. egg nog) and

foods from the natural health section of each store. Food products sold at multiple retailers were

only purchased once and when multiple sizes of a product were available, only one size was

purchased. Information recorded from the food labels into the FLIP database included the

46

product name, nutrition information, and FOP symbols used.

Nutrition information from the Nutrition Facts table and the ingredient list were used to

determine which products qualified for the FOP systems under study. Nutrients such as Vitamin

E, magnesium, potassium, and folate, which are not required in the Nutrition Facts table, as well

as the quantity of whole grains or servings of vegetables and fruit in the product were included in

some of the criteria for both the Health Check™ and Sensible Solutions™ systems as “or”

statements (i.e. products could qualify by being a source of one of these nutrients or food

components or by being a source of another nutrient listed on the Nutrition Facts table)

(144,145). In this study criteria were only applied to those nutrients and food components that

were available from the Nutrition Facts table or ingredient list.

All FLIP products were classified into the appropriate Health Check™ and Sensible Solutions™

food categories by a single coder and the nutrient criteria for each of these FOP systems were

applied. Food categories were verified by a second coder in a random sample of 5% of products

and less than 0.5% of verified products were found to be misclassified. For the Sensible

Solutions™ relative nutrient criteria, the mean calorie, fat, saturated fat, sugar, and sodium

content for each food category/subcategory was calculated to create the reference product used to

determine if a product is lower (25%) in these nutrients. It should be noted the Kraft’s relative

threshold criteria were designed to compare products to a base product (such as the original

product variant) or an appropriate reference product (not publicly identified by the

manufacturer). In the absence of information on the composition of the base or reference

products for all products in FLIP, category reference products were established based on means

for the category or subcategory.

Data were analyzed using SAS software (version 9.3, SAS Institute Inc., Cary, NC, 2011).

McNemar’s test was used to compare paired proportions, specifically testing whether the

proportion of products qualifying for FOP symbols was different from the proportion of products

carrying FOP symbols within the same food category or subcategory. Kappa coefficient was

calculated to measure the agreement between products qualifying for and carrying FOP symbols

(147). The kappa coefficient measures the difference between observed agreement and expected

agreement and lies on a scale of −1 to 1, where 0.0 is considered ‘poor’ agreement, 0.2 ‘slight’,

0.4 ‘fair’�, 0.6 ‘moderate’�, 0.8 ‘substantial’�, and 1.0 ‘almost perfect’ agreement. Statistical

47

significance level was set at p�<�.05, unless stated otherwise.

48

Table 4-1. Nutrient criteria for crackers to qualify for Health Check™ and Sensible Solutions™ symbols

FOP system Health Check™ (145) Sensible Solutions™ (9)

Food Category Crackers/Rusks Cookies & Crackers

Amount of Food Per 20 g serving and per on-pack serving Per serving

Must meet all of the following absolute threshold nutrient criteria:

Calories

Fat

Saturated and trans fat

Sodium

Added sugar

Other

No criteria

≤ 3 g

≤ 2 g + ≤15% of calories (combined)

≤5% of total fat

≤ 480 mg (per 50 g)

No criteria

No criteria

≤ 100 calories

≤ 30% of calories

≤10% of calories (combined)

≤ 290 mg

≤25% of calories

A “source of” Vitamin A, C, E, calcium magnesium, potassium, iron, protein, fibre; or,

Contain at least a half-serving of fruit, vegetable, or a nutritionally significant amount of whole grain; or,

Has a functional nutrition benefit.

Or must meet one of the following relative threshold nutrient criteria:

Calories

Fat

Saturated fat

Sugar

Not applicable* Must be free of, or low in, one of these nutrients, or must have 25% less of one of these in comparison to the base product or an appropriate reference product

Must be reviewed by the Nutrition

49

Sodium Department

* The Health Check™ system does not use relative threshold criteria.

50

4.4 Results Criteria were available to categorize 7503 (71.5%) and 3009 (28.7%) of the 10,487 food

products in FLIP under the Health Check™ and Sensible Solutions™ FOP systems, respectively.

Health Check™ and Sensible Solutions™ did not have nutrient criteria established for the

remaining, unclassified foods. FLIP had food products from 81 of Health Check™’s 85

subcategories [see Appendix A]. No food products were collected from the following

subcategories: Fresh fruit (unpackaged foods were not collected); Vegetarian terrines, spreads, or

pates; Egg substitutes; and, Nut and/or seed bars.

Details on food products qualifying for, and carrying, FOP symbols by product category are

found in Figure 4-2. Overall, 3364 (44.8%) of the food products for which Health Check™

criteria were available met the nutrient criteria required to carry the symbol, while only 560

(7.5%) of the food products actually carried the symbol. Similarly, significantly more food

products qualified for the Health Check™ symbol than carried the symbol in 56 of the 85

program subcategories. Full details of the proportion of products qualifying for, compared to

carrying, the Health Check™ symbol by subcategory, including significant differences, are

attached [see Appendix A]. In most subcategories where significant differences were not

observed, there were either a very low percentage of products qualifying or the database

contained very few products in the subcategory. For Sensible Solutions™, when the absolute

threshold nutrient criteria were used, 737 (24.5%) of products for which Sensible Solutions™

criteria were available met the eligibility criteria for the symbol; in contrast, when the relative

nutrient criteria were used, 2379 (79.1%) of the same products were eligible for the symbol.

Overall, only 122 (4.1%) of the products in a Sensible Solutions™ food category carried the

system’s symbol. Full details by subcategory are available [see Appendix B].

False positives (products that carried a symbol but did not meet the criteria) were rare, with more

than 92% of products carrying a Health Check™ symbol meeting the relevant systems criteria.

False positives were found in 13 Health Check™ subcategories with the majority of false

positives in the ‘Combination foods’ subcategories. In most instances, false positives occurred

because the food failed to meet the nutrient criteria per reference amount (a standard serving size

established for each food category), although they met the nutrient criteria per on-pack serving.

51

No false positives were observed with Sensible Solutions™.

There was substantial agreement (kappa >0.8) between the number of products qualifying for

and carrying the Health Check™ symbol in only four subcategories, ‘dried fruit snacks’,

‘croutons’, ‘canned legumes’, and ‘stuffed pasta’. Appendix A provides the level of agreement

(kappa statistic) for all Health Check™ subcategories. Poor agreement (kappa <0.2) between

qualifying and carrying products was observed in many subcategories: ‘vegetable and fruit’ (11

out of 17 subcategories); ‘grain products’ (13/19), ‘dairy products’ (7/10); ‘meat and alternative’

(14/23); ‘fats and oils’ (3/4); and, ‘combination foods’ (4/12). The highest levels of agreement

between products carrying and qualifying for Sensible Solutions™ were observed when absolute

threshold criteria were applied to ‘refreshment beverages’ and ‘cookies and crackers’, however

even within these categories the level of agreement was poor (kappa <0.2). Appendix B provides

the level of agreement (kappa statistic) for all Sensible Solutions™ categories and subcategories.

The Health Check™ and Sensible Solutions™ nutrient criteria were not equally discriminating

in identifying “healthier” choices across food categories (Figure 4-2 and Appendixes A and B).

In eight of the 18 ‘vegetable and fruit’, seven of the 23 ‘meat and alternative’, and two of the

four ‘oils and fats’ subcategories, more than 75% of the products qualified for the Health

Check™ symbol [see Appendix A]. In contrast, in seven of the 12 ‘combination foods’

subcategories, less than 25% of the products qualified for the Health Check™ symbol. With

respect to the Sensible Solutions™ system, fewer than 25% of the foods in each category

qualified for the symbol when the absolute threshold criteria were used, with the exceptions of

‘convenient meal products’ and ‘100% juice’ [see Appendix B]. However, when the relative

threshold criteria were used, more than 70% of foods in each food category qualified for the

Sensible Solutions™ symbol.

The FLIP database contained 409 products made by Sensible Solutions™’s proprietor, Kraft

Canada. Forty-five (11.0%) of Kraft’s products qualified for their symbol on the basis of the

system’s absolute threshold criteria and 361 (88.3%) of their products qualified on the basis of

the system’s relative threshold criteria. However, only 122 (29.8%) of the Kraft products in the

FLIP actually carried the Sensible Solutions symbol. The lower proportion of Kraft products

carrying the Sensible Solutions™, symbol relative to the number of their products that could

52

qualify based on relative threshold criteria suggests that a significant number of products are

disqualified at the required review phase by Kraft’s Nutrition Department.

53

Figure 4-2. Proportion of food products that qualified for, compared with the proportion of

food products that carried, the different front-of-pack symbols.

(A) Heart and Stroke Foundation Health Check™. Criteria were available to categorize 7503 of the 10,487 food products in the FLIP under the Health Check™ system. (B) Kraft Sensible Solutions™. Criteria were available to categorize 3009 of the 10,487 food products in the FLIP under the Sensible Solutions™ system. † Differ significantly from the proportion of food products carrying a front-of-pack symbol p<0.05.

54

4.5 Discussion The findings from the present study showed that significantly more products met the Health

Check™ and Sensible Solutions™ nutrition rating systems’ definition of ‘healthy/healthier’ (as

described in their respective nutrient criteria) than carried either of these FOP symbols in most

food categories. Past research has found that, given two similar foods, one carrying the Health

Check™ symbol and one without it, 80% of consumers would perceive the product with the

symbol as ‘probably a better choice’ while only 4% of consumers would perceive there was ‘no

real difference’ between the two products (121). Similarly, researchers found that consumers

exposed to a FOP symbol on a mousse cake perceived the cake as healthier than consumers who

were given the same cake without a symbol (p�=�0.004) (122). The magnitude of perceived

differences in the healthiness of foods appears to be influenced by the format of the FOP symbol

(60,63,64,109). Given the large number of products that qualify for, yet do not carry these

symbols, our findings suggest that the two FOP systems under study may give consumers the

erroneous impression that foods carrying the symbols are healthier than a similar product without

these symbols – contrary to the Canadian guidance regarding the use of FOP systems (26).

When absolute threshold nutrient criteria were used, a smaller proportion of products qualified

for Sensible Solutions™ than Health Check™. However, when Sensible Solutions™ relative

threshold nutrient criteria were used, a larger proportion of products qualified for the Sensible

Solutions™ than Health Check™. In fact, the relative threshold criteria appeared poor at

differentiating between healthy and less healthy products, with a large proportion of products

qualifying for the symbol in most food categories when these criteria were applied (Figure 4-2).

Based on our findings, relative threshold nutrient criteria were less able to discriminate between

products based on healthiness. However it should be noted that Kraft designed the relative

nutrient criteria to be applied relative to a base product (e.g. a reduced fat Oreo cookie compared

to a regular Oreo cookie) or matched with an appropriate reference product. Thus the use of

category means as the reference nutrient levels for determining which products qualify based on

relative thresholds is a weakness of this analysis. Furthermore, the secondary assessment by

Kraft’s Nutrition Department of products that qualify based on relative threshold criteria is not

documented, and could not be applied in this study.

With respect to the Health Check™ system, this study found only four subcategories where there

was substantial agreement between the number of products qualifying for and carrying symbol.

55

Considering consumers perceive products with the Health Check™ symbol as healthier than

similar products without the symbol (121), our findings suggest that Health Check™ may be a

useful guide to choosing healthier products for consumers in very few subcategories. Most

subcategories within each of the major Health Check™ categories showed only poor agreement

between products qualifying for and carrying this symbol. However, the consumer has no way to

determine in which food subcategories the Health Check™ symbol identifies most products that

meet the system’s definition of healthy, limiting its utility as a guide to healthier choices.

However, universal implementation of a FOP system like Health Check™ or similar threshold

based system to all products (not just those that have bought into the program) would allow

consumers to better differentiate between healthy and less healthy food choices within all food

categories. Indeed, in their 2011 report on FOP nutrition rating systems, the US Institute of

Medicine recommended that an ideal FOP system should be applied to all grocery products

(109).

Proponents of FOP systems suggest these systems have the potential to encourage product

reformulation by manufacturers to meet their nutrient criteria (19). The few studies that have

examined this issue, including one focused on Health Check™, found that FOP systems

successfully encouraged manufacturers to lower the sodium, saturated and trans fat, and calories

in their products (112-115). However, in 24 of 85 Health Check™ subcategories, greater than

75% of products already met the criteria, suggesting that options for reformulation would be

minimal [see Appendix A]. Similarly, when the Sensible Solutions™ relative nutrient threshold

criteria were applied, more than 70% of products in many food categories qualified for the

systems’ symbol. The results of this study would suggest that, within some product

subcategories, the nutrient criteria of Health Check™ and Sensible Solutions™ (especially the

relative thresholds) should be strengthened if they are to encourage the reformulation of more

food products in a healthful way. Indeed, the Health Check™ program has been continually

adjusting its nutrient criteria to encourage manufacturers to reformulate their products to reduce

the amount of nutrients such as sodium and trans fat in the food supply (145,148,149).

The US Institute of Medicine expert committee recommended in their 2011 report that the model

FOP system should be applied universally and be based on absolute thresholds for saturated and

trans fat, sodium, and sugar for two food categories, individual foods and main dishes/meal

products, to allow for the comparison of foods within and across categories (109). In contrast to

56

this recommendation, the two systems in this study, Health Check™ and Sensible Solutions™,

had different nutrient criteria for up to 85 different subcategories, thereby limiting consumers’

ability to use these FOP systems to compare products across categories. Furthermore, the

nutrient criteria used in both of these systems are based on additional nutrients beyond saturated

and trans fat, sodium, and sugar, despite the Institute of Medicine’s finding that there is

insufficient evidence to suggest that including such nutrients in a FOP system would be useful.

Finally, the expert committee proposed that the model FOP system should take a ranked

approach to nutritional guidance where, after meeting a minimum eligibility threshold, products

could earn and display additional nutritional “points” based on their content of those core three

nutrients. However neither system in this study offered additional ranking interpretation of the

nutritional quality of products, thereby limiting their full ability to inform consumers and to

promote continued product improvements.

In addition to the model FOP system proposed by the Institute of Medicine, a number of single,

standardized (mandatory or voluntary) FOP systems are being proposed or adopted by

governments, experts, and industry groups in countries such as the US, European Union member

states, Australia and New Zealand, and South Korea (15). The proposed systems are largely

nutrient-specific, including nutrients such as saturated fat, sodium, and sugar, and display the

amount per serving or per 100 g on the FOP. In addition, the use of traffic light colours to

identify high (red), moderate (amber), and low (green) amounts of nutrients is under

consideration in a subset of these countries. Summary indicator FOP systems, such as Health

Check™ and Sensible Solutions™, are not presently being considered in any jurisdiction for

universal implementation. If Canada were to consider adopting a single, standardized FOP

system they may want to consider an approach more consistent with what is being proposed

internationally, particularly by the Institute of Medicine. The Institute of Medicine proposes two

important features that are not covered by current FOP systems; 1) they recommend a graded

system with one to three stars or checkmarks awarded depending on nutritional composition;

and, 2) products that don't meet the basal criteria, would carry the FOP symbol with zero stars or

checkmarks.

Strengths of this study include the large number of food categories and subcategories examined,

as well as the inclusion of both a non-profit led and a manufacturer led system. In addition, the

FOP systems under study were quantitatively evaluated within the context of the entire food

57

supply and evaluation was not limited to the products of the FOP system’s proprietary

manufacturer or the products of manufacturers that have bought into the non-profit FOP system.

There are a few limitations to this study. First, the nutritional composition of products was based

on the Nutrition Facts table and data were only available on the 14 core nutrients found in the

nutrition label. Both Health Check™ and Sensible Solution™ included some criteria for

nutrients and food ingredients not included in the Nutrition Facts table, thus the present study

may have underestimated the number of products qualifying for these systems based on the

absence of data on these nutrients and food ingredients. Furthermore, the Nutrition Facts table

does not differentiate between total and added sugar, which is used in the Sensible Solutions™

system. As we were unable to differentiate between total and added sugar we applied this criteria

to total sugar and, as a result, may have underestimated the total number of products that would

qualify for Sensible Solutions™. In addition, reliance on nutrient content values reported in the

Nutrition Facts table instead of values determined through chemical analysis may have decreased

the precision of our results. However, one recent Canadian study of five food categories found

no significant differences between nutrient content values reported in the Nutrition Facts table

compared to values determined through chemical analysis for saturated and trans fat, indicating

that the Nutrition Facts table values are quite precise (150).

4.6 Conclusions Within Canada’s current labelling environment, where FOP systems are not universally applied,

nor subject to specific regulations, substantial agreement between the number of products

qualifying for and carrying symbols was only found in a minority of food categories. As a result,

health professionals should advise their clients that FOP symbols, in their current application,

cannot be reliably used to identify food products that meet higher nutritional standards than other

similar products. Overall, many more products qualified for FOP nutrition rating systems than

carried them, thus supporting concerns that FOP systems could mislead consumers into thinking

that products with a FOP symbol are healthier than those without when this is not actually the

case. Given the proliferation of FOP systems internationally, similar analyses should be

undertaken in other countries to determine the extent to which FOP systems highlight all

products that meet higher nutritional standards. As governments and industry groups implement

single, voluntary FOP systems the extent of uptake should be monitored; voluntary systems may

58

rarely be applied to products with poor nutritional quality, and as demonstrated here, if adoption

is not widespread FOP systems may mislead the consumer if they believe products with symbols

to be healthier than comparable products without an FOP. This analysis suggests consumers may

benefit from a single, standardized FOP symbol that identifies all food products that meet a

common set of nutritional standards – such as those proposed or under consideration in several

countries (15).

59

Chapter 5

5 Front-of-pack symbols are not a reliable indicator of products with healthier nutrient profiles

This manuscript has been submitted to Appetite for publication: Emrich TE, Qi Y, Cohen JE,

Lou WY, L’Abbe MR. Front-of-pack symbols are not a reliable indicator of products with

healthier nutrient profiles. Appetite (revision submitted May 21, 2014).

This study addressed objective #2 of my thesis, to:

• compare the calorie, saturated and trans fat, sodium, and sugar content of foods with FOP

symbols to foods without symbols by system type.

Student contribution:

I developed the original idea and design in collaboration with my supervisor Dr. Mary L’Abbe. I

was involved in the development of the FLIP database as part of the L’Abbe lab group, including

data collection at grocery retailers, as well as data entry and validation. I independently

reviewed each food label in FLIP for the presence of FOP systems (using the definitions

described in the manuscript) and classified each FOP system identified by type (with validation

from a second independent reviewer). I designed the analysis with the help of Dr. Wendy Lou

and her graduate student, Ying Qi, and carried out the analyses independently. I completed the

original interpretation of the data and independently prepared a draft of the manuscript prior to

engaging my co-authors in reviewing and revising the manuscript that was eventually submitted

to Appetite.

60

5.1 Abstract

Front-of-pack (FOP) nutrition rating systems and symbols are a form of nutrition marketing used

on food labels worldwide. In the absence of standardized criteria for their use, it is unclear if

FOP symbols are being used to promote products more nutritious than products without symbols.

OBJECTIVES: To compare the amount of calories, saturated fat, sodium, and sugar in products

with FOP symbols, and different FOP symbol types, to products without symbols. METHODS:

The median calorie, saturated fat, sodium, and sugar content per reference amount of products

with FOP symbols were compared to products without FOP symbols using data from the Food

Label Information Program, a database of 10,487 Canadian packaged food labels. Ten food

categories and 60 subcategories were analyzed. Nutrient content differences were compared

using Wilcoxon rank-sum test; differences greater than 25% were deemed nutritionally relevant.

RESULTS: Products with FOP symbols were not uniformly lower in calories, saturated fat,

sodium, and sugar per reference amount than products without these symbols in any food

category and the majority of subcategories (59/60). None of the different FOP types examined

were used to market products with overall better nutritional profiles than products without this

type of marketing. CONCLUSION: FOP symbols are being used to market foods that are no

more nutritious than foods without this type of marketing. Because FOP symbols may influence

consumer perceptions of products and their purchases, it may be a useful public health strategy to

set minimum nutritional standards for products using FOP symbol marketing.

MeSH key words: Food Labeling, Food Quality, Food Analysis

61

5.2 Introduction

Worldwide, chronic diseases account for 60% of deaths, and unhealthy diet is a preventable risk

factor shared by most chronic diseases (1,151). To reduce chronic disease risk, the World Health

Organization recommends that individuals and populations limit their intake of saturated and

trans fat, cholesterol, and simple and added sugars, while achieving energy balance (1,3). To

help consumers choose foods consistent with these recommendations, the World Health

Organization supports the provision of “accurate, standardized and comprehensible information

on the content of food items” on food packages (3). Indeed, in many countries around the world

standardized, voluntary or mandatory nutrition labels are found on the back-of-pack of some, or

all, prepackaged foods (15). For example, Canada has required the use of a mandatory Nutrition

Facts table (NFt) on most prepackaged foods since 2007 (5). Furthermore, voluntary claims that

describe the level of a nutrient in a food or the relationship between a food and health are also

permitted on products meeting prescribed conditions in many countries (119). For example,

Canada’s Food and Drug Regulations allow for the voluntary use of nutrient-content claims such

as “low in fat” and health claims such as “a healthy diet with adequate calcium and vitamin D,

and regular physical activity, help to achieve strong bones and may reduce the risk of

osteoporosis” on food labels. Besides nutrition labels and claims, a variety of front-of-pack

(FOP) nutrition rating systems and symbols have been providing simplified nutrition information

to consumers on the front of food packages since the 1980s (15,19). Standardized, voluntary

FOP systems have been introduced in some countries (16,17); however, multiple FOP systems

with their own unique symbols and underlying criteria can currently be found in most

marketplaces (19,56,106). Within Canada, there are presently no specific regulations governing

the use of FOP symbols, beyond that they may not be “false, misleading, or deceptive” (73).

Voluntary claims and FOP systems provide nutrition information beyond what is required on the

nutrition label in most jurisdictions and can therefore be defined as forms of nutrition marketing

(70).

The US Institute of Medicine has categorized FOP systems into general three types: nutrient-

specific systems, summary indicator systems, and food group information systems (19).

According to the Institute of Medicine, nutrient-specific systems typically either display the

amount of calories and select nutrients per serving (i.e. repeat some of the information required

by nutrition labels on the FOP) or use symbols based on claim criteria (i.e. ‘low in fat’ or ‘high

62

in fibre’). Summary indicator systems provide summary information on the nutrient content of a

food product using a single symbol, icon, or score and are based on nutrient thresholds or

algorithms. Finally, food group information systems use symbols to convey the presence of a

food group or ingredient (see Figure 5-1 for examples of each type of FOP system).

Nutrition marketing has the potential to influence consumer purchases at the grocery store, which

may impact consumption patterns and ultimately chronic disease risk. Consumers perceive

products with summary indicator systems (such as the Heart and Stroke Foundations’ Health

Check™ symbol shown in Figure 1) as more healthful and lower in ‘negative’ nutrients (121-

123). Moreover, it has been found that FOP claims (such as the nutrient-content and health

claims that form the basis of some nutrient-specific systems such as the General Mills’ Goodness

Corner found in Figure 1) exert a ‘halo’ effect whereby consumers tend to generalize the claim

to the entire product, believing the product is healthier with respect to other nutritional and health

elements not identified in the claim (82,83,89). While we are not aware of any study examining

consumers perceptions of products with food group information systems, it is possible that this

“halo” effect may extend to such systems given their similarities with nutrient-specific systems

based on claims criteria. While there is little research available examining the impact of the

different FOP systems on food purchases and consumption (126), 23% of consumers report

looking for better choice slogans, symbols or logos [FOP systems] on food labels (6), and

qualitative research has found that many consumers use FOP nutrition information more often

than back-of-pack nutrition labels (81).

Despite evidence that consumers perceive products with FOP systems as healthier or having

more favourable nutrient contents, it is presently not known if FOP systems are being used to

market products with overall better nutrient compositions. Most countries allow products to

carry claims (like those that form the basis of some nutrient-specific systems) without

considering their overall nutrient composition (119). Furthermore, while summary indicator

systems typically consider multiple nutrients in their criteria, the nutrients included and their

thresholds or algorithms vary from one system to the next (19,106), and food group information

systems typically only consider a single food group or ingredient and not overall nutrient content

(19). Considering the weaknesses in the present voluntary FOP systems, the Institute of

Medicine has suggested that to best promote health, FOP systems need to consider calories and

multiple nutrients, namely saturated fat, trans fat, sodium, and added sugar as these nutrients are

63

of greatest relevance to public health and chronic disease risk (109). However, since FOP

systems only currently consider selected nutrients and food components, they may be used to

market products that are no healthier in their content of these nutrients than products without

such marketing.

In the absence of standardized underlying nutrient criteria, we hypothesize that products

marketed with FOP symbols will provide similar levels of calories, saturated fat, trans fat,

sodium, and sugar to products without symbols. This study aimed to evaluate if FOP symbols

are being used on foods lower in calories, saturated fat, trans fat, sodium, and sugar than foods

without symbols. This study also compared foods with and without different types of FOP

symbols to determine which FOP types were most likely to identify products with more

favourable levels of the nutrients related to health risks.

Figure 5-1: Examples of different front- of

Nutrient Specific Systems

Nutrient-specific systems that display the amount of calories and select nutrients per serving

The Grocery Manufacturers Association and the Food Marketing Institute’s Facts Up Front

Nutrient- specific systems based on claim criteria

General Mills’ Goodness Corner

of-pack symbol types

Nutrient Specific Systems

the amount of calories

The Grocery Manufacturers Association and the Food

specific systems based on claim criteria

Summary Indicator Systems

Heart and Stroke Foundation of Canada’s Health Check™

Kraft’s Sensible Solutions™

64

Heart and Stroke Foundation of

Food Group Information

Systems

Whole Grain Council’s Whole Grain Stamp

65

5.3 Methods

A cross-sectional comparison of the calorie, saturated and trans fat, sodium, and sugar content of

foods with and without FOP symbols was completed using the Food Label Information Program

(FLIP). The FLIP is a Canadian database of food label information developed at the University

of Toronto that has quantified the use of nutrition marketing, such as FOP systems, nutrient

content claims, and health claims, on food labels and collected information on the nutritional

compositions of foods from the NFt (which includes information on calories and 13 core

nutrients such as fat, total carbohydrates, protein, sodium, and sugar) (78). Data were collected

in 2010-2011 and included information on 10,487 unique products. Data were sampled from the

four major Canadian grocery retailers that together accounted for 56% of the grocery sales in

Canada (152). Every product with a NFt available from national and private label brands was

purchased by systematically scanning the grocery store shelves in each aisle and collecting each

unique product with a NFt. Each product was purchased once and only in a single size.

FLIP data were collected for 23 pre-defined food categories and 153 subcategories, as described

in Schedule M of Canada’s Food and Drug Regulations [B.01.001] (26). Schedule M has

established reference serving sizes for each subcategory and is the basis of the criteria for

making nutrition and health claims. Nutrition information for each product was taken from the

NFt, which provides information on the amount of calories and 13 core nutrients in a

manufacturer defined serving size. Information from the NFt was used to derive the amount of

calories and nutrients per Schedule M reference amount in order to standardize serving sizes for

comparison.

Each food label was scanned for FOP systems using the definitions and categories defined by the

Institute of Medicine (19). As previously described by Schermel et al. (78), FOP systems in

FLIP were independently classified by two reviewers, with any disagreements being resolved in

consultation with the research team. This analysis included FOP systems that provided nutrition

information beyond what is required by the NFt (i.e. nutrition marketing): nutrient-specific

systems based on claim criteria, summary indicator systems, and food group information

systems. Nutrient-specific systems that displayed the amount of calories and select nutrients on

the FOP were excluded as they simply repeated NFt required information.

66

5.3.1 Data analysis

This analysis focused on the ten food categories that had the largest number of foods with FOP

systems: 1) bakery products, 2) cereals and other grain products (hereafter referred to as cereals

and grains), 3) combination dishes, 4) dairy products and substitutes (dairy products), 5) fats and

oils, 6) fruits and fruit juices, 7) meat, poultry, their products and substitutes (meat and poultry),

8) snacks, 9) soups, and 10) vegetables. A total of 92 subcategories were found within these ten

food categories. For example, the bakery products subcategory included subcategories such as

‘bread’, ‘bagels, tea biscuits, scones, rolls, buns, croissants, tortillas, soft bread sticks, soft

pretzels and corn bread’, and ‘crackers, hard bread sticks and melba toast’. Data were analyzed

using SAS software (version 9.3, SAS Institute Inc., Cary, NC, 2011). The distributions of all

continuous variables were first assessed and then differences in the amount of calories, saturated

and trans fat, sodium, and sugar per reference amount of foods, with and without FOP symbols,

were compared using Wilcoxon rank-sum test. Statistical significance level was set at p<0.05.

Differences in calorie or nutrient content per reference amount greater than 25% were deemed to

be nutritionally relevant as per the Food and Drug Regulations’ criteria for “reduced” or “lower”

claims (5). Under the Canadian Food and Drug Regulations, in order to carry a claim of being

“reduced” or “lower” (for example “Sodium reduced” or “Lower in saturated fat”), a food must

contain 25% less of the nutrient of interest per reference amount of the food than the reference

amount of a similar reference food. Data are reported as median followed by interquartile range.

Only statistically significant and nutritionally relevant differences are reported in the results and

tables. Additional sub-group analyses were conducted by subcategory and by FOP type

(nutrient-specific system based on claims criteria, summary indicator system, and food group

information system).

5.4 Results

Overall FOP symbols (excluding nutrient-specific systems that only displayed the amount of

select nutrients on the FOP) were found on 17.8% of the products in the FLIP (with some

products having two or more symbols). Nutrient-specific symbols based on claim criteria were

found on 3.4% of the products and, as reported by Schermel et al. (78), 7.5% of the products

carried a summary indicator system, 3.5% a food group information system, and 7.0% a hybrid

system that combined features of two of more of the FOP types. Of the 92 subcategories

67

examined, 60 contained products with FOP symbols. Trans fat was excluded from the results as

no differences >0.1 g were observed in any category.

5.4.1 Differences between products with and without FOP symbols by food category

Products with FOP symbols were not uniformly lower in calories, saturated fat, sodium, and

sugar per reference amount than products without symbols in any food category, and in some

instances were higher in one or more of these nutrients (Table 5-1). Products with FOP symbols

were only significantly lower in calories than products without symbols in the ‘cereals and

grains’ category (209 versus 300 cal), while products with FOP symbols had higher calories than

products without symbols in the ‘soups’ category (120 versus 90 cal) (p<0.05). In contrast,

products with FOP symbols were lower in saturated fat than those without in five categories

(‘bakery products’, ‘combination dishes’, ‘dairy products’, ‘meat and poultry’, and ‘snacks’)

with differences ranging from 0.8-2.3 g (p<0.05). ‘Meat and poultry’ products with FOP

symbols were lower in sodium than products without symbols (356 versus 522 mg) (p<0.05). In

half of the categories (‘cereals and grains, ‘combination dishes’, ‘dairy products’, ‘soups’, and

‘vegetables’), products with FOP symbols were higher in sugar than products without, with

differences as great as 6 g of sugar per reference amount (p<0.05). ‘Cereals and grains’ with

FOP symbols were significantly higher in saturated fat, sodium, and sugar than ‘cereals and

grains’ without symbols.

5.4.2 Differences between products with and without FOP symbols by food subcategory

No differences in calorie, saturated fat, sodium, and sugar content were observed between

products with and without FOP symbols in half of the subcategories examined (30/60) (see

Appendix C). Products with FOP symbols were higher in at least one nutritional component

(calories, saturated fat, sodium, or sugar) of public health relevance than products without

symbols in nine subcategories. For example ‘cookies and graham wafers’ with symbols had

25% more sodium than products without symbols (p<0.05). Products with FOP symbols were

found to be lower than products without symbols in one or two nutrients for 17 subcategories

(p<0.05), saturated fat and sodium were the nutrients most often lower. For example, ‘canned

68

meat and poultry’ with FOP symbol marketing had 0.4 g of saturated fat and 241 mg of sodium

compared to 1.6 g and 413 mg in products without symbols (p<0.5).

Products with FOP symbols were lower in three of the nutritional components analyzed in six

subcategories. For example, ‘coffee cakes, donuts, Danishes, sweet rolls, sweet quick-type

breads and muffins’ with FOP symbols had less calories (133 versus 217 cal), saturated fat (0.8

versus 2.2 g), and sugar (10 versus 17 g) than products without symbols (p<0.05) (Appendix C).

Only ‘meat and poultry with sauce’ with FOP symbol marketing was lower in all four nutritional

components compared to products without symbols: calories (134 versus 266 cal), saturated fat

(0.5 versus 4.2 g), sodium (442 versus 782 mg), and sugar (1 versus 4 g) (p<0.05).

5.4.3 Comparison of different FOP symbol types

None of the different FOP types examined was used to market products with overall better

nutritional profiles than products without this type of marketing (Figure 5-2). Although there

were cases of individual nutrient levels being improved, overall products with and without FOP

symbols had similar nutritional compositions, regardless of symbol type. The largest numbers of

significant differences were observed between products with and without hybrid or summary

indicator symbols (p<0.05). However, no food category with either of these FOP types was

lower in all four nutritional components than products without symbols. For example, while

‘soups’ with summary indicator symbols contained less calories, saturated fat, and sodium than

‘soups’ without symbols (by 74 cal, 1.5 g, and 214 mg, respectively), they were higher in sugar

by 14 g (p<0.05).

69

Table 5-1: Comparison of the nutrient content per r eference amount of products with and without front- of-pack

nutrition rating symbols *

Food Category FOP status

N Calories (kcal) Saturated fat (g) Sodium (mg) Sugar (g)

Bakery Products No 1254 144 [120, 183] 1.2 [0.3, 3.4] 169 [91, 250] 6 [1, 13]

Yes 382 126 [91, 140] 0.5 [0.3, 1.3]† 148 [92, 205] 2 [1, 8]†

Cereals and Other Grain Products

No 548 300 [164, 302] 0.2 [0.0, 0.3]† 1 [0, 57]† 2 [1, 4]†

Yes 229 209 [156, 300]† 0.3 [0.2, 0.5] 109 [2, 200] 6 [3,11]

Combination dishes No 852 300 [230, 340] 3.0 [1.3, 4.7] 709 [509, 880] 4 [2, 7]†

Yes 192 261 [212, 295] 1.5 [0.7, 2.5]† 561 [468, 660] 5 [3, 9]

Dairy Products and Substitutes

No 741 109 [81, 130] 3.5 [1.5, 6.0] 160 [100, 220] 1 [0, 9]†

Yes 98 110 [84, 132] 1.5 [0.4, 3.2]† 126 [100, 210] 7 [1, 14]

Fats and Oils No 395 80 [70, 100] 1.0 [0.8, 2.0] 120 [0, 300] 0 [0, 2]

Yes 81 70 [35, 80] 1.0 [0.3, 1.0] 135 [70, 260] 0 [0, 2]

Fruits and Fruit Juices No 611 120 [100, 130] 0.0 [0.0, 0.0] 10 [0, 25]† 25 [21, 29]

Yes 189 120 [86, 130] 0.0 [0.0, 0.0] 20 [6, 35] 25 [19, 28]

Meat, Poultry, Their Products and Substitutes

No 557 147 [90, 223] 2.9 [1.1, 4.9] 522 [425, 640] 1 [0, 1]

Yes 86 127 [91, 154] 0.6 [0.4, 1.5]† 356 [256, 437]† 1 [0, 1]

Snacks No 384 260 [230, 270] 1.5 [1.0, 3.0] 288 [151, 390] 2 [0, 3]

Yes 87 214 [179, 240] 0.7 [0.0, 1.3]† 233 [74, 360] 2 [0, 4]

Soups No 244 90 [60, 150]† 0.5 [0.0, 1.5] 740 [650, 898] 2 [1, 4] †

Yes 90 120 [90, 150] 0.5 [0.2, 1.0] 625 [480, 650] 4 [2, 7]

70

Vegetables No 489 25 [16, 40] 0.0 [0.0, 0.0] 180 [34, 290] 2 [0, 4] †

Yes 134 30 [20, 50] 0.0 [0.0, 0.0] 71 [15, 290]† 3 [1, 5]

*Front-of-pack nutrition rating symbol nutrition marketing includes: 1) nutrient-specific symbols based on claim criteria, 2) summary

indicator symbols, 3) food group information symbols, and, 4) hybrid symbols; and, excludes nutrient-specific systems that display the

amount of calories and select nutrients per serving. All data are presented as Median and Interquartile range [Q1, Q3]. Calorie and

nutrient amounts are expressed per reference amount and rounded to the number of decimal places provided in the Nutrition Facts table.

Reference amounts are reference serving size amounts found in Schedule M of Canada’s Food and Drug Regulations and are the basis of

the criteria for making nutrient content and health claims in Canada.

† Statistically significant (p<0.05) and nutritionally relevant (≥25%) difference between products with and without a front-of-pack symbol

in the amount of calories or nutrient of interest.

Full details for each subcategory are provided in Appendix C.

71

Figure 5-2: Difference in calorie, saturated fat, sodium, and s ugar content per

reference amount between products with and without front-of-pack symbols by

symbol type

*Statistically significant (p<0.05) and nutritionally relevant (≥25%) difference in calorie or

nutrient content per reference amount between products with and without FOP symbol. Products

without FOP symbols were the reference. Negative percentages indicate instances and amounts

where products with FOP symbols were lower in calories or nutrients than the reference.

Positive percentages indicate instances and amounts where products with FOP symbols were

higher in calories or nutrients than the reference. The dashed line represents a 25% difference in

calorie or nutrient content.

72

5.5 Discussion

The present study found that, while FOP symbols were being used as nutrition marketing on

17.8% of all products, these symbols were not being used to market products that were overall

lower in calories, saturated fat, sodium, and sugar than their counterparts without FOP symbols,

and in some instances were higher in one or more of these nutrients. These findings are

consistent with those of Colby et al (70), who found that nutrition marketing in the form of food

company symbols [FOP systems] were being used to market products high in saturated fat,

sodium, or sugar. Further, from this analysis it is clear that none of the different FOP types are

promoting healthier food choices when multiple nutrients are examined, despite certain FOP

types, such as summary indicator systems, considering multiple nutrients in their criteria. Thus,

when grocery shoppers are comparing similar products, choosing the product with FOP symbol

marketing is no guarantee of a selection consistent with a diet that reduces chronic disease risk.

For example, although consumers may perceive ‘combination dishes’ with summary indicator

symbols as healthier and lower in negative nutrients (121,123), we found that compared to

products without symbols, ‘combination dishes’ with symbols were lower in saturated fat and

sodium, but higher in sugar, and similar in calorie content.

The finding that FOP symbols are not being used to promote products with lower levels of all the

nutritional components of public health concern is worrisome because of the ‘halo’ effect

associated with FOP nutrition information (82,83). Although a nutrient-specific or food group

information symbol may only state a product is ‘low in fat’ or a ‘source of whole grains’,

because of the ‘halo’ effect, consumers may infer that a product is also lower in calories,

saturated fat, sodium, and sugar based on this information when it is not actually the case. For

example, we found that foods carrying food group information symbols based on claims criteria

were, for the most part, equal or higher than foods without these symbols in calories, saturated

fat, sodium, and sugar and thus were not a good guide to selecting products as part of a diet that

reduces the risk of chronic disease. Although FOP symbols have been found to impact

consumers’ perceptions of product healthiness and nutrient content less when used in

conjunction with the NFt (153), FOP nutrition information often leads consumers to truncate

their search for information and skip reading the standardized nutrition label (83). These results

suggest that requiring products to meet comprehensive minimum standards with respect to their

calorie, saturated fat, sodium, and sugar before being able to use a FOP symbol may be a useful

73

public health strategy to ensure that these symbols are being used to market products consistent

with a health promoting diet. Indeed, the Institute of Medicine has proposed that prior to

qualifying for a FOP symbol, products should meet a minimum standard with respect to

saturated and trans fat, sodium, and added sugar content (109).

One of the proposed benefits of FOP systems is that they may stimulate manufacturers to

reformulate existing products in order to qualify to carry a FOP symbol (19). Research has

found that FOP symbols developed by non-profits have led participating manufacturers to

reformulate products, changing their content of one or more nutrients, in order to qualify for a

symbol (112-115). However, we found that overall products with FOP symbols had similar

levels of calories, saturated fat, sodium, and sugar to products without symbols, regardless of

FOP type. This suggests that manufacturers wanting to use FOP symbol marketing in the present

unstandardized marketplace would have no impetus to reformulate their products to lower their

levels of calories and nutrients of public health concern as, in general, the criteria associated with

the different FOP systems are so diverse many products can qualify for one without

reformulation.

5.5.1 Limitations

There are a few limitations to this study, including the somewhat subjective nature of identifying

FOP symbols on food packages in the FLIP. However, every attempt has been made to

minimize any subjectivity by using pre-established definitions and two independent raters to

identify FOP symbols. Another limitation is the comparison of foods at the category level.

Categories contain a variety of foods and comparison of products with and without symbols may

not always be a true comparison of like products. We have attempted to increase the likelihood

of fair comparison by conducting detailed analysis at the subcategory level where similar

products are grouped (See Appendix C). Finally, nutrient content information was derived from

the food label rather than chemical analysis, however a 2011, Canadian study found the values

reported in the NFt to be a reliable indicator of the actual nutrient content as determined through

chemical analysis, at least for saturated and trans fat (150).

74

5.5.2 Conclusions

Although lower levels of either calories or one or more nutrients of public health concern were

identified in products with FOP symbols in some product categories and subcategories, in

general the results of this study demonstrated the calorie, saturated fat, sodium, and sugar content

of products being marketed with FOP symbols are no better than products without FOP

marketing, regardless of the FOP type, in a marketplace without standardized FOP labelling.

This suggests that such symbols are being applied more as a marketing feature in the interest of

selling products than promoting healthier food choices. If nutrition marketing influences

consumer perceptions of product healthiness and nutrient content, and ultimately their product

purchases, the results of this study suggest that minimum standards should be established

regarding the content of nutrients of public health significance for products with FOP symbols.

This will help ensure that consumers relying solely on FOP symbols to guide their selection of

healthier products, as part of a diet that reduces chronic disease risk are not being misled.

Furthermore, the establishment of minimum standards for products to carry FOP symbols would

incentivize manufacturers to reformulate their products to lower the content of nutrients of public

health significance, should they want to use this form of nutrition marketing.

75

Chapter 6

6 Consumer perceptions of the Nutrition Facts table and Front-of-Pack nutrition rating systems

This manuscript has been published: Emrich TE, Qi Y, Mendoza JE, Lou WY, Cohen JE,

L’Abbe MR. Consumer perceptions of the Nutrition Facts table and Front-of-Pack nutrition

rating systems. Appl Physiol Nutr Metab [Internet]. 2014 [cited 2014 Jan 7]; 39(4):417-424.

Available from: http://www.nrcresearchpress.com/doi/abs/10.1139/apnm-2013-

0304?journalCode=apnm#.UsxrK3lZW44

This study addressed objectives #3 and #4 of my thesis, to:

• describe Canadian’s attitudes to the regulation of FOP systems; and,

• determine experimentally which FOP system is most liked and understood by Canadians.

Student contribution:

I developed the original idea and design for this study in collaboration with my supervisor Dr.

Mary L’Abbe. I drafted the original survey used in the study prior to receiving input on its

design from my supervisor, thesis committee, and study collaborators. I completed the pilot test

of the survey in collaboration with Julio Mendoza, who administered the existing Guelph

Consumer Monitor and the Canadian Consumer Monitor (CCM) online panels. I created the

online version of this survey and administered it to the CCM panel with support from Julio

Mendoza. I designed the analysis for this survey in collaboration with Dr. Mary L’Abbe, Dr.

Wendy Lou, and Ying Qi, and carried out the analyses independently. I completed the original

interpretation of the data and independently prepared a draft of the manuscript prior to engaging

my co-authors in reviewing and revising the manuscript that was published in Applied

Physiology, Nutrition, and Metabolism.

76

6.1 Abstract

Preferences for, and consumer friendliness of, front-of-pack (FOP) nutrition rating systems have

not been studied with a Canadian population and studies comparing systems when accompanied

by mandatory labelling, like Canada’s Nutrition Facts table (NFt) are lacking. The purpose of

this study was to evaluate four FOPs relative to the NFt with respect to consumer friendliness

and their influence on perceptions of the healthiness and nutrient content of food. Canadian

consumers (n=3029) participating in an online survey were randomized to score the consumer

friendliness of one of five FOP conditions with or without an NFt and to score the healthiness

and nutrient content of two foods using the provided label(s). Mean differences in scores were

evaluated with ANCOVA controlling for age, gender, and education with Tukey-Kramer

adjustments for multiple comparisons. The NFt received the highest scores of consumer

friendliness with respect to liking, helpfulness, credibility, and influence on purchase decisions

(p<0.05); however consumers still supported the implementation of a single, standardized FOP

system, with the nutrient-specific systems (a traffic light and daily value FOP) being preferred

and scored as more consumer friendly than summary indicator systems. Without the NFt,

consumer ratings of healthiness and calorie and nutrient content differed by FOP. With the NFt

present, consumers rated healthiness, calorie and nutrient content similarly, except those who

saw the traffic light. Their ratings were influenced by the traffic light’s colours. The introduction

of a single, standard nutrient-specific FOP to supplement the mandatory NFt should be

considered by Canadian policy makers.

Keywords: nutrition labelling, front-of-pack nutrition rating systems and symbols, surveys

77

6.2 Introduction Front-of-pack (FOP) nutrition rating systems and symbols provide simplified information to

consumers on the nutritional characteristics of food products (19). Since being introduced in the

US and Sweden in the late 1980s, FOP systems have proliferated, with numerous symbols

developed by governments, expert groups, health organizations, food manufacturers, and

retailers currently in use (15,19). We recently found that 19% of foods in a database of Canadian

packaged food labels carried 1 or more FOP system (78). Four general types were found:

nutrient- specific systems that show the amount per serving of select nutrients or use symbols

based on the criteria used to make nutrient content or health claims on food packages; summary

indicator systems that use a single symbol or score to provide summary information about the

nutritional quality of a product; food group information systems that indicate the presence of a

food group or food ingredient; or hybrid systems that combine characteristics of 2 or more of the

preceding systems (19).

The proliferation of FOP systems has led to concerns that they may be confusing and misleading

to consumers because of their inconsistent appearance, application, and underlying nutrient

criteria, and this has prompted expert groups to call for a single, standardized FOP system for use

on all prepackaged food (13,109). However, consumer attitudes toward the adoption of a single,

standardized FOP system have not been evaluated. Furthermore, only a limited number of

studies have examined consumer preferences for different FOP systems and FOP systems’

overall consumer friendliness with respect to liking, helpfulness, credibility and trustworthiness,

and understanding (126). Most studies have identified a preference for nutrient-specific systems,

specifically Traffic Lights, over summary indicator systems (33,62,63,136,137,139). However,

results vary by country and ethnic group, suggesting a need for country-specific studies

(33,62,63,137). In Canada, where Traffic Lights are not in use and summary indicator systems

are the predominant FOP style (78), the consumer friendliness of different FOP systems has not

been tested. Moreover, although in most jurisdictions (including Canada) FOP systems usually

appear in conjunction with either a mandatory or a voluntary nutrition label on food packages,

the consumer friendliness of FOP systems has been tested only relative to the standard nutrition

label in a single study (33); the consumer friendliness of different FOP systems has not been

tested in situations with or without the standard nutrition label, such as the Nutrition Facts table

(NFt) used in Canada.

78

To use FOP systems to guide the selection of foods, consumers must understand them. This

understanding can be (i) substantial understanding (ability to interpret the label correctly to

identify healthier foods) or (ii) conceptual understanding (general understanding of the concept

behind an FOP system) (134,135). With respect to substantial understanding, several studies

have found that consumers are able to use FOP systems to identify the healthier of 2 products

(126), whereas a more limited number of studies have found that consumers perceived the same

product as healthier when it carried an FOP system than when it did not (122,123). The latter

studies suggest the potentially misleading nature of FOP systems; however, it is unclear whether

this effect occurs across a range of product categories and different FOP system types, such as

nonprofit compared with manufacturer-developed systems, or information-based compared with

interpretive systems. Conceptual understanding of FOP systems has been less studied. A 2010

study of a nutrient-specific system in 6 European countries found low levels of conceptual

understanding, with substantial variation among countries (134); conceptual understanding of

other FOP systems has not been tested.

This research sought to examine many of these research gaps, to:

• Identify consumers’ preferred FOP system for use on packaged foods based on self-

reported preferences and measures of consumer friendliness (including conceptual

understanding);

• Determine how different FOP systems, with or without the NFt, influence consumers’

perceptions of the healthiness of different types of prepackaged foods;

• Determine consumers’ attitudes towards the use and regulation of FOP systems.

6.3 Materials and Methods

6.3.1 Participants The study was conducted together with the Canadian Consumer Monitor (CCM) panel

(http://consumermonitor.ca/). The CCM is a nationally representative panel of Canadian

consumers between the ages of 20 and 69 years who complete on-line surveys related to food

and nutrition every 8 to 10 weeks (89,154). The CCM panel was recruited between March 2010

and January 2011; 31 223 individuals were contacted via email to join the panel, and 6665

completed the baseline questionnaire. The current study was conducted in July and August 2011.

79

Consent forms were sent to each participant’s email when he or she joined the panel, and a link

to the consent form was provided at the start of every survey. Ethics approval was received from

both the University of Toronto and the University of Guelph research ethics boards.

6.3.2 Survey Design and Procedure For the mock package experiment, each CCM participant was assigned randomly to 1 of the 5

test FOP systems (Figure 6-1). The Health Check and Smart Pick symbols were chosen to be

representative of summary indicator systems. The former represented a third-party nonprofit

health foundation symbol; the latter was designed for this study to be representative of

manufacturer-developed symbols (such as Kraft’s Sensible Solutions or Pepsi’s Smart Spot). The

Multiple Traffic Light and the Daily Value system were chosen to be representative of nutrient-

specific systems: the former, an interpretive system; the latter, an information-based system.

Participants were further randomized to view their assigned FOP system with or without an NFt,

for a total of 10 different treatment conditions. The no-NFt condition was designed to simulate

what consumers see on store shelves if they choose not to turn over the package to view the

mandatory NFt. Each participant viewed his or her assigned FOP system–NFt condition on the

mock packages of both a frozen meal and a breakfast cereal. These categories were chosen

because they are believed to be familiar to most participants, minimizing the potentially

confounding effect of participants learning about a new product (83).

For each treatment condition, dependent variables were measured on a 5-point Likert-type scale.

The consumer-friendliness measures of liking, helpfulness, credibility, understanding, and

influence on purchase intention were evaluated for each FOP system. Participants in the no-FOP

system–no-NFt condition were not asked to respond to the consumer friendliness measures,

whereas participants in the no-FOP system–with-NFt condition were asked to respond to the

consumer friendliness measures in relation to the NFt. Additionally, all respondents were asked

to rate their perceptions of the healthiness, as well as the calorie and nutrient content, of both the

frozen meal and the breakfast cereal. The consumer friendliness and perceived healthiness

measures were modelled after those used by Feunekes et al. (63) to test the effectiveness of

different FOP nutrition labelling formats in Europe.

Conceptual understanding of each of the FOP systems was further assessed by asking consumers

to rate, on a 5-point Likert-type scale, a series of statements describing the FOP systems (e.g.,

80

“This symbol is a guide to whether or not the food contains a little or a lot of a certain nutrient”,

“This symbol indicates that the food meets a certain nutritional standard”, etc., with anchors 1

(“does not describe at all”) and 5 (“describes completely”).

Finally, panelists completed additional questions about their preferences regarding the regulation

and use of FOP systems by food manufacturers in Canada. Respondents’ demographic variables

(age, gender, education, province, income) were collected during the recruitment process and the

first survey and were merged with this survey data. The survey was translated, face validated by

nutrition-labelling professionals, reviewed by a plain- language expert, and pilot tested with an

online consumer panel from Guelph, Ontario, prior to being administered to the CCM panel

(155).

6.3.3 Statistical Analyses All data analyses were carried out using SAS version 9.3 (SAS Institute, Cary, N.C., USA). Data

are presented as means (± SD) for continuous variables and count followed by percentage for

categorical variables. Analysis of covariance (ANCOVA) was conducted to compare the mean

rating differences in consumer friendliness, healthiness measures, calorie and nutrient content of

the products, and the importance of including calorie counts and different nutrients in the FOP

systems. The Tukey–Kramer method was used for multiple comparisons adjustment throughout

the study. The mean ratings of statements describing the FOP systems were compared using

ANCOVA with control of variables for age, gender, and education. For all the analyses, p < 0.05

was considered as statistically significant unless stated otherwise.

81

Figure 6-1. Front-of-Pack Nutrition Rating Systems evaluated in the randomized

mock package experiment.

Summary indicator systems:� (A) Health Check, a nonprofit-developed system (The Health Check logo, Health Check word mark, and the Heart and Stroke Foundation word mark are trademarks of the Heart and Stroke Foundation of Canada used under license), and (B) Smart Pick, modelled after a manufacturer- sponsored system. Nutrient-specific systems: (C) The Multiple Traffic Light, an interpretive system, and (D) The Daily Value system, an information-based system. B, C, and D were created by On Brand Design under contract with Dr. Mary L’Abbé at the Department of Nutritional Sciences, University of Toronto. In the fifth FOP condition, participants were exposed to mock packages without an FOP system. Participants were randomized to evaluate an assigned FOP system with or without a Nutrition Facts table, for a total of 10 treatment groups.

A

B C

D

82

6.4 Results

6.4.1 Participants Three thousand twenty-nine respondents (65% women) completed the survey and were included

in the analysis. Demographic characteristics of respondents are shown in Table 6-1. There were

no significant differences in the 10 conditions with respect to gender, age, or education among

the participants. The sample was older and better educated than the general Canadian population.

6.4.2 Consumer friendliness of FOP systems Mean scores of consumer friendliness differed by FOP system and by NFt (with the exception of

understanding, which did not differ by NFt) (Table 6-2). The NFt–no-FOP system label

combination received significantly higher mean scores on liking, helpfulness, credibility, and

influence on purchase decisions than all the FOP system–NFt combinations, except for the Daily

Value system–no-NFt combination, which received similar scores for liking (on both the frozen

meal and the breakfast cereal) as well as helpfulness and credibility (on the frozen meal only).

The 2 nutrient-specific systems (with or without the NFt) received significantly higher scores on

understanding than did the 2 summary indicator systems (with or without the NFt) on both the

frozen meal and the breakfast cereal. In addition, the 2 nutrient- specific systems (with or

without the NFt) received higher scores on liking, helpfulness, credibility, and influence on

purchase decisions than did Health Check (with the NFt) and Smart Pick (with or without the

NFt) (p < 0.05).

6.4.3 Effect of NFt on consumer friendliness of FOPs Pairwise comparison between the frozen meals with or without the NFt showed that, consumers

found the summary indicator systems (Health Check™ and Smart Pick) more helpful, credible,

and likeable when they were shown without the NFt (p<0.05) (Table 6-2A). In addition, the

Daily Value FOP was given significantly higher scores on helpfulness when shown without the

NFt. However, no significant differences in consumer friendliness scores were seen between the

FOP pairs (FOP system with and without NFt) on the breakfast cereal (Table 6-2B).

6.4.4 Conceptual Understanding With respect to conceptual understanding, when presented with a list of possible descriptors of

the NFt or their FOP system, consumers scored the correct descriptions of the concept behind the

83

NFt or specific FOP system the highest (Table 6-3). More than 70% of consumers randomly

assigned to either the NFt or the Daily Value system believed that their respective labels could be

used to compare foods both within (e.g., cereal to cereal) and across (e.g., cereal to crackers)

food categories. Seventy-one percent of consumers in the Traffic Light group believed that it

could help them compare foods within a food category. Fewer than 50% of the consumers in the

Health Check or Smart Pick groups believed these FOP systems could be used to compare foods

within or across food categories.

6.4.5 Effect of FOP on perceived overall healthiness and nutrient content

Mean scores of healthiness and nutrient content differed by NFt and FOP system (Table 6-4).

With respect to the frozen meal, consumers who saw the NFt, regardless of FOP system, gave

similar scores on healthiness and on calorie, fat, saturated fat, trans fat, fibre, and sugar content.

However, consumers who saw the Traffic Light and an NFt scored the frozen meal significantly

lower in sodium than did all the other FOP system–NFt groups.

When there was no NFt, consumers in the Health Check group scored the frozen meal as

significantly healthier than did the 2 nutrient-specific system groups and the no-FOP system

group. In addition, the Health Check group scored the meal lower in saturated and trans fat than

did all the other FOP system groups, but higher in sugar than did the nutrient-specific system and

no-FOP system groups. Consumers in the Health Check and Smart Pick groups without the NFt

scored the frozen meal as healthier, lower in saturated fat and sodium, and higher in fibre, but

also higher in calories and sugar, when compared with the same FOP system with the NFt. In

addition, consumers in the Health Check–no-NFt group rated the frozen meal as lower in fat and

trans fat than did the Health Check–NFt group.

With respect to the breakfast cereals, consumers who saw the NFt with no FOP system, Health

Check, Smart Pick, or Daily Value system scored the cereal similarly with respect to healthiness

and calorie, fat, saturated fat, trans fat, sodium, fibre, and sugar content (Table 6-4). However,

consumers who saw the Traffic Light scored the cereal lower with respect to healthiness and

higher with respect to fat and sugar content than did all other FOP system–NFt groups (p < 0.05).

The Traffic Light group also scored the cereal higher in sodium than did consumers who saw no

FOP system or a summary indicator system (p < 0.05). On the other hand, when the NFt was

84

absent, a number of differences in product ratings among the different FOP systems were seen.

Consumers with the Daily Value FOP system perceived the cereal as significantly healthier than

did consumers who saw the same cereal with no FOP system or Traffic Light. In addition,

consumers who saw the Daily Value FOP system scored the cereal as lower in fat and sugar than

did all other FOP system groups, and lower in sodium than did all other FOP system groups

except for the Health Check group. Consumers scored the cereal as lower in calories and

saturated fat when they saw a nutrient-specific system than did all other FOP system groups (p <

0.05). Finally, consumers who saw either 1 of the summary indicator systems or no FOP system,

without the NFt, scored the cereal as significantly higher in calories and all 7 nutrients than did

consumers exposed to the same FOP system with the NFt. Meanwhile, consumers who saw the

Traffic Light scored the cereal lower in trans fat, sodium, and sugar when the NFt was present

than when it was absent (p < 0.05).

6.4.6 Consumer preferences for FOP systems The majority of participants (86%) believed that there should be either a single mandatory or a

single voluntary FOP system used by all manufacturers, and most (68%) believed that a single

FOP system should be used on all food packages, not just on healthy products or on products

chosen at the manufacturer’s discretion. Participants were split as to who should be responsible

for overseeing the use of FOP systems, with 44% and 35% supporting oversight by the

government and the nonprofit sector, respectively; only 14% favoured oversight by

manufacturers or industry associations. With respect to the importance of including calories and

11 different nutrients in the FOP system, participants rated, on average, all nutrients as at least

somewhat important (mean rating >3.8 out of 5). Sodium, trans fat, saturated fat, and sugar

received the highest mean ratings of importance for inclusion in an FOP system. When asked

directly which FOP system they preferred (Health Check, Smart Pick, Traffic Light, or Daily

Value FOP system), 53% preferred the Traffic Light and 30% preferred the Daily Value FOP

system.

85

Table 6-1. Demographic characteristics of survey re spondents (n=3029)

Characteristics CCM

(N) (%)

Gender

Male

Female

1049

1980

34.6

65.4

Age

20-29

30-39

40-49

50-59

60-69

208

491

756

889

685

6.9

16.2

25.0

29.3

22.6

Education

< high school

High school

Trade school

College

University

32

537

276

1039

1145

1.1

17.7

9.1

34.3

37.7

Province

Alberta

British Columbia

Manitoba

New Brunswick

Newfoundland

Nova Scotia

Ontario

Prince Edward Island

Quebec

Saskatchewan

332

448

189

75

58

167

1078

66

404

172

11.0

16.1

6.2

2.5

1.9

5.5

35.6

2.2

13.3

5.7

Income

< $25,000

$25,000-49,999

182

438

6.0

14.5

86

Characteristics CCM

(N) (%)

$50,000-74,999

$75,000-99,999

$100,000-124,999

$125,000-149,999

$150,000-174,999

$175,000-199,999

$200,000+

Not provided

460

411

317

177

145

83

115

701

15.3

13.6

10.5

5.8

4.8

2.7

3.8

23.1

Note: CCM, Canadian Consumer Monitor.

87

Table 6-2. Mean scores of consumer friendliness by FOP system and NFt* (n=3029)

Part A: Frozen meals

Variables FOP System ANCOVA†

No FOP system‡ (n=600)

Summary Indicator Systems Nutrient-specific systems

Health Check™ (n =

642)

Smart Pick (n = 572)

Traffic light (n = 588)

Daily Value (n = 627)

FOP system

NFt FOP system X NFt

Liking§

No NFt

NFt

N/A

4.1a

3.2de

2.7fg

3.0ef

2.5g

3.5c

3.4cd

3.9ab

3.7bc

<.0001 <.0001 0.0535

Helpfulness||

No NFt

NFt

N/A

4.3a

3.2d

2.8e

2.7e

2.4f

3.6bc

3.3cd

4.1a

3.8b

<.0001 <.0001 0.9561

Credibility¶

No NFt

NFt

N/A

4.0a

3.1c

2.7d

2.4d

2.1e

3.2c

3.0c

3.7ab

3.6b

<.0001 <.0001 0.1086

Understanding**

No NFt

NFt

N/A

4.1bc

3.9cd

3.8de

3.6e

3.6de

4.4a

4.3ab

4.3ab

4.2ab

<.0001 0.0537 0.5808

Influence on purchase decisions††

No NFt

NFt

N/A

4.2a

2.7ef

2.4fg

2.3gh

2.1h

3.1cd

3.0de

3.7b

3.4bc

<.0001

0.0001

0.5327

88

Part B: Breakfast cereal

Variables FOP System ANCOVA†

No FOP system‡ (n=600)

Summary Indicator Systems Nutrient-specific systems

Health Check™ (n =

642)

Smart Pick (n = 572)

Traffic light (n = 588)

Daily Value (n = 627)

FOP system

NFt FOP system X NFt

Liking§

No NFt

NFt

N/A

4.1a

3.1d

3.0de

2.9de

2.7e

3.5c

3.4c

3.9ab

3.7bc

<.0001 0.0017 0.6880

Helpfulness||

No NFt

NFt

N/A

4.3a

3.0d

2.9de

2.6ef

2.5f

3.6c

3.5c

4.0b

3.8bc

<.0001 0.0020 0.5117

Credibility¶

No NFt

NFt

N/A

4.1a

3.1cd

2.9d

2.5e

2.4e

3.3c

3.2cd

3.8b

3.7b

<.0001 0.0182 0.6566

Understanding**

No NFt

NFt

N/A

4.2a

3.8b

3.8b

3.6b

3.6b

4.4a

4.4a

4.3a

4.3a

<.0001 0.5831 0.7284

Influence on purchase decisions††

No NFt

NFt

N/A

4.2a

2.6d

2.5de

2.4de

2.2e

3.2c

3.2c

3.6b

3.5bc

<.0001

0.0127

0.8258

Note: FOP, front-of-pack; NFt, Nutrition Facts table; FOP×NFt, the significance of the interaction between FOP and NFt on consumers ratings of consumer friendliness on the 5-point Likert scale; ANCOVA, analysis of covariance; NA, not applicable.

89

*Values that do not share lowercase letters within each variable are significantly different from each other using LSMeans multiple comparison with Tukey−Kramer adjustments (p < 0.05).

†Models controlled for gender, age, and education.

‡Respondents randomized to view no FOP system and no NFt were not asked to complete consumer friendliness measures; respondents randomized to view no FOP system with an NFt completed consumer friendliness measures with reference to the NFt.

§Question: “How much do you like the [FOP system] on this food?” Anchors: 1 = “do not like at all”, 5 = “like a lot”. Question: “How helpful is the [FOP system] in helping you choose a healthier food?” Anchors: 1 = “not very helpful”, 5 = “extremely helpful”. ¶Question: “How credible is the [FOP system] to you?” Anchors: 1 = “not at all credible”, 5 = “extremely credible”. **Question: “How difficult is it for you to understand the [FOP system]?” Anchors: 1 = “very difficult”, 5 = “very easy”. ††Question: “Would the [FOP system] influence your decision to buy this food?” Anchors: 1 = “not at all”, 5 = “extremely”.

90

Table 6-3. Mean scores of statements describing the NFt and FOP systems

FOP system Statements rated most descriptive of each FOP system Mean Score*

Nutrition Facts table

The Nutrition Facts table is a guide to whether or not the food contains a little or a lot of a certain nutrient.

The Nutrition Facts table is a guide to the amount of different nutrients in a food.

The Nutrition Facts table is a guide to the amount of nutrients a person should eat in a day.

4.0

3.8

2.9

Health Check™ This symbol indicates that the food meets a certain nutritional standard.

This symbol indicates that the food is healthy or good for me.

This symbol indicates that the food is more nutritious than other similar foods.

3.5

3.1

2.9

Smart Pick This symbol indicates that the food meets a certain nutritional standard.

This symbol indicates that the food is healthy or good for me.

This symbol indicates that the food is more nutritious than other similar foods.

2.8

2.7

2.7

Traffic Light This symbol is a guide to whether or not the food contains a little or a lot of a certain nutrient.

This symbol is a guide to the amount of different nutrients in a food.

This symbol indicates that the food is healthy or good for me.

3.5

3.0

2.7

Daily Value This symbol is a guide to whether or not the food contains a little or a lot of a certain nutrient.

This symbol is a guide to the amount of different nutrients in a food.

This symbol indicates that the food is healthy or good for me.

3.6

3.1

2.8

Note: Bolded statements are the correct descriptions of the FOP system of interest. FOP, front-of-pack.

*Anchors: 1 = “Does not describe at all”; 5 = “Describes completely”.

91

Table 6-4. Mean healthiness, calorie and nutrient c ontent scores of a frozen meal

and breakfast cereal by NFt and FOP system *

Characteristics and FOPS

Frozen Meal Breakfast Cereal

No NFt Mean

NFt Mean

P† (NFt*FOP)

No NFt Mean

NFt Mean

P† (NFt*FOP)

Healthiness‡

No FOP system

Health Check™

Smart Pick

Traffic Light

Daily Value

2.5cd

3.2a

3.0ab

2.7bc

2.4cd

2.3d

2.4cd

2.4cd

2.5cd

2.3d

<.0001

3.4cd

3.4bcd

3.6ac

3.2d

3.7ab

3.8a

3.9a

3.8a

3.4cd

3.7ab

0.0028

Calories§

No FOP system

Health Check™

Smart Pick

Traffic Light

Daily Value

4.0a

3.6b

3.6b

2.8d

2.9c

3.1c

3.0cd

2.9cd

2.9cd

3.1c

<.0001

3.4a

3.3a

3.3a

2.2b

2.1b

2.2b

2.1b

2.1b

2.3b

2.2b

0.0028

Fat§

No FOP system

Health Check™

Smart Pick

Traffic Light

Daily Value

3.7a

2.9e

3.2bcd

3.0d

3.0c

3.4ab

3.2bcd

3.3bc

3.2bcd

3.3bcd

<.0001

2.5a

2.2bc

2.4ab

2.2bc

1.7d

1.7d

1.7d

1.6d

2.1c

1.8d

<0.0001

Saturated fat§

No FOP system

Health Check™

Smart Pick

Traffic Light

Daily Value

3.5c

2.4e

2.8d

3.1d

3.7abc

3.9a

3.7abc

3.9ab

3.6bc

3.8abc

<.0001

2.2a

2.0a

2.2a

1.4b

1.3bc

1.2bc

1.2c

1.2c

1.3bc

1.3bc

<0.0001

Trans fat§

No FOP system

Health Check™

3.2a

2.2c

2.9ab

2.7b

0.0001

2.0ab

1.8bc

1.2d

1.2d

<0.0001

92

Smart Pick

Traffic Light

Daily Value

2.7ab

2.7abc

3.0ab

2.9ab

2.7b

2.9ab

2.1a

1.8abc

1.7c

1.2d

1.3d

1.3d

Sodium§

No FOP system

Health Check™

Smart Pick

Traffic Light

Daily Value

4.3a

3.2c

3.7b

3.3c

4.4a

4.4a

4.3a

4.4a

3.8b

4.3a

<.0001

3.0a

2.6bc

2.8ab

2.9ab

2.5c

2.1de

2.0e

2.1de

2.5c

2.3cd

<0.0001

Fibre§

No FOP system

Health Check™

Smart Pick

Traffic Light

Daily Value

2.2b

2.6a

2.6a

2.1b

2.0b

2.2b

2.2b

2.2b

2.1b

2.1b

<.0001

3.8ab

3.8ab

3.9a

3.5ab

3.4b

2.4c

2.3c

2.4c

2.4c

2.4c

0.0036

Sugar§

No FOP system

Health Check™

Smart Pick

Traffic Light

Daily Value

3.1a

2.8ab

2.7b

1.4c

1.2c

1.4c

1.5c

1.4c

1.4c

1.4c

<.0001

3.9a

3.6ab

3.6ab

3.7a

2.7c

2.7c

2.7c

2.7c

3.4b

2.8c

<0.0001

Note: NFt, Nutrition Facts table; FOP, front-of-pack; FOP×NFt, the significance of the interaction between NFt and FOP on consumer ratings of product healthiness and nutrient content.

*Values that do not share letters are significantly different from each other using least-squared means multiple comparison with Tukey−Kramer adjustment (p < 0.05).

†p value for interaction of FOP system and NFt; model controlled gender, age, and education.�‡Question: “How healthy is this food for you?” Anchors: 1 = “not at all healthy”, 5 = “very healthy”.�§Question: “Please indicate if this food has a little or a lot of the following nutrient” Anchors: 1 = “a little”, 5 = “a lot”.

93

6.5 Discussion Overall, we found that the NFt was more consumer friendly than were FOP systems with respect

to liking, helpfulness, credibility, and influence on purchase decisions. This is an encouraging

finding given the widespread use of standardized nutrition labels such as the NFt on food

packages in Canada and internationally (15). However, our findings are in contrast to those of

the only other study to compare a mandatory nutrition label with different FOP systems; Gorton

et al. (33) found that more consumers “liked” or “really liked” Traffic Lights than they did a

mandatory nutrition label. The high ratings of the NFt along the dimensions of consumer

friendliness studied here are perhaps a result of the widespread use of this mandatory label by

Canadian consumers; 71% of Canadians report looking for the NFt when reading food labels,

whereas only 23% report looking for “healthy–better choice” symbols or logos, such as FOP

systems (6). We also found a consumer preference for nutrient-specific over summary indicia-

tor systems, consistent with the findings of other studies (33,63,136). With respect to consumer

friendliness measures, in the majority of FOP system–NFt comparisons, the 2 nutrient-specific

systems (the Daily Value and the Traffic Lights) were scored higher than were the 2 summary

indicator systems (Health Check and Smart Pick) with respect to liking, helpfulness,

understanding, and influence on purchase decisions, and the Daily Value system was scored as

significantly more credible than all other FOP systems.

The widespread use of the NFt by consumers in Canada may also explain the higher levels of

conceptual understanding observed in relation to the NFt. Consumers gave higher scores to the

correct descriptions of the NFt (≥3.8 out of 5) than to the correct descriptions of the FOP

systems, indicating their greater understanding of the NFt. Scores for the correct descriptions of

nutrient-specific FOP systems were similarly high, perhaps because of their greater similarity to

the NFt than to the summary indicator systems.

However, our findings relating to consumer friendliness and conceptual understanding should be

considered in relation to the composition of our sample, which was predominantly female and

better educated than the average consumer. This is particularly relevant given previous research

that has shown that nutrition labels are more likely to be used by women than by men, and by

college-educated consumers than by those with a high school education (156). Furthermore, a

2011 systematic review identified greater understanding of nutrition labels among women and

94

consumers with higher levels of education, literacy, and numeracy; a lack of numeracy skills

appeared particularly problematic in the understanding and use of nutrition labels (157). Thus,

despite our finding that the NFt was superior to the FOP systems along several dimensions of

consumer friendliness and conceptual understanding, other research would suggest that certain

segments of the population may benefit from a more simplified presentation of nutrition

information on food packages in the form of FOP systems. Moreover, CCM panelists strongly

favoured the implementation of a single, standardized FOP system, either by government or a

nonprofit group, despite the higher scores of consumer friendliness and conceptual understanding

attributed to the NFt.

When used in conjunction with the mandatory NFt, FOP systems, with the exception of the

Traffic Lights, had a limited influence on consumers’ perceptions of the healthiness and calorie

and nutrient content of the breakfast cereal and frozen meal used in this study. Compared with

other FOP system groups that saw the NFt, consumers who saw the frozen meal with a “green”

sodium Traffic Light perceived the meal as lower in sodium, whereas consumers who saw the

breakfast cereal with “amber” fat and sodium and “red” sugar Traffic Lights perceived the meal

as higher in fat, sodium, and sugar and as less healthy overall. Thus, it appears that consumers

can correctly interpret the meaning of the Traffic Light colours and incorporate them into their

evaluations of the product.

We found that, as in previous studies, FOP systems influenced consumer perceptions of product

healthiness and nutrient content in the absence of a mandatory nutrition label (122,123).

Andrews et al. (123) found that a frozen dinner displaying a summary indicator system was

rated as healthier overall and lower in negative nutrients than was the same frozen dinner

displaying a Traffic Light nutrient-specific system, or a no-FOP system control. Similarly, we

found that, when the NFt was absent, a frozen dinner with a Health Check summary indicator

system was scored healthier overall and lower in saturated and trans fat than was the same frozen

dinner with no FOP system or a nutrient-specific FOP system. However, a different pattern

emerged when the breakfast cereal was shown without the NFt. Consumers exposed to the

nutrient-specific systems scored the cereal as lower in negative nutrients, demonstrating that the

influence of FOP systems differs by product type.

Finally, with respect to many nutrients, consumers who saw no FOP system or a summary

95

indicator system, without the NFt, scored the frozen meal more favourably, but the breakfast

cereal less favourably, than did consumers who saw the same FOP conditions with the NFt. This

finding is of particular concern because past research has found that the presence of nutrition

information on the FOP label may lead consumers to truncate their search for nutrition

information (i.e., not look at the NFt) and rely exclusively on the claims made on the FOP label

to evaluate products (83).

6.5.1 Conclusion In summary, consumers are in favour of a single, standardized FOP symbol for use on packaged

foods, despite the high levels of consumer friendliness associated with the existing mandatory

nutrition label. Consumers would prefer an FOP system that builds upon the well-liked, helpful,

credible, and influential NFt, providing nutrient-specific information in the form of a Traffic

Light or Daily Value system; however, prior to widespread implementation, such a system will

need to be evaluated as suitable for consumers with lower health-literacy skills. Our findings

suggest that consumers support the recommendation of the IOM that a single, standardized FOP

system appear on all products, and that this should therefore be considered by Canadian policy

makers.

96

Chapter 7

7 Discussion

7.1 General Discussion

Chronic non-communicable diseases are the leading cause of death in Canada and internationally

(23,24). Reducing the burden of chronic disease includes promoting healthy diets as a central

element (1,3,23). Proposed interventions to promote healthy diets have included food labelling

and restrictions on the marketing of unhealthy foods (25). FOP nutrition rating systems and

symbols exist at the intersection of food labelling and marketing and have been identified as a

potential intervention to prevent or reduce obesity and other diet-related chronic diseases (109),

by:

• Increasing the proportion of consumers who use nutrition information to make nutritious

choices for themselves and their families; and,

• Encouraging industry to reformulate their products.

Myriad FOP systems are currently found on packaged food labels in Canada (78). Despite their

pervasiveness within the country, Canada has no specific regulations governing the use of FOP

systems. Further, the present policy of the Canadian Government, not to pursue a single,

standardized FOP system (133), is at odds with the recommendations of experts (13,56,109), as

well as the policies being pursued in a number of other developed nations (15-17).

In 2010, at the outset of this thesis research, Canadian specific data to support the evolution of

Canadian FOP system policies was lacking. As a result, this thesis research explored FOP

systems within the Canadian context with the overarching objective of generating data related to

present and proposed FOP system policies to support the progression of food labelling

interventions in Canada. Three studies were conducted that addressed concerns related to

unregulated, multiple FOP system environments, and proposed single, standardized FOP

systems.

97

7.1.1 Evaluation of existing FOP system policy

The Canadian Government has described FOP symbols as a great source of additional

information for consumers on the foods they purchase (133). Previous investigations have

concluded that the additional information provided by FOP symbols, and summary indicator

symbols in particular, leads consumers to conclude that products with symbols are healthier than

those without (120-124). In Chapter 6, we tested the impact of four FOP systems on perceptions

of the overall healthiness and nutrient content of two foods against a no FOP system control

(153). We found further evidence that the presence of FOP symbols positively influenced

perceptions of healthiness and nutrient content, although the effect varied by FOP system type

and food. For a frozen meal without an NFt, compared to the no FOP system condition,

consumers perceived the food as significantly: healthier, lower in calories, fat, saturated fat, and

sodium, and higher in fibre in the summary indicator conditions; lower in calories, as well as

total fat, saturated fat, sodium, and sugar (nutrients to which the traffic light attributed amber or

green ratings) in the traffic light condition; lower in calories, fat, and sugar in the daily value

condition. For a breakfast cereal without an NFt, compared to the no FOP system condition,

consumers perceived the food as significantly: lower in sodium in the Health Check™ summary

indicator condition; lower in calories, as well as fat and saturated fat (nutrients to which the

traffic light attributed amber and green ratings, respectively) in the traffic light condition;

healthier, lower in calories, fat, saturated fat, trans fat, sodium, and sugar in the daily value

condition. Taken together, consumers clearly attribute more healthy characteristics to foods with

FOP symbols than to those without. These data provide evidence that FOP systems (including

summary indicator systems and daily value systems common in the Canadian marketplace)

contravene current Canadian guidance on the use of third-party FOP symbols described in

section 1.2.2.2 – FOP systems should not “give the impression that a single food or brand is

“healthier” than, or nutritionally superior to, other foods not bearing the … symbol … ” (26).

Prior our study described in Chapter 6, no studies examined the moderating effect of a

mandatory nutrition label, such as the NFt, on consumers’ perceptions of a foods’ healthiness

and nutrient content in the presence of a FOP system. However, Andrews et al. (123) did find

that that the availability of the US Nutrition Facts panel led to more accurate use of nutrition

information by consumers shown a frozen meal with or without a summary indicator system.

Consistent with this, we found that, when presented with an NFt, consumers rated the overall

98

healthiness and nutrient content of a food with a summary indicator system or daily value system

similarly to when no FOP system was present (153). These data suggest that the potentially

misleading impressions made by FOP systems may be attenuated if they are used in conjunction

with a mandatory nutrition label, such as the NFt. However, previous research has found that

FOP nutrition marketing information leads consumers to limit their search for nutrition

information to the FOP such that back-of-pack mandatory nutrition labels may go unused (83).

Therefore, the attenuating benefits of the NFt may be lost if the consumer does not turn the

package over or look for the NFt. Further, our results suggest that traffic light colours may in

fact override the moderating effect of a mandatory nutrition label (153). When the NFt was

given, consumers in the traffic light condition perceived 1) the frozen meal as lower in saturated

fat and sodium (nutrients attributed yellow and green ratings, respectively), and 2) the breakfast

cereal as less healthy, as well as higher in fat, sodium, and sugar (nutrients attributed yellow,

yellow, and red ratings, respectively) than consumers in the no FOP system condition.

In our investigations in Chapters 4 and 5, we used novel approaches to assess the value of FOP

systems in helping consumers make nutritious choices. Although the presence of a FOP symbol

gives consumers the impression that a food is healthier or nutritionally superior to foods without

symbols, prior to our studies there were no investigations that examined whether foods with FOP

symbols were in fact healthier and nutritionally superior to foods without symbols. In Chapter 4,

we compared the proportion of packaged food products that meet the nutrient profiling criteria of

two of Canada’s most common summary indicator systems (Health Check™ and Sensible

Solutions™) to the proportion of packaged food products that actually display these systems’

symbols on their food labels (158). We found that a large proportion of products (44.8-79.1%)

met the nutrient profiling criteria for one of these systems or the other, whereas these systems’

symbols were only found on the food labels of a minority of products (4.1-7.5%). This was the

first study to examine the proportion of packaged food products that meet the nutrient profiling

criteria a FOP system to the proportion of packaged food products that actually display its

symbol. The results of this study show that the use of such FOP systems may in fact mislead

consumers using such symbols to think that they are making healthier choices.

In Chapter 5, we compared the calorie and nutrient content (fat, saturated fat, trans fat, sodium,

and sugar) of foods with an FOP symbol on their label to foods without a symbol. We found that

foods with FOP symbols were not consistently lower in calories, fat, saturated fat, trans fat,

99

sodium, and sugar than foods without symbols. We interpreted the data from Chapters 4 and 5

as evidence that consumers cannot rely on the presence of a FOP symbol as an indicator of foods

that meet higher nutritional standards, despite the impression these symbols may give off.

Furthermore, these data suggest that FOP symbols may be in violation of subsection 5.1 of the

Food and Drugs Act that prohibits the labelling of foods in a way that is misleading or deceptive.

Experts have suggested that the potential for FOP systems to stimulate product reformulation

will contribute to their efficacy in preventing and reducing obesity and diet-related chronic

disease (109). Indeed numerous studies have found a positive effect of third-party summary

indicator systems on stimulating product reformulation among participating manufacturers (112-

115), although studies of third-party nutrient-specific systems have failed to demonstrate a

similar effect (116). However, a necessary pre-requisite for a FOP system to stimulate

reformulation is that the majority, or large proportions, of existing products wishing to carry the

system’s symbol do not already meet the system’s nutrient profiling criteria. The investigation

in Chapter 4 found that, depending on the product sub-category, up to 100% of products already

meet the nutrient profiling criteria of Canada’s only third-party summary indicator system (158)

– leaving little or no impetus for reformulation.

Further, in Canada’s present regulatory environment for FOP systems, manufacturer’s wishing to

market their products with a FOP symbol have the option of creating their own proprietary FOP

system, with their own nutrient profiling criteria, if their products do not qualify for an existing

third-party system, or if they do not wish to participate in a third-party system. This may

account for our Chapter 5 finding that products with FOP symbols have similar levels of calories,

total fat, saturated fat, trans fat, sodium, and sugar to products without symbols. This finding

held true for summary indicator systems, despite the fact that these systems typically consider

multiple nutrients in their nutrient profiling criteria.

In summary, our data provide the first empirical evidence that FOP systems in use in Canada

may be misleading and deceptive, in direct violation of existing guidelines and acts governing

their use. Moreover, Canada’s present regulatory environment for FOP systems does not appear

to support FOP systems’ dual purposes of 1) helping consumers make nutritious choices, and, 2)

encouraging industry to reformulate.

100

7.1.2 Evaluation of potential FOP system policy

In 2011, Canada’s Minister of Health asserted that Canadians have all “the tools they need to

make healthy food choices when they shop for groceries” by way of justifying the federal

government’s position not to pursue a single, standardized FOP system (133). In Chapter 6, we

surveyed Canadian consumers to gauge their support for the implementation of a single,

standardized FOP system (153). Overwhelming support (86%) was found for the adoption of

either a mandatory or voluntary single, standardized FOP system. Furthermore, most consumers

wanted the system to be universally applied to all food products. We interpret this as evidence

that, contrary to the position of the government, Canadians do not believe they have all the tools

they need to make healthy food choices. Indeed, Canadian consumers expressed a desire for a

“standardized system to rate the overall nutritional quality of a food product” in a recent

consultation conducted by Health Canada to support the modernization and improvement of

food labels (159).

Our third investigation, reported in Chapter 6, tested consumer liking and conceptual

understanding of four FOP systems to order to identify Canadians’ favoured FOP system for use

on packaged foods (153). Our survey, conducted in 2011, was the first to explore different FOP

systems with Canadian consumers. Consistent with international research (33,62,63,136-138),

we found nutrient-specific systems were preferred over summary indicator systems on a range of

attributes (153). Another survey of Canadian consumers conducted later in 2011 also found

nutrient-specific systems were rated more highly than summary indicator systems on a range of

attributes (160). In our Chapter 6 investigation, we also found that of the two nutrient-specific

systems, consumers rated the daily value system highest on attributes of consumer friendliness

such as liking, helpfulness, credibility, and influence on purchase decisions (153). However, in a

side-by-side comparison, more consumers in the study selected the traffic light system as their

preferred FOP symbol than selected the daily value system. Our Chapter 6 data suggest that, in

addition to being better liked by consumers, nutrient-specific systems are also better understood.

Nutrient-specific systems received higher ratings of self-reported understanding than did

summary indicator systems. Moreover, a test of conceptual understanding suggested that

nutrient-specific systems are better understood than manufacturer developed summary indicator

systems.

101

Our study, presented in Chapter 6, is the only one to date to examine the impact of a mandatory

nutrition label, the NFt, on how consumers perceive different FOP systems (153). Consumers’

ratings of the likability, helpfulness, and credibility of summary indicator systems were lower in

the presence of the NFt. Similarly, the daily value system was rated less helpful when the NFt

was present. In contrast, consumers rated the traffic light system similarly in both the presence

and absence of the NFt.

In summary, our data provide the first evidence that Canadians desire the adoption of a single

standardized FOP system. Further, the evidence demonstrates that nutrient-specific FOP systems

are Canadians’ preferred FOP system, as they were better liked and understood than summary

indicator systems. These findings are important should Canada pursue the adoption of a single,

standardized FOP system as, according to the theoretical model put forward by Grunert and

Wills (68) (Figure 1-2), liking and understanding of a FOP system are pre-requisites for its use

by consumers.

7.2 Future directions

7.2.1 Policy

Our work provides a foundation to support the development of better policy concerning the use

of FOP systems on food labels in Canada. At a minimum, it suggests the need for expansion and

enforcement of existing guidelines for FOP system use. However, the evidence also makes a

strong case for the development of new policies in this area.

7.2.1.1 Enforcing and expanding existing guidelines

FOP systems currently fall under a range of federal regulations and guidelines depending on the

FOP system category as described in sections 1.2.2 and 1.3. Of particular interest are guidelines

put forth in the Canadian Food Inspections Agency’s Guide to Food Labelling and Advertising

that state third-party [FOP systems] should not give the impression that (26):

1. a single food or brand of food is "healthier" than, or nutritionally superior to, other foods

not bearing the … symbol …

2. the food is a treatment, preventative of or cure for disease.

102

With respect to the first point, we – as well as other researchers – have presented evidence that

FOP symbols (third-party or otherwise) are giving consumers the impression that a single food is

“healthier” or nutritionally superior to other foods (120-124,153). In particular the evidence

most strongly supports the role of summary indicator systems in generating these impressions.

To avoid misleading consumers, the Canadian Food Inspection Agency suggests that in addition

to the FOP symbol, at least one of the following should appear on the food label:

1. A statement that clearly explains the reason for the appearance of the [FOP symbol].

2. The name of the third party [who owns the symbol] … is clearly shown, in conjunction

with its nutrition recommendations or dietary guidelines or those it endorses.

3. A clear indication that the [symbol] … does not constitute an endorsement of the food.

Indeed, the Heart and Stroke Foundation require that participating products carry an explanatory

message on-pack in conjunction with its Health Check™ symbol (the only third-party summary

indicator system in Canada), stating “The Heart and Stroke Foundation’s Registered Dietitians

have reviewed this product to ensure it meets the specific nutrient criteria developed by the

Health Check™ program based on the recommendations in Canada’s Food Guide. A fee is paid

by each participating company to help cover the cost of this voluntary, not-for-profit program”

(161). Manufacturer developed FOP systems are not covered under this guideline, and as a

result most manufacturer-developed systems come without such an explanatory message;

additionally, the criteria they use for their systems are often not readily available. As a result, we

recommend that the Canadian Food Inspection Agency expand the guidelines, currently only

applied to third-party systems, to all summary indicator systems and take action to enforce these

guidelines wherever summary indicator systems are found without additional information

stipulated to in the Guide to Food Labelling and Advertising.

Beyond maintaining the status quo with respect to FOP system policy, the expansion and

enforcement of existing guidelines represents the bare minimum that can be done to reduce the

misleading nature of the current FOP situation in Canada. However, this minimal policy

approach would still be lacking in three important aspects: 1) it would not ensure that consumers

could rely on FOP systems to help them identify healthier choices, 2) it would not provide

manufacturers with any impetus to reformulate, and 3) such a policy would not overcome the

confusion associated with multiple symbols and logos (each relying on different nutritional

103

criteria for their use), identified in previous consumer research (131,132). As a result, we do not

recommend expanding and enforcing existing guidelines as the preferred course of action.

Instead, we would recommend the government pursue one of the courses of action explored in

sections 7.2.1.2 or 7.2.1.3 such as establishing a regulated definition for ‘healthy’ for use in food

labelling as is done in the US, or introducing a single, standardized FOP system.

7.2.1.2 Establishing a definition for ‘healthy’ foods

We have presented evidence that foods with FOP symbols have similar calorie and nutrient

contents to foods without symbols. Further, our data suggest that specific FOP systems,

including third-party and manufacturer developed systems, are not a reliable indicator of foods

that meet a higher nutritional standard. In section 6.1 we presented the dual objectives of FOP

systems in promoting population health: supporting healthier choices by consumers and

reformulation by industry. Our findings suggest that to achieve these purposes the government

should establish a minimum standard for the nutrient profiling criteria behind FOP systems. The

development of minimum standards for nutrient profiling criteria would support two potential

FOP system policy alternatives: 1) the development of a single, standardized FOP system, or 2)

new regulatory requirements that establish prescribed conditions for FOP symbol use – like those

currently in place for nutrient content, disease risk reduction, and therapeutic claims.

A 2009 consultation with stakeholders from organizations with policies or programs supported

by nutrient profiles found that 2/3 of them were in favour of the Canadian government

establishing a standardized nutrient profile to define ‘healthy’ foods (94). Supporters believed

that a standard nutrient profiling criteria to define ‘healthy’ foods would help consumers to adopt

a healthier eating pattern and encourage manufacturers to reformulate or innovate healthier

products consistent with this definition. In the US, nutrient profiling criteria have already been

established for the implied claim implicit in the use of the word “healthy” on food labels (Table

7-1) (19). These criteria would also apply to implied “healthy” claims such as those

communicated by summary indicator systems. Further, these nutrient profiling criteria are also

used in the US in relation to nutrient-content claims. Foods carrying nutrient content claims that

do not meet these minimum criteria must carry disclaimers to alert consumers to negative aspects

of the product – i.e. nutrient levels that exceed the “healthy” threshold.

104

Canada currently has no definition of “healthy” in place. Although, Health Canada recently

introduced nutrient profiling criteria to classify foods from the four food groups into four tiers

based on their compliance with Canada’s Food Guide guidance, Health Canada describes these

criteria as exclusively a surveillance tool (111).

In their 2011 report, Front-of-Package Nutrition Rating Systems and Symbols: Promoting Health

Choices, the US Institute of Medicine (109) issued a series of recommendations for the

development of nutrient profiling criteria that may be edifying to Canadian policy makers. The

Institute of Medicine states that FOP system nutrient criteria should:

1. Target nutrients of public health concern (saturated and trans fat, sodium, and added

sugar).

2. Evaluate nutrient amounts consistent with existing science-based regulations.

3. Apply a single set of nutrient criteria of across all/most product categories (to facilitate

comparisons within and across food categories).

4. Allow compliance to be monitored.

With respect to point #2, the Institute of Medicine suggested that nutrient profiling criteria be set

using the nutrient levels set by the US Food and Drug Administration to define eligibility for

nutrient content and health claims. Similarly, Canada could use the nutrient levels set out in the

Food and Drug Regulations for “low” nutrient content claims to set nutrient profiling criteria.

However, criteria for sugar would need to be developed, as no claim currently exists for “low in

sugar” or “low in added sugar” although levels established in other jurisdictions could be applied

or adapted. Further, with respect to point #4, compliance with added sugar criteria cannot be

monitored through chemical analysis and there is no requirement for the declaration of added

versus total sugars in the NFt. Compliance with added sugar criteria would be difficult to

monitor and require review of the ingredient list. However, a recent Health Canada proposal for

modernizing the food label has proposed the addition of ‘added sugars’ to the NFt, and if

accepted would effectively eliminate this challenge (162). However, nutrient profiling criteria

could be established with respect to total sugar rather than added sugar to facilitate monitoring,

105

as was done in Health Canada’s recently introduced nutrient profiling criteria to classify foods

from the four food groups into tiers (111).

Establishing nutrient profiling criteria to define healthy in the current multiple FOP system

environment would help ensure FOP systems support consumers in identifying healthy foods by

ensuring that foods carrying a FOP system of any kind all meet a consistent standard.

Additionally, standard minimum criteria would incentivize manufacturers to reformulate their

products in order to use FOP systems as nutrition marketing on their food labels. However, as

with the policy option of expanding and enforcing existing guidelines, a policy that supports the

continued use of multiple FOP systems within a single marketplace, albeit with standardized

nutrient criteria, would do nothing to alleviate consumer confusion. The implementation of

some form of standardized FOP system supported by a standard definition of “healthy” is the

policy approach most likely to alleviate consumer confusion.

Table7-1. US Food and Drug Administration and US De partment of Agriculture

nutrient profiling criteria for “healthy” claims (109)

Food Individual Food Seafood/Game Meat Meal or Main Dish

Amount of Food Per reference amount and labeled

serving

Per reference amount and per 100

g

Per 100 g and labeled serving

Fat ≤ 3 g ≤ 5 g ≤ 3 g and ≤ 30% or less of calories

Saturated fat ≤ 1 g and ≤ 15% or less of calories

≤ 2 ≤ 1 g and ≤ 15% or less of calories

Sodium ≤ 480 mg* ≤ 480 mg* ≤ 600 mg

Cholesterol ≤ 60 mg ≤ 95 mg ≤ 90 mg

Beneficial nutrient† ≥ 10% daily value ≥ 10% daily value ≥ 10% daily value of two

Fortification In accordance with fortification policy

* Except for foods with a reference amount ≤ 30 g or 2 Tbsp. must contain ≤ 480 mg sodium per 50 g and ≤ 60 mg cholesterol per 50 g

†Beneficial nutrient could be vitamin A, vitamin C, calcium, iron, protein or fiber

106

7.2.1.3 Implementing a single, standardized FOP system: A discussion of policy options

The results from this thesis provide important evidence regarding a range of potential policy

options that could be applied to the FOP labelling situation. Our research has demonstrated a

strong preference for the implementation of a single, standardized FOP system among Canadian

consumers (153). This is in agreement with the most comprehensive review of FOP systems

completed to date. The Institute of Medicine concluded a model FOP symbol system should be

characterized by a single, standard symbol (109). Taken together, this information suggests

Canadian policy makers should revisit their current position regarding FOP systems.

Although our research indicated consumers prefer a mandatory system to a voluntary system, a

standard, voluntary FOP system might still be considered as a viable policy option. Policy

makers in the European Union have introduced a voluntary daily value FOP system, while

voluntary summary indicator systems are in use in countries such as Denmark, Norway, Sweden,

and the Netherlands (15). Most recently, the United Kingdom and Australia have introduced

voluntary FOP systems for use on packaged foods, with the intent to regulate if voluntary uptake

by industry is limited (16,17). All these programs are underpinned by standardized nutrient

profiling criteria. Similarly, Canada could introduce a voluntary FOP system with specific

nutrient profiling criteria and guidelines for the system’s appearance. This could increase

consistency in the marketplace, and potentially decrease consumer confusion. However, if

voluntary adoption of a standardized system by manufacturers is not widespread the persistence

of competing symbols in the marketplace may lead to continued confusion.

To reduce the confusion generated by multiple competing FOP systems, the Government of

Canada could prohibit the use FOP systems outside the standard voluntary FOP system. Indeed

the Institute of Medicine has recommended that a standardized FOP system be used to the

exclusion of proprietary FOP systems (109). However, a voluntary, exclusive FOP system

policy approach may still be misleading to consumers if they are unclear as to why some foods

carry the FOP system and others to do not.

To truly allow for comparison, the Institute of Medicine has recommended that a standardized

FOP system appear in a consistent location on all food products (109), leading to our final policy

option the implementation a mandatory, standardized FOP system. This approach would be akin

107

to current regulations in place related to the NFt with prescribed formatting and clear guidelines

describing which products must carry the system. A consistent and mandatory FOP system with

a standard format and nutrient profiling criteria would be the system most likely to facilitate the

selection of healthier choices by consumers, as well as stimulate product reformulation among

manufacturers. Further, it would ensure FOP systems are not misleading or deceptive, and avoid

the consumer confusion associated with multiple FOP systems or the inconsistent application of

FOP systems and their nutrient profiling criteria. However, in the current political climate,

where voluntary public health nutrition initiatives have been preferred over regulatory ones in

cases that involved the food industry in recent years, such a policy is unlikely (14,163).

Moreover, legal scholars have raised concerns that an outright ban on the use of proprietary FOP

systems may lead to challenges under the Canadian Charter of Rights and Freedoms (164).

7.2.1.3.1 Format With respect to a standardized FOP system for Canada, our Chapter 6 investigation has provided

evidence that Canadian consumers may be best served by a nutrient-specific FOP system.

Nutrient-specific systems have several strengths relative to the other FOP systems and should be

strongly considered if a single, standardized FOP system is pursued. First, they typically apply a

consistent approach across all or most food and beverage categories, facilitating consumer

product comparisons both within and across food categories (19). In addition, they are consistent

with existing regulations for declaring nutrient amounts per serving and for nutrient content

claims. Further, nutrient-specific systems that display the amount of calories and other nutrients

on the FOP, as in the Daily Value system we tested, allow consumers to make food selections

within the context of their total diet using the percent daily value. Furthermore, nutrient-specific

systems that characterize nutrient amounts as “high”, “medium”, and “low”, as in the traffic light

system we tested, would add further ‘interpretive’ information that helps consumers make food

choices consistent with dietary guidance. From a manufacturer perspective, a nutrient-specific

system may encourage product reformulation to meet the nutrient profiles targeted in the system.

Limitations associated with nutrient-specific systems relate to limited amounts of space available

on labels, which may restrict the amount of information that can be included in an FOP symbol.

Further, there is concern that the inclusion of too many nutrients may clutter the label and

decrease consumers’ ability to use the label.

In addition, the two nutrient-specific systems preferred in our Chapter 6 investigation are

108

consistent with a number of the conclusions reached by the US Institute of Medicine in their

2010 and 2011 reviews of FOP nutrition rating systems and symbols (19,109). Namely, these

systems declare calorie and serving size information, which the Institute of Medicine concluded

was important to facilitate the selection of lower-calorie foods and/or appropriate serving sizes to

combat overweight and obesity (19). Further, both systems highlight most of the nutrients

deemed to be of greatest public health significance –saturated fat, sodium, and sugar (19,109).

Trans fat was omitted from the labels we tested but could be substituted in place of total fat,

which the Institute of Medicine concluded lacked sufficient evidence to support its inclusion

(19). Finally, these nutrient-specific systems would be reinforcing of the existing mandatory

nutrition label, the NFt.

However, these formats diverge from the Institute of Medicine recommendations with respect to

several key points. First, the Institute of Medicine recommended that FOP systems translate

nutrition label information into “quickly and easily grasped health meaning, making healthier

options unmistakable” (109). However, Daily Value systems provide information only and

provide no translation. Traffic light systems translate nutrient information into “green”,

“amber”, and “red” colours and provide health meaning, however it is unclear if such an

approach would make healthier options “unmistakable”. For example, when comparing a

product with three green and one red traffic light to a product with two green and two yellow

traffic lights, the healthier option may not be evident. As an alternative, the Institute of Medicine

has suggested that the FOP system most likely to enable healthier food choices would be (109):

1. Simple (requiring no nutritional knowledge or sophistication on the part of the

consumer to understand).

2. Interpretive (providing guidance based on nutrition information without providing any

specific nutrition information).

3. Ordinal (using a scaled or ranked approach).

As such the Institute of Medicine has proposed a FOP system that uses 0-3 stars based on the

food’s content of nutrients of public health concern, but does not explicitly mention these

nutrients in the symbol or the amount of these nutrients a serving of the food provides. While

based on the best available evidence, the Institute of Medicine’s proposed FOP system has not

been tested with consumers.

109

7.2.2 Research

Although our research has pointed to several possible policy directions for FOP systems in

Canada, we have concluded that the preponderance of evidence supports the adoption of a single,

standardized, and mandatory FOP system with a common nutrient profiling scheme. However,

continued research is needed to support the development of policy in this area. Further, any

evolutions in FOP system policy will need to be followed up with evaluation. Our above

discussion leads to a number of valuable research questions.

From this research, it is clear that nutrient profiling criteria to define “healthy” to form the basis

of either a single, standardized FOP system or existing third-party and manufacturer developed

FOP systems is needed. Presently, there is no gold standard for defining “healthy” foods (95),

however as described in section 1.3.3, there are several important considerations in developing a

nutrient profiling system. Considerations include the nutrient profiling system’s purpose and

target audience, application, included nutrients and food components, amount of food, type of

model, and basis of the system (19,20,93-95,105,106). The Institute of Medicine has proposed

that nutrient profiling criteria to support FOP systems be developed on the basis of existing

nutrient-content and health claims (109). Regardless of whether criteria are developed based on

existing regulations as proposed, or are modeled after existing nutrient profiling systems

developed internationally, or are based on current healthy eating guidance, future investigations

should focus on the creation and validation of a nutrient-profiling criteria to support FOP

systems in Canada, or to support healthy eating by Canadian consumers.

Investigations into how proposed nutrient profiling criteria may impact the food supply and

nutrient intake of Canadians will also be needed. As in our Chapter 4 investigation, proposed

nutrient profiling criteria could be applied to the existing Canadian food supply to determine the

proportion of products already meeting the criteria based on their existing composition. This

would permit the evaluation of the potential of different nutrient profiling criteria to stimulate

reformulation. Further, investigations will be needed to determine if proposed nutrient profiling

criteria for Canada will support improved dietary intakes and health. The potential impact of

foods that meet third-party or industry summary indicator system nutrient-profiling criteria on

health risk factors and outcomes, and dietary intakes has been explored in a handful of studies,

with positive results (165-169). However, the potential impact of nutrient-specific systems –

110

such as traffic light nutrient profiling criteria or nutrient-content claim criteria – on dietary

intakes have not been studied.

Related to FOP system format, although our Chapter 6 investigation singled out nutrient-specific

systems as the preferred format these systems share many common features with the existing NFt

(153). Common features include complex nutrition terminology as well as high numeracy

requirements related to using serving size, percent daily value, and nutrient content information.

As described in section 1.2.1.1 these features have been an impediment to the use and

understanding the existing mandatory nutrition label and have even prompted some of the calls

to implement a single, standard FOP system. Research has shown that consumers with the

lowest levels of nutrition and health knowledge and infrequent label readers found nutrient

specific systems more challenging to understand than summary indicator systems (63), and these

consumers preferred summary indicator systems to nutrient specific systems (138). Mejean et al.

(138) found that, compared to consumers with higher levels of education, a greater proportion of

consumers with lower education levels preferred summary indicator systems over nutrient-

specific systems. The population surveyed in our Chapter 6 investigation had higher levels of

education as well as a demonstrated interest in nutrition and greater tech literacy through their

regular participation in online on food and nutrition surveys. As a result, our population’s

preference for nutrient-specific FOP systems may not be generalizable to population subgroups

with lower levels of education and nutrition knowledge – who may be in most need of help.

Investigations that explore preferences and understanding of FOP systems among consumers

with lower levels of education and health literacy will be needed prior to identifying an

appropriate standard FOP system for Canada. Focus group discussions may be a more

appropriate mean to get at this needs and concerns of these population subgroups. The nutrient-

specific systems evaluated in our study should also be tested against the simple, interpretive, and

ordinal FOP system proposed by the US Institute of Medicine, which was published after our

study was completed.

Research to date evaluating the effectiveness of FOP systems has been deemed to be of low

methodological quality (170). Self-reported, observational studies, and survey experiments, like

the one reported on in this thesis, need to be followed up with studies that examine end points

such as sales, reformulation, and health outcomes. However, the majority of research examining

these outcomes has been rated as low to mediocre (170). Natural experiments that include a

111

control are needed to build upon the existing FOP sales studies. Further, prospective studies of

reformulation in response to FOP systems are needed. Most existing studies have retrospectively

asked manufacturers participating in third-party FOP programs about reformulation in response

to involvement in the program (112-115). Ideally, prospective studies should include both

participating and non-participating manufacturers to determine if reformulation can be attributed

to participating in the FOP system. Moreover, a prospective study would provide evidence of

the impact of voluntary FOP systems on the entire food supply. Finally, Vyth et al. (170) stated

that the best evidence of FOP system effectiveness will be generated by a longitudinal,

randomized controlled trial in a real-life setting that examines the health effects of FOP systems

on biomarkers of health and nutritional status. While this would be the gold standard, the

feasibility of such an approach is questionable as it would be difficult both to establish an

appropriate control group and conduct a study of sufficient duration to demonstrate an effect and

attribution of FOP labelling to health outcomes would be difficult to quantify separately from

other risk factors.

112

Chapter 8

8 Summary and Conclusions

FOP systems are widespread in the Canadian marketplace, yet there are no specific regulations

governing their use. The Canadian Government has expressed its intentions to maintain the

status quo with respect to FOP systems in Canada. There is ample evidence regarding the

challenges of using the current NFt; however, data are lacking regarding FOP systems in the

Canadian context, and it is unclear if Canadians are being well served by the current minimal

policies regarding FOP systems in Canada. Further, Canadians’ liking and understanding of

different FOP systems have not been studied. Thus, the overall purpose of this research was to

generate food supply and consumer data to support the evolution of Canadian food labelling

policy as it relates to FOP systems.

We observed that although a large proportion of Canadian food products qualify for common

third-party and manufacturer developed summary indicator FOP systems, few products that meet

the systems criteria are identified by their symbols (Chapter 4). Given this, we concluded that

these symbols provide minimal impetus for reformulation for prospective participants. Further,

we concluded that consumers cannot rely on these summary indicator systems to identify

products of superior nutritional quality.

Additionally, in Chapter 5 we observed that products with FOP systems were similar in calorie

and nutrient composition to products without FOP systems. Further, we found that no specific

FOP system type was a better indicator of products of higher nutritional quality. Given the

similar nutrient profiles of foods with and without FOP symbols, we concluded such symbols

were deceptive and misleading consumers. Therefore, our results suggest that the introduction of

minimum nutrient-profiling criteria for all FOP systems may be an appropriate policy option to

minimize this problem.

In our third study (Chapter 6) we demonstrated that consumers perceived a frozen meal and

breakfast cereal as healthier and/or lower in negative nutrients and higher in positive nutrients

when they carried FOP systems. For the most part, this effect was attenuated in the presence of

the NFt – supporting the concerns that FOP systems have the potential to mislead when used

113

without the mandatory nutrition label, particularly when they truncate the search for additional

information found in the NFt. Our study was the first to test Canadians liking and understanding

of different FOP systems. These novel data suggest that nutrient-specific FOP systems should be

explored as possible single, standardized FOP labelling system for Canadians.

When our results are considered in relation to other studies of FOP systems, they suggest that

Canadians may not be well served by Canada’s existing policies on FOP systems. FOP system

policy in Canada needs to evolve to include standard minimum nutrient profile criteria, and

perhaps the implementation of a single, standardized nutrient-specific system for use on food

labels.

114

References

(1) World Health Organization. Diet, nutrition and the prevention of chronic diseases. World Health

Organization technical report series 2003;916.

(2) The Secretariat for the Intersectoral Healthy Living Network in partnership with the F/P/T Healthy

living Task Group and the F/P/T Advisory Committee on Population Health and Health Security. The

Integrated Pan-Canadian Healthy Living Strategy. 2005.

(3) World Health Organization. Global Strategy on Diet, Physical Activity and Health. 2004.

(4) Health Canada. Eating Well with Canada's Food Guide. 2007.

(5) Government of Canada. Regulations Amending the Food and Drug Regulations (Nutrition Labelling,

Nutrient Content Claims and Health Claims). The Canada Gazette, Part II 2003;137(1):154.

(6) Canadian Council of Food and Nutrition. Tracking Nutrition Trends VII. 2008.

(7) Ni Mhurchu C, Gorton D. Nutrition labels and health claims in New Zealand and Australia; a review

of use and understanding. Aust NZ J Public Health 2007;31:105-112.

(8) Gorton D. Nutrition labelling - Update of scientific evidence on consumer use and understanding of

nutrition labels and claims. 2007.

(9) Cowburn G, Stockley L. Consumer understanding and use of nutrition labelling: A systematic review.

Public Health Nutr 2005;8(1):21-28.

(10) van Kleef E, van Trijp H, Paeps F, Fernández-Celemín L. Consumer preferences for front-of-pack

calories labelling. Public Health Nutr 2007;11(2):203-213.

(11) The Strategic Counsel. Focus Testing of Creatives for the Nutrition Facts Education Initiative.

2010;HC POR 09-16.

(12) Emrich TE, L'Abbe MR. Nutrition Labelling Education Research Synthesis. 2014 Janaury 2014.

(13) The Standing Committee on Health. Healthy Weights for Healthy Kids. 2007.

(14) Sodium Working Group. Sodium Reduction Strategy for Canada: Recommendations of the Sodium

Working Group. 2010.

115

(15) European Food Information Council. Global Update on Nutrition Labelling. 2013 February 2013.

(16) Department of Health, Food Standards Agency, Welsh Government, The Scottish Government.

Guide to creating a front of pack (FoP) nutrition label for pre-packed products sold through retail outlets.

2013.

(17) Plibersek T, Neumann S. Star Rating System for Packaged Foods. Joint Media Release 2013 2013

Jun 14.

(18) Health Canada. Managing health claims for foods in Canada: Towards a modernized framework.

2007.

(19) Committee on the Examination of Front-of-Package Nutrition Rating Systems and Symbols, Institute

of Medicine. Examination of Front-of-Package Nutrition Rating Systems and Symbols: Phase I Report.

2010.

(20) Lobstein T, Davies S. Defining and labelling ‘healthy’ and ‘unhealthy’ food. Public Health Nutr

2009;12(3):331-340.

(21) Carlson LA. Front-of-the-Pack and On-Shelf Labeling: Tools for Spotting Nutritious Choices at the

Supermarket Shelf. Nutr Today 2010;45(1):15-21.

(22) Lupton JR, Balentine DA, Black RM, Hildwine R, Ivens BJ, Kennedy ET, et al. The Smart Choices

front-of-package nutrition labeling program: rationale and development of the nutrition criteria. Am J

Clin Nutr 2010;91(4):1078S-1089S.

(23) World Health Organization. Global status report on non-communicable diseases 2010. 2011;

Available at: Global status report on non-communicable diseases 2010. Accessed January 29, 2014.

(24) Mirolla M. The Cost of Chronic Disease in Canada. 2004.

(25) Beaglehole R, Bonita R, Horton R, Adams C, Alleyne G, Asaria P, et al. Priority actions for the non-

communicable disease crisis. Lancet 2011;377:1438-1447.

(26) Canadian Food Inspection Agency. Guide to Food Labelling and Advertising. 2010; Available at:

http://www.inspection.gc.ca/english/fssa/labeti/guide/ch1e.shtml. Accessed October 27, 2010.

(27) Codex Alimentarius Commission. Guidelines on Nutrition Labelling. 2012; Available at:

www.codexalimentarius.net/download/standards/34/CXG_002e.pdf. Accessed January 29, 2014.

116

(28) Environics Research Group. Post-Campaign Evaluation – Healthy Eating. 2008.

(29) The Strategic Counsel. Canadians' Understanding and Use of the Nutrition Facts table: Baseline

National Survey Results. 2011;POR 031-10(HCPOR-10-06).

(30) The Strategic Counsel. Canadians' Understanding and Use of the Nutrition Facts table: Survey

Results. 2012;POR-088-11(HCPOR-11-01).

(31) US Department of Health and Human Services, Centers for Disease Control and Prevention. Healthy

People 2000 Final Review. 2001.

(32) Drichoutis AC, Lazaridis P, Nayga RM. Who is looking for nutritional food labels? Eurochoices

2005;4:18-23.

(33) Gorton D, Ni Mhurchu C, Chen M, Dixon R. Nutrition labels: a survey of use, understanding and

preferences among ethnically diverse shoppers in New Zealand. Public Health Nutr 2008;12(9):1359-

1365.

(34) Miller C, Brown J. Knowledge and use of the food label among senior women in the management of

type 2 diabetes mellitus. Journal of Nutrition, Health & Aging 1999;3(3):152-7.

(35) Rothman RL, Housam R, Weiss H, Davis D, Gregory R, Gebretsadik T, et al. Patient understanding

of food labels: The role of literacy and numeracy. American Journal of Preventive Medicine

2006;31(5):391-398.

(36) Jay M, Adams J, Herring SJ, Gillespie C, Ark T, Feldman H, et al. A randomized trial of a brief

multimedia intervention to improve comprehension of food labels. Prev Med 2009 200901;48(1):25-31.

(37) Kristal AR, Levy L, Patterson RE, Li SS, White E. Trends in food label use associated with new

nutrition labeling regulations. American Journal of Public Health 1998;88(8):1212-1215.

(38) Pelletier AL, Chang WW, Delzell JEJ, McCall JW. Patients' Understanding and Use of Snack Food

Package Nutrition Labels. The Journal of the American Board of Family Practice 2004;17(5):319-323.

(39) Hawthorne KM, Moreland K, Griffin IJ, Abrams SA. An Educational Program Enhances Food Label

Understanding of Young Adolescents. J Am Diet Assoc 2006;106(6):913-916.

(40) Lewis CJ, Yetley Ea. Nutrition labels in bar graph format deemed most useful for consumer purchase

decisions using adaptive conjoint analysis. Journal of the American Dietetic Association 1992;92(1):62.

117

(41) Hrovat KB, Harris KZ, Leach AD, Russell BS, Harris BV, Sprecher DL. The new food label, type of

fat, and consumer choice. A pilot study. Archives of family medicine 1994;3(8):690-695.

(42) Dooley DA, Novotny R, Britten P. Integrating research into the undergraduate nutrition curriculum:

Improving shoppers’ awareness and understanding of nutrition facts labels. J Nutr Educ 1998;30:225-231.

(43) Levy AS, Fein SB. Consumers' Ability to Perform Tasks Using Nutrition Labels. J Nutr Educ

1998;30(4):210-217.

(44) Miller CK, Jensen GL, Achterberg CL. Evaluation of a Food Label Nutrition Intervention for

Women with Type 2 Diabetes Mellitus. J Am Diet Assoc 1999;99(3):323-328.

(45) Levy L, Patterson RE, Kristal AR, Li SS. How well do consumers understand percentage daily value

on food labels? American Journal of Health Promotion 2000;14(3):157-160.

(46) Sullivan AD. Determining how low-income food shoppers perceive, understand, and use food labels.

Canadian Journal of Dietetic Practice and Research 2003;64(1):25-30.

(47) Misra R. Knowledge, Attitudes, and Label Use among College Students. J Am Diet Assoc

2007;107(12):2130-2134.

(48) NRG Research Group. Qualitative Study on Use and Understanding of Nutrition Labelling. 2007.

(49) Ascentum. Nutrition Labelling Resources/Tools Consultation. 2009.

(50) Phase 5. Nutrition Facts Education Campaign Web site Usability Testing. 2010;HCPOR-10-05.

(51) Lubman N, Doak C, Jasti S. Food Label Use and Food Label Skills among Immigrants from the

Former Soviet Union. Journal of Nutrition Education and Behavior 2012 201209/10;44(5):398-406.

(52) Vanderlee L, Goodman S, Yang WS, Hammond D. Consumer understanding of calorie amounts and

serving size: implications for nutritional labelling. Canadian Journal of Public Health 2012 September;

2013/11;103:327+.

(53) Lando AM, Lo SC. Single-Larger-Portion-Size and Dual-Column Nutrition Labeling May Help

Consumers Make More Healthful Food Choices. Journal of the Academy of Nutrition and Dietetics 2013

201302;113(2):241-250.

118

(54) Lando AM, Labiner-Wolfe J. Helping consumers make more healthful food choices: Consumer

views on modifying food labels and providing point-of-purchase nutrition information at quick service

restaurants. Journal of Nutrition Education and Behaviour 2007;39(3):157-163.

(55) Levy AS, Fein SB, Schucker RE. More effective nutrition label formats are not necessarily preferred.

Journal of the American Dietetic Association 1992;92(10):1230-1234.

(56) Silverglade B, Ringel Heller I. Food Labelling Chaos - The Case for Reform. 2010.

(57) Turnbull J. Canadian Medical Associations's Presentation to the House of Commons Standing

Committee on Health Nutrition Labelling. 2011 March 3.

(58) Jeffery B. Letter to Federal Minister of Health: Ensure that the Food Label Modernization initiative

includes health and nutrition issues to protect families, save health care resources, and raise workforce

productivity by disease prevention. Centre for Science in the Public Interest 2013 October 16.

(59) Burton S, Andrews JC. Age, product nutrition, and label format effects on consumer perceptions and

product evaluations. The Journal of Consumer Affairs 1996 Summer 1996;30(1):68.

(60) Jones G, Richardson M. An objective examination of consumer perception of nutrition information

based on healthiness ratings and eye movements. Public Health Nutr 2007 Mar.;10(3):238-244.

(61) Miller C, Probart C, Achterberg C. Knowledge and Misconceptions About the Food Label Among

Women With Non-Insulin-Dependent Diabetes Mellitus. Diabetes Educ 1997;23(4):425-432.

(62) Signal L, Lanumata T, Robinson J, Tavila A, Wilton J, Ni Mhurchu C. Perceptions of New Zealand

nutrition labels by Maori, Pacific and low-income shoppers. Public Health Nutrition 2007;11(7):706-713.

(63) Feunekes GIJ, Gortemaker IA, Willems AA, Lion R, van den Kommer M. Front-of-pack nutrition

labelling: Testing effectiveness of different nutrition labelling formats front-of-pack in four European

countries. Appetite 2008;50(1):57-70.

(64) Borgmeier I, Westenhoefer J. Impact of different food label formats on healthiness evaluation and

food choice of consumers: A randomized-controlled study. BMC Public Health 2009;9:184.

(65) Kelly B, Hughes C, Chapman K, Louie JC, Dixon H, Crawford J, et al. Consumer testing of the

acceptability and effectiveness of front-of-pack food labelling systems for the Australian grocery market.

Health Promot Int 2009;24(2):120-129.

119

(66) McCullum C, Achterberg CL. Food shopping and label use behavior among high school-aged

adolescents. Adolescence 1997;32(125):181-197.

(67) Wansink B. How Do Front and Back Package Labels Influence Beliefs About Health Claims? The

Journal of Consumer Affairs 2003;37(2):305-316.

(68) Grunert KG, Wills JM. A review of European research on consumer response to nutrition

information on food labels. J Public Health 2007;15(5):385-399.

(69) Colby SE. Nutrition Marketing on Food Labels. Proceedings of the Academy of Marketing Studies

2008;13(1):17-21.

(70) Colby SE, Johnson L, Hoverson B. Nutrition marketing on food labels. J Nutr Educ Behav

2010;42(2):92-98.

(71) Health Canada. Substantiation of Health Claims (Scientific Evidence). 2014; Available at:

http://www.hc-sc.gc.ca/fn-an/label-etiquet/claims-reclam/assess-evalu/index-eng.php. Accessed Feb 12,

2014.

(72) Health Canada. Summary of Health Canada's Assessment of a Health Claim about Ground Whole

Flaxseed and Blood Cholesterol Lowering. 2014; Available at: http://www.hc-sc.gc.ca/fn-an/label-

etiquet/claims-reclam/assess-evalu/flaxseed-graines-de-lin-eng.php. Accessed Feb 12, 2014.

(73) Government of Canada. Food and Drugs Act. 2010.

(74) Health Canada. Guidance Document – The Use of Probiotic Microorganisms in Food 2009 April

2009.

(75) LeGault L, Brandt MB, McCabe N, Adler C, Brown A, Brecher S. 2000-2001 Food Label and

Package Survey; An Update on the Prevalence of Nutrition Labeling and Claims on Processed, Packaged

Foods. Journal of the American Dietetic Association 2004;104(6).

(76) Brandt M, Moss J, Ferguson M. The 2006-2007 food label and package survey (FLAPS): nutrition

labeling, trans fat labeling. Journal of Food Composition and Analysis 2009 12;22:S74-S77.

(77) Brandt MB, Moss J, Ellwood K, Ferguson M, Asefa A. Tracking Label Claims. Chicago 2010

20100101;64(1):34.

120

(78) Schermel A, Emrich T, Arcand J, Wong C, L’Abbe M. Nutrition Marketing on Processed Food

Packages in Canada: 2010 Food Label Information Program. Appl Physiol Nutr Metab 2013;38(6):666-

672.

(79) Williams P, Yeatman H, Ridges L, Houston A, Rafferty J, Roasler A, et al. Nutrition function, health

and related claims on packaged Australian food products - prevalence and compliance with regulations.

Aia Pac J Clin Nutr 2006;15(1):10-20.

(80) Bureau Europeen des Unions de Consommateurs. Report on European Consumers’ Perception of

Foodstuffs Labelling. 2005.

(81) Canadian Council of Food and Nutrition. Ethnographic Study: A New Perspective on Canadians'

Attitudes and Behaviours Toward Food and Nutrition. 2010.

(82) Andrews JC, Netemeyer R, Burton S. Consumer generalization of nutrient content claims in

advertising. Journal of Marketing 1998;62:62-75.

(83) Roe B, Levy AS, Derby BM. The impact of health claims on consumer search and product

evaluations outcomes: Results from FDA experimental data. Journal of Public Policy and Marketing

1999;18(1):89-105.

(84) Kozup JC, Creyer EH, Burton S. Making Healthful Food Choices: The Influence of Health Claims

and Nutrition Information on Consumers' Evaluations of Packaged Food Products and Restaurant Menu

Items. Journal of Marketing 2003;67(2):19-34.

(85) van trijp, Hans C. M., van der Lans, Ivo A. Consumer perceptions of nutrition and health claims.

Appetite 2007;48:305-324.

(86) Labiner-Wolfe J, Lin CJ, Verrill L. Effect of Low-carbohydrate Claims on Consumer Perceptions

about Food Products’ Healthfulness and Helpfulness for Weight Management. Journal of Nutrition

Education and Behavior 2010;42(5):315-320.

(87) Lähteenmäki L, Lampila P, Grunert K, Boztug Y, Ueland Ø, Åström A, et al. Impact of health-

related claims on the perception of other product attributes. Food Policy 2010;35(230):239.

(88) Zank GM, Kemp E. Examining Consumers’ Perceptions of the Health Benefits of Products with

Fiber Claims. The Journal of Consumer Affairs 2012;46(2):333-344.

121

(89) Wong CL, Arcand J, Mendoza J, Henson SJ, Qi Y, Lou W, et al. Consumer attitudes and

understanding of low sodium claims on food: An analysis of healthy and hypertensive individuals.

American Journal of Clinical Nutrition 2013;97(6):1288-1298.

(90) Government of Canada. Food and Drug Regulations. 2010.

(91) Verbekea W, Scholdererb J, Lahteenmakib LL. Consumer appeal of nutrition and health claims in

three existing product concepts. Appetite 2009;52:684-692.

(92) Dietitians of Canada. Diabetes, Obesity and Cardiovascular Disease Network. Evidence-Based

Background Paper on Point-of-Purchase Nutrition Programs. 2006.

(93) Stockley L. Review of 'front of pack' nutrition schemes. 2007.

(94) Reza Z. Defining “Healthy” Foods Environmental Scan of the Situation in Canada. 2009.

(95) World Health Organization. Nutrient Profiling: Report of a WHO/IASO Techinical Meeting. 2010

Oct 4-6.

(96) The 2007/08 Food Regulation Standing Committee, Front-of-Pack Labelling Working Group.

Information Prepared by The 2007/08 Food Regulation Standing Committee, Front-of-Pack Labelling

Working Group.

(97) Rayner M, Boaz A, Higginson C. Consumer use of health-related endorsements on food labels in the

United Kingdom and Australia. J Nutr Educ 2001;33(1):24-30.

(98) Vyth EL, Steenhuis IHM, Vlot JA, Wulp A, Hogenes MG, Looije DH, et al. Actual use of a front-of-

pack nutrition logo in the supermarket: consumers' motives in food choice. Public Health Nutr

2010;13(11):1882-1889.

(99) Williams P, Duncan R, de Agnoli K, Hull A, Owers A, Wang T. Introduction and use of front of

pack daily intake labelling on Australian packaged foods 2007-2009. Food Australia 2010;62(12):583-

588.

(100) Storcksdieck genannt Bonsmann S, Fernandez-Celemın L, Larranaga A, Egger S, Wills JM,

Hodgkins C, et al. Penetration of nutrition information on food labels across the EU-27 plus Turkey.

European Journal of Clinical Nutrition 2010;64:1379-1385.

122

(101) The Grocery Manufacturers Association, The Food Marketing Institute. Facts Up Front. Available

at: http://www.factsupfront.org. Accessed Feb 14, 2014.

(102) Heart and Stroke Foundation. Heart and Stroke Health Check Program. 2010; Available at:

http://www.healthcheck.org/page/what-health-check. Accessed July 12, 2011.

(103) NuVal LLC. NuVal Nutrition Made Easy. 2014; Available at: http://www.nuval.com/. Accessed

Feb 14, 2014.

(104) Whole Grains Council. Whole Grain Stamp. 2011; Available at:

http://www.wholegrainscouncil.org/whole-grain-stamp. Accessed July 13, 2011.

(105) Garsetti M, de Vries J, Smith M, Amosse A, Rolf-Pedersen N. Nutrient profiling schemes:

overview and comparative analysis<br />. European Journal of Clinical Nutrition 2007;46:15-28.

(106) Hawkes C. Defining “Healthy” and “Unhealthy”Foods: An International Review. 2009.

(107) Roodenburg AJC, Popkin BM, Seidell JC. Development of international criteria for a front of

package food labelling system: the International Choices Programme. European Journal of Clinical

Nutrition 2011;65:1190-1200.

(108) Trichterborn J, Harzer G, Kunz C. Nutrient profiling and food label claims: evaluation of dairy

products in three major European coutries. Eur J Clin Nutr 2011;65:1032-1038.

(109) Committee on Examination of Front-of-Package Nutrition Ratings Systems and Symbols (Phase II),

Institute of Medicine. Front-of-Package Nutrition Rating Systems and Symbols: Promoting Healthier

Choices. 2011 October 20.

(110) Brownell KD, Koplan JP. Front-of-Package Nutrition Labeling - An abuse of trust by the food

industry? N Engl J Med 2011;364(25):2373-2375.

(111) Health Canada. Summary of the Development and Use of a Surveillance Tool: The Classification of

Foods in the Canadian Nutrient File According to Eating Well with Canada's Food Guide. 2014;130517.

(112) Young L, Swinburn B. Impact of the Pick the Tick food information programme on the salt content

of food in New Zealand. Health Promotion International 2002;17(1):13-19.

123

(113) Williams P, McMahon A, Boustead R. A case study of sodium reduction in breakfast cereals and

the impact of the Pick the Tick food information program in Australia. Health Promotion International

2003;18(1):51-56.

(114) Vyth EL, Steenhuis IHM, Roodenburg AJC, Brug J, Seidell JC. Front-of-pack nutrition label

stimulates healthier product development: a quantitative analysis. Int J Behav Nutr Phys Act 2010;7.

(115) Dummer J. Sodium reduction in Canadian food products with the health check program. Can J Diet

Pract Res 2012;73(1):e227-e332.

(116) Louie JC, Dunford EK, Walker KZ, Gill TP. Nutritional Quality of Australian Breakfast Cereals.

Are They Improving? Appetite 2012;59:464-470.

(117) Hodgkins C, Barnett J, Wasowicz-Kirylo G, Stysko-Kunkowska M, Gulcan Y, Kustepeli Y, et al.

Understanding how consumers categorise nutritional labels: A consumer derived typology for front-of-

pack labelling. Appetite 2012;59:806-817.

(118) Sutherland LA, Kaley LA, Fischer L. Guiding Stars: the effect of a nutrition navigation program on

consumer purchases at the supermarket. Am J Clin Nutr 2010 20100401;91(4):1090S.

(119) Hawkes C. Nutrition labels and health claims: the global regulatory environment. 2004.

(120) Scott V, Worsley AF. Ticks, claims, tables and food groups: A comparison for nutrition labelling.

Health Promot Int 1994;9(1):27-37.

(121) Reid RD, Slovinec D'Angelo ME, Dombrow CA, Heshka JT, Dean TR. The Heart and Stroke

Foundation of Canada's Health Check food information program: modelling program effects on consumer

behaviour and dietary practices. Can J Public Health 2004 Mar-Apr;95(2):146-150.

(122) Steenhuis IHM, Kroeze W, Vyth EL, Valk S, Verbauwen R, Seidell JC. The effects of using a

nutrition logo on consumption and product evaluation of a sweet pastry. Appetite 2010;55(3):707-709.

(123) Andrews JC, Burton S, Kees J. Is simpler always Better? Consumer evaluations of front-of-package

nutrition symbols. JPP&M 2011.

(124) Newman CL, Howlett E, Burton S. Shopper Response to Front-of-Package Labeling Programs:

Potential Consumer and Retail Store Benefits. Journal of Retailing 2014;90(1):13-26.

124

(125) Roberto CA, Shivaram M, Martinez O, Boles C, Harris JL, Brownell KD. The Smart Choices front-

of-package nutrition label. Influence on perceptions and intake of cereal. Appetite 2012;58:651-657.

(126) Hawley KL, Roberto CA, Bragg MA, Liu PJ, Schwatz MB, Brownell KD. The science of front-of-

package labels. Public Health Nutr 2012.

(127) Temple JL, Johnson KM, Archer K, LaCarte A, Yi C, Epstein LH. Influence of simplified nutrition

labeling and taxation on laboratory energy intake in adults. Appetite 2011;57:184-192.

(128) Sacks G, Rayner M, Swinburn B. Impact of front-of-pack ‘traffic-light’ nutrition labelling on

consumer food purchases in the UK. Health Promot Int 2009 2009;24(4):344-352.

(129) Sacks G, Tikellis K, Miller L, Swinburn B. Impact of 'traffic-light' nutrition information on online

food purchases in Australia. Aust NZ J Public Health 2011;35(2):122-126.

(130) Neuhouser ML, Kristal AR, Patterson RE. Use Of Food Nutrition Labels is Associated with Lower

Fat Intake. J Am Diet Assoc 1999;99(1):45-53.

(131) Draper AK, Adamson AJ, Clegg S, Malam S, Rigg M, Duncan S. Front-of-pack nutrition labelling:

are multiple formats a problem for consumers. European Journal of Public Health 2011;23(3):517-521.

(132) Malam S, Clegg S, Kirwan S, McGinigal S. Comprehension and use of UK nutrition signpost

labelling schemes. 2009.

(133) Schmidt S. Aglukkaq pans U.S. nutrition recommendations. Postmedia News 2011 October 20.

(134) Grunert KG, Fernández-Celemín L, Wills JM, Bonsmann SSg, Nureeva L. Use and understanding

of nutrition information on food labels in six European countries. J Public Health 2010 Jun;18(3):261-

277.

(135) 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 Oct;55(2):177-189.

(136) Berning JP, Chouinard HH, McCluskey JJ. Consumer Preferences for Detailed versus Summary

Formats of Nutrition Information on Grocery Store Shelf Labels<br />. Journal of Agricultural & Food

Industrial Organization 2008;6(1):1542.

(137) Moser A, Hoefkens C, Van Camp J, Verbeke W. Simplified nutrient labelling: Consumers'

perceptions in Germany and Belgium. J Consum Prot Food Saf 2010 May;5(2):169-180.

125

(138) Mejean C, Macouillard P, Peneau S, Hercberg S, Castetbon K. Perceptions of front-of-pack labels

according to social characteristics, nutritional knowledge and food purchasing habits. Public Health Nutr

2012;16(3):392-402.

(139) Maubach N, Hoek J. A qualitative study of New Zealand parents' views on front-of-pack nutrition

labels. Nutrition & Dietetics 2010;67:90-96.

(140) Carter O, Mills B, Phan T. An independent assessment of the Australian food industry's Daily

Intake Guide 'energy alone' label. Health Promotion Journal of Australia 2011;22(1):63-67.

(141) Larsson I, Lissner L, Wilhelmsen L. The 'Green Keyhole' revisited: Nutritional knowledge may

influence food selection. European Journal of Clinical Nutrition 1999;52:777-780.

(142) Evidence for Health Claims on Food: How Much is Enough? Proceeding of the Symposium

Presented at the Canadian Nutrition Congress. ; June 21, 2007; ; 2008.

(143) Food Manufacturers - Heart&Stroke Health Check™ Program. 2010; Available at:

http://www.healthcheck.org/page/licensee-overview-0.

(144) Kraft Canada Inc. Sensible Solution. 2011; Available at: http://www.kraftcanada.com/en/healthy-

living-ideas/sensiblesolution/sensiblesolution.aspx. Accessed July 12, 2011.

(145) Heart and Stroke Foundation. Health Check™ Nutrient Criteria: December 2009. 2009.

(146) Old Dutch Foods I. Snack Wise. 2011; Available at: http://www.olddutchfoods.com/nutrition-

highlights/snack-wise. Accessed October 3, 2011.

(147) Viera AJ, Garrett JM. Understanding Interobserver Agreement: The Kappa Statistic. Family

Medicine 2005;37(5):360-363.

(148) Heart and Stroke Foundation. Health Check™ Nutrient Criteria: July 2009. 2009.

(149) Heart and Stroke Foundation. Health Check Nutrient Criteria: January 2011. 2011.

(150) Pantazaopoulos P, Kwong K, Lillycrop W, Gao Y, Samadhin M, Ratnayake WM, et al. Trans and

saturated fat on food labels in Canada: fact or fiction? Can J Public Health 2011;102(4):313-316.

(151) World Health Organization. Preventing chronic diseases: a vital investment. 2005.

(152) Mintel Global Market Navigator. Grocery Retailing - Canada - a snapshot. 2010.

126

(153) Emrich TE, Qi Y, Mendoza JE, Lou W, Cohen JE, L'Abbe MR. Consumer perceptions of the

Nutrition Facts table and Front-of-Pack nutrition rating systems. Appl Physiol Nutr Metab 2014;39:1-8.

(154) Arcand J, Mendoza J, Qi Y, Henson S, Lou W, L'Abbe MR. Results of national survey examining

Canadians' concern, actions, barriers, and support for dietary sodium reduction interventions. Can J

Cardiol 2013 May;29(5):628-631.

(155) Emrich TE, Mendoza JE, L'Abbe MR. Effectiveness of Front-of-pack nutrition symbols: A pilot

study with consumers. Can J Diet Pract Res 2012;73(4).

(156) Blitstein JL, Evans WD. Use of Nutrition Facts Panels among Adults Who Make Household Food

Purchasing Decisions. Journal of Nutrition Education and Behavior 2006;38(6):360-364.

(157) Campos S, Doxey J, Hammond D. Nutrition labels on pre-packaged foods: a systematic review.

Public Health Nutr 2011 August;14:1506.

(158) Emrich TE, Cohen JE, Lou WY, L'Abbe MR. Food products qualifying for versus carrying a front-

of-pack nutrition rating system: a cross-sectional study. BMC Public Health 2013;13(846).

(159) Health Canada. Consulting Canadians to Modermize and Improve Food Labels. 2014; Available at:

http://www.hc-sc.gc.ca/fn-an/alt_formats/pdf/label-etiquet/modernize-report-moderniser-rapport-eng.pdf.

Accessed July 12, 2014.

(160) Savoie N, Barlow K, Harvey KL, Binnie MA, Pasut L. Consumer Perceptions of Front-of-packahe

Labelling Systems and Healthiness of Foods. Canadian Journal of Public Health 2013;104(5):e359-e363.

(161) Heart and Stroke Foundation. Backgrounder - Health Check. 2010.

(162) Health Canada. Nutrition Labelling Consultation. 2014; Available at: http://www.hc-sc.gc.ca/fn-

an/label-etiquet/consultation/index-eng.php. Accessed July 16, 2014.

(163) The Trans Fat Task Force. TRANSforming the Food Supply: Report of the Trans Fat Task Force

Submitted to the Minister of Health. 2006.

(164) von Tigerstrom B. Food Labelling for Healthier Eating: Is Front-of-package Labelling the Answer?

Manitoba Law Journal 2009;33(1).

(165) Roodenburg AJC, Temme EHM, Davies OH, Seidell JC. Potential impact of the Choices

Programme on nutrient intakes in the Dutch population. Nutr Bull 2009 September 2009;34(3):318-323.

127

(166) Temme EHM, van der Voet H, Roodenburg AJC, Bulder A, van Donkersgoed G, van Klaveren J.

Impact of foods with health logo on saturated fat, sodium and sugar intake of young dutch adults. Public

Health Nutr 2010;14(4):635-644.

(167) Chiuve SE, Sampson L, Willet WC. Adherence to the Overall Nutritional Quality Index and Risk of

Total Chronic Disease. Am J Prev Med 2011 May;40(5):505-513.

(168) Vyth EL, Hendriksen MAH, Roodenburg AJC, Steenhuis IHM, van Raaij JMA, Verhagen H, et al.

Consuming a diet complying with front-of-pack label criteria may reduce cholesterol levels: a modeling

study. Eur J Clin Nutr 2012;66:510-516.

(169) Lichtenstein AH, Carson JS, Johnson RK, Kris-Etherton PM, Pappas A, Rupp L, et al. Food-intake

patterns assessed by using front-of-pack labeling program criteria associated with better diet quality and

lower cardiometabolic risk. Am J Clin Nutr 2014;99:454-462.

(170) Vyth EL, Steenhuis IHM, Brandt HE, Roodenburg AJC, Brug J, Seidell JC. Methodological quality

of front-of-pack labeling studies: a review plus identification of research challenges. Nutrition Reviews

2012;70(12):709-720.

128

Appendices

129

Appendix A

Proportion of food products that qualified for Heal th Check™ compared to the proportion of food produc ts that

carried the system's symbol by subcategory (N=7503)

Category N Products carrying Health Check™ N (%)

Products qualifying for Health Check™ N (%)

P-Value* Kappa Statistic (confidence interval)†

Vegetables & Fruit

Fruit Juices

Fresh Fruit

Frozen Fruit

Canned Fruit

Apple and other fruit sauces

Dried Fruit Pieces

Dried Fruit Snacks

Fresh and Frozen Vegetables (plain)

Canned Vegetables (plain)

Canned Tomatoes

Frozen and Canned Vegetables (seasoned, sauced)

Tomato Juice

Vegetable Juices and Blends

Tomato Paste

Minor Main Entrée Vegetable-Based Sauces

554

0

6

129

33

69

38

171

162

60

27

6

35

11

93

47

-

0

19

6

7

14

25

7

12

1

1

6

0

1

(8.5)

(-)

(0.0)

(14.7)

(18.2)

(10.1)

(36.8)

(14.6)

(4.3)

(20.0)

(3.7)

(16.7)

(17.1)

(0.0)

(1.1)

317

-

6

115

18

68

14

171

160

58

6

5

18

10

49

(57.2)

(-)

(100.0)

(89.2)

(54.6)

(98.6)

(36.8)

(100.0)

(98.8)

(96.7)

(22.2)

(83.3)

(51.4)

(90.9)

(52.7)

<.0001

-

0.0313

<.0001

0.0005

<.0001

.

<.0001

<.0001

<.0001

0.0625

0.1250

0.0005

0.0020

<.0001

0.1(0.1-0.2)

-

0.0 (0.0-0.0)

0.0 (0.0-0.1)

0.3(0.1-0.5)

0.0 (-0.0-0.0)

1.0 (1.0-1.0)

0.0 (0.0-0.0)

0.0 (-0.0-0.0)

0.0 (-0.0-0.0)

0.2 (-0.2-0.6)

0.1 (-0.1-0.3)

0.3 (0.1-0.6)

-0.0 (-0.0--0.0)

0.0 (-0.0-0.1)

130

Category N Products carrying Health Check™ N (%)

Products qualifying for Health Check™ N (%)

P-Value* Kappa Statistic (confidence interval)†

Vegetable-Based Dips

Frozen Fruit Bars

23

7

0

0

(0.0)

(0.0)

6

0

(26.1)

(0.0)

0.0313

-

0.0 (0.0-0.0)

-

Grain Products

Bread

Bread Products

Hot Breakfast Cereals

Breakfast Cereals (20-42 g per 250 mL)

Breakfast Cereals (≥43 g per 250 mL)

Very High Fibre Breakfast Cereals

Flour / Grains

Crackers / Rusks

Croutons

Rice Cakes

Waffles / Pancakes

Rice / Grains (plain)

Instant Rice (plain)

Pasta

Side Dishes – Rice, grains or potatoes (seasoned, sauced)

Side Dishes – Pasta or noodles (seasoned, sauced)

Grain-based Bars

185

228

57

79

145

7

6

261

53

62

73

78

5

383

132

104

178

20

17

4

7

7

2

0

28

5

2

2

8

2

32

7

0

7

(10.8)

(7.5)

(7.0)

(8.9)

(4.8)

(28.6)

(0.0)

(10.7)

(9.4)

(3.2)

(2.7)

(10.3)

(40.0)

(8.4)

(5.3)

(0.0)

(3.9)

112

91

40

29

53

7

3

115

6

18

9

78

5

242

31

1

52

(60.5)

(39.9)

(70.2)

(36.7)

(36.6)

(100.0)

(50.0)

(44.1)

(11.3)

(29.0)

(12.3)

(100.0)

(100.0)

(63.2)

(23.5)

(1.0)

(29.2)

<.0001

<.0001

<.0001

<.0001

<.0001

0.0625

0.2500

<.0001

1.0000

<.0001

0.0156

<.0001

0.2500

<.0001

<.0001

1.0000

<.0001

0.1 (0.1-0.2)

0.2 (0.1-0.3)

0.0 (-0.1-0.1)

0.3 (0.1-0.5)

0.1 (0.0-0.2)

0.0 (0.0-0.0)

0.0 (0.0-0.0)

0.2 (0.1-0.3)

0.9 (0.7-1.0)

0.2 (-0.0-0.3)

0.3 (-0.0-0.7)

0.0 (0.0-0.0)

0.0 (-0.0--0.0)

0.1 (0.1-0.1)

0.3 (0.1-0.4)

-0.0 (-0.0--0.0)

0.2 (0.1-0.3)

131

Category N Products carrying Health Check™ N (%)

Products qualifying for Health Check™ N (%)

P-Value* Kappa Statistic (confidence interval)†

Muffins / Muffin-Style Bars

Plain Popcorn

57

53

0

3

(0.0)

(5.7)

11

3

(15.8)

(5.7)

0.0010

1.0000

0.0 (0.0-0.0)

0.3 (-0.2-0.8)

Milk & Alternatives

Milk and Milk Based Drinks

Yogurts

Yogurt Based Drinks

Dairy-based dips

Fresh Cheese (plain and flavoured)

Cheese

Soy-based Cheese

Ricotta Cheese (plain)

Cottage Cheese (plain and flavoured)

Plant-based Beverages

82

95

17

48

63

390

5

10

23

76

1

12

0

0

0

9

0

0

0

8

(1.2)

(12.6)

(0.0)

(0.0)

(0.0)

(2.3)

(0.0)

(0.0)

(0.0)

(10.5)

57

49

11

37

0

46

0

1

18

61

(69.5)

(51.6)

(64.7)

(77.1)

(0.0)

(11.8)

(0.0)

(10.0)

(78.3)

(80.3)

<.0001

<.0001

0.0010

<.0001

-

<.0001

-

1.0000

<.0001

<.0001

0.0 (-0.0-0.0)

0.2 (0.0-0.3)

0.0 (0.0-0.0)

0.0 (0.0-0.0)

-

0.3 (0.1-0.5)

-

-0.0 (-0.0-0.0)

0.0 (0.0-0.0)

0.1 (0.0-0.1)

Meat & Alternatives

Meats / Poultry (plain, seasoned, coated)

Meats / Poultry (with sauce)

Ground Meats (plain, seasoned)

Patties, meatballs, etc.

Sausages

Deli Meats / Ham

Fish and Seafood (plain)

45

103

6

126

117

213

60

0

4

1

6

1

4

0

(0.0)

(3.9)

(16.7)

(4.8)

(0.9)

(1.9)

(0.0)

13

18

6

25

5

55

49

(28.9)

(17.5)

(100.0)

(19.8)

(4.3)

(25.8)

(81.7)

0.0002

0.0001

0.0625

<.0001

0.1250

<.0001

<.0001

-0.0 (-0.0--0.0)

0.3 (0.1-0.6)

0.0 (0.0-0.0)

0.3 (0.1-0.5)

0.3(-0.2-0.8)

0.1 (0.0-0.2)

0.0 (0.0-0.0)

132

Category N Products carrying Health Check™ N (%)

Products qualifying for Health Check™ N (%)

P-Value* Kappa Statistic (confidence interval)†

Fish and Seafood (seasoned or coated)

Fish and Seafood (sauced)

Canned Fish and Seafood (packed in broth or water)

Canned Fish and Seafood (seasoned, sauced)

Processed Fish

Dried Legumes

Frozen and Canned Legumes (plain)

Canned Legumes (prepared)

Tofu (plain)

Vegetarian Meat Alternatives

Vegetarian Terrines, Spreads or Pâtés

Eggs

Egg Substitutes

Nuts, Seeds or Ready to Eat Dried Legumes

Nuts and Seeds Butters

Legume-based dips

104

32

86

45

7

78

95

28

15

28

0

37

0

113

39

33

1

3

1

3

2

1

2

10

3

7

-

17

-

0

1

0

(1.0)

(9.4)

(1.2)

(6.7)

(28.6)

(1.3)

(2.1)

(35.7)

(20.0)

(25.0)

(-)

(46.0)

(-)

(0.0)

(2.6)

(0.0)

21

7

82

26

3

78

37

10

14

17

-

37

-

68

30

23

(18.3)

(21.9)

(95.4)

(57.8)

(42.9)

(100.0)

(39.0)

(35.7)

(93.3)

(60.7)

(-)

(100.0)

(-)

(60.2)

(76.9)

(69.7)

<.0001

0.1250

<.0001

<.0001

1.0000

<.0001

<.0001

1.0000

0.0009

0.0010

-

<.0001

-

<.0001

<.0001

<.0001

0.1 (-0.0-0.2)

0.5 (0.2-0.9)

0.0 (-0.0-0.0)

0.1 (-0.0-0.2)

0.7 (0.2-1.0)

0.0 (0.0-0.0)

0.1 (-0.0-0.2)

0.8 (0.6-1.0)

0.0 (-0.0-0.1)

0.4 (0.1-0.6)

-

-0.0 (-0.0--0.0)

-

-0.0 (-0.0--0.0)

0.0 (-0.0-0.0)

-0.0 (-0.0--0.0)

Oils & Fats

Oils

Margarines

Light Margarines

105

47

5

4

10

2

(3.8)

(21.3)

(40.0)

67

39

4

(63.8)

(82.9)

(80.0)

<.0001

<.0001

0.5000

0.0 (0.0-0.0)

0.1 (0.0-0.2)

0.3 (-0.3-0.8)

133

Category N Products carrying Health Check™ N (%)

Products qualifying for Health Check™ N (%)

P-Value* Kappa Statistic (confidence interval)†

Salad dressings 267 17 (6.4) 128 (47.9) <.0001 0.1 (0.0-0.2)

Combination Foods

Soups

Dinners and Entrees / Mixed Dishes

Pizza

Vegetarian or Meat Pies

Tofu or Meat or Fish with vegetables

Stuffed Pasta

Stuffed Meat

Major Main Entrée Sauce

Potato and Pasta Salads

Other Salads

Dried Fruit and Nut Mixture

Nut and/or Seed Bars (with or without dried fruit)

337

427

119

39

65

26

37

143

17

43

37

0

74

26

2

0

6

2

2

16

0

3

0

-

(22.0)

(6.1)

(1.7)

(0.0)

(9.2)

(7.7)

(5.4)

(11.2)

(0.0)

(7.0)

(0.0)

(-)

94

81

4

0

1

3

4

62

1

17

28

-

(27.9)

(19.0)

(3.4)

(0.0)

(1.5)

(11.5)

(10.8)

(43.4)

(5.9)

(39.5)

(75.7)

(-)

0.0005

<.0001

0.5000

-

0.1250

1.0000

0.6250

<.0001

1.0000

0.0001

<.0001

-

0.7 (0.7-0.8)

0.2 (0.1-0.3)

0.7 (.2-1.0)

-

-0.0 (-0.0-0.0)

0.8 (0.3-1.0)

0.3 (-0.2-0.8)

0.3 (0.2-0.4)

0.0 (0.0-0.0)

0.2 (-0.0-0.4)

0.0 (-0.0--0.0)

-

*P-Value for exact McNemar’s test to compare paired proportions; testing whether the proportion of products qualifying for the FOP

symbol is statistically different from the proportion carrying the FOP.

† The kappa coefficient measures the difference between observed agreement and expected agreement and lies on a scale of -1 to 1, where

0.0 is considered ‘poor’ agreement, 0.2 ‘slight’, 0.4 ‘fair’, 0.6 ‘moderate’, 0.8 ‘substantial’, and 1.0 ‘almost perfect’ agreement.

“- ” Indicates that there were no products available for a meaningful calculation.

134

Appendix B

Proportion of food products that qualified for Sens ible Solutions™ compared to the proportion of food products that

carried the systems symbol by subcategory (N=3009)

Category N Products Carrying Sensible Solutions™ N (%)

Absolute threshold criteria Based on Relative threshold criteria Products qualifying for Sensible Solutions™ N (%)

P-Value*

Kappa coefficient (confidene interval) †

Products qualifying for Sensible Solutions™ N (%)

P-Value*

Kappa coefficient†

Beverages

100% Juice‡

Refreshment Beverages

317

238

0

15

(0.0)

(6.3)

216

4

(68.1)

(1.7)

<.0001

0.0074

0.0 (0.0-0.0)

0.2 (-0.1-0.4)

-

238

(-)

(100.0)

-

<.0001

-

0.0 (0.0-0.0)

Cheese and Dairy

Natural and processed cheese

Cream cheese

Grated parmesan, Romano cheese, Cheese spreads and dip§

413

57

69

5

0

0

(1.2)

(0.0)

(0.0)

17

0

-

(4.1)

(0.0)

(-)

0.0169

-

-

-0.0 (-0.0--0.0)

-

-

404

40

64

(97.8)

(70.1)

(92.8)

<.0001

<.0001

<.0001

0.0 (-0.0-0.0)

0.0 (0.0-0.0)

-0.0 (-0.0--0.0)

Desserts 142 3 (2.1) 0 (0.0) 0.2500 -0.0 (-0.0--0.0) 103 (72.5) <.0001 0.0 (-0.0-0.0)

135

Category N Products Carrying Sensible Solutions™ N (%)

Absolute threshold criteria Based on Relative threshold criteria Products qualifying for Sensible Solutions™ N (%)

P-Value*

Kappa coefficient (confidene interval) †

Products qualifying for Sensible Solutions™ N (%)

P-Value*

Kappa coefficient†

Cookies and Crackers

536 77 (14.4) 67 (12.5) 0.3682 0.2 (0.1-0.3) 479 (89.4) <.0001 0.0 (0.0-0.0)

Salad Dressings 228 0 (0.0) 1 (0.44) 1.0000 0.0 (0.0-0.0) 215 (94.3) <.0001 0.0 (0.0-0.0)

Mayonnaise and Miracle Whip

38 3 (7.9) 0 (0.0) 0.2500 0.0 (0.0-0.0) 35 (92.1) <.0001 0.0 (-0.0-0.0)

Convenient Meals 945 13 (1.4) 432 (45.7) <.0001 0.0 (0.0-0.0) 777 (82.2) <.0001 0.0 (0.0-0.0)

Peanut Butter 26 6 (23.1) 0 (0.0) 0.0313 0.0 (0.0-0.0) 24 (92.3) <.0001 0.0 (-0.0-0.1)

*P-Value for exact McNemar’s test to compare paired proportions; testing whether the proportion of products qualifying for the FOP

symbol is statistically different from the proportion carrying the FOP.

†The kappa coefficient measures the difference between observed agreement and expected agreement and lies on a scale of -1 to 1, where

0.0 is considered ‘poor’ agreement, 0.2 ‘slight’, 0.4 ‘fair’, 0.6 ‘moderate’, 0.8 ‘substantial’, and 1.0 ‘almost perfect’ agreement.

‡Sensible Solutions™ does not have relative threshold criteria established for 100% Juice.

§ Sensible Solutions™ does not have absolute threshold criteria established for Grated parmesan, Romano cheese, Cheese spreads and dip.

“-” Indicates that there were no products available for a meaningful calculation.

136

Appendix C

Comparison of the nutrient content per reference am ount of products with and without front-of-pack nut rition rating

symbol nutrition marketing by product subcategory*

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

Bakery Products

Bread, excluding sweet quick, type rolls

No 147 127 [119, 135] 0.3 [0.0, 0.4] 234 [185, 264] 1 [0, 2]†

Yes 36 122 [118, 129] 0.3 [0.3, 0.5] 211 [179, 263] 2 [1, 2]

Bagels, tea biscuits, scones, rolls, buns, croissants, tortillas, soft bread sticks, soft pretzels and corn bread

No 179 147 [138, 162] 0.3 [0.2, 0.7] 266 [213, 324] 2 [1, 3]

Yes 48 146 [138, 154] 0.3 [0.1, 0.5] 217 [192, 256] 2 [1, 3]

Brownies

No 19 181 [160, 189] 2.3 [1.6, 4.2] 100 [76, 126] 15 [13, 16]

Yes 2 146 [133, 158] 1.6 [1.1, 2.1] 113 [111, 116] 15 [14, 15]

Cake [heavy weight] No 47 438 [388, 463] 12.5 [8.8, 15.0] 263 [225, 330] 33 [28, 39]

Yes 2 210 [210, 210]† 2.0 [2.0, 2.0]† 195 [190, 200] 18 [18, 18]†

Cake [medium weight] No 25 307 [288, 320] 2.7 [1.9, 10.1] 216 [107, 320] 22 [17, 36]

Yes 2 191 [145, 237]† 2.8 [2.6, 3.0] 238 [237, 238] 20 [14, 27]

Coffee cakes, donuts, Danishes, sweet rolls, sweet quick, type breads and muffins

No 82 217 [195, 236] 2.2 [1.5, 5.4] 216 [177, 249] 17 [14, 23]

Yes 7 155 [147, 173]† 0.8 [0.8, 2.0]† 165 [141, 220] 10 [9, 18]†

Cookies with or without coating or filling; graham wafers

No 231 144 [139, 155] 3.2 [1.5, 4.1] 83 [55, 110]† 10 [8, 12]

Yes 63 133 [130, 140] 1.4 [0.7, 2.0]† 104 [75, 150] 8 [7, 9]†

137

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

Crackers, hard bread sticks and melba toast

No 140 90 [86, 95] 0.4 [0.2, 1.3] 130 [100, 169] 1 [0, 2]

Yes 98 90 [82, 95] 0.5 [0.3, 1.1] 153 [110, 189] 1 [0, 1]

Dry breads, matzo, and rusks

No 10 110 [100, 126] 0.1 [0.0, 0.5] 177 [92, 252] 1 [0, 2]

Yes 13 130 [120, 130] 0.4 [0.3, 0.5] 170 [120, 200] 2 [1, 2]

Ice cream cones No 1 19 [19, 19] 0.0 [0.0, 0.0] 8 [8, 8] 0 [0, 0]

Yes 1 21 [21, 21] 0.0 [0.0, 0.0] 8 [8, 8] 1 [1, 1]

Croutons No 39 35 [30, 35] 0.0 [0.0, 0.1] 65 [55, 80] 0 [0, 0]

Yes 14 30 [28, 30] 0.0 [0.0, 0.5] 65 [33, 85] 0 [0, 1]

French toast, pancakes, and waffles

No 40 203 [193, 220] 1.1 [0.7, 1.6] 405 [361, 461] 6 [4, 8]

Yes 12 190 [167, 225] 0.7 [0.3, 1.6] 401 [296, 481] 6 [4, 6]

Grain, based bars with filling or partial or full coating

No 67 171 [160, 183] 2.3 [1.3, 3.2] 94 [65, 152] 14 [11, 15]

Yes 26 160 [141, 167] 1.5 [0.5, 1.9]† 98 [92, 131] 13 [9, 14]

Grain, based bars, without filling or coating

No 55 127 [118, 129] 0.5 [0.4, 0.7] 85 [69, 98] 9 [8, 10]

Yes 30 120 [111, 129] 0.5 [0.4, 1.1] 79 [60, 99] 9 [7, 10]

Rice cakes and corn cakes No 39 64 [60, 64] 0.2 [0.0, 0.3] 136 [59, 192] 1 [1, 1]

Yes 23 68 [64, 68] 0.2 [0.0, 0.3] 143 [56, 204] 1 [0, 4]

Pies, tarts, cobblers, turnovers, other pastries

No 90 331 [300, 380] 7.5 [6.1, 7.9] 204 [180, 259] 19 [17, 27]

Yes 4 158 [149, 171]† 0.0 [0.0, 0.5]† 270 [156, 270] 13 [10, 17]†

Pie crust No 12 143 [129, 150] 3.1 [2.3, 3.7] 107 [96, 142] 0 [0, 2]

Yes 1 133 [133, 133] 1.7 [1.7, 1.7] 100 [100, 100] 1.7 [1.7, 1.7]

Cereals and Hot breakfast cereals, such No 23 150 [149, 158] 0.4 [0.3, 0.4] 183 [115, 243] 12 [8, 13]

138

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

other grain products

as oatmeal, or cream of wheat

Yes 34 156 [150, 158] 0.4 [0.3, 0.5] 173 [111, 219] 8 [0, 11]†

Ready-to-eat breakfast cereals, puffed and coated, flaked, extruded, without fruit or nuts, very high fibre cereals

No 47 116 [111, 120] 0.0 [0.0, 0.3]† 130 [125, 200] 7 [4, 10]

Yes 38 113 [110, 120] 0.2 [0.1, 0.3] 168 [140, 214] 6 [4, 10]

Ready-to-eat breakfast cereals, fruit and nut type, granola

No 63 212 [200, 226] 0.4 [0.2, 1.4] 135 [50, 220] 11 [9, 14]

Yes 82 214 [202, 220] 0.4 [0.2, 0.6] 145 [50, 238] 11 [9, 14]

Bran and wheat germ No 1 50 [50, 50] 0.1 [0.1, 0.1] 0 [0, 0] 0 [0, 0]

Yes 1 35 [35, 35] 0.0 [0.0, 0.0] 0 [0, 0] 0 [0, 0]

Grains, such as rice or barley

No 73 160 [160, 162] 0.0 [0.0, 0.2] 0 [0, 1] 0 [0, 0]

Yes 12 165 [160, 185] 0.1 [0.0, 0.3] 3 [0, 13] 0 [0, 0]

Pastas without sauce No 322 300 [300, 307] 0.2 [0.0, 0.3]† 0 [0, 2]† 2 [1, 3]†

Yes 61 310 [310, 310] 0.3 [0.3, 0.5] 2 [0, 2] 3 [2, 3]

Stuffing No 13 167 [144, 208] 1.3 [1.3, 3.1] 557 [485, 613] 3 [2, 3]

Yes 1 167 [167, 167] 1.3 [1.3, 1.3] 359 [359, 359] 2 [2, 2]

Combination dishes

Measureable with a cup, such as casserole, hash, macaroni and cheese with or without meat, pot pie, spaghetti with sauce, stir fry, meat and poultry casserole, baked and refried beans, wieners and

No 442 316 [264, 350] 2.2 [0.8, 3.7] 820 [666, 940] 5 [2, 8]

Yes 129 275 [242, 301] 1.3 [0.6, 2.5]† 580 [528, 700]† 6 [4, 10]

139

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

beans, meat chilli, chilli with beans, creamed chipped beef, beef or poultry ravioli in sauce, beef stroganoff, poultry a la king, Brunswick stew, goulash, stew, ragout or poutine

Not measureable with a cup, such as burritos, egg rolls, enchiladas, pizza, pizza rolls, sausage rolls, pastry rolls, cabbage rolls, quiche, sandwiches, crackers and meat or poultry lunch-type packages, gyros, burger on a bun, frank on a bun, calzones, tacos, pockets stuffed with meat, lasagna ,chicken cordon bleu, stuffed vegetables with meat or poultry, shish kabobs, empanadas, fajitas, souvlaki, meat pie or tourtiere

No 310 306 [251, 345] 4.5 [3.1, 6.0] 644 [524, 799] 4 [2, 6]

Yes 59 195 [148, 272]† 1.9 [1.2, 2.7]† 468 [325, 558]† 4 [3, 6]

Hors d’oeuvres No 100 125 [96, 150] 2.1 [0.6, 3.8] 214 [188, 251] 1 [1, 2]†

Yes 4 88 [88, 88]† 0.1 [0.1, 0.2]† 156 [148, 177]† 2 [2, 2]

Dairy Cheese, including cream No 348 100 [86, 120] 5.0 [3.8, 6.0] 200 [160, 230] 0 [0, 1]

140

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

products and substitutes

cheese and cheese spread Yes 32 86 [65, 90] 3.5 [2.4, 3.6]† 210 [200, 240] 0 [0, 1]

Cottage cheese No 14 100 [100, 110] 1.0 [0.5, 1.5] 420 [305, 540] 6 [5, 6]

Yes 9 111 [100, 122] 0.6 [0.6, 0.6] 299 [277, 300]† 9 [5, 11]

Cheese used as an ingredient, such as dry cottage cheese or ricotta cheese

No 9 86 [80, 100] 4.0 [2.4, 4.7] 71 [70, 71] 2 [2, 2]

Yes 1 50 [50, 50] 1.5 [1.5, 1.5] 70 [70, 70] 2 [2, 2]

Quark, fresh cheese and fresh dairy desserts

No 62 286 [265, 300] 13.3 [11.7, 15.0] 900 [464, 1000] 0 [0, 0]

Yes 1 214 [214, 214] 7.1 [7.1, 7.1] 393 [393, 393] 4 [4, 4]

Cream and cream substitute

No 22 30 [20, 50] 1.0 [0.4, 3.0] 7.5 [5.0, 10.0] 1 [0, 3]

Yes 1 50 [50, 50] 3.0 [3.0, 3.0] 0 [0, 0] 0 [0, 0]

Milk, evaporated or condensed

No 16 20 [15, 60] 0.5 [0.2, 0.8] 16 [15, 20] 1 [1, 11]

Yes 1 65 [65, 65] 1.0 [1.0, 1.0] 18 [18, 18] 11 [11, 11]

Plant-based beverages, milk, buttermilk and milk-based drinks, such as chocolate milk

No 138 130 [100, 160] 1.0 [0.3, 2.5] 120 [100, 140] 12 [10, 19]

Yes 27 130 [110, 188] 0.6 [0.4, 2.5] 120 [110, 135] 12 [10, 29]

Yogurt No 69 140 [110, 158] 1.8 [0.1, 2.5] 93 [85, 105] 21 [9, 23]

Yes 26 125 [84, 158] 0.5 [0.0, 1.8]† 97 [88, 108] 20 [7, 23]

Fats and oils Butter, margarine, shortening, lard

No 72 70 [70, 70] 2.5 [1.5, 5.0] 60 [0, 70] 0 [0, 0]

Yes 19 70 [70, 70] 1.0 [1.0, 1.0]† 70 [60, 70] 0 [0, 0]

Vegetable oil No 97 80 [80, 80] 1.3 [1.0, 1.5] 0 [0, 0] 0 [0, 0]

141

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

Yes 8 80 [80, 80] 1.0 [0.8, 1.0] 0 [0, 0] 0 [0, 0]

Dressing for salad No 196 100 [60, 120] 1.0 [0.6, 2.0] 300 [240, 360] 2 [1, 4]

Yes 32 60 [30, 90]† 0.8 [0.0, 1.0] 280 [245, 320] 2 [2, 4]

Mayonnaise, sandwich spread and mayonnaise-type dressing

No 19 60 [40, 100] 0.5 [0.5, 1.0] 115 [90, 130] 1 [0, 2]

Yes 20 48 [30, 50] 0.5 [0.3, 1.0] 130 [115, 140] 1 [0, 2]

Oil, spray type No 11 4 [4, 4]† 0.0 [0.0, 0.0] 0 [0, 0] 0 [0, 0]

Yes 2 5 [5, 5] 0.0 [0.0, 0.0] 0 [0, 0] 0 [0, 0]

Fruit and fruit juices

Fruit, fresh, canned or frozen, except those listed as separate item

No 118 96 [84, 108] 0.0 [0.0, 0.0] 10 [0, 18] 18 [16, 23]

Yes 50 84 [64, 108] 0.0 [0.0, 0.0] 0 [0, 12]† 17 [12, 22]

Dried fruit, such as raisins, dates or figs

No 62 120 [110, 130] 0.0 [0.0, 0.0] 0 [0, 5] 23 [15, 26]

Yes 7 140 [140, 160] 0.0 [0.0, 0.0] 0 [0, 140] 26 [20, 26]

Juices, nectars and fruit drinks represented for use as substitutes for fruit juices

No 422 120 [110, 130] 0.0 [0.0, 0.0] 13 [5, 25]† 26 [23, 30]

Yes 132 120 [110, 140] 0.0 [0.0, 0.0] 25 [19, 35] 26 [22, 29]

Meat, poultry, their products and substitutes

Luncheon meats; pate, sandwich spread, potted meat food product; taco fillings; meat pie fillings and cretons

No 99 60 [55, 85] 0.5 [0.3, 1.3] 523 [480, 600] 1 [0, 1]

Yes 9 62 [57, 69] 0.2 [0.2, 0.3]† 314 [256, 322]† 0 [0, 1]

Sausage products No 118 148 [127, 173] 4.4 [3.0, 5.5] 528 [468, 602] 1 [0, 1]

Yes 6 85 [81, 99]† 1.3 [1.1, 1.8]† 412 [338, 587] 1 [1, 2]

142

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

Cuts of meat and poultry without sauce, and ready-to-cook cuts, with or without breading or batter, including marinated, tenderized and injected cuts

No 34 240 [200, 259] 3.0 [1.3, 4.0] 568 [450, 750] 0 [0, 1]

Yes 13 150 [125, 170]† 0.5 [0.4, 1.0]† 169 [106, 450]† 0 [0, 1]

Patties, cutlettes, chopettes, steakettes, meatballs, sausage meat and ground meat, with or without breading or batter

No 109 210 [160, 242] 4.2 [1.6, 7.0] 414 [311, 560] 0 [0, 1]

Yes 35 150 [120, 171]† 1.5 [0.6, 3.5]† 375 [301, 410] 1 [0, 2]

Cured meat products No 32 98 [70, 126] 2.0 [1.0, 2.9] 539 [476, 789] 1 [0, 1]

Yes 1 91 [91, 91] 0.8 [0.8, 0.8] 501 [501, 501] 0 [0, 0]

Canned meat and poultry No 25 74 [63, 85] 1.6 [1.1, 1.9] 413 [370, 487] 0 [0, 0]

Yes 6 69 [68, 70] 0.4 [0.2, 1.0]† 241 [167, 328]† 0 [0, 0]

Meat and poultry with sauce, such as meat in barbecue sauce or turkey with gravy, but excluding combination dishes

No 90 266 [210, 300] 4.2 [2.1, 5.6] 782 [605, 1092] 4 [1, 11]

Yes 16 134 [123, 145]† 0.5 [0.4, 0.6]† 442 [355, 475]† 1 [0, 2]†

Snacks Chips, pretzels, popcorn, extruded snacks, grain-based snack mixes and fruit-based snacks, such as fruit chips

No 297 250 [227, 268] 1.5 [1.0, 2.0] 330 [240, 425] 1 [0, 2]†

Yes 78 205 [176, 238] 0.7 [0.0, 1.0]† 261 [74, 360] 2 [0, 9]

Nuts or seeds for use as No 79 288 [240, 320] 3.5 [2.5, 5.0] 75 [10, 170] 3 [2, 13]

143

Food category Subcategory FOP status

N Calories [kcal] Saturated fat [g]

Sodium[mg] Sugar [g]

snacks Yes 9 311 [310, 311] 4.5 [4.4, 5.0] 156 [40, 156] 2 [2, 3]

Soups Soups No 244 90 [60, 150] † 0.5 [0.0, 1.5] 740 [650, 898] 2 [1, 4] †

Yes 90 120 [90, 150] 0.5 [0.2, 1.0] 625 [480, 650] 4 [2, 7]

Vegetables Vegetables without sauce No 250 30 [25, 50] 0.0 [0.0, 0.0] 170 [20, 270] 2 [1, 4]

Yes 84 35 [23, 50] 0.0 [0.0, 0.0] 65 [13, 290] 3 [1, 4]

Vegetables with sauce No 25 35 [21, 45] 0.0 [0.0, 0.0] 680 [486, 760] 1 [0, 4]

Yes 2 48 [35, 60] 0.0 [0.0, 0.0] 175 [0, 350] 4 [2, 6]

Lettuce and sprouts No 27 13 [10, 15] 0.0 [0.0, 0.0] 20 [13, 47] 1 [0, 1]

Yes 17 15 [11, 15] 0.0 [0.0, 0.0] 15 [11, 34] 1 [0, 2]

Vegetable juice and vegetable drink

No 25 60 [50, 60] 0.0 [0.0, 0.0] 640 [350, 640] 10 [8, 10]

Yes 16 50 [50, 60] 0.0 [0.0, 0.0] 415 [133, 565]† 8 [6, 9]

Olives No 46 20 [18, 25] 0.3 [0.2, 0.4] 237 [176, 280] 0 [0, 0]

Yes 1 17 [17, 17] 0.0 [0.0, 0.0] 131 [131, 131] 0 [0, 0]

Vegetable pastes, such as tomato paste

No 7 20 [20, 35] 0.0 [0.0, 0.0] 20 [20, 30] 3 [3, 5]

Yes 4 23 [20, 25] 0.0 [0.0, 0.0] 150 [20, 290] 3 [3, 3]

Vegetable sauce or puree, such as tomato sauce or tomato puree

No 6 24 [15, 30] 0.0 [0.0, 0.0] 200 [10, 250] 3 [2, 3]

Yes 10 29 [24, 30] 0.0 [0.0, 0.0] 200 [20, 226] 3 [3, 3]

*Front-of-pack nutrition rating symbol nutrition marketing includes: 1) nutrient-specific symbols based on claim criteria, 2) summary

indicator symbols, 3) food group information symbols, and, 4) hybrid symbols; and, excludes nutrient-specific systems that display the

amount of calories and select nutrients per serving. All data are presented as Median and Interquartile range [Q1, Q3]. Calorie and

nutrient amounts are expressed per reference amount rounded to the number of decimal places provided in the Nutrition Facts table.

144

Reference amounts are reference serving size amounts found in Schedule M of Canada’s Food and Drug Regulations and are the basis of

the criteria for making nutrient content and health claims in Canada.

† Statistically significant (p<0.05) and nutritionally relevant (≥25%) difference between products with and without a front-of-pack symbol

in the amount of calories or nutrient of interest.

145

Copyright Acknowledgements

™The Health Check logo, Health Check word mark, and the Heart and Stroke Foundation word

mark are trademarks of the Heart and Stroke Foundation of Canada used under license.